Study of Bevacizumab + mFOLFOX6 Versus Bevacizumab + FOLFIRI With Biomarker Stratification in Participants With Previously Untreated Metastatic Colorectal Cancer (mCRC) (MAVERICC)

• ClinicalTrials.gov Identifier: NCT01374425
• Recruitment Status: Completed
• First Posted: June 16, 2011
• Results First Posted: August 11, 2016
• Last Update Posted: August 11, 2016
• Sponsor: Genentech, Inc.
• Information provided by (Responsible Party): Genentech, Inc.

• Study Type: Interventional
• Study Design: Allocation: Randomized; Intervention Model: Parallel Assignment; Masking: None (Open Label); Primary Purpose: Treatment
• Condition: Colorectal Cancer
• Interventions: Drug: 5-Fluorouracil; Drug: Bevacizumab; Drug: Irinotecan; Drug: Leucovorin; Drug: Oxaliplatin; Drug: Capecitabine
• Enrollment: 376

Participant Flow

Recruitment Details
Pre-assignment Details	The trial included a 21-day Screening period during which participants provided information for demographics, medical history, and cancer/treatment history and completed urinalysis collection.

Arm/Group Title	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
Arm/Group Description	Participants with untreated metastatic colorectal cancer (mCRC) who were candidates for first-line therapy received bevacizumab plus leucovorin, 5-fluorouracil, and oxaliplatin (mFOLFOX6) until disease progression or unacceptable toxicity. Bevacizumab was given as 5 milligrams per kilogram (mg/kg), leucovorin as 400 milligrams per meter-squared (mg/m^2), oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via intravenous (IV) infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus leucovorin, 5-fluorouracil, and irinotecan (FOLFIRI) until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.
Period Title: Overall Study
Started	188	188
Completed	0	0
Not Completed	188	188
Reason Not Completed
Radiographic Disease Progression	86	81
Clinical Disease Progression	9	6
Adverse Event	29	18
Death (Progression of Disease)	1	0
Death (Adverse Event)	2	4
Withdrawal by Subject	6	7
Refused Treatment	10	24
Protocol Violation	2	3
Sponsor Decision	0	1
Lost to Follow-up	2	1
Other	41	43

Baseline Characteristics

Arm/Group Title		Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI	Total
Arm/Group Description		Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Total of all reporting groups
Overall Number of Baseline Participants		188	188	376
Baseline Analysis Population Description		Intent-to-Treat (ITT) Population: All randomized participants regardless of receiving any study drug.
Age, Continuous; Mean (Standard Deviation); Unit of measure: Years
	Number Analyzed	188 participants	188 participants	376 participants
		59.2 (10.88)	60.7 (10.66)	59.9 (10.78)
Sex: Female, Male; Measure Type: Count of Participants; Unit of measure: Participants
	Number Analyzed	188 participants	188 participants	376 participants
	Female	66 35.1%	71 37.8%	137 36.4%
	Male	122 64.9%	117 62.2%	239 63.6%

Outcome Measures

1. Primary Outcome

Title	Progression-Free Survival (PFS) According to Response Evaluation Criteria in Solid Tumors (RECIST) Version 1.1
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as greater than or equal to (≥) 20 percent (%) increase in sum of largest diameters (LD) of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 millimeters (mm). Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% confidence interval (CI) was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

Intent-to-treat (ITT) Population

Arm/Group Title	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.
Overall Number of Participants Analyzed	188	188
Median (95% Confidence Interval); Unit of Measure: months
	10.09; (8.80 to 11.56)	12.55; (10.48 to 14.29)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6, Bevacizumab + FOLFIRI
	Comments	Analysis stratified by high/low excision repair cross-complementing (ERCC)-1 level and region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.0555
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.79
	Confidence Interval	(2-Sided) 95%; 0.61 to 1.01
	Estimation Comments	Hazard ratio (HR) (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

2. Primary Outcome

Title	PFS According to RECIST Version 1.1 in Participants With High ERCC-1 Levels
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as ≥20% increase in sum of LD of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 mm. Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 High)	Bevacizumab + FOLFIRI (ERCC-1 High)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level greater than (>) 1.7 × 10^-3 ERCC-1/B-actin messenger ribonucleic acid (mRNA) at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	64	67
Median (95% Confidence Interval); Unit of Measure: months
	9.92; (8.51 to 12.48)	11.17; (9.10 to 17.84)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 High), Bevacizumab + FOLFIRI (ERCC-1 High)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.3944
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.84
	Confidence Interval	(2-Sided) 95%; 0.56 to 1.26
	Estimation Comments	HR (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

3. Primary Outcome

Title	PFS According to RECIST Version 1.1 in Participants With Low ERCC-1 Levels
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as ≥20% increase in sum of LD of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 mm. Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 Low)	Bevacizumab + FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level less than or equal to (≤) 1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	124	120
Median (95% Confidence Interval); Unit of Measure: months
	10.97; (8.54 to 12.29)	12.68; (10.48 to 14.49)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 Low), Bevacizumab + FOLFIRI (ERCC-1 Low)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.0786
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.76
	Confidence Interval	(2-Sided) 95%; 0.55 to 1.03
	Estimation Comments	HR (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

4. Primary Outcome

Title	PFS According to RECIST Version 1.1 in Participants With High ERCC-1 Levels Versus Participants With Low ERCC-1 Levels
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as ≥20% increase in sum of LD of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 mm. Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High)	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	131	244
Median (95% Confidence Interval); Unit of Measure: months
	10.87; (9.10 to 12.68)	11.56; (9.95 to 12.98)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High), Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.9576
	Comments	P-value (relative to ERCC-1 Low subgroup) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.99
	Confidence Interval	(2-Sided) 95%; 0.77 to 1.28
	Estimation Comments	HR (relative to ERCC-1 Low subgroup) was estimated by Cox regression.

5. Primary Outcome

Title	PFS According to RECIST Version 1.1 in Participants With High Vascular Endothelial Growth Factor (VEGF)-A Levels Versus Participants With Low VEGF-A Levels
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as ≥20% increase in sum of LD of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 mm. Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High)	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level >5 picograms per milliliter (pg/mL) at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	185	185
Median (95% Confidence Interval); Unit of Measure: months
	10.02; (8.80 to 11.17)	12.68; (10.9 to 14.26)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High), Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
	Comments	Analysis stratified by high/low ERCC-1 level and region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.1658
	Comments	P-value (relative to VEGF-A Low subgroup) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.19
	Confidence Interval	(2-Sided) 95%; 0.93 to 1.53
	Estimation Comments	HR (relative to VEGF-A Low subgroup) was estimated by Cox regression.

6. Secondary Outcome

Title	PFS According to RECIST Version 1.1 in Participants With High ERCC-1 and High VEGF-A Levels
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as ≥20% increase in sum of LD of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 mm. Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 High, VEGF-A High)	Bevacizumab + FOLFIRI (ERCC-1 High, VEGF-A High)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA and VEGF-A level >5 pg/mL at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA and VEGF-A level >5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	28	31
Median (95% Confidence Interval); Unit of Measure: months
	8.80; (5.91 to 10.02)	11.17; (8.11 to 18.07)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 High, VEGF-A High), Bevacizumab + FOLFIRI (ERCC-1 High, VEGF-A High)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.3019
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.73
	Confidence Interval	(2-Sided) 95%; 0.41 to 1.32
	Estimation Comments	HR (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

7. Secondary Outcome

Title	PFS According to RECIST Version 1.1 in Participants With High ERCC-1 and Low VEGF-A Levels
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as ≥20% increase in sum of LD of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 mm. Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 High, VEGF-A Low)	Bevacizumab + FOLFIRI (ERCC-1 High, VEGF-A Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA and VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA and VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	35	35
Median (95% Confidence Interval); Unit of Measure: months
	12.52; (9.43 to 15.64)	12.68; (8.18 to 20.83)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 High, VEGF-A Low), Bevacizumab + FOLFIRI (ERCC-1 High, VEGF-A Low)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.6035
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.86
	Confidence Interval	(2-Sided) 95%; 0.48 to 1.53
	Estimation Comments	HR (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

8. Secondary Outcome

Title	PFS According to RECIST Version 1.1 in Participants With Low ERCC-1 and High VEGF-A Levels
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as ≥20% increase in sum of LD of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 mm. Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 Low, VEGF-A High)	Bevacizumab + FOLFIRI (ERCC-1 Low, VEGF-A High)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA and VEGF-A level >5 pg/mL at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA and VEGF-A level >5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	66	59
Median (95% Confidence Interval); Unit of Measure: months
	9.76; (8.18 to 12.45)	11.07; (8.18 to 14.29)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 Low, VEGF-A High), Bevacizumab + FOLFIRI (ERCC-1 Low, VEGF-A High)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.4032
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.83
	Confidence Interval	(2-Sided) 95%; 0.54 to 1.28
	Estimation Comments	HR (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

9. Secondary Outcome

Title	PFS According to RECIST Version 1.1 in Participants With Low ERCC-1 and Low VEGF-A Levels
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as ≥20% increase in sum of LD of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 mm. Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 Low, VEGF-A Low)	Bevacizumab + FOLFIRI (ERCC-1 Low, VEGF-A Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA and VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA and VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	55	60
Median (95% Confidence Interval); Unit of Measure: months
	11.10; (8.54 to 13.08)	14.32; (11.56 to 14.98)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 Low, VEGF-A Low), Bevacizumab + FOLFIRI (ERCC-1 Low, VEGF-A Low)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.0647
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.65
	Confidence Interval	(2-Sided) 95%; 0.41 to 1.03
	Estimation Comments	HR (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

10. Secondary Outcome

Title	Overall Survival (OS)
Description	Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. OS was defined as the time from randomization to death. The median duration of OS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population

Arm/Group Title	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.
Overall Number of Participants Analyzed	188	188
Median (95% Confidence Interval); Unit of Measure: months
	23.85; (20.40 to 26.05)	27.47; (24.64 to 36.73)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6, Bevacizumab + FOLFIRI
	Comments	Analysis stratified by high/low ERCC-1 level and region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.0861
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.76
	Confidence Interval	(2-Sided) 95%; 0.56 to 1.04
	Estimation Comments	HR (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

11. Secondary Outcome

Title	OS in Participants With High ERCC-1 Levels
Description	Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. OS was defined as the time from randomization to death. The median duration of OS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 High)	Bevacizumab + FOLFIRI (ERCC-1 High)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	64	67
Median (95% Confidence Interval); Unit of Measure: months
	22.54; (17.02 to 26.05)	26.51; (19.09 to 36.73)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 High), Bevacizumab + FOLFIRI (ERCC-1 High)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.3295
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.80
	Confidence Interval	(2-Sided) 95%; 0.51 to 1.26
	Estimation Comments	HR (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

12. Secondary Outcome

Title	OS in Participants With Low ERCC-1 Levels
Description	Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. OS was defined as the time from randomization to death. The median duration of OS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 Low)	Bevacizumab + FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	124	120
Median (95% Confidence Interval); Unit of Measure: months
	25.53; (20.40 to 28.75)	27.93; (24.97 to 38.44)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 Low), Bevacizumab + FOLFIRI (ERCC-1 Low)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.1519
	Comments	P-value (relative to Bevacizumab + mFOLFOX6) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.74
	Confidence Interval	(2-Sided) 95%; 0.49 to 1.12
	Estimation Comments	HR (relative to Bevacizumab + mFOLFOX6) was estimated by Cox regression.

13. Secondary Outcome

Title	OS in Participants With High ERCC-1 Levels Versus Participants With Low ERCC-1 Levels
Description	Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. OS was defined as the time from randomization to death. The median duration of OS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High)	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	131	244
Median (95% Confidence Interval); Unit of Measure: months
	23.23; (19.09 to 26.94)	27.27; (24.34 to 31.28)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High), Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
	Comments	Analysis stratified by region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.2774
	Comments	P-value (relative to ERCC-1 Low subgroup) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.19
	Confidence Interval	(2-Sided) 95%; 0.87 to 1.62
	Estimation Comments	HR (relative to ERCC-1 Low subgroup) was estimated by Cox regression.

14. Secondary Outcome

Title	Percentage of Participants With Objective Response According to RECIST Version 1.1
Description	Objective response was defined as complete response (CR) or partial response (PR) according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to less than (<) 10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. The percentage of participants with CR or PR was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population

Arm/Group Title	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.
Overall Number of Participants Analyzed	188	188
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	61.2; (54.2 to 68.1)	65.4; (58.6 to 72.2)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6, Bevacizumab + FOLFIRI
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Objective Response Rates
	Estimated Value	4.3
	Confidence Interval	(2-Sided) 95%; -5.5 to 14.0
	Estimation Comments	The difference was calculated as the percentage of participants with objective response in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

15. Secondary Outcome

Title	Percentage of Participants With Objective Response According to RECIST Version 1.1 in Participants With High ERCC-1 Levels
Description	Objective response was defined as CR or PR according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. The percentage of participants with CR or PR was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 High)	Bevacizumab + FOLFIRI (ERCC-1 High)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	64	67
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	56.3; (44.1 to 68.4)	65.7; (54.3 to 77.0)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 High), Bevacizumab + FOLFIRI (ERCC-1 High)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Objective Response Rates
	Estimated Value	9.4
	Confidence Interval	(2-Sided) 95%; 7.2 to 26.1
	Estimation Comments	The difference was calculated as the percentage of participants with objective response in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

16. Secondary Outcome

Title	Percentage of Participants With Objective Response According to RECIST Version 1.1 in Participants With Low ERCC-1 Levels
Description	Objective response was defined as CR or PR according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. The percentage of participants with CR or PR was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 Low)	Bevacizumab + FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	124	120
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	63.7; (55.2 to 72.2)	65.8; (57.3 to 74.3)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 Low), Bevacizumab + FOLFIRI (ERCC-1 Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Objective Response Rates
	Estimated Value	2.1
	Confidence Interval	(2-Sided) 95%; -9.9 to 14.1
	Estimation Comments	The difference was calculated as the percentage of participants with objective response in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

17. Secondary Outcome

Title	Percentage of Participants With Objective Response According to RECIST Version 1.1 in Participants With High ERCC-1 Levels Versus Participants With Low ERCC-1 Levels
Description	Objective response was defined as CR or PR according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. The percentage of participants with CR or PR was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High)	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	131	244
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	61.1; (52.7 to 69.4)	64.8; (58.8 to 70.7)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High), Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Objective Response Rates
	Estimated Value	-3.7
	Confidence Interval	(2-Sided) 95%; -14.0 to 6.6
	Estimation Comments	The difference was calculated as the percentage of participants with objective response in the ERCC-1 High subgroup minus the ERCC-1 Low subgroup. The 95% CI was computed using normal approximation to the binomial distribution.

18. Secondary Outcome

Title	Percentage of Participants With Disease Control According to RECIST Version 1.1
Description	Disease control was defined as CR, PR, or stable disease (SD) according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. SD was defined as neither sufficient shrinkage to quality for PR nor sufficient increase to qualify for disease progression (≥20% increase in sum of LD of target lesions plus absolute increase ≥5 mm) in reference to the smallest sum of LD on study. The percentage of participants with CR, PR, or SD was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population

Arm/Group Title	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.
Overall Number of Participants Analyzed	188	188
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	93.1; (89.5 to 96.7)	91.0; (86.9 to 95.1)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6, Bevacizumab + FOLFIRI
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Disease Control Rates
	Estimated Value	2.1
	Confidence Interval	(2-Sided) 95%; -7.6 to 3.3
	Estimation Comments	The difference was calculated as the percentage of participants with disease control in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

19. Secondary Outcome

Title	Percentage of Participants With Disease Control According to RECIST Version 1.1 in Participants With High ERCC-1 Levels
Description	Disease control was defined as CR, PR, or SD according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. SD was defined as neither sufficient shrinkage to quality for PR nor sufficient increase to qualify for disease progression (≥20% increase in sum of LD of target lesions plus absolute increase ≥5 mm) in reference to the smallest sum of LD on study. The percentage of participants with CR, PR, or SD was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 High)	Bevacizumab + FOLFIRI (ERCC-1 High)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	64	67
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	93.8; (87.8 to 99.7)	85.1; (76.5 to 93.6)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 High), Bevacizumab + FOLFIRI (ERCC-1 High)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Disease Control Rates
	Estimated Value	-8.7
	Confidence Interval	(2-Sided) 95%; -19.1 to 1.7
	Estimation Comments	The difference was calculated as the percentage of participants with disease control in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

20. Secondary Outcome

Title	Percentage of Participants With Disease Control According to RECIST Version 1.1 in Participants With Low ERCC-1 Levels
Description	Disease control was defined as CR, PR, or SD according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. SD was defined as neither sufficient shrinkage to quality for PR nor sufficient increase to qualify for disease progression (≥20% increase in sum of LD of target lesions plus absolute increase ≥5 mm) in reference to the smallest sum of LD on study. The percentage of participants with CR, PR, or SD was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 Low)	Bevacizumab + FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	124	120
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	92.7; (88.2 to 97.3)	95.0; (91.1 to 98.9)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 Low), Bevacizumab + FOLFIRI (ERCC-1 Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Disease Control Rates
	Estimated Value	2.3
	Confidence Interval	(2-Sided) 95%; -3.7 to 8.3
	Estimation Comments	The difference was calculated as the percentage of participants with disease control in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

21. Secondary Outcome

Title	Percentage of Participants With Disease Control According to RECIST Version 1.1 in Participants With High ERCC-1 Levels Versus Participants With Low ERCC-1 Levels
Description	Disease control was defined as CR, PR, or SD according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. SD was defined as neither sufficient shrinkage to quality for PR nor sufficient increase to qualify for disease progression (≥20% increase in sum of LD of target lesions plus absolute increase ≥5 mm) in reference to the smallest sum of LD on study. The percentage of participants with CR, PR, or SD was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High)	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	131	244
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	89.3; (84.0 to 94.6)	93.9; (90.8 to 96.9)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High), Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Disease Control Rates
	Estimated Value	-4.5
	Confidence Interval	(2-Sided) 95%; -10.6 to 1.5
	Estimation Comments	The difference was calculated as the percentage of participants with disease control in the ERCC-1 High subgroup minus the ERCC-1 Low subgroup. The 95% CI was computed using normal approximation to the binomial distribution.

22. Secondary Outcome

Title	Percentage of Participants With Liver Metastasis Resection
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. R1 was defined as the presence of exposed tumor or histologically detected tumor cells at the line of transection, or <1 mm microscopic margins. In the case of use of radiofrequency ablation or cryotherapy, the resection was considered as R1. R2 was defined as macroscopic positive margins or incomplete resection at time of surgery. The percentage of participants with resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis.

Arm/Group Title	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.
Overall Number of Participants Analyzed	67	55
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	14.9; (6.4 to 23.5)	10.9; (2.7 to 19.1)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6, Bevacizumab + FOLFIRI
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Resection Rates
	Estimated Value	-4.0
	Confidence Interval	(2-Sided) 95%; -15.9 to 7.8
	Estimation Comments	The difference was calculated as the percentage of participants with resection in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

23. Secondary Outcome

Title	Percentage of Participants With Complete Liver Metastasis Resection
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. The percentage of participants with R0 resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis.

Arm/Group Title	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.
Overall Number of Participants Analyzed	67	55
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	11.9; (4.2 to 19.7)	5.5; (0.0 to 11.5)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6, Bevacizumab + FOLFIRI
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Complete Resection Rates
	Estimated Value	-6.5
	Confidence Interval	(2-Sided) 95%; -16.3 to 3.3
	Estimation Comments	The difference was calculated as the percentage of participants with complete resection in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

24. Secondary Outcome

Title	Percentage of Participants With Liver Metastasis Resection in Participants With High ERCC-1 Levels
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. R1 was defined as the presence of exposed tumor or histologically detected tumor cells at the line of transection, or <1 mm microscopic margins. In the case of use of radiofrequency ablation or cryotherapy, the resection was considered as R1. R2 was defined as macroscopic positive margins or incomplete resection at time of surgery. The percentage of participants with resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 High)	Bevacizumab + FOLFIRI (ERCC-1 High)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	17	15
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	17.6; (0.0 to 35.8)	6.7; (0.0 to 19.3)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 High), Bevacizumab + FOLFIRI (ERCC-1 High)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Resection Rates
	Estimated Value	-11.0
	Confidence Interval	(2-Sided) 95%; -33.1 to 11.1
	Estimation Comments	The difference was calculated as the percentage of participants with resection in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

25. Secondary Outcome

Title	Percentage of Participants With Complete Liver Metastasis Resection in Participants With High ERCC-1 Levels
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. The percentage of participants with R0 resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 High)	Bevacizumab + FOLFIRI (ERCC-1 High)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	17	15
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	17.6; (0.0 to 35.8)	6.7; (0.0 to 19.3)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 High), Bevacizumab + FOLFIRI (ERCC-1 High)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Complete Resection Rates
	Estimated Value	-11.0
	Confidence Interval	(2-Sided) 95%; -33.1 to 11.1
	Estimation Comments	The difference was calculated as the percentage of participants with complete resection in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

26. Secondary Outcome

Title	Percentage of Participants With Liver Metastasis Resection in Participants With Low ERCC-1 Levels
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. R1 was defined as the presence of exposed tumor or histologically detected tumor cells at the line of transection, or <1 mm microscopic margins. In the case of use of radiofrequency ablation or cryotherapy, the resection was considered as R1. R2 was defined as macroscopic positive margins or incomplete resection at time of surgery. The percentage of participants with resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 Low)	Bevacizumab + FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	124	120
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	10.5; (5.1 to 15.9)	7.5; (2.8 to 12.2)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 Low), Bevacizumab + FOLFIRI (ERCC-1 Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Resection Rates
	Estimated Value	-3.0
	Confidence Interval	(2-Sided) 95%; -10.1 to 4.2
	Estimation Comments	The difference was calculated as the percentage of participants with resection in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

27. Secondary Outcome

Title	Percentage of Participants With Complete Liver Metastasis Resection in Participants With Low ERCC-1 Levels
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. The percentage of participants with R0 resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX6 (ERCC-1 Low)	Bevacizumab + FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	124	120
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	8.9; (3.9 to 13.9)	3.3; (0.1 to 6.5)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX6 (ERCC-1 Low), Bevacizumab + FOLFIRI (ERCC-1 Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Complete Resection Rates
	Estimated Value	-5.5
	Confidence Interval	(2-Sided) 95%; -11.5 to 0.4
	Estimation Comments	The difference was calculated as the percentage of participants with complete resection in the Bevacizumab + FOLFIRI arm minus the Bevacizumab + mFOLFOX6 arm. The 95% CI was computed using normal approximation to the binomial distribution.

28. Secondary Outcome

Title	Percentage of Participants With Liver Metastasis Resection in Participants With High ERCC-1 Levels Versus Participants With Low ERCC-1 Levels
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. R1 was defined as the presence of exposed tumor or histologically detected tumor cells at the line of transection, or <1 mm microscopic margins. In the case of use of radiofrequency ablation or cryotherapy, the resection was considered as R1. R2 was defined as macroscopic positive margins or incomplete resection at time of surgery. The percentage of participants with resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High)	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	131	244
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	4.6; (1.0 to 8.2)	9.0; (5.4 to 12.6)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High), Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Resection Rates
	Estimated Value	-4.4
	Confidence Interval	(2-Sided) 95%; -9.5 to 0.6
	Estimation Comments	The difference was calculated as the percentage of participants with resection in the ERCC-1 High subgroup minus the ERCC-1 Low subgroup. The 95% CI was computed using normal approximation to the binomial distribution.

29. Secondary Outcome

Title	Percentage of Participants With Complete Liver Metastasis Resection in Participants With High ERCC-1 Levels Versus Participants With Low ERCC-1 Levels
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. The percentage of participants with R0 resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High)	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level >1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with ERCC-1 level ≤1.7 × 10^-3 ERCC-1/B-actin mRNA at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	131	244
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	4.6; (1.0 to 8.2)	6.1; (3.1 to 9.2)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 High), Bevacizumab + mFOLFOX/FOLFIRI (ERCC-1 Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Complete Resection Rates
	Estimated Value	-1.6
	Confidence Interval	(2-Sided) 95%; -6.2 to 3.1
	Estimation Comments	The difference was calculated as the percentage of participants with complete resection in the ERCC-1 High subgroup minus the ERCC-1 Low subgroup. The 95% CI was computed using normal approximation to the binomial distribution.

30. Secondary Outcome

Title	PFS According to RECIST Version 1.1 in Participants With Wild-Type V-Ki-ras2 Kirsten Rat Sarcoma Viral Oncogene Homolog (KRAS) Versus Participants With Mutant KRAS
Description	Tumor assessments were performed according to RECIST Version 1.1. Disease progression was defined as ≥20% increase in sum of LD of target lesions in reference to the smallest sum of LD on study, in addition to an absolute increase ≥5 mm. Disease progression was further defined as the last documented progression determined by the Investigator no later than 1 day before initiation of second-line therapy. Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. PFS was defined as the time from randomization to death or disease progression, whichever occurred first. The median duration of PFS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death or disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (KRAS Wild-Type)	Bevacizumab + mFOLFOX/FOLFIRI (KRAS Mutant)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with wild-type KRAS at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with mutant KRAS at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	208	128
Median (95% Confidence Interval); Unit of Measure: months
	12.45; (10.48 to 14.06)	10.94; (8.77 to 12.35)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (KRAS Wild-Type), Bevacizumab + mFOLFOX/FOLFIRI (KRAS Mutant)
	Comments	Analysis stratified by high/low ERCC-1 level and region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.0593
	Comments	P-value (relative to KRAS Wild-Type subgroup) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.29
	Confidence Interval	(2-Sided) 95%; 0.99 to 1.69
	Estimation Comments	HR (relative to KRAS Wild-Type subgroup) was estimated by Cox regression.

31. Secondary Outcome

Title	OS in Participants With High VEGF-A Levels Versus Participants With Low VEGF-A Levels
Description	Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. OS was defined as the time from randomization to death. The median duration of OS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High)	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level >5 pg/mL at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	185	185
Median (95% Confidence Interval); Unit of Measure: months
	22.83; (18.76 to 27.27)	27.93; (24.97 to 36.01)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High), Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
	Comments	Analysis stratified by high/low ERCC-1 level and region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.0020
	Comments	P-value (relative to VEGF-A Low subgroup) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.64
	Confidence Interval	(2-Sided) 95%; 1.20 to 2.24
	Estimation Comments	HR (relative to VEGF-A Low subgroup) was estimated by Cox regression.

32. Secondary Outcome

Title	OS in Participants With Wild-Type KRAS Versus Participants With Mutant KRAS
Description	Death on study included death from any cause occurring no later than 3 months after the last component of study treatment. OS was defined as the time from randomization to death. The median duration of OS was estimated by Kaplan-Meier analysis and expressed in months. The 95% CI was computed using the method of Brookmeyer and Crowley.
Time Frame	From Baseline until death (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (KRAS Wild-Type)	Bevacizumab + mFOLFOX/FOLFIRI (KRAS Mutant)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with wild-type KRAS at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with mutant KRAS at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	208	128
Median (95% Confidence Interval); Unit of Measure: months
	28.75; (24.77 to 36.73)	24.64; (19.98 to 26.94)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (KRAS Wild-Type), Bevacizumab + mFOLFOX/FOLFIRI (KRAS Mutant)
	Comments	Analysis stratified by high/low ERCC-1 level and region of enrollment.
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Statistical Test of Hypothesis	P-Value	0.0955
	Comments	P-value (relative to KRAS Wild-Type subgroup) based on two-sided stratified log rank test and not adjusted for multiple comparisons.
	Method	Log Rank
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.34
	Confidence Interval	(2-Sided) 95%; 0.95 to 1.88
	Estimation Comments	HR (relative to KRAS Wild-Type subgroup) was estimated by Cox regression.

33. Secondary Outcome

Title	Percentage of Participants With Objective Response According to RECIST Version 1.1 in Participants With High VEGF-A Levels Versus Participants With Low VEGF-A Levels
Description	Objective response was defined as CR or PR according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. The percentage of participants with CR or PR was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High)	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level >5 pg/mL at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	185	185
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	60.5; (53.5 to 67.6)	66.5; (59.7 to 73.3)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High), Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Objective Response Rates
	Estimated Value	-5.9
	Confidence Interval	(2-Sided) 95%; -15.7 to 3.8
	Estimation Comments	The difference was calculated as the percentage of participants with objective response in the VEGF-A High subgroup minus the VEGF-A Low subgroup. The 95% CI was computed using normal approximation to the binomial distribution.

34. Secondary Outcome

Title	Percentage of Participants With Objective Response According to RECIST Version 1.1 in Participants With Wild-Type KRAS Versus Participants With Mutant KRAS
Description	Objective response was defined as CR or PR according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. The percentage of participants with CR or PR was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (KRAS Wild-Type)	Bevacizumab + mFOLFOX/FOLFIRI (KRAS Mutant)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with wild-type KRAS at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with mutant KRAS at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	208	128
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	66.3; (59.9 to 72.8)	60.9; (52.5 to 69.4)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (KRAS Wild-Type), Bevacizumab + mFOLFOX/FOLFIRI (KRAS Mutant)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Objective Response Rates
	Estimated Value	-5.4
	Confidence Interval	(2-Sided) 95%; -16.0 to 5.2
	Estimation Comments	The difference was calculated as the percentage of participants with objective response in the KRAS Mutant subgroup minus the KRAS Wild-Type subgroup. The 95% CI was computed using normal approximation to the binomial distribution.

35. Secondary Outcome

Title	Percentage of Participants With Disease Control According to RECIST Version 1.1 in Participants With High VEGF-A Levels Versus Participants With Low VEGF-A Levels
Description	Disease control was defined as CR, PR, or SD according to RECIST Version 1.1. CR was defined as disappearance of all target lesions and short-axis reduction to <10 mm of any pathological lymph nodes. PR was defined as ≥30% decrease in sum of LD of target lesions in reference to sum of LD at Baseline. Confirmation of response at a consecutive assessment was not required. SD was defined as neither sufficient shrinkage to quality for PR nor sufficient increase to qualify for disease progression (≥20% increase in sum of LD of target lesions plus absolute increase ≥5 mm) in reference to the smallest sum of LD on study. The percentage of participants with CR, PR, or SD was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	From Baseline until disease progression; assessed every 6 weeks (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High)	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level >5 pg/mL at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	185	185
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	91.9; (88.0 to 95.8)	93.0; (89.3 to 96.7)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High), Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Disease Control Rates
	Estimated Value	-1.1
	Confidence Interval	(2-Sided) 95%; -6.5 to 4.3
	Estimation Comments	The difference was calculated as the percentage of participants with disease control in the VEGF-A High subgroup minus the VEGF-A Low subgroup. The 95% CI was computed using normal approximation to the binomial distribution.

36. Secondary Outcome

Title	Percentage of Participants With Liver Metastasis Resection in Participants With High VEGF-A Levels Versus Participants With Low VEGF-A Levels
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. R1 was defined as the presence of exposed tumor or histologically detected tumor cells at the line of transection, or <1 mm microscopic margins. In the case of use of radiofrequency ablation or cryotherapy, the resection was considered as R1. R2 was defined as macroscopic positive margins or incomplete resection at time of surgery. The percentage of participants with resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High)	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level >5 pg/mL at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	185	185
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	5.9; (2.5 to 9.4)	9.2; (5.0 to 13.4)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High), Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Resection Rates
	Estimated Value	-3.2
	Confidence Interval	(2-Sided) 95%; -8.6 to 2.1
	Estimation Comments	The difference was calculated as the percentage of participants with resection in the VEGF-A High subgroup minus the VEGF-A Low subgroup. The 95% CI was computed using normal approximation to the binomial distribution.

37. Secondary Outcome

Title	Percentage of Participants With Complete Liver Metastasis Resection in Participants With High VEGF-A Levels Versus Participants With Low VEGF-A Levels
Description	The timing of resective surgery was not defined in the protocol and was left at the discretion of the Investigator. Resection was classified as R0, R1, or R2 following surgery. R0 was defined as complete resection with clear margins ≥1 mm. The percentage of participants with R0 resection of liver or liver plus lymph node metastases was reported. The 95% CI was computed using normal approximation to the binomial distribution.
Time Frame	At time of resective surgery during study (maximum up to 45 months overall)

Outcome Measure Data

Analysis Population Description

ITT Population; only those participants with liver or liver plus lymph node metastases were included in the analysis. The "Number of Participants Analyzed" reflects the number of evaluable participants for the outcome measure.

Arm/Group Title	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High)	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
Arm/Group Description:	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level >5 pg/mL at Baseline were included in separate analyses.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 or bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Depending on the regimen to which the participant was randomized, bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin or irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles. Participants with VEGF-A level ≤5 pg/mL at Baseline were included in separate analyses.
Overall Number of Participants Analyzed	185	185
Measure Type: Number; Number (95% Confidence Interval); Unit of Measure: percentage of participants
	3.2; (0.7 to 5.8)	8.1; (4.2 to 12.0)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A High), Bevacizumab + mFOLFOX/FOLFIRI (VEGF-A Low)
	Comments	[Not Specified]
	Type of Statistical Test	Superiority or Other
	Comments	[Not Specified]
Method of Estimation	Estimation Parameter	Difference in Complete Resection Rates
	Estimated Value	-4.9
	Confidence Interval	(2-Sided) 95%; -9.6 to -0.2
	Estimation Comments	The difference was calculated as the percentage of participants with complete resection in the VEGF-A High subgroup minus the VEGF-A Low subgroup. The 95% CI was computed using normal approximation to the binomial distribution.

Adverse Events

Time Frame	Serious Adverse Events (AEs): Baseline until 3 months after last dose or study discontinuation/termination (up to 45 months overall); Other AEs: Baseline until 28 days after last dose or study discontinuation/termination (up to 45 months overall).
Adverse Event Reporting Description	Safety Population: All randomized participants who received at least one partial or complete dose of study drug.
Arm/Group Title	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
Arm/Group Description	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus mFOLFOX6 until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, oxaliplatin as 85 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to oxaliplatin and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.	Participants with untreated mCRC who were candidates for first-line therapy received bevacizumab plus FOLFIRI until disease progression or unacceptable toxicity. Bevacizumab was given as 5 mg/kg, leucovorin as 400 mg/m^2, irinotecan as 180 mg/m^2, and 5-fluorouracil as 400 mg/m^2 bolus followed by 2400 mg/m^2 continuous 46-hour infusion. All treatments were administered via IV infusion and started on Day 1 of each 2-week cycle. Participants could be transitioned to oral capecitabine in the event of unacceptable toxicity to irinotecan and given as 850 or 1000 mg/m^2 twice a day on Days 1 to 14, with bevacizumab continued in 3-week cycles.
All-Cause Mortality
	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
	Affected / at Risk (%)	Affected / at Risk (%)
Total	--/--	--/--
Serious Adverse Events
	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
	Affected / at Risk (%)	Affected / at Risk (%)
Total	79/185 (42.70%)	87/183 (47.54%)
Blood and lymphatic system disorders
Thrombocytopenia * 1	3/185 (1.62%)	0/183 (0.00%)
Febrile neutropenia * 1	2/185 (1.08%)	4/183 (2.19%)
Anaemia * 1	1/185 (0.54%)	1/183 (0.55%)
Immune thrombocytopenic purpura * 1	1/185 (0.54%)	0/183 (0.00%)
Neutropenia * 1	1/185 (0.54%)	2/183 (1.09%)
Pancytopenia * 1	1/185 (0.54%)	1/183 (0.55%)
Cardiac disorders
Atrial flutter * 1	1/185 (0.54%)	0/183 (0.00%)
Cardio-respiratory arrest * 1	1/185 (0.54%)	0/183 (0.00%)
Left ventricular dysfunction * 1	1/185 (0.54%)	0/183 (0.00%)
Myocardial infarction * 1	1/185 (0.54%)	1/183 (0.55%)
Supraventricular tachycardia * 1	1/185 (0.54%)	0/183 (0.00%)
Atrial fibrillation * 1	0/185 (0.00%)	3/183 (1.64%)
Atrial thrombosis * 1	0/185 (0.00%)	1/183 (0.55%)
Gastrointestinal disorders
Diarrhoea * 1	6/185 (3.24%)	2/183 (1.09%)
Small intestinal obstruction * 1	5/185 (2.70%)	6/183 (3.28%)
Rectal haemorrhage * 1	4/185 (2.16%)	0/183 (0.00%)
Abdominal pain * 1	3/185 (1.62%)	7/183 (3.83%)
Intestinal obstruction * 1	3/185 (1.62%)	3/183 (1.64%)
Vomiting * 1	3/185 (1.62%)	4/183 (2.19%)
Constipation * 1	2/185 (1.08%)	0/183 (0.00%)
Dysphagia * 1	2/185 (1.08%)	0/183 (0.00%)
Large intestine perforation * 1	2/185 (1.08%)	1/183 (0.55%)
Nausea * 1	2/185 (1.08%)	1/183 (0.55%)
Oesophagitis * 1	2/185 (1.08%)	0/183 (0.00%)
Upper gastrointestinal haemorrhage * 1	2/185 (1.08%)	0/183 (0.00%)
Abdominal pain lower * 1	1/185 (0.54%)	0/183 (0.00%)
Ascites * 1	1/185 (0.54%)	0/183 (0.00%)
Duodenal ulcer haemorrhage * 1	1/185 (0.54%)	0/183 (0.00%)
Enteritis * 1	1/185 (0.54%)	0/183 (0.00%)
Gastrooesophageal reflux disease * 1	1/185 (0.54%)	0/183 (0.00%)
Ileal perforation * 1	1/185 (0.54%)	0/183 (0.00%)
Large intestinal haemorrhage * 1	1/185 (0.54%)	0/183 (0.00%)
Neutropenic colitis * 1	1/185 (0.54%)	0/183 (0.00%)
Oesophageal motility disorder * 1	1/185 (0.54%)	0/183 (0.00%)
Oesophageal varices haemorrhage * 1	1/185 (0.54%)	0/183 (0.00%)
Pneumoperitoneum * 1	1/185 (0.54%)	0/183 (0.00%)
Small intestinal perforation * 1	1/185 (0.54%)	0/183 (0.00%)
Abdominal distension * 1	0/185 (0.00%)	1/183 (0.55%)
Abdominal pain upper * 1	0/185 (0.00%)	2/183 (1.09%)
Anal fistula * 1	0/185 (0.00%)	1/183 (0.55%)
Colitis * 1	0/185 (0.00%)	2/183 (1.09%)
Duodenitis * 1	0/185 (0.00%)	1/183 (0.55%)
Enterovesical fistula * 1	0/185 (0.00%)	1/183 (0.55%)
Gastric haemorrhage * 1	0/185 (0.00%)	1/183 (0.55%)
Gastrointestinal haemorrhage * 1	0/185 (0.00%)	1/183 (0.55%)
Haemorrhoidal haemorrhage * 1	0/185 (0.00%)	1/183 (0.55%)
Ileus * 1	0/185 (0.00%)	2/183 (1.09%)
Large intestinal obstruction * 1	0/185 (0.00%)	2/183 (1.09%)
Pancreatitis * 1	0/185 (0.00%)	1/183 (0.55%)
Reflux gastritis * 1	0/185 (0.00%)	1/183 (0.55%)
General disorders
Chest pain * 1	3/185 (1.62%)	4/183 (2.19%)
Pyrexia * 1	3/185 (1.62%)	4/183 (2.19%)
Device dislocation * 1	2/185 (1.08%)	1/183 (0.55%)
Asthenia * 1	1/185 (0.54%)	1/183 (0.55%)
Chills * 1	1/185 (0.54%)	0/183 (0.00%)
Mucosal inflammation * 1	1/185 (0.54%)	0/183 (0.00%)
Multi-organ failure * 1	1/185 (0.54%)	0/183 (0.00%)
Peripheral swelling * 1	1/185 (0.54%)	0/183 (0.00%)
Systemic inflammatory response syndrome * 1	1/185 (0.54%)	0/183 (0.00%)
Device malfunction * 1	0/185 (0.00%)	1/183 (0.55%)
Fatigue * 1	0/185 (0.00%)	3/183 (1.64%)
Gait disturbance * 1	0/185 (0.00%)	1/183 (0.55%)
Localised oedema * 1	0/185 (0.00%)	1/183 (0.55%)
Hepatobiliary disorders
Cholecystitis * 1	1/185 (0.54%)	1/183 (0.55%)
Cholecystitis acute * 1	1/185 (0.54%)	0/183 (0.00%)
Hyperbilirubinaemia * 1	1/185 (0.54%)	0/183 (0.00%)
Ischaemic hepatitis * 1	1/185 (0.54%)	0/183 (0.00%)
Biloma * 1	0/185 (0.00%)	1/183 (0.55%)
Cholangitis * 1	0/185 (0.00%)	1/183 (0.55%)
Cholecystitis chronic * 1	0/185 (0.00%)	1/183 (0.55%)
Cholelithiasis * 1	0/185 (0.00%)	1/183 (0.55%)
Hepatic steatosis * 1	0/185 (0.00%)	1/183 (0.55%)
Hepatitis * 1	0/185 (0.00%)	1/183 (0.55%)
Jaundice * 1	0/185 (0.00%)	1/183 (0.55%)
Infections and infestations
Sepsis * 1	4/185 (2.16%)	3/183 (1.64%)
Pneumonia * 1	3/185 (1.62%)	7/183 (3.83%)
Cellulitis * 1	2/185 (1.08%)	0/183 (0.00%)
Septic shock * 1	2/185 (1.08%)	1/183 (0.55%)
Upper respiratory tract infection * 1	2/185 (1.08%)	0/183 (0.00%)
Abdominal abscess * 1	1/185 (0.54%)	0/183 (0.00%)
Abdominal infection * 1	1/185 (0.54%)	0/183 (0.00%)
Appendicitis * 1	1/185 (0.54%)	0/183 (0.00%)
Bronchitis * 1	1/185 (0.54%)	0/183 (0.00%)
Candida sepsis * 1	1/185 (0.54%)	0/183 (0.00%)
Diverticulitis * 1	1/185 (0.54%)	0/183 (0.00%)
Endocarditis * 1	1/185 (0.54%)	0/183 (0.00%)
Infection * 1	1/185 (0.54%)	1/183 (0.55%)
Lung infection * 1	1/185 (0.54%)	0/183 (0.00%)
Respiratory tract infection * 1	1/185 (0.54%)	0/183 (0.00%)
Subcutaneous abscess * 1	1/185 (0.54%)	0/183 (0.00%)
Urinary tract infection * 1	1/185 (0.54%)	0/183 (0.00%)
Urosepsis * 1	1/185 (0.54%)	1/183 (0.55%)
Device related infection * 1	0/185 (0.00%)	2/183 (1.09%)
Gastroenteritis * 1	0/185 (0.00%)	1/183 (0.55%)
Gastroenteritis viral * 1	0/185 (0.00%)	1/183 (0.55%)
Kidney infection * 1	0/185 (0.00%)	1/183 (0.55%)
Lower respiratory tract infection * 1	0/185 (0.00%)	1/183 (0.55%)
Neutropenic sepsis * 1	0/185 (0.00%)	1/183 (0.55%)
Perineal abscess * 1	0/185 (0.00%)	1/183 (0.55%)
Pneumonia bacterial * 1	0/185 (0.00%)	1/183 (0.55%)
Pneumonia parainfluenzae viral * 1	0/185 (0.00%)	1/183 (0.55%)
Injury, poisoning and procedural complications
Infusion related reaction * 1	1/185 (0.54%)	0/183 (0.00%)
Laceration * 1	1/185 (0.54%)	0/183 (0.00%)
Stoma site haemorrhage * 1	1/185 (0.54%)	0/183 (0.00%)
Toxicity to various agents * 1	1/185 (0.54%)	0/183 (0.00%)
Wound dehiscence * 1	1/185 (0.54%)	0/183 (0.00%)
Abdominal wound dehiscence * 1	0/185 (0.00%)	1/183 (0.55%)
Intestinal anastomosis complication * 1	0/185 (0.00%)	1/183 (0.55%)
Spinal compression fracture * 1	0/185 (0.00%)	1/183 (0.55%)
Investigations
Platelet count decreased * 1	2/185 (1.08%)	0/183 (0.00%)
Weight decreased * 1	1/185 (0.54%)	0/183 (0.00%)
Blood bilirubin increased * 1	0/185 (0.00%)	2/183 (1.09%)
Metabolism and nutrition disorders
Dehydration * 1	8/185 (4.32%)	5/183 (2.73%)
Decreased appetite * 1	1/185 (0.54%)	0/183 (0.00%)
Diabetes mellitus * 1	1/185 (0.54%)	0/183 (0.00%)
Failure to thrive * 1	1/185 (0.54%)	1/183 (0.55%)
Hyperglycaemia * 1	1/185 (0.54%)	0/183 (0.00%)
Hypokalaemia * 1	1/185 (0.54%)	1/183 (0.55%)
Hyponatraemia * 1	1/185 (0.54%)	2/183 (1.09%)
Malnutrition * 1	1/185 (0.54%)	0/183 (0.00%)
Hypoalbuminaemia * 1	0/185 (0.00%)	1/183 (0.55%)
Hypophosphataemia * 1	0/185 (0.00%)	1/183 (0.55%)
Metabolic alkalosis * 1	0/185 (0.00%)	1/183 (0.55%)
Musculoskeletal and connective tissue disorders
Muscular weakness * 1	1/185 (0.54%)	1/183 (0.55%)
Polyarthritis * 1	1/185 (0.54%)	0/183 (0.00%)
Rhabdomyolysis * 1	1/185 (0.54%)	0/183 (0.00%)
Neoplasms benign, malignant and unspecified (incl cysts and polyps)
Meningioma * 1	0/185 (0.00%)	1/183 (0.55%)
Nervous system disorders
Aphonia * 1	1/185 (0.54%)	0/183 (0.00%)
Ataxia * 1	1/185 (0.54%)	0/183 (0.00%)
Cerebrovascular accident * 1	1/185 (0.54%)	1/183 (0.55%)
Seizure * 1	1/185 (0.54%)	0/183 (0.00%)
Syncope * 1	1/185 (0.54%)	2/183 (1.09%)
Haemorrhage intracranial * 1	0/185 (0.00%)	1/183 (0.55%)
Ischaemic stroke * 1	0/185 (0.00%)	1/183 (0.55%)
Thrombotic stroke * 1	0/185 (0.00%)	1/183 (0.55%)
Psychiatric disorders
Confusional state * 1	1/185 (0.54%)	1/183 (0.55%)
Mental status changes * 1	0/185 (0.00%)	2/183 (1.09%)
Renal and urinary disorders
Acute kidney injury * 1	1/185 (0.54%)	2/183 (1.09%)
Proteinuria * 1	1/185 (0.54%)	1/183 (0.55%)
Renal failure * 1	0/185 (0.00%)	1/183 (0.55%)
Respiratory, thoracic and mediastinal disorders
Pulmonary embolism * 1	9/185 (4.86%)	9/183 (4.92%)
Acute respiratory failure * 1	2/185 (1.08%)	1/183 (0.55%)
Respiratory distress * 1	2/185 (1.08%)	0/183 (0.00%)
Chronic obstructive pulmonary disease * 1	1/185 (0.54%)	0/183 (0.00%)
Dyspnoea * 1	1/185 (0.54%)	1/183 (0.55%)
Epistaxis * 1	1/185 (0.54%)	1/183 (0.55%)
Hiccups * 1	1/185 (0.54%)	0/183 (0.00%)
Pulmonary thrombosis * 1	1/185 (0.54%)	0/183 (0.00%)
Asthma * 1	0/185 (0.00%)	1/183 (0.55%)
Cough * 1	0/185 (0.00%)	1/183 (0.55%)
Interstitial lung disease * 1	0/185 (0.00%)	1/183 (0.55%)
Pulmonary artery thrombosis * 1	0/185 (0.00%)	1/183 (0.55%)
Surgical and medical procedures
Colostomy * 1	0/185 (0.00%)	1/183 (0.55%)
Vascular disorders
Hypotension * 1	5/185 (2.70%)	1/183 (0.55%)
Hypertension * 1	3/185 (1.62%)	6/183 (3.28%)
Arterial thrombosis * 1	1/185 (0.54%)	0/183 (0.00%)
Deep vein thrombosis * 1	1/185 (0.54%)	4/183 (2.19%)
Embolism venous * 1	1/185 (0.54%)	1/183 (0.55%)
Venous thrombosis * 1	1/185 (0.54%)	1/183 (0.55%)
Jugular vein thrombosis * 1	0/185 (0.00%)	1/183 (0.55%)
Thrombosis * 1	0/185 (0.00%)	1/183 (0.55%)
* Indicates events were collected by non-systematic assessment; 1 Term from vocabulary, MedDRA (18.1)
Other (Not Including Serious) Adverse Events
Frequency Threshold for Reporting Other Adverse Events	5%
	Bevacizumab + mFOLFOX6	Bevacizumab + FOLFIRI
	Affected / at Risk (%)	Affected / at Risk (%)
Total	185/185 (100.00%)	181/183 (98.91%)
Blood and lymphatic system disorders
Neutropenia * 1	65/185 (35.14%)	86/183 (46.99%)
Anaemia * 1	25/185 (13.51%)	34/183 (18.58%)
Thrombocytopenia * 1	44/185 (23.78%)	13/183 (7.10%)
Eye disorders
Vision blurred * 1	13/185 (7.03%)	12/183 (6.56%)
Lacrimation increased * 1	4/185 (2.16%)	10/183 (5.46%)
Gastrointestinal disorders
Diarrhoea * 1	99/185 (53.51%)	123/183 (67.21%)
Nausea * 1	106/185 (57.30%)	109/183 (59.56%)
Constipation * 1	63/185 (34.05%)	65/183 (35.52%)
Vomiting * 1	58/185 (31.35%)	57/183 (31.15%)
Abdominal pain * 1	39/185 (21.08%)	54/183 (29.51%)
Stomatitis * 1	35/185 (18.92%)	38/183 (20.77%)
Gastrooesophageal reflux disease * 1	17/185 (9.19%)	18/183 (9.84%)
Dyspepsia * 1	19/185 (10.27%)	14/183 (7.65%)
Rectal haemorrhage * 1	11/185 (5.95%)	18/183 (9.84%)
Abdominal pain upper * 1	8/185 (4.32%)	15/183 (8.20%)
Oral pain * 1	8/185 (4.32%)	14/183 (7.65%)
Haemorrhoids * 1	6/185 (3.24%)	16/183 (8.74%)
Proctalgia * 1	7/185 (3.78%)	13/183 (7.10%)
Dry mouth * 1	7/185 (3.78%)	11/183 (6.01%)
Abdominal distension * 1	5/185 (2.70%)	12/183 (6.56%)
Dysphagia * 1	10/185 (5.41%)	2/183 (1.09%)
General disorders
Fatigue * 1	105/185 (56.76%)	104/183 (56.83%)
Mucosal inflammation * 1	37/185 (20.00%)	51/183 (27.87%)
Pyrexia * 1	18/185 (9.73%)	29/183 (15.85%)
Asthenia * 1	21/185 (11.35%)	18/183 (9.84%)
Oedema peripheral * 1	18/185 (9.73%)	18/183 (9.84%)
Temperature intolerance * 1	34/185 (18.38%)	1/183 (0.55%)
Chest pain * 1	15/185 (8.11%)	10/183 (5.46%)
Chills * 1	10/185 (5.41%)	10/183 (5.46%)
Influenza like illness * 1	10/185 (5.41%)	6/183 (3.28%)
Immune system disorders
Hypersensitivity * 1	10/185 (5.41%)	0/183 (0.00%)
Infections and infestations
Urinary tract infection * 1	18/185 (9.73%)	20/183 (10.93%)
Upper respiratory tract infection * 1	15/185 (8.11%)	18/183 (9.84%)
Sinusitis * 1	9/185 (4.86%)	13/183 (7.10%)
Injury, poisoning and procedural complications
Infusion related reaction * 1	15/185 (8.11%)	8/183 (4.37%)
Investigations
Weight decreased * 1	33/185 (17.84%)	29/183 (15.85%)
Neutrophil count decreased * 1	33/185 (17.84%)	23/183 (12.57%)
Platelet count decreased * 1	24/185 (12.97%)	3/183 (1.64%)
White blood cell count decreased * 1	15/185 (8.11%)	9/183 (4.92%)
Metabolism and nutrition disorders
Decreased appetite * 1	56/185 (30.27%)	55/183 (30.05%)
Hypokalaemia * 1	36/185 (19.46%)	33/183 (18.03%)
Dehydration * 1	28/185 (15.14%)	27/183 (14.75%)
Hyperglycaemia * 1	9/185 (4.86%)	23/183 (12.57%)
Hyponatraemia * 1	8/185 (4.32%)	15/183 (8.20%)
Musculoskeletal and connective tissue disorders
Arthralgia * 1	22/185 (11.89%)	25/183 (13.66%)
Back pain * 1	21/185 (11.35%)	23/183 (12.57%)
Pain in extremity * 1	22/185 (11.89%)	19/183 (10.38%)
Musculoskeletal pain * 1	17/185 (9.19%)	11/183 (6.01%)
Muscle spasms * 1	7/185 (3.78%)	11/183 (6.01%)
Nervous system disorders
Neuropathy peripheral * 1	85/185 (45.95%)	23/183 (12.57%)
Headache * 1	31/185 (16.76%)	38/183 (20.77%)
Dysgeusia * 1	33/185 (17.84%)	31/183 (16.94%)
Paraesthesia * 1	35/185 (18.92%)	15/183 (8.20%)
Peripheral sensory neuropathy * 1	35/185 (18.92%)	12/183 (6.56%)
Dizziness * 1	15/185 (8.11%)	28/183 (15.30%)
Psychiatric disorders
Insomnia * 1	29/185 (15.68%)	35/183 (19.13%)
Depression * 1	24/185 (12.97%)	18/183 (9.84%)
Anxiety * 1	19/185 (10.27%)	20/183 (10.93%)
Renal and urinary disorders
Proteinuria * 1	38/185 (20.54%)	39/183 (21.31%)
Respiratory, thoracic and mediastinal disorders
Epistaxis * 1	41/185 (22.16%)	59/183 (32.24%)
Cough * 1	31/185 (16.76%)	29/183 (15.85%)
Dyspnoea * 1	28/185 (15.14%)	26/183 (14.21%)
Rhinorrhoea * 1	12/185 (6.49%)	21/183 (11.48%)
Dysphonia * 1	14/185 (7.57%)	18/183 (9.84%)
Hiccups * 1	8/185 (4.32%)	14/183 (7.65%)
Oropharyngeal pain * 1	11/185 (5.95%)	8/183 (4.37%)
Skin and subcutaneous tissue disorders
Alopecia * 1	26/185 (14.05%)	61/183 (33.33%)
Dry skin * 1	27/185 (14.59%)	26/183 (14.21%)
Rash * 1	19/185 (10.27%)	26/183 (14.21%)
Palmar-plantar erythrodysaesthesia syndrome * 1	24/185 (12.97%)	18/183 (9.84%)
Skin hyperpigmentation * 1	18/185 (9.73%)	13/183 (7.10%)
Pruritus * 1	11/185 (5.95%)	9/183 (4.92%)
Hyperhidrosis * 1	2/185 (1.08%)	10/183 (5.46%)
Vascular disorders
Hypertension * 1	61/185 (32.97%)	54/183 (29.51%)
* Indicates events were collected by non-systematic assessment; 1 Term from vocabulary, MedDRA (18.1)

Limitations and Caveats

[Not Specified]

More Information

Certain Agreements

Principal Investigators are NOT employed by the organization sponsoring the study.

There IS an agreement between Principal Investigators and the Sponsor (or its agents) that restricts the PI's rights to discuss or publish trial results after the trial is completed.

The Study being conducted under this Agreement is part of the Overall Study. Investigator is free to publish in reputable journals or to present at professional conferences the results of the Study, but only after the first publication or presentation that involves the Overall Study. The Sponsor may request that Confidential Information be deleted and/or the publication be postponed in order to protect the Sponsor's intellectual property rights.

Results Point of Contact

• Name/Title: Medical Communications
• Organization: Hoffmann-La Roche
• Phone: 800-821-8590
• EMail: genentech@druginfo.com

• Responsible Party: Genentech, Inc.
• ClinicalTrials.gov Identifier: NCT01374425
• Other Study ID Numbers: ML25710; 2011-004755-39 ( EudraCT Number )
• First Submitted: June 14, 2011
• First Posted: June 16, 2011
• Results First Submitted: June 30, 2016
• Results First Posted: August 11, 2016
• Last Update Posted: August 11, 2016