This phase III trial studies how well combination chemotherapy, bevacizumab, and/or atezolizumab work in treating patients with deficient deoxyribonucleic acid (DNA) mismatch repair colorectal cancer that has spread to other places in the body (metastatic). Chemotherapy drugs, such as fluorouracil, oxaliplatin, and leucovorin calcium, work in different ways to stop the growth of tumor cells, either by killing the cells, by stopping them from dividing, or by stopping them from spreading. Immunotherapy with monoclonal antibodies, such as bevacizumab and atezolizumab, may help the body's immune system attack the cancer, and may interfere with the ability of tumor cells to grow and spread. Giving combination chemotherapy, bevacizumab, and atezolizumab may work better in treating patients with colorectal cancer.
Name: Atezolizumab
Name: Bevacizumab
Name: Fluorouracil
Name: Leucovorin
Name: Oxaliplatin
Name: Quality-of-Life Assessment
Name: Questionnaire Administration
Description: The analysis set is intent-to-treat (ITT). The experimental arms will be compared to the control arm by the log-rank test stratified by BRAF status (V600E mutation or not), metastatic disease: (liver-only, extra-hepatic), and prior adjuvant therapy for colon cancer (yes, no). Hazard ratios and associated confidence intervals from a stratified Cox regression model will also be reported along with estimates of the distributions of time to PFS event by the method of Kaplan and Meier. Sensitivity analyses accounting for 2 or more consecutively missed scheduled tumor imaging scans before progression/death will also be conducted.
Measure: Progression free survival (PFS) Time: From the time from randomization until first confirmed progression or death from any cause, assessed up to 5 yearsDescription: Will be analyzed using the stratified log rank test and the ITT population. Kaplan-Meier plots will illustrate the distribution of these endpoints by treatment. Stratified Cox regression models will be used to estimate hazard ratios and associated confidence intervals.
Measure: Overall survival (OS) Time: The time from randomization to death from any cause, assessed up to 5 yearsDescription: Assessed by Response Evaluation Criteria in Solid Tumors (RECIST) 1.1. Will be analyzed by a logistic regression models that control for the stratification factors (BRAF status, liver involvement, and adjuvant chemotherapy [chemo]) using the ITT population. Observed proportions along with confidence intervals will be presented by treatment.
Measure: Objective response rate (ORR) (complete response [CR] or partial response [PR]) Time: Up to 5 yearsDescription: Defined by Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. The safety profile will be described by tabulating the maximum observed grade of adverse event for each individual adverse event, for each system organ class, and overall using the Safety population.
Measure: Incidence of adverse events Time: Up to 30 days after last cycleDescription: Assessed by RECIST 1.1. Will be analyzed by a logistic regression models that control for the stratification factors (BRAF status, liver involvement, and adjuvant chemo) using the ITT population. Observed proportions along with confidence intervals will be presented by treatment.
Measure: Disease control rate (CR + PR + stable disease [SD]) Time: At 12 monthsDescription: Assessed by RECIST 1.1. Will be analyzed using the stratified log rank test and the ITT population. Kaplan-Meier plots will illustrate the distribution of these endpoints by treatment. Stratified Cox regression models will be used to estimate hazard ratios and associated confidence intervals.
Measure: Duration of overall response (CR or PR) Time: From the time of first response to first confirmed progression by the study investigator or death from any cause, assessed up to 5 yearsDescription: Assessed per RECIST 1.1. Will be analyzed using the stratified log rank test and the ITT population. Kaplan-Meier plots will illustrate the distribution of these endpoints by treatment. Stratified Cox regression models will be used to estimate hazard ratios and associated confidence intervals.
Measure: Duration of SD Time: From the time of first on-study assessment of SD to first progression by the study investigator or death from any cause, assessed up to 5 yearsDescription: Will use the ITT population. Measured by the Patient Reported Outcomes Measurement Information System (PROMIS) Fatigue questionnaire. Will be compared between the control arm (Chemo-bevacizumab [Bev]) and the experimental arm by means of ordinal logistic regression with adjustment for the corresponding baseline measurement and stratification variables. The comparison will be performed at the significance level of 0.05 (two-sided) and the clinical meaningfulness of the comparison will be considered.
Measure: Severity of fatigue Time: At 16 weeksDescription: Will use the ITT population. Measured by the European Organization for Research and Treatment of Cancer (EORTC) Quality-of-Life questionnaire (QLQ)-C30 physical functioning scale. Will be compared between the control arm (Chemo-Bev) and the experimental arm by means of ordinal logistic regression with adjustment for the corresponding baseline measurement and stratification variables. The comparison will be performed at the significance level of 0.05 (two-sided) and the clinical meaningfulness of the comparison will be considered.
Measure: Physical functioning Time: At 16 weeksDescription: Will use the PRO population. The corresponding item (PRO-CTCAE severity, PRO-CTCAE frequency, or QOL scale) will be compared between the control arm (Chemo-Bev) and the experimental arm using a mixed regression model for repeated measures with the response being ordinal for the PRO-CTCAE items and linear for QOL scales. The model will also include the corresponding baseline measurement, time, and stratification variables. Presence of treatment-by-time interaction will be investigated for each model. Each comparison will be performed at the significance level of 0.05 (two-sided) and the clinical meaningfulness of the comparison will be considered.
Measure: Severity and frequency of quality of life (QOL) and patient-reported outcomes (PRO) Time: Up to 5 yearsDescription: Will use the ITT population. Will be measured using the EuroQoL 5 Dimensions 5 Levels (EQ-5D-5L) questionnaire.
Measure: Health utility scores Time: Up to 5 yearsDescription: Will use the ITT population.
Measure: Proportion of patients reporting each response option at each assessment timepoint by treatment arm for item GP5 from the Functional Assessment of Cancer Therapy - General (FACT-G) Time: Up to 5 yearsDescription: Will use the ITT population.
Measure: Intratumoral lymphocyte PD-L1+ expression (>= 1 % is positive) by immunohistochemistry (IHC) as a predictive biomarker of efficacy Time: Up to 5 yearsDescription: Will use the ITT population.
Measure: Efficacy dependent on the number of somatic mutations Time: Up to 5 yearsDescription: Will use the ITT population.
Measure: Efficacy in tumors with MLH1 silencing Time: Up to 5 yearsDescription: Will use the ITT population.
Measure: Change in quantification of cell free deoxyribonucleic acid (cfDNA) mutations Time: Baseline up to 5 yearsDescription: Will use the ITT population.
Measure: Development of progression or relapse to treatment in cfDNA Time: Up to 5 yearsDescription: Will use the ITT population.
Measure: Changes in T-cell repertoire diversity as a predictive biomarker of efficacy Time: Baseline up to 5 yearsDescription: Compared to patients with low levels of diversity. Will use the ITT population.
Measure: PFS of patients with high levels of diversity Time: Up to 5 yearsDescription: Will use the ITT population.
Measure: Change in T-cell diversity Time: Baseline to first restaging between immunotherapy arms and the standard of care armDescription: Will use the ITT population.
Measure: Mechanism of immune resistance to PD-1 blockade in mismatch repair deficient (dMMR)/microsatellite instability-high metastatic colorectal cancer (mCRC) by comparative analysis of tumor samples collected Time: Baseline up to 5 yearsAllocation: Randomized
Parallel Assignment
There is one SNP
The experimental arms will be compared to the control arm by the log-rank test stratified by BRAF status (V600E mutation or not), metastatic disease: (liver-only, extra-hepatic), and prior adjuvant therapy for colon cancer (yes, no). --- V600E ---