Patients with mismatch repair–deficient and/or microsatellite instability metastatic colorectal cancer experienced a progression-free survival benefit with longer disease control following treatment with avelumab compared with standard second-line chemotherapy.
Patients with mismatch repair–deficient and/or microsatellite instability (dMMR/MSI) metastatic colorectal cancer (CRC) experienced a progression-free survival (PFS) benefit with longer disease control following treatment with avelumab (Bavencio) compared with standard second-line chemotherapy, according to findings from the phase 2 SAMCO-PRODIGE 54 trial (NCT03186326) published in JAMA Oncology.
At a median follow-up of 33.3 months (95% CI, 28.3-34.8), patients who received the anti–PD-L1 monoclonal antibody (n = 61) experienced a median PFS of 4.1 months (range, 2.31-5.68) compared with 6.2 months (range, 4.11-7.29) among patients who received standard chemotherapy (n = 61). However, study authors noted that because the Kaplan-Meier curves crossed at 7.3 months corresponding to a PFS rate of 36%, the log-rank test and the hazard ratio of PFS analyses were not sufficient (log-rank P = .30).
Using the Qiu and Sheng statistical test that appeared better suited for this analysis, investigators determined that avelumab was superior to chemotherapy in terms of PFS (P = .03); the estimated 12- and 18-month PFS rates were 31.2% (95% CI, 20.1%-42.9%) and 27.4% (95% CI, 16.8%-39.0%), respectively, in the avelumab arm compared with 19.4% (95% CI, 10.6%-30.2%) and 9.1% (95% CI, 3.2%- 18.8%), respectively, in the control arm. Additionally, the estimated restricted mean survival time for PFS was 12.3 months (95% CI, 8.7-15.8) vs 8.1 months (95% CI, 6.2-10.0) in the avelumab and chemotherapy arms, respectively, after 36 months of follow-up (P = .04).
The overall response rates (ORR) were comparable with 29.5% of patients in the avelumab arm vs 26.2% in the chemotherapy arm experiencing a response. The disease control rate was 70.5% in the avelumab arm vs 77.0% in the chemotherapy arm, and among these patients the rate of ongoing disease control at 18 months was 75.7% vs 19.1%, respectively. The median duration of disease control was 16.7 months (interquartile range [IQR], 5.7-33.4) vs 7.3 months (IQR, 4.9-11.9), respectively (P < .001).
SAMCO-PRODIGE 54 was an open-label trial that was conducted at 40 centers in France. To be included in the study, patients needed to be at least 18 years of age with unresectable dMMR/MSI stage IV CRC, a World Health Organization (WHO) performance status score of 1 or less, and adequate organ function. All patients experienced disease progression after receiving a first-line standard chemotherapy regimen with or without a targeted agent based on RAS status.
Patients who were enrolled on the study were randomly assigned in a 1:1 manner to receiveavelumab at a dose of 10 mg/kg every 2 weeks intravenously or investigator’s choice of second-line chemotherapy. In the chemotherapy arm, patients were also allowed to receive a targeted agent according to first-line treatment regimen and RAS/BRAF status; if progression occurred, patients could receive an immune checkpoint inhibitor at the investigator’s discretion. Treatment proceeded until disease progression, unacceptable toxicity, or withdrawal. Stratification occurred by center, WHO performance status, BRAF status, and age.
The primary end point was PFS by RECIST v1.1 criteria, and patients who were alive without progression were censored on the date of last news. Secondary end points included overall survival (OS), ORR, time to best response, duration of disease control, and safety.
The baseline patient characteristics were well balanced between the avelumab and chemotherapy arms; the median age was 66 years (IQR, 54-75) and 67 years (IQR, 60-75), respectively. Most patients in both arms had right-sided primary tumors (87% vs 77%) and had previously received FOLFOX/CAPOX with or without targeted therapy (65.6% vs 70.5%). BRAF V600E (41% vs 44%) and RAS (23% vs 21%) mutations were present in patients in both arms. Additionally, patients had more than 5 metastases at rates of 41% in the avelumab arm vs 44% in the chemotherapy arm.
Additional findings showed that patients in both the avelumab and chemotherapy arms experienced complete response (6.6% vs 4.9%), partial response (23.0% vs 21.3%), stable disease (41.0% vs 50.8%), and progressive disease (27.9% vs 16.4%). The time to best response was 3.5 months (IQR, 2.0-8.0) vs 2.0 months (IQR, 1.8-2.5), respectively.
The median duration of treatment in the avelumab and chemotherapy arms was 7.4 months (range, 0.03-46.5) vs 5.1 months (range, 0.03-19.7), respectively, and the median OS was 25.8 months (95% CI, 14.1-not reported [NR]) vs 23.4 months (95% CI, 13.0-NR), respectively (HR, 0.94; 95% CI, 0.57- 1.53; P = .79). At the May 23, 2022, data cutoff, 32 patients in each arm had died.
Most patients in the chemotherapy arm (50.8%) were subsequently treated with an immune checkpoint inhibitor. Overall, 83.8% of patients in the chemotherapy arm who received a subsequent line of therapy were treated with an immune checkpoint inhibitor. In the avelumab arm, 29.5% of patients were still being treated with the agent and 53.5% of patients who experienced disease progression received a subsequent anticancer therapy.
In terms of safety, in the 63-patient avelumab arm and the 64-patient chemotherapy arm treatment-related adverse effects (TRAEs) of any grade occurred at rates of 88.9% vs 98.4%, respectively. TRAEs of grade 3 or 4 severity occurred at rates of 31.7% vs 53.1%, respectively, and included abnormal liver test results (7.9% vs 1.6%), diarrhea (4.8% vs 7.8%), neurotoxicity (1.6% vs 3.1%), and hypertension (1.6% vs 10.9%), among others. Grade 3 or 4 neutropenia was reported in 18.8% of patients in the chemotherapy arm.
Immune-mediated AEs occurred in 17.5% of patients in the avelumab arm, including grade 1 to 2 hypothyroidism (n = 6), grade 1 to 2 hyperthyroidism (n = 5), grade 3 colitis (n = 1), grade 2 infusion-related reactions (n = 2), and a grade 3 infusion-related reaction (n = 1). In the chemotherapy arm, 3.1% of patients experienced cetuximab (Erbitux) infusion–related reactions.
Six patients in the avelumab arm and 7 patients in the chemotherapy arm discontinued treatment due to an AE. No grade 5 AEs were reported in either arm.
Taïeb J, Bouche O, André T, et al. Avelumab vs standard second-line chemotherapy in patients with metastatic colorectal cancer and microsatellite instability: a randomized clinical trial. JAMA Oncol. Published online August 3, 2023. doi:10.1001/jamaoncol.2023.2761