O-0015 Survival According to Prognostic Risk Factors in Patients Receiving First-Line Chemotherapy Plus Cetuximab in the Crystal and Opus Studies

ABSTRACT Introduction Analysis of randomized trials has shown that patients with metastatic colorectal cancer (mCRC) can be divided into prognostic risk groups according to baseline clinical parameters including Eastern Cooperative Oncology Group (ECOG) performance status, white blood cell count (WBC), alkaline phosphatase (ALP) and number of metastatic sites. The effect of adding cetuximab to first-line chemotherapy on overall survival (OS) in these groups of KRAS wild-type mCRC patients was investigated in the CRYSTAL and OPUS studies. Methods Patient risk groups were: low-risk (LRG= ECOG 0/1, 1 metastatic site), intermediate-risk (IRG= ECOG 1 metastatic site, ALP 1, low WBC, 1 metastatic site) and high-risk (HRG= ECOG 1 metastatic site, ALP u003e300 U/L or ECOG u003e1, high WBC, or ECOG u003e1, low WBC and u003e2 metastatic sites). Exploratory analyses comprised estimates of effects using the Cox proportional hazards regression model for OS on individual patient data, and comparison of treatment arms using the log-rank test. Results In the pooled analyses, in both treatment arms OS was longest in LRG patients (median 27.0 months, 95% CI 24.8-29.5 with chemotherapy + cetuximab [n = 173] and 25.7 months, 95% CI 21.9–28.7 with chemotherapy [n = 164]), and shortest in HRG patients (median 17.7 months, 95% CI 13.1–19.3 with chemotherapy + cetuximab [n = 46]; 12.6 months, 95% CI 9.2–14.4 with chemotherapy [n = 59]). In IRG patients, median OS was 16.4 (95% CI 14.9–20.0) and 22.2 months (95% CI 19.5–25.7) with chemotherapy (n = 213) and chemotherapy + cetuximab (n = 166), respectively. Adding cetuximab to chemotherapy led to marked improvements in OS in the HRG (hazard ratio [HR] 0.765, 95% CI 0.506–1.157, p = 0.203) and IRG (HR 0.781, 95% CI 0.622–0.981, p = 0.033), while improvements in LRG patients (HR 0.869, 95% CI 0.672–1.124, p = 0.287) were observed only after prolonged follow up. In CRYSTAL, median OS in LRG patients was 28.2 months (95% CI 24.9-31.5) with chemotherapy + cetuximab (n = 132) and 26.1 months (95% CI 21.2–29.9) with chemotherapy (n = 127). Median OS in HRG patients was 16.7 months (95% CI 8.3–19.0) with chemotherapy + cetuximab (n = 31) and 12.8 months (95% CI 9.2–14.6) with chemotherapy (n = 43). In IRG patients, median OS was 16.6 (95% CI 14.8–20.4) and 23.5 months (95% CI 20.0–29.6) with chemotherapy (n = 172) and chemotherapy + cetuximab (n = 141), respectively. In OPUS, median OS in LRG patients was 26.0 months (95% CI 22.9-not estimable [NE]) with chemotherapy + cetuximab (n = 41) and 23.9 months (95% CI 19.5–NE) with chemotherapy [n = 37]). For HRG patients, median OS was 19.9 months (95% CI 13.1–30.4) with chemotherapy + cetuximab (n = 15) and 11.8 months (95% CI 4.8–18.5) with chemotherapy (n = 16). In IRG patients, median OS was 16.2 (95% CI 12.0–21.6) and 18.4 months (95% CI 11.0–19.8) with chemotherapy (n = 41) and chemotherapy + cetuximab (n = 25), respectively. Conclusion These analyses confirm the concept of prognostic risk groups for OS according to baseline clinical parameters in patients with KRAS wild-type mCRC. Benefit from the addition of cetuximab to first-line chemotherapy in terms of OS appears to be more pronounced in the IRG and HRG.