Díaz‐Rubio, Eduardo

PurposeProgression-free survival (PFS) has previously been established as a surrogate for overall survival (OS) for first-line metastatic colorectal cancer (mCRC). Because mCRC treatment has advanced in the last decade with extended OS, this surrogacy requires re-examination.MethodsIndividual patient data from 16,762 patients were available from 22 first-line mCRC studies conducted from 1997 to 2006; 12 of those studies tested antiangiogenic and/or anti–epidermal growth factor receptor agents. The relationship between PFS (first event of progression or death) and OS was evaluated by using R2statistics (the closer the value is to 1, the stronger the correlation) from weighted least squares regression of trial-specific hazard ratios estimated by using Cox and Copula models.ResultsForty-four percent of patients received a regimen that included biologic agents. Median first-line PFS was 8.3 months, and median OS was 18.2 months. The correlation between PFS and OS was modest (R2, 0.45 to 0.69). Analyses limited to trials that tested treatments with biologic agents, nonstrategy trials, or superiority trials did not improve surrogacy.ConclusionIn modern mCRC trials, in which survival after the first progression exceeds time to first progression, a positive but modest correlation was observed between OS and PFS at both the patient and trial levels. This finding demonstrates the substantial variability in OS introduced by the number of lines of therapy and types of effective subsequent treatments and the associated challenge to the use of OS as an end point to assess the benefit attributable to a single line of therapy. PFS remains an appropriate primary end point for first-line mCRC trials to detect the direct treatment effect of new agents.

Purpose This study addressed whether age is prognostic for overall survival (OS) or progression-free survival (PFS) in patients with metastatic colorectal cancer (mCRC). Patients and Methods A total of 20,023 patients from 24 first-line clinical trials in the ARCAD (Aide et Recherche en Cancérologie Digestive) database were analyzed. Primary age effects and interactions with age, sex, performance status (PS), and metastatic site were modeled using Cox proportional hazards stratified by treatment arm within study. Results Of total patients, 3,051 (15%) were age ≤ 50 years. Age was prognostic for both OS (P < .001) and PFS (P < .001), with U-shaped risk (ie, highest risk was evident in youngest and oldest patients). Relative to patients of middle age, the youngest patients experienced 19% (95% CI, 7% to 33%) increased risk of death and 22% (95% CI, 10% to 35%) increased risk of progression. The oldest patients experienced 42% (95% CI, 31% to 54%) increased risk of death and 15% (95% CI, 7% to 24%) increased risk of progression or death. This relationship was more pronounced in the first year of follow-up. Age remained marginally significant for OS (P = .08) when adjusted for PS, sex, and presence of liver, lung, or peritoneal metastases, and age was significant in an adjusted model for PFS (P = .005). The age effect did not differ by site of metastatic disease, year of enrollment, type of therapy received, or biomarker mutational status. Conclusion Younger and older age are associated with poorer OS and PFS among treated patients with mCRC. Younger and older patients may represent higher-risk populations, and additional studies are warranted.

Purpose Six randomized phase II and III trials have investigated the role of oxaliplatin (OX) in combination with capecitabine (CAP) or infusional fluorouracil (FU) in metastatic colorectal cancer. This meta-analysis compared the efficacy of CAP/OX compared with infusional FU/OX. Patients and Methods This analysis compared all published CAP/OX versus infusional FU/OX regimens. A total of 3,494 patients (FU, n = 1,737; CAP, n = 1,757) were analyzed for response rate (RR), progression-free (PFS), overall survival (OS), and toxicity. Results The fixed-effect pooled estimate for RR showed higher RR for FU-based regimens (Odds ratio [OR] = 0.85; 95% CI, 0.74 to 0.97; P = .02) whereas the analysis of chemotherapy-only trials, excluding the bevacizumab containing NO16966 and TREE 2 trials, led to an OR of 0.74 (95% CI, 0.60 to 0.92; P = .007). However, for PFS (hazard ratio [HR] = 1.04; 95% CI, 0.96 to 1.12; P = .17) and OS (HR = 1.04; 95% CI, 0.95 to 1.12; P = .41) all models suggested similar outcome within the range of noninferiority. Grade 3/4 toxicities (thrombocytopenia—HR = 2.07, 95% CI, 1.42 to 3.03; P < .0002; diarrhea—HR = 1.34; 95% CI, 1.08 to 1.66; P < .0009; and grade 2/3 hand-foot-syndrome [HFS]—HR = 3.54; 95% CI, 2.07 to 6.05; P < .00001) were less prominent with FU-based regimens whereas neutropenia (HR = 0.15; 95% CI, 0.11 to 0.19; P < .00001) was lower in the CAP regimens. Conclusion The combination of CAP and OX resulted in lower RR, but this did not affect PFS and OS, which were similar in both treatment arms. The toxicity analysis showed the characteristic toxicity of each of the different FU schedules, with thrombocytopenia and HFS consistently more prominent in the CAP regimens.