In Reply: We agree with Dr Shitara and colleagues that the RASCAL II study did not show the same association between p.G13D KRAS mutations and worse prognosis1 as our analysis did. Unlike RASCAL II, which did not report the prognostic effect of the various KRAS mutations specifically in the metastatic setting, our data set was limited to this stage and specifically to chemotherapy refractory patients. In the PETACC-3 trial, no significant differences in prognostic value between the various types of KRAS mutations after univariate or multivariate survival models were observed.2 We agree that there were small numbers of patients with p.G13D-mutated tumors in the CO.17 data set, but a strict analysis of prognostic effect (as opposed to predictive effect) can only be performed on patients who received best supportive care alone, so a mixed comparison of some treated patients with all patients who received best supportive care would not properly address this issue but rather add additional confounders.
Tejpar S, De Roock W, Jonker D. KRAS Genotypes and Outcome in Patients With Chemotherapy-Refractory Metastatic Colorectal Cancer Treated With Cetuximab—Reply. JAMA. 2011;305(6):564-566. doi:10.1001/jama.2011.88