Response to Hu, Zhou, and Sun

We thank Dr Hu and colleagues for their letter in response to our submission “Combining anti-vascular endothelial growth factor and anti-epidermal growth factor receptor antibodies: randomized phase II study of irinotecan and cetuximab with/without ramucirumab in second-line colorectal cancer (ECOG-ACRIN E7208).” We appreciate the opportunity to further expand on these aspects of our study, which have been addressed to some degree in the paper itself.

1. We agree with the correspondents that the study showed a Progression Free Survival (PFS) benefit but not overall survival, and possibly this was due to subsequent treatments. In the ECOG system, we did not have the option to track subsequent therapies so cannot adequately answer this question.

2. The study was limited to KRAS wild-type tumors, because this was the first molecularly directed GI study in the NCTN group. After the study was open to enrollment, further data arose on the broader “extended RAS” phenotype, but we did not feel it would be useful to amend this eligibility criterion because there was uncertainty as to the broad availability in community settings at the time. Therefore, as many as 20% of the patients enrolled could have the non-KRAS RAS mutations. Finally, the issue as to sidedness for colon cancer has only been fully elucidated as a factor for EGFR-directed agents in the last 5 years. We did seek to obtain this information on the patients enrolled in E7208, but these data were not available and had not been captured when the study was designed. It was not possible for ECOG to retrieve this information retrospectively, so we are not able to analyze this factor for the study.

3. We agree that patients entering the study were heterogenous with respect to prior treatment, because they would be in any follow-on Phase III trial that might be performed. We feel this was a strength of the study, as prior attempts at second-line treatment in the NCTN system, with more constrained eligibility criteria, were not successful. We felt this approach is more representative of the spectrum of patients we see. We used these factors as stratification and did analyze the subsets by these factors specifically to show precisely that patient history of treatment to date can influence response to subsequent therapy. In a larger phase III trial, these issues could be dissected with greater numbers

4. We also agree that a randomized phase II trial is not definitive and, in fact, call for phase III trials using the correct patient population, namely in patients with RAS wild-type, left-sided tumors in the final paragraph. Our goal in the trial was to demonstrate that combined anti-VEGF and anti-EGFR agents can be combined with safety and efficacy in a pilot study format. We join you in looking forward to such phase III trials.

Acknowledgments

This response was written by Dr Hochster on behalf of the authors of the E7208 study: Howard S. Hochster, Paul Catalano, Michelle Weitz, Edith P. Mitchell, Deirdre Cohen, Peter J. O’Dwyer, Bryan A. Faller, Jeremy S. Kortmansky, Mark H. O’Hara, Sheetal M. Kricher, Jill Lacy, Heinz-Josef Lenz, Udit Verma, and Al B. Benson.

Author contributions

Howard S. Hochster, MD (Conceptualization; Data curation; Formal analysis; Investigation; Project administration; Writing—original draft).

Funding

None.

Conflicts of interest

None.

Data availability

No data were collected or analyzed for the writing of this response.