Early Tumor Shrinkage in Colorectal Cancer Suggests Long-Term Cetuximab Benefit

Colorectal cancer patients who have early tumor shrinkage after first-line treatment with chemotherapy plus cetuximab may be more likely to have long-term response to therapy, according to an analysis of two large clinical trials.

Hubert Piessevaux, MD, PhD, of the Cliniques Universitaires Saint-Luc in Brussels, and colleagues show that a more robust tumor response at 8 weeks after start of therapy was associated with improved progression-free and overall survival. An early tumor shrinkage of 20% or more could identify patients, receiving a combination of chemotherapy plus cetuximab, who had both longer progression-free and overall survival. For patients who received chemotherapy alone, there was only a weak association between early tumor shrinkage and progression-free and overall survival.

The results provide evidence that tumor shrinkage can be used as a prognostic biomarker. The results are published in the Journal of Clinical Oncology.

The patients analyzed for the study included the 1,289 patients compiled from two randomized first-line colorectal cancer trials of cetuximab whose tumors could be analyzed for KRAS mutation status.

“If confirmed in a prospective trial, [early tumor shrinkage] may be useful to guide on-treatment decisions including continuation or discontinuation of therapy,” conclude the authors.

While colorectal tumors with a mutation in the KRAS gene generally do not benefit from EGFR-targeted therapy such as cetuximab, there are no current biomarkers to select patients who are more likely to respond. In their study, Piessevaux analyze the data from the CRYSTAL (Cetuximab Combined With Irinotecan in First-Line Therapy for Metastatic Colorectal Cancer) phase III trial and the OPUS (Oxaliplatin and Cetuximab in First-Line Treatment of Metastatic Colorectal Cancer) phase II trial that accrued patients between 2004 and 2006. A previous single-arm colorectal cancer trial with cetuximab suggested that early tumor shrinkage could predict the long-term response to the antibody-chemotherapy combination treatment.

The current results provide further evidence that early tumor shrinkage is an indication that a patient with wild-type KRAS colorectal cancer is not only sensitive to cetuximab but is likely to have a long-term response from therapy. Patients were followed for a median of 45 months in the CRYSTAL trial and 32 months in the OPUS trial.

The authors found a significant interaction of early tumor shrinkage and treatment type with progression-free survival (P = .028, P = .004 for CRYSTAL and OPUS data sets, respectively) but not for survival (P = .573 and P = .546, respectively). According to the authors, the lack of association with survival could be due to the trial designs which may not have been powered to show a survival benefit or it may reflect the relationship between early tumor shrinkage and survival for this cancer and therapy type.

The extent of early tumor shrinkage correlated with outcomes, leading the authors to suggest that “treatment-specific outcome measures might be investigated in future clinical trials, as the information carried by tumor size reduction may reflect different biologic processes.” For those patients who received FOLFIRI chemotherapy plus cetuximab, early tumor shrinkage of 40% or more led to a better outcome compared to shrinkage between 10% and 40%.

The current analysis used data from these two trials which measured tumor burden at 8 weeks, but further studies are needed to better understand if a tumor response beyond 8 weeks is also a marker for response.

As Geoffrey R. Oxnard, MD, of the Dana-Farber Cancer Institute in Boston, and Lawrence H. Schwartz, MD, of Columbia University point out in their editorial of this study, previous analyses have mostly analyzed response phenotypes to therapies in single-arm cancer trials. “The novelty of [the current study] lies in their attempt to use early tumor shrinkage to identify a subset of patients who gain maximal benefit from a targeted therapy (cetuximab) when added to chemotherapy in a randomized trial,” state the editorial authors.

Besides suggesting that future prospect trials should include an early response criteria that allows patients to switch therapy if no evidence of response occurs, Oxnard and Schwartz also highlight that parsing patients into early responders and non-responders rather than by Response Evaluation Criteria in Solid Tumors (RECIST) criteria, may also facilitate the identification of new tissue or blood biomarkers.

“This work represents the beginning of a period in which, for those therapies without adequate predictive tissue biomarkers, response phenotype plays an increasing role in guiding our delivery of targeted therapies,” the editorial authors conclude.