Long-Term Cetuximab Therapy Shows Early Tumor Shrinkage in Colorectal Cancer

November 2013, Vol 4, No 9 - In the Literature


Although colorectal tumors with a mutation in the KRAS gene generally do not benefit from epidermal growth factor receptor (EGFR)-targeted therapy such as cetuximab (Erbitux), there are no current biomarkers to select patients who are more likely to respond to EGFR therapy. In a new study, researchers analyzed data from 2 trials to determine if early tumor shrinkage was associated with long-term outcomes in patients with colorectal cancer receiving first-line treatment with cetuximab (Piessevaux H, et al. J Clin Oncol. 2013;31:3764-3775).

The researchers combined data from the CRYSTAL and the OPUS trials. CRYSTAL was a randomized, open-label, multicenter, phase 3 trial comparing fluorouracil, leucovorin, and irinotecan (FOLFIRI) plus cetuximab with FOLFIRI alone. OPUS was a randomized, open-label, multicenter, phase 2 trial comparing fluorouracil, leucovorin, and oxaliplatin (FOLFOX-4) plus cetuximab with FOLFOX-4 alone. The analysis included 1289 patients whose tumors could be analyzed for KRAS mutation status. Patients were followed for a median of 45 months and 32 months in the CRYSTAL and OPUS trials, respectively. Response was assessed every 8 weeks until disease progression or withdrawal.

Both trials showed that a more robust tumor response at 8 weeks after the start of therapy was associated with improved progression-free survival (PFS) and overall survival (OS) in patients with KRAS wild-type tumors. Early tumor shrinkage of ≥20% could identify patients who were receiving chemotherapy plus cetuximab with longer PFS and OS. In the CRYSTAL and OPUS trials, respectively, the cutoff value of early tumor shrinkage of ≥20% versus <20% for median PFS was 14.1 versus 7.3 months and 11.9 versus 5.7 months, and median survival was 30 months versus 18.6 months and 26 months versus 15.7 months. The interaction between early tumor shrinkage and the treatment group (with cetuximab) was significantly associated with PFS (P = .027 and P = .004 for CRYSTAL and OPUS data, respectively), but not for survival (P = .573 and P = .546, respectively).

These results suggest that tumor shrinkage can be used as a prognostic biomarker.

In their accompanying editorial, Oxnard and Schwartz state that the study suggests several potential applications, including a response-guided treatment approach in which patients without early tumor response to chemotherapy plus cetuximab are switched to another therapeutic agent (Oxnard GR, Schwartz LH. J Clin Oncol. 2013;31:3739-3741). The findings also have the potential to improve tissue and biomarker discovery, they say.