POSEIDON Trial: Second-Line Abituzumab Combined with Cetuximab plus Irinotecan in Patients with KRAS Wild-Type Metastatic Colorectal Cancer

Conference Correspondent - ESMO 2014 - Gastrointestinal and Head & Neck Cancer


Abituzumab (EMD 525797, DI17E6) is a humanized monoclonal antibody directed specifically against the αγ subunit of human integrin receptors. This novel therapy demonstrated promising antitumor activity in experimental models, which was further potentiated when combined with cetuximab. These early results were the rationale for the phase 1/2 open-label, randomized, controlled, multicenter POSEIDON clinical trial that evaluated 2 doses of abituzumab ?1000 mg plus standard of care (Hohler et al. POSEIDON Trial. ESMO 2014: Abstract 507PD).

In this trial, 216 patients with confirmed KRAS wild-type metastatic colorectal cancer that failed oxaliplatin/fluoropyrimidine-containing treatment were randomized in a 1:1:1 ratio to the standard of care of cetuximab plus irinotecan (n = 72) at a standard dose and schedule alone, or that regimen in combination with intravenous abituzumab 500 mg (n = 73) or 1000 mg (n = 71) every 2 weeks.

The results of the phase 2 portion of the trial showed no difference in investigator- assessed median progression-free survival or in response rates among the 3 treatment arms. Although there was a trend for prolonged overall survival (OS) in the 2 combination arms compared with the standard of care, that difference was not statistically significant (15.0 vs 14.4 vs 11.6 months). Further prognostic and predictive biomarker analysis of this patient population was performed and was also reported.

High integrin ?vß6 expression was found in 98 of the 197 tumor tissues analyzed, which was a negative prognostic factor for OS compared with patients with low ?vß6 expression in the standard-of-care arm (10.2 vs 13.8 months; hazard ratio [HR], 1.96; P = .037). Although there was a trend toward the prognostic value of high ?vß5 and ?v expression levels, these did not reach statistical significance. Similarly, high expression levels of ?vß6 (15.3 vs 10.2 months; HR, 0.48; P = .008) and ?v (18.8 vs 10.2 months; HR, 0.53; P = .025), but not of ?vß5, were predictive of improved OS compared with standard of care.

Exploratory analyses identified pretreatment plasma proteins of CCL23 as a candidate predictive and prognostic biomarker for OS. Clearly, these provocative results need to be validated in further research to better define their utility as prognostic and predictive tools for guiding treatment decision-making.