Median overall survival also almost doubled in the CDX-011-treated patients to a median 10 months versus 5.5 months for patients in the control arm.
Results were statistically significant although the benefit was derived from an analysis that included just 12 patients treated with CDX-011 and 4 patients treated with investigator's choice chemotherapy.
Based on these data, Celldex would like to run a confirmatory trial for CDX-011 in patients with triple-negative breast cancer that also contains high levels of GPNMB, pending FDA agreement. Approximately 5,000 to 9,000 breast cancer patients in the U.S. fit this category, according to various estimates.
CDX-011 may benefit a broader group of 30,000 to 40,000 breast cancer patients in the U.S. with tumors that contain high levels of GPNMB.
Thirty-two percent of GPNMB high expressers responded to CDX-011 treatment compared to a 13% response rate in patients treated in the control arm in the phase II study. CDX-011 delayed the re-growth of tumors by a median 2.7 months compared to 1.5 months for the control arm. Median overall survival was 10 months for CDX-011 versus 5.7 months for the control arm. These results, culled from a subgroup of 38 patients, trended in favor of CDX-011 but were not statistically significant.
CDX-011 does not appear to benefit breast cancer patients with tumors that contain minimal levels of the GPNMB protein.
When all 122 patients in the phase II study were analyzed, treatment with CDX-011 resulted in a 16% response rate compared to a 14% response rate in patients treated with investigator's choice. The time to tumor re-growth and overall survival were essentially identical between both groups of patients.
Celldex Chief Medical Officer Tom Davis said different cutoff levels of GPNMB expression were examined in the phase II study, with 25% and above determined to be the level at which CDX-011 showed the greatest efficacy.
With relatively few patients to examine in the phase II study, however, the margin of error is greater. This raises the risk that the benefit of CDX-011 seen in the small phase II study may disappear when a larger study of the drug is conducted.