Biotech Values

[pcrutch]

From Alex To -- CLDX: Statistically Significant Overall Survival An Upside Surprise

• At the San Antonio Breast Cancer Symposium on Saturday, Celldex presented the mature Phase 2 data on CDX-011. In triple negative breast cancer patients who were also high GPNMB expressers, the study achieved statistically significant differences in both overall survival (OS) and progression free survival (PFS). The median OS in the CDX-011 group was 10.0 months (n=10), while the median OS in the control group was 5.5 months (n=6). The p value was 0.003. The median PFS in the CDX-011 patients was 3.0 months (n=12), and the median PFS in the control group was 1.5 months (n=6). The p value was 0.008. The control group was the investigator's choice of salvage therapy. The patient population was heavily pre-treated, and the study was conducted as last-line salvage therapy.

• In line with a Phase 2 exploratory trial, the data set is naturally small. Therefore, one should exercise caution in interpreting the statistics. However, the rest of the results in this Phase 2 trial overwhelmingly corroborate the impression that GPNMB is a highly promising new target for the treatment of breast cancer and that CDX-011 is a highly efficacious drug in this setting. Here is the key supporting evidence we gleaned from the trials. (1) In all GPNMB high expressers with all receptor subtypes - meaning not just the triple negative breast cancer patients, but also HER2+, or ER+, or PR+ patients - the median OS in the CDX-011 group was 10.0 months (n=22), while the median OS in the control group was 5.7 months (n=11), p=0.18. The PFS comparison was 2.7 months (n=27) versus 1.5 months (n=11), p=0.14. This piece of data not only correlates with the triple negative results, it also points to the promise that CDX-011 will not just be a drug for triple negative patients only, but will be able to address the 22% of all breast cancer patients whose tumors express this antigen.

This is a patient population about the same size as that of HER2+, which is the target for Herceptin. Roche (ROG.VX) realizes $6 billion annual sales from Herceptin. We view this as the market opportunity for CDX-011, not counting the other cancer types outside of breast cancer that can potentially benefit from CDX-011. (2) There is a nice correlation between the degree of response to CDX-011 and the degree of expression of GPNMB, which again proves the validity of the drug and the target. In all patients in the trial with less than 5% tumor expression of GPNMB, about 13% of the patients achieved a partial response or unconfirmed partial response (PR/uPR) after receiving CDX-011 (n=34), while PR/uPR in patients with more than 5% GPNMB expression was about 19%, in patients with more than 10% tumor GPNMB expression was 27% and in patients with more than 25% tumor GPNMB expression was 32%. There is no such correlation in the control group. (3) While the duration of response in triple negative patients is not overwhelming, the OS and PFS results make the concern over the duration of response moot. This is also corroborated by the findings in disease control rate: 75% versus 25% in triple negative patients with high GPNMB expression, and 64% versus 38% in all GPNMB high expressers.

• We think these Phase 2 results clearly demonstrated the promise of CDX-011. A partnership with a larger company in the near term will be beneficial. While it is entirely feasible for CLDX to run a small registrational trial in triple negative patients and to get the drug on the market as soon as possible, to realize the commercial potential of this drug, a large comprehensive pivotal program to explore CDX-011 in all GPNMB high expresser breast cancer patients, from the salvage setting all the way to front-line combination usages, plus exploring the drug in other GPNMB expressing cancer types, will be needed. With this data set, we will be surprised if no partner steps up to the plate.