Bad Day for Genentech: Tarceva and Avastin Doubtful
Gregory Frykman, M.D.
Genentech’s (DNA 37) quarterly conference call yesterday heightened our concern about the future clinical trial
results for both their antiangiogenic drug Avastin (bevacizumab) and their epidermal growth factor receptor
(EGFR) blocker Tarceva (erlotinib) anticipated mid-2003.
We have already seen one EGFR blocker, Iressa (gefitinib), fail miserably in two well-conducted clinical trials
designed to demonstrate clinical benefit. However, we also believe that there are forces within the Food and
Drug Administration (FDA) that are still leaning toward approval of Iressa. Thus, the impending discussion
about the regulatorily-acceptable endpoints for developmental studies in non-small lung cancer (NSCLC) to be
held on April 15 is especially timely and raises the stakes for the next-in-class EGFR blocker, Tarceva.
Tarceva (of Genentech, Roche Holdings (RHHBY 65)² and OSI Pharmaceutical (OSIP 16.32)², which has
many similarities to Iressa and has undergone similar development, will be the subject of great interest when the
results from the first line NSCLC study are unblinded in mid-2003. Because of Tarceva’s similarity to Iressa in
terms of target, mechanism, and disease, we went into Genentech’s call with a negative bias toward Tarceva. We
learned that the data safety monitoring board (DSMB) recommended to Genentech to drop the Tarceva
continuation component of its trial because of a unspecified "safety signal."
This strongly suggests to us that the DSMB did not find any therapeutic advantage to Tarceva when added to
chemotherapy. Moreover, because of the Tarcerva’s inherent toxicity (most prominently rash), the DSMB likely
felt it was unethical to treat patients with the drug if Tarceva was not providing any clinical benefit. Unless
strong, credible, positive data emerges soon, we reiterate our pessimism that Tarceva will demonstrate clinical
benefit in NSCLC, though its ultimate regulatory fate will be dependent on how Iressa fares at the FDA (to be
known on or before May 6) and Tarceva’s own clinical trial results in NSCLC and pancreatic cancer.
On a similar note, we reiterate our pessimism about Avastin, Genentech’s prototype anti-VEGF antibody for
cancer. The field of angiogenesis inhibition has witnessed several disappointments in recent years and Avastin
(which has already failed in breast cancer) may be the next drug to disappoint. Results from a critical colon
cancer trial are due in mid-2003 and we are not expecting them to be positive. Moreover, with a 10% overall
response rate in renal cell carcinoma (RCC) from an earlier study, it is unlikely that a survival advantage will be
demonstrated in the two RCC studies Genentech is planning. Although results from these two studies will not be
known for two to three years, we also expect negative results. |
