Some comments from Btechnews on the Iressa results.
Ice
DISAPPOINTING IRESSA RESULTS
Yesterday AstraZeneca (NYSE: AZN) announced that Iressa plus platinum-based chemotherapy did not increase survival rates of patients with non-small cell lung cancer (NSCLC). The Iressa results, coming after the well-publicized problems with ImClone (NASDAQ: IMCL) and Bristol-Myers Squibb's NYSE: BMY) epidermal growth factor receptor (EGFR) inhibitor Erbitux, raised concerns on Wall Street about the effectiveness of other EGFR inhibitors such as OSI (NASDAQ: OSIP), Roche (Swiss: ROG), and Genentech's (NYSE: DNA) Tarceva as well as Abgenix (NASDAQ: ABGX) and Amgen's (NASDAQ: AMGN) ABX-EGF. While it would be nice to be able to make a clear statement that these results either significantly lessen the probability of the entire class of EGFR inhibitors being approved or that Iressa's results have no relevance to Tarceva or ABX-EGF, the story is much more complicated.
The science underlying EGFR inhibitors is very strong. The problems with the individual drugs in this class highlight the difficulty of translating scientific advances into actual drugs. In addition, cancer is an extremely complex set of diseases; only rarely is a drug clearly effective in treating cancer, as was the case with Novartis's (NYSE: NVS) Gleevec. The more usual scenario is that a new drug will demonstrate enough of a survival advantage in certain, specific situations to be approved. As the medical community learns more about the drug and how to utilize it in a comprehensive treatment protocol, the drug expands into wider use. Therefore, results in which a class of drugs shows promise in some scenarios but then fails to demonstrate efficacy in others is the norm. In addition, factors such as trial design, specific molecule design, and drug-drug interactions in combination therapy can have a large impact on trial results for a given drug. Therefore, making judgments about the entire class of EGFR inhibitors is difficult based on the limited information that has been released by AstraZeneca.
Having said that, AstraZeneca's statements were certainly negative for Iressa. As AstraZeneca stated in its press release and conference call, the trials were well designed. The company's announcement concerned the first substantive clinical survival endpoint data generated by Iressa trials. Previous results, superficially more positive, have demonstrated effects only on surrogate or palliative endpoints. Surrogate endpoint data in anti-cancer agent trials rely on measurements of tumor shrinkage, response rates, and so forth rather than clinical endpoints such as survival, disease-free survival, or progression-free survival. Therapeutic benefit in terms of surrogate endpoints, though fine for the typical phase II-III transition, is only on rare occasion enough for FDA drug approval. Palliative endpoints rely on measurements of symptomatic relief leading to comfort as well as freedom from pain and distress.
As we stated in our May 28th report, Iressa's tumor response data from its previous phase II trials, which AstraZeneca has used to file for U.S. regulatory approval, are not strong enough to support a typical cancer drug' s approval. However, Iressa did demonstrate sufficient palliative results - i.e. relief of shortness of breath, coughing, and chest tightness - to warrant FDA consideration in terminal NSCLC. In this conservative FDA environment, much will rest on the recommendation of the oncologic drug advisory committee (ODAC), which meets on September 24.
AstraZeneca indicated today that it will not abandon this application because of the disappointing phase III data. However, approval based on palliative effects in a relative small phase II trial was already risky. The preliminary phase III results of Iressa in combination with chemotherapy further decrease the odds of approval in 2002. At the very least, the FDA will likely want to examine more substantive phase III results and may require additional trials of Iressa.
How Iressa's disappointing results impact assessments of other EGFR inhibitors is an important question. The results could be specific to Iressa, specific to small molecule EGFR inhibitors such as Tarceva but not antibodies like Erbitux and ABX-EGF, or general for all EGFR inhibitors in certain clinical scenarios such as EGFR inhibitors in combination with platinum-based chemotherapy.
Iressa's results increase the probability that Tarceva monotherapy will be significantly superior to Iressa therapy. OSI, Roche, and Genentech are conducting high-powered phase III trials with both Tarceva alone and in combination with chemotherapy in NSCLC patients. Well-designed phase II trials have already demonstrated survival advantages of Tarceva monotherapy in chemo-refractory NSCLC patients. 40% of recipients were alive at 12 months (compared to a historical chemotherapy control of on the order of 25-35%). Other evidence of significant survival advantages was observed, and is likely to be confirmed in higher-powered studies.
Even if Tarceva, like Iressa, does not increase the benefits of chemotherapy, survival advantages with Tarceva alone would remain important. The only shortcoming of the Tarceva data to date is a lack of substantive quality-of-life and symptomatic information. Confirming phase II efficacy in NSCLC would set up more secure off-label use in head and neck cancer (currently in phase II study) and ovarian cancer (currently in phase II study). Even given the concern that Iressa results bode poorly for Tarceva, the agent continues to have above average odds of FDA approval. Yesterday's 57% sell-off of OSI shares overestimates the risk associated with investment in the company based upon this drug.
Erbitux has recently begun higher-powered phase III trials that will address the concerns of the FDA coming after its December "Refuse To File" letter and further concerns relating to data presented in May. Erbitux has a slightly lower than average chance of being approved by the FDA compared to other drugs in phase III trials, and will not be approved before 2005 at best. The phase III Iressa results do not significantly impact our view of Erbitux's odds of approval.
ABX-EGF has not generated enough clinical data thus far to warrant attempts to translate Iressa's disappointing results into predictions of ABX-EGF's probability of success. However, Abgenix's investment thesis is not based on the success of any one of its current pipeline drugs. Rather, it is based on an assessment that Abgenix will be successful in using its drug development platform to generate a robust clinical pipeline and its ability to efficiently sort through this pipeline and invest in the drug candidates that are most likely to be successful.
btechnews.com |
