Merrill Lynch and Rachel on Ariad:
Ariad Pharmaceuticals, Inc.
Expert lung cancer feedback
?Key themes in lung cancer drug development
We highlight key takeaways based on our recent discussion with a key opinion leader in non-small cell lung cancer.
Novartis' LDK398 first of second gens to market, but tolerability concerns detract
As data matures for second-generation ALK inhibitors, Novartis' LDK398 has notable side-effects that are likely to relegate it to Xalkori failures, including higher rates of transaminitis and quality of life-impacting chronic GI events that force substantial dose reductions. Expect Novartis to file for accelerated approval by YE in Xalkori failures, suggesting approval in 3Q14; a proposed Phase 3 frontline study is being designed vs. chemo, not Xalkori, for fear of not generating a substantial enough benefit head to head.
'113 looking more competitive, shot at best-in-class ALK inh
'113 is earlier stage in development than NVS, but data generated to date show no apparent liver or meaningful GI issues. The most notable '113 adverse events are respiratory (shortness of breath), and new data at the ESMO meeting (Sept 27-Oct 1) should clarify whether these events are drug- or disease-related. Brain mets activity shows a strong signal for '113, and ARIA will have the benefit of collecting these data in Phase 2 (NVS can't get meaningful data retrospectively). Ability to maintain patients on high active doses should make '113 a better choice vs. LDK398 in Xalkori failures, and bolster the potential to show superiority over Xalkori in future front-line studies. Our expert doesn't expect Chugai's 90%+ response rate to stand over time when tested outside of Japan.
CLVS '1686 groundbreaking for T790M EGFR, '113 unlikely to show benefit
Our expert believes that CLVS CO-1686 data, while limited, shows ground-breaking activity for patients with T790M EGFR mutated NSCLC. However, the 75% response rate may weaken over time, not only due to small numbers tested initially, but also because at least some of the patients tested may have had tumors less dependent on T790M, since they were not dosed immediately after failing Tarceva. As data matures, a response rate of >40% with 6+ months duration of response is considered clinically meaningful. While it is worth testing '113 in T790M, our expert felt (and we agree) that if '113 were to be active in T790M, the signal in the initial Phase 1 would have been strong. First data from AZN's 9291 in T790M is expected at the ESMO meeting as well, and worth watching to understand how the competitive landscape is evolving.
AI
