Clovis (CLVS) was one of ASCO’s clear losers following safety issues for its lead program, CO-1686. Despite demonstrating robust efficacy in T790M+ lung cancer, hyperglycemia associated with the drug raised concerns about its ultimate market positioning vs. its close competitor, AstraZeneca’s (AZN) AZD9291.
At ASCO, the two drugs generated similar response rates of 58% and 64%, respectively, but with different safety profiles. While Astra’s safety profile was in line with expectations (skin toxicities, lung inflammation), Clovis disclosed an increase in hyperglycemia events which occurred in the majority of patients. The company also disclosed that although in most cases blood sugar levels were controlled with oral drugs, a handful of patients started injecting insulin. This triggered a shift in sentiment favoring AZD9291 solely based on safety and patient preference.
On a standalone basis, side effects with CO-1686 are manageable and clinical risk/reward is still favorable but given AZD9291’s “cleaner” profile, AstraZeneca now has the upper hand. Importantly, the market’s focus on side effects relies on the assumption that the two drugs are comparable in terms of efficacy. If one of the drugs proves to have superior efficacy, this would overshadow any safety differences and guide physician preference. Going forward, investors will compare PFS of the two drugs across their respective studies. At ASCO, Clovis’ PFS curve appeared optically better but data sets were immature and therefore unreliable.
CO-1686 has value even as an underdog
Both companies are aggressively pursuing Tarceva/Iressa failures with T790M+ as a first indication. At its ASCO analyst event, AstraZeneca disclosed they will file for accelerated approval in 1Q15, slightly ahead of Clovis’ timelines (mid-2015). For first line EGFR mutated patients, Clovis has a 6-month lead as it expects to start a head-to-head trial vs. Tarceva around mid-2015. Both companies intend to pursue combinations with PD-1 antibodies later this year whereas AstraZeneca has an entire pipeline to choose from regarding additional combination studies.
Acknowledging AZD9291’s milder safety profile, CO-1686 may still capture a meaningful market share assuming efficacy is comparable. For diabetic patients, AZD9291 will probably be the drug of choice but some patients may prefer taking CO-1686 with metformin (which is an extremely safe drug) in order to avoid the EGFR-related toxicities (GI and skin toxicities) associated with AZD9291. EGFR-related side effects are not life threatening but they are viewed as a major issue for some patients. This is particularly relevant for 2nd line patients who suffered from EGFR-related side effects while on Tarceva for almost a year.
Assuming 70/30 split in favor of AZD9291, CO-1686 still represents a $350M and $700M commercial opportunity for 2nd and 1st line, respectively. As a reference, Tarceva and Iressa generate ~$2B annually.
A diversified pipeline
Despite the disappointment at ASCO, I intend to keep Clovis and take advantage of future price fluctuations for adding more shares. Clovis has 3 drugs with clear activity in humans and a multitude of risk/reward profiles:
CO-1686 has a high likelihood of approval and although it will probably not become the market leader, it still represents a significant commercial opportunity.
Rucaparib (PARP inhibitor) has clear activity in breast and ovarian cancer with BRCA+ tumors but the PARP arena is highly competitive. A potential differentiator might be Clovis’ approach for patient selection beyond BRCA+ tumors. An ongoing phase II in ovarian cancer is designed to identify additional subsets that may be sensitive to rucaparib.
Lucitanib (VEGFR/FGFR) is still early in development but preliminary data in FGF-aberrant breast cancer are very encouraging. At ASCO, Clovis presented already published results showing a PR in 50% 6/12 of patients and disease stabilization in the remaining patients. The lack of hyperphosphatemia (increased phosphate levels in the blood) led the discussant (David B. Solit) to conclude that the effect is not attributed to FGFR inhibition. Nevertheless, it is plausible that lucitanib does not cause hyperphosphatemia because it does not inhibit FGFR4, in contrast to other FGFR inhibitors in development. This is supported by a recent publication that shows that hyperphosphatemia emerges only when both FGFR1 and 4 are inhibited.
Increased focus on lucitanib in 2014
With a market cap of $1.55B, the market ascribes most of its valuation relies predominantly on CO-1686 and ascribes limited value to rucaparib or lucitanib. Lucitanib (which is why I got into Clovis 6 months ago) may become an important part of the Clovis story later this year if the drug continues to show efficacy in FGF-aberrant breast cancer. Servier, which has rights to the drug outside of US and Japan, could have initial data from a breast cancer phase II it opened in December 2013. The trial is recruiting both FGFR1+ and FGFR1- patients in order to assess whether lucitanib’s activity is more pronounced in biomarker-positive cases. Clovis will start a US trial in FGFR-driven lung cancer.
If lucitanib continues to demonstrate a >30% response rate in FGF-aberrant breast cancer, it may become a new stand of care for FGF-aberrant breast cancer. Given that FGF-aberrations (FGFR1+ or 11q+) are observed in 25% of cases, this segment could be as big as the HER-2 market, which represents a $1.5B global opportunity in advanced breast cancer (excluding adjuvant). Activity in lung cancer (data expected in 2015) could open a similar commercial opportunity based on a ~15% incidence.
From a competition perspective, there are other FGFR inhibitors in clinical development from Novartis (NVS) (BGJ398), AstraZeneca (AZD4547) and J&J (JNJ) (JNJ-42756493) but all are FGFR selective and do not inhibit VEGFR. Based on the limited available data generated by these molecules, their activity as monotherapy is limited to FGFR-fusions, which are rare and occur across multiple tumor types. Of the dual FGFR/VEGFR inhibitors, lucitanib has a clear lead.
In summary, Clovis should be viewed as a core oncology holding based on a pipeline of 3 targeted therapies with clinical proof-of-concept, biomarker-defined patient populations and a capable management team with a proven track record. Although the company has significant commercial rights for all three programs, its market cap reflects only CO-1686’s potential in 2nd line T790M+ lung cancer assuming AstraZeneca gets the lion share of the market. Updates from lucitanib may validate the efficacy signal seen for lucitanib in breast cancer and put Clovis in the lead in a $1.5B race.
We are initiating a position in Foundation Medicine (FMI) based on the growing need of genome profiling solutions in oncology (discussed in my post-ASCO write-up). We are also selling Curis at a 36% profit. We plan to add an additional position in Clovis during 2014 based on price fluctuations.
Portfolio holdings – June 15th, 2014