Last week, Clovis Oncology (CLVS) announced the acquisition of EOS, a small Italian company with an oncology program in phase II. To me, this is the smartest deal in 2013 so far given the relatively low price ($200M) Clovis paid for a mid stage oncology asset with a strong clinical proof of concept and a clear differentiation from competition. Clovis has commercialization rights for US and Japan while Servier has EU+ROW according to a previous licensing agreement with EOS.
EOS’ drug, lucitanib (E-3810) is a dual VEGF/FGF pathway inhibitor. The FGF pathway became a popular target based on the high incidence of FGF pathway activating mutations in many tumor types. In addition, the FGF family is implicated in general cancerous processes such as angiogenesis and migration. While the VEGF pathway has been successfully pursued with small molecules (Sutent, Inlyta, Nexavar), unlocking the therapeutic potential of the FGF pathway proved more challenging.
One of the reasons for the limited success lies in the complexity of the FGF network which includes 4 receptors and over 20 different ligands with diverse biological functions. Another reason may be the fact that even in patients whose tumors have FGF pathway activation, inhibiting the pathway alone is not sufficient for a pronounced clinical effect. In other words, “FGF-positive” cells are not addicted to FGFR signaling the way cells with other mutations (ALK, EGFR, BRAF) are addicted to their respective mutations.
Lucitanib has the right selectivity profile
There are many (>10) small molecule FGFR inhibitors in development. Some (Novartis, Ariad, Boehringer Ingelheim) developed broad spectrum drugs that inhibit multiple targets on top of FGFR (in some cases FGFR was not the primary target) whereas others (Astra Zeneca, Novartis, Lilly) developed more selective FGFR inhibitors. Although, both approaches have their merits, the general trend today favors selective inhibitors (especially in biomarker defined populations) due to their ability to maximally hit the desired target without unnecessary toxicity.
Lucitanib falls somewhere in the middle. It is not a pure FGFR inhibitor but it inhibits only a handful of additional targets (predominantly VEGFR).Within the FGFR and VEGFR families, lucitanib inhibits FGFR1 and VEGFR2 (both considered the primary receptors in their respective network) with similar potency. This appears to be a distinguishing property that may explain why lucitanib is the only FGFR inhibitor with good clinical activity.
Another unique feature of lucitanib is strong inhibition of CSF1R, which was noted in EOS’ presentations but not in Clovis’ recent presentation. Nevertheless, it is not clear whether CSF1R is involved in the drug’s activity.
Lucitanib’s clinical proof of concept comes from a small cohort of breast cancer patients whose tumors had activation of the FGF pathway (FDF+).Of 12 evaluable patients 6 (50%) achieved a partial response (5 confirmed) and 5 achieved disease stabilization with a median progression-free survival (PFS) of 9.4 months. Despite the small sample size, this response profile cannot be coincidental and compares very favorably with other FGFR inhibitors which have demonstrated little to no effect in similar patients. As a benchmark, Kadcyla, which is considered a breakthrough drug in HER2+ breast cancer, had a response rate of 38.4% and a PFS of ~5 months in phase II.
The most provocative finding was the impressive activity lucitanib had in some patients who had previously failed other FGFR inhibitors. Clovis’ presentation reports 4 such patients, all of whom were on lucitanib much longer than on prior FGFR inhibitors. One patient, who progressed on Novartis’ (NVS) selective FGFR inhibitor after 2 weeks, is still on lucitanib for 21 months. This provides the ultimate proof that lucitanib is a best-in-class FGFR inhibitor.
Lucitanib’s addressable market is significant based on the high incidence of FGF+ tumors across multiple tumor types. The opportunity in breast cancer alone (23.7% of cases) is comparable to the HER2+ segment, which represents a $1B market in advanced stage disease (excluding adjuvant therapy). FGF activation has been documented in other tumors including lung, ovarian and head and neck cancer. FGFR1-amplified lung cancer appears particularly promising, as there are efficacy signs in this indication with other FGFR inhibitors. FGFR1 amplification occurs predominantly in squamous-NSCLC, which are not candidates for Avastin due to safety concerns. The risk-reward profile of a dual VEGFR/FGFR inhibitor in these patients is yet to be evaluated.
There are over 10 active programs for FGFR inhibitors as well as multiple antibodies in development. To date, lucitanib remains the only drug with compelling activity (others have shown limited to no activity in FGF+ tumors) and it may be the only drug to be approved as monotherapy.
Results to date imply that selective FGFR inhibitors are not sufficiently active in humans and that VEGFR inhibition is required to potentiate the effect. There are only 2 other drugs with a similar selectivity spectrum (selective FGFR1/VEGFR2): Lilly’s (LLY) LY2874455 and a compound developed by Advenchen, which discovered lucitanib before licensing it to EOS.
Selective FGFR inhibitors such as Novartis’ BGJ398 and AstraZeneca’s (AZN) AZD4547 must be combined with VEGFR inhibitors in order to replicate lucitanib’s activity in FGF+ patients. Although technically this can be achieved even through co-formulation in the same capsule, finding the exact regimen and dose ratio is not always straightforward. On the other hand, using 2 distinct drugs allows for flexibility in dosing regimen and ratios.
Most of the VEGFR inhibitors (approved or in development) are not selective and combinations may be too toxic. The only truly selective VEGFR inhibitor is Aveo’s (AVEO) tivozanib. Its selectivity and ideal safety profile make tivozanib an ideal candidate for combination with selective FGFR. Novartis and AstraZeneca are pursuing combinations with several drugs but not with VEGFR inhibitors.
We are initiating a position in Clovis, which now has 3 potential blockbusters with clinical proof of concept and biomarker-defined target populations. We are also selling Gilead (GILD) for a profit of 250%.
Portfolio holdings – December 1st 2013