Array Biopharma – Following Puma’s footsteps
Last month Array Biopharma (ARRY) announced a licensing deal with Oncothyreon (ONTY) for ARRY-380, a selective HER2 kinase inhibitor for the treatment of HER2+ breast cancer. ARRY-380 is regarded as an insignificant program, evidenced by the modest deal size ($10M upfront) and the lack of market reaction, but this could change once investors make the connection between ARRY-380 and Puma Biotechnology’s (PBYI) neratinib (EGFR/HER2 inhibitor). Although neratinib is more advanced, backed by more clinical data and probably has broader potential, ARRY-380’s selectivity profile could differentiate it in certain clinical settings. As the market is clearly excited with neratinib (Puma has a market cap of ~$1.6B), some of the excitement could eventually be tunneled toward Array as well.
Renewed interest in HER2 kinase inhibitors
HER2 kinase inhibitors are making a comeback largely thanks to Puma Biotechnology, who is developing neratinib, a dual EGFR/HER2 inhibitor for HER2+ breast cancer. Today, targeting HER2 with kinase inhibitors is playing a minor role in the HER2+ breast cancer arena that has been dominated by HER2 antibodies from Roche: Herceptin, Perjeta and Kadcyla. GSK’s (GSK) Tykerb is the only approved kinase inhibitor for HER2+ breast cancer but it is not commonly used due to an unfavorable clinical profile (mild efficacy and safety issues). It is typically reserved for patients who progress on HER2 antibodies (3rd/4th line patients).
Puma’s neratinib (licensed from Pfizer in 2011) is similar to Tykerb in its spectrum of activity (both drugs are dual EGFR/HER2 inhibitors) but as an irreversible inhibitor, it is considered more potent (also more toxic) with potentially broader clinical applications.
Puma is positioning neratinib as an improved next-gen HER2 kinase inhibitor with the intention of replacing Tykerb in 3rd/4th line HER2 breast cancer as well as getting the drug approved in settings where Tykerb is considered ineffective. These include HER2+ breast cancer with brain metastases and 2 niche indications where HER2 is not amplified but mutated (small subsets in breast and lung cancer). Puma is also evaluating neratinib in early-stage localized HER2+ breast cancer.
What ARRY-380 has to offer?
As a selective HER2 inhibitor, ARRY-380 has the potential to be a safer and more effective alternative to neratinib in certain clinical settings. Its main advantage is in the selective targeting of HER2 with limited activity on EGFR. This is an important differentiator in HER2+ breast cancer, where HER2 plays a significant role while adding EGFR inhibition has a questionable benefit.
Historically, when Tykerb and neratinib were developed, targeting both EGFR and HER2 made sense as a way to overcome cross resistance. Today we know that inhibiting both targets is problematic in HER2+ breast cancer because it adds toxicity and precludes complete HER2 inhibition. This explains Tykerb’s modest benefit (no survival benefit) and problematic safety profile (gastrointestinal [GI] toxicities) compared to the HER2 antibodies, all of which extend survival and are well tolerated. Interestingly, there is no approved combination of selective EGFR and HER2 drugs as past results showed increased toxicity and no clear added value. A recent study which evaluated Perjeta (HER2 antibody) with Erbitux (EGFR antibody) concluded the combination was too toxic.
Although Puma’s neratinib looks more potent than Tykerb (cross-trial comparison) and has potential utility in new indications (tumors with non-amplified HER2 but with HER2 mutations) it is basically more of the same with respect to HER2+ breast cancer. Like Tykerb, it is associated with a high degree of GI toxicities that may limit market adoption (especially with the regimen pursued in the current phase III). ARRY-380, on the other hand, is very safe and can inhibit HER2 potently without causing side effects. This also makes it more suitable for combination regimens, which is the preferred route for HER2 kinase inhibitors.
Is ARRY-380 potent enough?
The biggest question regarding ARRY-380 is whether it is efficacious enough relatively to Tykerb and neratinib. Array reported phase I results which demonstrated activity in several advanced HER2+ breast cancer patients but it is still unclear whether the theoretical ability to hit HER2 harder coupled with improved safety profile will lead to clinical superiority.
ARRY-380 only has a limited data set from a phase I trial. The trial recruited 50 patients with HER2+ tumors, the majority of whom had breast cancer. As expected, the drug appears to have a differentiated safety profile with minimal skin and GI toxicities. Activity at the highest dose (20 evaluable patients with HER2+ breast cancer) included 2 confirmed PRs, 1 unconfirmed PR and multiple cases of tumor shrinkage.
How does a modest objective response rate of 10% stack up against competition? As monotherapy, Tykerb and neratinib led to response rates of 4-8% and 24%, respectively. At first glance, ARRY-380 seems to fall at the lower end of the spectrum but it is important to remember that its phase I included patients who were more advanced. In particular, all patients failed prior Herceptin and the vast majority also failed Tykerb. In contrast, Tykerb’s and neratinib’s results were generated in far less heavily pretreated patients, all of whom were Tykerb-naïve.
Oncothyreon and Array are expected to start several combination trials which could have top line data as early as Q4 2014. The most straightforward route to registration is replicating neratinib’s phase III design in 3rd/4th line patients in combination with Xeloda. Another potential route could be HER2+ breast cancer brain metastases.
In summary, ARRY-380 clearly has activity in last line patients and future combination trials are needed for showing its true clinical relevance. Puma’s neratinib, which just started phase III, has a 2-year lead but ARRY-380 unique selectivity profile might make it a best-in-class HER2 inhibitor (At least in HER2+ breast cancer). As ARRY380 starts to generate results, investors might start to assign more value to the drug, especially in light of the excitement around Puma.
From a valuation perspective, ARRY-380’s deal adds another phase II program to Array’s partnered pipeline, which is becoming increasingly hard to track. I counted 10 active partnered programs, including 2 programs in phase III. With additional 4 proprietary programs, Array is one of the most diversified biotech companies and probably one of the cheapest (~$580M market cap).
Morphosys signs lucrative deal with Celgene
Morphosys (MOR.DE) announced a huge licensing deal with Celgene (CELG) for MOR202, a CD38 antibody for the treatment of multiple myeloma and other blood cancers. The attractive economics ($92 upfront payment and $60M equity investment) demonstrate the excitement around CD38 antibodies following promising results with Genmab’s (GEN.CO) CD38 antibody, daratumumab. Daratumumab, which recently recveived breakthrough designation from the FDA, is leading the race, followed by Morphosys’ MOR202 and Sanofi’s/Immunogen’s (IMGN) SAR650984.
At the moment, daratumumab is the only anti-CD38 antibody with published results. The other 2 antibodies are in phase I for 2-3 years but no data have been made available from these trials. As a result, it is unclear whether they also have promising activity. Genmab insists its antibody has superior properties and management has been using every opportunity to downplay competition. Genmab even produced its own versions of MOR202 and SAR650984 in order to show daratumumab’s superiority in certain preclinical assays.
The deal with Celgene, which has the largest myeloma franchise in the world, is a very positive indication since Morphosys had to share all the data with them. Nevertheless, until results are available this is only a hypothesis.
For Morphosys’ investors, the deal with Celgene is transformational as MOR202 is the company’s most important asset. If one assumes MOR202’s results are as good as those of daratumumab, Morphosys is still undervalued. If results are not as good, it will have an impact on the stock but the downside is still limited thanks to Morphosys’ large partnered pipeline and the strong cash flow from licensing deals.
Amgen’s bid for Onyx could start a bidding war
Last week, Amgen (AMGN) disclosed a $120 a share offer for Onyx, which was immediately rejected by Onyx’s board of directors. Onyx also announced it is seeking alternative offers with better terms, a standard step following an initial acquisition bid.
As I discussed earlier this year, Onyx is in a unique position with 3 commercial assets and a royalty stake in a promising late stage assets (Pfizer’s [PFE] palbociclib). It is clear that for a company like Amgen, Kyprolis represents the most attractive asset as the drug is expected to grow rapidly in the coming years and Onyx has global (Ex Japan) rights for the drug. Kyprolis’ current run rate is ~$250M and is expected to be a $1.5-$2B drug by 2018 providing positive readouts from 3 pivotal trials.
For Amgen, Kyprolis is a perfect fit given its attempts to branch out from supportive care hematological treatments to therapeutic hematology. In 2011, Amgen acquired Micromet for its lead agent blinatumomab for ALL which is expected to reach the market next year. To date, Amgen has been unsuccessful in introducing new cancer therapies and despite some positive phase III readouts for 2 programs (T-Vec and trebananib), the market is skeptic about their value.
Amgen’s initial bid is likely to encourage other bidders to try and acquire Onyx. This dynamic typically leads to a slightly higher price tag, which, is predicted to be in the $140-150 range ($12B valuation) based on a flurry of analyst reports. Even assuming Kyprolis is approved for earlier lines of multiple myeloma and factoring in a better than expected launch for Stivarga, a valuation of $12B (including ~$7 per share cash position) looks like a good offer.
Synta – Good news, bad news
Following the sharp decline in Synta’s (SNTA) shares, it was encouraging to see insider buying including almost 3 million shares that were bought by board member Bruce Kovner and 50 thousand shares bought by the CEO, Safi Bahcall. This does not guarantee anything but when insiders buy shares it usually means they think the price is attractive.
Last week, Jefferies’ Thomas Wei issued a note with several interesting observations following a meeting with Synta’s management. With respect to the phase II (GALAXI-1) results, he brought up geographic imbalances between the two arms as a possible explanation for the “diminishing” survival benefit. The control arm had a higher proportion of patients from eastern geographies, who appeared to be less advanced than patients in western geographies (survival of 8-10 vs. 5 months for the 2 groups respectively). In the ongoing phase III, the two arms will be geographically balanced.
As I discussed on last month’s post, the most reliable finding is always a straightforward ITT (intent-to- treat) analysis, which is clean and unbiased. ITT analysis of GALAXI-1 showed a clear survival trend on the verge of statistical significance as well as a statistically significant PFS benefit. If replicated in the ongoing phase III, ganetespib’s survival benefit will resemble that of Avastin (see figure below). This assumes no impact from Synta’s decision to limit enrollment in the phase III to patients with less aggressive disease who appeared to derive a more substantial benefit in the phase II.
2 additional points from the report are a complete response from the company’s ongoing triple-negative breast cancer trial (we already know there is another PR in that trial) and a forthcoming licensing deal in Asia.
On the bad news front – It appears that ganetespib is not potent enough as a single agent in ALK-mutated lung cancer. Although no results have been published from an ongoing monotherapy phase II, I conclude this based on: (i) the company’s decision to develop ganetespib only in combination regimens and (ii) the fact that read out from the CHIARA trial (1st line ALK+) is no longer listed on the clinical milestones slide in Synta’s presentation. Importantly, ganetespib still has value as an add-on to ALK inhibitors, which is being evaluated in an investigator-sponsored trial.
We are selling a third of our position in Celldex (CLDX). I am still bullish on the company’s phase III programs but this looks like a good opportunity to lock in some profit (660%) as well as minimize exposure to a single position that accounted for over 10% of the entire portfolio.
Portfolio holdings – July 7th 2013