It was nice to see M&A picking up in the 2nd half of 2019, with multiple high profile acquisitions. Two sectors of interest were kinase inhibitors for oncology (Array Biopharma, Loxo, ArQule) and Gene therapy for rare diseases (Spark, Audentes). siRNA came back to the limelight with Novartis’ $9.7B bid for MDCO for its twice-yearly PCSK9 treatment, marking a radical change in sentiment towards siRNA.
This uptick is great news for investors in small/mid-cap biotechs for which acquisitions are the primary exit route. Looking at some of the deals, I am having a hard time justifying valuations or seeing the acquirers making a decent ROI in the long run. What bothers me is not the fact that many acquired assets eventually fail but the amount some biopharmas are willing to spend for high risk assets, sometimes even preclinical ones.
To me, Dec 9 provided a good example with two acquisitions (ArQule and Synthorx) similar in size but very different in risk profile. I realize there might be some personal bias here as Pontifax is one of the largest shareholders in ArQule, but the fact that both companies were acquired with similar valuations is a real head-scratcher to me.
I try not to discuss ArQule (ARQL) for obvious reasons but suffice it to say that their lead drug has solid efficacy in humans with a clear route to market. Opportunity size is always debatable, ranging from hunderds of millions to several billions, but the drug has a high likelihood of reaching the market and generate revenues based on a straightforward clinical program. In other words, Merck is likely to see more than $2.5B in sales from this acquisition.
Synthorx’s (THOR) is on the other end of the spectrum as its lead agent (THOR707) just entered P1 with very limited visibility towards the market. THOR707 is an engineered IL2 designed to retain the cytokine’s beneficial effect on T cell activation while sparing side effects and undesirable activity (Treg activation and vascular leak syndrome). The drug’s potential is significant, of course, as with any IO (immuno-oncology) agent that is heralded as the “next PD1” but given the lack of clinical validation, the poor translation from animals to humans in IO and recent experience with other IO agents (Nektar’s next-gen IL2 in particular) one has to admit technical risk is high.
I like the drug and Synthorx’s platform and completely understand why it is an attractive acquisition target. Nevertheless, I am struggling to understand how someone can pay $2.5B for an asset that just started P1, knowing that most drugs fail and that success rate in IO is one of the worst in the industry. I realize Sanofi’s management was under a lot of pressure to unveil something new at its R&D event but if I were a Sanofi shareholder I would expect the company to treat shareholders’ money with a little more respect.
In general, I never understood lucrative preclinical deals where a pharma pays hundreds of millions upfront for something that hasn’t been proven in humans. These deals typically don’t age well, especially in areas of low technical success like IO.
A recent example can be seen with Novartis’ decision to terminate its STING program it had licensed from Aduro at the preclinical stage for an upfront of $225M (and they didn’t even get full commercialization rights). Just like engineered cytokines today, STING was a super-hot space in 2015, but that doesn’t necessarily change drug development statistics. Another notorious preclinical IO deal was the BMS/Flexus acquisition, in which BMS paid 800M upfront for a preclinical IDO inhibitor.
The trend seems to continue as earlier this year, Novartis acquired another company with no clinical data (lead program just started P1 in healthy volunteers) and paid $310M upfront. Novartis will probably write off these $310M, which still looks like an excessive amount for an asset with no clinical data but at least they didn’t pay $2.5B…
2019 as a turnaround year for ADCs
After years of stagnation, antibody-drug conjugates are finally gaining momentum.
Seattle Gentics (SGEN) and its partners won FDA approvals for Polivy (Anti CD79b for lymphoma) and Padcev (Anti-nectin4 for bladder cancer). Despite relying on legacy ADC technology used for Adcetris, both agents demonstrated unequivocal efficacy with a reasonable safety profile, and are likely to become commercially successful products.
On the next-gen ADC front, Daiichi presented updated results for its two lead programs (Enhertu targeting HER2 and DS1062 targeting TROP2). Despite demonstrating unprecedented efficacy in breast and lung cancer, respectively, safety profile remains challenging (especially lung toxicity). Enhertu just received FDA approval in last line HER2 breast cancer, implying the risk/benfit is still attractive (the waterfall chart below from SABCS 2019 is truly impressive).
Enhertu was the subject of a huge deal between Daiichi and AstraZeneca, which paid $1.35B upfront (out of a total of $6.9B) for co-promotion rights. It is too early to assess this deal but given the strong clinical data, the multibillion opportunity in HER2 breast cancer and the potential to expand to other HER2+ tumors (gastric cancer), this deal looks like a calculated risk from Astra’s point of view. This is in contrast to their 2015 decision to buy 55% of Acerta for $4B. Enhertu’s clinical profile and mechanism are radically different from approved HER2 agents, as can be seen in the cross-trial comparison below from Daiichi’s recent R&D event.
The same could not be said about Acerta’s Btk inhibitor (Calquence), which was another covalent inhibitor like Imbruvica with slightly better selectivity and overlapping resistance mechanisms. Reversible Btk inhibitors like ARQ531 and LOXO305 are mechanistically different from Imbruvica, which is why they (in contrast to Calquence) actually work in Imbruvica failures.
The Astrazeneca/Daiichi deal also compares favorably to the Abbvie/Stemcentrx acquisition, where Abbvie bought an ADC with dubious efficacy (22% response rate, median PFS of 3.8 months) and a problematic safety profile for $5.8B.
Beyond Enhertu, I am still excited about Daiichi’s TROP2 program, DS-1062. Despite seeing significant lung toxicity, it still looks like a viable product with potential utility in multiple solid tumors. A recent update from the company continues to show clear efficacy. Although the spider plot chart at the 8 mg/kg cohort is not as clean and striking as the initial one from WCLC2019, most patients experience tumor shrinkage already at the first scan so a 50% response rate is still realistic.
Other ADC programs to watch are Zymeworks’ (ZYME) bispecific HER2 ADC and Sutro’s (STRO) FRa ADC, both expect to have readouts in 2020. I like both programs as they utilize new technologies that might solve issues with current ADC technologies.
Zymeworks’ ZW49 binds two different sites on HER2 which may make it more potent to target cells, potentially expanding the therapeutic window. Sutro is using site specific conjugation to generate a more homogeneous ADC product (so far, approved ADCs are a mixture of several entities with different drug-antibody ratios). The company’s CD74 program (STRO-001) appears to have a narrow therapeutic window, which may be related to the payload it employs. The FRa ADC (STRO-002) utilizes a different payload and appears much better tolerated but efficacy is still too early to assess.
Portfolio holdings – Dec 22, 2019