Crucial ADC readouts at WCLC 2019
The coming months will be very important for the ADC industry which has been struggling to bring forward effective treatments, especially for solid tumors. While there have been some positive news (mostly utilizing Seattle Genetics’ technology like the recent update on GSK’s BCMA program in multiple myeloma), existing platforms are still limited by side effects. In order to address this need, the industry has to find new technologies that would minimize off -target toxicities.
To me, the most intriguing ADC platform today is that of Daiichi Sankyo, which has been quietly developing its next generation ADC platform at a time when most of its peers de-prioritized their ADC programs.
Daiichi’s bet started to pay off with DS-8201, now partnered with AstraZeneca (AZN), which demonstrated unprecedented efficacy in breast cancer. The company has 6 additional ADC programs, including two in clinical testing: DS-1062 (anti-TROP2) and U3-1402 (anti-HER3). Both programs will report data at WCLC 2019 next month that, based on previous results, could demonstrate impressive efficacy.
The TROP2 program looks particularly promising based on preliminary findings presented at ASCO 2019. The study was able to reach high doses not typically achieved with ADCs (8 mg/kg) without severe toxicity issues (although side effects were clearly observed) and demonstrate very encouraging dose-dependent efficacy. If the trend continues and no safety show stoppers emerge, the 6-8 mg/kg cohorts may demonstrate a >50% response rate, which is unprecedented for an ADC in NSCLC. Response rate could actually be higher in TROP2 expressors as the trial did not include TROP2 expression as an entry criterion.
Data for DS-1062 and U3-1402 will be crucial for the entire ADC field as they could potentially validate Daiichi’s ADC platform and provide the first differentiated next-gen ADC platform in over a decade. Most importantly, they might mark the start of the next innovation cycle in the ADC space.
Positive results for DS-1062 will have a material impact on Immunomedics (IMMU), which has a TROP2 ADC. While Immunomedics’ lead indication is TNBC (triple negative breast cancer), their ADC could become irrelevant even before an expected approval next year as both ADCs use Topo-1 inhibitors as payloads. If DS-1062’s data are as good as expected, it will be very hard to justify Immunomedics’ valuation of ~$2.5B. Therefore, I would stay out of the stock going into WCLC 2019 given the asymmetric risk/reward profile.
AVROBIO – Back in the game
I was previously very skeptic about AVROBIO’s (AVRO) Fabry program (AVR-RD-01) given the decline in enzyme (AGA) levels and VCN seen 3-6 months post treatment. Based on the trajectory at the time I thought transduced cells will eventually vanish after ~2 years which could make AVR-RD-01 irrelevant.
I am now much more optimistic following last month’s update that demonstrated stabilization of the slope of AGA levels and VCN in two patients 18 and 28 months post treatment, respectively (three additional patients did not have long enough follow up).
While more data are needed (# of patients and longer follow up) the new data suggests AVR-RD-01 may have a long lasting “tail” of enzyme activity. Enzyme levels generated by AVR-RD-01 appear clinically meaningful based on substrate reduction to levels below what had been achieved with ERT in these patients. Importantly, patient #1 is ERT free for 10 months with no significant changes in substrate levels, which unequivocally demonstrates the treatment’s activity with ERT’s “masking” effect.
Substrate levels are considered a good surrogate for long term tissue damage (mainly kidney and heart) but the clinical impact will have to be evaluated in the long run. A read-across from AVROBIO’s P2 in ERT-naïve patients further supports the notion that levels achieved in P1 will lead to clinical improvement.
The company presented biopsy data from an ERT-naive patient that had a dramatic reduction in substrate in the kidney, which is the approvable endpoint for Fabry. Enzyme levels in this patient were similar to long term expression levels seen in patients in P1. In theory, long term clinical outcomes should be better with AVR-RD-01 vs. ERT as the prior produces constant, stable enzyme levels vs. the typical peak/trough profile with ERT.
Things look much better for AVROBIO today but long term durability will remain an overhang for a long time as the company aspires to provide life-long treatments. Risk is still high but with the new data demonstrating stabilization or near-stabilization of enzyme levels in P1 coupled with kidney biopsy data in P2 and experience with other bone marrow transplant technologies, the company is attractive at the current valuation.
AVR-RD-01 is competing with AAV-based treatments for Fabry from Amicus (FOLD), Abeona (ABEO) and Freeline. These treatments from are more straightforward to administer as a “drug in a vial” but their distribution profile and durability are still unclear.
Chiasma – Positive P3 but market unimpressed
Chiasma (CHMA) reported positive P3 data for its oral octreotide program, Mycapssa, in acromegaly. Despite the clear efficacy across every endpoint, the stock price is now lower than the price prior to the announcement.
Overall, results look very good to me with a clear superiority over placebo, which they demonstrate for the first time (previous P3 was a single arm study). The study, which involved switching stabilized patients from an injectable to Mycapssa, clearly demonstrated that in a significant portion of patients the disease can be controlled with Mycapssa: 58% of patients in the Mycapssa maintained an IGF-1 response (≤ 1.0 x ULN) vs. 19% in the placebo arm.
The drug’s eventual market penetration is a matter of debate but given the pain and inconvenience associated with injectables, Mycapssa should be able to capture a significant market share. It is easy to understand why patients and physicians will prefer to start with oral and only if patients are not controlled, switch or add an injectable drug. Initial uptake is hard to predict but the eventual conversion to an oral alternative is highly likely.
Looking at rescue treatment and drop off rates during the study provides more color Mycapssa’s use in real life. 25% of patients in the active arm received rescue treatments vs. 68% in the control arm. Importantly 90% of patients who completed 36 weeks on Mycapssa opted to stay on the drug as part of an open label extension study.
The lackluster stock reaction can be attributed to future competition from Crinetics (CRNX), which has an oral agent (CRN00808) with encouraging P1 data including good PD modulation. As a highly selective and potent small molecule dosed once daily, CRN00808 represents a major threat to Chiasma but it is still in P2 and has no data in patients. One important advantage Chiasma has is the fact it utilizes a known and approved product with years of clinical experience. As a novel small molecule, Crinetics faces off target risks that will have to be assessed in long term studies.
In summary, Chiasma is now a de-risked story with a decent chance of capturing 25-50% of the acromegaly market. The indication represents a ~$900M global opportunity so even a 20% chunk that translates to $180M in annual sales makes Chiasma a good risk/reward bet with a market cap of $216M.
I am starting a new position in AVROBIO based on the recent update in Fabry. I am also adding follow-on positions in Madrigal (MDGL) and Viking Therapeutics (VKTX), and selling Nightstar (NITE) following the Biogen (BIIB) acquisition.
Portfolio holdings – Aug 25, 2019