September 08, 2019 - 1:42pm EST by
2019 2020
Price: 87.27 EPS 0 0
Shares Out. (in M): 39 P/E 0 0
Market Cap (in $M): 3,429 P/FCF 0 0
Net Debt (in $M): -485 EBIT 0 0
TEV ($): 2,944 TEV/EBIT 0 0
Borrow Cost: General Collateral

Sign up for free guest access to view investment idea with a 45 days delay.



We recommend shorting Mirati Theraputics (NASDAQ:MRTX), which has a market cap of $3.4bn and $485m. Mirati had a huge run up since ASCO in June 2019 when the market got excited over Amgen’s release of an early cut of phase 1 data with AMG 510. Mirati’s share price jumped from $70 to $100+ in June on the hype adding around $1bn to the market cap. The bump allowed Mirati to raise $204m in a secondary priced at $97 per share.
Mirati’s KRAS product candidate MRTX849 is an oral KRAS G12C inhibitor similar to Amgen’s AMG510 and given lofty expectations, we expect results to disappoint. At $87 we think that MRTX shares are still a good short and should sell off based on the Amgen data and on the release of clinical data in Q4 2019.
Mirati is a pre-revenue clinical stage oncology company, whose clinical and pre-clinical programs consist of two main product candidates: itravatinib, a multi-kinase inhibitor and MRTX849, and an oral KRAS G12C inhibitor. Similar to AMG510, MRTX 849 is an orally available small molecule designed to target the KRAS G12C mutated protein. Mirati received an FDA approval of its IND for MRTX849 in November 2018 and in January 2019 started dosing the first patient the first patient in the dose escalation phase of a phase 1/2 clinical trial. Mirati's Phase 1/2 clinical trial will evaluate MRTX849 as a single agent in patients with advanced solid tumors that harbor KRAS G12C mutations. A dose expansion phase is planned to follow the selection of a recommended Phase 2 dose. Mirati plans to release a clinical data update in Q4 2019. 
The KRAS G12C market opportunity is estimated to be $7 bn in the US and the EU ($4.9 bn for NSCLC, $2.2bn for CRC, and $400m for Pancreatic). Following the release of the Amgen data in June, SunTrust raised the probability adjusted peak sales for MRTX849 from $1.3bn to $1.65bn with a probability of success of 52%. The peak sales estimate would assume that MRTX849takes almost 30% of the market, which is lofty given the number of competitors in the space.
Amgen Update
Over the weekend, Amgen released underwhelming data on the ongoing Phase 1 study evaluating AMG 510 in patients with previously treated KRAS G12C mutated solid tumors. AMG510 is an investigational oral therapy designed to target the KRAS G12C protein.
Amgen’s Phase 1 study for AMG510 enrolled 35 patients with various tumor types (14 non-small cell lung cancer [NSCLC], 19 colorectal cancer [CRC] and two other). The primary treated tumor types were non-small cell lung cancer (NSCLC) and colorectal cancer (CRC), which have the highest incidence of KRAS G12C mutations. KRASG12C accounts for approximately 13% of NSCLC and 3-%-5% of CRC cases.Unlike patients with other mutations, such as HER2 or BRAF, patients with KRAS-mutated cancers still don’t have a treatment tailored to their mutation, largely due to the lack of obvious binding sites on the KRAS proteins. Patients in the trial were enrolled in four dose cohorts - 180 mg, 360 mg, 720 mg and 960 mg, taken orally once a day.
The initial headline findings presented at ASCO back in June (https://www.amgen.com/media/news-releases/2019/06/amgen-announces-first-clinical-dataevaluating-novel-investigational-krasg12c-inhibitor-amg-510-at-asco-2019/) showed that in 10 evaluable NSCLC patients, 5 experienced a partial response (PR), and another four had stable
disease (SD), for an objective response rate (OCR) of 90 percent (9/10). In other words, the drug shrank tumors in 50% (5/10) of the NSCLC patients and stopped tumor growth in 40% (4/10). All of the patients given the 960 mg dose (3/3) had a partial response or a 100% OCR for this subset.
The additional follow-up in a larger group of patients being presented at WCLC next week shows that thirteen of the evaluable NSCLC patients received the target dose of 960 mg (10 additional patients since ASCO), of which 54% (7/13) achieved a partial response and 46% (6/13) achieved stable disease.
However, compared to the earlier June results we see for this larger sample of patients a (1) significant drop in the objective response rate for patients given the highest 960mg dose to 54% (7/13) from 100% (3/3) at ASCO in June and only 40% (4/10) for the new patients treated since ASCO with the 960mg dose and (2) a high relapse rate for patients with partial responses (PR) with 3 patients relapsing in a relatively short time periods (blue and pink dots on the graph).