Tempest Therapeutics (NASDAQ: TPST) 

May 2, 2023 

Summary: 

Tempest Therapeutics (TPST) is an under-the-radar biotech company developing small molecule inhibitors for unique oncologic targets. Their lead program, TPST-1120 (a PPAR⍺ antagonist), recently released initial data from a clinical trial that is partnered with Roche. This showed substantial superiority to the current standard of care in 1L hepatocellular carcinoma (HCC), a >$700mm market in the US by our estimates (and multiples larger abroad). TPST has the opportunity to seek partners for this program (or deepen their partnership with Roche) to expand into a larger registrational trial in HCC, along with other indications such as renal cell carcinoma and cholangiocarcinoma. Despite promising efficacy and a large commercial opportunity, the stock currently trades at a market cap of ~$30mm and an enterprise value of ~$21mm. We believe that the company is undervalued and underfollowed because it went public quietly via reverse merger in 2021, and this is the first comprehensive data update that they have provided to date. The equity is mostly held by a small group of biotech specialist funds, and it receives very little attention from other investors. We see significant asymmetry to an investment in TPST at its current valuation and look forward to updates from the company on plans for a partnership and future development of TPST-1120. Key risks to the thesis include the size/early stage of data, potential to fail to replicate efficacy, low cash/dilution risk, and illiquidity. 

Background on Tempest: 

Tempest Therapeutics was founded in 2017 and went public via reverse merger with MLND in the summer of 2021. The merger closed on June 28, 2021 and the company changed its ticker to TPST, receiving a $30mm concurrent PIPE financing. Notable participants in this financing include Versant Ventures, Rock Springs Capital, and Lilly Asia Ventures, among others. The company’s pipeline consisted of TPST-1495 (dual EP2/4 antagonist), TPST-1120 (PPAR⍺ antagonist), and some preclinical development programs. Since the listing, TPST-1120 has become the company’s lead asset and will be the focus of this note. 

In terms of management, the founding CEO, Tom Dubensky (experience at several biotechs including Immune Design), shifted into the role of President in 2021 and the President/COO Stephen Brady moved into the CEO role. Brady’s background also consists of an executive role at Immune Design (acquired by Merck for $300mm in 2019) and experience at several other biotechs. Both Dubensky and Brady have board seats, along with a representative of Versant Ventures and several other biotech executives. 

Of note in 2021 was the establishment of a partnership with Roche for TPST-1120. This is a unique partnership, in that TPST retains all development and commercial rights to the drug, but Roche is progressing it through a global randomized Ph1b/2 trial in 1L hepatocellular carcinoma. 

In April of 2022, the company announced another PIPE financing with participation from existing investor Versant Ventures and new investor EcoR1 Capital, another biotech specialist fund. The financing raised $15mm via common stock and pre-funded warrants and resulted in a ~20% ownership position by Versant and ~10% by EcoR1. 

Later in 2022, TPST released an update at ASCO, covering several of their programs. Of note for TPST-1120, the company showed PRs in several patients with pre-treated renal cell carcinoma and cholangiocarcinoma, indicating a potential new avenue of development for this drug in addition to the Roche partnership. The company also updated that the Roche partnership was continuing and to expect a readout of ORR (first) and PFS/OS (second) in 2023. 

We received the ORR update on April 28, 2023, which is the focus of this note and the driver of our bullishness on the stock. At a high level, TPST-1120 improved the response rate and survival of patients with 1L HCC relative to the current standard of care. Below we will provide more background on TPST-1120, take a closer look at the data, and evaluate potential next steps for the company.  

TPST-1120: 

TPST-1120 is a small molecule antagonist of PPAR⍺ (Peroxisome Proliferator-Activated Receptor alpha), which is a transcription factor and master regulator of fatty acid oxidation, a metabolic adaptation that supports tumor growth and metastasis. The company states that the drug has a dual mechanism of action, both killing tumor cells directly and causing immune activation in the TME by reversing immune cell exhaustion. In their preclinical work, they found that PPAR⍺ antagonism sensitizes tumor bearing animals to treatment with angiogenesis inhibitors. This creates synergies with both PD-1 and VEGF inhibitors, providing the rationale for the Roche partnership and creating important competitive implications that we will explore later. Other research has shown that HCC has been known to occur when PPAR⍺ agonists were administered for an extended period to mice or rats, further validating the mechanism (PMC3533465). According to the company’s 10-K, TPST-1120 is the first and only PPAR⍺ antagonist in development. 

At ASCO 2022, the company released a data update on TPST-1120 in combination with nivolumab (anti-PD-1). This update showed two PRs in patients with late-line RCC who had previously progressed on anti-PD-1 therapy without achieving a response. Additionally, another patient attained a PR with late-line, heavily pre-treated CCA. All three responders were treated at the two highest dose levels of TPST-1120. This update provided a first look at clinical data while waiting for a more fulsome update from the Roche partnership. 

The primary value driver for the TPST-1120 program is the partnership with Roche, who is investigating the drug in combination with atezolizumab + bevacizumab vs atezolizumab + bevacizumab (current SoC in 1L HCC) alone. Roche has led the randomized global Ph1b/2 since 2021 as a part of their MORPHEUS program, which is a series of Phase 1b/2 clinical trials designed to evaluate the safety and efficacy of combining various immunotherapies and targeted therapies in patients with advanced solid tumors. Roche dosed the first patient in September of 2021, and the recent update in April of 2023 is the first data cut we have seen from the program. 

Initial Ph1b/2 HCC Data: 

In April of 2023, TPST released initial data from this program, which showed substantial outperformance of TPST-1120 + SoC relative to SoC alone. The data is early and in a 70-patient trial, but it provides clear proof of concept and leads us to believe that TPST is significantly undervalued, as we expect that Roche (or a competitor) would like to advance TPST-1120 into a larger trial. 

Source: Company presentation 

Starting with ORR, the TPST-1120 arm substantially outperformed the SoC arm with 17.5% confirmed ORR vs 10.3%. On unconfirmed ORR, TPST-1120 delivered 30% vs 17.2% for SoC. 

Source: Company presentation 

We also saw improvements in the number of patients still on treatment in each arm (47.5% vs 23.3%), along with a mortality benefit for the TPST-1120 arm. The mortality benefit was particularly notable, with 43.5% of SoC patients deceased vs only 17.5% on the TPST-1120 arm.  

Source: Company presentation 

This mortality benefit is critical for future expectations of the final ORR number and performance on the OS and PFS endpoints. Since ORR tends to increase over time and a larger percentage of patients remained alive and on treatment in the TPST-1120 arm, we can expect that the ORR gap between these two arms will likely widen over time, resulting in an even larger treatment benefit at the next update. 

Secondly, the mortality benefit provides a very promising early indication of how these arms will compare on the survival endpoints of OS/PFS, which will read out later this year. Therefore, we expect fairly high probability that the OS/PFS update later this year will be positive. 

Understanding the SoC Arm: 

While the response rates and mortality data are both very promising, observant investors will notice that the ORR for the SoC arm is lower than what it produced in the registrational IMbrave150 trial. Below, we will explore why SoC underperformed and if there are any significant baseline imbalances between the arms of the trial. 

Source: Company presentation 

Starting with the two arms of the Ph1b/2 trial, we can see consistent baseline characteristics that may even slightly disadvantage TPST-1120. For example, higher ECOG status, higher rates of metastasis, and high baseline NLR ratio all disadvantage TPST-1120. The only baseline characteristic that advantages the SoC arm is the higher rate of alpha-feto protein. As far as region of enrollment and whether the disease was due to viral hepatitis, the data on these characteristics are mixed. An analysis of the Chinese subpopulation of the IMbrave150 trial showed generally consistent findings with the full population (PMC8339481). Similarly, as the company notes, IMbrave150 showed similar performance in viral vs non-viral disease. Therefore, the baseline characteristics look relatively balanced between these arms, and may, in fact, marginally disadvantage TPST-1120. 

The second question that we need to answer is why the SoC arm appears to be underperforming the IMbrave150 study on an ORR and survival basis. That study showed ORR for atezo + bev of 27%, and an update to the study later showed 30%. In the TPST trial, the SoC arm did 17.2% ORR at the first look. However, there are still 7 patients on treatment in this arm, and we would expect to see more responses in this group over time. The company provided the slide below, which shows the changing ORR over time in the IMbrave150 study: 

Source: Company presentation 

What this shows for the IMbrave150 trial is an ORR of 14.6% at 4 months, 24.6% at 6 months, and 31% at no landmark. The ORR increased as patients spent more time on drug. Given that Roche began dosing patients in the TPST trial in September of 2021 and the data cut was in early February 2023 (17 months running), we can assume that the median follow-up for the SoC arm is somewhere in the 8-9 month range. Therefore, it is reasonable to expect that another 1-2 responses could occur in the SoC arm, which would bring the ORR to 20-23%. This is still lower than the 27-30% that we would expect, but it’s a smaller gap. Of course, this is an estimate based on several assumptions, but it can help to explain some of the delta between the ORR numbers. 

The second metric that we can look at is the mortality rate. At 12 months, the OS number for IMbrave150 was 67.2%, indicating that ~33% of patients had died on the trial. By comparison, the TPST update showed mortality of 43% at an estimated 8-9 months. This correlates fairly closely with the ORR delta, indicating that the SoC arm did underperform expectations, but not by an enormous margin. 

So why did the SoC arm likely underperform? 

It’s difficult to say. We can look back at the baseline characteristics slide to see what the biggest differences were between the TPST trial and IMbrave150. Notably, on the TPST trial, we see younger patients, lower ECOG status, and lower metastasis. These should all improve efficacy for the SoC arm. The only characteristic that was worse for the TPST trial SoC arm was the alpha-feto protein level. It is possible that the increased rate of extrahepatic spread for patients on the IMbrave150 trial may have had an impact, as atezo + bev might work better on metastases outside of the liver (PMC9122051). This could disadvantage the SoC arm on the TPST trial, which had lower rates of metastasis relative to the IMbrave150 trial. We also don’t have data on the NLR for the IMbrave150 trial to make a comparison on this metric. Finally, the rates of viral disease (53% for SoC in TPST trial vs 70% for IMbrave150) could have something to do with it, but the analysis of IMbrave150 didn’t support this thesis. So, it’s hard to make a firm conclusion on why SoC underperformed. There may be additional characteristics that aren’t reflected in this table, or the expected ORR for patients today may simply be lower than it was in 2019 due to exogenous factors. 

As one final way to look at this data, since the SoC arm in this trial underperformed expectations, let’s compare directly to IMbrave150. To determine an expected mature ORR for TPST-1120, we can assume that ORR continues to improve over time. Based on the analysis of the timing of responses in the IMbrave150 trial provided by TPST, we could expect to increase the number of responses by ~25% (19% of responses came after >6 months). That would bring the mature ORR to 37.5%, which outperforms the 27-30% benchmark. Secondly, the IMbrave150 trial showed a 6-month and 12-month OS rate of 85% and 67%. If we interpolate these numbers to estimate the 9-month OS rate, we get to 76%. This indicates a mortality rate of 24% at 9 months. Compare this to TPST’s 17.5% mortality rate at an estimated 8-9 months of follow up, and we see superiority here too. Caveat that this is a highly speculative comparison, but it shows that, if ORR continues to increase in line with IMbrave150 rates, we should still expect superiority to the registrational trial for SoC on both ORR and mortality. 

It is important to note that durvalumab + tremelimumab was approved in October of 2022 for all patients with 1L HCC based on a median OS improvement of 2.6 months (16.4 vs 13.8) over sorafenib. The mPFS endpoint was not statistically significant and favored the sorafenib arm by 0.3 months. ORR was 20.1% in the tremelimumab + durvalumab arm vs 5.1% for sorafenib and statistically significant. This appears to put TPST-1120 in the ballpark for approval due to superiority, assuming ORR continues to mature and we see positive survival data later this year.  

This analysis does not obviate the risk that we are dealing with relatively small numbers in this trial and that we may be seeing an anomaly due to random chance as opposed to a true treatment effect. This is a risk that any investor in an early-stage biotech company must get comfortable with. But based on the data that we have seen so far, this is a drug that shows promise and should be evaluated in additional patients. In our opinion, the magnitude of these risks is significantly outweighed by the low valuation of TPST and the probability that they will find a partner to take the drug forward into a registrational trial. 

What’s Next for TPST-1120: 

With this promising initial data in-hand, TPST has several options for advancing this program. First, it is important to note that Roche does not have an option or a right of first refusal on this program. So, TPST is free to shop around for a partner, and management noted on their conference call that “multiple companies have been interested in talking with us about this program.” Company filings have also indicated that strategic opportunities are likely to occur this year. Given the highly competitive nature of the 1L HCC market and TPST-1120's synergism with PD-1 inhibitors, it makes sense that several large pharmaceutical companies with PD-1 franchises could be interested in taking TPST-1120 forward into a pivotal trial as a combination therapy to protect and grow their market share. The simplest way to progress the program would be to simply expand the protocol that Roche is using for the Ph1b/2 trial into a registrational trial.  

The company also proposed the use of beta-catenin as a biomarker that could help to improve efficacy of TPST-1120. After looking at additional data later this year, TPST could consider pursuing this subpopulation. This provides a free call option for the company to increase the magnitude of the efficacy that they have already displayed, since the recent data was in all-comers. This could open the potential for an accelerated approval pathway. 

Next, Roche may be interested in exploring TPST-1120 in adjuvant HCC. Since Roche recently presented Ph3 data for atezo +bev in this indication, they could have an interest in adding TPST-1120 to the doublet. 

And finally, RCC and CCA provide two additional indications where TPST-1120 could be effective based on the data that the company presented at ASCO in 2022 in pre-treated patients. 

Most likely, TPST is interested in partnering one or more of these indications or geographies and using this cash to finance other trials in-house where they would maintain full economics. Given HCC’s prevalence in China, pharmaceutical companies with a presence in Asia are the most logical choice for partners. 

Sizing the HCC Market: 

From the paper “Epidemiology and surveillance for hepatocellular carcinoma: New trends” published in the Journal of Hepatology in 2020, there were approximately 854,000 new liver cancer cases in 2015. HCC represents about 75-85% of these cancers. It is estimated that 72% of cases occur in Asia (more than 50% in China), 10% in Europe, 7.8% in Africa, 5.1% in North America, 4.6% Latin America and 0.5% in Oceania. Liver cirrhosis is the primary risk factor for HCC, often coinciding with HBV, HCV, NASH, and alcoholic cirrhosis. Between 1990 and 2015, liver cancer incidence increased by 75% worldwide, largely a result of etiology, population growth, and age distribution. While it’s hard to pin down the exact percentage of patients that receive systemic therapy, studies indicate that "most patients receive a diagnosis when their tumors are already too advanced for curative approaches such as surgical resection, orthotopic liver transplantation, or local percutaneous tumor ablation” (PMC6340758). Additionally, nearly 70% of patients develop recurrent HCC after resection (PMC6813818). 

For a rough estimate of potential sales in the US, we can assume 854,000 global cases of liver cancer with 80% of these HCC and 5.1% of these in the US, resulting in 34,843 cases in the US. Assuming that 60% of these cases will need systemic therapy and 50% of these receive the current recommended 1L treatment, we get 10,453 patients per year. With an expected PFS of 6.8 months from the IMbrave150 study (expected time on treatment) and an estimated price of $125k/year, we get a $740mm market opportunity in the US. Considering that the vast majority of HCC cases occur in Asia, we can expect a much larger market opportunity abroad. 

As far as other treatment options for HCC go, the primary non-pharmaceutical approaches are surgical resection, liver transplantation, and ablation. Transarterial embolization and radiotherapy is another option, along with systemic pharmacological treatment. The NCCN guidelines are shown below, with atezo + bev as the preferred 1L therapy for Child-Pugh Class A patients. Of note, durvalumab + tremelimumab was recently approved in October or 2022 for all 1L HCC patients. 

Source: NCCN 

If TPST decides to enrich for the beta-catenin biomarker, we estimate a range of around 25-30% of patients with HCC would be eligible (PMC7349517). For this opportunity, we would expect ~$200mm/year in peak sales potential in the US with multiples of this abroad. 

Lastly, the company could pursue RCC and CCA due to the promising data that they presented at ASCO 2022 and the mechanistic rationale for TPST-1120 working in these indications as well. 

Financials and Upside/Downside: 

Cash: $31.2mm as of 12/31/22 and the company burned $34.6mm in 2022, so we can assume ~$19.6mm cash today. Company has guided runway through 2Q24. $100mm ATM with Jefferies 

Debt: $10mm loan balance outstanding with Oxford Finance maturing August 1, 2025 

Market Cap: $30.3mm at $2.20/share (as of May 1 close). Includes 10.6mm shares outstanding and 3.2mm pre-funded warrants from the April 2022 PIPE 

Enterprise Value: $20.7mm 

Institutional Ownership: Versant 20%, EcoR1 10%, Rock Springs 8% 

Insider trades: None of note other than the Versant/EcoR1 PIPE from April 2022 

Short Interest: None of note other than significant day trader activity on April 28 

When considering what the upside/downside case for the stock could look like, we can provide several scenarios based on the outcome of partnership discussions and plans for future trials. 

Big upside case: Roche or another pharma company partners worldwide rights for $125mm upfront and $750mm in milestones, and TPST takes RCC and CCA alone in-house. The stock should trade to reflect the cash and NPV of the opportunity, which is likely somewhere around $250mm market cap (~$18/share). 

Moderate upside case: TPST gets some interest in a partnership, but not enough for significant negotiating leverage. They take $50mm upfront with $500mm milestones for worldwide rights, then do an offering to provide financing for RCC/CCA trials in-house. The stock trades up to $150mm market cap (~$11/share) and then trades down on a large $50mm offering to ~$7.50/share. 

Downside case: The ORR data is not convincing to potential partners, who want to wait for OS/PFS readout later this year. TPST surprisingly misses on both (we think this is low probability), and then can’t find a partner. TPST drops the program and must dilute to finance the remaining pipeline, resulting in 80% downside to a $6mm market cap. This represents the truly disastrous downside scenario. 

Clearly risk/reward is skewed to the upside given both the magnitude of the upside relative to downside and the probability of these events taking place. We estimate a ~70% chance that TPST will find a partner to progress the TPST-1120 program in the near term. 

Key Risks: 

The primary risk is that the efficacy does not replicate in a larger trial. As mentioned before, the recent data update came from a Ph1b/2 trial with 70 patients. The efficacy could be skewed due to a number of high/low responders in either treatment arm. Accordingly, the company could fail to sustain their superior efficacy at the OS/PFS readout later this year (although we believe that the mortality imbalance in the initial update reduces this risk). If the company does proceed to a registrational trial, that trial could fail to achieve statistical significance.  

Secondly, potential partners may not be interested in financing a registrational trial without seeing additional data. That could push a partnership announcement to later this year when TPST would be running low on cash, impacting their negotiating position. There is also no guarantee that partners will be convinced by this data and interested in financing a larger trial.  

While the company notes that safety was similar between both arms of the trial, there is a chance that toxicity could arise over time or in a future trial that would render TPST-1120 inviable.  

And finally, as with any small cap company, there is always management/execution risk along with illiquidity in the stock. TPST has less than one year of cash remaining and could pursue a dilutive capital raise to extend their runway.