Madrigal: Buy And Hoard Before Q4 Data Readout (NASDAQ:MDGL)

Madrigal (NASDAQ:MDGL) is down again substantially, 30% from my original buy price nearly 3 years ago, and I have previously used such drops to add to my position and average down. So should I do that now – that is the question.

Before we go into that, let me first briefly discuss the trial plan here. The phase 3 trial is called MAESTRO, and it has 3 components. The first is MAESTRO-NAFLD-1, which is a trial of 1200+ patients with presumed NASH. The trial primarily measures safety and tolerability of resmetirom. Topline data was announced in January. I have covered this before. From my notes at that time:

Strong data in “presumed NASH” patients in ph 3 trial, good safety, which was the primary endpoint in this placebo controlled NAFLD trial. Met secondary endpoints in participants with presumed NASH, which were reductions in liver fat and in atherogenic lipids, such as LDL-C which can clog arteries….Liver fat, as measured using magnetic resonance imaging, fell 48% after 16 weeks of treatment with the study drug, compared to a 6% fall in the placebo group. By Week 52, liver fat in the high-dose resmetirom cohort was still down 48%, while it was up 8% in the placebo group…. What I absolutely like about this data is the outstanding safety.

Readers of this article pointed out two red flags to the safety data:

1) For all the other safety measurements, the results for 80 mg, 100 mg and placebo are separated, except for the “AE discontinuations from study”, they are all combined, n=21; (2.17%). So we don’t know the source of the 21 discontinuations. If they are all or mostly from the high dose, “Houston, we’ve a problem!”

2) For the AEs >10%, we see that the diarrhea in the treatment arms are from 23% to 31% for the 80 mg and 100 mg, respectively. The diarrhea caused by the 100 mg more than double that for the placebo.

Given that this drug needs to be taken on a daily basis for years and years, the fact that almost one third of the patients have diarrhea is not fun.

(thanks to NDHT for another outstanding comment)

Both points are important, and we need more granularity in the data to understand their implications. Here’s the EASL 2022 data for your reference:

At baseline the DB safety population (n = 969) was age 55.9 (11.8); female, 54.4%, white 88.6%; hispanic 34.7%; BMI 35.3 (6.0) type 2 diabetes 49%, hypertension 76.1%, dyslipidemia 87.9%; FS 7.4 (4.7) kPa. Discontinuations (22.5%) did not differ by treatment, most patient decision (pandemic related). DB compliance was impacted by COVID drug kit delays. AE withdrawals were 80 mg, 2.4%; 100 mg, 2.8%; PBO, 1.3%. The primary objective was met. TEAEs were 80 mg, 88.4%; 100 mg, 86.1%; PBO, 81.8%. TEAEs ≥grade 3 severity were 80 mg, 7.6%; 100 mg, 9.0%; PBO, 9.1%. AEs in excess of PBO were grade 1-2 AEs of diarrhea (80 mg, 23.5%; 100 mg, 31.2%; PBO, 13.8%) and nausea (80 mg, 11.9%; 100 mg, 18.2%; PBO, 7.9%), in the first few weeks. ALT increases ≥3XULN were 80 mg, 0.61%; 100 mg, 0.31%; PBO, 1.6%. There were no changes in body weight or HR. BP decreased by 2-3 mmHg in the RES arms. Key 2o end points were met (Table). Comparative mean reduction in FS VCTE was not significant; a responder analysis of FS and MRE showed significant reductions with RES treatment.

Two points are important – “TEAEs ≥grade 3 severity were 80 mg, 7.6%; 100 mg, 9.0%; PBO, 9.1%,” and AE diarrhea figures “in the first few weeks.” This sort of tells me that the diarrhea occurred only in the first few weeks (the company stated – “Further analysis has shown that these increases in GI adverse events, diarrhea and nausea, did not occur after the first few weeks of therapy.”) and was resolved, and did not lead to higher grade treatment emergent adverse events. Indeed, such TEAEs were quite similar across all three arms.

The second trial is MAESTRO-NASH, which is the most important trial of the three. The design is interesting. There are two endpoints, and either one can be achieved for a successful trial outcome:

The dual primary surrogate endpoints on biopsy are NASH resolution, with at least a 2-point reduction in NAS (NASH Activity Score), and with no worsening of fibrosis OR a one point decrease in fibrosis with no worsening of NASH. Either primary endpoint can be achieved for a successful trial outcome.

Here I note recent comments from an analyst:

• B Riley analyst points to the study’s dual primary endpoints: NASH resolution and fibrosis improvement. While the former could lead to a favorable outcome, the analyst is not so convinced about the likely effect size on fibrosis improvement.

This is a faulty understanding of the situation. First, Resmetirom does not need to meet both primary endpoints. Either will do. Second, early data is predictive of success with both endpoints. Third, given that if approved, resmetirom will be the first approved NASH therapy of any consequence, and company sponsored studies that have shown the interest of the medical community and willingness to prescribe, I think approval with meeting any one of the two goals will not obstruct commercial success. As the company noted in their previous earnings call:

As we’ve gained increasing confidence that we’ll achieve both NASH resolution and the fibrosis improvement endpoints in the MAESTRO-NASH biopsy study, we are moving 1-point fibrosis reduction up the hierarchy to a primary endpoint along with NASH resolution.

The trial will have around 2000 patients with NASH with significant fibrosis. The Subpart H population, i.e. the study population that will produce biopsy data for an accelerated approval application, is ~900. This is the data that will come out in Q4 2022 – and constitutes the most important catalyst for Madrigal. The dual primary endpoints are constituted in such a way that for the trial to be successful, either resmetirom resolves NASH or it helps with fibrosis. The NAFLD trial and an earlier phase 2 trial has already shown early indications that resmetirom can do both. For example, in an open label extension part of the NAFLD study, Resmetirom improved magnetic resonance elastography (MRE) and FibroScan vibration controlled transient elastography (VCTE) at week 52 (MRE and FibroScan are measures of liver stiffness, a surrogate for liver fibrosis). In the earlier phase 2 trial as well, we saw the following:

▪ Resmetirom achieved reductions in liver fat that we believe are predictive of NASH resolution and fibrosis improvement on biopsy in the Phase 3 MAESTRO-NASH trial

▪ Resmetirom demonstrated fibrosis improvement in an exploratory analysis of biopsies by second harmonic generation, an automated, fully quantitative assessment of fibrosis on liver biopsy slides based on unique architectural features of collagen.

This is the reason we are optimistic.

There’s an outcomes portion to this study, and a separate outcomes study as well. These are not important right now, but will eventually become very much so at the time for a full approval.

Madrigal’s goal here is not to target biopsy-identified NASH patients, because in real world clinical practice, such biopsies are seldom performed. Their goal is primarily to show a relation between non-invasive tests or NITs and biopsy data as a way for patient selection, and then to use the same relation to demonstrate successful patient treatment. There are literally millions of such patients in the US alone.

Financials

MDGL has a market cap of $1.04bn and a cash balance of $220mn. They also have a $200mn ATM facility from Hercules Capital, of which $159mn is available; and a $200mn available long term debt. Their cash burn in Q1 was $49mn, which gives them a runway of 3 quarters; however, the two loan facilities together extend that to 8 more quarters, providing useful cash for commercial launch.

Bottomline

MDGL is down 30% from my original buy price. Their most important catalytic event is 2 quarters away. Earlier trial data indicates signs of success. Given all of that, MDGL is a buy.