Just when the liver disease Nash scored its third failure in short succession, that of Galectin’s GR-MD-02 yesterday, Madrigal today delivered what look to be very promising mid-stage data with the relatively low-key asset MGL-3196.
While Galectin stock closed off 31% on the disappointment, Madrigal opened up 100% today. Both companies’ projects use novel mechanisms of action, but their data highlight the lack of agreement over what the US FDA might deem to be relevant endpoints in Nash, and the fact that improvement in liver fibrosis remains a tough gold standard.
Neither Madrigal nor Galectin measured fibrosis as a primary endpoint in their respective phase II studies. Madrigal instead looked at change in liver fat, measured via imaging, which it said MGL-3196 reduced to a highly statistically significant degree versus placebo after 12 weeks in the phase II trial in question.
Further analysis was impossible since the study remains blinded. As such, investors must wait until data on Nash resolution by biopsy, and serum biomarkers of fibrosis, are assessed at the 36-week point. Recent Nash setbacks have included Allergan’s cenicriviroc and Intercept’s Ocaliva.
Meanwhile, Galectin’s Nash-CX trial was studying GR-MD-02 for treating liver fibrosis and resultant portal hypertension in patients with cirrhosis presumed to be due to Nash.
This too was not directly measuring fibrosis. Instead, its primary endpoint looked to show a reduction in hepatic venous pressure gradient (HVPG) versus placebo. There is some support for measuring HVPG, with Roth analysts stating that the US FDA had suggested this as a surrogate endpoint specifically in Nash cirrhosis.
For its part Madrigal said recently published data had shown a high correlation between liver fat reduction of 30% or more and improvement in Nash, though relevant Nash endpoints remain a contentious issue. Patients in Madrigal’s trial had to have Nash confirmed by liver biopsy – another important point.
None of this has helped Galectin, which said yesterday that, in the 161 evaluable cirrhotic subjects recruited into Nash-CX, GR-MD-02 had failed to reduce HVPG. The only crumb of comfort was that if it measured the primary efficacy measure only in the 81 subjects without oesophageal varices, an indicator of less-advanced disease, it saw a meaningful benefit.
However, even this does not stand up to scrutiny: the purported HVPG-lowering effect was seen only with the lower, 2mg dose of GR-MD-02 (nominal p=0.01), and not with 8mg. Galectin said an effect across both doses was seen on mild portal hypertension – a measure not listed in the trial’s secondary endpoints – again only in patients without oesophageal varices.
The only secondary endpoint Galectin highlighted related to liver biopsy, which it said showed a statistically significant reduction in hepatocyte ballooning. Secondary endpoints listed on Clinicaltrials.gov do mention liver biopsy, but only related to fibrosis staging; hepatocyte ballooning is one of three components of the non-alcoholic fatty liver disease activity score.
Such faint glimmers of efficacy will disappoint those who had held out hope for GR-MD-02 based on its mechanism of inhibiting galectin-3, a protein whose levels appear to be increased in patients with liver fibrosis, the main driver of cirrhosis.
Meanwhile, Madrigal claims that MGL-3196 is the first orally administered, small-molecule, liver-directed, truly β-selective thyroid hormone receptor agonist. The rationale is backed by the finding that hypothyroidism is particularly common in Nash patients, and with MGL-3196 Madrigal aims to treat the disease’s underlying cause.
Despite Galectin’s failure with GR-MD-02, management indicated on a call that pressing on into phase III was a real possibility. Specifically, the company will consider a pivotal study in Nash cirrhosis patients without varices, measuring development of varices, as well as change in HVPG, as clinical outcomes.
Anyone wishing to back such a strategy must bear in mind the lamentable track record of phase III studies designed based on purported phase II subgroup benefits, and the fact that with only $7m in the bank at the end of the third quarter Galectin will not be able to proceed without a significant cash injection.
For Madrigal, of course, the chances of a partnership just improved significantly.
|Selected trials in Nash-related cirrhosis|
|Project||Mechanism||Company||Study||Trial ID||Primary completion|
|GR-MD-02||Galectin-3 inhibitor||Galectin||Nash-CX||NCT02462967||Failed 5 Dec 2017|
|Emricasan||Caspase inhibitor||Conatus/Novartis||Encore-PH||NCT02960204||Oct 2018|
|Selonsertib||Ask-1 inhibitor||Gilead||Stellar-4||NCT03053063||Jan 2020|