Gilead’s first big Nash bash ends in the trash

Nash’s first big readout has fallen flat with the failure of Gilead’s Ask1 inhibitor selonsertib in its initial pivotal trial. But any fears that the flop could burst the Nash bubble appear to be unfounded: companies such as Madrigal and Viking, which are testing candidates with different mechanisms, saw their stock rise today.

Gilead itself sank 4% this morning, despite widespread scepticism about selonsertib’s chances before the Stellar-4 readout. A second phase III study of the project is set to report soon, but this now looks like an even longer shot.

Less severe

Stellar-4 tested selonsertib in subjects with F4 fibrosis, representing the most severe Nash population short of those with liver failure. Meanwhile, the upcoming Stellar-3 trial looks at selonsertib in patients with less severe F3 disease. Fibrosis, which is traditionally measured using biopsy, has five stages, with a score of 0 representing no scarring and 4 representing cirrhosis.

Nash (nonalcoholic steatohepatitis) is an advanced form of fatty liver disease. In both disorders fat accumulates in the liver, but in Nash there is also inflammation and liver cell damage.

It is this scarring, or fibrosis, that Gilead is attempting to reverse with selonsertib; other Nash candidates take a different approach, for example by clearing bile acid or acting on metabolic targets.

Stellar-3 and 4 are evaluating the proportion of patients achieving at least a one-point improvement in fibrosis score, without worsening of Nash, at 48 weeks. The other co-primary endpoint in both trials is event-free survival at week 240.

Stellar-4 found that 14.4% of patients treated with selonsertib at 18mg (p=0.56 versus placebo) and 12.5% treated at the lower 6mg dose (p=1.00) achieved at least a one-stage improvement in fibrosis, without worsening of Nash, compared with 12.8% of placebo recipients.

Data from Stellar-3 are due in the second quarter, but the latest results do not bode well unless there is a hope that less severe fibrosis could be more easily halted or reversed.

The evidence so far does not provide much support to this hypothesis. Selonsertib’s phase II Nash trial enrolled patients with moderate to severe fibrosis, stages F2 and F3, but produced mixed results: there appeared to be a benefit in terms of fibrosis scores, but data on fibrosis biomarkers were unconvincing.

Even if Stellar-3 does hit, it is unlikely that Gilead would be able to file for approval with just one positive pivotal trial.

Fibrosis failings

Targeting fibrosis has proven difficult in other diseases in the past, and Nash appears to be no exception. Thus selonsertib’s failure could be bad news for other companies looking at Nash candidates designed to tackle liver scarring. 

There do not appear to be any other Ask1 inhibitors in clinical development, according to EvaluatePharma, but Genkyotex has previously described its NOX 1 & 4 inhibitor GKT831 as analogous to selonsertib (Event – Genkyotex gets a second chance, October 25, 2018). Genkyotex was down 3% today.

Gilead’s recent deal with Scholar Rock, to develop TGF-beta antibodies for fibrotic diseases, could also be on shaky ground, although the groups did not say whether they were planning to look at Nash.

And selonsertib’s setback could bode ill for the next Nash candidate to report pivotal data, Intercept’s Ocaliva. That candidate, an FXR agonist, is designed to regulate bile acid production, but Intercept has included a fibrosis co-primary endpoint in its phase III trial.

Still, that study, Regenerate, only needs to hit one of its co-primaries to be considered a success. Interim data are due later this quarter (Intercept’s big Nash day draws near, February 11, 2019). As for Gilead, its attempts to become a player in Nash have now fallen short three times.

Gilead’s best hope for selonsertib now seems to be as part of a combination with its other Nash assets. The company is already testing the Ask1 inhibitor alongside its FXR agonist GS-9674 and ACC inhibitor GS-0976 in the Atlas trial, with data due in the fourth quarter.

Still, at last year's EASL meeting, Gilead presented "lacklustre" non-invasive data from a 12-week study of the same triple combo, Evercore ISI analysts noted.

Selonsertib’s failure is yet another example of Gilead’s poor R&D track record outside its core areas of HIV and hepatitis C, said Leerink analysts. They also questioned the wisdom of moving selonsertib into phase III based on “murky” early-stage data. 

Other groups in Nash will hope that Gilead’s failures in the disorder are down to choosing the wrong candidates or targets, rather than a reflection of a poorly understood disease. For now, there is still plenty of enthusiasm around Nash, although this might change with a few more big failures.