The year of Nash just keeps firing blanks. After the unequivocal failure of Gilead’s selonsertib and the highly debatable success of Intercept’s Ocaliva, Cymabay’s PPAR agonist seladelpar has shown itself to be numerically worse than placebo.
This shockingly bad result has an important read-across to another of 2019’s key Nash readouts, namely the phase III Resolve-It trial of Genfit’s own PPAR agonist, elafibranor, which of course had earlier failed phase II. Little wonder that Genfit lost 14% this morning; Cymabay crashed 50%, meanwhile, despite insisting that seladelpar was not dead and buried.
The setback means that, by default, Ocaliva stands as the Nash winner of the year so far. Intercept stock crept up 4%, though the company is itself flirting with the PPAR approach, having licensed Aralez’s benzafibrate in January to complement a Nash combo.
Investors nervously tracking Nash developments will, of course, point to mechanistic differences: seladelpar is a PPAR delta agonist, while Genfit’s elafibranor hits PPAR alpha and delta; bezafibrate and a fourth compound, Inventiva’s lanifibranor, are described as pan-PPAR agents.
Be that as it may, Cymabay has set a very poor precedent. The initial readout from its phase II study focused on change in liver fat, as measured by MRI-PDFF, for 10mg, 20mg and 50mg seladelpar doses versus placebo in 171 Nash subjects.
On three key measures – relative and absolute change in liver fat, and proportion of subjects with ≥30% fat reduction – all three doses did numerically worse than placebo. Against sellside expectations of a 30-45% relative fat reduction seladelpar actually achieved a puny 9.8-14.2% versus baseline, while placebo scored a 20.8% effect.
The company pointed out that there was no statistical evidence suggesting that its project actually was worse, but the take-home message is that, on liver fat at least, seladelpar is biologically inert.
On an analyst call today Cymabay bemoaned the fact that its findings in mice and human hepatocytes, backing an effect on liver fat, did not translate into a benefit in the clinic. However, in a U-turn it also moved to cast doubt on the importance of this marker in Nash, and insisted that seladelpar could still hit a fibrosis endpoint.
Importantly, the current trial will proceed to a 52-week liver biopsy readout in the first half of next year, generating fibrosis and Nash activity score data on the basis of which a phase III study might still be designed. “The fat story is not the end of the story,” the company told analysts.
To back its claim it presented exploratory cuts of the dataset, suggesting seladelpar’s effect on reducing liver enzymes; here there was an apparently strong numerical effect favouring active treatment, with a numerical dose response in the cases of ALT and GGT to boot.
Investors must therefore grapple with the following questions: did Cymabay focus on a hopeless endpoint? Are liver enzyme changes much better than fat reduction as a marker of liver injury? And is a purported effect on liver enzymes likely to translate into histological Nash improvement?
In the meantime, Genfit holders will squirm some more. They do not have long to wait now.
|Selected PPAR agonists in development|
|Elafibranor||Genfit||Alpha & delta agonist||Resolve-It, NCT02704403||Topline data due 2019|
|Seladelpar||Cymabay||Delta agonist||NCT03551522||Failed to reduce liver fat by MRI-PDFF|
|Lanifibranor||Inventiva||Pan agonist||NCT03008070||Data due H1 2020|
|Bezafibrate||Aralez/Intercept||Pan agonist||None in NASH||Intercept plans ph II trial plus Ocaliva in PBC|