89Bio Enlivened by Nash success

Today’s mid-stage data on 89Bio’s nonalcoholic steatohepatitis project pegozafermin make the FGF21 analogue look highly competitive in what is finally shaping up to be a real market. On the twin endpoints of Nash resolution and fibrosis improvement the higher doses of pegozafermin beat almost all potential rivals on a cross-trial basis. 

A couple of Nash projects are closer to market, but arguably 89Bio’s real competitor is Akero, whose similarly acting efruxifermin hit in phase 2 last September. But 89Bio’s trial design and statistical analysis both appear more robust than Akero’s, and pegozafermin could have a dosing advantage to boot. 89Bio’s shares soared 30% in early trade.

The phase 2b Enliven trial tested three subcutaneous doses of pegozafermin but the lowest, 15mg once-weekly, did not achieve significance versus placebo on the fibrosis improvement co-primary endpoint. On the other endpoint, Nash resolution, the low-dose group posted better data than the others, with a 35-point advantage over placebo. The company said that the lack of dose-response here was an artefact of the low patient numbers – just 14 patients received the low dose. 

Consequently 89Bio is concentrating on the two higher doses – 30mg weekly and 44mg every fortnight – and played up a cross-trial comparison of these against other companies’ phase 2 Nash studies, as well as with phase 3 data on projects from Intercept and Madrigal.

On both endpoints Akero’s efruxifermin looks like the one to beat. But it should be noted that both these graphs use the completers population for Akero’s Harmony trial, as that company itself reported in September. A more stringent intention-to-treat analysis reduces Akero’s effect size; SVB analysts pointed out that on fibrosis improvement the higher efruxifermin dose was not statistically significant at on an ITT basis.

By contrast, 89Bio’s data on the two higher doses are identical whether a completers analysis or multiple imputation, a form of ITT analysis, is used. The p values differ slightly, but both doses remain highly significant versus placebo.

Diagnosis and dosing

89Bio also contended that the method used to diagnose Nash in Enliven was more robust than that used in Harmony, and indeed in almost all the other trials. Biopsies were scored independently by three expert blinded pathologists, something 89Bio said reduced variation between different readings and gave an “accurate representation of true placebo rates in this condition”.

Moreover, Akero’s agent is injected once a week, whereas one of the successful doses in 89Bio’s trial was given fortnightly. And this dose was actually safer than the weekly dose; it also caused lower rates of GI toxicity than efruxifermin on a cross-trial basis.

A second mid-stage trial of efruxifermin, Symmetry, is set to read out in the second half of this year.

Whether the efficacy of 89Bio’s FGF21 analogue trails Akero’s or exceeds it, the Enliven data are clearly good enough to justify heading into phase 3. Speaking to Evaluate Vantage before the Enliven readout, the group's chief medical officer, Hank Mansbach, said the next step would be a phase 3 in the non-cirrhotic population, perhaps in around 1,000 subjects. 

“If you look at the recent data from Madrigal, 966 was their sample size,” Mansbach said. 89Bio’s phase 3 will have to be sufficiently large to be able to support accelerated approval; in Mansbach’s words, it would be “a big undertaking”.

Madrigal is also gunning initially for accelerated approval.

Mansbach said 89Bio could conduct the trial itself, but “we would be open to looking at partnership opportunities, should those arise”. 

He added that a successful showing in Enliven would, when considered alongside Akero’s Harmony readout, suggest that the FGF21 analogue category is the most promising approach to Nash. “Two different molecules, two different studies – it reinforces the class as the class to be in in Nash.” Certainly the market is big enough to support both.