FibroGen and AstraZeneca already endured a delay in the U.S. review of their anemia candidate roxadustat. Just as the pair was expecting to snag its first-in-class approval, as the revised date draws near, a capricious FDA has stunned the companies with a second blow.
The FDA will convene an advisory committee to review roxadustat’s application for treating anemia caused by chronic kidney disease, both in patients who require dialysis and those who don’t, the two companies said Monday.
The new request came as a shock, given that roxadustat’s FDA decision date was already pushed back from December to March 20. But because the FDA hasn’t set a date for the meeting, industry watchers widely suspect that another delay could be imminent, with at least one analyst significantly lowering his expectations for an eventual approval.
By FibroGen CEO Enrique Conterno’s account, the FDA had a surprising change of mind. The FDA had told FibroGen on three separate occasions that the agency didn’t plan to hold an AdComm meeting for roxa, he said during a conference call Monday.
If approved, roxa could be the first in the oral HIF-PH inhibitor class to land a U.S. green light and challenge current stand-of-care erythropoietin injectable therapies such as Amgen and Johnson & Johnson’s Epogen/Procrit. For that reason, industry watchers had pegged the drug as a potential blockbuster. SVB Leerink analyst Geoffrey Porges had estimated it could be bringing in $3 billion to $3.5 billion by the mid-2020s. But now, with the uncertainties triggered by the AdComm, he has lowered the number to $719 million in 2025 on a risk-adjusted basis.
Conterno acknowledged that it’s not unusual for a first-in-class med to go through an AdComm; in fact, the companies were preparing for that scenario last spring, but then stopped the work after consulting with the FDA, he said. Now, with the about-face, the two are scrambling to resume the preparation.
The FDA had previously pushed back its review of roxa, asking for additional analyses of existing data sets.
There were some concerns about the drug’s heart safety profile, especially in nondialysis-dependent patients, but FibroGen and AstraZeneca largely dispelled those doubts with a pooled analysis of heart-related events in its clinical programs. Epogen and Procrit are not allowed in this patient group because of their increased cardiovascular risks. Roxa's CV profile matched up to that of a placebo in nondialysis patients and to Epogen and Procrit in the dialysis population.
But a citizen petition filed in November requested that the FDA require more cardiovascular safety data before approving roxa. The petitioners argued that the non-inferiority comparison with placebo reflected negatively on roxa, because correcting anemia should result in better heart outcomes.
In a Tuesday note, Porges suggested that the FDA might have launched the AdComm in response to the various citizens petitions, including one from Amgen. If that’s the case, then “the roxa data package stands on its own and the AdComm is not a smoking gun pointing to threatened approval,” he wrote in the note.
Porges also pointed out that such last-minute notification is “very unusual.” With no details on what exactly the FDA is concerned about, Porges has significantly lowered his confidence in the approval of roxa to 30% and 65% for nondialysis and dialysis populations, respectively, from the previous 80% to 100%. His colleague Andrew Berens lowered his estimates to 50% for both populations, saying that AstraZeneca management doesn’t expect a complete response letter, but rather a delay of the decision beyond the March 20 action date.
Mizuho analyst Difei Yang, in a Tuesday note, also predicted the drug will get its approval in both patient groups, but perhaps with a black box warning.Yang expects the expert panel meeting will happen around June or July, pushing back a decision to the third quarter and a commercial launch to the fourth. Porges and Berens had similar timelines.
Roxa was first approved in China, where it generated $29.2 million sales in the fourth quarter of 2020, up from $22.7 million in the third quarter. The drug recently earned an approval under the brand Evrenzo in Japan, where FibroGen is partnered with Astellas, and an EU decision is expected later this year.
In the HIF-PH inhibitor class, Roxa could compete with Akebia and Otsuka's vadadustat, which is approved in Japan as Vafseo. But the Street has low expectations after that drug turned up increased heart risks compared with Amgen’s erythropoietin therapy Aranesp in nondialysis patients in the phase 3 PRO2TECT trial. GlaxoSmithKline also competes in Japan with its drug Duvroq (daprodustat).