Dive Brief:

• FibroGen on Tuesday said study results the company has touted to support the safety of its anemia pill roxadustat weren’t accurate in a surprise announcement that raised further doubts about the future of the drug and damaged the biotech’s credibility.

• For two years, Fibrogen has said roxadustat was as safe, and in some cases safer, than a standard injectable medicine used as a comparator in clinical studies of certain people with chronic kidney disease. But senior executives claimed they recently learned those results included “post hoc changes” that made them appear better than they would have under the original analysis plan, which Fibrogen disclosed for the first time Tuesday.

• On a conference call, executives stressed the newly disclosed data still show roxadustat is as safe as the injectable drug or a placebo on measures of heart safety. Nonetheless, the news is the latest twist in a series of setbacks the drug has faced and sent FibroGen shares down by nearly a third in pre-market trading.

Dive Insight:

The prospects for anemia pills, a new class of medicines meant to be safer and more convenient than the injectable drugs used to treat chronic kidney disease, have taken a dramatic turn for the worse over the past year.

The drugs work by tricking the body into responding as if it’s in a low-oxygen environment, stimulating the production of blood cells and boosting hemoglobin levels in people with anemia. Two leading candidates from FibroGen and Akebia Therapeutics have steadily accrued data over the years showing they can match drugs like Amgen’s Aranesp and Epogen, which remain lucrative and widely used medicines despite links to potential heart problems.

FibroGen’s roxadustat is already cleared for use in Japan and China, and analysts had expected the two drugs to be headed in the same direction in the U.S.

But both now have significant problems to overcome. Akebia reported in September that its drug, vadadustat, appeared worse than Aransep on a measure of heart safety in one of its Phase 3 trials. The company has since filed for U.S. approval, pointing to the full data accrued in multiple studies, but some analysts are skeptical whether the Food and Drug Administration will agree.

Similar uncertainty now surrounds FibroGen. Roxadustat appeared headed for an OK late last year, only for the FDA to delay its decision from December to March. Then, early last month, FibroGen executives expressed surprise as they revealed the agency plans to convene an advisory committee to review the drug, further forestalling the decision.

Tuesday’s stunning revelation compounds the company’s issues. FibroGen admitted pooled safety results it’s presented at medical meetings and in press releases since 2019 were misleading. They included post-hoc changes to patient “stratification factors” — variables like race or body mass — that artificially inflated the drug’s performance on a composite measure of heart safety.

FibroGen has used those inflated result to claim roxadustat was as good as a placebo or an injectable drug — and, in one particular patient subgroup, better — at reducing the risk of heart problems like strokes. But the numbers in a pre-specified statistical analysis plan agreed on by the FDA, released Tuesday, were less impressive in each patient group.

On a conference call, neither FibroGen CEO Enrique Conterno nor chief medical officer Marker Eisner explained why or when the changes were made, or who was responsible.

“We still want to understand how this happened,” Conterno said, adding the company has started a “comprehensive internal review to ensure this doesn’t occur in the future.”

Conterno said FibroGen executives recently became aware of the two data sets while preparing for the advisory meeting and that the data submitted to the FDA included both sets of results. The company met with the agency last week to ensure the difference between the two analyses was clear. Conterno claimed there was “no change” in the underlying efficacy data and stressed confidence in its safety.

“I’m not going to let this overshadow the incredible work that we have put into this treatment,” he said.

Nonetheless, FibroGen can no longer claim roxadustat reduces the risk of cardiovascular events compared to an injectable drug. And the incident could diminish not only the drug’s approval prospects, but its outlook even if cleared, analysts suggested.

Geoffrey Porges, an analyst at SVB Leerink, noted that the various delays and missteps will leave investors questioning whether the drug “has any chance” of approval. If the statistical analysis is new to the FDA, he wrote, the agency may call for a full audit of FibroGen’s results, which could take months.

Meanwhile, Jefferies analyst Peter Welford wrote that roxadustat’s risk-benefit profile “now seems less favorable” and its approval case is more about convenience than safety.

The incident may also damage the drug’s credibility with kidney disease specialists, according to Christopher Raymond, of Piper Sandler. The post-hoc results were “presented or published in multiple forums, and it’s our sense that a walking back or correction of this data is not going to reflect well on the drug in the eyes of the nephrology community,” Raymond wrote.

An advisory committee meeting has been tentatively scheduled for July 15. FibroGen shares rights to the drug with AstraZeneca.