AstraZeneca said it has voluntarily withdrawn the U.S. FDA’s accelerated approval for immunotherapy Imfinzi (ingredient: durvalumab) after failing to prove its efficacy in the phase-3 study.

The withdrawal was second of its kind after BMS pulled a conditional nod for Opdivo (nivolumab) in the U.S. last year.

Critics said the voluntary cancellation exposed the loophole in the FDA’s system of granting conditional approval based on insufficient phase-1/2 trial results.

On Monday, AstraZeneca announced the voluntary withdrawal of the Imfinzi indication for advanced bladder cancer that it had obtained from the FDA in accelerated approval.

The FDA granted the conditional nod for Imfinzi in May 2017, based on tumor response rates and duration of response data from the Phase-1/2 Study 1108 trials.

The Study 1108 trials tested Imfinzi’s efficacy and safety in advanced solid tumors, including previously treated bladder cancer. AstraZeneca had to confirm Imfinzi’s effect in the first-line treatment for metastatic bladder cancer in the phase-3 DANUBE study.

However, the DANUBE trial failed to meet the primary endpoints, leaving AstraZeneca with no option to maintain conditional approval for Imfinzi.

AstraZeneca’s withdrawal of conditional nod was the second after BMS canceled Opdivo indication in small-cell lung cancer in the U.S.

Earlier in 2018, BMS won FDA’s accelerated approval for Opdivo to treat small-cell lung cancer patients who have been previously treated with platinum-based chemotherapy and first-line therapy. The nod was based on surrogate endpoints of the Phase-1/2 CheckMate 032 trials in patients with advanced or metastatic solid tumors.

BMS had to confirm Opdivo’s effect in confirmatory phase-3 CheckMate 451 and CheckMate 331 trials. However, the two trials failed to improve overall survival (OS), the primary endpoint. Consequently, BMS withdrew the Opdivo indication in small-cell lung cancer voluntarily on Dec. 29 last year.

The second consecutive withdrawal hurt credibility in FDA’s conditional approval system, which is based on insufficient surrogate endpoints, observers said.

Evaluate Vantage, a news service by market research firm Evaluate Pharma, said that the FDA gave conditional approvals based on poor data more frequently than ever.

“Only 44 of the 130 accelerated approvals granted in 2010-2020 have been converted to formal green lights backed by confirmatory data,” Evaluate Vantage said.

Surprisingly, only one such approval was withdrawn during the cited period -- BMS’ Opdivo in small-cell lung cancer -- Evaluate Vantage said, criticizing the FDA’s tolerance with failures in confirmatory studies.

Immunotherapies such as Keytruda (pembrolizumab), Tecentriq (atezolizumab), Opdivo, and Imfinzi failed in various indications in confirmatory trials. Still, the FDA has rarely imposed regulatory measures or took minimal measures such as indication reduction, according to Evaluate Vantage.

Due to the failure of the Keynote 361 study of Keytruda in the first-line treatment of bladder cancer, the FDA narrowed the U.S. label for Keytruda. However, the FDA did not impose any regulatory action on confirmatory trials' failure (Keynote 240, Keynote 061~062, Keynote 604) in liver cancer, gastric adenocarcinoma, and small-cell lung cancer, Evaluate Vantage said.

The Imvigor 211 study failure on Tecentriq led to a narrowed label, but no regulatory outcome was found in the failure of Impassion 131 trial in triple-negative breast cancer, it went on to say.

The failure of the Checkmate 459 study of Opdivo in liver cancer also did not result in any regulation from the FDA, it added.