After numerous twists and turns Pharmamar will no longer have to wait for next year’s binary outcome of the pivotal Atlantis trial before filing its small-cell lung cancer project Zepsyre. Instead, based on the just-revealed result of a call with the US FDA, filing for accelerated approval will take place in the fourth quarter.
The basis for this will be remission data Zepsyre has generated in an uncontrolled basket study that had been upsized at the agency’s request, and was last updated at Asco. The move is a U-turn from earlier guidance, and looks likely to have been spurred by the recent failures of Rova-T and Opdivo in progressed SCLC.
It also means that Pharmamar might not have to run the risk of seeing Atlantis fail – a possibility, given what now look to be outdated assumptions for survival in its control cohort – thus derailing Zepsyre's path to market. Investors sent the company's stock up 30% today.
Last year, with two seemingly impressive cuts of data from the SCLC cohort of Zepsyre’s basket study already in hand, the company was still saying that results of Atlantis would be needed for filing. Atlantis tests Zepsyre plus doxorubicin versus topotecan or the CAV chemo regimen, and should yield overall survival data early next year.
However, at a meeting in December the FDA apparently warmed to the early submission idea, pending some additional details from the basket trial. And on August 7 Pharmamar had a call with the agency, the result of which it told Vantage was “unequivocal”, leading to today’s statement that accelerated approval would be pursued.
Anyone wondering why the FDA might have changed its tune should look at the competitor pipeline in advanced SCLC.
Though progress has been made first-line, with Roche’s Tecentriq securing approval and Astrazeneca’s Imfinzi reportedly scoring a win in the Caspian trial, others have been less lucky. Last October Bristol-Myers Squibb’s Opdivo failed the Checkmate-331 study, and two months later Abbvie’s Rova-T flunked the Tahoe trial, both in the second-line setting.
As it stands, Zepsyre is the only other industry asset in a pivotal SCLC study apart from Unituxin (dinutuximab), the low-key project being developed by United Therapeutics, and Merck & Co’s Keytruda, whose front-line Keynote-604 trial reads out late this year.
|Basket trial||105-pt 2nd-line SCLC cohort: single-agent ORR 35.2%, mOS 9.3mth||Basis of US accelerated approval filing||NCT02454972|
|Atlantis||613 2nd-line SCLC pts, stratified by PD-L1 status: doxo combo vs topotecan or CAV, mOS primary endpoint, data early 2020||Confirmatory for US approval||NCT02566993|
This is quite the turnaround for Pharmamar, which had initially not expected Zepsyre to work as monotherapy. But at Asco 2018 it reported a 39% response rate, and 11.8 months of median overall survival, in the basket trial’s then 61-patient SCLC cohort.
The SCLC arm was expanded, and two months ago Pharmamar revealed the latest cut, in 105 subjects, showing 35.2% ORR and mOS of 9.3 months. It also highlighted the result in resistant patients, who effectively have no options, where survival was 5 months, though overall the slippage in OS disappointed and the stock sold off.
This is more or less the dataset that Pharmamar will now file. The FDA asked for clarification, including assessment by central review, but Vantage understands that while this lowered the ORR by a few points there was apparently no material difference.
Not having to rely for approval on Atlantis, now over a year since the last patient was dosed, avoids another potential stumbling block: the placebo effect. The basket trial’s OS result looks competitive against topotecan’s SCLC label, which cites 5.8 months’ OS, but recent studies have put the chemo’s actual benefit at closer to 10 months.
Indeed, it is this that has been blamed for the demise of Checkmate-331, where the topotecan control arm scored 8.4 months’ survival. Atlantis data will still be needed to formalise Zepsyre’s potential accelerated approval, but the pressure is no longer on it to generate a knockout result.