Cargo can't replicate firi-cel's academic success

In the end neither the presence of the high-profile cell therapy scientist Dr Crystal Mackall as a co-founder nor the academic data that provided clinical backing could save Cargo Therapeutics and its lead project, firicabtagene autoleucel. As of today Cargo is cutting 50% of its staff, and firi-cel looks to be dead in the water.

The retrenchment has been prompted by the failure of Firce-1, a study of firi-cel in lymphoma patients who relapsed after standard anti-CD19 Car-T therapy. Firce-1 was billed as potentially backing an accelerated approval pathway for firi-cel, but it's now been scrapped after showing efficacy woefully below that seen in an earlier academic trial, as well as significant toxicities.

Cargo had $368m in cash at the end of 2024, and it now plans to put this towards developing its next Car-T asset, CRG-023, which hits CD19, CD20 and CD22. But the market took a dim view of the scale of the firi-cel setback, and Cargo, which floated in late 2023 at a $600m valuation, fell 75% on Thursday.

Why?

A key question is why firi-cel, an autologous anti-CD22 Car, should have performed so much worse in Cargo's phase 2 Firce-1 study than it had in a phase 1 trial conducted by Mackall's Stanford University; the latter's results were published in The Lancet last year, and underscored an investment case based on the success of Firce-1.

The Stanford study saw a 47% complete response rate at three months, and a 3% rate of grade 3 or higher immune effector cell-associated hemophagocytic lymphohistiocytosis-like syndrome (IEC-HS). Firce-1, on the other hand, saw 22 of its 51 evaluable patients develop a CR, but just four of these were still ongoing at three months, implying a three-month CR rate of only 8%.

Cargo will surely have been aware of how efficacy was progressing before now, but there was no reason to halt Firce-1 without a sudden toxicity signal. That apparently has now been seen, with Cargo revealing that 18% of Firce-1 patients developed IEC-HS at grade 3 or higher, "including grade 4 and grade 5 serious adverse events".

The company used the same Car construct as Stanford, though it seems to have modified manufacturing to reflect an intended commercial process rather than the academic one used earlier. It's possible that Stanford recruited fitter patients, but data showing the patient disposition in Firce-1 aren't yet available.

A full post mortem of Firce-1 won't be possible until the complete results are presented, which Cargo says will happen at a future medical conference.

Hero to zero: how firi-cel data deteriorated

Note: *immune effector cell-associated hemophagocytic lymphohistiocytosis-like syndrome. Source: The Lancet & company statement.

It's notable that this isn't the first time firi-cel has been discontinued. The construct is the result of work undertaken at the NCI and Stanford, and was once owned by Bristol Myers Squibb (back then it was coded JCAR018), through the legacy companies Celgene, Juno and Opus Bio.

But all that work ended at some point in the past five years, apparently as a result of an unpromising profile in paediatric ALL, and an unwillingness to test firi-cel in lymphoma. It was Cargo that seized on the promise Stanford showed in relapsed DLBCL, though in the event this too has now come to nothing.

Other companies also developing anti-CD22 Car-T therapies for lymphoma include Sana (with an allogeneic version of firi-cel), Umoja Biopharma and Essen Biotech. Expect all of these now to be casting a nervous eye over the Firce-1 data.