Adaptimmune puts fludarabine back on the agenda

If Adaptimmune’s only problem was the lack of responses in the first six patients enrolled into its ovarian cancer study of NY-ESO Spear then its 8% slide yesterday could easily be called an overreaction.

However, the initial data hint at a repeat of Juno’s troubles earlier this year with the use of fludarabine preconditioning, and raise the key question as to whether a therapeutic window exists. This is not the first setback for Adaptimmune’s engineered T-cell receptor therapeutics, and with this approach moving at snail’s pace investors have chosen to take a step back.

One issue is that the market already priced in a lot for Adaptimmune, which within the past year had been valued at nearly $1bn despite only having one engineered TCR asset in early clinical trials. In the meantime, two clinical holds, slow patient recruitment and manufacturing changes have added to existing doubts about the complexity of the TCR approach.

No response

In fact yesterday’s announcement on NY-ESO Spear in ovarian cancer was nothing new: the data had actually been presented at Asco in May, showing no objective clinical responses in six ovarian cancer and four melanoma patients.

These cohorts had had cyclophosphamide-only preconditioning, and Leerink analysts subsequently suggested that fludarabine needed to be added to enable the patients to be lymphodepleted sufficiently to allow strong cell expansion. This is precisely what Adaptimmune’s protocol amendment confirmed yesterday.

The use of fludarabine in chemo preconditioning is a live issue, since in July a trial of Juno’s CAR-T project JCAR015 was suspended owing to patient deaths. The company put this down to the Cy/Flu regimen, and the hold was lifted when it proposed preconditioning patients only with cyclophosphamide (Neurotoxicity only short-term toxic to Juno stock, July 13, 2016).

But in a separate trial CAR-T efficacy was clearly superior in patients given Cy/Flu versus non-Cy/Flu preconditioning – something that Adaptimmune’s experience seems to support. The big questions are whether Adaptimmune can give sufficient fludarabine to allow strong cell expansion while not also causing severe toxicity, and whether Juno can obtain sufficient efficacy without fludarabine.

To be fair, the problem is more complex than fludarabine just causing toxicity; indeed, plenty of groups working on adoptive cell therapy use Cy/Flu as standard. But Adaptimmune has already seen several setbacks.

A pivotal trial of NY-ESO Spear in myxoid round cell liposarcoma was recently put on hold after a US FDA request on chemistry and manufacturing controls. A year ago a phase II study in gastro-oesophageal cancer was halted for four months after a patient death. And much earlier two deaths with a TCR against Mage-A3 caused a complete redesign of that asset.

In addition to preconditioning, concomitant blocking of checkpoints like PD-1 might be necessary to increase the efficacy of cell-based approaches, and one idea is to edit out this protein on the T cells. Adaptimmune plans the simpler approach of a combination with an anti-PD-1 MAb.

Three clinical assets

Adaptimmune now also has a third project in the clinic: a TCR against Mage-A10. It touts the asset’s key strength as its ability to generate high-affinity TCRs – putting it at odds with groups that use natural TCRs, which Adaptimmune claims have insufficient affinity or hit antigens that fail to be presented in vivo.

While many groups cite the potential for TCRs to target far more antigens than CARs, the approach is limited by the need to match each project to a patient's HLA haplotype – a concept somewhat analogous to matching patients' blood groups in blood transfusions.

This and other difficulties have resulted in just a few groups dabbling in TCR therapeutics, including Juno, Kite, Bellicum, and Catapult Therapy TCR, as well as Adaptimmune. Medigene and Bluebird Bio last month struck an alliance focusing on TCRs.

After the initial wave of CD19-targeting CAR-T projects, follow-up CARs have moved slowly through the pipeline; it is now clear that investors need to be patient with engineered TCRs, too.

To contact the writer of this story email Jacob Plieth in London at [email protected] or follow JacobPlieth on Twitter