Cytokine interest shows no sign of waning

It might have come as a surprise to learn that Nektar and Bristol Myers Squibb’s IL-2 project bempegaldesleukin is still in play. Even more surprisingly, lacklustre results seen so far with other IL-2 assets have not put off a host of other companies from pursuing projects based on this cytokine in oncology.

These all aim to improve on Proleukin, which is approved but has well-known toxicity issues. A look at the pipeline shows various ways that groups hope to do so, and one major approach is trying to avoid the alpha unit of the IL-2 receptor.

“Not alpha”

The rationale is that the alpha chain is a component of the high-affinity IL-2 receptor, which is expressed on immune-suppressing T regulatory cells (Tregs). Meanwhile, the beta and gamma units form the intermediate-affinity receptor, which is found on cancer-killing CD8-positive T cells and natural killer cells.

Hitting IL-2 can therefore lead to both immune stimulation and suppression; indeed, various companies are testing low-dose IL-2 in autoimmune disorders.

It is thought that the efficacy of Proleukin, a recombinant human IL-2, could be limited by its stimulation of Tregs, making it hard to find a therapeutic window for the drug.

Still, the jury is still out on the “not alpha” strategy after recent disappointing updates, first with Sanofi’s Thor-707 at AACR, and then with Alkermes’s nemvaleukin alfa at Asco.

Previous data with the latter, at last year’s Esmo meeting, had raised hopes that Alkermes might have cracked the IL-2 conundrum. The group is pressing on in mucosal melanoma, an underserved niche, but ultimately wants to find a partner for nemva.

The next big event in the cytokine space might not answer the question of whether alpha is indeed the enemy. Data are due soon from the phase 2 Propel study of Nektar/Bristol’s bempegaldesleukin, which is best described as a partial alpha agonist, albeit one designed to act preferentially on the intermediate-affinity receptor (Nektar again looks to bempeg to sweeten its valuation, August 23, 2021). 

It is worth remembering that Bristol paid $1.85bn up front to license bempegaldesleukin, while Sanofi shelled out $2.5bn for Thor-707’s developer, Synthorx. It is fair to say that investor interest in IL-2 has cooled since these deals, but there are still plenty of companies trying to get in on the act, presumably in the hope that they can do better.

IL-15

One of the more advanced players in the space is Immunitybio, the company formerly known as Nantcell, whose lead project is the “not alpha” IL-15 fusion protein Anktiva. IL-15 assets are included in this analysis as this cytokine is structurally similar to IL-2, and its beta and gamma chains are identical to IL-2’s.

At the Asco-GU congress in February Immunitybio presented promising data from the phase 2/3 Quilt-3.032 trial in non-muscle invasive bladder cancer in which Anktiva was given locally with BCG. The company has said it hopes to file for accelerated approval this year.

The results with Anktiva raise the question of whether direct delivery to the tumour could improve the efficacy of IL-2/IL-15 therapies.  

Other groups are looking at different ways of homing in on tumours. But one such approach that has fallen by the wayside is Roche’s anti-FAP/IL-2 fusion protein RG7461, which was designed to concentrate IL-2 in FAP-expressing tumour tissue. That company is now focused on RG6279, an anti-PD-1/IL-2 fusion protein. A similar strategy is being pursued by Pfizer with PF-07209960 and Kadmon with KD033.

Another targeted project comes from Cue Biopharma; the group’s Cue-101 is a fusion protein designed selectively to activate tumour-specific T-cells via the E7 protein of human papillomavirus. An early-stage monotherapy trial of this asset in head and neck cancer initially yielded disappointing early data last year, but since then a partial response has been seen in the same study, and the group has selected the dose it plans to take into phase 2.

Meanwhile, Biontech is taking a different tack: the group is developing two projects using mRNA to encode IL-2. BNT151 is a modified “not alpha” IL-2 that the group plans to combine with anti-PD-(L)1 therapy, while BNT153 is a native IL-2 being trialled alongside BNT152, an IL-7 project. Both started phase 1 this year.

The table below shows that IL-2/IL-15 is being attacked from many other angles. And this analysis does not include a glut of preclinical assets, including projects from the likes of Medicenna, Ascendis and the newly public Werewolf Therapeutics that are set to hit the clinic soon.

This story has been updated to remove Xencor's XmAb564 from the table.