Cell therapy is of course a very hot area, and has been for some years now. There’s a lot going on in that area, and I wanted to mention some early data from Caribou Biosciences, a gene-editing company founded by CRISPR pioneer Jennifer Doudna and colleagues. One of their key technologies is a new guide RNA method for CRISPR, which is in fact a mixture of RNA and DNA that they call chRDNA. It seems to have increased specificity in editing the genomes of primary human T cells, and doing that is of course one of the big areas of cell therapy in oncology.
Caribou has been working on a therapy for B-cell non-Hodgkin’s lymphoma (B-NHL), targeting patients who have gone through several therapies already and are now in need of something new. These relapsed/refractory patients are a feature of drug trials for new mechanisms in oncology, and those trials are swords with two very sharp and well-defined edges. On the one hand, this is obviously a tough patient population to treat. They have advanced disease, with tumor cells that have already been through the obvious forms of therapy and have mutated out around them, and not to beat around the bush, these are people who are certain to die from their cancer (and in the not too distant future at that) unless something new is available. So there can be a high bar to getting something to work in this population. But the flip side of that is that if you do manage to get a result, it is often quite significant and of immediate medical interest. These new modalities tend toward a death-or-glory readout, more so than in most clinical trials anyway, and these patients are the very definition of “unmet medical need”.
The latest results involve taking T cells that target CD19 (via chimeric antigen receptors, CARs), going after this protein which is expressed in all B-cell lineage types. This is the targeting that has led to three FDA-approved cell therapies for B-cell lymphoma, but all of these involve individual donor-derived T cells that are removed, engineered, and then infused back into each patient. When this works, the results can be dramatic indeed, but it’s an expensive and laborious process. If you could use allogenic T cells (“off the shelf”, as it were, and not derived from each individual patient), the who process could speed up considerably and be less costly as well. A number of groups have been working on this, and results so far indicate that the idea is probably feasible, but no one has quite realized its potential, either.
Caribou appear to be the first people in the clinic trying this allogenic anti-CD19 route with cells that also have their PD-1 receptors knocked out. That wrinkle has been proposed because one of the reasons that CAR T-cell therapy fails is thought to be because of increased PD-L1 expression on the targeted lymphocyte tumor cells. That interacts with PD-1 present on the CAR-T cells, and that signaling pathway is thought to partially cancel out the cytotoxic effect that you’re trying to induce. To that end, there have been quite a few attempts to damp down PD-1 on these T cells, or strip it out entirely, and that’s where Caribou’s CRISPR technology comes in.
They’ve reported treatment of five B-NHL patients so far, all of whom had gone through anywhere from two to eight prior therapies and had relapsed to their present uncontrolled state. All the patients got the standard lymphodepletion drugs for several days before treatment. That’s a small group, but they had four complete responses and one partial response, which is just that sort of notable effect size mentioned above. All four of the complete responses at 28 days were still holding up at three months. Side effects were what you’d expect from this sort of thing, and in line with existing CAR-T treatments. The next step is to double the number of cells in the dose, with data expected later this year.
That’s a really interesting start, and I hope that this holds up with further clinical work (both in the efficacy and safety ends). There’s some cell-biology interest as well in the whole mechanism, with this recent paper from a group in Russia that cast some doubt on the whole idea of lowering PD-1 in these cells. That team expressed an anti-PD-1 nanobody in their anti-CD19 T cells to knock down its function, but this led to decreased cell survival and decreased cytotoxicity against the targeted B cells. They concluded that lowering PD-1 might be a bad idea in general in this system, but these clinical results are a strong counterargument, at least so far. Immunology being what it is, and cell biology being what it is, there may well be rather different effects in stopping expression of PD-L1 completely (as in this CRISPR route) versus having it expressed but cancelled out by another engineered protein (as with the Russian paper). Let’s see what happens in the patients! (You will recognize the time-honored motto of all clinical oncology work there).
