Trouble from CAR-T Treatment?

BY DEREK LOWE

 The FDA took everyone by surprise recently by announcing that they were looking into 19 cases of lymphoma that they believe were actually brought on by the use of CAR-T cell therapy to treat other blood cancers. CAR-T is pretty much the last technological resort for resistant forms of leukemia, lymphoma, and myeloma; you turn to this method after all others have failed.

But that’s not because it doesn’t work, of course. Chimeric antigen receptor T cell therapy is often dramatically effective, but it’s a very expensive and laborious way to go. That’s because the treatment has to be customized for each individual patient. T cells from a patient are genetically modified to express receptors that target specific antigens (like CD19 and BMCA) that are not found in wild-type cells. These modified cells are expanded to large numbers and gradually infused back into the patient, where they start attacking their targets. That attack is a stressful process that can involve cytokine release syndrome and other side effects, but it can do a spectacular job of clearing the overabundant blood cell types that are the hallmark of those malignancies. The CAR-T cells multiply after infusion, and the effects can last for years. (In childhood syndromes like ALL, this is often used to buy time for eventual stem-cell transplants into the bone marrow, which can be curative).

The FDA says it’s aware of 19 cases of T-cell lymphoma, however, that may have been brought on by the modified T cells themselves. One way that could happen would be during the engineering to get the chimeric antigen receptors into their genomes - if these land in some specific (and specifically wrong!) places in the DNA sequence, it could set off carcinogenesis. At first, that might even manifest itself as a strong response to therapy - the CAR-T cells are multiplying away and killing off the B-cells (or other target) of the cancer to be treated. But what if they don’t stop, because they are experiencing cancerous growth themselves?

Interestingly, from news reports (here’s Stat and FiercePharma) the companies involved in making these therapies say that they have not received enough reports to add up to the FDA figure, and practitioners are expressing surprise as well. It’s certainly possible, though, that the agency has reports of its own. The total number of patients treated with CAR-T therapy so far would appear to be in the low tens of thousands. So that’s where you start working on the risk assessment. Are there particular antigens that are more likely to lead to T-cell malignancy? Particular ways of inserting them into the T-cell genomes? How many of the affected patients have had these cells sequenced to see if they really are a result of the CAR-T engineering, versus wild-type malignancies? These questions will all have to be sorted out to understand what we’re looking at.

The calculation would be easier if they patient population stayed in the “last resort” category, because those patients are going to die anyway, unfortunately, and do so rather quickly if something like CAR-T isn’t applied. But there are reports that CAR-T when administered earlier can spare patients years of chemotherapy - which is good, but now a possible risk of T-cell trouble may have to be factored in. If it turns out that chimeric antigen receptor therapy (as currently practiced) can lead to rare cases of T-cell lymphoma, the answer may be to change the way we do it - the technology for cell engineering is getting better all the time, and we should be able to lower the incidence of inadvertently producing a cancer cell line in the process, if that’s what’s going on. 

https://www.science.org/content/blog-post/trouble-car-t-treatment