Putting the CART before the horse: could CAR-T cells be a last-resort therapy in MS to rival HSCT?

First off, apologies my long winter hibernation from the blog. As you may have gathered (if anyone is paying close attention to my blogging habits) I'm no longer full-time at BartsMS, and have been working in mental health down the road at Mile End. There are loads of interesting things to say about the overlap between MS and mental health, lots of which have been discussed before on the blog. 

As well as the obvious overlap - disorders like depression, anxiety and cognitive impairment are more common in pwMS - I think there is a lot to learn from psychiatry about the whole approach to doctoring. From my limited experience, psychiatrists are better than us physical docs at thinking holistically about health, and at making proper shared decisions with clients - clients, rather than 'patients'. But I'll write more about this another time.

As you may be bored of reading by now, in the past few years it's become clear that B cells - the cells of the immune system that make antibodies - are heavily involved in the disease process. The most vivid piece of evidence for this is the effectiveness of ocrelizumab in dampening disease activity and disability progression. Ocrelizumab is pretty much a selective B cell-killing agent, so this implies that knocking out B cells is sufficient to disrupt the disease process. 

The problem with any drug that targets the immune system is that it will inevitably lead to 'off target' side effects. We think that a small proportion of B cells recognise something in the brain as 'foreign' and are responsible for driving the disease process - the ideal drug for MS would target these cells while leaving the vast majority of innocent B cells alone. This would presumably lead to fewer problems with long-term cancer risk and immunity against bugs.

While this is some way off, there is lots of exciting work going into selectively targeting particular cells of the immune system. An amazing technology which has been spearheaded by the cancer world caught my eye this week: the technology is called 'Chimeric Antigen Receptor T cells' or CAR-T cells. The idea is that each normal T cell carries a unique receptor which means that they are geared up to respond maximally to a specific particle, such as a bit of a virus or a bit of a bacterium. You can now make T cells in the lab to carry whatever receptor you want, and can therefore generate T cells that recognise and respond to anything. These cells are grown in the lab and then infused into the patient, stimulating a massive immune response against whatever target has been chosen. 

Several clever groups of scientists have generated CAR-T cells which specifically recognise and kill B cells. They've done this by making CAR-T cells which recognise CD19 - a marker present on the surface of all B cells. There are now two versions of these cells approved for use in the US for children with B-cell cancers. I read this very interesting paper this week about using CAR-T cells for a horrible B cell cancer which had not responded to first-line treatment.  It was extremely effective, producing remission in 81% of people at 3 months. As expected for a last-ditch therapy, it was also incredibly dangerous. About 3/4 of people had serious adverse events. 8/10 had what's called a cytokine storm - a severe systemic illness which often requires admission to intensive care. What is very interesting is that although the therapy led to lots of nasty short-term complications, people did not die of these side effects.

I'm blogging about this here because CAR-T cells, although dangerous, have the potential to be used for pwMS who do not respond to any of the normal highly-active therapies. I wonder if in the next few years this will come to rank alongside stem cell transplantation as an effective, albeit very dangerous, last-resort option for people who have tried everything else. 

I should emphasise that at the moment there is no evidence whatsoever that anti-CD19 CAR-T cells work in MS. Equally, these cells are still pretty broad-brush sledgehammers: they do not discriminate between different B cell subsets. They are particularly useful in B cell cancers in which the disease-causing cells are much more numerous in the blood than in MS. 

My hunch is anti-CD19 CAR T cells would be very effective in MS, but it is going to be difficult to justify doing a trial given how many safe and highly-effective alternatives are out there. It's likely that in the next few years more refined versions of this tech will come out that can target only certain subsets of B cells. 

Would you ever consider CAR-T cells if this kind of thing were available in the next 20 years?

Labels: , , ,