Steroid Refractory GVHD: Scope of the Problem

Yi-Bin Chen, MD: In this last part, we are going to discuss the exciting recent advances in the therapy of steroid refractory acute and chronic graft-vs-host disease [GVHD]. As Corey mentioned, there has been a lot of industry activity that has resulted in a lot of large trials in these populations. Fortunately for us and our patients, there have been approvals and will be more approvals to give our patients more options for therapy, ultimately to improve their outcomes.

Even for our best efforts in prevention and initial therapy, I think we all hold close our patients with steroid refractory graft-vs-host disease because they do the worst; we see them the most. And they’re the ones who ultimately pass away from this complication that we inherently feel we’ve had a part in causing from the prescribed therapy we’ve given. For acute graft-vs-host disease, we have discussed a little about why trials haven’t succeeded. I think the big reason why they have not succeeded is the biological heterogeneity of the population. If we all look to historical trials for steroid refractory acute GVHD, we can see the criteria include 3 distinct populations. No. 1 is those who progress on the first 3 to 5 days of high-dose steroids; No 2, those who do not get better after a week; and then No. 3, those who respond first, we taper steroids, and they flare. That third population is probably more steroid dependent rather than refractory, but because of numbers and the need to get trials done, we’ve somewhat lumped them together into steroid refractory. We have increased the heterogeneity to get these trials done, and as a result, we sometimes don’t make progress that way.

I think chronic graft-vs-host disease is a bit messier; the timelines are longer, the response metrics are a bit more difficult, and they are a bit more subjective given the tools we have in terms of the NIH [National Institutes of Health] scale and the numerical scale. Month to month, you try to grade; you don’t remember the number you gave the time before and you probably don’t have access to it. These are inherent to the systems we’ve created. Then as Corey mentioned, even those who are not steroid refractory, they’re on long courses of steroids; the amount of morbidity they experience, especially the older population, is significant.

I think we can all agree that large studies show that over half of patients with both acute and chronic [GVHD] who start systemic therapy will require another agent at some point during their course. At the end of the day, to me that tells us where we are for steroid refractory acute and chronic graft-vs-host disease.

Richard Maziarz, MD: Some people consider a fourth version of steroid refractory GVHD. This has affected end points of some studies: when you have a primary response and an ongoing response in some organs, but you have progression in a single organ. It’s a mixed response as we would have in treatment of lymphoma.

For steroid refractory GVHD, it’s been the bane of existence for most of us. Health resource utilization goes up. If you look at cost analyses, something I’ve been examining for the last number of years, that’s the outliers. We can see that 20% or 25% of patients will clearly add hundreds of thousands of dollars of extra hospital expenses in the allogeneic setting; that’s for steroid refractory GVHD. It’s refractory to our primary therapy and it’s fraught with clinical failure, a high mortality rate. A lot of the issues in the past have been primarily because our therapies were poorly designed. If we take HIV as a model, AZT [azidothymidine] made a big difference. However, it was a single biologic attack and a virus that could mutate and then become resistant. It wasn’t until we had multimodality viral pathway interruption that we could actually make a difference in HIV.

With GVHD, many of the studies we’ve done for 20 years trying to say can we impact steroid refractory acute GVHD with single-agent studies, added on to patients who were already resistant to steroids, you basically are hitting 1 molecule, whether it be a cytokine or a homing molecule, it is a single intervention for a complex pathophysiology.

Transcript edited for clarity.