Focusing on acute graft versus host disease, panelists consider tools and methods that help to overcome challenges in making a differential diagnosis.
Corey Cutler, MD, MPH, FRCPC: We’ve mentioned a lot about acute and chronic [graft-vs-host disease]. Let’s jump into acute GVHD [graft-vs-host disease] and discuss it in some degree of detail. Acute GVHD, as those of us in the field know, is an alloimmune response generally directed against 3 organs and 3 organs only. The reasons for that aren’t entirely clear, but we tend to recognize acute GVHD affecting the skin, the gastrointestinal [GI] tract, and the liver in that order of frequency. The skin will present with an erythematous—often raised, or maculopapular—rash affecting small body surface areas all the way up to 100% BSA [body surface area] when there’s very severe desquamation.
When we think about the clinical differential of acute graft-vs-host disease, drug toxicity or things like Stevens-Johnson syndrome are the main things we think about. Acute GVHD of the skin will occur in about 70% of people who get GVHD. The GI tract is the next most commonly affected organ. Predominantly it presents with profuse watery diarrhea, often more than 5 to 10 bowel movements a day. In fact, we quantify the number of bowel movements and the volume of diarrhea when we’re staging and grading GVHD. Although it’s not the most common organ involved with acute GVHD, it’s the most morbid organ and leads to the most adverse outcomes in GVHD altogether.
The liver is the third most common organ, is affected the least, and rarely presents as a solitary organ involvement. When we think about hepatic GVHD, we think about drug toxicities, veno-occlusive disease of the liver, and other early complications of transplantation that might be the differential. For the gastrointestinal tract, because our patients are often hospitalized, we think about things like cytomegalovirus or other infections, like Clostridioides difficile, as the main causes of a differential or diarrhea in the transplant patient.
We often do diagnostic biopsies. We do the skin very rarely because it has a fairly characteristic appearance, but for the gastrointestinal tract, the majority of us will perform flexible sigmoidoscopy or colonoscopy to get a tissue biopsy of the distal colon for diagnostic purposes, particularly to exclude things like cytomegalovirus or adenovirus colitis. The liver is the most tricky organ to diagnose because a liver biopsy is not without its own risks. There are no gold standards, so how do you go about deciding whether somebody has GVHD when the clinical scenario was a little less clear? Yi-Bin, what do you do?
Yi-Bin Chen, MD: As you said, it’s a clinical diagnosis. Presenting symptoms of rash and diarrhea and LFT [liver function test] abnormalities are nonspecific, so it’s our job to try and put it together with all the information in front of us and maybe a diagnostic tissue biopsy to put it together. A lot of the time, you’re excluding other causes. In certain patients, you’re doing a little watchful waiting to see what happens as you withdraw certain drugs or other things to make the diagnosis.
I won’t say we have 100% confidence every time. Sometimes it’s the best guess about what this appears to be. There’s a motivation to have other noninvasive tests for graft-vs-host disease. What’s that rooted in? Part of it is what Corey mentioned, which is that sometimes the diagnostic picture is unclear. If we had a blood test such as a troponin, which exists for cardiac ischemia, for graft-vs-host disease, it would be superuseful. The other motivations are that graft-vs-host disease is an adaptive immunological response. It doesn’t start 1 day and present the next day. An immune response takes a couple of weeks to develop from what we’re taught in basic immunology. It would be helpful for us to have a picture of what’s going on before the clinical morbidity ensues.
The third motivation, as Corey mentioned, is for someone with lower GI disease. We’re judging response by measuring the volume of diarrhea and the number of bowel movements. That’s incredibly primitive. Many things affect that, such as the use of antibiotics, the use of narcotics, the oral intake the patient is having, and so forth. It would be helpful to have much more accuracy to understand the biological underpinnings.
Transcript edited for clarity.