<p>Incorporating patient-reported outcomes (PROs) may enhance the precision of response assessments in chronic graft-versus-host disease (cGVHD). In this discussion, our expert panelists provide updates on response assessment in the cGVHD clinical trial setting and the potential of PROs to inform decision making and personalized treatment strategies in the clinic.</p>

cGVHD is a disease that, in many ways, is driven by patient symptoms. So, there is always an aspect of subjectiveness that affects how we approach treatment. If a patient with cGVHD expresses that they are suffering from symptoms, then that may push us to engage in another line of therapy. For a patient who has ongoing low-level symptoms, we may keep their current treatment as is and reassess at a later date.

 

The National Institutes of Health (NIH) consensus criteria for cGVHD provide a framework for assessing therapeutic response and are used in clinical trials. Recent clinical trials have identified successful therapeutic agents with overall response rates of approximately 70%, but the vast majority of those responses are partial responses.

 

These partial responses vary by patient. For example, there are times when a patient with a partial response may feel like a new person, whereas another patient with a partial response may feel that their symptoms are only a little bit better. Both of those scenarios might be adjudicated as a partial response in a clinical trial setting. The question is: Can we find a way to better integrate PROs into clinical trials? We have different tools and techniques to assess patient symptoms. However, additional research is needed to determine the best way to incorporate a patient symptom scale into our response criteria in a way that can be validated across all trials.

I think that this is an area where there is a significant difference between clinical trial and clinical practice settings. Of course, there are other areas in transplant medicine where the clinical trial tools very much mimic what we do clinically. For instance, assessing response to acute GVHD treatments is often based on the same criteria that would be used in a clinical trial setting, although we may switch therapies more aggressively sometimes. In cGVHD, however, as Dr DeFilipp highlighted, much of it is based on patient symptoms, which are much more subjective for us to measure because we are relying on the patient, and patients may experience what we would otherwise see as objectively similar responses differently.

 

In the clinical practice setting, I think that much of determining a patient’s response to treatment is centered on talking to the patient, trying to make some objective assessments when we can, and making clinical decisions on that basis. However, there are certain situations for which we do have hard metrics, such as for pulmonary function testing. Those are things that we can measure, but many of the other things, while we could technically measure them the same way as we would in a clinical trial (at least in my practice), are not something that I do routinely. Essentially, we follow the NIH response criteria, which are meant for clinical trials but are adapted to the realities of clinical practice. In clinical practice, we cannot perform the full formal assessment and grading, but we can at least assess most of the organs and combine that information with patient symptoms.

 

What I have found in some cases is that I can see objective responses (eg, areas of skin tightening starting to resolve with the introduction of a new drug), but the patient does not appreciate them until you point them out. The patient has become so used to living with their disease that they do not notice the difference the drug is making in improving their symptoms. One of our challenges in the clinical setting is determining how to integrate these tools. I think that it is important to have good metrics calculating the patient experience that we can all agree on and that can guide patient care, not only in clinical trials but also in clinical practice.

The NIH global scoring system for cGVHD is a very blunt instrument, and, to some degree, asking every patient the same 15 or 20 questions in a survey is just as blunt of a way to assess patient experience. It is an important first step, and I think that we all applaud the US Food and Drug Administration (FDA) for including PROs as part of the pathway to drug approval, but we are not there yet in terms of a sophisticated way of assessing PROs.