This patient has very high risk, steroid-refractory chronic GVHD, since he has progressed in spite of appropriate doses of steroids administered for at least two  weeks.¹  Therefore continuing current therapy would not be a desirable therapeutic option in this patient.  Some manifestations in our patient are reminiscent of acute GVHD and such late presentations of acute GVHD manifestations are associated with worse survival in chronic GVHD studies.²  Unfortunately, the prognosis for this patient is grim and the best therapy is not known.³  If a clinical trial is available, we would strongly recommend enrolling this patient to help us determine the best treatment for patients with multiple high risk features not responding to primary steroid-based therapy.  The options listed are not optimal.  In this poll 17% of respondents recommended PUVA.  PUVA therapy has been shown to be helpful in patients with skin (lichen planus-like) disease only, but not systemic disease as seen in this patient.⁴  Phototherapy (not given as an option in this poll) is more commonly delivered in the form of extracorporeal photopheresis (ECP) and could be considered as it may control systemic GVHD while theoretically maintaining the desired graft-versus-leukemia (GVL) effect.  But the rapid progression of the patient’s symptoms is concerning since ECP therapy often requires several weeks to see improvement.  Likewise, he would require careful management with his degree of thrombocytopenia through the line placement and the procedure.  At least in one trial, patients with thrombocytopenia also had a lower rate of response to ECP.⁵  

54% of respondents recommended mycophenolate mofetil (MMF).  MMF has been used in steroid-refractory chronic GVHD and is the most commonly used second line agent.  There have been a number of small phase II trials reported using MMF with response rates in steroid refractory disease ranging from 45-75%.⁶  Although the drug is well tolerated in general, in our experience it would be difficult to administer in this patient given his mouth sores, weight loss and thrombocytopenia.  Mycophenolic acid (MPA) drug levels may prove useful in ensuring he is receiving an adequate trial.  Rituximab (10% of responders) is also appealing given its low toxicity, positive effects in immune thrombocytopenia and unique mechanism of action.  However, rituximab may be less effective for visceral disease, such as the rapidly progressive hepatic involvement seen in this patient.⁷ Sirolimus (15% of responders) has been studied also in phase II trials of chronic GVHD.⁸   When combined with tacrolimus and steroids, response rates of more than 60% have been reported.  The major toxicity seen with this combination was renal.  Again, obtaining blood levels may help to minimize toxicities and ensure an adequate trial of treatments.

We would encourage the patient to enroll in a clinical trial, if one is available.  If the patient is unable to participate in a trial, then of the potential treatments listed above, we would first try sirolimus because of its mechanism of action and encouraging results reported so far.  However, both MMF and rituximab are also reasonable choices.  Since this patient is showing some worrisome features of acute GVHD it would be reasonable concurrently with starting the new therapy also to administer a 3-7 day course of prednisone at 2 mg/kg/day or a four-day pulse of intravenous methylprednisolone at 10 mg/kg/day for more rapid disease control and then go back immediately to previous dose of prednisone.⁹

As stated above, the best therapy is not known and it is unfortunately a matter of trial and error based on toxicity profiles, past experience and patient choice. Chronic GVHD remains a great challenge and major community efforts will need to continue in order to better address this late effect of transplantation.