Severe Allograft Rejection and Autoimmune Hemolytic Anemia After Anti-PD1 Therapy in a Kidney Transplanted Patient
A 73-year-old man received a diagnosis of a superficial spreading melanoma at a metastatic stage during 15 months posttransplant. Immunosuppression was reduced, and tacrolimus was switched to everolimus (2.5 mg/d). After 2 injections of 3 mg/kg of body weight of anti-PD-1 antibody at 30 days interval, he presented with acute renal failure 25 days after the last dose, with an enlarged kidney transplant and grade 2A acute cellular rejection on biopsy. No donor-specific antibodies were identified; the rejection was resistant to high doses of steroid, and the patient returned to dialysis. Simultaneously, he developed AIHA with a Coombs assay positive for complement and severe thrombocytopenia (nadir platelet count, 38 000 per mm3). There was no evidence of thrombotic microangiopathy, no platelet antibodies, and no local invasion on bone marrow histology. It resolved in 1 week, associated with use of high-dose steroids and cessation of nivolumab. The patient died from superficial spreading melanoma dissemination 3 months later.
There have been 7 other reports of anti-PD-1 administration in kidney transplanted patients (6 nivolumab and 1 pembrolizumab). Among the total of 8 patients, 6 presented with acute cellular rejection (Banff grade 2A or 2B), including the 2 patients with areas of cortical necrosis. No patient had donor-specific antibodies, and the timing of rejection a few days or weeks after nivolumab suggest causality, especially in patients on stable immunosuppression for 10 years or more in whom reduction of immunosuppression would not be expected to result in immediate rejection.
Mycophenolate had been stopped in all patients at the time of cancer diagnosis, and calcineurin inhibitors had been withdrawn, minimized, or switched to low-dose everolimus as in our patient. Two patients remained on full-dose immunosuppression and had no rejection. Tacrolimus was switched for full-dose everolimus in 1 patient, and the authors advocated for the maintenance of significant immunosuppression associated with the use of anti-PD-1 administration.2
Anti-PD-1 antibodies have also been associated with autoimmune manifestations with AIHA reported in 4 cases.3 It occurred after a mean of 4 cycles, was Coombs positive for immunoglobulin G and/or C3, and responded well to high dose of corticosteroids. Our case is the first description of AIHA in a transplant patient.
The indication for anti-PD-1 therapy will likely increase in transplanted patients with cancer. The transplant community must therefore be aware of the risk of allograft rejection and transplant loss. Immunosuppression minimization, although it is the recommended approach to active neoplasia in a transplant recipient, must be reconsidered when the patient is treated with anti-PD-1 therapy.