Transplantation-associated thrombotic microangiopathy (TA-TMA) represents a challenge after allogeneic haematopoietic stem cell transplantation (HSCT) because of diagnostic uncertainties, lack of established treatment, and an overall poor prognosis (Laskin et al, 2011). We report the case of a 61-year-old man who was diagnosed with multiple myeloma in 2009. He was initially treated with a combination of bortezomib, doxorubicin and dexamethasone, followed by high-dose melphalan with autologous HSCT. Following disease relapse in 2011, he received three courses of bortezomib, thalidomide dexamethasone, and was then included in a sequential autologous-allogeneic tandem approach that comprised high-dose melphalan and auto-HSCT followed by two Gray total body irradiation-conditioned allo-HSCT (Karlin et al, 2011). Two months post-HSCT, the patient was diagnosed with graft-versus-host disease (GvHD) of the skin (stage 3) and gut (stage 1), which responded to methylprednisone 2 mg/kg. Three months later, GvHD of the gut reappeared during steroid tapering. Although the GvHD responded to the increased doses of steroids, the patient developed a severe TA-TMA (Fig 1). As already reported in such a situation, acute renal failure was absent (Cho et al, 2010). A concomitant diagnosis of cytomegalovirus (CMV) & Epstein Barr virus (EBV) infections was made. Ciclosporin was stopped, treatment with Foscanet was initiated and the patient received one injection of mabthera. Seven days later, he developed a progressive involvement of the central nervous system, with confusion and peripheral facial paralysis, while the biological characteristics of TA-TMA remained stable (Fig 1). Ciclosporin was no longer detectable in the blood. CMV & EBV infections responded to treatment at that time. Magnetic resonance imaging and a lumbar puncture were normal. Complement proteins (C3, C4, Factor H, Factor I) were in the normal ranges. Testing for anti-complement Factor H antibodies was negative. No decrease in ADAMTS 13 (a disintegrin and metalloproteinase with a thrombospondin type 1 motif, member 13) activity was identified. Given the devastating prognosis, we administered eculizumab to this patient accordingly to atypical haemolytic and uraemic syndrome dosage (900 mg weekly for four weeks, followed by 1200 mg every two weeks). The patient's neurological status improved dramatically within 48 h after the first infusion of eculizumab. Clinical improvement was associated with rapid normalization of disease activity markers: platelet counts increased, and lactate dehydrogenase levels decreased quickly (Fig 1). Complete C5 blockage, defined by a 50% haemolytic complement (CH50) activity below 10% was observed 24 h after the first infusion of eculizumab and during the entire study period. Concomitantly, the number of circulating endothelial cells (CEC), a known prognostic marker in thrombotic microangiopathy (Erdbruegger et al, 2006), decreased drastically. The levels of CEC were about 1200/ml before eculizumab infusion compared with 512/ml at 36 h and 5/ml at 3 days post-injection. The patient was discharged two weeks later with normal neurological status. Three months later, eculizumab treatment was stopped because there were no signs of TMA. After a 3-month follow-up period, the patient had completely recovered and continues to do very well. Given the progressive course of the disease after the mabthera injection, it seemed unlikely that the recovery was due to this treatment. The rapid clinical response to eculizumab supports the concept that TA-TMA might involve aberrant and autonomous complement activation despite the control of the potential causes (calcineurin inhibitors, GvHD & herpes virus infections). The dramatic resolution of symptoms after eculizumab administration suggests that TA-TMA is an area deserving further careful investigation of therapeutic complement blockade.
Conflict of interest
The authors do not have any potential conflict of interest to declare.