Autologous non-myeloablative haemopoietic stem-cell transplantation compared with pulse cyclophosphamide once per month for systemic sclerosis (ASSIST): an open-label, randomised phase 2 trial

Background Non-randomised studies of haemopoietic stem-cell transplantation (HSCT) in systemic sclerosis have shown improvements in lung function and skin flexibility but high treatment-related mortality. We aimed to assess safety and efficacy of autologous non-myeloablative HSCT in a phase 2 trial compared with the standard of care, cyclophosphamide. Methods In our open-label, randomised, controlled phase 2 trial, we consecutively enrolled patients at Northwestern Memorial Hospital (Chicago, IL, USA) who were aged younger than 60 years with diffuse systemic sclerosis, modified Rodnan skin scores (mRSS) of more than 14, and internal organ involvement or restricted skin involvement (mRSS <14) but coexistent pulmonary involvement. We randomly allocated patients 1:1 by use of a computer-generated sequence with a mixed block design (blocks often and four) to receive HSCT, 200 mg/kg intravenous cyclophosphamide, and 6.5 mg/kg intravenous rabbit antithymocyte globulin or to receive 1.0 g/m(2) intravenous cyclophosphamide once per month for 6 months. The primary outcome for all enrolled patients was improvement at 12 months' follow-up, defined as a decrease in mRSS (>25% for those with initial mRSS >14) or an increase in forced vital capacity by more than 10%. Patients in the control group with disease progression (>25% increase in mRSS or decrease of >10% in forced vital capacity) despite treatment with cyclophosphamide could switch to HSCT 12 months after enrolment. This study is registered with ClinicalTrials.gov, number NCT00278525. Findings Between Jan 18, 2006, and Nov 10, 2009 we enrolled 19 patients. All ten patients randomly allocated to receive HSCT improved at or before 12 months' follow-up, compared with none of nine allocated to cyclophosphamide (odds ratio 110, 95% CI 14.04-infinity; p=0.00001). Eight of nine controls had disease progression (without interval improvement) compared with no patients treated by HSCT (p=0.0001), and seven patients switched to HSCT. Compared with baseline, data for 11 patients with follow-up to 2 years after HSCT suggested that improvements in mRSS (p<0.0001) and forced vital capacity (p<0.03) persisted. Interpretation Non-myeloablative autologous HSCT improves skin and pulmonary function in patients with systemic sclerosis for up to 2 years and is preferable to the current standard of care, but longer follow-up is needed.