Autologous Nonmyeloablative Hematopoietic Stem Cell Transplantation for Diffuse Cutaneous Systemic Sclerosis: Identifying Disease Risk Factors for Toxicity and Long-Term Outcomes in a Prospective, Single-Arm Trial

Abstract

Objective: Two randomized trials for patients with diffuse systemic sclerosis (SSc) demonstrated an overall survival (OS) and event-free survival (EFS) advantage of autologous hematopoietic stem cell transplantation (AHSCT) using CD34+ selected peripheral blood stem cells (PBSCs) compared with monthly cyclophosphamide (CY). We asked if an unmodified PBSC graft followed by maintenance mycophenolate mofetil (MMF) after AHSCT, instead of a CD34+ selected graft, could provide comparable AHSCT outcomes.

Methods: Twenty patients with high-risk SSc were enrolled in a prospective, single-arm trial with CY 200 mg/kg and horse antithymocyte globulin (ATG; CY200/ATG), followed by unmanipulated autologous PBSC, and then MMF maintenance starting at 2 months after AHSCT.

Results: Point estimates of OS and EFS at 5 years after AHSCT were 85% (95% confidence interval [CI] 60.4%-94.9%) and 75% (95% CI 50%-88.7%), respectively. Median follow-up was 7.5 years (range 5.6-11.6) after transplant for living patients. Eight patients (40%) required intensive care unit treatment early after transplant. Early transplant-related mortality occurred in two patients (10%). Five patients developed relapse/progression of SSc after AHSCT. Four of nine patients with anti-RNA polymerase III antibodies had prior scleroderma renal crisis and the lowest quartile of estimated glomerular filtration rate (eGFR) on study entry; all four patients developed prolonged organ failure/death early after transplant.

Conclusion: We observed favorable OS and EFS after AHSCT for patients with SSc, using CY200/ATG, unmanipulated PBSCs, and MMF posttransplant maintenance, which was comparable to trials with CD34+ graft selection. We identified a possible risk factor, pretransplant low eGFR, for adverse outcomes after AHSCT.

Figure 1

Swimmer plot of survival and events for 20 patients with diffuse cutaneous systemic sclerosis enrolled in this autologous HSCT trial. The patient number is shown as identified in Table 1 in ascending sequence of trial enrollment. A red font patient number indicates anti‐RNA‐polymerase III seropositive before HSCT. The exclamation mark indicates a history of scleroderma renal crisis, “↓” indicates baseline pre‐HSCT decreased renal function (estimated glomerular filtration rate <65 mL/min/1.73m²), and the asterisk indicates death. The blue bar border indicates survival without event. The red bar border indicates survival after meeting the protocol‐defined organ failure event endpoint. The green bar border indicates survival after the addition of a DMARD for the treatment of clinical systemic sclerosis progression or disease activity. The purple fill indicates treatment duration with protocol‐prescribed maintenance MMF or MPA. All patients received an unmodified (not CD34+ selected) autologous peripheral blood stem cell graft at time point 0, and the follow‐up is shown in years. DMARD, disease‐modifying antirheumatic drug; HSCT, hematopoietic stem cell transplantation; MMF, mycophenolate mofetil; MPA, mycophenolic acid.

Figure 2

Kaplan‐Meier survival curves of (A) OS, (B) EFS, and (C) DEFS for the entire cohort of 20 systemic sclerosis patients after HSCT. The EFS endpoint included death or meeting the defined organ failure criteria for longer than 6 months. The DEFS endpoint included death, organ failure, or initiation of additional disease‐modifying antirheumatic drugs (indicator of disease relapse/recurrence after transplant). Tick marks show the time of censorship and status at time of last contact. The dotted lines indicate the 95% confidence interval. DEFS, composite of additional disease‐modifying antirheumatic drug and event‐free survival; EFS, event‐free survival; HSCT, hematopoietic stem cell transplantation; OS, overall survival.

Figure 3

Rate of change in skin score, lung function, and HAQ‐DI score after nonmyeloablative autologous hematopoietic stem cell transplantation. The Bsl values and statistical analysis of change over time is shown in Table 2. The colored lines are spaghetti plots of change from the pretransplant Bsl for each patient, and the bold black lines are the overall change trend based on the linear mixed effect model. (Upper left) Change from the pretransplant Bsl in mRSS reported at timepoints after autologous HSCT. (Upper right) Change in FVC% at time points after autologous HSCT. Eight of 18 evaluable patients had a greater than 10% increase in predicted value from the pretransplant Bsl. (Lower left) Change in DLco at timepoints after autologous HSCT. Seven of 18 evaluable patients had a greater than 15% increase in predicted value from the pretransplant Bsl. (Lower right) Change in HAQ‐DI value (range 0–3.0) from pretransplant baseline up to 4 years after HSCT. Bsl, baseline; DLco%, diffusion lung capacity carbon monoxide corrected for hemoglobin percent predicted value; FVC%, forced vital capacity percent predicted value; HAQ‐DI, Health Assessment Questionnaire Disability Index; HSCT, hematopoietic stem cell transplantation; mRSS, modified Rodnan skin score.

Figure 4

Kaplan‐Meier survival of patients based on ARA status (neg or pos) before HSCT. (A) EFS of ARA‐neg (n = 11) and ARA‐pos (n = 9) patients (difference not significant). (B) ‐ DEFS of ARA‐neg and ARA‐pos patients. Black lines indicate ARA neg, and red lines indicate ARA pos. Tick marks show the time of censorship and status at time of last contact. The dotted lines indicate the respective 95% confidence intervals. ARA, anti‐RNA polymerase III autoantibody; DEFS, composite of additional disease‐modifying antirheumatic drug and event‐free survival; EFS, event‐free survival; HSCT, hematopoietic stem cell transplantation; neg, negative; pos, positive.