Abstract
Introduction
Systemic sclerosis (SSc) is a chronic autoimmune disease that affects skin and internal organs. Severe progressive cutaneous SSc with heart, lung, digestive tract or kidney involvement has poor prognosis with mortality rates up to 30% 5 years after diagnosis despite standard therapies. Autologous Hematopoietic Stem Cell Transplantation (AHSCT) has been shown effective for treating severe or rapidly progressive SSc. Improved skin score and quality of life and at least stabilization of lung function were consistently reported in early phase II trials. Both the phase II ASSIST and the larger phase III ASTIS prospective randomized trials showed better survival rates and long-term outcomes in patients (pts) treated with AHSCT compared to monthly iv cyclophosphamide. Among the current strategies to further improve safety and efficacy of AHSCT, the benefit of ex vivo CD34+ cell selection of the graft remains debated. The potential risk of re-injecting pathological T-cells with non-selected grafts may be outweighed by high cost, reduced numbers of injected cells and increased immunosuppression in pts receiving CD34+selected grafts. We therefore aimed to evaluate the influence of ex vivo CD34+ cell selection on the outcome of SSc pts treated with AHSCT.
Methods
This Autoimmune Disease Working Party (ADWP) multicenter retrospective study followed the European Bone Marrow Transplant Association (EBMT) guidelines. All EBMT-affiliated centers with eligible SSc patients were invited to participate. Inclusion criteria were: diagnosis of SSc according to the American College of Rheumatology criteria and all pts, but those from ASTIS, having received a first AHSCT between 2000 to 2012, using any type of conditioning except irradiation. For those centres that agreed to participate, primary data was collected, where available using a standardized SSc minimal essential data ‘B’ form. A specific added Excel questionnaire was elaborated to include clinical and laboratory data at diagnosis, before and after AHSCT. Filled questionnaires were sent by local investigator to ADWP- EBMT data manager, evaluated for completion and consistency.
Results
Out of 131 SSc pts (70% female, n = 92) with the inclusion criteria enrolled in the study, median aged at SSc diagnosis 41 (17-68) years, ex vivo CD34+ selection of the graft had been performed in 63 (48%). Before AHSCT, 120 (92%) pts had diffuse skin involvement and 69 (53%), 27 (21%) and 103 (79%) pts respectively had gastrointestinal, heart and lung involvement. CD34+ selected and non-selected transplant pts were similar for median age at diagnosis (34 vs 40 years), female gender (76 vs 63 %), interval from SSc diagnosis to AHSCT (21 vs 26 months), modified Rodnan’s skin score at mobilization (25 vs 22) and for the number of infused CD34+ cells/µl (4.1 vs 4.7). Presence of Raynaud’s phenomenon before SSc diagnosis (92% vs 74%, p=0.005) and of heart involvement before mobilization (29% vs 13%, p=0.03) were more frequent in pts receiving CD34+ versus non selected grafts. With a median follow-up of 39 (1-161) months after AHSCT, the 5 year Progression Free Survival (PFS) and Overall Survival (OS) were 59 (49-69) and 81 (73-89) months for the whole study group. Two year PFS and OS were respectively 68% [95% CI: 60-76] and 86% [80-92]. Non Relapse Mortality (NMR) was 3.9% at day 100 and 5.5% at 2 years. There was no significant difference between between pts receiving CD34+ and non selected grafts in terms of 100 d Non Relapse Mortality (NMR) (3% vs 4%), 2 year OS (80% vs 85%), PFS (54% vs 67%) and overall NRM (5% vs 6%). By multivariate analysis, OS was lower in the presence of SSc heart involvement before mobilization (HR: 6.36; p=0.0002) and relapse/progression increased in presence of SSc gastrointestinal involvement (HR: 2.39; p=0.02) before mobilization .
Conclusion
This ADWP EBMT retrospective study demonstrates that ex vivo CD34+ selection of the graft does not add benefit to the OS, PFS and NRM in SSc pts after AHSCT. It also shows that SSc heart involvement at baseline decreases overall survival and that gastrointestinal involvement contributes to disease progression after AHSCT. Further prospective randomized trials, under controlled and more homogeneous conditions, are warranted to confirm these findings.
Disclosures
No relevant conflicts of interest to declare.
Principled selection of baseline covariates to account for censoring in randomized trials with a survival endpoint