Abstract
Background. Diffuse cutaneous systemic sclerosis (SSc) is a multi-system chronically progressive and debilitating autoimmune connective tissue disease affecting internal organs, characterized by a dysregulated immune system and the production of autoantibodies. Patients experience progressive cumulative mortality despite currently available immunosuppressive therapies.
Objective. To assess the safety and efficacy of hematopoietic stem cell transplantation (SCT) with the research question: "what is the relative effect on overall mortality in patients with SSc and internal organ involvement treated with SCT compared to those treated with other available immunosuppressive therapies?"
Design. Systematic review with meta-analysis of randomized controlled trials (RCTs), with trial sequential analysis (TSA) adjustment for random error and repetitive testing of sparse and accumulating data (Patel A et al BMJ 2014;349:g4561).
Data sources. PubMed, PubMed Central, Web of Science, OvidSP, Cochrane central register of controlled trials (CENTRAL), clinicaltrials.gov, eudract.ema.europa.eu, controlled-trials.com, free Google search, relevant conference proceedings, hand searching of reference lists, and contact with authors as necessary. Search terms "systemic sclerosis" and "transplant". The search was last updated on 21 July 2018.
Eligibility criteria. RCTs of adult patients with SSc randomized to receive SCT compared with those who were randomized to receive a non-SCT treatment were included, if overall mortality outcome data were available.
Data extraction. Two reviewers independently assessed articles for inclusion, extracted data to assess risk of bias, trial methods, patients, interventions, comparisons, and outcome. The relative risk (RR) of overall mortality was calculated using a random effects model accounting for clinical heterogeneity.
Primary outcome measure. Overall mortality, using intension to treat (ITT) and per protocol (PP) analyses.
Risk of Bias. Assessed using the Cochrane Risk of Bias Assessment Tool.
Results. There were three eligible RCTs (ASSIST 2011, ASTIS 2014, SCOT 2018), including 250 patients followed for up to 7.8 years (156 patients in Europe and 94 in North America). All trials were open label and assessed as at unclear risk of bias. Peripheral blood stem cell collection: 2-4 g/m2 cyclophosphamide before reinfusion of thawed cryopreserved 2 million CD34+ cells/kg, or GCSF and CD34+ cell enrichment before reinfusion of thawed cells (median of 5.6 million cells/kg). Conditioning was cyclophosphamide 200 mg/kg with 6.5-7.5 mg/kg rabbit anti-thymocyte globulin (ATG) or 120 mg/kg with 8 Gy total body irradiation (TBI) (lung and kidney shielding with 2 Gy limit) and 90 mg/kg equine ATG. Supportive care included corticosteroids, angiotensin converting enzyme inhibitors (usually lisinopril), and aciclovir. Control group treatments comprised monthly cyclophosphamide, 0.5-1 g/m2, 6-12 infusions.
Significant overall mortality benefit was observed in the SCT group, with both an ITT (RR 0.60, 95% confidence interval (CI) 0.40-0.90; P=0.01) and PP analysis (RR 0.52, 95% CI 0.33-0.83; P=0.006). Statistical heterogeneity was absent (I2=0%). SCT benefit was robust to TSA adjustment for both the ITT (RR 0.60, 95% CI 0.36-0.98; P=0.014; diversity adjusted information size (D2=0%) and PP analyses (RR 0.51, 95% CI 0.31-0.86; P=0.004; D2=0%). The information size for an ITT model had >70% power (Fig 1), while the PP model had >90% power (Fig 2). The corresponding number needed to treat (NNT) to prevent 1 death was 7 (95% CI 4-29) for the ITT analysis and 5 (95% CI 3-11) for the PP analysis.
However, significantly higher CTCAE grade 3-4 infective (P<0.01) and hematological (P<0.01) adverse events were reported in the SCT group compared to control, with no difference for cardiovascular (P=0.92) and respiratory (P=0.24) events. More viral infection or reactivation events were reported in the SCT group compared to control (P=0.05): CMV (P=0.01), EBV (P=0.08), HSV (P=0.59). Importantly, subsequent malignancy was similar for both groups (P=0.96), 4-5%.
Conclusions. Our analysis has sufficient information size to indicate that autologous SCT for patients with SSc and internal organ involvement is safe and efficacious. The magnitude of the reduction in long-term mortality of autologous SCT over other available therapies may be practice changing.
No relevant conflicts of interest to declare.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal