CHICAGO – Subgroup categorization of patients with severe scleroderma by their gene-expression profile correlated with responses to a newly proven treatment for the disease that involves myeloablation and autologous hematopoietic stem cell transplantation (HSCT).
Patients who fell into the “fibroproliferative” scleroderma subgroup, roughly one-third of patients enrolled in the treatment study, showed a high level of benefit from myeloablation and autologous HSCT,, said at the annual meeting of the American College of Rheumatology.
In contrast, the roughly one-third of patients in the study with a gene-expression profile that placed them into the “normal-like” subgroup had outcomes that closely matched the normal-like patients in the control group, who were treated with cyclophosphamide, which suggests that the normal-like patients are probably not good candidates for HSCT, said Dr. Whitfield, a professor of molecular and systems biology at the Geisel School of Medicine at Dartmouth in Hanover, N.H.
Thisstarts to get at the question of “How do you do personalized medicine in a disease like scleroderma?” he explained. “HSCT may be a game changer for patients with the fibroproliferative type of scleroderma,” who did relatively poorly in the control group of the trial when they received cyclophosphamide. Categorization of patients by their gene-expression profiles is a way to find order among scleroderma patients in what is otherwise “a very heterogeneous disease, where some patients improve on a treatment and others do not,” Dr Whitfield said.
The study used data collected in the(Scleroderma: Cyclophosphamide or Transplantation) trial, which enrolled 75 patients with severe scleroderma at any of 26 sites in the United States and Canada. SCOT compared the safety and efficacy of myeloablation followed by autologous HSCT with that of treatment with cyclophosphamide, and it followed patients for a median of 54 months. The results showed that overall HSCT was superior for the primary endpoint and several secondary endpoints, including event-free survival, which was 79% after HSCT and 50% in control patients by the end of follow-up ( ).
Dr. Whitfield’s group took peripheral blood cells from 30 of the patients treated by HSCT and from 33 of the patients treated with cyclophosphamide per protocol and analyzed the gene-expression profiles of the cells to categorize patients into the gene-expression subtypes of scleroderma that had been previously defined by Dr. Whitfield and his associates: fibroproliferative, inflammatory, limited, or normal-like (
Analysis of event-free survival out to 6 years following enrollment showed that, in the fibroproliferative subgroup, roughly 90% of patients treated with HSCT remained alive and event free, compared with about 35% of the cyclophosphamide patients, a highly statistically significant difference. In the inflammatory subgroup, event-free survival persisted in about 90% in the HSCT recipients, compared with about 50% of those in the control arm, a difference that did not reach statistical significance. Among patients with normal-like gene expression, the event-free survival rate was about the same regardless of treatment, about 60% in each treatment arm. The results suggest that patients with normal-like disease “are probably not good candidates for a treatment as intensive as HSCT,” Dr. Whitfield said in an interview.
Although the HSCT and cyclophosphamide treatment groups included relatively small numbers of patients, when the researchers subdivided the trial cohort into three different scleroderma types, the analysis remained “powered well enough to see a difference; the difference was very clearly statistically significant,” Dr. Whitfield declared.
“Now that we have a treatment [HSCT] to tie to the [gene-expression analysis], we can think about using this in routine practice,” he concluded.
Dr. Whitfield is a cofounder of Celdara, and he has been a consultant to Bristol-Myers Squibb, Corbus, UCB, and Third Rock Ventures.
SOURCE: Franks J et al. Arthritis Rheumatol. 2018;70(suppl 10): Abstract .