Markers of Responsiveness to ATG/CsA Therapy in Patients with Severe Aplastic Anemia.

While ATG/CsA therapy results in a significant response rate in severe aplastic anemia (sAA), a substantial minority of patient remains refractory. Lack of response to intense immunosuppression (IS) may be a consequence of “non -immune” pathogenesis or be due inadequate intensity/duration of IS or exhaustion of stem cell reserves. Identification of markers of IS responsiveness would allow for a better patient selection and early application of stem cell transplantation (SCT). Various parameters of immune activation have been investigated but suitable makers of responsiveness to IS were not identified. Previously, we have shown a potential utility of flow cytometric analysis of T cell VB repertoire to identify expanded T cell clones that are involved in the autoimmune attack on hematopoietic progenitor and stem cells in immune-mediated bone marrow failure. Among 110 AA cases treated in our institution, 88 had sAA; 12 of them underwent early SCT while 74 were initially treated ATG/CsA. Of those patients, 60 received equine ATG (H-ATG) as a primary (N=56), or salvage modality (N=4) after rabbit ATG (R-ATG) with overall response rate (defined by transfusion independence and not fulfilling severity criteria) of 65% (66% as a first line). R-ATG was used in 23 patients as an initial (N=15) or salvage treatment (N=7) with overall response rate of 60% (61% as a first line, P=.88 as compared to H-ATG). For 24 patients (3 with mAA) we performed serial flow cytometric VB utilization profiling within CD3 CD4 and CD8 cells. All of these patients were minimally transfused and not infected at the time of initial testing. Prior to therapy, VB expansions (≥2 VB families) within CD4 and CD8 cells (defined as >mean+2xSD value for a given VB family in controls) indicative of the presence of potentially pathogenic T cell clones were present in 7/21 and 13/21 patients, respectively, P=.26. Nominally, CD4 and CD8 VB overrepresentations were 7% >mean+2xSD (1.7%-17.3%) and 9.26% (1.8%-24.5%), respectively. Within this specific sub-cohort of patients, the overall response rate measured at 6 months was 14/19 (73%). Response correlated with the presence of VB expansions prior to therapy; responders showed on average more and larger CD8 expansions (2 VB/patient, 0–3; 6.16% av. size, 1.8%-24.5%) and larger expansions vs. lower number of and smaller CD8 expansions (0.6 VB/patient, 0–3; 1.7% av. size, 5.7–11.4%) were seen in refractory patients (only 1/5 refractory patients had VB expansions). Conversely, 10/14 (71%) responses were seen in patients in with ≥2 CD8 VB expansion, only 1 patient responded but had no CD8 VB expansions. In contrast, only 1/5 (20%) ≥2 CD8 VB expansion were seen in refractory patients. Remarkably, hematologic improvement was associated with resolution of at least one significant VB expansion during monthly evaluations. In patients who relapsed reappearance of the original or new VB expansion within CD8 cells was seen. We conclude that CD8 VB expansions involving ≥2 VB families may predict of subsequent response to IS suggesting that VB typing may be a suitable monitoring tool for patients with immune-mediated AA.