Early Tapering of Immunosuppression in MRD Positive Myelofibrosis Patients Induces Molecular Response and Excellent Outcome after Hematopoietic Stem Cell Transplantation

Background Driver mutations (JAK2, CALR, and MPL) are the pathophysiologic hallmark of myelofibrosis. Around 90% of myelofibrosis patients harbor one of the three mutations, which can be measured in peripheral blood. Driver mutation clearance on day 30 after allogeneic hematopoietic cell transplantation (HCT) is a strong predictor of disease progression and overall survival. However, a significant number of patients can achieve mutation clearance even 3 or 6 months after HCT, and in contrast to other hematological malignancies, the time between driver mutation detection and clinical progression is significantly longer, allowing to plan interventions. Whether early tapering of immunosuppression in patients at high risk of disease progression can improve patient outcomes is unknown. Here, we evaluated the role of preemptive therapy with early immunosuppression tapering for patients with persistent molecular disease at day 30 after HCT. Methods All patients received a homogenous HCT platform with reduced intensity conditioning and anti-T-lymphocyte globulin. Post-transplant graft-versus-host disease (GVHD) prophylaxis consisted of mycophenolate mofetil (until day 28) and cyclosporine A, which was tapered preemptively after day 30 within 2 months driven by cyclosporine levels and molecular monitoring, targeting complete molecular clearance. Patients had to be GVHD-free at time of tapering start. If a patient developed GVHD during tapering, immunosuppression was increased according to current recommendations. Primary endpoint was molecular remission after tapering start at any time point after day 30 post-transplant. Secondary endpoints were the incidence of acute GVHD within 100 days after tapering start, incidence of relapse and non-relapse mortality, and overall survival. Results This study included 55 patients with myelofibrosis (62% primary and 38% secondary myelofibrosis), with a median age of 59 years, and with persistent driver mutation positivity at day 30 after HCT. Median time from HCT to start of immunosuppression tapering was 51 days (range, 25-97). 41 patients had JAK2, 12 patients had CALR, and 2 patients had MPL driver mutation genotype. The median mutation burden before HCT was significantly higher for CALR (45.2%) and MPL (75.7%) compared with JAK2 (33.2%; P<0.001). The median mutation burden at day 30 after HCT was similar between the driver mutations (P=0.50), being 1% for JAK2, 0.9% for CALR, and 0.2% for MPL. In terms of outcomes, molecular remission was achieved in 35/55 patients (64%). Median time to remission was 67 days (range, 10-168 days). Mutation burden at day 30 after HCT was associated with molecular remission after preemptive therapy, showing higher likelihood of remission for patients with lower mutation burden (beta coefficient 0.56; P=0.046). Acute GVHD grade II-IV was observed in 43% (95% CI, 27-59%) of patients achieving molecular remission after tapering, within a median of 63 days. Incidence of severe acute GVHD grade IV in responders was 11%. All patients without remission did not develop acute GVHD. Median follow-up was 4.4 years from time of tapering start. In terms of relapse and non-relapse mortality, cumulative incidence was 6% (95% CI, 0-17%) and 9% (95% CI, 0-18%) for patients with molecular remission after preemptive tapering. The only 2 deaths without relapse in patients with molecular remission occurred early after development of GVHD (grade III and IV, respectively). Overall survival was 91% (95% CI, 82-100%) for patients with molecular remission versus 39% (95% CI, 14-64%) for patients without response after tapering. Lastly, we compared patients achieving molecular remission after tapering in matched analysis with 112 patients (out of a total cohort of 324) who were always driver mutation negative from day 30 until day 180 post-transplant, finding similar disease-free and overall survival. Conclusion This is the first study to show that early tapering of immunosuppression as a means to target molecular response can overcome initial poorer prognosis of patients with persistent driver mutations at day 30 post-transplant in myelofibrosis.