Background Allogeneic hematopoietic cell transplant (alloHCT) remains the only curative therapy for myelofibrosis (MF). Consensus-based definitions of disease response from the International Working Group-Myeloproliferative Neoplasms Research and Treatment/European LeukemiaNet (IWG-MRT/ELN) were developed for pharmacological clinical trial assessments and are utilized by the Center for International Blood and Marrow Transplant Research (CIBMTR) for disease response reporting post-alloHCT (Tefferi et al, 2013).

Given the biologic and clinical distinctions between pharmacologic therapy and allo-HCT, it is logical to hypothesize that the markers of disease response and predictors of long-term success after transplant may differ from those used in the non-transplant setting.

In this study, we sought to retrospectively assess post-transplant response based on clinical, molecular and morphological findings - referred as clinical-Complete Response (CR) - and compare these to the IWG-MRT/ELN criteria defined responses.

Methods This was a single-center retrospective cohort study of patients who underwent allo-HCT for primary or secondary myelofibrosis between January 1, 2018 and June 30, 2023.

We extracted clinician-documented response assessments from the medical record and compared these with responses per IWG-MRT/ELN criteria at 6, 12 and 24 months post-transplant.

Results 57 patients underwent first alloHCT indicated for MF with a median age of 63 years (range 29 – 78 years). High molecular risk mutations were present in 44% (n=25).

At 24 months post-alloHCT, 45/57 (79%) patients remained alive. 31/57 (54%) patients were deemed by their treating clinician to be in remission (clinical-CR). However, only 51% (n = 16/31) of these met IWG-MRT/ELN CR criteria. The remaining 15 patients were classified by IWG-MRT/ELN as Partial Response (n=14) or Clinical Improvement (n=1) due to abnormal marrow cellularity (n=8), persistent thrombocytopenia with platelet count in the range 50-100 x10^9/L (n=5) or could not be classified as CR due to lack of bone marrow morphology reporting.

Among the 16 patients who ultimately met IWG-MRT/ELN CR criteria at 2 years post-transplant, only 25% (N=4) were in CR according to the IWG-MRT/ELN criteria by 6 months, and 50% (N=8) were in CR by 12 months post-alloHCT. Reasons for failing to meet IWG-MRT/ELN CR criteria by 6 months in this cohort included abnormal marrow cellularity (n=8), persistent splenomegaly (n=2) and moderate cytopenias (n=2). By 12 months, 5 patients failed to meet CR due to abnormal marrow cellularity, 1 due to abnormal cellularity plus persistent splenomegaly, 1 due to moderate thrombocytopenia and 1 due to abnormal cellularity plus moderate thrombocytopenia. Among the 31 patients who were in clinical-CR by 24 months, 58% (n=18) were in clinical-CR by 6 months, and 100% (n=31) by 12 months.

All patients who met IWG-MRT/ELN CR by 24 months achieved molecular remission of driver mutations and cytogenetic remission by 6 months. All patients in the clinical-CR group had achieved molecular driver mutation remission by 6 months. However, 4 of these patients demonstrated new non-driver mutations at 6 months that persisted through to 24 months with stable variant allele frequency.

Conclusion This study highlights discrepancies between IWG-MRT/ELN and clinician-defined criteria of successful outcomes after alloHCT in patients with MF. Notably, all patients who ultimately achieved clinical- or IWG-MRT/ELN-defined CR by 24 months had reached molecular remission of driver mutations by 6 months, underscoring the value of early molecular assessment in response evaluation. Many experienced delays in meeting morphological and laboratory parameters required by IWG-MRT/ELN, despite being destined for CR by 24 months. These findings call into question the applicability of IWG-MRT/ELN criteria in the post-transplant setting.

Future efforts should aim to prospectively validate transplant-specific response criteria that incorporate molecular response, transfusion independence, chimerism, and durable count recovery. Such frameworks may better predict long-term outcomes, inform surveillance strategies and personalize post-transplant care. This is particularly important since interventions to treat relapse post-alloHCT often rely on augmenting graft vs leukemia effect which carries significant risks such as graft vs host disease.

This content is only available as a PDF.
2025
Sign in via your Institution