An integrative prognostic model for transplant outcomes in myelofibrosis

Allogeneic hematopoietic cell transplantation (allo-HCT) is the only curative option for patients with myelofibrosis, yet transplant outcomes remain heterogeneous, with survival influenced by a complex interplay of disease-related, molecular, and transplant-specific factors.

Existing prognostic tools were primarily designed for pre-transplant risk stratification while transplant-specific systems lack molecular granularity and thus may not fully capture variables critical to post-transplant outcomes. Furthermore, most models were based on primarily patients without JAK inhibitor exposure. To address this gap, the authors conducted a large international multicenter study using up-to-date machine learning approaches to develop a novel, comprehensive prognostic model in the modern era of molecular analysis and JAK inhibition.

The model aims to more accurately predict overall survival and transplant-related risk in patients with myelofibrosis undergoing allo-HCT, thereby guiding clinical decision-making and improving individualized risk assessment. The authors analyzed 1,258 patients with myelofibrosis in the era of JAK inhibition undergoing allo-HCT across multiple international centers (60% primary, 40% secondary myelofibrosis). Clinical, molecular, and transplant-related variables were collected.

The dataset was randomly split into a 60% training cohort and a 40% independent validation cohort. Recursive partitioning was used to identify potential hierarchies of factors such as age and leukocyte counts. To develop an integrated prognostic model for overall survival (OS) and non-relapse mortality (NRM), the authors applied both penalized Cox regression and ensemble survival methods. Variable selection and model tuning were performed via cross-validation in the training set. Final model performance was assessed in the validation cohort using concordance indices and calibration metrics. The best-performing model was selected based on predictive accuracy and clinical applicability.

Older age (consistent cutoff identified by non-supervised recursive portioning was 60 years, HR 1.39; p<0.001) and poor performance status (HR 1.52; p<0.001) were independently associated with reduced OS and NRM. Anemia, thrombocytopenia, and both severe leukocytosis and leukopenia predicted OS andNRM. Other disease-related risk factors were shorter time to secondary myelofibrosis transformation (continuous HR 1.05; p<0.001). Looking at cytogenetics, regularization models identified complex karyotype as the only predictive factor (HR 1.51; p<0.001).

On a molecular level, CALR/MPL mutations were protective, while ASXL1, TP53, U2AF1, and RAS pathway mutations conferred high-risk features. In terms of donor relations, haploidentical (HR 1.49; p<0.001), and mismatched unrelated donors (HR 1.67; p<0.001) consistently showed worse OS and NRM compared with matched related or unrelated transplants. An ensemble survival model was trained on 700 patients and validated in an independent cohort of 558 patients. Model performance was strong (C-index: 0.74 for OS, 0.68 for NRM in training; 0.70 and 0.64, respectively, in validation), with good calibration. The model stratified patients into 4 risk groups based on predicted OS. Five-year OS and NRM rates were 90% and 7% (low risk), 71% and 20% (intermediate), 46% and 34% (high), and 25% and 52% (very high risk) (p<0.001). The 4-tiered model outperformed existing prognostic tools (disease-related: DIPSS, MIPSS70 and MIPSS70v2.0; and transplant-related: MTSS, EBMT scores) by incorporating transplant-specific and molecular risk features.

This improved and integrative myelofibrosis transplant scoring system (iMTSS) supports more personalized transplant risk assessment and informed clinical decision-making. The authors developed and validated a robust, integrative myelofibrosis transplant scoring system (iMTSS) undergoing allo-HCT, incorporating clinical, molecular, and transplant-specific variables. The model effectively stratifies patients into four distinct risk groups with significantly different survival outcomes, outperforming existing pre-transplant scoring systems.

This tool enables more precise, individualized risk assessment and has the potential to guide transplant decision-making, donor selection, and post-transplant management strategies in clinical practice. A web-based calculator will be developed and presented at the meeting.

Impact of ruxolitinib on corticosteroid treatment patterns in 1147 patients with chronic graft-versus-host disease in real-world practice in the United States: A long-term follow-up analysis

Ruxolitinib (RUX) is an oral, selective Janus kinase (JAK)1/JAK2 inhibitor approved for chronic graft-versus-host disease (cGVHD) after failure of 1–2 lines of systemic therapy in patients ≥12 years of age. Treatment patterns of RUX for cGVHD have evolved since approval, and real-world treatment characteristics and impact on corticosteroid (CS) management need further description.

The authors report long term data describing real-world characteristics in a large cohort treated with RUX to understand treatment patterns and impact on CS dosing in patients with cGVHD after allogeneic hematopoietic stem cell transplantation (allo-HSCT) in the United States.

This retrospective analysis of administrative claims data included commercial, Medicare/Medicare Advantage, and Medicaid health plan members of any age who received allo-HSCT, had cGVHD diagnosis, received RUX as cGVHD therapy (≥1 claim for RUX between July 1, 2021, and September 30, 2023; earliest claim was the index date), and had health insurance coverage ≥6 months before and ≥6 months after index (or less due to death). Follow-up was until earliest of death, end of continuous plan enrollment, or end of study period. Prescription refills in pharmacy claim records were used to determine RUX and prednisone-equivalent CS dosing and treatment length.

Line of therapy for cGVHD medications was based on claims for CS, belumosudil, abatacept, alemtuzumab, RUX, etanercept, hydroxychloroquine, ibrutinib, imatinib, interleukin-2, methotrexate, mycophenolate mofetil, pentostatin, and rituximab after cGVHD diagnosis. Calcineurin inhibitors (CNIs), mTOR inhibitors, and extracorporeal photopheresis (ECP) were measured separately from the line of therapy. This analysis included 1147 patients who received RUX for cGVHD. Median (IQR) age was 52 (31– 62) years (<18 years, 13.0%). Overall, 56.9% were male, 51.5% had commercial insurance, 31.0% had Medicaid, and 17.4% had Medicare. Patients were followed for a median (IQR) of 458 (254–740) days from initiating RUX. Patients initiated RUX a median (IQR) of 90 (20–256) days from cGVHD diagnosis (70 days in patients aged <18 years; 96 days in adults). Overall, RUX was administered as first-line therapy with/without CNIs/mTOR inhibitors in 29.0% (n=333) of patients, and second-line or later therapy in 71.0% (n=814). RUX monotherapy with/without CS was the most common regimen in patients receiving RUX as first- (84.7%; n=282), second- (93.7%; n=680), or third-line (77.3%; n=68) therapy. Other common therapies during RUX treatment were CNIs (52.2% of patients), mTOR inhibitors (14.5%), and ECP (12.6%). Median (IQR) RUX starting dosage was 10 mg/day (10–20) overall, in patients aged <18 years (5–10), and in adult patients (10–20). Among patients who had a RUX refill, 59.5% (n=615/1034) had a dose change; of those, 51.5% (n=317) had an increase and 48.5% (n=298) had a decrease in their first dose change.

Median (IQR) time to first dose change was 71 (29–188) days. Kaplan-Meier analysis showed median (95% CI) length of RUX treatment was 242 days (218–279); 40.9% of patients remained on RUX at 1 year and 21.3% at 2 years Median (IQR) CS dosing was 40.0 (20.0–70.0) mg/day pre-index. By day 90, 71% of patients remained on CS; among these, median (IQR) CS dose decreased by 62.5% (15.0 [5.0–30.0] mg/day). By day 180, 55% had CS refill, with a median (IQR) CS dose of 15.0 (5.6–37.5) mg/day. By 360 days post-index, 35% had CS refill, with a median (IQR) dose of 17.5 (7.5–40.0) mg/day. Median (95% CI) time from earliest CS fill for cGVHD to discontinuation of CS (presence of a ≥90-day gap in supply) or reduction to low-dose (10 mg) CS was 77 (68–86) days overall, 84 (65–107) days in pediatric patients, and 74 (65–86) days in adult patients. By the end of 1-year follow-up, discontinuation or reduction to low-dose CS was achieved in 90% of patients (pediatrics, 87%; adults, 91%).

Patients treated with RUX for cGVHD in real-world clinical practice remained on treatment for a median of 8 months; 40.9% remained on RUX for 1 year and 21.3% for 2 years, suggesting long-term persistence of safety and ongoing clinical benefit. Median time from earliest CS fill for cGVHD to discontinuation or reduction to low-dose CS was 77 days. By the end of 1-year follow-up, 90% of patients achieved discontinuation or reduction to low-dose CS. These outcomes support RUX as a long-term and CS-sparing treatment for cGVHD.

Peri-transplant ruxolitinib therapy significantly improves GVHD-free, relapse-free survival rates in transplantation for myelofibrosis

Myelofibrosis (MF) is an advanced haematological malignancy manifested by the presence of splenomegaly, B-symptoms and cytopenias. The only curative approach is allogeneic stem cell transplantation (HSCT) however outcomes are significantly restricted by high rates of graft failure and graft-versus-host-disease (GVHD). Following changes to local conditioning protocols based upon clinical trial data (Ali et al, Blood Advances 2022), the authors evaluated peri-transplant predictors of long-term success in a uniformly managed cohort. Consecutive patient records were extracted from a comprehensive single-centre database. Key variables included high resolution HLA-matching for DP permissivity based upon the T-cell-epitope model, peritransplant ruxolitinib use during conditioning, genetic risk scoring and stem cell dose. Multivariate Cox and Fine-Gray models were used to assess factors influencing overall survival (OS), progression-free survival (PFS), GVHD-free relapse-free survival (GRFS), and cause-specific mortality.

37 patients receiving transplants between 2008-2023 were identified. Median age at transplant was 54 (range 32-71), and 65% were male. 68% (25/37) of patients had primary myelofibrosis and 45% had a JAK2 driver mutation (15/33). Most patients were DIPSS+ intermediate-2 (68%) or high risk (19%). 54% received ruxolitinib pre-transplant (20/37), of whom 35% (7/20) continued ruxolitinib 5mg twice daily during transplant conditioning until engraftment and then ceased treatment. Most patients (86%) received fludarabine/busulfan-based reduced intensity conditioning, with ATG-based GVHD prophylaxis in all but 3 cases. 12 patients received matched sibling donors, 21 10/10 matched unrelated donors, 3 9/10 HLA-matched and 1 haploidentical transplant. 95% of cases (35/37) achieved day 28 neutrophil engraftment, with 2 cases of primary graft failure. Median time to neutrophil and platelet engraftment was 17 and 28 days respectively. Poor graft function was seen in 33% of patients (12/36), defined as 2-3 cytopenias lasting over 2 weeks after day +28 in the presence of donor chimerism >5%, requiring blood or platelet support or GCSF/EPO. At a median follow-up of 5 years, median survival post-HSCT was 17 months. One-, two-, and three-year OS were 66%, 55% and 52% respectively. GRFS at 1 and 3 years was 24% and 17%, highlighting the high burden of GVHD and relapse.

In multivariate models, a CD34+ stem cell dose >7 x 10^6/kg was associated with a reduced risk of death (HR 0.28, p=0.03), while HCT-CI comorbidity ≥3 was associated with a higher risk of death (HR 4.19, p<0.01), driven by increased non-relapse mortality risk (p=0.04). Combined donor-recipient CMV seropositivity predicted improved OS and PFS compared with other serotypes (HR 0.12 & 0.23, p=0.02 for both). Letermovir prophylaxis had no independent impact. The authors then focused on GRFS as a valuable composite endpoint particularly relevant for myelofibrosis outcomes, encompassing survival without severe grade III-IV acute GVHD, chronic GVHD requiring systemic therapy, relapse or death. No benefit was noted for high resolution HLA-matching for DP permissivity based upon the T-cell-epitope model. Continuation of ruxolitinib during conditioning was the only variable significantly associated with improved GRFS (HR 0.29, p=0.03), with 1-year GRFS of 42.9% vs 15.4% in those stopping ruxolitinib pre-conditioning (p=0.02). No significant interaction was seen for transplant year, engraftment kinetics or infection risk. One-year OS for patients continuing vs ceasing ruxolitinib was 85.7% and 61.5% respectively (p≥0.05).

This long-term real world analysis underscores the substantial transplant-related morbidity in myelofibrosis. Major barriers to long-term safety and quality of life include high rates of GVHD, immunosuppression-related morbidity and poor graft function. Crucially, these results demonstrate that peri-transplant ruxolitinib continuation appears to significantly improve composite outcomes such as GRFS without compromising engraftment. These findings support ongoing prospective evaluation of JAK inhibition as part of transplant preparation and emphasise the need for optimised donor selection and graft composition in MF.