Time without Transfusion Reliance or Anemia Worsening: A Novel Method for Integrating Transfusion Dependence, Anemia, and Survival with Myelofibrosis Therapies Based on the Phase 3 Simplify-1 and Simplify-2 Trials

Introduction: While approved Janus kinase inhibitors (JAKi) have improved overall survival (OS) in myelofibrosis (MF), traditional survival analyses do not consider individual patient (pt) experience, which may be negatively impacted by factors such as persistent symptoms, adverse events (AEs), and red blood cell (RBC) transfusion dependence. In particular, RBC transfusions, an important component of anemia management in MF with currently approved JAKi, are associated with negative quality of life (QOL), increased healthcare resource utilization, and poor prognosis.

Momelotinib (MMB), a JAK1/JAK2/activin A receptor type 1 inhibitor, has demonstrated consistent anemia benefits, including increased transfusion independence and reduced transfusion burden, as well as spleen and symptom improvements across 3 phase 3 trials in MF. In the SIMPLIFY-1 (NCT01969838) and SIMPLIFY-2 (NCT02101268) trials, achieving transfusion independence with MMB was associated with prolonged OS (Mesa R, et al. Leukemia. 2022). Here we apply time without transfusion reliance (TWiTR) analyses, integrating transfusion-dependent (TD) status and OS, to further characterize the benefits of MMB on survival quality.

Methods: Analyses were performed in the safety populations, by treatment arm, over the randomized 24-wk treatment period of SIMPLIFY-1 (n/N=430/432; MMB vs ruxolitinib [RUX], JAKi naive) and SIMPLIFY-2 (N=156; MMB vs best available therapy [BAT; 88.5% RUX], JAKi experienced).

TWiTR (Zeidner JF, et al. Haematologica. 2023) was based on the established time without symptoms and toxicity (TWiST) analysis, replacing toxicity with TD status. The analysis partitioned survival duration into 3 health states: (1) TD, defined as the number of days on which pts were TD (defined per protocol in the SIMPLIFY studies as requiring ≥4 units of RBC transfusions or having a hemoglobin level <8 g/dL in the previous 8 weeks) before experiencing a leukemia-free survival (LFS) event (leukemic transformation or death); (2) Post-LFS, defined as the number of days from an LFS event to death or censoring (OS; Post-LFS = OS − LFS); and (3) TWiTR, defined as the number of days without Post-LFS or TD (TWiTR = LFS − TD). An additional analysis incorporated the number of days on which pts had any-grade anemia events in the TD state (TD-Anemia [TD-An]) to define time without transfusion reliance or anemia worsening (TWiTR-An). Health state durations were estimated via Kaplan-Meier curves of the survival distribution function, with mean duration in each state calculated as the area under the curve.

Results: In SIMPLIFY-1, pts in the MMB arm spent less time in the TD state and more time in the TWiTR state vs pts in the RUX arm (Figure 1). Mean duration in the TD state was 28.3 vs 62.7 days (difference with MMB, −34.4 days), while mean duration in the TWiTR state was 163.4 vs 131.1 days with MMB vs RUX (difference with MMB, 32.3 days). Fewer pts in the MMB arm had treatment-emergent anemia AEs, and incorporation of anemia events into the analysis increased the mean time spent in the TD-An state compared with the TD state in both treatment arms, although this increase was smaller in the MMB arm, at 32.0 vs 74.8 days in the RUX arm (difference with MMB, −42.8 days). Mean duration in the TWiTR-An state was 159.7 vs 119.0 days with MMB vs RUX (difference with MMB, 40.7 days).

Similar results were observed in SIMPLIFY-2, although incorporating anemia events had less impact. Mean duration in the TD state was slightly shorter with MMB vs BAT, at 69.4 vs 89.9 days (difference with MMB, −20.5 days); mean duration in the TD-An state was 69.6 vs 89.9 days (difference with MMB, −20.3 days). Consistently, mean duration in the TWiTR state was slightly longer with MMB vs BAT, at 118.3 vs 95.9 days (difference with MMB, 22.4 days); mean duration in the TWiTR-An state was 118.0 vs 95.9 days (difference with MMB, 22.1 days).

Conclusions: Pts treated with MMB spent less time reliant on RBC transfusions and more time free from transfusions and anemia events, without the cost of increased AEs, compared with those treated with RUX or BAT. Given the negative QOL and prognostic impact of RBC transfusion dependence, these analyses suggest that the anemia benefits of MMB lead to improved quality of survival for pts with MF.