An exploratory analysis of myelofibrosis (MF) patient subgroups by baseline hemoglobin levels in the gecacitinib phase 3 trial.

6571 Background: Anemia is a key prognostic indicator of MF. In the double-blind, randomized phase 3 ZGJAK016 trial, gecacitinib (GCA), a dual JAK/ACVR1 inhibitor, demonstrated superior spleen response over hydroxyurea, and a trend toward improvement in constitutional symptom and anemia associated with MF in JAK inhibitor-naive patients (pts). To further elucidate the impact of GCA on anemia, we conducted a post-hoc analysis of the data from this trial, examining outcomes in relation to the severity of anemia. Methods: Pts in the GCA group were categorized post subgroups based on their baseline hemoglobin (Hb) levels: less than 100 g/L (moderate to severe anemia), 100 g/L to lower limit of the normal (LLN) (mild anemia), LLN to upper limit of the normal (ULN) (normal), and more than ULN for (Hb elevated). The primary endpoint was the proportion of pts with a spleen volume reduction of ≥ 35% from baseline (SVR35) at week (wk) 24. Secondary endpoints included the proportion of pts with a ≥ 50% reduction in Total Symptom Score (TSS50), and transfusion independence (TI) rate at wk 24 (no red blood cell transfusions and no Hb levels of <80 g/L in the last 12 wks before wk 24). Results: Of all the 71 pts randomly assigned to GCA in the intent-to-treat (ITT) population, 47 (66.2%) were moderately/severely anemic at baseline (including 16 pts [22.5%] with severe anemia [Hb levels of < 80 g/L]), 11 (15.5%) were mildly anemic and 13 (18.3%) were nonanemic (including three pts [3.8%] with Hb levels of > ULN). In the moderately/severely anemic subgroup, a higher proportion of pts were classified as DIPSS high risk, transfusion dependent, and had a diagnosis with primary MF at baseline. Most pts in the mildly anemic group were TI, as were all in the nonanemic subgroup. The three pts with elevated Hb all had post- polycythemia vera MF. Mean Hb levels increased by wk 2 across all GCA subgroups, then remained stable in the anemic subgroups, or slightly decreased but still > 110 g/L in the normal subgroup and markedly decreased in the Hb elevated subgroup. Mean platelet counts decreased by wk 2 except in the Hb elevated subgroup and then maintained stability. In the moderately/severely and mild anemic subgroup, 24 of 36 (66.7%) pts who were TI at baseline maintained this status at wk 24 and 8 of 22 (36.4%) who were non-TI at baseline achieved TI at wk 24. The SVR35 rates and TSS50 rates at wk 24 were comparable across subgroups and consistent with those observed in the ITT population. Conclusions: In summary, GCA delivers a threefold benefit in terms of spleen and symptom management, as well as anemia, to JAK inhibitor-naive pts with MF who have mild, moderate/severe, or no anemia at baseline, particularly those with Hb levels below LLN. Clinical trial information: NCT04617028 . ITT(n =71) Hb <100 g/L(n =47) Hb ≥100 g/L to LLN(n =11) LLN to ULN(n =10) >ULN(n =3) SVR35 rate at wk 24 64.8% 59.6% 81.8% 70.0% 66.7% TSS50 rate at wk 24 62.0% 57.4% 90.9% 50.0% 66.7% TI rate at wk 24 60.6% 48.9% 81.8% 90.0% 66.7%