Cytokine Modulation Correlates Strongly with Symptom Improvement in Patients with Myelofibrosis Treated with Nuvisertib (TP-3654), an Investigational Selective PIM1 Kinase Inhibitor

Background: PIM1 expression is upregulated in myelofibrosis (MF) and has been shown to modulate cytokine induced PI3K/AKT and JAK/STAT pathways. Clinical features of MF are partly attributed to circulating cytokines, a subset of which have been shown to correlate with poor prognosis. Nuvisertib (TP-3654), an oral investigational highly selective PIM1 kinase inhibitor, is being evaluated in a phase 1/2 study in patients (pts) with relapsed/refractory MF. In preclinical murine MF models, TP-3654 alone and in combination with ruxolitinib, showed reduced spleen size, bone marrow (BM) fibrosis and, importantly, cytokine response genes.

Methods: Data were collected from the ongoing global phase 1/2 study evaluating the safety and efficacy of TP-3654 monotherapy in pts with MF (NCT04176198, Arm 1). Key eligibility included primary or secondary MF; previously treated with or ineligible for JAK inhibitor treatment; DIPSS intermediate 1, 2 or high-risk MF; platelet ≥25x10⁹/L; splenomegaly (≥450 cm³ by imaging); and ≥2 measurable symptoms per MFSAF v4. We conducted a comprehensive analysis of 64 cytokine changes in plasma from pts treated with TP-3654 after ≤4-week (wk) intervals to Wk 12, and 8-wk intervals past Wk 96. Individual cytokine longitudinal analysis and a modified Kaplan-Meier (KM) assessment was conducted to predict symptom score improvement.

Results: As of June 10, 2024, 65 pts were enrolled in 5 dose levels. At baseline, 52% of pts were anemic (Hgb <10 g/dL), 55% had low platelet counts (<100x10⁹/L), 35% had both anemia and thrombocytopenia, and 26% had high molecular risk mutations. In pts treated with 720 mg QD or BID dose levels for ≥12 weeks, ≥25% spleen volume reduction (SVR) was observed in 31% of pts (SVR25 response) and ≥50% reduction in symptoms in 50% of pts (TSS50 response) at any time. The observed overall TSS50 response, as well as each individual symptom domain, were durable for ≥12 wks and showed a strong correlation with cytokine changes. Longitudinal analyses showed significant (p <0.0001) decreases in pro-inflammatory cytokines IL-18, MIP1β and PAI, among others, and an increase of the anti-inflammatory adipokine, adiponectin. Cytokine changes were observed within 1 day of TP-3654 treatment and continued past Wk 48, showing early- and long-term pharmacodynamic modulation.

A modified KM analysis showed changes in ENRAGE, PAI1, MIP1β and adiponectin levels at multiple timepoints, significant (p <0.0001) predicting Wk 12 and Wk 24 TSS50 response probability. Additionally, individual symptom improvement at Wk 12 and Wk 24 were linked to cytokine changes, such as fatigue and itching for ENRAGE, and fatigue, satiety and SVR for adiponectin. Individual cytokine-related symptoms analyzed at Wk 12 (baseline individual score ≥3 per MFSAF v4) were linked to cytokine changes: fatigue for IL1RA, EGFR, and CD40L; night sweats for TNF-RI, TIMP1, EPO and CD40L; and itching for MIP1β and MCP1; and similar correlations observed at Wk 24. In addition, 6 of 18 (33%) evaluable pts showed ≥1 grade reduction in BM fibrosis correlating with 3 of 6 (50%) showing anemia and platelet response, and 5 of 6 (83%) were on active treatment for ≥1 year. Reductions in TSS, spleen volume, inflammatory cytokines and BM fibrosis with hematological responses in those evaluable pts suggest potential disease modification.

Conclusions: The preliminary data of TP-3654 in relapsed/refractory MF pts showed promising clinical activity, including SVR, TSS improvement, and correlating cytokine modulation. ENRAGE and adiponectin levels were significant predictors of SVR, TSS50 responses and specific symptom improvement for fatigue, itching and satiety. Translational studies are ongoing to further explore potential disease modification activity of TP-3654.