Addition of navitoclax to ongoing ruxolitinib treatment in patients with myelofibrosis (REFINE): a post-hoc analysis of molecular biomarkers in a phase 2 study

Background: Primary analyses of cohort 1a of the REFINE trial showed that addition of navitoclax to ruxolitinib induced a 35% or greater reduction in spleen volume (SVR₃₅) and reduced symptoms in patients with myelofibrosis no longer benefiting from ruxolitinib. Here, we report the exploratory post-hoc biomarker analyses from cohort 1a. Methods: REFINE is a phase 2, multicentre, open-label trial designed to assess the activity and safety of navitoclax alone or in combination with ruxolitinib in patients with primary or secondary (post-polycythaemia vera or post-essential thrombocythaemia) myelofibrosis. Cohort 1a of the study included patients who had disease progression or suboptimal response on stable ruxolitinib monotherapy. Patients in cohort 1a, who had previously received ruxolitinib for 12 weeks or more, continued their current stable dose, and navitoclax was orally administered at 50 mg per day and escalated weekly to a maximum of 300 mg per day, based on tolerability. The primary activity endpoint was SVR₃₅ at week 24 from baseline. Secondary endpoints were a 50% or greater reduction in total symptom score (TSS₅₀) at week 24 from baseline as measured by the Myelofibrosis Symptom Assessment Form (version 4.0), anaemia response assessed according to International Working Group-Myeloproliferative Neoplasms Research and European LeukemiaNet criteria, and change in grade of bone marrow fibrosis according to the European consensus grading system; and exploratory endpoints included overall survival and changes in inflammatory cytokines. Exploratory analyses investigated potential prognostic biomarkers of the benefit of navitoclax-based combination treatment, including bone marrow fibrosis and variant allele frequency, in patients with a suboptimal response to ruxolitinib. This study is registered with ClinicalTrials.gov (NCT03222609) and is ongoing. Findings: Between Nov 14, 2017, and April 10, 2019, 34 patients in cohort 1a received at least one dose of navitoclax plus ruxolitinib. 23 (68%) patients were male, with 32 (94%) being White. At data cutoff (May 6, 2021), the median follow-up for survivors was 26·2 months (IQR 21·9–32·3). 33 patients were evaluable for biomarker analyses; 19 (58%) had high molecular risk mutations. Five (31%) of 16 patients had SVR₃₅ at week 24 in the high molecular risk group, as did four (31%) of 13 in the non-high molecular risk group. Four (36%) of 11 patients in the high molecular risk group had TSS₅₀ at week 24 compared with two (25%) of eight in the non-high molecular risk group; seven (39%) of 18 in the high molecular risk group had an improvement in fibrosis by at least one grade compared with five (36%) of 14 in the non-high molecular risk group; and four (28%) of 14 had reductions in variant allele frequency of 20% or greater in the high molecular risk group compared with two (17%) of 12 in the non-high molecular risk group. Patients with improvements in fibrosis of one grade or more and a reduction of 20% of more in variant allele frequency had improved overall survival (median overall survival not reached) compared with those who did not achieve fibrosis improvement or a reduction in variant allele frequency (median overall survival 28·5 months [95% CI 19·6–not estimable] for both), suggesting potential disease modification. Additionally, changes in concentrations of β-2-microglobulin (week 12: r=0·57; week 24: r=0·57), TIMP metallopeptidase inhibitor 1 (week 12: r=0·47; week 24: r=0·54), TNF receptor type II (r=0·55; week 24: r=0·40), and vascular cell adhesion molecule-1 (r=0·58; week 24: r=0·50) were positively associated with changes in spleen volume. Interpretation: These biomarker analyses reveal clinically meaningful splenic responses independent of high molecular risk mutation status in patients treated with navitoclax plus ruxolitinib who were not benefiting from ruxolitinib monotherapy. Furthermore, the overall survival benefit observed in those with an improvement in fibrosis or a reduction in variant allele frequency is suggestive of disease modification, implying the therapeutic potential of adding navitoclax to ruxolitinib for patients with myelofibrosis who had disease progression or suboptimal response to ruxolitinib monotherapy. Funding: AbbVie.