GB2064 (lenumlostat) shows preliminary evidence of bone marrow collagen fibrosis reduction with manageable tolerability in JAK inhibitor-refractory myelofibrosis: results from the MYLOX-1 phase IIa study

Myelofibrosis is a rare myeloproliferative neoplasm with limited treatment options beyond JAK inhibitors, which provide symptomatic benefit and adequate spleen volume reduction but do not consistently modify underlying disease, including bone marrow fibrosis (BMF).¹ Progressive BMF drives cytopenias and poor prognosis.² Lysyl oxidase-like 2 (LOXL2) is a copper-dependent amine oxidase that crosslinks extracellular matrix collagens, contributing to tissue fibrosis.³ LOXL2 expression is elevated in the bone marrow of patients with primary myelofibrosis but absent in healthy tissue.^4,5 GB2064 is a selective LOXL2 inhibitor with preclinical antifibrotic effects.⁶ We report results from MYLOX-1 (NCT04679870), a phase II study evaluating GB2064 in participants with primary or secondary myelofibrosis who were refractory, intolerant, or ineligible for JAK inhibitor therapy.

This open-label, multicenter study enrolled 18 adults with confirmed primary or secondary myelofibrosis (N=9 each), predominantly intermediate-2 risk (67%), and Eastern Cooperative Oncology Group (ECOG) performance status 0-2 (Table 1; Online Supplementary Figure S1). The study was conducted in accordance with the Declaration of Helsinki, approved by relevant institutional review boards/ethics committees, and all patients provided written informed consent. Participants included a mix of JAK inhibitor–naïve and previously treated patients, with the majority having refractory or intolerant disease. All harbored at least one driver mutation (JAK2^V617F 67%, CALR 22%, MPL 22%), with 38% carrying high molecular risk mutations (ASXL1, EZH2, SRSF2, IDH). Participants received GB2064 1,000 mg twice daily orally for 9 months, with an optional extension allowing up to 3 additional years of treatment if participants derive clinical benefit. The selected dose provided greater than 90% inhibitory concentration drug levels at maximum concentration and for 16 hours per 24-hour dosing interval, supported by nonclinical and healthy volunteer data.

The primary endpoints were safety and tolerability. Secondary endpoints included centralized, blinded expert assessment of BMF by bone marrow biopsies at months 6 and 9, hematologic parameters, spleen volume by magnetic resonance imaging, and symptom burden using the MPN10 questionnaire (Online Supplementary Figure S3). Recruitment stopped when the predefined target of reduction of collagen fibrosis in at least three of 16 evaluable participants was exceeded.

Seven of the 18 participants completed the study, having remained on treatment for the full 9-month core phase. Twelve participants discontinued from the study due to progressive disease (N=3), withdrawal of consent (N=3), adverse events (N=3), death (N=1), lack of efficacy (N=1), and physician’s decision (N=1). One non-treatment-related fatality occurred at month 3 (sepsis with multiorgan failure). The mean duration of treatment was 172.8 days (median 207.0 days, range 4-288).

Table 1.Baseline characteristics of the study participants (N=18).

Among the ten participants evaluable for BMF at month 6, six (60%) demonstrated at least a one-grade reduction in bone marrow collagen fibrosis compared with baseline (Table 3; Online Supplementary Figure S4). Notably, this represents a reduction in type 1 collagen, the advanced, therapy-resistant fibrotic component required for the highest fibrosis grade (MF-3) and characteristic of malignant rather than reactive BMF.⁷ Two of the ten (20%) showed at least a one-grade reduction in reticulin fibrosis at month 9 (Table 3). All six participants with reduced collagen fibrosis exhibited disease stabilization with stable spleen volume and hematologic parameters over 9 months; none required red blood cell transfusions.

Clinical activity was demonstrated, with responses of modest magnitude. At month 6, clinical responses included one of 18 participants (6%) achieving an anemia response with a 1.8 g/dL hemoglobin increase and one participant (6%) achieving ≥35% spleen volume reduction. Additionally, two participants (11%) attained ≥50% reduction in total symptom score, and two participants (11%) demonstrated improved ECOG performance status. Overall, five participants were deemed by investigators to have derived clinical benefit: of these, four continued into the extension phase, with three maintaining treatment for 20-28 months.

GB2064 demonstrated bone marrow penetration by matrix-assisted laser desorption/ionization mass spectrometry (Online Supplementary Figure S2). Plasma pharmacokinetics showed stable concentrations with no additional accumulation beyond steady state. Target engagement was demonstrated with the mean free LOXL2 decreasing from 96.9% pre-dose to 46.6% at 2 hours post-dose. A high inter-individual variability in GB2064 concentration was observed between all analyzed tissues with a coefficient of variation of 89% (N=12). While GB2064 appears to have successfully penetrated the bone marrow, there was no conclusive concentration-response relationship linked to histopathological bone marrow changes.

Treatment-emergent adverse events (TEAE) occurred in 17 of 18 participants (94%) and were predominantly mild to moderate. Gastrointestinal disorders were most common (72%); these were generally self-limiting or responsive to anti-emetic therapy. Four gastrointestinal TEAE in three participants led to discontinuation: grade 3 diarrhea, grade 2 nausea (2 events), and grade 1 vomiting. Nine participants (50%) experienced a grade 3 or higher TEAE. One treatment-related serious adverse event (a fall in an elderly participant) occurred. No significant worsening of anemia or thrombocytopenia was observed. No clinically significant hepatic adverse events occurred except one case of transient grade 2 elevated gamma-glutamyl transferase and alkaline phosphatase (month 1, cause unknown, resolved while continuing treatment). During the extension phase, GB2064 was well tolerated with no serious adverse events and no new gastrointestinal events or transfusion requirements (Table 2).

Table 2.Safety overview for the main study and extension phase.

Table 3.Bone marrow fibrosis response and clinical outcomes.

Several design considerations warrant acknowledgment. First, the sample size of the study was small (N=18), while minimizing experimental exposure in this JAK inhibitor-refractory, intolerant and ineligible population. Second, while clinical responses were modest in magnitude and frequency, the finding that six out of ten evaluable patients achieved meaningful collagen fibrosis reduction, accompanied by disease stabilization and acceptable tolerability, provides encouraging preliminary evidence of disease-modifying potential in this treatment-refractory population and supports further investigation in adequately powered, controlled trials. The extension phase study design did not include bone marrow biopsies in order to prioritize collection of comprehensive safety data; consequently, the duration of any improvement in fibrosis was not evaluable. Fibrosis assessment was planned for the subsequent study with anticipated participant rollover.

The reduction specifically in collagen fibrosis is clinically significant and distinguishes GB2064 from existing therapies. While JAK inhibitors provide symptomatic benefits,^8,9 their effects on BMF are limited and inconsistent,¹⁰ particularly pertaining to collagen fibrosis, which represents an advanced, irreversible-appearing fibrotic change in myelofibrosis. The ability of GB2064 to reverse type 1 collagen deposition, not merely reticulin fibrosis, may suggest true disease modification rather than symptomatic palliation alone.

Emerging evidence suggests that BMF score changes may serve as a surrogate endpoint for disease modification, with worsening fibrosis correlating with shorter survival.^11-14 Disease stabilization in fibrosis responders, in whom progression would typically occur in this high-risk population, supports potential disease-modifying effects of LOXL2 inhibition.

In summary, GB2064 demonstrated acceptable safety and tolerability with manageable gastrointestinal adverse events and no significant cytopenias. Bone marrow penetration, target engagement, and reduction of collagen fibrosis in 60% of evaluable participants suggest potential disease-modifying activity. Despite a small sample size and modest clinical response rates, these findings support further investigation of LOXL2 inhibition in myelofibrosis, particularly in combination with JAK inhibitors or other agents.

Footnotes

• Received September 8, 2025
• Accepted February 24, 2026

Correspondence

R.F SCHLENK - Richard.schlenk@nct-heidelberg.de

Disclosures

HTS, BJ, VA, JAR, RJS, and BS were employed by Galecto Inc during the conduct of this study. JM has received research funding and consulting fees from multiple pharmaceutical companies, including Galecto Inc. FI has received honoraria from multiple pharmaceutical companies. RFS or his institution has received grants and consulting fees from multiple pharmaceutical companies. FP, MDG, CNH, and DT have no conflicts of interest to disclose.

Contributions

Funding

This study was funded by Galecto Inc (Copenhagen, Denmark). Medical writing and editorial support in the form of developing the draft outline and full manuscript, assembling tables, editing figures and collating authors’ comments were provided by Dr Kathryn White of Cathean Ltd. and was funded by Galecto Inc.