Evaluation of a Pan-Lysyl Oxidase Inhibitor, Pxs-5505, in Myelofibrosis: A Phase I, Randomized, Placebo Controlled Double Blind Study in Healthy Adults

BACKGROUND: Myeloproliferative neoplasms (MPNs) are clonal stem cell neoplasms characterized by terminal expansion of the myeloid cell lineage and include essential thrombocythemia (ET), polycythemia vera (PV), and myelofibrosis (MF). MF carries the worst prognosis and is characterized by reactive bone marrow fibrosis. Currently bone marrow transplant is the only known therapy that can reverse fibrosis and alter the disease course.

Lysyl oxidases (LOX, LOXL1-4) are copper amine oxidase enzymes that facilitate the cross-linking of collagen and elastin through deamination and oxidization of lysine residues, yielding highly reactive aldehydes. This is essential for fibrotic tissue formation. An earlier study identified an important role for LOX in the development of MF in mice (Eliades et al, J. Biol. Chem. 2011; PMID: 21665949). Furthermore, small molecule pan-LOX inhibitors reduced spleen size and bone marrow fibrosis in mouse models of MF (Leiva et al, Int J. Hemat. 2019; PMID: 31637674). This provides clinical rationale for the use of pan-lysyl oxidase inhibition in MF.

A Phase 1 study was conducted to establish the safety, tolerability, pharmacokinetics (PK), and pharmacodynamics (PD) of PXS-5505, a pan-lysyl oxidase inhibitor. Plasma levels of LOX and LOXL2 were investigated in a separate cohort of MPN patients, with the ultimate goal to develop PXS-5505-based therapy in MF patients.

METHODS: PXS-5505 was dosed orally in a Phase I, randomized, placebo-controlled, single ascending (Part A) and multiple ascending dose (Part B) double-blind study in healthy male volunteers. Plasma LOX and LOXL2 levels were also measured in a separate cohort of MPN patients using ELISA-based Single Molecule Array technology (Simoa; Quanterix).

RESULTS: Forty subjects were enrolled in Part A and 16 were enrolled in Part B. In Part A, there were 5 cohorts consisting of 8 subjects (6 active, 2 placebo) with doses of 10, 50, 100, 200 or 300 mg administered once. In Part B, there were 2 cohorts of 8 subjects each (6 active, 2 placebo) at doses of 100 or 200 mg daily for 14 days. There were no significant treatment-related adverse events. Mean age in Part A was 32.0 years (SD 11.14) and 32.1 years (SD 12.14) in Part B.

Cmax and AUC increased linearly across the dose range of 10 to 300 mg for single dose administration. In multiple dosing at 200 mg daily, Cmax was 916 ng/mL and AUC0-24 was 7421 hr*ng/mL on Day 14, with median Tmax of 1 hour and t1/2 of 7 hours. When measured in plasma, LOX was dose-dependently inhibited and achieved very strong inhibition (median 80%) with a 200 mg dose around Tmax. The 300 mg dosing did not significantly increase inhibition. Multiple daily doses of 100-200 mg resulted in 60-70% and 50-60% inhibition of plasma LOX at 12 and 24 hours. Average LOX plasma levels were 3.61 ng/mL (range: 3.20-13.42 ng/mL; SD: 2.92 ng/mL).

We measured plasma LOX and LOXL2 levels in a separate MPN cohort of 9 ET, 8 PV, and 13 MF patients (mean age 61.4, range:24-84, 65% males). LOXL2 levels were higher in MF (mean 415 pg/mL) compared to ET (mean 209 pg/mL) and PV (mean 322 pg/ml), although this was not significant. However, LOXL2 levels in all MPN patients (mean 333 pg/mL) were significantly higher than LOXL2 levels in normal controls (mean 152 pg/mL, p<0.01). There were no significant differences in LOX levels in the two groups.

DISCUSSION: PXS-5505 demonstrated an excellent safety profile and was well tolerated in healthy human subjects. PK/PD properties are consistent with preclinical data and support once or twice daily >100 mg dosing over 14 days. PXS-5505 achieves long-lasting, strong inhibition of lysyl oxidases. Plasma LOXL2 levels are higher in MF patients compared to healthy controls, and we found no significant disease associations of LOX or LOXL2 among MPN subtypes in our small cohort. LOXL2 is likely a more sensitive MF biomarker as it is present at low concentrations in the blood, while LOX is constantly produced from major organs making detection due to disease more difficult. Based on previous mouse studies, it is possible that LOX levels would be higher in MF patients when compared to age-matched controls, which we will investigate further. We will open a Phase IB/II study of PXS-5505 in MF patients resistant to ruxolitinib.