Abstract
Myeloproliferative neoplasms (MPN's) including myelofibrosis (MF) are neoplastic diseases that initiate in the bone marrow from abnormal regulation of hematopoietic stem cells primarily harboring somatic mutation activating Janus Kinase 2 (JAK2). Central phenotypes resulting from uncontrolled JAK2 signaling in MF have been established, leading to pronounced extramedullary hematopoiesis, splenomegaly, and bone marrow fibrosis; however, therapeutic resistance, durability of remission, and assessment of treatment response remain significant limitations to inhibitor treatment in many MF patient populations. Mechanisms linked to compensatory phosphoinositide 3-kinase (PI3K) and mitogen-activated protein kinase (MAPK) signaling pathways in MF highlight the frontier for combinatorial multi-kinase inhibition therapies. Striking a balance between positive clinical outcomes, dose-limiting toxicity, and monitoring of pharmacological efficacy represents an unmet need for evaluation strategies capable of effective translation from bench to bedside.
Despite reliance of disease staging on the grade of fibrosis in bone marrow sampled from just a single biopsy site (iliac crest), oncologists have rarely utilized radiological imaging to analyze the global bone marrow architecture for monitoring disease progression and response to treatment. Uncontrolled hematopoiesis during MF progression results in displacement of normal marrow leading to fibrosis and alterations in the bone marrow microenvironment. Here, we evaluated multiparametric magnetic resonance imaging (MRI) for detection of bone marrow phenotypes and treatment normalization in the tibia of murine Jak2+/V617F transplant models of MF. Diffusion-weighted MRI (DWI) quantified changes in cellularity as the dynamics of water mobility through apparent diffusion coefficient (ADC) and was validated longitudinally by repeat test-retest measurements in Jak2+/V617F mice over a 10wk period post-bone marrow transplant (post-BMT), and used to generate spatially resolved repeatability coefficients of the murine tibia marrow space. Parametric response mapping (PRM) revealed an ~20% increase in ADC in the proximal tibia (2-9 mm region) between 5 and 10 w post-BMT, and detected advancement of a cellular/adipose interface towards the distal tibia indicative of increased proliferation and displacement of marrow adiposity, presenting the ability to non-invasively identify and monitor progression MF in the marrow space using PRM. Assessment of extramedullary hematopoiesis by abdominal MRI revealed that animals treated with an experimental orally bioavailable multi-kinase inhibitor LP-182 (MEKi/PI3Ki/mTORi) in combination with the JAK inhibitor Ruxolitinib displayed ~35% reduction in spleen volume as compared to either treatment alone, and ~60% reduction compared to vehicle, with concomitant on-target inhibition of downstream kinase signal transduction evidenced by immunochemistry. Combination treatment of LP-182 and Ruxolitinib between 5 and 10 w post-BMT prevented any significant increase in ADC in the tibia marrow space of Jak2+/V617F mice compared to wild-type untreated mice, whereas untreated Jak2+/V617F mice displayed ~30% increase in ADC over the same duration highlighting the ability to monitor both disease progression and response to treatment non-invasively using PRM. Furthermore, treatment of mice with LP-182 resulted in maintenance of hematologic markers, reduction in allele burden, and was well tolerated at the administered dosage (400 mg/kg; daily) for >2 months, with no observable pharmacotoxicity as evaluated in a blinded study by a board-certified veterinary pathologist.Taken together, MRI biomarker changes correlated with changes in the bone marrow space during disease progression and response treatment with novel agents designed to target key, co-activated oncogenic drivers of MF progression. Parametric response mapping revealed important discoveries: 1) imaging of bone marrow provides spatiotemporal monitoring of disease progression and heterogeneity; 2) temporal marrow changes in response to therapy intervention can be resolved; 3) LP-182 prevented changes in cellularity to normalize marrow and provide a translational opportunity for patient allele burden reduction. While our efforts center on MF, quantitative MRI of bone marrow will be applicable to other hematologic cancers providing broad oncological patient impact.
