No abstract

Classical myeloproliferative neoplasms (MPNs) are chronic, phenotypically diverse malignancies 1 associated with significant morbidity, shortened survival, 2 and limited treatment options. 3 Development of lifeprolonging, potentially curative drugs in MPNs has been more challenging than expected for neoplasms harboring highly recurrent driver mutations that activate targetable tyrosine kinases. Decades of monitoring may be required for analysis of the most important clinical outcomes in MPNs: thrombosis, progression, and mortality. This has prompted use of more convenient clinical trial endpoints with unclear connection to these long-term events. Short-term biomarkers linking MPN biology to event risk may help identify and develop disease-modifying agents with greatest potential to improve event-free survival (EFS). Unfortunately, no such biomarkers are presently available.

Introduction: Polycythemia vera (PV) progression to myelofibrosis (MF) is associated with significant morbidity and mortality. Recombinant interferon alfa (rIFN-α) is the only disease-modifying treatment shown to reverse marrow fibrosis, but the myelofibrosis-free survival (MFS) and overall survival (OS) benefit associated with rIFN-α treatment is yet to be determined. In our preliminary analysis of 306 PV patients (pts) (Abu-Zeinah et al. ASH 2019), those treated with rIFN-α had a longer MFS and OS compared to those treated with hydroxyurea (HU) or phlebotomy-only (PHL-O). Herein, we present the final analysis comparing MFS and OS of 470 PV pts treated with rIFN-α, HU, or PHL-O. Methods: This single-center retrospective study at Weill Cornell Medicine (WCM) was approved by the WCM institutional review board. Pts were identified by an automated query of the electronic medical record system for the diagnosis of PV and were included if they met the PVSG criteria (1974-2007), WCM criteria (2008-2016) (Silver et al. Blood 2013), and the WHO 2016 criteria (2016-2020). Post-PV MF (PPVMF) diagnosis was made according to IWG-MRT criteria. Both intention-to-treat (ITT) and on-treatment analyses compared the MFS and OS by treatment group and duration, respectively. In the ITT analysis, pts were grouped to rIFNa or HU according to the first treatment they received for at least one year or to PHL-O if no cytoreductive treatment was given. MFS and OS were estimated using Kaplan-Meier methods and the log-rank test compared survival between treatment arms in ITT analysis. Multivariable analysis of PPVMF risk and mortality was performed using a Cox proportional hazards model. Results: The 470 PV pts identified had a median age of 54 years (range 20-94) at diagnosis; 229 (49%) were women. The median follow-up was 10 years (range 0-45). The primary treatment was rIFN-a in 94 (20%), HU in 188 (40%), other drugs or combinations in 55 (12%), and PHL-O in 133 (28%). The median age at diagnosis for rIFN-a, HU and PHL-O groups was 49, 58 and 54 years respectively (p <0.001) (Table 1). The median MFS for all pts was 23.8 years and for rIFN-a, HU, and PHL-O groups was 32.5, 22.6, and 20.5 years respectively (p <0.001) (Fig 1A-B). The median OS for all pts was 26.7 years and for rIFN-a, HU, and PHL-O groups was 27.7, 25.9 and 21.3 respectively (p<0.01) (Fig 1C-D). In multivariable analysis including age, sex, CV risk factors and thrombosis history, the rIFN-a group had a lower risk of MF compared to the HU group (HR = 0.42, p =0.008) and the PHL-O group (HR = 0.24, p<0.001). Likewise, the rIFN-a group had a lower risk of death compared to the HU group (HR = 0.33, p=0.01) and the PHL-O group (HR 0.3, p=0.001) independent of age. The HU group had a lower risk of PPVMF compared to PHL-O (HR 0.45, p=0.005) but no OS difference. Analysis of longitudinal bone marrow samples showed significantly less MF 2-3 fibrosis in rIFN-a compared to the HU and PHL-O groups (Figure 2). Time on treatment multivariable analysis showed that rIFN-a reduced PPVMF by 6% and all-cause mortality by 8% per year (HR = 0.94, p <0.001 and HR = 0.92, p<0.001 respectively), independent of age, thrombosis history, CV risk factors, or other treatments. On the other hand, HU was not associated with a risk reduction of either MF or mortality. The discontinuation rate of rIFN-a or HU due to toxicity was similar at 2.2 and 2.8 per 100 patient-years. Discussion: This is the largest single center study in PV using universal diagnostic criteria showing the MFS and OS advantage with rIFN-a over HU or PHL-O, and its similar rates of toxicity to HU. The evidence suggests that early cytoreductive treatment rather than PHL-O is advantageous. Treatment with PHL-O was associated with a higher risk of MF than either rIFN-a or HU. Conclusion: Our results support early use of rIFN-a as a safe, disease-modifying treatment of rigorously defined PV to delay and potentially prevent PPVMF, and prolong overall survival of PV pts. Disclosures Ritchie: Jazz pharmaceuticals: Honoraria, Research Funding; Pfizer: Honoraria, Research Funding; Abbvie: Honoraria; Sierra Oncology: Honoraria; Incyte: Speakers Bureau; Novartis: Honoraria. Silver:PharmaEssentia: Speakers Bureau. OffLabel Disclosure: Interferon alfa has been used off label in the treatment of MPN for decades

Myeloproliferative Neoplasms (MPN) harbor highly recurrent driver mutations affecting targetable kinases yet treatment options for these phenotypically diverse diseases are limited, and patients experience significant morbidity and shortened survival. The most important disease-related complications—thrombosis, transformation and death—are not used as clinical trial endpoints due to the long follow-up required to assess such disease modifying activity. A reliable monitoring biomarker linking MPN biology with these important clinical outcomes is missing. MPN driver mutation allele frequency (MAF) from whole blood or marrow (WB) does not faithfully predict MPN phenotype, clinical progression or response. This is likely because WB MAF is a composite measure of alleles from a heterogenous and variable mixture of mature leukocytes and, as such, does not report any information about the critical MPN stem and progenitor cells (MPN-SPCs). Driver mutations allow MPN cells to outcompete their normal hematopoietic counterparts and this competitive advantage—increased “fitness”—underlies core biology of MPN pathogenesis. We developed an approach to directly measure MPN fitness from samples. We measured fitness in 115 samples from 84 patients with JAK2V617F MPNs by quantifying MAF of 11 well-defined and strictly validated hematopoietic stem, progenitor and mature cell populations purified from routinely collected blood and marrow specimens. Unsupervised, hierarchical clustering of MPN fitness revealed 4 major fitness levels: F1, F2, F3, and F4 with significantly different but overlapping clinical features and diagnoses. Notably, these four fitness levels were associated with significantly different event-free survival (EFS): 95% (F1), 81% (F2), 73% (F3), 50% (F4) at 24 months (log-rank p=0.017). In contrast, WB MAF quartile failed to predict EFS. Multivariable models showed that fitness was associated with event risk independent of age, sex, duration of disease, MPN diagnosis and WB MAF. Principal component analysis allowed convenient projection of the 11-component MAF fitness measures to reduce dimensionality and develop a model for relative risk (RR) of event that could be used to assess individual or serial samples. Serial samples with more than a year of follow-up was available for 13 patients. We found that a reduction of this RR score was associated with a therapeutic response (p=0.045). In contrast, increasing RR overtime portended a disease-related event (p=0.045). Changes in WB MAF did not correlate with RR (r2=0.022) possibly explaining why WB MAF failed to predict events. These data demonstrate that fitness dynamics from serial blood samples can be used as a monitoring biomarker to assess changes in RR over time. Thus, fitness risk is a promising endpoint alongside corresponding clinical parameters such as blood counts, spleen size and marrow fibrosis grade. Our study offers a feasible approach to monitor the MPN biology central to disease progression and can be used in clinical trials to efficiently identify disease-modifying, potentially life-prolonging treatments.