Abstract
Pulmonary hypertension (PH) is an under-recognized complication of myelofibrosis (MF) occurring in 30% of MF patients and associated with poor survival. Echocardiographic diagnostic findings include; elevated right ventricular systolic pressure (RVSP)>35 mmHg, right atrial (RA) enlargement and increased tricuspid regurgitation velocity (TRV) ≥2.5 m/sec. The pathophysiology of PH in MF has not been elucidated, although in idiopathic PH, the proliferation of pulmonary artery endothelial cells has been linked to activation of STAT3 pathway. Dysregulation of JAK-STAT pathway has been implicated in the pathogenesis of MF. Ruxolitinib, a JAK1/2 inhibitor, was approved for management of splenomegaly and cytokine-mediated symptoms in MF. Furthermore, no specific therapy in the management of MF-associated PH has been established. Given the association between MF and PH and the possible pathophysiologic link mediated by JAK signaling, we prospectively followed 19 patients with MF-associated PH and compared their echocardiographic findings and PH relevant serum biomarker levels (nitric oxide [NO], NT-pro brain natriuretic peptide [NT-proBNP], von Willebrand antigen (vWB), ristocetin co-factor (RCA), and uric acid (UA) pre- and post-ruxolitinib therapy. All categorical data were summarized for frequency, counts and percentages, and the comparison between two groups was performed by two-sample Wilcoxon signed rank test. Among 19 patients (pts), 9 had PMF, 5 post-ET MF, 4 post-PV MF and one CMML-1. In this cohort, 11 were females and 8 were males. The median age of the cohort was 68 years (range, 50-81 years). Fifteen pts were JAK2 V617F positive and 4 were wild-type, 8 were intermediate-1, 4 intermediate-2 and 6 high risk per Dynamic International Prognostic Scoring System-Plus risk grouping. The mean ruxolitinib dose was 10 mg BID (range: 5 mg QOD-20 mg BID]. Median duration of disease was 32 mos (6-164 mos), ruxolitinib duration of treatment was 10 mos (4 -17 mos) and follow-up was 11 mos (6-22 mos). Prior to the initiation of ruxolitinib treatment, NT-pro BNP levels, were measured and found to be elevated in 90% (17/19) of pts. In addition, UA, vWB, and RCA levels were all elevated in 47% (9/19), 24% (4/17), and 12% (2/17) of pts respectively. The strongest correlation among serum biomarkers was between plasma vWB and RCA levels (r2=-0.89, P=<.001). The biomarker most closely associated with elevated NT-pro BNP was UA both in the pre- (r2=-0.53, P=.065) and post-treatment (r2=-0.64, P=.019) settings. Echocardiographic findings by TTE pre- and post ruxolitinib therapy were available for 12 pts (63%). All 12 had documented PH with a mean RVSP of 47.5 mm Hg (42-68) [normal pressure ≤30 mmHg]. Echocardiographic evidence correlated with RCA (r2=-0.64, P= .045) and plasma NT-pro BNP levels (r2=-0.8, P=.013). Ruxolitinib resulted in reductions in NT-pro BNP level (88%) (p=.013), plasma UA levels in (71%), vWB (71%), and RCA (71%) (P=.045). Nitric oxide, a primary regulator of vascular endothelial function is reduced in MF patients with PH compared to normal individuals (median NO, 36 vs 65 pM). Treatment with ruxolitinib resulted in marked increase in NO levels compared to baseline (68 pM vs 36 pM; P=0.04) while no changes in NO levels were observed after treatment with hydroxyurea and lenalidomide (N=10). Treatment with ruxolitinib also resulted in reduction of key cytokines (TNF-α, IL-4, IL-10) that inhibit NO production and induction of cytokines (IFN-γ) that lead to increase in NO synthesis supporting the role of cytokines in PH pathogenesis in MF. Murine studies further supported the role of ruxolitinib in induction of NO levels. Eight normal CD-1 mice were treated with ruxolitinib (50 mg/kg p.o. daily for 5 days for three consecutive cycles with 14 day intervals between each cycle). After the first cycle, NO levels were significantly higher compared to baseline followed by significant increase compared to baseline at cycle 3 (P=.04). In addition, PH mice (Caveoline-1 mice) have been bred and undergoing treatment with ruxolitinib to assess changes in NO levels and its impact in improving of PH. In conclusion, aberrant JAK-STAT signaling in MF mediates PH by dysregulation of NO and cytokines levels which can be restored by therapy with JAK inhibitors. This suggests that inhibition of the JAK-STAT signaling pathway is a novel and viable target for the management of patients with MF-associated PH.
Disclosures:
No relevant conflicts of interest to declare.
C1q/TNF-related protein-9 ameliorates hypoxia-induced pulmonary hypertension by regulating secretion of endothelin-1 and nitric oxide mediated by AMPK in rats
