FDA Approval of the JAK Inhibitor Ruxolitinib and the Management of Patients with Myelofibrosis

Five-year follow-up data are now available for the COMFORT-I and II trials of ruxolitinib.^1,2  In COMFORT-II, 53 percent of patients treated with ruxolitinib achieved a ≥ 35 percent reduction in spleen volume from baseline at any time during treatment.¹  Improvement in spleen size was maintained with long-term therapy, with the median duration of maintenance of spleen volume reduction being 3.2 years. The Kaplan-Meier probability of maintaining this reduction was 51 percent at three years and 48 percent at five years. Median OS was not reached in the ruxolitinib arm and was 4.1 years in the best available therapy (BAT) arm. There was a 33 percent reduction in risk of death with ruxolitinib compared with BAT (HR, 0.67; 95% CI, 0.44-1.02; p = .06). The probability of survival at five years was 56 percent with ruxolitinib and 44 percent with BAT. There were no new or unexpected adverse effects (AEs) with long-term ruxolitinib exposure. The most commonly reported AEs in patients who received ruxolitinib any time were thrombocytopenia (52%), anemia (49%), diarrhea (36%), and peripheral edema (33%); grade 3/4 AEs included anemia (23%), thrombocytopenia (15%), and pneumonia (6%). There were similar rates of evolution to leukemia between the ruxolitinib (5.5%) and BAT (6.8%) arms. The long-term efficacy and safety results of COMFORT-1²  are very similar to COMFORT-II, and were presented at the ASCO annual meeting in June 2016 at the time of this writing.² 

Ruxolitinib has immunosuppressive properties that are in part related to impairment of dendritic cell, NK-cell, and T-cell function.^3-5  Atypical and/or opportunistic infections have been observed in MF patients treated with ruxolitinib, including reactivation of tuberculosis,⁶  viral hepatitis,⁷  and herpes zoster. Additionally, there have been case reports of toxoplasmosis and cytomegalovirus retinitis,^8,9  as well as pneumocystis and fungal pneumonias. Although there are no consensus guidelines regarding use of anti-infective prophylaxis (nor is the live zoster vaccine recommended), treating physicians should perform risk assessments for these infections in patients before commencing ruxolitinib, with tailored laboratory screening where appropriate.

Phase III clinical development of another JAK inhibitor, fedratinib, was discontinued because of the development of several cases of Wernicke’s encephalopathy. In vitro studies indicate that oral absorption of dietary thiamine update is inhibited by fedratinib, which blocks the human thiamine transporter.¹⁰  Inhibition of the thiamine transporter does not appear to be an issue with the other JAK inhibitors currently being tested. Another JAK inhibitor, pacritinib, has been evaluated in two phase III clinical trials vs. BAT: PERSIST-1 (JAK inhibitor-naive MF patients) and PERSIST-2 (prior JAK2 inhibitors treatment allowed and platelet count < 100 × 10⁹/L required). In the PERSIST-1 trial, pacritinib produced reductions of spleen volume and symptoms and a 26 percent rate of red blood cell transfusion-independence with otherwise minimal myelosuppression.¹¹  However, in February 2016, the FDA placed a clinical hold on pacritinib due to an interim analysis of the PERSIST-2 trial which demonstrated a detrimental effect on survival in the ruxolitinib compared to BAT arm. Specifically, a relatively higher rate of intracranial hemorrhage, cardiac failure, and cardiac arrest was observed with ruxolitinib therapy.¹¹  Momelotinib remains the only other JAK inhibitor currently in phase III clinical trial testing, with results expected soon from the SIMPLIFY-1 trial (momelotinib vs. ruxolitinib; 1:1 randomization; no prior JAK inhibitor therapy) and SIMPLIFY-2 trial (momelotinib vs. BAT; 2:1 randomization; a key inclusion criterion is anemia and/or thrombocytopenia with prior ruxolitinib treatment).

Although a limited proportion of patients treated with ruxolitinib or the other JAK inhibitors may demonstrate a reduction in bone marrow fibrosis and/or JAK2 V617F allele burden over time,¹  disappearance of fibrosis or complete molecular remissions is exceedingly rare. More rigorous data are needed to establish the long-term effects of JAK inhibitors on marrow fibrosis improvement or stabilization compared to conventional therapies. Meanwhile, phase I/II trials of ruxolitinib in combination with other therapies with different mechanisms of action are being undertaken to assess whether response rates and/or drug-related anemia can be improved. Such medications include immunomodulatory agents (e.g., lenalidomide; pomalidomide); hypomethylating agents (e.g., azacitidine/decitabine); phosphoinositide-3 kinase inhibitors (e.g., buparlisib; idelalisib); histone deacetylase inhibitors (e.g., panobinostat); hedgehog pathway inhibitors (e.g., sonidegib); and agents to stimulate erythropoiesis (erythropoietin; danazol).¹²  Phase II monotherapy trials of antifibrotics (e.g., PRM-151) and the telomerase inhibition (imetelstat) are in progress in intermediate or high risk MF patients who did not respond to, or did not tolerate ruxolitinib. Type II JAK inhibitors, which only inhibit mutant JAK2 (e.g., CHZ868), are in pre-clinical development and may offer the hope of more in-depth responses since currently available JAK inhibitors inhibit both wildtype and mutant JAK2.¹³  Lastly, the role of ruxolitinib in the pre-and post-transplant setting is the subject of several ongoing clinical trials.

Updated Conflict of InterestDr. Gotlib has received (or is pending) research funding from Incyte (manufacturer of ruxolitinib), Sanofi-Aventis (fedratinib), Gilead (momelotinib; idelalisib), Promedior (PRM-151), and imetelstat (Janssen).

How does the recent FDA approval of the JAK inhibitor ruxolitinib influence your management of patients with myelofibrosis?

Myelofibrosis (primary and post-PV/ET MF) is a Philadelphia chromosome-negative myeloproliferative neoplasm with a natural history characterized by progressive anemia, spleen enlargement due to extramedullary hematopoiesis, and potential for evolution to acute myeloid leukemia (AML). Impairment of quality of life is due to both massive splenomegaly (e.g., early satiety and abdominal discomfort) and inflammatory cytokines, which mediate debilitating symptoms such as night sweats, fevers, muscle/bone pain, and cachexia. The activating JAK2 V617F mutation is present in 50 to 60 percent of patients, and the MPL mutation (W515L/K), resulting in ligand-independent activation of the thrombopoietin receptor, is identified in an additional 5 to 10 percent of individuals. It has become abundantly clear, however, that MPNs such as MF are more genetically complex. Molecular alterations in additional genes and dysregulation of the epigenetic machinery also contribute to disease pathogenesis.

Prognostic scoring systems based on clinical and laboratory factors obtained either at the time of diagnosis (IPSS)¹  or during the disease course (dynamic IPSS, or DIPSS)²  have been developed in order to estimate both overall survival and risk of progression to AML. The IPSS uses five adverse prognostic factors: age >65, hemoglobin <10 g/dL, white blood cell count >25,000/mm³, constitutional symptoms, and peripheral blood blasts >1 percent. The DIPSS-Plus refines prognosis assessment by incorporating three additional adverse risk factors: platelet count <100,000/mm³, the need for red blood cell transfusions, and poor-risk cytogenetics. Using the IPSS as an example, patients can be stratified into one of four risk groups: low (score 0), intermediate-1 (score 1), intermediate-2 (score 2), or high (score >3), with a median overall survival among the groups ranging from approximately 11 years to just over two years.¹ 

Age, performance status, and prognostic risk group drive decision making about treatment options. The Table (below) illustrates this point using three patients. Patient 1 is a lowrisk patient with excellent performance status, minimally abnormal blood counts and splenomegaly, and no constitutional symptoms. Such patients do not warrant immediate treatment and a watch-and-wait approach may be undertaken. At the other end of the spectrum, Patient 3 is a younger, high-risk patient characterized by abnormal blood counts, marked splenomegaly, and poor-risk cytogenetics. The DIPSS-Plus estimate of overall survival is 16 months, and the five- and 10-year leukemia rates are 18 and 31 percent, respectively.²  Given this patient’s younger age and relatively poor prognosis, evaluation for a potentially curative myeloablative hematopoietic stem cell transplant would be encouraged.

For patients who require treatment and are not candidates for transplantation, available therapies for MF-related cytopenias, splenomegaly, and symptoms are considered palliative. These options have included chemotherapy such as hydroxyurea; erythropoiesis-stimulating agents; immunomodulatory drugs such as thalidomide or lenalidomide, with or without corticosteroids; splenectomy; splenic irradiation; and clinical trials. In November 2011, only four years after commencing clinical trial evaluation, ruxolitinib became the first JAK inhibitor approved by the FDA for MF patients (intermediate- and high-risk).

The registration trials for ruxolitinib consisted of two large phase III trials: COMFORT-I was a randomized (1:1), double-blind, multicenter study comparing ruxolitinib 15 or 20 mg twice daily (dose stratified according to baseline platelet count) versus placebo,³  and COMFORT-II was a randomized (2:1), open-label, multicenter trial comparing ruxolitinib 15 or 20 mg bid versus best available therapy (BAT; investigator-selected including no treatment).⁴  Both trials met the primary endpoint of the percentage of ruxolitinib versus control patients achieving > 35 percent reduction in spleen volume at week 24 (COMFORT-I: 41.9% vs. 0.7%) and week 48 (COMFORT-II: 28.5% vs. 0%). After 24 weeks in the COMFORT-I trial, the proportion of patients with ≥50 percent improvement in total symptom score (using the myelofibrosis symptom assessment form) was 45.9 percent versus 5.3 percent (ruxolitinib vs. placebo, p<0.0001). Anemia and thrombocytopenia were common ruxolitinib-related adverse events but rarely led to drug discontinuation, and the drug was otherwise well tolerated. In an updated analysis of COMFORT-I, there was a significant overall survival benefit with ruxolitinib; at a median follow-up of 51 weeks, there were 13 (8.4%) deaths in the ruxolitinib group and 24 (15.7%) deaths in the placebo arm.⁵  The implications of these short-term data are unclear since JAK inhibitors exert modest or no impact on fundamental disease-related features such as JAK2 mutant allele burden or marrow fibrosis.

Ruxolitinib’s potency as a “spleen shrinker” and “symptom mitigator” is shared by other JAK inhibitors currently in clinical trials (e.g., SAR302503 [formerly TG101308], CYT387, SB518, etc.). Patients with or without the JAK2 V617F mutation respond similarly. Therefore, firstline treatment with a JAK inhibitor would be an ideal choice for Patient 2 described in the table and could be considered a bridging option for Patient 3 until a transplant is performed.

In an ad hoc analysis of the COMFORT-I and COMFORT-II trials, worsening of spleen size as well as symptoms and quality-of-life scores were similar between the placebo and BAT control groups.⁶  Although it may be premature to abandon conventional treatments such as hydroxyurea, the comparative superiority of ruxolitinib in these trials justifies its frontline use for intermediate- and high-risk MF patients in whom the primary goal is improvement of splenomegaly and constitutional symptoms. For MF patients with anemia or RBC transfusion-dependence as the predominant clinical issue, no standard of care currently exists. In this regard, differences among JAK inhibitors may prove informative in tailoring particular agents to specific patient presentations. For example, the JAK 1/2 inhibitor CYT387 has demonstrated improvements in hemoglobin/transfusion-dependence in an ongoing phase II trial.⁷  Lenalidomide (or pomalidomide, on a trial basis) may also elicit benefits in anemia in ~20 to 30 percent of MF patients.

Given the rapid “on/off” action of JAK inhibitors, caution must be undertaken when stopping these agents because of the potential for return of symptoms in a short period of time. During therapy, disease “persistence” (incomplete regression or return of splenomegaly and symptoms) can occur and may be partly explained biologically by reactivation (phosphorylation) of JAK2 through heterodimerization with JAK family members JAK1 and TYK2.⁸  Future directions will therefore be focused on combination trials of JAK inhibitors with other targeted targets (e.g., histone deacetylase inhibitors, anti-fibrotics, PI3 kinase/ AKT inhibitors) to improve the quality and duration of responses.

Table: MF Patient Profiles

PB=peripheral blood

*below left costal margin by palpation