Ruxolitinib combined with dexamethasone is a promising prophylaxis method for differentiation syndrome in acute promyelocytic leukemia (APL2022)

Abstract

Abstract Introduction: Acute promyelocytic leukemia (APL) has been highly cured with all-trans retinoic acid (ATRA) and arsenic trioxide (ATO), but they still bring differentiation syndrome (DS) in about 25% of cases. Patients with intermediate-to-high-risk APL are more prone to severe DS, which is closely associated with early mortality. The mechanism of DS remains unclear, besides current prevention methods are limited and unsatisfactory. This study aims to optimize prophylaxis strategy of DS among patients with intermediate-to-high-risk APL. Patients and Methods: The APL2022 study was a prospective, single-arm clinical trial in patients with newly diagnosed intermediate-to-high-risk APL (according to Sanz risk model, white blood cell (WBC) count > 10 × 109/L or platelet (PLT) count < 40 × 109/L) aged between 14 and 75. From June 2023 to July 2025, 114 eligible patients were treated with ATRA and ATO based induction therapy, combining limited dosage of chemotherapy if WBC count > 10 × 109/L. Seventeen of them were excluded from the study because of initially fatal bleeding, PLT count < 10 × 109/L, or refusal. And 98 patients received Ruxolitinib 5mg bid and dexamethasone 10mg per day for prophylaxis of DS after the initiation of ATRA. DS scores were measured by the following 7 clinical manifestations, each for 1 point: unexplained fever, dyspnea, pleural or pericardial effusion, pulmonary infiltration, renal failure, hypotension, and a weight gain of ≥ 5 kg. Mild DS was measured as 1-2 points, moderate DS as 3 points, and severe DS ≥ 4 points. When the DS score continued to be 0 and the WBC count less than 5 × 109/L for more than 3 days, Ruxolitinib should be discontinued and the dosage of dexamethasone gradually reduced. If symptoms reoccurred, they could be reused. If the DS score was ≥ 3, ATRA was suspended and dexamethasone should be added up to 10mg q12h. For safety concern, when PLT count was less than 10 × 109/L or any other grade III-IV adverse effect which was considered possibly related to Ruxolitinib occurred, it should be discontinued. Results: The median use of Ruxolitinib was 13 days (range 1 - 23). Among the 98 per-protocol intermediate-to-high-risk patients, 15 (15.3%) developed DS, which was significantly reduced compare to the rate in our historical study (26.3% (169 of 642), P = 0.019). High-risk patients were remarkably benefited from this strategy, with the incidence of DS reduced from 30.6% to 12.2% (P = 0.015). Dyspnea (67%), fever (67%), pleural or pericardial effusion (60%), and weight gain (53%) were the most common symptoms of DS. Once acute renal dysfunction or hypotension occurred, patients all developed severe DS. Mild-to-moderate DS occurred in 12 patients, while severe DS in 3 (3.1%) patients. Seven patients died within 30 days from diagnosis, with an early death rate of 7.1%. Only 1 died from severe DS, 4 from cerebral bleeding, 1 from hemoptysis, and 1 from infection. The following factors at diagnosis were analyzed through univariate analysis to evaluate the risk for DS (P < 0.05): age, gender, FLT3-ITD, FLT3-TKD, NRAS, WT1, WBC count, hemoglobin, PLT count and fibrinogen. Only NRAS mutation was significantly related to the onset of DS (OR = 5.319, 95% CI = 1.381-19.126, P = 0.015). We collected the data of cytokines or receptors at different time points after ATRA treatment and fitted the dynamic trajectory through Locally Weighted Regression (LOESS). Multivariate Logistic regression showed that IL-2R was the only independent predictive marker for DS (P = 0.021) among all the 13 detectable cytokines or receptors, and not affected by the infection status (using mixed effect model, P = 0.222). Patients with IL-2R elevation greater than 2.5 times the upper limit of normal within 7 days of ATRA treatment had a significantly increased risk of DS (OR = 7.31, 95% CI = 1.13-47.70, P = 0.032). Moreover, NRAS mutation would accelerate the risk of DS, manifested as an increased trend in DS risk when IL-2R reached a 1.5-fold upper limit within 5 days (OR=10.00, p=0.086), and a significantly increased risk if reaching a 2.5-fold upper limit within 5 days (OR = 34.00, P = 0.043). Conclusion: Ruxolitinib and dexamethasone combination strategy could be an effective and safe recommendation for preventing DS during ATRA-ATO induction therapy for APL patients. NRAS mutation and changes in IL-2R after ATRA treatment might serve as risk predictors for DS under this protocol. (www.chictr.org.cn, ChiCTR2300072086)