Ruxolitinib discontinuation syndrome in patients with myelofibrosis

Ruxolitinib is a Janus kinase (JAK) 1/2 inhibitor that is licensed for the treatment of patients with myelofibrosis (MF) and is effective in controlling MF-related splenomegaly and inducing hematological responses. The approval was based on results from the COMFORT-I and -II (NCT00952289 and NCT00934544) phase III trials.¹ In the original phase I/II study of ruxolitinib (NCT00509899), 92% of patients discontinued treatment, mainly due to associated toxicities or lack of response/disease progression. Of these patients, 11% developed severe withdrawal symptoms during ruxolitinib discontinuation.²

Ruxolitinib discontinuation syndrome (RDS) occurs within 21 days of treatment stop and is characterized by an acute relapse of disease symptoms, accelerated splenomegaly, worsening of cytopenias, and occasional hemodynamic decompensation, including acute respiratory distress syndrome and shock.² RDS occurs regardless of the ruxolitinib dose, tapering, and steroid usage.¹

In order to investigate RDS in a real-world setting, a large cohort (N = 700) of patients with MF, treated with ruxolitinib across 21 mostly Italian hematology centers, were assessed. These results were presented by Francesca Palandri as oral presentation during the 25th European Hematology Association Annual Congress.¹ 

Aims¹

• Outline the modalities of ruxolitinib discontinuation
• Describe the incidence, timing, and severity of RDS
• Investigate the outcome and risk factors associated with RDS 

Methods¹

• A survey was conducted in all 21 participating centers to obtain comprehensive information regarding the timing and modalities of ruxolitinib discontinuation and clinical/laboratory data during treatment, at the time of discontinuation, and after discontinuation

• Data cutoff: May 1, 2020
• RDS was defined as follows:
  • Mild: If symptoms did not require any medical interventions
  • Moderate: If symptoms required medical interventions, such as administering steroids, oral analgesics, or restarting ruxolitinib
  • Severe: If symptoms required intravenous medications, hospital admission, splenectomy, or caused the delay of hematopoietic stem cell transplantation (HSCT)
  • Fatal: If death was attributable to RDS

Results¹

• After a median follow-up time of 36.1 months from the start of ruxolitinib, 251 patients (35.9%) survived ≥ 30 days after discontinuation and were evaluable for analysis
• Patient characteristics at the time of ruxolitinib discontinuation can be seen in Table 1
  • Most patients were > 70 years of age and presented with anemia and a large splenomegaly
  • The ruxolitinib dose was < 10 mg twice daily in over 50% of patients

Table 1. Patient characteristics at the time of ruxolitinib discontinuation¹

Management of ruxolitinib discontinuation^1,2

• Ruxolitinib was immediately stopped in 64.5% of patients; in the remaining 35.5% of patients, the ruxolitinib dose was gradually decreased by 5 or 10 mg/day at variable intervals (range, 3–30 days)
• The median duration of tapering was 14 days
•  In 38.2% of patients that received ruxolitinib tapering, it was carried out in combination with prednisone (12.5 mg/day) ± hydroxyurea (1 g/day) administration
• In 13 centers, tapering was performed in < 20% of cases
• Only four centers regularly performed tapering, and tapering use was consistent within individual centers

RDS incidence¹

• 13.5% of patients developed RDS, mostly mild (Figure 1)
  • The median time to all grades of RDS was 7 days (range, 2–21 days)

Figure 1. RDS incidence¹

RDS, ruxolitinib discontinuation syndrome

• The most common symptom in 21 patients with mild RDS was an increase of the spleen (62% of patients), followed by fatigue, itching, bone pain, and abdominal discomfort (29% of patients). Night sweats, fever, and weight loss were seen in 9% of patients
  • The median time to mild RDS was 10 days (range, 3–21 days)
• An increased spleen was also seen in ten patients (70%) with moderate RDS, while two patients had night sweats, fever, and weight loss. One patient had extreme fatigue
  • The median time to moderate RDS was 8.5 days (range, 3–20 days) 
• Severe RDS occurred only in three patients (1.2%), and no fatal cases were reported

Outcomes¹

• Mild RDS required no intervention by definition, however, 38% of patients received various therapies, such as hypomethylating agents, chemotherapy, splenectomy, or allo-HSCT, and 15% received JAK2-inhibitors (Figure 2A)
• Of the patients with moderate RDS, eight received corticosteroids and two received investigational non-JAK2 inhibitors. Three patents subsequently received ruxolitinib re-challenge after an average time of 2.6 years (Figure 2B)  

Figure 2. Outcomes and treatments for patients with mild and moderate RDS¹

allo-HSCT, allogeneic hematopoietic stem cell transplantation; HMA, hypomethylating agent; HU, hydroxyurea; JAK2, Janus kinase 2; RDS, ruxolitinib discontinuation syndrome

• Severe RDS occurred in three patients (Table 2) within 3 days of discontinuation and caused various symptoms
  • All patients rapidly recovered after ruxolitinib re-challenge at 10 mg twice daily

Table 2. Treatment and outcome of patients with severe RDS¹

 Risk Factors for RDS¹

• Table 3 shows that the factors at the time of ruxolitinib stop that were significantly associated with increased risk of RDS (univariate analysis) were platelet count < 100 × 10⁹/L (HR, 2.83; p = 0.01), ruxolitinib dose ≤ 10 mg twice daily (HR, 4.95; p = 0.01), and spleen size ≥ 10 cm (HR, 2.15; p = 0.03)

Table 3. Risk factors for RDS (univariate analysis)¹

• Multivariable analysis revealed that two factors at the time of ruxolitinib discontinuation were significantly associated with a higher probability of developing RDS:
  
  • Platelet count < 100 × 10⁹/L (HR, 2.98; 95% CI, 1.29–6.90; p = 0.01)
  • Spleen size ≥ 10 cm (HR, 2.03; 95% CI, 1.01–4.17; p = 0.04)

Conclusion

Severe RDS is rare and occurs in approximately 1% of patients. Diagnosis of RDS is crucial, as ruxolitinib re-challenge may rapidly improve the clinical status of the patient. As mild or moderate RDS mainly occurred within a few days from discontinuation and affected > 10 % of patients, any other therapy should be started as close as possible to ruxolitinib discontinuation. RDS is often associated with an increase in spleen size and may therefore indicate residual JAK disease control activity, allowing to identify patients who could still benefit from second-line JAK2 inhibition.

A limitation of the study was that RDS prevention strategies were infrequent and inconsistent across the different centers. This may have prevented the detection of a correlation between tapering and RDS reduction. The implementation of consistent discontinuation strategies may allow for better RDS prevention in the future.  

Expert opinion

For more information on how to manage RDS, watch our video with Francesca Palandri: