Ruxolitinib discontinuation syndrome: incidence, risk factors, and management in 251 patients with myelofibrosis

Ruxolitinib (RUX) is the first JAK1/JAK2 inhibitor (JAKi) approved for the treatment of splenomegaly and symptoms related to myelofibrosis (MF). The current study aims to investigate in a real-world context: 1) modalities of RUX discontinuation; 2) incidence, timing, and severity of RDS; 3) outcome and risk factors associated with RDS. In 2016, a clinical network was established to collect information about RUX therapy in MF. This network now includes 22 academic hematology centers where MF patients are followed. A specific survey was conducted in all participating Centers with the scope to obtain comprehensive information regarding timing/modalities of RUX discontinuation, and subsequent outcome. RDS included all new symptoms that occurred within 21 days from RUX discontinuation and were interpreted by the treating Hematologist as caused by RUX discontinuation. 700 RUX-treated MF patients were included in the database. After a median follow-up from RUX start of 36.1 months, 251 (35.9%) patients discontinued RUX and were evaluable for RDS. In multivariable Cox regression analysis, only platelet count <100 x109/l (HR 2.98, 95%CI 1.29-6.90) and spleen ≥10 cm BCM (HR 2.03, 95%CI 1.01-4.17) at RUX stop were significantly associated with higher probability of RDS. RDS was significantly associated with the need of RUX rechallenge, with 8/34 (23.5%) RDS patients eventually resuming RUX (p<0.001). Notably, the occurrence of RDS did not significantly influence overall survival. Overall, this study shows that symptoms and/or splenomegaly significantly increase in around 15% of patients soon after RUX stop, with sometimes considerable clinical deterioration in already frail patients. Re-expansion of MF after RUX can be sudden and potentially life-threatening, and a rapid diagnosis of RDS is critical, as the reintroduction of RUX can quickly improve clinical status in most cases. implementation and standardization of discontinuation strategies should be pursued for a better RDS prevention in the future. Finally, we observed that the risk of RDS was significantly higher in patients with greater burden of the disease at the time of discontinuation. In this context, the re-use of a JAKi may be particularly reasonable. In conclusion, these results confirm the need for a careful surveillance of MF patients at the time of RUX discontinuation. A quick switch to alternative treatments, if clinically indicated, should be planned particularly for patients who stop RUX with large splenomegaly. In the absence of available alternatives, the occurrence of RDS may indicate a residual disease control activity and identify a population that can still benefit from JAK2i.