Ruxolitinib is a selective JAK 1/2 inhibitor, which shows an excellent treatment outcome in myelofibrosis (MF) patients. Main side effect of JAK 1/2 inhibitors is an increased risk of infection. JAK1/2 inhibition may interfere with the differentiation of interferon-γ (IFN-γ) producing Th1 cells and IFN-γ is a key cytokine involved in protective immunity against Mycobacterium tuberculosis(TB). During COMPORT-II trial, a case of disseminated TB with ruxolitinib was reported. Here, we analyze the efficacy and safety of ruxolitinib in Korean MF patients and report cases of TB lymphadenitis during the treatment.


Forty-nine patients diagnosed with PMF, PPV-MF or PET-MF have been enrolled and at this time twenty patients are evaluable (median age; 63 years, 37-80). Starting dose of ruxolitinib was determined based on each patient’s baseline platelet count (20 mg bid/d for a baseline platelet more than 200,000/µL, 15 mg bid/d for 100,000-200,000/uL). To determine the efficacy of ruxolitinib, we serially assessed the spleen size by palpation, myelofibrosis symptom assessment using MFSAF and BM examination with JAK2V617Fmutation allele burden. Among 20 evaluable patients, 16 assessed IFN-γ release assays (IGRAs, quantiFERON-TB Gold test) before starting ruxolitinib.


Of total twenty patients, 12 (60.0%), 3 (15.0%) and 5 (25.0%) were PMF, PPV-MF and PET-MF, respectively. By DIPSS, 13 (65.0%) was Int-2 risk, 3 (15.0%) and 4 (20.0%) were Int-1 and high risk. Eleven patients started with 20 mg bid/d (median baseline platelet: 302,000/uL, 206,000-814,000) and nine were 15 mg bid/d (median; 139,000/uL, 100,000-194,000). Median baseline total symptom score (TSS) was 12 (1-36) and palpable spleen length was 19 cm (1-30). JAK2V617Fmutation was positive in 13 (65.0%) patients (median allele burden; 87.1%, 26.2-93.7). Median time of ruxolitinib exposure was 2.0 ms. (0.8-6.2). Two patients increased TSS following ruxolitinib treatment, however, median maximal reduction in TSS was above 90.9% (27.8-100) and 64.7% of patients showed more than 50% reduction of TSS with ruxolitinib. In an aspect of spleen length, all except two patients showed decreased palpable spleen length. Median maximal reduction in spleen length was 70.2% (0-100) and 72.2% of patients showed more than 35% reduction in spleen length with ruxolitinib. Three patients (15.0%) experienced gr. 3/4 thrombocytopenia and one (5.0%) gr. 3 neutropenia. Among patients who assessed pre-treatment IGRAs, only one revealed positive IGRAs. Since there was no evidence of active TB in symptom and radiologic examination, he was diagnosed as latent TB infection (LTBI) and started 9 ms.-isoniazid (INH) treatment. He had a huge hepatosplenomegaly combined with large amount of ascites which needed frequent paracentesis, hence, we started ruxolitinib with INH treatment. He showed no evidence of active TB and achieved negative IGRAs result on 5 ms. of ruxolitinib treatment. On 1 m. and 5 ms. of ruxolirinib treatment, two patients developed pyrexia and neck masses which were diagnosed as TB lymphadenitis. All of them had no previous history of TB and showed negative results in pre-treatment IGRAs and radiologic examinations. First patient discontinued ruxolitinib by herself and eventually died of MF progression 2 ms. later. Second patient continued ruxolitinib treatment with TB medication and there was no evidence of active TB or MF progression on 5 ms. of ruxolitinib treatment.


Ruxolitinib was generally well tolerated and showed an excellent treatment outcome in Korean MF patients. By 2008 WHO report, intermediate burden of TB cases exist in Korea, hence, TB is still endemic in Korea. According to 2011 Korean Guidelines for TB, LTBI should be treated in patients receiving immunosupressive agents including TNF-α inhibitors. Although further prospective investigations on the incidence of TB with JAK 1/2 inhibitors in TB endemic countries are warranted, it seems to be reasonable to check the possibility of LTBI before starting JAK 1/2 inhibitors. LTBI confirmed patients receiving JAK 1/2 inhibitors may be deemed a high risk of active TB and consider LTBI treatment. Furthermore, it is necessary to use a caution for active TB infection during the treatment of JAK 1/2 inhibitors in such countries.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.