Osteoarticular Pain after Discontinuation of Tyrosine Kinase Inhibitors (TKI): A French Cohort

Context: The Tyrosine Kinase Inhibitors (TKIs) have revolutionized the treatment of chronic myeloid leukemia (CML) increasing dramatically the survival of CML patients and leading to a residual disease with a sustained and deep molecular response. In this subset of very good responder patients, the attempts of stopping treatment in different clinical trials were successfully achieved without relapse. The Swedish team in the EURO-SKI protocol already reported cases of musculoskeletal pain occurring after cessation of TKI (Richter et al., JCO, 2014). Since several clinical trials regarding TKI discontinuation have been also run in France, we decided to retrospectively collect data using the pharmacovigilance system of the different Trials collected prospectively.

Method: 428 patients from STIM2 (n=204) and EURO-SKI (n=224) trials were systematically analyzed from the case report from each trial. For the EURO-SKI only French patients were included. Statistical analysis was performed using Stata 13 software (StataCorp LP, College Station, TX, US). Comparisons between the independent groups were realized using the Chi-squared or Fisher's exact tests for categorical variables, and using Student t-test or Mann-Whitney test for quantitative. Multivariate analyses were performed to take into account adjustment on covariates fixed according to univariate results and clinically relevance.

Results: Among the 428 patients the main characteristics were as follow i,e; 208 (48.6%) men and 220 (51.4%) women, with a median age of 77.5 years (24-93). Sokal scores (n=449) were low in 187 (41.6%) patients, intermediate in 188 (41.9%) patients and high in 74 (16.5%) patients. A withdrawal TKI syndrome (WS) was reported for 102 (23.8%) patients (100 after imatinib and 2 after nilotinib). 2). The WS consists in bone and articular pains and arthritis and affects the upper limbs, shoulders and cervical rachis, with a grade 1 or 2 in most patients and grade 3 in 22% of patients . The prevalence of WS depends on the trials, 34.8% in EURO-SKI group and 13.8% in STIM2 group (p<0.001). The WS was treated by non-steroidal anti-inflammatory drugs, corticosteroids or by local infiltration. The median duration of WS was 7 months (range: 3-30 months, 24 exploitable cases).

We did not observe any difference between WS group and the group without painful syndrome in terms of sex ratio (p=0.92), age (p=0.33), sokal score (p=0.15), BCR-ABL transcript (p=0.42) or duration of CML (p=0.24). However the median duration of TKI therapy appeared longer in this subgroup (median: 88.8 months vs 79.8 months (p=0.02). There was no biological inflammatory syndrome and the results of medical imaging were inconclusive. However, a medical history of osteoarticular pains or disease appeared as predisposing to withdrawal syndrome (22.9% in WS group vs 9.8% in control group; p=0.002). Finally the two factors, duration of treatment and medical history were confirmed using multivariate analysis (RR=1.73 and 1.76 respectively). Among 19 exploitable cases suffering CML relapse and requiring further TKI treatment, pain disappeared in 7 patients (37%) within a median period of 3.5 weeks.

Conclusion: About 23% of patients who stopped TKIs experienced a TKI WS and all TKI seems to be concerned. The predisposing factors were a medical history of osteoarticular pain or disease, and the duration of treatment. So patients and physicians should be aware and recommendations should be proposed for patients who have treated longtime with a history of arthritis.