Publication Only: Chronic myeloid leukemia - Clinical
A musculoskeletal pain after discontinuation of tyrosine kinase inhibitors (TKI) in patients (pts) with chronic myeloid leukemia (CML) and deep molecular response (DMR) has been described as a withdrawal syndrome (WS). According to studies of treatment-free remission (TFR) in CML the frequency of WS is nearly 23,8-31% (Berger et al, Blood 2015, Saussele et al, Lancet Oncol., 2018). The predisposing factors of WS and its nature are being studied.
To evaluate the clinical features of WS and factors associated with the WS in CML pts after TKI cessation in the RU-SKI study.
The key inclusion criteria of the RU-SKI trial were as follows: CML chronic phase, TKI therapy ≥ 3 years, DMR duration ≥ 2 years. DMR was defined as at least MR4 (BCR-ABL ≤ 0,01% IS). TKI were restarted if a loss of major MR (MMR - BCR-ABL>0,1% IS) was observed. In total 98 pts were included. Median (Me) time of observation was 25 months (mo) (range 12-42). High Sokal score was in 13% pts, 48% were males. Me age was 46 years (range 22-80). Me duration of TKI therapy and Me duration of DMR before treatment cessation were 8,3 years (range 3-16) and 3,2 years (range 2-11) respectively. MMR loss and TKI resuming was in 46(47%) pts. Me time till MMR loss was 3 months (range 1-12). The molecular relapse free survival (MRFS) was 51% at 12 mo. WS was defined as a musculoskeletal pain newly observed in CML pts after TKI cessation or as a worsening of previously observed musculoskeletal symptoms in pts with medical history of locomotion system diseases. We have evaluated age, gender, weight, body mass index, Sokal score, duration of TKI therapy, presence of locomotion system diseases in pts with and without WS. Comparison was done with Mann-Whitney U-test and Chi-squared test. MRFS was evaluated by Kaplan-Meier metod, log-rank test was used for comparison.
WS was observed in 41(42%) of 98 pts. TKI before discontinuation were as follows: imatinib (33), nilotinib (7) and dasatinib (1). Me time to WS appearance was 2 mo (range 1-7). Newly observed and worsening musculoskeletal pain were in 29(71%) and 12(29%) pts respectively. WS was presented as arthralgia, myalgia, ossalgia or combination of symptoms in 19(46%), 2(5%), 1(2,4%) and 17(41%) pts. Atypical WS (headache, paresthesia) was in 2 pts. WS grade 1-2 was in 39(95%) pts, grade 3 in 2 pts. WS was resolved in 35(85%)/41 pts: spontaneously or after anti-inflammatory therapy in 24 pts; after TKI restart in 11 pts. Me duration of WS was 5 mo (range 1-35). An older age (p = 0,039) and longer duration of TKI therapy (p = 0,001) were observed in pts with WS, other factors were not significant (table 1). MMR loss occurred in 11(27%) of pts with WS vs 35(61%) of pts without WS (p = 0,0009). The development of WS was observed both in the early and in the late period after TKI cessation, when the majority of the relapses have already occurred. A Landmark analysis is reasonable for a more accurate assessment of WS on MRFS relationship. Considering the time of WS and of molecular relapse, the 2-month landmark point was selected. MRFS at 12 mo was 42% and 66% in pts with and without WS respectively (p = 0,095) (figure 1).
The rate of WS in the CML pts population of the RU-SKI trial was 42% that is higher than in other TFR studies. The clinical features and duration of WS were comparable to usually described. A longer period of TKI exposure and older age of pts are associated with the development of the WS. We found a tendency of lower rate of MRFS in pts with WS. The relation between MMR loss and WS is desirable to be studied in larger pts cohorts.