The TKI-Free Duration after a First Discontinuation Attempt That Failed in CP CML Patients Is a Predictive Factor of TKI-Free Remission after a Second Attempt

Background: Tyrosine kinase inhibitors (TKIs) are able to induce, in some chronic myeloid leukemia (CML) patients in chronic phase (CP), long-term molecular response 4.5 (MR4.5) and several studies have now demonstrated that TKIs could be safely discontinued in those patients with a Treatment-Free Remission (TFR) rate reaching ~50%. The French CML group had recently demonstrated that a failure of the first TKI discontinuation attempt does not preclude a 2^nd successful attempt (RE-STIM study, Legros et al. Cancer 2017).

Methods: The RE-STIM study is a national observational multicentre study collecting all cases of 2nd TKI discontinuation attempt of regardless the type, the duration of TKI, the duration of MR4.5 and the reason of discontinuation. CP-CML Patients in failure of a 1st attempt, had to recover a 2^nd sustained MR4.5 on TKI to be eligible for this new analysis of the enlarged database (n=106). Loss of MMR loss was the trigger for therapy re-introduction.

Results: At the time of analysis (1^st June 2019), 106 patients (median age: 55 years (range: 25-81 years)) were included with 41 months (2-131) of follow-up after 2^nd discontinuation. Fifty males and 56 females were enrolled. The Sokal risk score was low in 45%, intermediate in 26.5%, high in 20% and unknown in 8.5% of patients. The majority of patients (95%) were treated with imatinib as first-line, and the others with a 2^nd generation TKI. The median total time on TKI prior to a 2^nd discontinuation was 104 months (range: 38-235) and the median duration of a 2^nd MR4.5 prior to a 2^nd discontinuation was 68 months (range: 20-176). After a 1st discontinuation attempt, the reason for TKI re-challenge was in majority a loss of MMR (66%), a loss of MR4.5 in 33% of patients (missing data in 1%).

The TFR rates after a 2^nd discontinuation attempt were 44.3% [95% CI 35.48-55.41] at 24 months, 38.5% [95% CI 29.65- 50.09] at 36 months and 33.2% [95% CI 24.31- 45.39] at 48 months. In univariate analysis, we failed to find any association between TFR and: age, gender, Sokal score, prior exposure to IFN, TKI in combination versus monotherapy, TKI type, TKI treatment duration and uMR4.5 duration before the 1^st and 2^nd discontinuation attempts, and type of molecular relapse after the 1^st discontinuation attempt (MR4.5 versus MMR loss). However, the speed of molecular relapse after the 1^st TKI discontinuation remains a factor significantly associated with outcome. In patients who remained in uMR4.5 at 3 months after the 1^st discontinuation, the TFR rate at 48 months was 53% [95% CI: 35.32-79.31] and 26% [95% CI: 16.88-40.28] for others. Another factor significantly associated with outcome is the TKI-free duration after the 1^st attempt (Figure). The TFR rate at 48 months was 45 % [95% CI: 28.64- 69.62] in patients who remained without treatment more than 6 months after their 1^st attempt and 27% [95% CI: 17.57- 41.34] for others.

All patients are alive at last follow-up except 2 who died from CML-unrelated reasons. One patient developed a sudden blast crisis at 4 years from 2^nd discontinuation. The last previous molecular biology 3 months before transformation was MR4. In patients in TKI re-challenge (n=63), median TKI-free duration was 6 months (2-64), 55% of patients regained their MMR within 3 months (0-35) and 41% regained MR4.5 within 5 months (2-53).

Conclusions: This larger cohort confirms that TKIs could safely and successfully be discontinued a 2^nd time in CP CML patients despite a 1^st failure. The speed of molecular relapse after the 1^st TKI discontinuation and TKI-free duration remain major factors significantly associated with TFR outcome.

Figure: TFR according TKI-free duration after the 1^st attempt of discontinuation

Figure

View largeDownload slide

Figure

View largeDownload slide

Close modal