Paper Details

ABSTRACT

Background: Treatment-free remission (TFR) is an established treatment goal for selected patients with chronic-phase chronic myeloid leukemia (CP-CML) who achieve sustained deep molecular response (DMR). Most patients who lose major molecular response after tyrosine kinase inhibitor (TKI) discontinuation regain molecular response after TKI re-initiation. Nevertheless, rare but potentially fatal cases of accelerated- or blast-phase progression have been reported during TFR or after TKI resumption. Because previous evidence has largely been limited to case reports or small case series without an appropriate comparator, it remains uncertain whether TKI discontinuation is associated with an increased risk of disease progression. We compared the incidence of advanced-phase progression between patients who discontinued TKIs and TFR-eligible patients who continued TKI therapy.

Methods: This multicenter retrospective cohort study included consecutive adult patients with chronic-phase CML who met prespecified, uniform TKI discontinuation criteria based on the ELN Guidelines at participating tertiary hematology centers. Patients who did not undergo regular molecular monitoring after TKI discontinuation were excluded. All patients with documented loss of MMR after discontinuation resumed TKI therapy. Patients were classified according to the treatment strategy actually adopted: a discontinuation cohort comprising patients who stopped TKI therapy and a reference cohort comprising TFR-eligible patients who continued TKI therapy. Follow-up began at TKI discontinuation in the discontinuation cohort and at the date of confirmed TFR eligibility in the reference cohort and continued until advanced-phase progression, death, last follow-up, or data cutoff. Molecular relapse was defined as loss of major molecular response (MMR). Advanced-phase progression was defined as transformation to accelerated phase (AP) or blast phase (BP) according to ICC criteria. The primary endpoint was the occurrence of advanced-phase progression after TKI discontinuation. Secondary analyses descriptively evaluated the clinical characteristics, molecular findings, treatment, and outcomes of patients who developed AP or BP. Crude progression proportions were compared using Fisher’s exact test. Incidence rates were calculated per 1,000 person-years with exact 95% confidence intervals. Given the small number of progression events, no multivariable risk-factor analysis was performed.

Results: A total of 3,573 TFR-eligible patients with CP-CML were included, of whom 1,450 discontinued TKI therapy and 2,123 continued treatment. The discontinuation and reference cohorts contributed 5,177 and 9,662.3 person-years of follow-up, respectively.

Five patients in the discontinuation cohort developed advanced-phase progression after TKI cessation, corresponding to a crude proportion of 0.34% (95% CI, 0.11%–0.80%) and an incidence rate of 0.97 per 1,000 person-years (95% CI, 0.31–2.25). These events comprised 1 AP and 4 lymphoid BP transformations. In contrast, no advanced-phase progression occurred among the 2,123 patients who continued TKI therapy during 9,662.3 person-years of follow-up (0%; upper 95% confidence limit, 0.17%). The crude progression proportion was significantly higher in the discontinuation cohort (Fisher’s exact p=0.011).

Among the 5 patients with advanced-phase progression, 4 experienced sudden progression without preceding loss of MMR or TKI re-initiation. The remaining patient lost MMR, resumed TKI therapy, and subsequently developed advanced-phase progression more than 2 months after treatment re-initiation. Molecular testing at progression detected a SETD2 mutation in 1 patient and TGFBR2 and PCNT mutations in another. Two patients died during follow-up, whereas the remaining 3 were alive at the data cutoff.

Conclusions: TKI discontinuation was associated with a higher risk of advanced-phase progression than continued therapy among TFR-eligible patients, although the absolute risk remained very low. Importantly, progression was not always preceded by loss of MMR. These findings support sustained long-term molecular and clinical surveillance after TKI discontinuation and warrant confirmation in prospective studies.