Treatment-Free Remission Outcomes in a BCR::ABL1 Digital PCR Selected Clinical Cohort of CML Patients

Abstract

Approximately 40%-60% of patients reaching a stable deep molecular response during TKI treatment will maintain a state of remission after TKI discontinuation, denoted as treatment-free remission (TFR). Depth of molecular response assessed by BCR::ABL1 digital PCR prior to TKI discontinuation has demonstrated its significance as a reliable predictive parameter for TFR. A clinically applicable prediction cutoff of 0.0023%IS has been established and externally validated. In this study, BCR::ABL1 digital PCR, as most sensitive and stable assay of its kind, was investigated as a TFR prediction tool in the Netherlands, and evaluated for its predictive value to stop TKI treatment below the aforementioned cutoff. The primary endpoint of molecular recurrence (MolR, BCR::ABL1 > 0.1%IS) at 12 months was prospectively assessed. Overall, 67 discontinued patients below the set BCR::ABL1 digital PCR cutoff were included. The overall MolR probability was 50% (95% CI, 36%-62%). In 38 patients treated for more than 6 years as commonly recommended as desirable treatment duration before TFR attempt, the MolR probability dropped to 36% (95% CI, 18%-50%). Patients attempting an early TKI discontinuation (treated for less than 6 years) had a high MolR probability of 76% (95% CI, 65%-89%). BCR::ABL1 digital PCR was successfully used in Dutch clinical practice. Our study indicates that in patients with a low BCR::ABL1 digital PCR result, a total TKI treatment duration of six or more years remains associated with a lower MolR rate and should generally be pursued. In patients treated for more than 6 years, BCR::ABL1 digital PCR was capable to identify stop candidates with a higher probability of TFR success.

Keywords: BCR::ABL1 digital PCR; chronic myeloid leukemia; treatment‐free remission; tyrosine kinase inhibitor treatment discontinuation.

FIGURE 1

Flow chart of sample assessments and patient inclusion. A total of 167 samples in 135 patients were assessed. Six samples were excluded because of dPCR failure, four samples were excluded because of ABL1 too low (= less than 100 000 ABL1 transcripts). Some patients had serial dPCR assessments (Sample 2 and Sample 3) with a minimum interval of 6 months. dPCR LOW = BCR::ABL1 dPCR < 0.0023%IS, dPCR HIGH = BCR::ABL1 dPCR > 0.0023%IS. *Two patients decided to not discontinue their TKI treatment despite a low dPCR result. **Four patients were excluded because of time discrepancy, that is, their sample was taken > 4 months prior to TKI stop or after TKI stop.

FIGURE 2

Probability of molecular recurrence in patients with a BCR::ABL1 digital PCR (dPCR) result below the prediction cutoff of 0.0023%IS. (A) In patients treated with a tyrosine kinase inhibitor (TKI) for more or less than 6 years. (B) In patients treated with imatinib or second generation TKI (2GTKI).