Molecular Response at 3 Months Measured with Genexpert BCR-ABL (IS) Platform Predicts Further Outcome in Chronic Myeloid Patients but the Cutoff Differs from the 10% Commonly Used with BCR-ABL (IS) EUTOS Method

INTRODUCTION: In chronic myeloid leukemia (CML) patients in chronic phase (CML-CP), BCR-ABL levels ≤10% at 3 months measured by RT-qPCR (IS) has been consistently correlated with probabilities to obtain an optimal response at 12 months. Monitoring molecular response with automated cartridge-based detection system GeneXpert BCR-ABL (Cepheid^®) method has shown an optimal correlation with standardized BCR-ABL (IS) EUTOS method in patients with complete cytogenetic response (CCyR). However, is not known if both methods are also equivalent when measuring BCR-ABL levels above 1%, and therefore, the utility of GeneXpert  in order to evaluate response at 3 months must be confirmed.

AIMS: To validate the predictive value of molecular response at 3 months with GeneXpert method

METHODS: We have studied 125 new consecutive CML-CP patients treated with tyrosine kinase inhibitors (TKIs) followed in 13 centers. Median age at diagnosed was 55 years. The percentage of low, intermediate and high risk Sokal groups were 42%, 40% and 18% . First line treatment was imatinib (IM), nilotinib (NI), dasatinib (DA) or bosutinib (BO) in 58%, 28%, 13% and 1% of the patients, respectively. BCR-ABL level was measured by GeneXpert platform, where all necessary steps to measure BCR-ABL levels are automatically performed. ABL was used as gene control. The study was approved by the Ethics Committee.

RESULTS: Median follow up was 43 months. The proportion of patients that achieved CCyR by 12 months, analyzed by intention to treat, was 84% (108/123). Probabilities for each specific TKI were 78%, 93%, 100% and 100% for IM, NI, DA and BO respectively. 23% (96/125) of patients required treatment changed due to resistance or intolerance. Treatment discontinuation probabilities were 32%, 11%, 5% and 0% for IM, NI, DA and BO respectively. Only 4% (5/125) did not achieve an optimal response at 3 months (BCR-ABL ≤10%), which is significant lower compare to results obtain with historical series when using EUTOS IS method. 10% cut-off at 3 month was unable to identify patients that achieved an optimal response in further evaluations. By 12 months, this cutoff did not correlate with probabilities to obtain CCyR (50% vs 86% (p=0.1) or major molecular response (MMR) (60% vs 79% (p=0.21)). In order to find a cutoff that could correlate with optimal response at 12 months, we used a receiver operating characteristic curve to identify the optimal cutoff in transcript level that would allow us to classify the patients as high risk or low risk with maximal sensitivity and specificity for each individual outcome. At 3 months, patients with transcript levels ≤ 1.6% had significantly better probabilities to obtain an optimal response by 12 months, with  81% and 94% sensitivity and specificity for CCyR. With this new cutoff, probabilities for CCyR and MMR at 12 months were 98% vs 54% (p<0.001) and 88% vs 56% (p<0.001) respectively (OR: . Finally, this cutoff has also been correlated with probability for treatment changed at any time (46% vs 16% (p=0.005))

CONCLUSIONS: The results of our study seem to show that the  10% threshold, commonly  used to evaluate response at 3 months when using BCR-ABL (IS) EUTOS method, is not associated with probabilities to achieve further optimal responses when using the GeneXpert platform . We have shown how a new cutoff of 1,6% % at 3 months when using GeneXpert could better identify patients with lower risk to achieve an optimal response at 12 months.