A Multi-Institutional Retrospective Analysis of Tyrosine Kinase Inhibitor (TKI) Clinical and Preclinical Efficacy According to BCR-ABL Mutation Status in CP-CML Patients

Introduction

Dozens of secondary mutations in BCR-ABL have been identified in CP-CML patients (pts) whose disease becomes resistant to imatinib. Four TKIs (dasatinib, nilotinib, bosutinib and ponatinib) are potentially available as second (2L) or later line treatment, but no comprehensive clinical dataset or validated preclinical methodology exists to inform the choice of therapy based on BCR-ABL mutation status. While an extensive clinical dataset is available that describes response rates observed with dasatinib 2L treatment according to the specific BCR-ABL mutation present at baseline (eg cytogenetic response rates based on at least 10 pts available for 15 mutants), minimal data are available for the other TKIs, including, nilotinib and, especially, bosutinib and ponatinib.¹ A generally more practicable approach is to compare sensitivities of BCR-ABL mutants to each TKI based on in vitro assessments. We have previously compared the in vitro potencies of all approved BCR-ABL TKIs against 21 BCR-ABL mutants, with preliminary results suggesting that TKI potency in vitro could be related to clinical efficacy if clinical pharmacokinetic parameters, and functional effects of protein binding, were taken into account.² However, confirmation that this, or an alternative, methodology can produce reliable associations between in vitro data and clinical efficacy across all TKIs requires a more extensive clinical dataset than is currently available. To address this gap, we sought to leverage the combined resources of multiple institutions to collect outcome data for all available TKIs being used 2L in CP-CML pts who had a single BCR-ABL mutation detected at baseline.  Uniform inclusion criteria were used to enhance the chance that the mutant BCR-ABL clone is the predominant driver of disease, so clinical outcomes could be optimally associated with preclinical data. Here we describe the design and initial results of this effort.

Methods

De-identified data were collected retrospectively based on chart review of appropriately consented pts. Key inclusion criteria include chronic phase disease being treated 2L and the presence of a single BCR-ABL mutation detected by Sanger sequencing. Additional demographic data collected include the 1L TKI and cytogenetic and molecular response status at baseline. Clinical outcome data collected include whether specific cytogenetic (MCyR or CCyR) and/or molecular (MR1, MR2, MMR) responses were achieved on 2L TKI treatment. To facilitate integration of data across multiple institutions, we examined the concordance between cytogenetic and molecular response outcomes. Univariate and multivariate regression will be used to examine associations between clinical responses and in vitro data.

Results

Results are reported as of July 24, 2015, based on data compiled from 157 pts who met all of the inclusion criteria (collected from 5 of the more than 10 institutions participating in this study). The median duration of prior TKI treatment was 3.0 years, with the vast majority of pts having received imatinib 1L (154 pts).  Pts had 37 unique mutations at baseline, with the most common being T315I (N=19), M244V (N=18), M351T (N=13), G250E (N=12), and F359V (N=12). Cytogenetic and/or molecular response data were available for all 157 pts according to treatment 2L with dasatinib (N=81), nilotinib (N=61), ponatinib (N=14), and bosutinib (N=1). The use of MR2 as a surrogate for CCyR was examined in 121 pts evaluable for this comparison: 59 pts achieved both MR2 and CCyR, 58 achieved neither, and 4 achieved MR2, but not CCyR.  MR2 demonstrated 100% sensitivity and 93.5% specificity at determining CCyR in the interim data.

Conclusion

A broad collaborative effort has been successfully initiated to greatly expand the available clinical data relating response outcomes to TKI treatment in CP-CML pts according to baseline mutation status.  Analysis of a substantially expanded clinical dataset and the association with preclinical data will be presented.

References:

1. Baccarani, M., et al, Molecular monitoring and mutations in chronic myeloid leukemia: how to get the most out of your tyrosine kinase inhibitor. Am Soc Clin Oncol Educ Book, 2014: p. 167-75.

2. Gozgit, M.J., et al., Comprehensive Analysis of the In Vitro Potency of Ponatinib, and all Other Approved BCR-ABL Tyrosine Kinase Inhibitors (TKIs), Against a Panel of Single and Compound BCR-ABL Mutants. ASH, 2013: p. Abstract: 3992.