Individual Decrease of BCR-ABL Transcript Level Within 3 Months of Imatinib Treatment Predicts Survival in Patients With Chronic Myeloid Leukemia
• BCR-ABL level analysis of CML-Study IV[1]
Summary of Key Conclusions
• Levels of BCR-ABL on the International Scale (IS) used for assessing patient progress do not reflect actual BCR-ABL transcript levels at baseline or after 3 months on imatinib treatment
• In this study population, individual reduction of BCR-ABL transcripts to 0.35-fold of baseline gives superior risk group stratification compared with 3-month BCR-ABLIS < 10%
o 5-year PFS difference: 19%
o 5-year OS difference: 15%
• No prognostic significance for survival identified from baseline BCR-ABL
• Decline of BCR-ABL transcripts from baseline to 3 months may provide better identification of high-risk patients in need of treatment intensification
Background
• BCR-ABL > 10%IS at 3 and 6 months currently recommended to assess patient response and determine treatment course[2]
o Does not account for baseline transcript levels
o Slope of change from baseline could provide more information about response to tyrosine kinase inhibitor therapy and biology of disease than endpoint analysis alone
• Exact individual decline of BCR-ABL measured as reduction ratio: BCL-ABL transcript at 3 months/BCR-ABL at diagnosis
• Current analysis evaluated predictive and prognostic value of individual baseline BCR-ABL transcript levels and slope of decline over 3 months of imatinib treatment in patients with chronic myeloid leukemia
Summary of Study Design
• Analysis of gene transcripts in blood samples obtained at baseline and 3 months after starting imatinib
o CML-Study IV patients[3]
o Median follow-up: 4.8 years
• Patients treated before obtaining baseline blood sample excluded from analysis
• Patients received imatinib-based treatment
o Imatinib 400 mg/day
o Imatinib 400 mg/day plus interferon alfa
o Imatinib 400 mg/day plus cytarabine
o Imatinib 800 mg/day
Baseline Characteristics
• 301 evaluable patients
Description of Current Analysis
• BCR-ABL levels at baseline and 3 months assessed by quantitative real-time polymerase chain reaction
o BCR-ABL converted to IS with lab-specific conversion factor
o Total ABL
o Beta-glucuronidase (GUS) as reference gene rather than ABL sequence to ensure linearity of BCR-ABL scale
• Predictive cutoff for transcript reduction defined by stepwise comparing HR in 0.05 steps
o HR 5.6 identified as optimal cutoff for OS prediction
o Translates to 0.35-fold reduction in baseline BCR-ABL transcript levels (or 0.46 log reduction)
• Exploratory landmark analyses of PFS and OS to evaluate prognostic factors
o Baseline BCR-ABL/GUS ratio
o Reduction of BCR-ABL/GUS from baseline to 3 months
o BCR-ABLIS < 10% at 3 months
Main Findings
• Actual BCR-ABL transcript levels vary widely between patients
o Range BCR-ABLIS: 1% to 230% of GUS reference gene expression
o Median BCR-ABLIS at baseline: 33%
o Median BCR-ABLIS at 3 months: 1.4% for 0.04-fold or 1.4 log reduction
• 0.35-fold BCR-ABL reduction at 3 months allowed identification of high- and low-risk groups for 5-year survival (HR: 5.6; 95% CI: 2.3-13.4; P < .001)
o In this patient sample, stratification based on 3-month BCR-ABL reduction predicts 5-year PFS and OS better than BCR-ABLIS at 3 months
• No prognostic significance found for PFS or OS identified from BCR-ABL baseline
References
1. Hanfstein B, Shlyakhto V, Lauseker M, et al. Optimizing early prediction of outcome in CML using the exact decline of BCR-ABL transcript levels within 3 months of imatinib treatment as a prognostic marker. Program and abstracts of the 55th American Society of Hematology Annual Meeting and Exposition; December 7-10, 2013; New Orleans, Louisiana. Abstract 253.
2. Baccarani M, Deininger MW, Rosti G, et al. European LeukemiaNet recommendations for the management of chronic myeloid leukemia: 2013. Blood. 2013;122:872-874.
3. Hehlmann R, Lauseker M, Jung-Munkwitz S. et al. Tolerability-adapted imatinib 800 mg/d versus 400 mg/d versus 400 mg/d plus interferon-alpha in newly diagnosed chronic myeloid leukemia. J Clin Oncol. 2011;29:1634-1642.