Molecular Predictive Markers in Dose Escalation Treatment for Suboptimal Responders to Standard Dose Imatinib in CML

Introduction Imatinib mesylate (IMT) dose escalation has been proposed as a therapeutic option in patients (Pts) with chronic myeloid leukemia (CML) who failed to achieve optimal response with standard dose IMT. We report the results of prospective multi-center single arm phase ¥≥study evaluating efficacy of escalated dose IMT. We intended to identify patterns of molecular change using serial quantitative RT-PCR and its relationship with clinical outcome. We also planned to find predictive markers for outcome with array comparative genomic hybridization (aCGH) and epigenetic study of bcr gene in addition to BCR/ABL mutation.

Patient and methods Pts in chronic phase (CP) CML who failed to achieve optimal response by European LeukemiaNET with adequate organ function were enrolled. Pts in accelerated phase (AP) or blast crisis (BC) who failed to achieve complete hematologic response after 3 months of IMT were also eligible. CP Pts received 600mg daily, while Pts in AP or BC received 600 or 800mg IMT daily. Pts received IMT for at least 12 months or until the appearance of a progressive disease, intolerable toxicity. Along with cytogenetic response (CyR), molecular response (MR) was assessed with BCR-ABL/ABL gene ratio of peripheral blood or bone marrow aspirate. Baseline BCR/ABL gene mutation test was performed using Matrix-assisted laser desorption/ionization time of flight mass spectrometry. Genome-wide screening for regions of genetic gains and losses with baseline blood samples was performed for 38 Pts using aCGH. Methylation status of 4 CpG sites in bcr gene promoter region was tested for 40 Pts and average methylation level was used for analysis. Blood samples at baseline and 6 months after dose escalation were tested. 29 optimal responders to standard dose IMT and 38 healthy donors were also tested for bcr methylation status for additional comparison.

Results 71 Pts (median age 49.0 years, M:F=50:21) received escalated dose IMT. Median time to treatment failure (TTTFx) was 18.0 months and toxicities were manageable. 44 and 52 Pts were evaluable for FISH at 6 months and 1 year, where 16 and 17 Pts showed complete CyR (CCyR) respectively. For 61 Pts with serial MR data, TTTFx was longer in Pts who achieved molecular reduction of more than 50% within 6 months (Molecular early responder: MER) than who didn’t (p<0.001). MER’s achieved CCyR more frequently at 6 months and 12 months (p=0.010, <0.001 respectively). Of 24 Pts who had mutational status data, 4 had mutation. They experienced TFx within 12 months and all failed to achieve CCyR. aCGH revealed significant copy number (CN) gain in chromosome 16p11.2 in MER’s compared to non-MER’s (p=0.034). Tendency for increased CN in 22q11.23 and decreased CN in 17q12 was observed in MER’s without reaching statistical significance (p=0.072 and 0.070 respectively). 4 candidate genes within the above regions – GSTT1, SULTA1A, PYCARD, TADAZL – were evaluated for CN variation. GSTT1 CN loss was more frequently observed in MER’s (p=0.035). GSTT1 CN loss also predicted the longer TTTFx without reaching statistical significance (p=0.086). In epigenetic study, Pts in PCyR at the time of study enrollment had increased baseline bcr methylation compared to Pts in less than PCyR (p<0.001). Pts who had increased amount of bcr methylation at 6 months compared to baseline had longer TTTFx compared to who did not (p=0.012). Baseline bcr methylation amount of study Pts was lower when compared to that of optimal responders and healthy donors (p=0.001 and p<0.001 respectively). bcr methylation decreased with increased duration of standard dose IMT both in study Pts and optimal responders (p=0.042 and 0.004 respectively), although the pattern of decrease was different between the two groups (p<0.001). In multivariate analysis baseline bcr methylation status was the only variable related to TTTFx (p=0.047).

Conclusion Escalated dose IMT is a reasonable option for CML Pts showing less than optimal response to standard IMT. MER after escalated dose IMT is a useful early predictive marker for long term response. Mutational status of BCR-ABL at baseline is possibly important for response. Chromosome 16p11.2, 22q11.23 and 17q12 are potential locations related to IMT response and GSTT1 CN loss may be a genetic change affecting clinical outcome. bcr methylation status is an epigenetic marker associated with IMT response, where decreased bcr methylation status is related to poor IMT response.