Early Detection of Imminent Hematologic Relapse Due to BCR-ABL Kinase Domain Mutations in CML Patients on Imatinib Therapy by D-HPLC.

Mutations of the BCR-ABL tyrosine kinase domain constitute the major cause of resistance in chronic myeloid leukemia (CML) patients (pts) on imatinib monotherapy. Sensitivity of conventional sequencing may allow the detection of a proportion of 10% mutated cells. We sought to improve the diagnostic armamentarium to screen for mutations prior to frank relapse. A total of 95 pts (chronic phase, CP, n=47; accelerated phase, AP, n=27; myeloid blast crisis, BC, n=19; lymphoid BC, n=2) who relapsed during imatinib therapy were screened for BCR-ABL kinase domain mutations applying denaturing high performance liquid chromatography (D-HPLC) and direct sequencing. Mutations were detectable in 47/79 (59%) pts with hematologic relapse and in 8/16 (50%) pts with cytogenetic relapse. 22 different point mutations affecting 18 amino acids and a novel deletion of 81bp of ABL exon 4 mapping to the P-loop region of the kinase domain were observed. To investigate the dynamics of the mutated clones D-HPLC was applied to 453 cDNA samples tracking back from relapse towards start of imatinib therapy. D-HPLC was optimized to detect 0.1–0.5% BaF3^BCR-ABL cells harboring various mutations in a background of unmutated cells. Hematologic relapse occurred after a median of 12.9 mo (range 0.9–44.2) of imatinib therapy. However, BCR-ABL mutations became first detectable by D-HPLC at a median of 5.8 mo (range 0.0–30.5) after commencing imatinib (p<0.0001). Nine pts (18%; CP, n=4; AP, n=3, myeloid and lymphoid BC, n=1 each) showed evidence for BCR-ABL mutations even prior to imatinib therapy (T315I, n=4; M351T, n=3; M244V, Y253H, n=1 each). The median interval between first detection of the mutation and relapse was 5.6 mo in CP, 8.1 mo in AP, and 2.4 mo in myeloid BC. P-loop mutations were revealed at a median of 2.8 mo, activation loop mutations 2.9 mo, and T315I 6.3 mo prior to relapse. Cytogenetic relapse occured at a median of 19.2 mo (range 10.3–36.3) after start of imatinib therapy. BCR-ABL mutations became first detectable at a median time of 15.8 mo (range 0.0–26.4) after commencing imatinib (difference n.s., p=0.061). In two pts BCR-ABL mutations (M244V in AP, L324Q in CP) were observed in a small clone prior to imatinib therapy. Twenty-five CP pts in continuous complete cytogenetic remission were screened for BCR-ABL mutations to determine the predictive value of minor clones harboring mutations for consecutive relapse. No mutation was found in any of these pts. We conclude that (i) D-HPLC is a reliable and sensitive method to screen for BCR-ABL mutations before and during therapy with tyrosine kinase inhibitors. (ii) The observation of BCR-ABL mutations during imatinib therapy is predictive for relapse. (iii) Mutations may be detectable several months before hematologic relapse, and (iv) the sensitive detection of small mutated clones could provide clinical benefit by triggering early therapeutic interventions, which should be demonstrated in prospective clinical trials.