Frequency and Clonality of BCR-ABL Compound Mutations in Chronic Myeloid Leukemia,

BCR-ABL kinase domain (KD) mutations are a common mechanism of chronic myeloid leukemia (CML) resistance to tyrosine kinase inhibitor (TKI) therapy. It is well known that some patients harbor more than one KD mutation in the same sample, but the frequency of true compound mutations (defined as two or more mutations in the same allele) and the clonal relationships between mutant clones have not been established.

The first group of samples (Group 1) was selected based on evidence of more than one BCR-ABL KD mutation by Sanger sequencing. Samples from a second group of patients (Group 2) who had one mutation in BCR-ABL KD by Sanger sequencing were also analyzed. The BCR-ABL KD was amplified using nested RT-PCR, the final amplicon was shotgun-cloned and 10 clones from each sample were sequenced.

Thus far, 18 samples from group 1 have been analyzed. Sequencing of 180 colonies from this group revealed >80 different mutations. Compound mutations were confirmed in 13 samples (Table 1), while 5 samples revealed the two mutations were in separate clones. T315I, F359V, V299L and M351T were over-represented among compound mutations and the T315I/F359V and V299L/M351T compound mutations were each detected in samples from two patients. On average, each patient sample had 7 different mutations per 10 clones (range: 3–14), some were silent or not previously shown to be associated with TKI resistance in biochemical or cell-based assays.

In the 13 samples from Group 1 with confirmed compound mutations (Table 1), all respective individual mutations had been shown to confer TKI resistance in biochemical and/or in vivo studies. In 4 patients, the compound mutant clone was non-dominant (≤40% of clones). In 3 of these 4 patients (11, 13, 16; Table 1), both mutations were also detected independently in co-existent clones, while 1 patient (18) showed 2 compound mutant clones (T315/F359V and V299L/M351T) at comparable levels (40% each). In the remaining 9 patients from Group 1, the compound mutant represented >70% of clones.

More than 2 mutations were observed in 41/102 screened compound mutant clones from Group 1. The number of mutations in these clones ranged from 3 to 6. Except V299L/M351T/E329G, which was observed in 5/10 screened clones of one patient (1), no clone with more than 2 mutations was detected in more than 1 clone per sample (≤10%). Notably, none of these additional mutations has yet been associated with TKI resistance in biochemical or cell-based assays.

So far, samples from 7 patients in Group 2 have been analyzed. Sequencing of the clones from these patients also revealed the presence of low-level compound mutants. On average 30% of the clones had compound mutations (range: 20%-40%) and an average of 5 different mutations were detected among the 10 sequenced clones per patient (range: 3–6), some of which were silent or not previously shown to be associated with TKI resistance in biochemical or cell-based assays.

1. True compound mutations are common in patients with evidence of multiple mutations by direct sequencing.

2. Mutant clones evolve sequentially as well as in parallel, suggesting complex clonal relationships, in which the identical phenotype (mutation) may be acquired independently by multiple clones.

3. Clones harboring compound mutations comprised of more than 2 mutations rarely achieve dominance, suggesting that the number of different mutations compatible with maintenance of BCR-ABL kinase activity may ultimately be limited.

4. Low-level compound mutations are also detected by cloning and sequencing in the samples from patients who have had one detected mutation in screening (Group 2).

Lange:Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; BMS: Honoraria; Pfizer: Honoraria, Membership on an entity's Board of Directors or advisory committees.