Molecular Responses to Dasatinib and Nilotinib in Patients with Chronic Myeloid Leukemia in Chronic Phase (CML-CP).

Background: Nilotinib (NIL) and dasatinib (DAS) have a 1- and 2-log higher inhibitory potency than imatinib against ABL1 kinase respectively, and are active against all BCRABL1 kinase mutants, except T315I. We investigated the molecular responses to nilotinib and dasatinib in patients (pts) with CML-CP receiving these agents either in the frontline (FL) or the post-imatinib failure (PIF) settings.

Methods: 205 pts with CML-CP received either NIL (n=87; 48 FL, 39 PIF) or DAS (n=118; 48 FL, 70 PIF) at various doses in phase II studies. Quantitative reverse transcription PCR in peripheral blood samples was performed prior to NIL or DAS start, after 1 mo of therapy, and every 3 mo thereafter.

Results: Table 1 illustrates the BCR-ABL1/ABL1 ratios at different time-points during therapy. No significant differences were observed regarding the median baseline BCRABL1/ ABL1 ratio across groups (p=0.88) or the median at 12 mos between the NIL and the DAS cohorts (p=0.14). We developed an exponential model with one population-based shape parameter and a subject-specific slope and intercept. We estimate the shape parameter, after merging all data from a treatment group, using non-linear least squares. With the shape parameter fixed, we estimate the subject-specific parameters using traditional least squares. A statistically significant difference between the shape parameters of the NIL-FL and the DAS-FL treatments was observed (p=0.005), with NIL inducing a sharper early decline in BCR-ABL1 transcripts. Our exponential model does not fit as well for NIL-PIF vs DAS-PIF, particularly for the latter, possibly due to differences in the number and types of mutations at the start of therapy. Indeed, DAS-resistant mutations (L248V/R, Q252H, E255K, V299L, T315I/A, and F317L/C/I/S/V) were detected in 7 (32%) of 22 patients carrying mutations (all PIF), while NIL-resistant mutations (Q252H, Y253H/F, E255K/V, T315I/A, F359V, V379I) were only found in 1/16 (6%) carrying mutations (all PIF). BCR-ABL1/ABL1 ratio reductions occurred in 73 (87%)of 84 pts who had at least 2 PCR analyses during NIL therapy: <1-log in 7 (8%) pts, >1-log in 16 (19%) pts, >2-logs in 18 (21%) pts, and >3-logs in 32 (38%) pts. BCR-ABL1/ABL1 ratio reductions occurred in 102 (92%) of 111 pts who had at least 2 PCR analyses during DAS therapy: <1-log in 20 (18%) pts, >1-log in 17 (15%) pts, >2-logs in 24 (22%) pts, and >3-logs in 41 (37%) pts. Major molecular response (MMR) and complete molecular response (CMR; undetectable BCR-ABL1 transcripts) rates were 39%/12% and 38%/7% for the NIL and DAS cohorts, respectively (50%/17% and 58%/7% for pts receiving NIL and DAS as frontline therapy). The MMR and CMR response rates for NIL and DAS among pts with mutations at the start of therapy were 19%/6% and 20%/0%, respectively.

Conclusion: NIL and DAS induce molecular responses in a significant number of pts, particularly when used in the frontline setting. Although molecular responses occur across a broad variety of BCR-ABL1 kinase mutations, CMR in this setting is a rare occurrence. NIL appears to induce faster molecular responses than DAS, al least in the FL setting. Longer follow-up is necessary to establish the clinical consequences of this phenomenon.

Table 1.BCR-ABL1transcript dynamics by treatment and cohort