Abstract 3751

Background and Purpose:

The introduction of imatinib to the treatment regimen of chronic myeloid leukemia (CML) has markedly changed its outcome. Several reports have shown that the efficacy of imatinib correlates with the trough plasma concentration in CML patients. However, little is known about the correlation between plasma concentration and long-term outcomes, including adverse effects. We analyzed the correlation between long term tolerability and plasma concentration in Japanese patients.


Forty-three patients with chronic phase CML who had good adherence with daily 300mg (n=8) or 400mg (n=35) imatinib administration at hospitals in our study group were enrolled in this study. The median age of patients starting imatinib was 57 years (range 15–82). Imatinib was started between Dec. 2001 and May 2007. Plasma trough concentration was examined on a day between Nov. 2007 and Aug. 2008. Median time from the start of imatinib until the plasma assay was 1548 days (range 91–2390).


Clinical data were reviewed in Sep 2011. We excluded patients who had been switched from imatinib to a second tyrosine kinase inhibitor (TKI) despite major molecular response (MMR). Median plasma concentration was 925 ng/ml (range 200–3280). Median follow-up was 2632 days (range 884–3496) and median follow-up after the plasma assay was 1216 days (range 361–1323). We divided the patients into two groups: the low plasma concentration group (<925ng/ml n=22) and high plasma concentration group (>925ng/ml n=21). Overall survival and achievement of complete molecular response (negative by RT-PCR) did not differ between the two groups. We defined all death, progression, switch to a second TKI and imatinib dose reduction as events. Also we defined death by adverse effect, imatinib dose reduction or switch to a second TKI due to adverse effects as intolerance-related events and progression, switch to a second TKI due to failure of achieve MMR or loss of MMR as efficacy-related events. Event-free survival (EFS) showed no difference between the two groups. However cumulative incidence of intolerance-related event was significantly higher in the high plasma concentration group than in the low plasma concentration group (whole observation period, p=0.02 by log-rank test; observation period after the plasma assay, p=0.01 by log-rank test). Grade 3 or 4 anemia occurred in 4 patients only in the high plasma concentration group. Cumulative incidence of efficacy-related event was slightly higher in the low plasma concentration group than the high plasma concentration group (whole observation period, p=0.06 by log-rank test; observation period after the plasma assay p=0.05 by log-rank test).


Trough imatinib plasma concentration at steady-state after initiation of therapy has been demonstrated to have correlation with the early efficacy of imatinib. Our findings suggest that imatinib plasma concentration also correlates with intolerance or adverse effect in long-term clinical observation. Although further studies are required to confirm the current data prospectively, imatinib plasma level testing seems to be helpful to optimize clinical outcomes for patients with CML.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.