Background: Imatinib plasma levels have been reported to correlate with response to therapy and possibly toxicity (Larson et al, Blood 2008; Picard et al, Blood 2007). These studies included patients taking standard dose, once daily imatinib. However, patients are frequently receiving imatinib at different doses and schedules that do not permit extrapolation of these results. Aims: To investigate the significance of plasma levels in patients receiving imatinib at various doses and schedules. Methods: Patients with CML in chronic (CP) or accelerated phase (AP) receiving imatinib at daily doses of 300 to 800 mg, on a once or twice daily schedule had their plasma levels measured. Patients were asked not to take imatinib the day of the assessment until after the levels were drawn. Plasma levels were measured by high performance liquid chromatography. Results: Imatinib plasma levels have been measured in 120 patients. Fifty-nine levels were trough levels: 35 on a once daily (QD) schedule (2 at 300mg/d, 25 at 400mg/d, 7 at 600mg/d 1 at 800mg/d) measured 24 hrs after last dose, and 24 on a twice daily (BID) schedule (16 at 800mg/d, 5 at 600mg/d, 3 at 400mg/d) measured 12 hrs after last dose. In 39 pts receiving once daily dosing, levels were measured 12 hrs after last dose as pts took imatinib in the evening because of convenience/better tolerance. Thirteen pts had levels measured at 18 hrs, and 9 at other times. Among pts with trough levels measured, median plasma levels were 1300 ng/ml (range, 388–3740). There was a strong correlation between dose and level (median in ng/ml: 943 for 300mg, 1190 for 400mg, 1310 fro 600mg, and 2180 for 800mg; p=0.0003). There was no correlation between plasma levels and gender (median in ng/mL for 28 female pts 1310, for 31 male 1300; p=0.46). Interestingly, pts receiving (n=7; 943ng/mL) concomitant CYP3A4 inhibitors had a trend for lower plasma levels than those not receiving such agents (n=52; 1335ng/ml) (p=0.06).Trough levels by response at time of assessment are shown in table 1: Response No. Median trough levels (ng/mL) p CCyR 52 1310 0.82 No CCyR 7 1240 MMR 31 1350 0.22 No MMR 28 1230 Regression analysis showed no correlation between plasma levels and age (p=0.58), BCR-ABL/transcript levels (p=0.50), neutrophil count (p=0.58), hemoglobin (0.74), or platelet count (0.47). Among 39 pts receiving QD schedule with measures at 12 hrs, median plasma levels were 1680 ng/mL (range, 600 to 3610). There was no correlation between levels and response in this group either. Repeat plasma levels were obtained in 25 pts (18 trough, 7 12 hr non-trough) with a median variability of 13.9%. Median plasma levels for pts measured 36–48 hrs after last dose were 388ng/ml (range, 280–418). Conclusion: Plasma levels measured in patients receiving different doses and schedules of administration in a common practice setting have no correlation with response or toxicity. Further studies are required to determine the applicability of imatinib plasma levels with different schedules of administration.

No abstract