A Phase I Study of Midostaurin (PKC412) Investigating Cardiac Interval Effects In Healthy Subjects

Abstract 4362

Midostaurin (PKC412) is a multi-targeted tyrosine kinase inhibitor (TKI) of several receptors, including wild-type and mutant variants of KIT and the FMS-like tyrosine kinase 3 (FLT3) receptor, and has known roles in hematopoiesis and leukemia. Midostaurin has demonstrated activity in acute myeloid leukemia (AML) and myelodysplastic syndrome in phase 1 and 2 trials, and is currently under investigation in a randomized phase 3 AML study at 50mg twice daily (bid) in combination with chemotherapy and a phase 2 monotherapy study of aggressive systemic mastocytosis (ASM) at 100mg bid. Despite the absence of specific midostaurin-related cardiac toxicity issues, we conducted a dedicated phase 1 study to directly investigate the effect of midostaurin on QTc interval. Healthy subjects were randomized to 3 treatment arms: placebo; midostaurin administered orally at 75mg bid on days 1 and 2 and once daily (od) on day 3; or an active control arm of moxifloxacin administered orally at 400mg od on day 3. The primary variable was QTcF interval on day 3 corrected for baseline and placebo in the midostaurin and moxifloxacin treatment arms. Drug exposure levels at each time point were confirmed for moxifloxacin, midostaurin, and its two metabolites – CGP62221 and CGP52421. Of 192 subjects enrolled, 166 completed the study. 24 of 80 subjects discontinued in the midostaurin arm: 19 (23%) due to adverse events (AEs), 17 of which encompassed expected gastrointestinal events. No patients were discontinued for AEs in the other 2 treatment groups. Discontinued patients were not included in the ECG analysis. In time-matched analysis of QTcF interval change, the maximum mean change in the midostaurin arm corrected for baseline and placebo was 0.72ms with a 90% confidence interval (CI) upper bound of 4.71ms, which excluded 10ms. At each nominal time point, the mean change from baseline placebo-corrected for midostaurin was <0ms. The QTcF change point estimate corrected for time-averaged baseline and placebo also showed a lack of QTc prolongation for midostaurin. Moxifloxacin had a maximum mean change corrected for baseline and placebo of 10.7ms with a lower unadjusted 90% CI of 6.4ms 1 hour post-dose on day 3. Plasma concentration vs QTcF change from baseline analysis confirmed a negative or lack of QT effect by midostaurin but a positive correlation for moxifloxacin. No symptomatic, clinically significant new post-baseline morphological abnormalities were identified in the study. 3 patients in the midostaurin group at a single time point or evaluation experienced new post-baseline T-wave abnormalities, as did 1 and 4 patients in the placebo and moxifloxacin groups, respectively, some at multiple time points. No subject had a new >30ms or >480ms change from baseline for QTcF or QTcI. For QTcB the only occurrences of change were in the 30–60ms category: 1 (1.3%) of the subjects on midostaurin met this non-specific outlier criterion; 7 (15.9%) on moxifloxacin; and 1 (1.5%) on placebo. 1 new U-wave abnormality was noted in the moxifloxacin group. The peak plasma concentration of midostaurin achieved in the present study (mean 2273ng/mL) covered the peak and trough plasma exposure observed at 50mg bid (2220ng/mL and 1005ng/mL, respectively) in AML patients. The peak level achieved for midostaurin was also above the steady-state trough level of 1060ng/mL, but below the peak concentration of 3500ng/mL, for the 100mg bid dose. Midostaurin was safe and generally well tolerated: 97% of the AEs noted in subjects while on study drug (n=61; 40%) were reported as grade 1. No grade 3/4 AEs were reported. While some TKIs exert pharmacologic effects on QTc interval, this carefully conducted trial demonstrates that midostaurin at 75mg bid has no effect on heart rate, AV conduction, or cardiac depolarization. The midostaurin exposure achieved in this study exceeds the peak and trough levels for the 50mg bid dose regimen under investigation in the AML phase 3 trial. The midostaurin exposure achieved also exceeds the steady state trough level, but not the peak level, of the 100mg bid dose regimen under investigation in the phase 2 ASM trial. Further, the effects of the long-acting metabolite CGP52421 cannot be fully addressed by this short study. Due to the lack of QT prolongation observed in this trial, we recommend reduced but continued ECG monitoring and omission of QT-related exclusion criteria in future midostaurin clinical trials.