A Study to Evaluate Corrected QT Interval and Drug-Drug Interaction of Trastuzumab on Carboplatin in the Presence of Docetaxel in Patients With HER2-Positive Metastatic or Locally Advanced Inoperable Cancer

Triplicate 12-lead electrocardiogram (ECG) measurements (each recording separated by approximately 2 minutes) were performed and average was calculated. The time corresponding to beginning of depolarization to repolarization of the ventricles (QT interval) was adjusted for RR interval using QT and RR from each ECG by Fridericia's formula (QTcF = QT divided by cube root of RR). Trastuzumab steady state was defined as the average of the 2 ECG measurements collected on Cycle 1 Day 8 (C1D8) and Cycle 2 Day 1 (C2D1) after the trastuzumab infusion.

AUC0-6hr = Area under the plasma concentration versus time curve from 0 to 6 hours post-infusion.

Dose normalized Cmax is the maximum observed concentration of carboplatin in plasma normalized for different dose levels.

The geometric mean ratio of Cmax of carboplatin was defined as the Cmax/D of carboplatin on Cycle 1 Day 1 (in the absence of trastuzumab) divided by Cmax/D of carboplatin on Cycle 2 Day 1 (in the presence of trastuzumab).

AUC0-6hr/D = Area under the plasma concentration versus time curve from 0 to 6 hours post-infusion, normalized by carboplatin dose level.

The geometric mean ratio of AUC0-6hr/D of carboplatin was defined as the AUC0-6hr/D of carboplatin on Cycle 1 Day 1 (in the absence of trastuzumab) divided by AUC0-6hr/D of carboplatin on Cycle 2 Day 1 (in the presence of trastuzumab).

Plasma decay half-life is the time measured for the plasma concentration to decrease by one half.

Triplicate 12-lead ECG measurements (each recording separated by approximately 2 minutes) were performed and average was calculated. The time corresponding to beginning of depolarization to repolarization of the ventricles (QT interval) was adjusted for RR interval using QT and RR from each ECG by Bazette's formula (QTcB = QT divided by square root of RR). Trastuzumab steady state was defined as the average of the 2 ECG measurements collected on Cycle 1 Day 8 and Cycle 2 Day 1 after the trastuzumab infusion.

For each postbaseline timepoint, a participant's corresponding baseline measure was subtracted from his or her average of the triplicate ECG measure to create a "baseline-adjusted" corresponding ECG measure for each participant at each postbaseline timepoint.

For each postbaseline timepoint, a participant's corresponding baseline heart rate was subtracted from his or her average of the triplicate heart rate to create a "baseline-adjusted" corresponding heart rate for each participant at each postbaseline timepoint.

Triplicate 12-lead ECG measurements (each recording separated by approximately 2 minutes) were performed and average was calculated. The time corresponding to beginning of depolarization to repolarization of the ventricles (QT interval) was adjusted for RR interval using QT and RR from each ECG by Fridericia's formula (QTcF = QT divided by cube root of RR) and by Bazette's formula (QTcB = QT divided by square root of RR). Participants with maximum QTc less than or equal to (<=) 450 msec, greater than (>) 450 to <=470 msec, >470 to <= 500 msec, or >500 msec were reported.

Triplicate 12-lead ECG measurements (each recording separated by approximately 2 minutes) were performed and average was calculated. The time corresponding to beginning of depolarization to repolarization of the ventricles (QT interval) was adjusted for RR interval using QT and RR from each ECG by Fridericia's formula (QTcF = QT divided by cube root of RR) and by Bazette's formula (QTcB = QT divided by square root of RR). Participants with maximum increase from baseline of =>30msec, 30 to <60 msec (borderline) and >=60 msec (prolonged) were summarized.

The incidence of abnormal U-wave changes from baseline was determined based on centrally read ECG tracings comparing each of the three triplicate readings from the post baseline ECG time points to the baseline ECG reading. At each time point, if at least one of the three triplicate readings was abnormal, the participant was counted as abnormal for that ECG timepoint as follows: a large U wave, inverted U wave, or T-U fusion compared with baseline was considered an abnormal significant change from baseline.

The incidence of abnormal T-wave changes from baseline was determined based on centrally read ECG tracings comparing each of the three triplicate readings from the post baseline ECG time points to the baseline ECG reading. At each time point, if at least one of the three triplicate readings was abnormal, the participant was counted as abnormal for that ECG timepoint as follows: an inverted T, flat T, or biphasic T compared with baseline was considered an abnormal significant change from baseline. Additionally, nonspecific T-wave changes from baseline were considered as abnormal nonsignificant changes from baseline. T-wave changes from baseline due to ventricular conduction or left ventricular hypertrophy strain were considered not evaluable.

Criteria for abnormal changes in PR interval were defined as: =>25 percentage (%) change from baseline, an absolute value >200 msec, or >=25% change from baseline and an absolute value >200 msec.

Criteria for abnormal changes in QRS interval were defined as: >=25% change from baseline, an absolute value >110 msec, or >=25% change from baseline and an absolute value >110 msec.

As per planned analysis, separate population pharmacokinetic analysis results are not available for the current study as this analysis is based on pooled data from multiple studies.