We evaluated the area under the plasma concentration-time curve (AUC) of afatinib required to avoid the onset of grade 2 or higher diarrhea. The C0 and AUC0-24 of afatinib were significant higher in patients with grade 2 diarrhea than in those with grade 0-1 diarrhea. The areas under the receiver operator curves were 0.795 with the highest sensitivity (89%) and specificity (74%) at an AUC0-24 threshold of 823.5 ng·h/mL, and 0.754 with the highest sensitivity (89%) and specificity (74%) at a C0 threshold of 28.5 ng/mL. In Kaplan-Meier analysis based on these cut-off AUC0-24 and C0 values, the median time to the incidence of grade 2 diarrhea was 16 days. The predicted AUC0-24 of afatinib from the single point of C6 showed the highest correlation with the measured AUC0-24 (r2 = 0.840); however, a significant correlation between the AUC0-24 and C0 was also observed (r2 = 0.761). C0 could be used as a marker of therapeutic drug monitoring because afatinib C0 was related to AUC0-24. Therefore, afatinib C0 should be monitored on day 8 after beginning therapy, and the daily dose of afatinib should be adjusted as an index with a cut-off value of 28.5 ng/mL.
Keywords: afatinib; diarrhea; limited sampling strategy; plasma concentration; therapeutic drug monitoring.