To explore the value and threshold of steady-state trough plasma concentration (trough concentration) of imatinib mesylate (IM) and its active metabolite N-desmethyl imatinib (NDI) in predicting the risk of moderate to severe adverse reactions in patients with gastrointestinal stromal tumors (GIST).

The subjects were selected from GIST outpatient who received IM treatment and re-visited doctor in the First Affiliated Hospital of Soochow University form July 2020 to March 2021. On the day of re-visiting, the relevant clinical information and occurrence of IM-related adverse reactions within 28 d prior to the trial of selected patients was asked and recorded and blood were collected (22 to 24 hours after the last medication) to determine IM and NDI trough concentration. Twenty-eight days after blood collection, telephone follow-up was conducted to record IM-related adverse reactions occurrence. Blood routine and blood biochemical examination results within 28 days before and after blood collection were collected through the hospital information system. After evaluating causality for adverse reactions and grading their severity, patients without or with grade I adverse reactions were regarded as the no/mild group and those with grade Ⅱ~Ⅴadverse reactions were regarded as the moderate-severe group. The risk factors of moderate-severe adverse reactions were analyzed by comparing the main clinical characteristics of patients in the 2 groups, and the value and threshold of IM and NDI trough concentrations in predicting the risk of moderate-severe adverse reactions were analyzed by receiver operating characteristic (ROC) curve.

A total of 119 patients were recruited in this study and 113 (95.0%) had adverse reactions. There were 65 patients in the no/mild group and 54 in the moderate-severe group. The differences in the gender and dose distribution of patients in the 2 groups were statistically significant (χ²=19.772, P<0.001; χ²=9.817, P=0.020); proportions of females, patients at dose of 300 mg/d, and patients at dose of 600 mg/d in the moderate-severe group were greater than those in the no/mild group. The trough concentration of IM and NDI of patients in the moderate-severe group were significantly higher than those of patients in the no/mild group[1 695 (1 258, 2 261) μg/L vs. 1 360 (938, 1 643) μg/L, P<0.001; 324(223, 379) μg/L vs. 264(217, 338) μg/L, P=0.042]. ROC curve analysis results showed that the breakpoints of IM and NDI trough concentrations for moderate-severe adverse reactions in patients with GIST were 1 539 μg/L (sensitivity 62.3%, specificity 70.3%) and 303 μg/L (sensitivity 56.6%, specificity 68.7%) respectively. The 119 patients were grouped according to the breakpoint concentrations. The incidences of moderate-severe adverse reactions were 63.0% (34/54) and 30.8% (20/65) in patients with IM trough concentration >1 539 μg/L and ≤1 539 μg/L, respectively and 59.6% (31/52) and 34.3% (23/67) in patients with NDI trough concentration >303 μg/L and ≤303 μg/L, respectively. The differences were statistically significant ( P<0.001, P=0.006).

The trough concentrations of IM and NDI are of some value in predicting the risk of moderate-severe adverse reactions in patients with GIST. Drug monitoring should be strengthened in patients with IM trough concentration>1 539 μg/L and NDI trough concentration >303 μg/L to ensure the safety of IM use.

Gastrointestinal stromal tumors; Imatinib mesylate; Drug-related side effects and adverse reactions; Active metabolites; Steady-state trough plasma concentration; N-desmethyl imatinib