Evaluation Of C-KIT Exon 11 Mutation Status In Gastrointestinal Stromal Tumors By Energy Spectrum CT

Objective: This study aims to evaluate the value of conventional CT indexs and quantitative parameters of energy spectrum CT in predicting the mutation status of cKIT exon 11 in gastrointestinal stromal tumors(GISTs),construct predictive models,and draw nomograms to assist in clinical treatment decision-making.Method: 187 patients with confirmed GISTs by pathology at Lanzhou University Second Hospital from January 2018 to December 2022 were retrospectively enrolled.All patients underwent three phase CT enhanced scans and c-KIT gene testing.Among them,61 patients underwent three phase CT enhanced scans of the entire abdomen using energy spectrum imaging(GSI)scanning mode.u Chisquare test,Fisher’s exact probability test,Mann-Whitney U test and independent sample t test were used to compare and analyze the clinical,pathological,conventional CT indexes and energy spectrum CT quantitative parameters of patients.Draw the receiver operating characteristic curve(ROC),calculate the area under the curve(AUC),and determine the optimal threshold,sensitivity,and specificity based on the maximum Jordan index to evaluate the diagnostic efficacy of parameters with statistically significant differences on c-KIT exon 11 mutation.The statistically significant parameters in the above univariate analysis were incorporated into Logistic regression,screen independent predictors for predicting the mutation status of c-KIT exon 11.Draw nomogram of clinical+conventional CT index model and clinical+spectral CT quantitative parameter model for predicting the mutation status of c-KIT exon 11,and evaluate the discrimination,calibration,and clinical net benefit of nomograms using C-index,calibration curve,and decision curve analysis..Results: The mutation rate of c-KIT exon 11 in 187 GISTs patients was 64.17%(120/187).a significant statistical difference in age in the mutation status of c-KIT exon 11(P=0.006).There was no significant statistical difference between the two groups in gender,tumor risk grade,Ki-67 expression and mitotic index(P>0.05).In routine CT index,there were significant statistical differences between the GISTs cKIT exon 11 mutation group and the wild group in terms of maximum lesion diameter(P=0.032),boundary(P=0.013),ulcer(P=0.047),necrosis/cystic change(P=0.016),enhancement pattern(P=0.010),venous phase CT value(P=0.001),delayed phase CT value(P=0.013),venous phase-plain scan CT value(P=0.004),and delayed phase-plain scan CT value(P=0.035).There was no statistically significant difference in the site of the tumor,growth pattern,morphology,calcification,intratumoral bleeding,and thickening of supplying vessels between the two groups in GISTs(P>0.05).In the quantitative parameters of energy spectrum CT,the CT values at a single energy level of 40 ke V–120 ke V(P<0.05),energy spectrum curve slope K(P=0.002),iodine concentration(IC)(P=0.002),normalized iodine concentration(NIC)(P=0.004)in the venous phase,and the CT values at a single energy level of 40 ke V–60 ke V(P<0.05),energy spectrum curve slope K(P=0.002),IC(P=0.002),NIC(P=0.004)in the delay phase of c-KIT exon 11 mutant type is lower than c-KIT exon 11 wild type,The AUC of the CT values at a single energy level of 60 ke V in the venous phase was thr highest(AUC=0.764).The quantitative parameters of arterial phase derived from energy spectrum CT,the CT values at a single energy level of 130 ke V–140 ke V,and water concentrations(WC)in venous phase,the CT values at a single energy level of70 ke V–140 ke V,and WC in delayed phase had no significant difference between the two groups(P>0.05).In different models for predicting the mutation status of c-KIT exon 11 before surgery,the C index of nomogram of the clinical+conventional CT index model was 0.703(95% CI: 0.652-0.799),the sensitivity and specificity were44.2% and 88.1%,respectively.The C-index of nomogram of the clinical+energy spectrum CT quantitative parameter model was 0.827(95% CI: 0.723-0.932),the sensitivity and specificity were 72.5% and 81.0%,respectively.The calibration curve showed the predicted values of the two models are in good consistency with the actual values(P values were 0.668 and 0.724,respectively).Decision curve analysis showed that both nomogram models achieved high clinical net benefits.Conclusion: Age,conventional CT index,the quantitative parameters of energy spectrum CT in venous and delayed phase can be used as preliminary evaluation indicators for the mutation status of GISTs c-KIT exon 11.The clinical combined energy spectrum CT quantitative parameter model has high efficiency in predicting the mutation status of GISTs c-KIT exon 11,and can assist in clinical risk stratification and determining treatment plans for patients.