[The (18)F-FDG positron emission tomography integrated computed tomography associate with intravoxel incoherent motion in prediction of EGFR expression in lung cancer].

ABSTRACT

Objective: To discuss the diagnostic value potential of (18)F-FDG PET/CT and intravoxel incoherent motion (IVIM) in genotype EGFR of lung cancer. Methods: A total of 116 cases proved pathologically pulmonary space occupying lesions (72 males, 44 females; age range was 31-85 years; 37 cases of high differentiation; 53 cases of middle differentiation, 26 cases of low differentiation) were prospectively collected from August 2017 to March 2019 in Henan Provincial People's Hospital. EGFR gene detection was performed in 50 patients (wild type 20 cases, mutant 30 cases). Whole body PET/CT and lung MR-imaging were performed before treatment in all patients. Comparison of PET/CT, IVIM semi-quantitative parameters between positive and negative of EGFR by Student-t test or Mann-Whitney U test. Evaluation diagnostic efficacy of each parameter to EGFR by drawing ROC curves. The optimum diagnostic threshold, specificity, sensitivity, positive predictive value, negative predictive value and the diagnostic accuracy were calculated. The diagnostic accuracy of (18)F-FDG PET/CT combined with IVIM was calculated. Results: The quantitative parameters standard intake(SUV), apparent diffusion coefficient (ADC), true diffusion coefficient (D), pseudo-diffusion coefficient (D(*)), perfusion fraction (f) were 8.7±3.5, (1.59±0.26) ×10(-3) mm(2)/s, (1.04±0.12)×10(-3) mm(2)/s, (3.9±3.0)×10(-2) mm(2)/s, 0.43±0.16 and the above parameters of the wild group were 14.3±3.3, (1.34±0.26)×10(-3) mm(2)/s, (0.89±0.12)×10(-3) mm(2)/s, (3.5±2.5)×10(-2) mm(2)/s, 0.40±0.11, respectively. The difference of quantitative parameters SUV (t=4.196, P=0.0001), ADC (t=2.502, P=0.0018), D (t=3.158, P=0.006) between the EGFR mutant group and the wild group was statistically significant. The difference of quantitative parameters D(*) (t=0.361, P=0.721), f (t=0.627, P=0.536) was not statistically significant. ROC curve was drawn and the area under the curves for diagnosis of EGFR mutations by SUV, ADC, D, D(*), f was 0.880, 0.755, 0.820, 0.575, 0.550. The diagnostic accuracy of these parameters above mentioned was 80.0%, 76.7%, 73.3%, 66.7%, 53.3% respectively. The diagnostic accuracy of SUV combined with ADC, D values was 83.3% and 80.0%. Conclusion: PET/CT and IVIM have certain diagnostic value for EGFR typing of lung cancer gene.