Objective  To investigate the importance of ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) PET/CT metabolic parameters combined with serum carcinoembryonic antigen (CEA), carbohydrate antigen (CA) 19-9, and CA72-4 levels in the prediction of microsatellite instability (MSI) status in patients with colorectal cancer (CRC).

Methods  A retrospective analysis of 96 patients (62 males and 34 females, aged (60.1±13.8) years) who underwent ¹⁸F-FDG PET/CT examination and were confirmed as colorectal cancer (CRC) via postoperative histopathology at the First Affiliated Hospital of Zhengzhou University from June 2018 to December 2022 were conducted. The patients were divided into the MSI and microsatellite stability (MSS) groups based on the results of immunohistochemistry. The maximum standardized uptake value (SUV_max) and peak standardized uptake value of region of interest (ROI) were measured, and the SUV of ROI set at 2.5 and 40% of the SUV_max as thresholds, respectively, to obtain mean standardized uptake value (SUV_mean) and SUV_mean40%, tumor metabolic volume (MTV) and MTV_40%, and total lesion glycolysis (TLG) and TLG_40%. Two independent sample student's t and Mann-Whitney U tests were conducted for the comparison of differences in the measurement data between groups. Statistical data were compared using χ² test. The independent predictor of MSI status was analyzed via multivariate Logistic regression. The metabolic parameters and serum tumor markers in the prediction of MSI status were analyzed using the receiver operating characteristic curve.

Results   Of 96 patients, 78 (81.25%) and 18 (18.75%) were in the MSI and MSS groups, respectively. Significant differences were observed in tumor primary location, differentiation degree and serum levels of CEA, CA19-9, and CA72-4 between the MSS and MSI groups (χ²=4.287, 5.103, Z=−3.467 to –2.412; all P<0.05). In addition, the MSI group exhibited significantly higher MTV_40% and TLG_40% values than the MSS group (27.03(12.07, 53.15) vs.13.39 (8.35, 21.46), 294.19(140.79, 679.66) vs. 141.36(81.11, 276.12)), and the differences were statistically significant (Z=−2.351, −2.104; both P<0.05). Multivariate Logistic regression analysis revealed that CA72-4 level (OR=1.147, 95%CI: 1.048–1.255, P=0.003), TLG_40% (OR=1.092, 95%CI: 1.004–1.187, P=0.041) and MTV_40% (OR=0.568, 95%CI: 0.345–0.936, P=0.026) was an independent predictor of MSI status in CRC patients. When CA72-4 level combined with MTV_40% predicted MSI status, the area under the curve (AUC) was 0.79(95%CI: 0.643–0.926, P<0.001), sensitivity was 72%, specificity was 87%. When CA72-4 level combined with TLG_40% were used to predict MSI status, the AUC was 0.81(95%CI: 0.676–0.933, P<0.001), and the sensitivity and specificity were 67% and 92%, respectively. When CA72-4 level combined with MTV_40% and TLG_40% were used to predict MSI status, the AUC was 0.98(95%CI: 0.931–0.999, P<0.001), and the sensitivity and specificity respectively reached 94% and 98%.

Conclusions   MTV_40%, TLG_40%, and serum CA72-4 level can be applied in the prediction of the MSI status in CRC. The use of ¹⁸F-FDG PET/CT metabolic parameters combined with serum tumor markers can achieve the noninvasive assessment of MSI status in patients with CRC for improved guidance on the immunotherapy .