Noninvasive MRI imaging feature-based prediction of intratumoral tertiary lymphoid structure maturity in hepatocellular carcinoma: a multicenter retrospective study
摘要
To develop a noninvasive MRI-based model for assessing intratumoral tertiary lymphoid structures (iTLSs) maturity in hepatocellular carcinoma (HCC) to guide therapeutic strategies.
Materials and methodsRetrospective analysis of 388 HCC patients from two centers included 338 in the training cohort and 50 in the external validation cohort. Patients were pathologically stratified into iTLSs-0 (n = 113), iTLSs-1 (n = 173), and iTLSs-2/3 (n = 32). External validation cohort included 20, 15, and 15 cases in each corresponding group. MRI features were analyzed using logistic regression to identify predictors of iTLSs maturity. The outcome cohort (n = 64) was used to assess the model’s ability to identify candidates for tyrosine kinase inhibitors combined with immune checkpoint inhibitors (TKI-ICI) treatment. Nomogram performance was evaluated via area under the curve (AUC) with 95% confidence intervals (CIs).
ResultsiTLSs-1 exhibited lower intratumoral hemorrhage (p < 0.0001) and higher tumor in vein (p = 0.001) versus iTLSs-0 (AUC = 0.713, 95% CI: 0.661–0.764). iTLSs-2/3 showed higher complete capsule (p < 0.001), lower satellite nodules (p = 0.049), and lower intratumoral artery (p < 0.001) compared to iTLSs-1 (AUC = 0.750, 95% CI: 0.701–0.799). External validation confirmed generalizability (AUC = 0.750 and 0.764). The integrated nomogram demonstrated good calibration and potential clinical utility. A nomogram-based threshold (probability > 0.5) was used to stratify iTLSs, and the stratification effectively predicted iTLSs-2/3 patients who benefited from postoperative TKI-ICI treatment compared to iTLSs-0 and iTLSs-1 (p = 0.013 and 0.010).
ConclusionMRI features effectively stratify iTLSs maturity in HCC, offering a noninvasive tool to personalize treatment.
Key Points