Application of quantitative imaging parameters combined with tertiary lymphoid structures in recurrence risk assessment of soft tissue sarcoma: a retrospective study.

Background: Soft tissue sarcoma (STS) is a heterogeneous group of malignant tumors with a high risk of postoperative recurrence, and accurate risk stratification remains challenging. Emerging evidence suggests that quantitative imaging parameters and tumor immune microenvironment features, such as tertiary lymphoid structures (TLS), may provide complementary prognostic information. This study aimed to investigate the value of quantitative imaging parameters combined with TLS in assessing recurrence risk among patients with STS. Methods: In this retrospective study, 183 patients with STS who underwent surgical treatment at Sichuan Cancer Hospital between April 2019 and May 2023 and had complete follow-up data were included. Based on postoperative outcomes, patients were categorized into a recurrence group (n=56) and a non-recurrence group (n=127). Clinicopathological features, conventional quantitative imaging parameters derived from computed tomography (CT) and magnetic resonance imaging (MRI), and TLS status were collected and compared between groups. Multivariate logistic regression analysis was performed to identify independent predictors of recurrence. Receiver operating characteristic (ROC) curves were constructed to evaluate the predictive performance of individual indicators and the combined model. Results: Patients in the recurrence group were more frequently to present with tumor diameter >5 cm, stage III-IV disease, deep invasion, high histological grade, and distant metastasis (all P<0.05). Significant differences were also observed in surgical margin status and TLS expression between the two groups (P<0.05). Imaging features associated with recurrence included hypodense lesions, liquefactive necrosis, peritumoral invasion, and larger lesion diameters (P<0.05). MRI quantitative parameters showed that apparent diffusion coefficient (ADC), rate transfer constant (Kep), and extracellular extravascular volume fraction (Ve) were significantly lower in the recurrence group (P<0.05). TLS positivity was more frequently observed in tumors with smaller size, superficial location, and lower histological grade (P<0.05). Multivariate logistic regression analysis identified histological grade [odds ratio (OR) =3.377, 95% confidence interval (CI): 1.297-8.794], TLS expression (OR =0.387, 95% CI: 0.152-0.984), and CT-derived lesion diameter (OR =5.028, 95% CI: 3.023-8.362) as independent predictors of recurrence. ROC analysis showed that lesion diameter demonstrated relatively good predictive performance (AUC =0.842), while the combined model achieved superior performance (AUC =0.877), with a sensitivity of 92.13% and specificity of 73.21%. Conclusions: The integration of quantitative imaging parameters and TLS status significantly improves the prediction of recurrence risk in STS. This combined model may enhance risk stratification and support individualized management strategies.