MRI-derived radiomics model for predicting intratumoral tertiary lymphoid structures in soft tissue sarcoma

To develop and validate an MRI-derived radiomics model for the prediction of intratumoral tertiary lymphoid structures (TLSs) status of soft tissue sarcoma (STS) and explore its prognostic value.

This study retrospectively included 302 patients of three cohorts who underwent surgical resection of STS from two medical centers. Radiomics features were derived for both intratumoral and peritumoral regions from preoperative axial fat-suppressed T2-weighted and T1-weighted imaging sequences. Intratumoral, peritumoral, and combined radiomics models were constructed using a logistic regression algorithm. The area under the receiver operator characteristic curve (AUC) and the DeLong test were utilized to assess and compare the performances of three radiomics models. By applying a linear combination of the chosen features, the Rad-score for the optimal radiomics model was computed.

TLS positivity was identified in 114 (38%) of the 302 patients. No clinical, radiological, or pathological variable was found to show a statistically significant association with TLSs status. The combined radiomics model showed superior performance compared to both the intratumoral and peritumoral models, with an AUC of 0.878 (95% CI 0.812-0.927) in the development cohort, 0.778 (95% CI 0.649-0.876) in the internal validation cohort, and 0.772 (95% CI 0.679-0.850) in the external validation cohort. In the cohort for all patients, the 36-month cumulative PFS rate was 66.1% in the high Rad-score (≥ 0.5) group vs. 37.2% in the low Rad-score group (p < 0.05, log-rank test).

An MRI-derived radiomics model could predict intratumoral TLS status in patients with STS and demonstrated a correlation with PFS.

The MRI-derived radiomics model could predict intratumoral TLSs status in patients with STS accurately, which may help to screen patients who will benefit from immunotherapy and have a better prognosis.

Intratumoral tertiary lymphoid structure status in patients with soft tissue sarcomas was accurately predicted by an MRI-derived radiomics model. The combined radiomics model showed superior performance compared to both the intratumoral and peritumoral radiomics models. Progression-free survival was significantly longer in patients with a high Rad-score (≥ 0.5) in the development, internal validation, and external validation cohorts.