Tumor habitat characteristics derived from intravoxel incoherent motion for early response assessment in soft tissue sarcoma undergoing neoadjuvant radiotherapy and targeted therapy: a phase II study.

BACKGROUND

Soft tissue sarcomas (STSs) are heterogeneous malignancies with limited reliable criteria for early response assessment. Commonly used criteria, such as Response Evaluation Criteria in Solid Tumors (RECIST) 1.1, often underperform because treatment-related changes may not translate into tumor shrinkage. This study aimed to assess the potential of intravoxel incoherent motion (IVIM)-derived tumor habitat characteristics as biomarkers for assessing the treatment responses of STS patients who received neoadjuvant radiotherapy (neo-RT) along with targeted therapy.

METHODS

Patients with STS of the limbs or trunk from a prospective phase II trial were selected. All patients received magnetic resonance imaging (MRI) examinations pre- and post-neo-RT and targeted therapy. K-means clustering was employed to classify lesions into subregional habitats based on IVIM parameters. Postoperative pathology classified patients into good- and poor-response groups, defined as <50% and ≥50% residual tumor cells, respectively. IVIM parameters, including mean true-diffusion coefficient (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f) value in the whole tumor and subregions, were extracted and compared between groups. Individual parameter models and combined model were constructed, and receiver operating characteristic (ROC) analysis was utilized to evaluate the diagnostic efficacy.

RESULTS

Twenty-one participants (11 males, 10 females; mean age 52.1±17.4 years) were enrolled, including 13 individuals in the good-response group and 8 in the poor-response group. K-means clustering identified three clusters based on pre- and post-treatment IVIM features. Five features that significantly differed between the good-response and poor-response groups were incorporated into individual parameter models and combined model. The area under the ROC curve of the combined model was 0.904, greatly outperforming that of RECIST 1.1 [area under curve (AUC) 0.567] and individual parameter models.

CONCLUSIONS

Tumor subregional IVIM parameters can well assess the pathological response degree in STS patients undergoing neo-RT and targeted therapy.