# MRI-Based Delta Necrosis as a Prognostic Marker Following Neoadjuvant Chemotherapy in Soft Tissue Sarcoma

## TL;DR

This study found that changes in tumor necrosis seen on MRI scans after chemotherapy in soft tissue sarcoma patients do not reliably predict treatment success or survival outcomes.

## Contribution

The study is the first to evaluate MRI-based delta necrosis as a potential prognostic marker in soft tissue sarcoma following neoadjuvant chemotherapy.

## Key findings

- MRI-based delta necrosis showed a weak and nonsignificant correlation with pathologic necrosis.
- Neither MRI-based nor pathologic necrosis thresholds were associated with survival outcomes in soft tissue sarcoma patients.
- The study suggests that radiologic changes in necrosis may not serve as reliable markers of therapeutic response.

## Abstract

Soft tissue sarcomas (STS) are rare and diverse tumors for which neoadjuvant chemotherapy is often used to improve local control and survival. However, accurately determining treatment effectiveness remains challenging. This study evaluated whether changes in tumor necrosis measured on MRI scans before and after chemotherapy, referred to as delta necrosis, could predict outcomes or mirror the amount of necrosis seen in surgical pathology. We found that MRI-based delta necrosis did not correlate with pathologic necrosis or patient survival, suggesting it may not serve as a reliable prognostic marker in STS. Standardized imaging and pathology assessment protocols, together with future exploration of molecular tools such as ctDNA, may help refine response evaluation and improve prognostic accuracy in this heterogeneous disease.

Background: The prognostic value of treatment-induced necrosis in soft STS remains uncertain. This study evaluated whether MRI-based changes in necrosis (Δ necrosis) between pre- and post-neoadjuvant chemotherapy scans correlate with pathologic necrosis and clinical outcomes. Methods: In this retrospective cohort, 27 patients with STS who received neoadjuvant chemotherapy and underwent pre- and post-treatment MRI were analyzed. Necrosis was graded categorically (<5%, 5–25%, 25–50%, 50–75%, 75–95%, and >95%), and Δ necrosis was calculated as the change in estimated necrosis between scans. Correlations between MRI-derived and pathologic necrosis were assessed using Spearman’s rank coefficient. Survival analyses (progression-free, local recurrence-free, and disease-specific overall survival) were performed using Kaplan–Meier and log-rank tests. Results: Post-treatment MRI necrosis moderately correlated with pathologic necrosis (ρ = 0.44, p = 0.028), whereas Δ necrosis showed a weaker, nonsignificant correlation (ρ = 0.24, p = 0.24). Neither MRI-based nor pathologic necrosis thresholds were associated with survival outcomes. Conclusions: MRI-based Δ necrosis did not predict pathologic necrosis or oncologic outcomes in STS, suggesting that radiologic changes in necrosis may not serve as reliable markers of therapeutic response. Future studies integrating quantitative imaging and standardized pathology protocols together with future exploration of molecular tools such as ctDNA are needed to refine treatment assessment in STS.