Histological response to neoadjuvant chemotherapy in localized Ewing sarcoma of the bone: A retrospective analysis of available scoring tools

Abstract

Aim

The aim is to evaluate which of the existing scoring systems of histological response to neoadjuvant chemotherapy best stratifies the clinical outcome of patients with localized Ewing sarcoma of bone.

Methods

474 patients with diagnosis of localized Ewing sarcoma of bone were included. The median follow-up was 13.5 years.

Results

The overall survival and the disease-free survival (DFS) were 70.8% and 63.9% at 5 years. The percentage of histological response to neoadjuvant chemotherapy ranged between 5% and 100% (mean 83%). The agreement between Bologna System and the different percentual cut-offs of histological response to neoadjuvant chemotherapy was high, with kappa statistics of 0.83 for a cut-off of ≥90%; 0.86 for a cut-off of ≥95%; 0.79 for a cut-off of ≥96% and 0.61 for a cut-off of 100%. Statistically higher DFS rates for good responders compared to poor responders were found when using each given system. Model performance indicators showed that Bologna system had a lower AIC score and a higher c-statistics to predict DFS. When the patients classified as good responders using the different percentual cut-offs of histological response to neoadjuvant chemotherapy, were instead re-classified using the Bologna system, statistical differences were noted in DFS within each specific group.

Conclusions

All scoring tools to evaluate histological response to neoadjuvant chemotherapy offer good predictive value for DFS in localized Ewing’s sarcoma of bone. The Bologna system better stratifies those patients with histological response to neoadjuvant chemotherapy between 90 and 99%, representing a more reliable scoring tool in this subset.

Introduction

Ewing sarcoma, a round cell sarcoma harbouring in the vast majority of cases an EWS-FLI1 fusion gene [1], is the second most common bone malignancy in children and adolescents, with an annual incidence of 200–500 cases in the United States and of 120–150 cases in Italy [2,3].

With advances in multidisciplinary care, the 5-year survival rate for Ewing sarcoma patients have improved from approximately 50% between 1983 and 1990 to 70% between 2004 and 2016 [[2], [3], [4]]. Prolonged disease-free survival (DFS) after treatment with multiagent chemotherapy and local treatment of the primary tumour, is achieved in the majority of patients with localized Ewing sarcoma [4]. Histological assessment of response to neoadjuvant chemotherapy determined on resection specimens of primary tumours represents an important prognostic factor. The histological response to chemotherapy is also used to identify patients who are candidates for more intensive or novel therapies [[4], [5], [6], [7]].

Histologic response to neo-adjuvant chemotherapy has been defined according to different scoring systems variably based on the percentage of tumour necrosis and/or repair fibrous tissue with 4 different cut-off points, considering as poor responders the cases with <90%, <95%, <96%, and <100% of histological response to CHT, respectively (Table 1). The “Bologna System” represents an alternative scoring system first reported by Picci et al. [ref], that considers as "poor responders" patients with macroscopic foci of viable tumor (grade 1) and as "good responders" the cases with microscopic evidence of viable tumor (grade 2) or with no evidence of viable tumor (grade 3) [[5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24]] (Table 1).

The aim of the present study is to evaluate which of the existing scoring systems of histological response to neoadjuvant chemotherapy offers the best performance in terms of prediction of clinical outcome in patients affected by localized Ewing sarcoma of bone.