MRI for evaluation of preoperative chemotherapy in osteosarcoma

Abstract

Introduction: Our study aims to highlight the role of Magnetic Resonance Imaging (MRI) in monitoring the therapeutic response after neoadjuvant chemotherapy in osteosarcoma of the long bones.

Methods: In this retrospective study, data from the Orthopaedics and Internal Medicine Department of Istanbul University Cerrahpasa Hospital was used. We selected the study cohort from our departmental database of patients with biopsy-proven osteosarcoma initially treated with preoperative chemotherapy at Istanbul University Cerrahpasa Hospital from 2010 to 2017. MRI images of 21 patients (male/female ratio: 2.5 with a mean age of 22) were analysed before and after neoadjuvant chemotherapy. The histological response to chemotherapy was graded according to The Huvos classification. Computed volumetry was performed to determine the size of the intramedullary component, largest enhancing component, and tumour volume. P < 0.05 was considered to denote a significant difference.

Results: The mean tumour volume before chemotherapy was 409 cm³. After chemotherapy, however, the tumor volume increased to 701 cm³ (p = 0.10). The mean intramedullary component size of the tumours before chemotherapy was 10.5 cm³ while after chemotherapy was 11.2 cm³ (p = 0.06). The mean largest enhancing component size was 3.09 cm³ and after chemotherapy, decreased to 2.34 cm³ (p = 0.01). Neoadjuvant chemotherapy significantly changed the tumour composition. Tumour volume and intramedullary component size measurements failed to demonstrate a significant correlation and could not be used as a prognostic factor for tumour response to preoperative chemotherapy. We suggest that the largest enhancing component of a tumour can be a potential prognostic marker for assessing the tumour response.

Conclusion: MRI can help predict histological necrosis after the administration of preoperative chemotherapy to osteosarcoma via measuring the largest enhancing component. Hence, it is a promising preoperative indicator of response to neoadjuvant chemotherapy. However, tumour volume and intramedullary component size measurement are not effective predictors of histological necrosis. The increased volume and intramedullary component of the tumour were attributed to the increased central necrotic component of the tumour after chemotherapy.

Implications for practice: In this study, we showed that MRI can help predict histological necrosis and thus, prognosis after the administration of preoperative chemotherapy to osteosarcoma via the measurement of the largest enhancing component of the tumour. This is significant because histological necrosis is currently the gold standard method for assessing the treatment response. However, this requires an invasive procedure, and a non-invasive method would be beneficial. Assessing the treatment response through imaging after the completion of the initial chemotherapy will also help determine the final surgical approach and thus predict survival.

Keywords: Computed tomography; Histological response; Magnetic resonance imaging; Neoadjuvant chemotherapy; Osteosarcoma; Prognosis.