Modification of RECIST 1.1 criteria for assessing response in breast tumours treated with radiation therapy using multiparametric breast MRI: Radiology and oncology perspective

The traditional, size-based, RECIST 1.1 guideline is the standard for assessing tumour response, but has notable limitations, particularly for breast tumours treated with radiotherapy (RT). RT often induces fibrosis, leading to a persisting measurable abnormality on T1W/T2W MRI sequences despite an underlying pathological response. As pre-operative RT trials expand and omission of surgery is being tested, having anatomico-functional MRI response criteria may facilitate more accurate response evaluation. We propose modified RECIST 1.1 criteria based on 3 sequences of multiparametric MRI (mpMRI): unenhanced T2W, contrast-enhanced (CE) T1W and diffusion-weighted imaging (DWI). Key recommendations include complete response defined as 1) resolution of tumour mass on all sequences, or 2) evidence of residual T2 abnormality with CE images showing no enhancement above background parenchymal enhancement (BPE) and DWI signal consistent with necrosis/fibrosis, or 3) minimal evidence of enhancement above BPE with DWI in keeping with fibrosis. Criteria are also defined for other response categories (partial response/stable disease/progression). These recommendations can be tested in future trials and offer guidance on the interpretation of mpMRI sequences when breast tumours are treated in situ with radiation, a scenario in which little published data currently exists. They highlight how DWI can aid RT response assessment and help overcome some limitations with current RECIST criteria.