Overview

Head and neck squamous cell carcinoma is the sixth most common malignant tumor in the world. Neoadjuvant therapy, including neoadjuvant chemotherapy and immunotherapy, is recommended for patients with locally advanced head and neck cancer. The response to neoadjuvant therapy varies among the patients. It is reported that about 37% of the patients achieve pathological complete response after receiving neoadjuvant therapy, who would achieve a better prognosis compared with the patients with non-pathological complete response. It is significant to predict and assess response to neoadjuvant therapy for the patients with head and neck cancer accurately, which could assist in formulating individualized therapeutic regimens. MRI has good soft tissue resolution and is a common preoperative examination method. However, this method lacks the ability to accurately predict the probability of patients achieving pathological remission after neoadjuvant therapy. At present, it is a novel and effective method to construct a model to predict the efficacy of neoadjuvant therapy based on MRI image omics analysis, and certain achievements have been made in breast cancer and rectal cancer. In this study, multi-sequence MRI was combined with clinical risk factors to construct an imaging omics model to predict the probability of pathological complete remission of patients with head and neck squamous cell carcinoma after neoadjuvant therapy, and to accurately identify diagnostic imaging remission, so as to better assist clinical decision-making.

Eligibility

Inclusion Criteria:

• (a) patients pathologically diagnosed as head and neck squamous cell carcinoma;
• (b) patients receiving two or three courses of neoadjuvant therapy (traditional chemotherapy plus PD-1 inhibitor);
• (c) MR scan within 30 days before the first course of neoadjuvant therapy;
• (d) patients undergoing radical surgical dissection following neoadjuvant therapy;
• (e) complete clinical data available.

Exclusion Criteria:

• (a) previous head and neck treatment history;
• (b) obvious motion or metal artifacts on the MRI image;
• (c) distant metastasis;
• (d) concurrent malignancies.