Search

Purpose: This study was designed to investigate the prognostic significance of artificial intelligence (AI)-based quantification of myxoid stroma in patients undergoing esophageal squamous cell carcinoma (ESCC) surgery after neoadjuvant chemotherapy (NAC) and to verify its significance in an independent validation cohort from another hospital., Methods: We evaluated two datasets of patients with pathological stage II or III ESCC who underwent surgery after NAC. Cohort 1 consisted of 85 patients who underwent R0 surgery for the primary tumor after NAC. Cohort 2, the validation cohort, consisted of 80 patients who received same treatments in another hospital. AI-based myxoid stroma was evaluated in resected specimens, and its area was categorized by using the receiver operating characteristic curve for overall survival (OS) of cohort 1., Results: The F1 scores, which are the degree of agreement between the automatically detected myxoid stroma and manual annotations, were 0.83 and 0.79 for cohorts 1 and 2. The myxoid stroma-high group had a significantly poorer prognosis than the myxoid stroma-low group in terms of OS, disease-specific survival (DSS), and recurrence-free survival (RFS) in cohort 1. Comparable results were observed in cohort 2, where OS, DSS, and RFS were significantly affected by myxoid stroma. Multivariate analysis for RFS revealed that AI-determined myxoid stroma-high was one of the independent prognostic factors in cohort 1 (hazard ratio [HR] 1.97, p = 0.037) and cohort 2 (HR 4.45, p < 0.001)., Conclusions: AI-determined myxoid stroma may be a novel and useful prognostic factor for patients with pathological stage II or III ESCC after NAC., (© 2024. Society of Surgical Oncology.)

Aims: Desmoplastic reaction (DR) categorisation has been shown to be a promising prognostic factor in oesophageal squamous cell carcinoma (ESCC). The usual DR evaluation is performed using semiquantitative scores, which can be subjective. This study aimed to investigate whether a deep-learning classifier could be used for DR classification. We further assessed the prognostic significance of the deep-learning classifier and compared it to that of manual DR reporting and other pathological factors currently used in the clinic., Methods and Results: From 222 surgically resected ESCC cases, 31 randomly selected haematoxylin-eosin-digitised whole slides of patients with immature DR were used to train and develop a deep-learning classifier. The classifier was trained for 89 370 iterations. The accuracy of the deep-learning classifier was assessed to 30 unseen cases, and the results revealed a Dice coefficient score of 0.81. For survival analysis, the classifier was then applied to the entire cohort of patients, which was split into a training (n = 156) and a test (n = 66) cohort. The automated DR classification had a higher prognostic significance for disease-specific survival than the manually classified DR in both the training and test cohorts. In addition, the automated DR classification outperformed the prognostic accuracy of the gold-standard factors of tumour depth and lymph node metastasis., Conclusions: This study demonstrated that DR can be objectively and quantitatively assessed in ESCC using a deep-learning classifier and that automatically classed DR has a higher prognostic significance than manual DR and other features currently used in the clinic., (© 2022 John Wiley & Sons Ltd.)