J Clin Invest. 2021 Apr 15;131(8):e145488. doi: 10.1172/JCI145488.
BACKGROUND: p16 positive oropharyngeal squamous cell carcinoma (OPSCC) patients are potentially cured with definitive treatment. However, there are currently no reliable biomarkers of treatment failure in p16 positive OPSCC. Pathologist-based visual assessment of tumor cell multinucleation has been shown to be independently prognostic of disease-free survival in p16 positive OPSCC. However, its quantification is time-intensive, subjective, and at risk of interobserver variability.
METHODS: We present a deep learning-based metric, the multi-nucleation index (MuNI), for prognostication in p16 positive OPSCC. This approach quantifies tumor multi-nucleation from digitally scanned hematoxylin eosin (H&E)-stained slides. Representative H&E whole slide images from 1,094 previously untreated p16 positive OPSCC patients were acquired from six institutions for optimizing and validating MuNI.
RESULTS: MuNI was prognostic for disease-free (DFS), overall (OS), or distant metastasis-free (DMFS) survival in p16 positive OPSCC with HRs of 1.78(95%CI:1.37-2.30), 1.94(1.44-2.60), and 1.88(1.43-2.47), respectively, independent of age, smoking status, treatment type, and T/N-categories in multivariable analyses. It was also prognostic for DFS, OS, and DMFS in OPSCC patients at stages I and III.
CONCLUSIONS: MuNI holds promise as a low-cost, tissue non-destructive, H&E stain based digital biomarker test for counseling, treatment, and surveillance of p16 positive OPSCC patients. These data support further confirmation of MuNI in prospective trials.
PubMed ID: 33651718
Article Size: 7.84 MB