A Quantitative Image Analysis Tool for Prognosticating Disease-Free Survival of Patients with Oral Cavity Squamous Cell Carcinoma
Abstract
Objective and Rationale: Head and neck squamous cell carcinoma (HNSCC) is the sixth most common malignancy and one of the costliest cancers to treat in the U.S. Oral cavity squamous cell carcinoma (OCSCC) is the most common type of HNSCC. Over the past few years, its incidence is increasing worldwide, and it is associated with major morbidity and mortality. However, unlike other type of HNSCC, OCSCC patients are considered to have a poor prognosis. About 30% of patients treated for oral cancer developed recurrence within 5 years, suggesting that traditional treatment strategy was not sufficient. Even though The American Joint Committee on Cancer 8th edition states staging system is considered as a gold standard for treatment guideline in OCSCC, recent studies revealed that it was insufficient to accurately guide therapeutic planning. In current treatment guidelines, all patients are treated with surgical resection initially. Necessity of the post-operative therapy is determined by the staging criteria. The criteria use tumor stage (T), regional lymph node metastases (N), and metastases (M) based on the clinical and pathological examination, such as the depth of primary tumor invasion and extracapsular extension of lymph node metastases. Unfortunately, some of the parameters, such as tumor grade and lymphocytic response have, despite years of application, failed to serve as effective risk predictor, especially in early stages of this cancer. For instance, tumors with T1-2 and N0-1 are considered as low risk and are treated with surgery without any post-operative radiation therapy. However, recent findings suggest that there are patients in the low risk group who actually have intrinsic tumor properties potentially driving tumor recurrence after surgery. Considering treatment escalation strategies for such patients could improve the outcome by preventing recurrence, morbidity, or mortality. Thus, there is crucial unmet clinical need for accurate biomarkers to identify which patients could receive added benefit from additional therapy. On the other hand, a one-size-fits-all approach to guide the treatment management may disservice for specific patient populations. For instance, African American (AA) OCSCC patients have significantly shorter overall survival than Caucasian (CA) ones. Similarly, Veteran (VA) patients constitute a population at high risk of death and complications during application of therapy. Interestingly, younger OCSCC patients who are not exposed to smoke and alcohol have poor survival, evidencing the need of studying intrinsic tumor biology and understanding the molecular features driving cancer presentation that are unique and specific to the populations, which will pave the way for building population-specific models. Understanding the real factors affecting such populations could help in the development of more specific and accurate prognostic and predictive risk models. The advent of whole-slide digital scanners enables cost-effective digital replication of traditional glass slides. It also brought up evolution of artificial intelligence (AI) to characterize tissue morphology. Computer vision and pattern recognition tools can aid pathologists to unlock sub-visual attributes about tumor behavior. By means of such tools, in the present project, the Principal Investigator (PI) seeks to identify interpretable imaging biomarkers such as TILs, multinucleations, keratin pearls, and tumor buds. The PI also seeks to identify and understand the unique and novel molecular features driving cancer presentation to improve outcomes across different key populations, such as AA and VA patients, aiming for building population specific risk models. PI’s Career Goal: The PI aims to pursue an academic career in AI-driven analysis of head and neck cancer leading to a full-time faculty position with a focus on developing novel biomarkers for predicting prognosis and to work on translational research
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Dec 28, 2022
- Source ID
- W81XWH2210236
Entities
People
- Rakesh Shiradkar
Organizations
- Emory University
- United States Army