Extracapsular extension is a poor predictor of disease recurrence in surgically treated oropharyngeal squamous cell carcinoma

6559 Background: While overall, patients with p16+ oropharyngeal squamous cell carcinoma (OPSCC) have a favorable prognosis, subsets of patients experience disease recurrence (DR) and death despite aggressive multimodality treatment. Aside from routine staging criteria, there are no biomarkers of tumor behavior routinely employed in OPSCC to identify patients at higher risk of DR. In this study we sought to evaluate whether the interplay between tumor-infiltrating lymphocytes (TILs) & cancer cells, in both stromal and epithelial compartments from digitized H&E-stained slides, can predict DR in OPSCC patients. Methods: OPSCC resected specimens from 354 patients (66 with DR) were retrospectively collected from 3 different sites. 107 (16 DR) patients from site 1 formed the training set and 247 (50 DR) patients from sites 2 & 3 formed the independent validation cohort. Computerized algorithms automatically identified 4 types of nuclei (TILs & non-TILs in both stromal & epithelial regions), defined clusters for each nuclei type based on cell proximity, and used network graph concepts to capture measurements relating to the arrangement of these clusters. The top 10 features determined by a statistical selection method (LASSO) were used to train a Cox regression model that assigns a risk of DR to each patient on the training set. The median risk score was used as threshold for stratifying patients on the validation set into low and high-risk of DR. Survival analysis was used to evaluate the stratification given by the trained model. Results: Patients identified by the TIL interplay model as high risk for DR had statistically worse disease specific survival. Univariate analysis yielded an HR=2.49 (95% CI: 1.22-5.07, p=0.04) for site 2 and HR=3.62 (95% CI: 1.39-9.43, p=0.03) for site 3. Multivariate analysis controlling the effect of different clinical variables is shown in the attached table. Conclusions: We introduce a prognostic model based on the automated quantification of the interplay between tumor microenvironment cells that is able to help distinguish OPSCC patients with higher DR risk from those who will experience longer disease-free survival. [Table: see text]

Download Full-text

An Analysis of the Potential Benefits of Metformin on Disease Recurrence in Oral and Oropharyngeal Squamous Cell Carcinoma

Journal of Cancer Therapy ◽

10.4236/jct.2013.45109 ◽

2013 ◽

Vol 04 (05) ◽

pp. 961-965 ◽

Cited By ~ 1

Author(s):

Christopher F. Thompson ◽

Marilene B. Wang ◽

Yas Sanaiha ◽

Chi Lai ◽

Tristan Grogan ◽

...

Keyword(s):

Squamous Cell Carcinoma ◽

Cell Carcinoma ◽

Squamous Cell ◽

Disease Recurrence ◽

Oropharyngeal Squamous Cell Carcinoma ◽

Potential Benefits

Download Full-text

Computer extracted features of nuclear architecture in H&E sections to predict disease recurrence in oropharyngeal squamous cell carcinoma patients.

Journal of Clinical Oncology ◽

10.1200/jco.2017.35.15_suppl.e17536 ◽

2017 ◽

Vol 35 (15_suppl) ◽

pp. e17536-e17536

Author(s):

Cheng Lu ◽

James Lewis ◽

Anant Madabhushi

Keyword(s):

Squamous Cell Carcinoma ◽

High Risk ◽

Cell Carcinoma ◽

Squamous Cell ◽

Characteristic Curve ◽

Strong Association ◽

Nuclear Architecture ◽

Disease Recurrence ◽

Oropharyngeal Squamous Cell Carcinoma ◽

Cell Cluster

e17536 Background: Most patients with oropharyngeal squamous cell carcinoma (OPSCC) are cured of their disease but 10-20% will develop recurrence. Identifying high risk patients is key to tailoring therapy and reducing treatment-related morbidity. In this study, we present initial findings using computerized analysis of local nuclear architecture from digitized H&E tissue sections of OPSCC via a new algorithm called cell run-length features (CRF) to distinguish patients with more versus less aggressive disease. Methods: H&E stained tissue microarray sections of 160 primary p16+ OPSCC were digitally scanned. Our CRF analysis involved 1) nuclei segmentation, 2) creation of local cell cluster graphs based on proximity of nuclei, and 3) computing CRF for each cell cluster graph. The CRF reflects the total number of different ways/runs that the graph vertices can be traversed. A higher CRF reflects more complex nuclear spatial architecture, while a lower CRF reflects the converse. A series of CRF features relating to statistics of the graph runs across the image are then extracted,. The top 5 predictive features were identified via Wilcoxon Rank Sum Test and then evaluated in conjunction with a binary quadratic classifier via 5-fold cross validation. Results: The top ranked features were related to CRF non-uniformity (i.e. reflecting the variation of complexity of different nuclei clusters) and the classifier yielded a mean area under the receiver operating characteristic curve of 0.75 in discriminating patients who did and did not have disease progression. Patients with high risk CRF values were 5.7 times more likely to suffer disease recurrence than those without. While the whole cohort had a disease recurrence rate of 12.5%, those with low CRF values developed recurrence in only 1.875%. Conclusions: CRF features demonstrate a strong association with disease recurrence in patients with p16 positive OPSCC. Our preliminary findings suggest that measurements related to local arrangement of nuclei have potential in the risk assessment of patients with p16 positive OPSCC. The findings will need to be independently validated in a separate, powered cohort of patients.

Download Full-text