Abstract B102: Salivary markers, demographic, and risk factor data in head and neck cancer: Towards a new combined approach to predict head and neck squamous cell carcinoma

Abstract

Abstract Head and neck squamous cell carcinoma (HNSCC) affects 50,000 people in the United States and 600,000 people world-wide each year. The primary risk factors are tobacco and alcohol use. Early detection tests are needed because the majority of patients present in late stage when cure rates reach only 40%. Furthermore, minority patients and those of low socioeconomic status (SES) suffer disproportionately from this disease. Our group has developed a simple, inexpensive, noninvasive diagnostic test based on soluble CD44 (solCD44) and total protein levels that distinguishes cancer patients from controls with up to 85% accuracy. The test appears to be particularly promising for detection of HNSCC in blacks. Our study uses a case-control design to evaluate soluble markers for HNSCC in 123 HNSCC patients and 78 controls frequency matched for age, gender, race, ethnicity, tobacco use and socioeconomic status (SES). There were no significant differences between cases and controls for Hispanic ethnicity, education, income, health care status, tobacco use or oral health. There are significant differences with respect to age (cases are older, mean 58.9 versus 55.6 years, p=0.026), gender (cases are more likely male, 78% vs. 64%, p=0.031), race (cases are more likely white, 83% vs. 60%, p=0.001). SolCD44 shows that levels are elevated in cases compared to controls (p<0.001). There are significant differences in solCD44 levels based on race in the lip/oral cancer group (blacks have higher levels, 3.21 vs. 2.10, p=0.048). There are also significant differences based on income in the control group (lower income associated with higher levels, 1.55 vs. 0.98, p=0.028). Furthermore, there is a borderline interaction with race (p=0.057) and income (p=0.074). Protein levels are also significantly higher in cases compared to controls (p=0.02). There are significant differences in protein level with respect to age in the control group (older subjects have higher levels 0.90 vs. 0.65 p=0.05), gender for oropharyngeal tumors (males have higher levels, 1.05 vs. 0.72, p=0.032), education for lip/oral cavity tumors (lower education is associated with higher levels, 1.18 vs. 0.87, p=0.046) and smoking in lip/oral cavity (current smokers have lower levels than never smokers, 0.81 vs. 1.33, p=0.006). There is a significant interaction between protein levels and race (p=0.041). For solCD44 in oral cavity tumors, there are significant differences with respect to stage, T stage and N stage with more aggressive or larger tumors having higher levels (p<0.007). There is also an interaction between each of the staging variables and CD44 level. The multivariate logistic regression models showed that in black males, there is a significant effect of either marker; when both markers are used, none of the markers is significant but there is an improvement in AUC. In black females, for which sample size is small, there is no significant effect of markers either individually or together, however the corresponding odds ratio appears similar to those for black males. Among white males, the effect of log2 CD44 was significant by itself or when protein was added. Finally, among white females, there was no effect of either marker separated or together, with AUCs near 0.50. In conclusion, we see that solCD44 and protein expressed in oral rinses from cases and controls vary across racial and gender groups. Our findings strongly suggest that race and gender are very important components for HNSCC early detection studies. Citation Information: Cancer Epidemiol Biomarkers Prev 2011;20(10 Suppl):B102.