Epidemiological studies of risk factors could aid in designing risk stratification tools

The latest issue of Cancer Research, Statistics and Treatment contained an interesting article by Michaelraj et al., an epidemiological study on risk factors of head-and-neck squamous cell carcinoma (HNSCC) in south India.[1] I would like to discuss some of the crucial findings of this study and their implications in developing risk stratification models and beyond. This study cross-sectionally analyzed the epidemiological profile of 150 consecutive patients with primary treatment-naïve HNSCC recruited over three years at a tertiary care hospital in Tamil Nadu.[1] As seen in most of the other regions of India, the most common primary site of HNSCC was the oral cavity (40.7%) in this cohort.[2] There were three times more men than women among the diagnosed cases, and only just about a quarter of the study cohort had no exposure to smoking, tobacco chewing, or alcohol (27.3%). Although there existed a significant variability among the men and women regarding the distribution of these risk factors, as per Table 4 in the paper,[1] more than half the men with HNSCC had exposure to multiple risk factors. Furthermore, 66.7% of the overall cohort had exposure to at least one tobacco product. These observations of Michaelraj et al.[1] align with the existing consensus, as they suggest a possible etiopathological role of these known carcinogenic elements, particularly tobacco, the exposure to which is significantly higher among men than women.[3] In Table 5,[1] the authors analyzed the proportional distribution of risk factors in different age groups and found it statistically significant by two-way ANOVA. This table also showed that 91.8% (100/109) of patients with HNSCC exposed to a known risk factor(s) were aged between 41 and 70 years, and only a few patients in the exposed group were outside this range. Additionally, the distribution of all these risk factors (including various combinations of these factors) was seen to peak around the sixth decade of life. Lastly, the patients in the sixth decade of life or older had a higher degree of exposure to multiple risk factors than those in the fifth decade or younger, who had either one risk factor or no exposure at all. All these findings suggest that the putative role of tobacco and alcohol in the carcinogenesis of HNSCC seems to be more relevant in older adults, in their fourth, fifth, and sixth decades of life than in the younger population, a notion that has also been supported by other recent studies.[4] While the relative risk attributable to these known carcinogens is not always predictable, it is understandable that the risk increases with an increase in the duration and severity of exposure to these factors.[5] By these observations, the elderly male with a long-standing use of tobacco, with or without alcohol, would seem to have a higher risk of developing HNSCC, which includes oral cancer. Accordingly, these groups of people, if targeted, would be more likely to benefit from screening programs and preventive interventions, as applicable.[6] By providing the epidemiological profile of risk factors involved, studies like the one by Michaelraj et al. form the basis for developing pre-screening risk-stratification tool(s) aimed at defining an appropriate high-risk population who could benefit from screening programs.[6] A prototype of a risk stratification tool for oral cancer called OraCLE has been proposed recently, which is based on exposure levels to the risk factors, and is awaiting validation studies.[7] Lastly, the status and distribution of human papillomavirus in this study are also in line with the present consensus, as they are known to affect only a small proportion of oropharyngeal tumors in the Indian context.[8,9] Overall, although the present study by Michaelraj et al.[1] does not establish a direct causal association between the studied risk factors and HNSCC, by defining the epidemiological profile of HNSCC in a cohort of patients from south India, it could form the basis for further studies in this regard, and could eventually aid in the design of appropriate screening and preventive strategies as relevant to the study population. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.