Risk Modelling in Cancer Prediction: Are we falling behind in Head and Neck Cancer?

Objectives: Risk prediction models incorporating multiple risk factors have been recognised as a method of identifying individuals at high risk of developing cancer.
Our objectives are to:
●Discuss the concept of risk models and their use in healthcare
● Provide examples of risk models currently in use in medical practice, and specifically cancer.
● Propose the need for these models within the field of Head and Neck Cancer (HNC)
Methods: Two of the authors (CEM and MWM) will independently conduct a systematic search of PubMed and Embase in accordance with the MOOSE consensus statement to identify risk models developed within the field of cancer in the last twenty years. Retrieved studies from both Pubmed and Embase will be imported into Endnote where duplicate studies will be excluded. Titles and abstracts of the residual studies will be interrogated by the two aforementioned authors. The extracted studies will be compared, and inconsistencies will be resolved by consensus. The full texts of the remaining studies will then be read to determine whether they met our inclusion criteria. In addition, the reference lists from all identified studies will be examined.
Results:
Risk prediction models are statistical algorithms used to predict population or individual risk of a particular disease, within a specified time frame. Models have been developed for lung, breast, ovarian and colorectal cancer, as well as cardiovascular disease. To date, there are no risk prediction models in use within the field of HNC. If developed, these models would hold promise for guiding selection of individuals at the population level, for screening. The challenge lies in identifying this high risk cohort of patients, who are most at risk of HNC and for whom screening programmes would be most cost-effective.
Conclusions: Risk prediction models have been developed in many areas of healthcare but they are lacking in oral and dental research. Head and Neck Cancer prediction models have the potential to inform screening programmes and facilitate early detection. Because of the public health significance, the National Cancer Institute have recognised risk modelling as an area of extraordinary opportunity. A cost-effective robust risk prediction algorithm is required to identify the cohort of patients at highest risk of developing HNC.