# Distinct Second Primary Tumor Phenotypes in Oral Squamous Cell Carcinoma According to Exposure Status and Immune Background

## TL;DR

This study finds that second primary tumors in oral cancer patients differ based on lifestyle factors and immune conditions, suggesting tailored surveillance strategies.

## Contribution

The study identifies distinct second primary tumor phenotypes in oral squamous cell carcinoma linked to exposure status and immune background.

## Key findings

- Smoking/drinking patients more frequently develop extra-oral second primary tumors.
- Non-smoking/non-drinking patients with immune conditions like oral lichen planus are more likely to have multifocal oral SCC.
- Oral lichen planus strongly predicts multifocal oral SCC in non-smoking/non-drinking patients.

## Abstract

Background: Second primary tumors (SPTs) are a major survivorship challenge in oral squamous cell carcinoma (OSCC), yet their biological phenotypes may differ according to exposure status and immune background. Methods: In this retrospective cohort (2011–2020), 242 surgically treated primary OSCC patients were classified as non-smoking, non-drinking (NSND; never smokers/never drinkers) or smoking and/or drinking (SD; any history of smoking and/or alcohol consumption). SPTs were categorized as extra-oral SPTs (eoSPTs) or multifocal oral SCC (mOSCC), with mOSCC (≥3) denoting ≥3 oral primaries. Immune background was assessed by documenting immune-modulating conditions (including oral lichen planus as an immune-mediated mucosal disorder). Multivariable logistic regression was used to evaluate predictors of eoSPTs and mOSCC. Results: SPT occurred in 82/242 (33.9%), comprising 54 eoSPT (22.3%) and 28 mOSCC (11.6%). Overall SPT prevalence was similar in NSND and SD patients (29.8% vs. 36.1%), but phenotype composition differed significantly (chi-square p = 0.004): eoSPTs were more common in SD (27.8% vs. 11.9%), whereas mOSCC was more common in NSND (17.9% vs. 8.2%); mOSCC (≥3) occurred in 10.7% of NSND versus 1.3% of SD patients. Immune-modulating conditions were associated with mOSCC but not eoSPTs. Within the immune-modulating spectrum, OLP showed strong phenotype specificity (0/20 eoSPTs; mOSCC in 7/20 [35.0%), particularly among NSND patients (38.9% with OLP vs. 12.1% without). In adjusted models, NSND status was associated with lower odds of eoSPT (OR 0.37, 95% CI 0.15–0.96), while OLP independently predicted mOSCC (OR 3.47, 95% CI 1.04–11.52). Conclusions: SPTs in OSCC comprise distinct phenotypes: SD patients predominantly develop eoSPTs consistent with carcinogen-associated aerodigestive field effects, whereas NSND patients exhibit an immune-associated, oral-restricted pattern with frequent mOSCC, supporting phenotype-tailored surveillance.