Abstract
Esophageal Squamous Cell Carcinoma (ESCC) remains a significant health burden in China. This study investigates the combined impact of dietary patterns and symptom clusters on the risk of ESCC and its precursor lesions. Utilizing Latent Class Analysis (LCA) on a multicenter cross-sectional dataset of 34,707 individuals, we identified five dietary patterns and three symptom severity classes. Results indicate that non-healthy dietary patterns and increased symptom severity are strongly associated with higher risks of esophageal disease. A developed nomogram demonstrated robust predictive ability (C-index = 0.746), offering a potential tool for early screening in high-incidence regions.
Introduction
Esophageal cancer is the sixth most common malignancy in China, with Esophageal Squamous Cell Carcinoma (ESCC) accounting for over 90% of cases. While traditional risk factors like smoking, alcohol consumption, and low fruit intake are well-documented, dietary habits and symptoms often occur in complex clusters rather than in isolation. Understanding these patterns is crucial for identifying high-risk individuals during the early, treatable stages of the disease, such as esophagitis and intraepithelial neoplasia. This study aims to categorize these patterns and develop a predictive model for esophageal disease progression.
Methodology
This multicenter cross-sectional study analyzed data from 34,707 participants aged 40–69 years across five high-incidence provinces in China (Hebei, Henan, Shandong, Shanxi, and Gansu).
Latent Class Analysis (LCA)
LCA was employed to identify hidden subgroups based on:
- Dietary Patterns: Frequency of intake across 10 food groups (e.g., vegetables, red meat, pickles).
- Symptom Severity: Presence of six symptoms, including oral health indicators (tooth loss, gingival bleeding) and gastrointestinal issues (dysphagia, reflux).
Statistical Modeling
Multivariable logistic regression was used to calculate Odds Ratios (ORs) and 95% Confidence Intervals (CIs) for various disease stages: esophagitis, low-grade intraepithelial neoplasia (LGIN), and high-grade lesions (HGIN)/ESCC. All diagnoses were confirmed via endoscopy and biopsy. Finally, a nomogram was constructed to visualize the risk prediction model.
Results
Dietary and Symptom Classification
Five distinct dietary patterns were identified:
- Healthy (28.8%): High intake of vegetables, fruits, and quality protein.
- Western (9.3%): High red meat, low vegetable intake.
- Lower consumers-combination (29.1%): Low overall food diversity and quantity.
- Medium consumers-combination (28.7%)
- Higher consumers-combination (4.1%)
Symptom severity was categorized into three classes: Asymptomatic, Mild symptoms, and Overt symptoms.
Risk Associations
Compared to the "Healthy" pattern, all other dietary clusters showed an increased risk of esophageal disease. Notably, the "Lower consumers-combination" pattern was associated with an OR of 2.33 for esophagitis. The "Western" pattern showed a significant association with HGIN/ESCC (OR = 1.69).
Symptom progression also correlated with disease severity. "Mild symptoms," particularly oral health issues like tooth loss, were linked to early-stage lesions (esophagitis and LGIN). "Overt symptoms" such as dysphagia were characteristic of late-stage HGIN and ESCC.
Model Performance
The risk prediction nomogram achieved a C-index of 0.746 for predicting HGIN and ESCC, indicating strong discriminatory power for clinical applications.
Discussion
The findings highlight the importance of dietary diversity and oral hygiene in preventing esophageal malignancy. The link between "Mild symptoms" (oral health) and early lesions suggests that poor oral hygiene may facilitate the conversion of nitrates into carcinogenic nitrosamines via bacterial overgrowth. Furthermore, the risk associated with the "Lower consumers-combination" pattern suggests that malnutrition and lack of dietary variety are as detrimental as high-fat "Western" diets in this population.
Conclusion
This study provides robust epidemiological evidence for the role of dietary and symptom patterns in ESCC risk. The identified clusters and the predictive nomogram offer valuable insights for targeted screening and public health interventions in high-risk regions of China. Future longitudinal studies are needed to validate these findings and establish causal relationships.
Summary updated