Comparison of five-year survival rates among patients with oral squamous cell carcinoma with and without association with syphilis: a retrospective case-control study

Abstract

Background: Syphilis is an infectious disease that is at least discussed to be premalignant. This potential, combined with its general pathological impact, raises the question if syphilis increases mortality in oral cancer patients. The aim of the study was to assess if the five-year survival rates among patients suffering from oral squamous cell carcinoma (OSCC) with (cohort I) and without association with syphilis (cohort II) differ.

Methods: Retrospective clinical data of patients diagnosed with OSCC (International Classification of Diseases [ICD]-10 codes C01-06) within the past 20 years from the access date September 25, 2021 were retrieved from the TriNetX network (TriNetX, Cambridge, Massachusetts, USA) to gain initial cohort 0. Subjects also diagnosed with syphilis (ICD-10 codes A51-53) were assigned to cohort I. Cohort II was comprised of the remaining individuals of cohort 0 by creating a group with the same number of patients as cohort I, and by matching for age and gender. Subsequently, Kaplan-Meier analysis and Cox proportional hazards regression were performed, and risk, odds and hazard ratios were calculated.

Results: Of a total of 73,736 patients in cohort 0, 199 individuals were each assigned to cohort I and II. During the five-year period after tumor diagnosis, 39 and 30 patients in cohort I and II died. The five-year survival probabilities did not significantly differ between the cohorts (I vs. II = 74.19% vs. 75.01%; p = 0.52; Log-Rank test), nor the risk of dying (I vs. II = 19.6% vs. 15.08%; risk difference = 4.52%; p = 0.23). The calculated risk, odds and hazard ratios were 1.3 (95% confidence interval [CI] = 0.84; 2.00), 1.37 (95% CI = 0.81; 2.31) and 1.17 (95% CI = 0.73; 1.88), respectively.

Conclusions: The obtained results indicate that the survival rate of individuals with OSCC might not be negatively influenced if syphilis is present/associated. However, the results need to be interpreted cautiously due to limitations related to the retrospective approach, especially as data on the tumor staging were not accessible.