Demographic and clinicopathological characteristics
Between 2004 and 2015, 1365 PSCC patients who met the inclusion criteria were identified from the SEER database. The demographic and clinicopathological characteristics of the entire cohort and the subgroups stratified by tumor size were presented in Table 1.
Among the 1365 eligible patients, the median age was 65 (55-75) years. There were 1141 (83.6%) white, and 769 (56.3%) of the population were married. PSCC was grade Ⅰ to Ⅳ in 414 (30.3%), 682 (50.0%), 260 (19.0%), and 9 (0.7%) patients, respectively. When the TNM classification system was used, T classifications were as follows: T1 in 700 (51.3%), T2 in 400 (29.3%), T3 in 249 (18.2%), and T4 in 16 (1.2%), respectively. N0 stage (80.1%) predominated in this cohort. The median tumor size was 30 (20-45) mm. Concerning treatment, only 139 of 1365 (10.2%) received chemotherapy, and 98 of 1365 (7.2%) underwent radiotherapy. The patients in the two subgroups were significantly different in marital status, tumor grade, tumor size, T classification, N classification, and the receipt of chemotherapy and radiotherapy.
In this study, the median follow-up time was 62 (95% CI: 58-69) months. By the end of follow up, 557 patients had died, of which 294 (52.8%) patients died from PSCC, and 264 (47.2%) patients died from other causes.
Effect of tumor size on CSM among PSCC patients
The distribution of tumor sizes was presented in Figure 2. 52.3% of tumors in the study population were at most 30 mm, and 47.7% of tumors were larger than 30 mm. Figure 3 exhibited the boxplots of tumor sizes in subgroups based on T and N classification, and significant differences could be observed (all P<0.001). The curves for the cumulative incidence functions of CSM stratified by tumor size was displayed in Figure 4. After controlling competitive risk events, a significant difference was observed for the cumulative incidence of CSM between two subgroups (P<0.001).
The effects of tumor size as both a continuous and categorical variable on CSM were separately analyzed. As shown in Figure 5, restricted cubic splines were used to flexible model and visualize the association between tumor size and CSM. The plot depicted an apparent tendency that the risks of CSM increased as tumors enlarged. Multivariable competing-risks regression analyses of CSM were displayed in Table 2. When tumor size was incorporated into the competing-risks regression model as a categorical variable, the estimates indicated that the PSCC patients with tumors >30 mm were more likely to succumb to CSM than those with tumors ≤30 mm (HR=1.57, 95%CI: 1.23-2.01, P<0.001). Race, Tumor grade, T classification, and N classification were independent predictors for CSM. However, no significant association with CSM was found with age at diagnosis, marital status, chemotherapy, and radiotherapy.
Sensitivity analyses for the association between tumor size and CSM
In the current study, subgroup analyses stratified by T classification were performed to verify the association between tumor size and CSM. As shown in Figure 4, there were statistical differences in cumulative incidence of CSM across the subgroups stratified by T classification (all P<0.05). The 5-year cumulative incidences of CSM for PSCC patients with tumors ≤30 mm and >30 mm in T1 subgroup were 13.2% and 19.7%, respectively. The corresponding 5-year cumulative incidences of CSM in T2 subgroup were 24.8% and 32.0%. For T3 subgroup, 5-year cumulative incidences of CSM for patients with tumors ≤30 mm and >30 mm were 25.5% and 39.5%, respectively. As Table 3 showed, after adjusting other variables, PSCC patients with tumors >30 mm had a 1.56-fold higher chance of CSM than those with tumors ≤30 mm in T1 subgroup (P=0.036). Similarly, for T3 subgroup, the patients with tumors > 30 mm had worse cancer-specific survival relative to those with tumors ≤30 mm (HR=2.51, 95%CI: 1.41-4.45, P=0.002). In the multivariable analysis involving T2 classification, tumors ≤30 mm seemed to be a predictor of better cancer-specific survival (HR=1.49, 95%CI: 0.97-2.27, P=0.067), though the statistical significance of the association was of borderline value. Due to the limitation of sample size (n=16), the subgroup analysis on T4 classification was not performed.
Association of tumor size with lymph node status.
Tumor size as a categorical variable was analyzed by using multivariable logistic regression models. Compared with tumors ≤30 mm, the odds ratio (OR) for lymph node metastasis was 1.46 (95% CI: 1.03-2.07, P=0.034) for tumors >30 mm. Additionally, age at diagnosis, tumor grade, T classification, chemotherapy, and radiotherapy were all significantly associated with lymph node metastasis. However, no statistical significance was reached with respect to race and marital status. The results of multivariable logistic regression analysis were presented in Table 4.