Reevaluating the prognostic significance of male gender for papillary thyroid carcinoma and microcarcinoma: a SEER database analysis

The prognostic significance of gender remains controversial for papillary thyroid carcinoma (PTC). In this study, we investigated the associations between gender and prognosis in a large cohort of patients with PTC or PTMC that was diagnosed in 2010–2013 and recorded in the Surveillance, Epidemiology, and End Results cancer registry. The mean ± standard deviation duration of survival for all patients with PTC during the study period was 21.47 ± 14.04 months. In Kaplan-Meier analyses of the entire cohort of PTC patients, survival curves for all-cause death and cancer-specific death declined more sharply for men than for women. Similar results were observed in analyses of patients with PTCs > 1 cm and PTMC. After adjusting for potential confounders, hazard rates indicated significantly elevated all-cause mortality for men in analyses of all PTCs, PTCs > 1 cm, and PTMCs. However, in a confounder-adjusted analysis of patients with PTMC, the hazard rate did not indicate significantly higher mortality for men than for women. Our study demonstrated that male gender is an independent poor prognostic factor for all PTCs and for PTCs > 1 cm. However, gender is not an independent prognostic factor for cause-specific survival in PTMC.

gender for PTC and PTMC. Our analysis was based on the records of 43712 patients with PTCs diagnosed during 2010-2013.

Results
Demographic and clinical features. Baseline clinicopathologic characteristics are summarized and compared between men and women with PTC in Table 1. The mean ± standard deviation (SD) durations of survival during the study period was 21.47 ± 14.04 months for the entire cohort of patients with PTC. Mean age was higher (53.03 ± 15.30 vs 48.68 ± 15.14 years, p < 0.001) and tumor size was larger (22.48 ± 21.79 vs 16.93 ± 17.75 mm, p < 0.001) in men than in women. In patients with PTC, gender was associated with race, T stage, N stage, distant metastasis, extrathyroidal extension, multifocality, radiation, and surgical method (Table 1).
Baseline clinicopathologic characteristics are summarized and compared between men and women with PTC > 1 cm in Table 2. The mean ± SD duration of survival for patients with PTC > 1 cm was 21.52 ± 14.00 months. Mean age was higher (52.26 ± 15.98 vs 47.31 ± 15.83 years, p < 0.001) and tumor size was larger (25.53 ± 19.32 vs 30.90 ± 22.18 mm, p < 0.001) in men than in women. PTC > 1 cm in male patients was more frequently associated with race, T stage, N stage, distant metastasis, extrathyroidal extension, multifocality, radiation, and surgical method (Table 2).
Next, we focused on the baseline clinicopathologic characteristics of patients with PTMC, which are summarized in Table 3. The mean ± SD survival months for PTMC patients was 21.38 ± 14.09 months. Mean age was higher (50.50 ± 13.96 vs 54.61 ± 13.71 years, p < 0.001) and tumor size was larger (5.49 ± 2.94 vs 5.49 ± 2.98 mm, p = 0.0909) in men than in women. PTMC in male patients was more frequently associated with race, T stage, N stage, distant metastasis, extrathyroidal extension, multifocality, and radiation (all p < 0.05), but not with surgical method (p = 0.189) ( Table 3).

Survival analysis of patients with PTC.
In a Kaplan-Meier analysis of the entire cohort of PTC patients, all-cause survival declined sharply for men, but declined more modestly for women (Log-rank test, p < 0.001) (Fig. 1A). Similar results were observed for cause-specific survival (Log-rank test, p = 0.005) (Fig. 1B). Among patients with PTC > 1 cm, all-cause and cause-specific survival curves were poorer for men than women (Log-rank test, both p < 0.001) ( Fig. 2A,B). Among patients with PTMC, all-cause and cause-specific survival curves were also poorer for men than women (Log-rank test; all-cause death, p < 0.001; cancer-specific death, p = 0.005) (Fig. 3A,B). In the entire cohort of patients with PTC, the rate of all-cause death per 1000 person-years was 7.16 (95% confidence interval [CI], 6.52-7.87) for men and 18.28 (95% CI, 16.40-20.37) for women (Table 4). Comparing male vs. female patients, the hazard ratio (HR) for all-cause mortality was 2.53 (95% CI, 2.20-2.90). After adjustment for patient age at diagnosis, race, multifocality, tumor size, extrathyroidal invasion, lymph node metastasis, distant metastasis, radiation, and surgical treatment, these HRs for all-cause and cause-specific mortality became 1.63 (95% CI, 1.39-1.91) and 1.67 (95% CI, 1.42-1.97), respectively. Therefore, male gender had a robust effect on all-cause death in the overall cohort of PTC patients; this effect remained even after additional adjustment for conventional clinicopathological risk factors and treatment methods (Table 4). We subsequently investigated all of the abovementioned associations in subgroup analyses of patients with PTC > 1 cm and PTMC. Similar results were observed (Table 4).

Discussion
Several studies have reported conflicting results regarding the effects of male gender on the prognosis of PTC. Kruijff and colleagues demonstrated the risk of structural recurrence in men was 2.44 times that in women, with a median follow-up period of 31 months 15 . A meta-analysis also showed that male gender was an independent   These inconsistent results introduce a dilemma into the treatment of PTC. More aggressive treatment options (such as total thyroidectomy, or lymph node dissection followed by radioactive iodine ablation) should be considered for men if male gender is a negative independent prognostic factor in PTC. The purpose of this study was to investigate whether male gender was a poor prognostic factor in PTC. Throughout our analyses of the SEER database, male gender was associated with elevated risks of all-cause death and cancer-specific death, both for PTC as a whole and for PTCs > 1 cm. Therefore, based on our results, relatively more aggressive treatment should be considered for these subgroups of patients.
Among patients with PTMC, male gender was associated with significantly elevated risks of all-cause death. However, we found that this association did not remain statistically significant after adjusting for patient age at diagnosis, race, multifocality, and other risk-associated clinicopathological characteristics and treatment methods. This may be attributable to the extremely good prognosis of PTMC, which as 15-year disease-specific survival rates as high as 99% and overall survival of 97.5% 19,20 .
Recent studies have reported that estrogen can regulate thyroid cell proliferation by binding to both ERα, and ERβ. These studies attempted to investigate the associations between estrogen receptor (ER) expression and tumor aggressiveness and prognosis in PTC, but obtained inconsistent conclusions [21][22][23] . Yi et al. investigated the clinical implications of ER coding genes in PTC and found that ESR1 expression and the ESR ratio were  associated with aggressive clinicopathological factors and poor overall survival in women 21 . Huang et al. further observed that ERα stimulates PTC growth and progression, whereas ERb1 inhibits PTC growth and progression 23 . Since ER expression levels differ between men and women, we speculate that ER level may be one of the underlying reasons for the gender differences in PTC tumor aggressiveness and prognosis. However, further research (including both genomic and epidemiologic investigations) should be performed to reveal the source of the gender disparity.
Some inherent limitations must be taken into consideration in the interpretation of our results. The major limitations of this study are that data regarding recurrence are not captured in SEER, and that designation of cancer-specific death is susceptible to overestimation bias, particularly for diseases such as PTMC. Furthermore, vascular invasion, family history, and other histologic findings were not evaluated or included in our study. In addition, the molecular markers such as BRAF point mutation and TERT promotor point mutation, were not obtained in our study or adjusted for in our analyses. Furthermore, given the generally favorable prognosis of PTC, the relatively short study period and follow-up (2010-2013) is a limitation to our analysis.

Conclusion
Our study demonstrated that male gender is an independent poor prognostic factor for all PTCs and for PTCs > 1 cm. However, gender could actually have substantial prognostic relevance for PTMC, but that you do not have enough data or a long enough follow-up period to determine this reliably. These results may be instructive for clinicians. In general, more aggressive treatments should be provided to men with all PTCs and PTCs > 1 cm.

Methods and Materials
Database. We investigated PTC and PTMC in a large cohort of patients from SEER. The SEER project is a United States population-based cancer registry that began in 1973 and is supported by the National Cancer Institute and the Centers for Disease Control and Prevention. It covers approximately 30% of the population of the United States and contains data across multiple geographic regions on incidence, prevalence, mortality, population-based variables, primary characteristics of the tumor, and other attributes.  Data collection and analysis. We examined SEER data for 2010-2013 and selected patients with a diagnosis of PTC, as defined by a combination of ICD-O site code of C73.9 (i.e., thyroid) and papillary histology.
The following diagnostic codes were included in the study: "papillary carcinoma, " "papillary adenocarcinoma, " "oxyphilic adenocarcinoma", "papillary adenocarcinoma, " and "papillary cyst-adenocarcinoma. " Demographic information, age, sex, tumor size, extrathyroidal extension, multifocality, nodal, metastasis, surgical treatment and radiation treatment were compiled from the SEER dataset, and a survival analysis was performed to evaluate the associations between gender and prognosis.
Statistical analysis. Patients were followed up until December 2013. The outcomes measures were thyroid carcinoma-specific mortality and all-cause mortality. Patient survival curves were investigated using Kaplan-Meier analyses, log-rank tests, and Cox proportional hazards regression analyses. HRs were used to show the magnitude of the effect of stages on cancer-specific mortality. Ninety-five percent CIs were used to indicate the significance of the risks. All P values were 2-sided and P less than .05 was regarded as indicating statistical significance. Analyses were performed using SPSS version 19.0 (IBM Corp, Armonk, NY, USA), Stata/SE version 12 (Stata Corp, College Station, TX, USA), and GraphPad Prism version 6 (GraphPad Software Inc, La Jolla, CA, USA).