Investigation of genetic sex-specific molecular profile in well-differentiated thyroid cancer: Is there a difference between females and males?

Background: Although more common in females, thyroid cancer is deemed to be more aggressive in males. The reasons for sex disparities in thyroid cancer are not well understood. We hypothesised that differences in molecular mutations between females and males contribute to this phenomenon. Methods: Retrospective multicentre multinational study of thyroid nodules that underwent preoperative molecular profiling between 2015 and 2022. The clinical characteristics and mutational profiles of tumours in female and male patients were compared. Collected data included demographics, cytology results, surgical pathology, and molecular alterations. Results: A total of 738 patients were included of which 571 (77.4%) were females. The extrathyroidal extension was more common in malignancies in males (chi-squared, p = 0.028). The rate of point mutations and gene fusions


| INTRODUCTION
Cancer gender disparity in incidence, disease aggressiveness and prognosis has been observed for a variety of malignancies, including thyroid cancer. 1 Thyroid cancer is more frequent in women, with a ratio of about three women to one man. [1][2][3][4][5] This gender disparity in thyroid cancer incidence is specific to the histologic subtype of thyroid cancer, with papillary and follicular thyroid cancer being more common in women than men, meanwhile medullary and anaplastic thyroid cancer have similar incidence in both genders. 6 Female hormones do appear to explain the higher incidence of thyroid cancer seen in females. 3,7 As a matter of fact, the difference in thyroid cancer incidence is more pronounced during the reproductive years of a women, that is from 15 until approximately 50 years of age. 8 The higher incidence of thyroid cancer seen in females may also be due to earlier and more frequent screening in women, as suggested by some studies. 1 While the gender disparity in thyroid cancer incidence is welldocumented and understood, the reason why thyroid cancer seems to be more aggressive in males is unclear. 1 Among the several traditional staging systems proposed for thyroid cancer, MACIS, AGES, the EORTC, and the Bergen scoring system include the male sex as a negative prognostic factor. 9,10 Neither radiation exposure, environmental nor dietary factors seem to have a role in the genetic sex disparity seen in thyroid cancer. 1 While hormonal factors would appear to be a logical hypothesis to account for the genetic sex disparity, there is no conclusive evidence that hormones may lead to more aggressive behaviour in thyroid cancer. 1,3 In consequence, authors have suspected that a contributor to gender disparity in thyroid cancer prognosis is a difference in molecular alterations. Early studies examining the prevalence of common somatic mutations or gene fusions (e.g., BRAF, RET-PTC, and NTRK) in thyroid cancer showed similar incidence among men and women. 1,3,7,[11][12][13][14] However, studies with large patients' population with comprehensive mutational profiling are scarce. We, therefore, felt appropriate to compare the molecular profile in a large cohort of thyroid cancer patients that underwent preoperative molecular testing and determine whether a difference exists in the molecular profiles of females and males.

| MATERIALS AND METHODS
This study was conducted and reported according to the Equator Guidelines, specifically according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. 15

| Study population
We conducted a retrospective multi-institutional multi-national cohort study of consecutive patients with papillary and follicular thyroid cancer treated at the Sir Mortimer B. Davis-Jewish General Hospital (JGH) and Royal Victoria Hospital (RVH), McGill University, Montreal, QC, Canada, and the Sheba Medical Center (SMC), affiliated with Technion University, Tel-Aviv, Israel, between 1 January 2015 and 1 June 2022. The study was approved by the local Research Ethics Committee (protocol number 2023-3312). Ethical guidelines were followed and all patients provided informed consent for the procedures. Relevant clinicopathologic data was extracted (Appendix A) and data were handled in a coded fashion. Eligibility criteria included previously untreated patients over 18 years of age at diagnosis with papillary and follicular thyroid cancer that underwent fine-needle aspiration biopsy and molecular testing. All patients underwent surgery for their thyroid nodules and all surgical pathologies were reviewed by a dedicated thyroid pathologist. Patients with poorly differentiated thyroid cancer, anaplastic thyroid cancer, and/or medullary thyroid cancer were excluded.
The primary outcome of the study was to determine and compare the genetic sex-associated frequency of known driver mutations among females and males.
2. Retrospective multinational study on molecular mutations in thyroid cancer.
3. The extrathyroidal extension was more frequent in males than females. to perform molecular test and its type was determined by the patient following a discussion with the physician.

| Operative approach and histopathological analysis
Patients underwent either total thyroidectomy or hemithyroidectomy with or without a sentinel lymph node biopsy and neck dissection as required according to the American Thyroid Association guidelines, 20 and the institutions practice. 21,22 Histological diagnosis, tumour size, margins, extra-capsular extension, and extra-thyroidal extension were then examined by experienced thyroid pathologists at our institutions. 23 The tumours were classified according to the 2017 WHO classification of thyroid tumours. 21

| Power analysis
A power analysis was conducted with a power value of 80% and a two-sided confidence level of 0.95. We assumed, based on the available literature, 1,3-5 an overall prevalence of any detectable mutation (due to the heterogeneity of molecular test used, below) of about 80%, with a sex difference of 10% in the mutation rate accountable for the difference in prognosis. The calculated sample size was 656 patients.

| Statistics
For continuous variables, distribution will be evaluated for normality according to Gauss' theorem. For normally distributed variables, mean and standard deviations are given, and comparisons among study groups were done using the t-test. For non-normally distributed variables, median, and interquartile range are given. To compare the distribution among samples, the nonparametric Mann-Whitney U test was used for two samples. Binary variables were associated in contingency tables using the two-tailed Pearson chi-squared test. A p value lower than 0.05 was considered to indicate statistical significance. Statistical analyses will be performed using SPSS 26.0.0.1 software (IBM).

| Descriptive statistics
The study included a total of 738 patients from the three centres was the most commonly used molecular test (63.9%). Upon final surgical pathology, PTC was the most common histology (81.9%). Perineural and lymphovascular invasion were present in 7/690 (1.0%) and 16/691 (2.3%) of cases, respectively. The rate of perineural and lymphovascular invasion did not differ between females and males (chi-squared test, p = 0.646 and p = 0.535, respectively). Extrathyroidal extension was more common in males than in females (chi-squared test, p = 0.028). One hundred fifty-seven (157/695; 22.6%) patients had positive nodal disease upon final pathological examination. These rates did not differ significantly between females and males (chi-squared test, p = 0.44). Further detailed clinicopathological characteristics are shown in Table 1.

| Mutational analysis
Identified mutation rates and each mutation's distribution among the study's groups are presented in Table 2.
Overall, there was no statistically significant difference in the rate of each of the mutations between both genders ( Table 2, p > 0.05).
Similarly, the number of mutations per nodule did not differ between genders. Likewise, the rate of mutation (for each mutation analysed) and the total number of mutations did not differ between each study centre (not shown, p > 0.05). Patients with a TERT promoter mutation were significantly older than those without a TERT promoter mutation [mean age 67.4 (SD 10.8) vs. 50.8 (SD 14.5)] (t-test, p < 0.0001) ( Figure 1B). For all other mutations, there was no statistically significant difference in the age of diagnosis ( p > 0.05, not shown).
For patients harbouring both BRAF V600E and TERT promoter mutations, the age at presentation was significantly greater than for those with only one or none of these mutations [mean age 63.3 years (SD 9.8) vs. 51.1 (SD 14.6)] (t-test, p = 0.009) ( Figure 1C). Interestingly, when stratifying by sex, our analysis showed that this difference was driven by females only. In other words, female patients with BRAF V600E and TERT mutations were significantly older than their wild-type or single mutation counterpart (t-test, p = 0.003) meanwhile the age did not differ between males with or without BRAF V600E and TERT mutations (t-test, p = 0.433).

| DISCUSSION
In the current study, we performed a genetic sex-specific analysis of molecular profiles in patients with thyroid cancer. We found that the absolute rate of mutations is not significantly different between genetic sexes. This is in keeping with a large meta-analysis showing that BRAF V600E mutations do occur at a similar rate in females and males. 13 Other mutations were also found to occur at the same rate in females and males. 1,14 Our study findings, based on a greater number of tested mutations and a larger cohort, are consistent with previous literature, showing the mutation distributions were not significantly different for most mutations. 1,13,14 On further analysis, we discovered interesting trends related to the age of the patients, namely that younger patients have a higher rate of BRAF V600E mutations, meanwhile, TERT promoter mutations occur most commonly in older patients, and this is independent of the sex. This is in line with previous studies in the literature, 24 although some studies did not find BRAF V600E mutation to be associated with age. 13 When both mutations occur simultaneously, we found that BRAF and TERT mutations occur more frequently in older females, but are distributed evenly among males of all ages. This is to our knowledge a finding unique to this study.
The gender differences in thyroid cancer incidence and prognosis is a heavily debated subject.
On the molecular level, studies show that oestrogens have a stimulative effect on proliferation and may contribute to mechanisms leading to DNA damage, thus participating in the initiation of tumorigenesis in thyroid cancer 1 and explaining the higher incidence of thyroid cancer in female patients. Changes in oestrogen receptor (ER) and androgen receptor (AR) expression have been observed in PTC, with ERα and AR expression being associated with worse prognosis and ERβ expression with better prognosis. 25 A study specifically T A B L E 1 Baseline characteristics according to genetic sex of patients. looking at the prognostic risk of male sex in BRAF V600E and wild-type BRAF PTC showed that male sex could be a robust independent risk factor for PTC-specific mortality in BRAF V600E patients but not in wild-type BRAF patients. 26 An analysis of the immune landscape of cancer of TCGA dataset found that programmed death receptor ligand-1 (PD-L1) was more highly expressed among females than in males in thyroid carcinoma and three other cancer types (head and neck squamous cell, renal cell carcinoma, and lung adenocarcinoma). 27 Androgens downregulate PD-L1 in thyroid cancer cells in an AR-dependent mechanism, thus providing another potential explanation on how thyroid cancer escapes the immune system of male patients and results in poorer prognosis. 28,29 This is in line with our findings since concomitant BRAF and TERT mutations have been found to correlate with PD-L1-status in thyroid cancer. 30 On the epidemiological level, data from the SEER database confirm that females have a higher incidence and better prognosis than males. 4 When stratifying for tumour type and reproduction status,  data showed that the genetic sex difference in prognosis was only seen when comparing premenopausal females to age-matched males. 4 However, some studies correcting for age show that genetic sex difference in prognosis is not explained by oestrogen alone. In other words, the worse prognosis seen in post-menopausal females seems to be linked mostly to age-related factors, not to the effect of (missing) female hormones. 1 Furthermore, a large study combining data from the SEER database and prevalence data from autopsies showed that genetic sex disparity is mostly confined to the detection of small subclinical PTCs, which are equally common in both sexes at autopsy but identified during life much more often in females than males. This phenomenon may be associated with gender differences in health care utilisation and patterns of clinical thinking and can harm both females, who are subject to over detection, and males, who may be at risk of underdetection. 7 The over detection of clinically indolent thyroid cancer in females may explain the apparent better prognosis of thyroid cancer in women compared to men. On the other hand, men may present clinically at a later stage with more aggressive disease. In our study, men had more commonly extrathyroidal extension upon surgical pathology and aggressive mutations such as BRAF and TERT at a young age.
Our study has some important limitations. First, we only included patients who had available preoperative molecular testing, therefore introducing a selection bias. Further, the majority of molecular testing was performed with ThyGenX/ThyGeNext, and a minority of patients had other molecular tests including ThyroSeq V3, which is a more comprehensive test. However, since the most common mutations are included in both panels, the introduced bias should be relatively small.
Due to the high cost of the molecular tests used in this study, a selection bias for patients of higher socioeconomic status is present as well. Finally, our study does not account for false-negative molecular profiling studies on cytology. Here the standard of care would have been to perform molecular profiling on surgical pathology samples, however with great additional costs. The rate of false negative is also dependent on the quality of the cytology specimen. In our study, the senior authors performed the vast majority of USFNAC, therefore limiting the false-negative rate. In addition, given that our surgeons and pathologists were not blinded to the molecular test results, there are performance and detection biases to account for. Furthermore, the type of surgery did somewhat differ between the centres, as some practiced sentinel lymph node biopsy meanwhile others did not.
Finally one should notice that we performed molecular testing not only on Bethesda 3 and 4 nodules, but also on Bethesda 5 and 6 nodules, which is not according to the current ATA recommendation. 20 As recently reported however, 31 the main expected clinical benefit of molecular testing for Bethesda 5 and 6 nodules is not to better predict the risk of malignancy but to identify those with higher risk to be aggressive and treat them optimally, in one surgical setting.
In conclusion, our study shows that while the absolute rate and

FUNDING INFORMATION
This study did not receive any funding.

CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT
The datasets generated for this study can be obtained upon reasonable request by email to the corresponding author.

ETHICS STATEMENT
The retrospective study was approved by the local ethics review