Abstract
Background:
Studies conflict on whether sex influences survival in gastroenteropancreatic neuroendocrine tumors (GEP-NETs). GEP-NETs express receptors and genes responsive to female hormones. We hypothesized that women would have improved survival and this difference would be greater in premenopausal age women compared to older women.
Materials and Methods:
The National Cancer Database (NCDB) from 2004 to 2016 was queried for patients with GEP-NETs based on histologic code. Demographic, tumor, treatment, and socioeconomic characteristics were compared between men and women and age ≤45 or >65 years using Fisher exact and Wilcoxon tests as appropriate. The primary endpoint was overall survival (OS), assessed by Kaplan-Meier survival analysis.
Results:
Included in the study were 73,521 patients with small bowel (SBNETs), gastric (GNETs), or pancreas (PNETs) neuroendocrine tumors (36,197 female, 37,324 male). Women lived longer regardless of primary site, with the largest difference in GNETs (median OS 139 vs. 85 months) and smallest in SBNETs (121 vs. 116, p<0.001 for both). While male patients more often had high grade and metastatic disease, female sex remained independently associated with improved OS after adjusting for confounders (HR 0.84, p<0.001). In GNETs and SBNETs, female sex had a larger beneficial effect on OS in premenopausal than postmenopausal patients.
Conclusion:
Women with GEP-NETs have improved survival over men, especially in the premenopausal age group. This may be due to a protective effect of female hormones, however further studies are necessary to uncover the biologic basis of this difference.
Keywords: Gastroenteropancreatic Neuroendocrine Tumor, Overall Survival, Sex-based Differences, National Cancer Database
Introduction
Gastroenteropancreatic neuroendocrine tumors (GEP-NETs) are a heterogeneous group of neoplasms arising from neural crest-derived cells. Overall survival in GEP-NETs depends on the histological grade and stage of the disease. Epidemiologic studies have revealed differences in GEP-NETs between male and female patients. The prevalence of GEP-NETs, overall, is higher in females and there are slight differences in the location of primary tumors, with stomach and appendiceal NETs more common in women, and ileal NETs more common in men.[1, 2]
Many GEP-NETs express estrogen receptors (ER) and progesterone receptors (PR).[3–5] A recent study by Estrella et al. demonstrated that the majority of pancreatic neuroendocrine tumors (PNETs) express estrogen receptor beta (ER-B) and that higher expression of ER-B was associated with more favorable prognostic features, including being well-differentiated, low grade and of smaller size.[6] In small bowel neuroendocrine tumors (SBNETs), mesenteric metastases more commonly occur in men than women, with a possible association with increased expression of androgen receptors and estrogen receptor alpha.[7]
While preliminary data demonstrate that GEP-NETs show differences in ER and PR expression and primary site incidence by sex, it is not clear whether these biologic variations lead to survival differences.[3, 6] Studies focusing on sex disparities in neuroendocrine tumors (NETs) have yielded conflicting results regarding differences in sex-specific survival. A SEER study from 1973 to 2007 noted differences in overall survival by sex in SBNETs by univariate analyses.[8] A more recent SEER study spanning 2004 to 2015 found that sex was an independent risk factor for survival in both GNETs and SBNETs.[2] On the other hand, another study of the National Cancer Database (NCDB) from 2004 to 2015 concluded that sex was not an independent factor in overall survival for low-grade PNETs.[9] Smaller, single-institutional studies of GEP-NETs have not found sex to be associated with overall survival, however these may not have had the power to identify small differences in survival with limited samples.[10–13] One possible reason why these studies have conflicting results is that analysis was performed solely by lumping all ages together rather than sub-grouping by pre and post-menopausal ages, thus diluting out a possible effect.
This study seeks to investigate the presence of sex-based survival differences for gastroenteropancreatic neuroendocrine tumors (GEP-NETs), with attention to primary tumor location. The NCDB was examined to determine whether sex impacts upon survival in GEP-NETs, hypothesizing that women would have improved survival, and that since hormonal difference could be responsible for these effects, survival differences would be greater in pre-menopausal age women compared to post-menopausal age women.
Methods
Study Population
A retrospective review of the NCDB from 2004 to 2016 was conducted. The NCDB is a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society. The data used in the study are derived from a de-identified NCDB file. The American College of Surgeons and the Commission on Cancer have not verified and are not responsible for the analytic or statistical methodology employed, or the conclusions drawn from these data by the investigator. The NCDB is HIPAA-compliant with publicly available data. Patient consent was waived due to the use of National Cancer Database (NCDB) de-identified data and this study was Institutional Review Board-Exempt. Tumors from the small intestine, stomach, and pancreas were extracted using the International Classification of Disease (ICD) anatomical classification (codes: C16.0–9, C17.0–3, C17.8, C17.9, C25.0–4, C25.7–9). Colon and rectal neuroendocrine tumors were excluded due to their unique biology (higher rates of high grade in colon primary and small, benign tumors in rectal).[14–19] Patients were identified to have NETs based on histopathological identification (ICD Oncology Histology codes: 8013, 8150, 8151, 8152, 8153, 8155, 8156, 8157, 8240, 8241, 8242, 8246, and 8249). Given new findings regarding the biological and clinical differences between duodenal NETs and other small intestinal NETs, duodenal NETs were analyzed separately.[20, 21] Patient demographics (i.e. race, sex, age, insurance status, treatment facility type, Charlson-Deyo Score), tumor characteristics (i.e. differentiation and TNM stage), and treatment (i.e surgical resection and chemotherapy) were abstracted. The NCDB characterization of tumor grade was used as follows: 1) Grade 1 – Well-differentiated, 2) Grade 2 - Moderately differentiated, 3) Grade 3 – Poorly differentiated, and 4) Grade 4 – Undifferentiated, with NCDB Grade 3 and Grade 4 grouped as “High Grade” tumors. Overall survival was the primary endpoint. A date range of 2004 to 2016 was chosen to allow for more accurate analysis of overall survival as GEP-NETs in general have an indolent course.[22, 23] For survival analysis, CoC/NCDB expects participating institutions to provide a 90% rate of follow up over a 5-year period, and follow-up in this dataset extended to approximately 2018.[24]
As NCDB does not capture date of menopause, female patients aged 45 or less were compared to those older than 65 to approximate pre- and post-menopausal groups to evaluate potential effects of female hormone exposure on survival. While other studies have used cutoffs younger than 65 to designate postmenopausal status, these wider cutoffs were chosen to minimize presence of confounding perimenopausal patients.
Statistical Analysis
Summary statistics describe the study population. Fisher’s exact test and Wilcoxon test were utilized to compare categorical variables and continuous variables for the study population, respectively. Kaplan-Meier method was used to analyze median overall survival times and Kaplan-Meier curves were compared with log rank test.[25] Univariable hazard ratio analysis was performed to identify risk factors for OS. To account for confounders, factors showing significant association with survival on univariate analysis were included in multivariable Cox proportional hazard ratio regression models.[26] Initial analysis performed included all ages. Subsequent subgroup analysis stratifies patient age as a surrogate for pre and postmenopausal status. An alpha significance level of ≤ 0.05 was utilized to indicate statistical significance and inclusion in multivariate models. Statistical analysis was performed using RStudio (RStudio, Boston, MA).
Results
Patient Data
In the NCDB from 2004–2016, 73,521-NET patients were identified including 38,978 small intestine, 24,068 pancreas and 10,477 gastric primaries (Table 1). Cases of small intestinal NETs also included 9,837 duodenal NETs. Duodenal NETs were distinguished from jejuno-ileal (SBNETs) for analysis. Among patients with SBNETs, 51.7% were male and 48.3% female, in pancreatic NETs (PNETs), 54.0% were male and 46.0% female, and in gastric NETs (GNETs), males accounted for 39.9% of cases and females 60.1%. In gastric, pancreatic and duodenal NETs, males were older than females at diagnosis (63.0 vs 61.5 years in GNETs, p<0.001; 61.2 vs 59.6 in PNETs, p<0.001; 64.1 vs 63.4 in DNETs, p<0.001). Women with SBNETs were slightly older at presentation than men. There was no difference in Charlson-Deyo score between males and females for GNETs, DNETs and SBNETs (p=0.3 for all), however for PNETs, a larger proportion of females had Charlson-Deyo scores of 0 compared to males, suggesting fewer comorbid conditions. Median follow-up duration [95% CI] for the study was 57.2 [56.7 – 57.7] months, 63.9 [63.1 – 64.7] months for SBNETs, 49.5 [48.8 – 50.3] months for PNETs, 56.2 [54.7 – 57.7] months for GNETs and 57.4 [55.9 – 58.8] months for DNETs.
Table 1.
Summary of Study Population
| Gastric Neuroendocrine Tumors (GNETs) | Pancreatic Neuroendocrine Tumors (PNETs) | Small Bowel Neuroendocrine Tumors (SBNETs) | Duodenal Neuroendocrine Tumors (DNETs) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Male (n=4,183) | Female (n=6,294) | P-value | Male (n=13,005) | Female (n=11,063) | P-value | Male (n=15,092) | Female (n=14,047) | P-value | Male (n=5,044) | Female (n=4,793) | P-Value | |
|
| ||||||||||||
| Age | 63.0 [62.6 – 63.4] | 61.5 [61.2 – 61.9] | <0.01 | 61.2 [61.0 – 61.4] | 59.6 [59.4 – 59.9] | <0.01 | 63.2 [63.0 – 63.4] | 63.3 [63.1 – 63.6] | 0.39 | 64.1 [63.8 – 64.5] | 63.4 [63.0 – 63.8] | 0.01 |
| Race | ||||||||||||
| White Non-Hispanic | 2994 (71.6%) | 4254 (67.6%) | <0.01 | 10443 (80.3%) | 8153 (73.7%) | <0.01 | 12442 (82.4%) | 11274 (80.3%) | <0.01 | 3286 (65.1%) | 3055 (63.7%) | 0.06 |
| Black | 571 (13.6%) | 983 (15.6%) | 1223 (9.4%) | 1731 (15.6%) | 1778 (11.8%) | 1989 (14.2%) | 1242 (24.6%) | 1247 (26.0%) | ||||
| Asian | 165 (3.9%) | 118 (1.9%) | 395 (3.0%) | 308 (2.8%) | 105 0.7%) | 71 (0.5%) | 141 (2.8%) | 109 (2.3%) | ||||
| Hispanic | 350 (8.4%) | 804 (12.8%) | 667 (5.1%) | 642 (5.8%) | 551 (3.7%) | 518 (3.7%) | 257 (5.1%) | 275 (5.7%) | ||||
| Stage | ||||||||||||
| 0 | 88 (2.1%) | 218 (3.5%) | 8 (0%) | 10 (0%) | 11 (0%) | 12 (0%) | 17 (0.3%) | 18 (0.4%) | ||||
| I | 985 (23.5%) | 1922 (30.5%) | 4187 (32.2%) | 3970 (35.9%) | 668 (4.4%) | 759 (5.4%) | 1664 (33.0%) | 1614 (33.7%) | ||||
| II | 448 (10.7%) | 534 (8.5%) | 2293 (17.6%) | 1930 (17.4%) | 1456 (9.6%) | 1087 (7.7%) | 569 (11.3%) | 446 (9.3%) | ||||
| III | 281 (6.7%) | 198 (3.1%) | 315 (2.4%) | 199 (1.8%) | 5125 (34.0%) | 4684 (33.3%) | 362 (7.2%) | 372 (7.8%) | ||||
| IV | 708 (16.9) | 371 (5.9%) | <0.01 | 4186 (32.2%) | 3139 (28.4%) | <0.01 | 2910 (19.3%) | 2917 (20.8%) | <0.01 | 245 (4.9%) | 236 (4.9%) | 0.88 |
| Unknown | 1673 | 3051 | 2016 | 1815 | 4922 | 4588 | 2187 | 2107 | ||||
| Grade | ||||||||||||
| Well-differentiated | 1521 (36.4%) | 2710 (43.1%) | 5667 (43.6%) | 5225 (47.2%) | 7641 (50.6%) | 7038 (50.1%) | 1521 (30.2%) | 2710 (56.5%) | ||||
| Moderately differentiated | 333 (8.0%) | 453 (7.2%) | 1535 (11.8%) | 1299 (11.7%) | 1600 (10.6%) | 1624 (11.6%) | 333 (6.6%) | 453 (9.5%) | ||||
| Poorly or undifferentiated | 676 (16.2%) | 315 (5.0%) | <0.01 | 1137(8.7%) | 745 (6.7%) | <0.01 | 260 (1.7%) | 228 (1.6%) | 0.52 | 676 (13.4%) | 315 (6.6%) | <0.01 |
| Unknown | 1651 | 2816 | 4666 | 3794 | 5591 | 5157 | 1653 | 2816 | ||||
| Received Surgery | 2406 (57.5%) | 4009 (63.7%) | <0.01 | 6853 (52.7%) | 6407 (57.9%) | <0.01 | 13756 (91.1%) | 12870 (91.6%) | 0.15 | 3448 (68.4%) | 3220 (67.2%) | 0.21 |
| Received Chemotherapy | 613 (14.7%) | 288 (4.6%) | <0.01 | 2979 (22.9%) | 2092 (18.9%) | <0.01 | 814 (5.4%) | 757 (5.4%) | 0.980 | 154 (3.1%) | 119 (2.5%) | |
| Charlson-Deyo Score | ||||||||||||
| 0 | 2817 (67.3%) | 4294 (68.2%) | 0.34 | 9221 (70.9%) | 8106 (73.3%) | <0.01 | 11285 (74.8%) | 10606 (75.5%) | 0.15 | 3306 (65.5%) | 3089 (64.4%) | 0.26 |
| 1 | 899 (21.5%) | 1317 (20.9%) | 2750 (21.1%) | 2230 (20.2%) | 2776 (18.4%) | 2600 (18.5%) | 1093 (21.7%) | 1142 (23.8%) | ||||
| 2 | 299 (7.1%) | 445 (7.1%) | 644 (5.0%) | 500 (4.5%) | 699 (4.6%) | 588 (4.2%) | 390 (7.7%) | 354 (7.4%) | ||||
| 3+ | 168 (4.0%) | 238 (3.8%) | 390 (3.0%) | 227 (2.1%) | 332 (2.2%) | 253 (1.8%) | 255 (5.1%) | 208 (4.3%) | ||||
Several factors potentially influencing overall survival, including presence of metastasis (i.e. Stage IV disease), presence of nodal disease, tumor grade, and receipt of surgery or chemotherapy showed differences between male and female patients. For GNETs and PNETs, more male patients presented with Stage IV disease than female patients, 16.9% vs. 5.9% (p<0.001) and 32.2% vs. 28.4% (p<0.001), respectively. For SBNETs, female patients had slightly higher rates of Stage IV disease, 20.8 vs. 19.3% (p=0.002). While most patients with PNETs or GNETs presented with earlier stage disease (Stage I or II), a majority of SBNETs presented with later stage disease (Stage III or IV). Male patients with GNETs, PNETs, and DNETs were more likely to present with high-grade disease, while rates of high-grade disease were not different between male and female SBNET patients. In GNETs and PNETs, 16.2 and 8.7% of male patients had high-grade tumors compared to 5.0 and 6.7% of female patients. There was no significant difference in the rate of poorly differentiated tumors between males and females for SBNETs (p=0.5). In GNETs and PNETs, female patients had surgical resection more often than male patients (63.7 vs. 57.5%, p<0.001 and 57.9 vs. 52.7%, p<0.001), while there was no difference in surgery rates between males and females with SBNETs (p=0.9). Patients with SBNETs had the highest rates of surgical resection while PNETs had the lowest rates. In GNETs and PNETs, males had higher rates of chemotherapy use than female patients (14.7% vs. 4.6%, p<0.01, 22.9% vs. 18.9%, p<0.01).
Identification of Factors Affecting Overall Survival
To assess sex-based survival differences, Kaplan-Meier Survival curves for each primary tumor location were compared (Figure 1). When assessing all GEP-NETs in patients of all ages, male patients had significantly shorter overall survival than females with median survival of 98.8 vs. 118.8 months (p<0.01). Regardless of primary site, female patients were noted to have a better overall survival than male patients (p<0.01 for all primary sites). This difference in survival was largest for GNETs (138.9 vs. 85.2 months) and smallest for SBNETs (126.0 vs. 119.0 months). In comparison with patients having SBNETs, DNETs, and GNETs, patients with PNETs had worse overall survival.
Figure 1. Kaplan-Meier Curves for Overall Survival with corresponding number at risk based on Sex for:
A) Gastric Neuroendocrine Tumors (GNETs) (Median survival 138.9 vs 85.2 months for females and males, respectively, p<0.01), B) Pancreatic Neuroendocrine Tumors (PNETs) (Median survival 87.0 vs 67.4 months, p<0.01), C) Small Bowel Neuroendocrine Tumors (SBNETs) (Median survival 126.0 vs 119.0 months, p<0.01), and D) Duodenal Neuroendocrine Tumors (DNETs) (Median survival 136.0 vs 117.0 months, p<0.01). Shaded areas of curves represent 95% confidence intervals. *Number at risk at time 0 months may differ from n in table 1 due to exclusion of patients with missing data.
To determine whether imbalanced clinical factors between male and female patients could account for survival differences, univariate and multivariate Cox hazard ratio regression analysis was performed to identify risk factors for OS. In the entire cohort of 73,521 NETs, age, sex, Charlson-Deyo Score of 0, receipt of chemotherapy or surgery, stage IV, and high-grade disease, private insurance, and care at an academic institution were independently associated with a difference in OS. Worse OS was associated with older age, receipt of chemotherapy, stage IV disease, and high-grade disease. Of note, high histological grade had a larger magnitude of effect than high stage disease (HR 4.9 vs. 2.9). Improved overall survival was associated with female sex, receiving surgical treatment, private insurance, receiving care at an academic institution, and Charlson-Deyo Score of 0. The largest effect size was seen with receiving surgical management (HR 0.28 [0.27–0.28] p<0.01). Subgroup analysis was performed based on primary tumor location, with the same associations seen for all primary sites (Table 2). After adjustment for these characteristics, female sex continued to significantly correlate with improved overall survival for all NETs analyzed together (HR 0.85, p<0.01) as well as each of the individual primary tumor locations: GNETs (HR 0.75, p<0.001), PNETs (HR 0.92, p<0.01), SBNETs (HR 0.90, p<0.01) and DNETs (0.80, p<0.01).
Table 2.
Multivariate Hazard Ratio Analysis of Risk Factors for Overall Survival
| Gastric Neuroendocrine Tumors (GNETs) | Pancreatic Neuroendocrine Tumors (PNETs) | Small Bowel Neuroendocrine Tumors (SBNETs) | Duodenal Neuroendocrine Tumors (DNETs) | |||||
|---|---|---|---|---|---|---|---|---|
| HR [95% CI] | P-value | HR [95% CI] | P-value | HR [95% CI] | P-value | HR [95% CI] | P-value | |
|
| ||||||||
| Female Sex | 0.76 [0.70 – 0.82] | <0.01 | 0.92 [0.88 – 0.96] | <0.01 | 0.90 [0.87 – 0.95] | <0.01 | 0.80 [0.73 – 0.86] | <0.01 |
| Race (Ref: Black) | ||||||||
| Non-Hispanic White | 0.82 [0.75 – 0.92] | <0.01 | 1.01 [0.91 – 1.09] | 0.64 | 0.97 [0.90 – 1.04] | 0.38 | 0.91 [0.83 – 0.99] | 0.04 |
| Asian | 0.78 [0.61 – 1.00] | 0.05 | 0.97 [0.83 – 1.13] | 0.65 | 0.97 [0.70 – 1.33] | 0.83 | 0.67 [0.49 – 0.91] | 0.01 |
| Hispanic | 0.54 [0.47 – 0.63] | p<0.01 | 0.87 [0.74 – 1.04] | 0.12 | 0.84 [0.65 –1.08] | 0.17 | 0.80 [0.56 – 1.15] | 0.23 |
| Age | ||||||||
| Age<45 | 0.60 [0.51 – 0.72] | p<0.01 | 0.74 [0.69 – 0.80] | p<0.01 | 0.46 [0.40 – 0.54] | p<0.01 | 0.50 [0.40 – 0.63] | p<0.01 |
| Age>65 | 1.87 [1.70 – 2.05] | p<0.01 | 1.41 [1.34 – 1.50] | p<0.01 | 2.14 [2.02 – 2.28] | p<0.01 | 1.64 [1.49 – 1.81] | p<0.01 |
| Had Surgery | 0.46 [0.42 – 0.50] | p<0.01 | 0.27 [0.26 – 0.29] | p<0.01 | 0.51 [0.47 – 0.54] | p<0.01 | 0.45 [0.41 – 0.49] | p<0.01 |
| Had Chemotherapy | 1.12 [0.99 – 1.28] | 0.08 | 1.23 [1.17 – 1.30] | p<0.01 | 1.42 [1.31 – 1.54] | p<0.01 | 1.70 [1.38 – 2.08] | p<0.01 |
| Private Insurance | 0.65 [0.59 – 0.72] | p<0.01 | 0.79 [0.75 – 0.84] | p<0.01 | 0.64 [0.60 – 0.69] | p<0.01 | 0.61 [0.55 – 0.68] | p<0.01 |
| Academic Institution | 0.83 [0.77 – 0.90] | p<0.01 | 0.86 [0.83 – 0.90] | p<0.01 | 0.93 [0.88 – 0.97] | p<0.01 | 0.81 [0.75 – 0.88] | p<0.01 |
| Charlson-Deyo Score = 0 | 0.69 [0.64 – 0.74] | p<0.01 | 0.80 [0.76 – 0.84] | p<0.01 | 0.67 [0.64 – 0.70] | p<0.01 | 0.64 [0.59 – 0.69] | p<0.01 |
| Nodal Disease | 1.19 [0.94 – 1.50] | 0.16 | 1.10 [0.94 – 1.27] | 0.23 | 0.84 [0.73 – 0.96] | 0.01 | 0.89 [0.65 – 1.23] | 0.49 |
| Metastatic Disease | 2.68 [2.38 – 3.01] | p<0.01 | 1.88 [1.79 – 1.98] | p<0.01 | 1.51 [1.32 – 1.60] | p<0.01 | 2.05 [1.73 – 2.43] | p<0.01 |
| Grade 3 or 4 Disease | 3.05 [2.69 – 3.47] | p<0.01 | 2.56 [2.40 – 2.72] | p<0.01 | 2.85 [2.52 – 3.22] | p<0.01 | 3.36 [2.72 – 4.15] | p<0.01 |
If hormonal effects account for the sex-specific survival difference, these survival differences might be more pronounced in pre-menopausal women. Patients ≤45 years old were analyzed as a “pre-menopause” group and compared to those age >65 (“post-menopause”) (Table 3 and 4). Male patients were compared to female patients of the same ages, adjusting for confounders. For patients aged ≤45, female sex was associated with better OS in comparison to male sex for GNETs and SBNETs, but not PNETs or DNETs. For patients aged >65, female sex was associated with an improved overall survival for all primary tumor locations. In GNETs and SBNETs as well as all NETs analyzed together, the beneficial effect of female sex on OS was more pronounced in the younger age group vs. the older age group (HR 0.74 vs. 0.88, p<0.01, for all NETs). Kaplan Meier survival analysis was additionally performed on sex and age subgroups. For SBNETs, GNETs, and DNETs, median survival was not reached for females with age ≤45. For GNETs, PNETs, and SBNETs, but not DNETs, Kaplan Meier survival analysis demonstrated improved survival for females age ≤45 compared to males age ≤45 (Figure 2). Females with age >65 had improved OS in comparison with their male counterparts regardless of primary tumor location: GNETs (77.6 vs 41.0 months, p<0.01), PNETs (52.6 vs 43.5 months, p<0.01), SBNETs (78.8 vs 75.1 months, p=0.01) and DNETs (95.9 vs 77.7, p<0.01).
Table 3.
Multivariate Cox Hazard Ratio Analysis for Overall Survival in Age ≤ 45
| Gastric Neuroendocrine Tumors (GNETs) (N= 870 F, 415 M) | Pancreatic Neuroendocrine Tumors (PNETs) (N= 1787 F, 1576 M) | Small Bowel Neuroendocrine Tumors (SBNETs) (N= 1247 F, 1207 M) | Duodenal Neuroendocrine Tumors (DNETs) (N= 475 F, 375 M) | |||||
|---|---|---|---|---|---|---|---|---|
| HR [95% CI] | P-value | HR [95% CI] | P-value | HR [95% CI] | P-value | HR [95% CI] | P-value | |
|
| ||||||||
| Female Sex | 0.52 [0.32 – 0.65] | <0.01 | 0.94 [0.82 – 1.09] | 0.46 | 0.57 [0.43 – 0.76] | <0.01 | 0.70 [0.44 – 1.11] | 0.13 |
| Race (Ref: Black) | ||||||||
| Non-Hispanic White | 0.97 [0.65 – 1.44] | 0.87 | 1.02 [0.84 – 1.23] | 0.84 | 1.38 [0.92 – 2.08] | 0.12 | 0.94 [0.58 – 1.51] | 0.80 |
| Asian | 1.63 [0.02 – 1.20] | 0.07 | 0.99 [0.65 – 1.52] | 0.98 | 1.67 [0.38 – 7.22] | 0.50 | 0.51 [0.06 – 4.03] | 0.53 |
| Hispanic | 0.38 [0.15 – 0.96] | 0.04 | 0.78 [0.53 – 1.13] | 0.19 | 4.73 [0.11 –1.92] | 0.29 | 0.80 [0.56 – 1.15] | 0.23 |
| Had Surgery | 0.75 [0.51 – 1.10] | 0.14 | 0.27 [0.22 – 0.32] | <0.01 | 0.35 [0.24 – 0.52] | <0.01 | 0.45 [0.41 – 0.49] | <0.01 |
| Had Chemotherapy | 3.94 [2.16 – 7.13] | <0.01 | 1.86 [1.58 – 2.18] | <0.01 | 2.56 [1.77 – 3.70] | <0.01 | 1.70 [1.38 – 2.08] | <0.01 |
| Private Insurance | 0.39 [0.28 – 0.55] | <0.01 | 0.76 [0.66 – 0.89] | <0.01 | 0.44 [0.33 – 0.59] | <0.01 | 0.61 [0.55 – 0.68] | <0.01 |
| Academic Institution | 1.05 [0.71 – 1.54] | 0.80 | 1.06 [0.91 – 1.24] | 0.48 | 0.99 [0.72 – 1.38] | 0.99 | 0.81 [0.75 – 0.88] | <0.01 |
| Charlson-Deyo Score = 0 | 0.70 [0.47 – 1.03] | 0.07 | 1.00 [0.82 – 1.22] | 0.98 | 0.74 [0.49 – 1.10] | 0.14 | 0.64 [0.59 – 0.69] | <0.01 |
| Nodal Disease | 0.73 [0.25 – 2.14] | 0.57 | 1.04 [0.64 – 1.67] | 0.88 | 0.78 [0.30 – 2.06] | 0.62 | 0.89 [0.65 – 1.23] | 0.49 |
| Metastatic Disease | 3.59 [2.29 – 5.64] | <0.01 | 2.14 [1.80 – 2.55] | <0.01 | 1.78 [1.27 – 2.49] | <0.01 | 2.05 [1.73 – 2.43] | <0.01 |
| Grade 3 or 4 Disease | 2.59 [1.46 – 4.60] | <0.01 | 2.60 [2.14 – 3.16] | <0.01 | 6.77 [4.01 – 11.44] | <0.01 | 3.36 [2.72 – 4.15] | <0.01 |
Table 4.
Multivariate Cox Hazard Ratio Analysis for Overall Survival in Age > 65
| GNETs (N= 2577 F, 1853 M) | PNETs (N= 3978 F, 5152 M) | SBNETs (N= 6225 F, 6630 M) | DNETs (N= 2251 F, 2424 M) | |||||
|---|---|---|---|---|---|---|---|---|
| HR [95% CI] | P-value | HR [95% CI] | P-value | HR [95% CI] | P-value | HR [95% CI] | P-value | |
|
| ||||||||
| Female Sex | 0.78 [0.71 – 0.85] | <0.01 | 0.92 [0.86 – 0.98] | 0.01 | 0.93 [0.88 – 0.98] | 0.01 | 0.81 [0.74 – 0.90] | <0.01 |
| Race (Ref: Black) | ||||||||
| Non-Hispanic White | 0.77 [0.67 – 0.88] | <0.01 | 1.06 [0.95 – 1.19] | 0.29 | 0.97 [0.89 – 1.07] | 0.57 | 0.98 [0.87 – 1.10] | 0.68 |
| Asian | 0.89 [0.66 – 1.21] | 0.47 | 1.06 [0.83 – 1.36] | 0.62 | 0.88 [0.58 – 1.33] | 0.54 | 0.76 [0.52 – 1.12] | 0.16 |
| Hispanic | 0.88 [0.64 – 1.21] | 0.43 | 0.94 [0.73 – 1.22] | 0.65 | 0.90 [0.64 –1.25] | 0.53 | 1.00 [0.67 – 1.48] | 0.99 |
| Had Surgery | 0.46 [0.42 – 0.51] | <0.01 | 0.32 [0.29 – 0.34] | <0.01 | 0.56 [0.51 – 0.61] | <0.01 | 0.45 [0.41 – 0.50] | <0.01 |
| Had Chemotherapy | 0.72 [0.61 – 0.86] | <0.01 | 1.03 [0.96 – 1.12] | 0.42 | 1.15 [1.02 – 1.29] | 0.02 | 1.25 [0.95 – 1.67] | 0.12 |
| Private Insurance | 0.80 [0.69 – 0.93] | <0.01 | 0.96 [0.87 – 1.06] | 0.96 | 0.84 [0.77 – 0.93] | <0.01 | 0.61 [0.55 – 0.68] | <0.01 |
| Academic Institution | 0.79 [0.71 – 0.87] | <0.01 | 0.82 [0.77 – 0.88] | <0.01 | 0.86 [0.81 – 0.92] | <0.01 | 0.81 [0.75 – 0.88] | <0.01 |
| Charlson-Deyo Score of 0 | 0.71 [0.64 – 0.77] | <0.01 | 0.77 [0.72 – 0.83] | <0.01 | 0.67 [0.63 – 0.71] | <0.01 | 0.64 [0.59 – 0.69] | <0.01 |
| Nodal Disease | 1.10 [0.82 – 1.48] | 0.51 | 0.96 [0.79 – 1.19] | 0.76 | 0.80 [0.68 – 0.94] | 0.01 | 0.89 [0.65 – 1.23] | 0.49 |
| Metastatic Disease | 2.50 [2.16 – 2.90] | <0.01 | 1.91 [1.77 – 2.06] | <0.01 | 1.37 [1.27 – 1.47] | <0.01 | 2.05 [1.73 – 2.43] | <0.01 |
| Grade 3 or 4 Disease | 3.22 [2.75 – 3.77] | <0.01 | 2.49 [2.27 – 2.73] | <0.01 | 2.60 [2.22 – 3.04] | <0.01 | 3.36 [2.72 – 4.15] | <0.01 |
Figure 2. Kaplan Meier Curves for Overall Survival for Age ≤ 45:
A) Gastric Neuroendocrine Tumors (GNETs) (Median survivals not reached, p<0.01), B) Pancreatic Neuroendocrine Tumors (PNETs) (Median survival 142.0 vs 117.0 months, p<0.01), C) Small Bowel Neuroendocrine Tumors (SBNETs) (Median survivals not reached, p<0.01), and D) Duodenal Neuroendocrine Tumors (DNETs) (Median survivals not reached, p=0.12). Shaded areas of curves represent 95% confidence interval.
Discussion
Most GEP-NETs have relatively indolent growth, with tumor behavior and survival largely dictated by grade and cellular differentiation.[27] While low and intermediate grade tumors have a 5-year survival of 75% and 62%, respectively, the survival of patients with high grade tumors is markedly lower at 24% over 5 years.[28, 29] Although medical therapies for symptoms and tumor growth control for late stage and metastatic tumors have improved, better understanding of tumor genomics and biology could allow for more individualized treatments.[30] Understanding sex-based cancer differences both biologically and clinically may support development of personalized therapeutic strategies in the future.[31, 32]
For neuroendocrine tumors, opportunities for therapeutics addressing sex hormones have been described. In 1980, Keshgegian et al. first reported estrogen receptor positivity in 2 cases of malignant carcinoid tumor and hypothesized that these tumors may be responsive to hormonal therapy.[33] The earliest use of hormonal therapy in NETs was in 1981 when tamoxifen was given to manage symptoms associated with carcinoid syndrome with additional successful anecdotal cases cited throughout the 1980s.[34–36] While tamoxifen was a relatively common therapy for carcinoid syndrome in the early 1980s, its usage dwindled with increased adoption of synthetic somatostatin analogues like octreotide.
This study sought to evaluate sex-specific survival difference in GEP-NETs using a retrospective review of 73,521 GEP-NETs from the National Cancer Database from 2004 to 2016. Regardless of tumor primary site, female sex was found to be associated with improved overall survival relative to male sex and this association held after multivariable adjustment for confounding baseline characteristic differences. To investigate the role of differing circulating sex hormone levels and account for longer lifespan of females, patients were further subgrouped based on age, with age ≤45 as a surrogate for “pre-menopause” and age >65 as a surrogate for “post-menopause”. This revealed that female sex appeared more protective against death among younger women versus younger men, than for older women versus older men for gastric and small bowel NETs, and for all NETs analyzed together. In the younger age group, not all subgroups showed significantly different survival by sex, possibly due to limited power resulting from much smaller patient numbers among patients ≤45 years old, yet in all NETs combined and in the large GNET and SBNET subgroups, women had significantly better survival. While survival differences by sex, particularly in the older age group, could be due to inherent longer median life span of females, the fact that the survival discrepancies persist and are more pronounced in the younger group (HR for OS among all NETs 0.74 in younger versus 0.88 in older females) argues that sex-based, and possibly hormone-based tumor biologic differences exist. The larger protective effect of sex in the ≤45 age group further suggests that baseline differences in mortality between genders does not fully explain this difference. This supports the hypothesis that increased levels of circulating sex hormones could contribute to improved GEP-NET survival in female patients.
This study is limited by inability to assess tumor grade/Ki-67 index as these data are not available in NCDB. Information on grade is included in NCDB, but the reported grade is not necessarily based on updated WHO guidelines, and in many cases likely includes only histologic differentiation. While the multivariable analysis model in this study included staging and differentiation, it remains unknown whether differences existed in tumor Ki-67 index or mitotic rate between male and female study population were present that could account for survival differences. Another limitation is that NCDB captures only overall survival. Cancer-specific survival or time to recurrence cannot be assessed, and thus whether noncancer mortality difference between males and females in the study group accounts for differences in overall survival cannot be determined. Lastly, NCDB records only age at the time of diagnosis. With a median follow up period of 49.5 to 63.9 months depending on tumor location, some patients over the course of the study may cross over from premenopausal to perimenopausal or post-menopausal and this confounding factor is unable to be analyzed.
These results agree with previous studies which demonstrated that for GEP-NETs, male sex was a negative prognostic factor.[2, 37, 38] More recently, a study by Greenberg et al. investigated PNETs using the NCDB and found that men and women had differences in OS for early stage disease rather than late stage disease, further supported by our results.[39] This, however, is one of few to evaluate for survival differences in the context of multiple different tumor primary locations and the first to incorporate age as a surrogate for sex hormone levels. Given the heterogeneous nature of neuroendocrine tumor biology, it is important to understand how the impact of sex on OS could differ based on primary location.
Estrogen receptor and progesterone receptor are well described members of the nuclear receptor family of intracellular transcription factors. Once activated by steroid hormone binding, they translocate into the nucleus and, as a complex, function as a DNA-binding transcription factor.[40, 41] Given the past efficacy of tamoxifen, a selective estrogen receptor modulator, in treatment of neuroendocrine tumors, it is possible that differing levels of circulating sex hormones have an effect on overall survival and patient outcomes, however the mechanism remains unknown. While in breast cancer pathophysiology, tamoxifen is best known as an ER alpha antagonist, its ability to act as an ER agonist in other organ systems could have other effects. One potential mechanism could be differing roles and expression levels of ER subtypes, ER-alpha and ER-beta.[42] In breast and prostate cancer, ER-alpha is proliferative and tumorigenic, while ER-beta is antiproliferative, anti-inflammatory, and potentially tumor suppressive. ER-alpha and beta expression has been studied in PNETs but not other GEP-NETs. In an investigation of 131 low and intermediate grade PNETs, Estrella et al. were unable to identify any ER-alpha expression, however 96% of PNETs in their cohort expressed ER-beta to differing levels.[6] A second study by Zimmerman et al. investigated expression patterns of ER and PR in foregut (including pancreatic), midgut and hindgut NETs as well as metastases.[3] This showed that PR was positive in 32% of primary tumors and ER was positive in 27% of primary tumors, with sex hormone receptor positive tumors more commonly occurring in females than males.[3] In their cohort, a small number of PNETs (7%) had ER-alpha expression.[3] There was no difference between primary tumors and metastases. These hormone receptor positive tumors presented with a more favorable disease course.[3] A retrospective study analyzed PR and phosphatase and tensin homologue (PTEN) expression in 160 low and intermediate grade pancreatic neuroendocrine tumors, finding that patients with PR-negative and PTEN-low tumors had shorter overall survival and metastasis-free survival.[43] PR expression has been found to incrementally decrease during the progression from normal islets to neuroendocrine microadenoma and finally to pancreatic neuroendocrine tumors, suggesting that loss of PR may play a role in tumorigenesis.[44] Strong PR expression has been associated with lower disease stage and more favorable clinical prognosis.[45] In addition to specific regulation of oncogenes and tumor suppressor genes, differences in sex hormone receptor expression have been suggested to play a role in tumor chemotherapy toxicity and thus efficacy.[32]
Alternatively, there may be a hormonal reason to why males do worse rather than why females do better, especially given that this difference is more prominent in the younger age group. Most of the knowledge of androgen receptor (AR) functions in cancer have been derived from studies in prostate cancer. In prostate cancer, high expression of androgen receptor leads to development of neuroendocrine “features” which are associated with resistance to androgen deprivation therapy and poor prognosis.[46] More recently, a few studies have sought to investigate AR in GEP-NETs, however results are conflicting. Immunohistochemistry performed on a tissue microarray of 120 pancreas, ileal and rectal NETs which did not identify any AR expression, however all tumors included in this study were well-differentiated.[47] Blazevic et al. in 2022 characterized androgen receptor (AR) expression in SBNETs and found that SBNETs had increased expression of AR in comparison to healthy intestine and that men were more likely to have mesenteric metastasis.[48] If AR expression is associated with detrimental outcomes, the fact that testosterone levels decrease with age could contribute to less pronounced sex-based survival differences in the older population.[49]
One recent study used open source data from the American Association for Cancer Research and Genomics Evidence Neoplasia Information Exchange (AACR GENIE) to investigate whether mutational signatures differ between males and females with PNETs.[39] They identified that mutations in Multiple Endocrine Neoplasia Type 1 (MEN1) and Death Domain Associated Protein (DAXX) more frequently occurred in males with PNETs, while mutations in TP53 were more common in female patients, however, this difference did not persist when accounting for multiple testing.[39] DAXX, MEN1 and TP53 mutations have been found in poorly differentiated tumors, and are associated with late stage disease and worse OS. This represents interesting preliminary data further suggesting different biology in these tumors.
These results support further investigation to elucidating the role of ER/PR and downstream transcription factors to explain the biochemical basis of these sex-specific survival differences, with the hope that this research could lead to novel therapeutic targets.[50, 51] HORMONET (Study of Tamoxifen in Well Differentiated Neuroendocrine tumors and Hormone Receptor Positive Expression) is a current ongoing phase II clinical trial to evaluate the effect of tamoxifen in NETs.[52] Identifying mechanisms at work in antihormonal therapy in NETs could further our understanding of the role of ER/PR in NET tumorigenesis.
5. Conclusion
Understanding sex-based differences in cancer has the potential to provide additional avenues for personalized cancer therapy. Regardless of primary site, female sex was associated with improved overall survival and this effect was more pronounced in younger patients for gastric and small bowel primary tumors. Further investigation is necessary to understand the biochemical and genetic basis for this survival difference.
6. Disclosure / Acknowledgements
The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article. This work was supported by NIH grants CA148062-01 to CGT, P50 CA174521-01 to PHE and JRH.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Katz LH, et al. , Risk factors associated with gastroenteropancreatic neuroendocrine tumors in a cohort of 2.3 million Israeli adolescents. Int J Cancer, 2018. 143(8): p. 1876–1883. [DOI] [PubMed] [Google Scholar]
- 2.Abdel-Rahman O and Fazio N, Sex-Based Differences in Prognosis of Patients With Gastroenteropancreatic-Neuroendocrine Neoplasms: A Population-Based Study. Pancreas, 2021. 50(5): p. 727–731. [DOI] [PubMed] [Google Scholar]
- 3.Zimmermann N KP, Keck T, Billmann F, Schmid S, Brabant G, Thorns C, Expression Pattern of CDX2, Estrogen and Progesterone Receptors in Primary Gastroenteropancreatic Neuroendocrine Tumors and Metastases. Anticancer Res, 2016. 36(3): p. 921–4. [PubMed] [Google Scholar]
- 4.Satyaswaroop PG ZR, Mortel R, Estrogen-like effects of tamoxifen on human endometrial carcinoma transplanted into nude mice. Cancer Res, 1984. 44(9): p. 4006–10. [PubMed] [Google Scholar]
- 5.Viale G DC, Gambacorta M, Zamboni G, Coggi G, Bordi C, Progesterone receptor immunoreactivity in pancreatic endocrine tumors. An immunocytochemical study of 156 neuroendocrine tumors of the pancreas, gastrointestingal and respiratory tracts, and skin. Cancer Res, 1992. 70(9): p. 2268–77. [DOI] [PubMed] [Google Scholar]
- 6.Estrella JS, et al. , Expression of estrogen-induced genes and estrogen receptor beta in pancreatic neuroendocrine tumors: implications for targeted therapy. Pancreas, 2014. 43(7): p. 996–1002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Blazevic A, et al. , Sexual dimorphism in small-intestinal neuroendocrine tumors: lower prevalence of mesenteric disease in premenopausal women. J Clin Endocrinol Metab, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Modlin IM, et al. , A nomogram to assess small-intestinal neuroendocrine tumor (‘carcinoid’) survival. Neuroendocrinology, 2010. 92(3): p. 143–57. [DOI] [PubMed] [Google Scholar]
- 9.Assi HA, et al. , Surgery Versus Surveillance for Well-Differentiated, Nonfunctional Pancreatic Neuroendocrine Tumors: An 11-Year Analysis of the National Cancer Database. Oncologist, 2020. 25(2): p. e276–e283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Senturk M, et al. , Clinicopathological Characteristics of Gastroenteropancreatic Neuroendocrine Tumors: 10 Years of Experience From a Single Center. Pancreas, 2022. 51(2): p. 159–163. [DOI] [PubMed] [Google Scholar]
- 11.Pommergaard HC, et al. , Surgery of the primary tumour in 201 patients with high-grade gastroenteropancreatic neuroendocrine and mixed neuroendocrine-non-neuroendocrine neoplasms. J Neuroendocrinol, 2021. 33(5): p. e12967. [DOI] [PubMed] [Google Scholar]
- 12.Schmocker RK, et al. , Duodenal, ampullary, and pancreatic neuroendocrine tumors: Oncologic outcomes are driven by tumor biology and tissue of origin. J Surg Oncol, 2021. 123(2): p. 416–424. [DOI] [PubMed] [Google Scholar]
- 13.Lewkowicz E, et al. , Gastroenteropancreatic neuroendocrine neoplasms: a 10-year experience of a single center. Pol Arch Med Wewn, 2015. 125(5): p. 337–46. [DOI] [PubMed] [Google Scholar]
- 14.Cope J and Srirajaskanthan R, Rectal Neuroendocrine Neoplasms: Why Is There a Global Variation? Curr Oncol Rep, 2022. 24(3): p. 257–263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Modlin IM, Lye KD, and Kidd M, A 5-decade analysis of 13,715 carcinoid tumors. Cancer, 2003. 97(4): p. 934–59. [DOI] [PubMed] [Google Scholar]
- 16.Srirajaskanthan R, et al. , Optimising Outcomes and Surveillance Strategies of Rectal Neuroendocrine Neoplasms. Cancers (Basel), 2023. 15(10). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Conte B, et al. , High-Grade Neuroendocrine Colorectal Carcinomas: A Retrospective Study of 100 Patients. Clin Colorectal Cancer, 2016. 15(2): p. e1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Starzynska T, et al. , Colorectal neuroendocrine neoplasms - management guidelines (recommended by the Polish Network of Neuroendocrine Tumours). Endokrynol Pol, 2017. 68(2): p. 250–260. [DOI] [PubMed] [Google Scholar]
- 19.Rinke A, et al. , European Neuroendocrine Tumor Society (ENETS) 2023 guidance paper for colorectal neuroendocrine tumours. J Neuroendocrinol, 2023. 35(6): p. e13309. [DOI] [PubMed] [Google Scholar]
- 20.Gamboa AC, et al. , Duodenal neuroendocrine tumors: Somewhere between the pancreas and small bowel? J Surg Oncol, 2019. 120(8): p. 1293–1301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Tran CG, et al. , Management of Duodenal Neuroendocrine Tumors: Surgical versus Endoscopic Mucosal Resection. Ann Surg Oncol, 2022. 29(1): p. 75–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sonbol MB, et al. , Survival and Incidence Patterns of Pancreatic Neuroendocrine Tumors Over the Last 2 Decades: A SEER Database Analysis. Oncologist, 2022. 27(7): p. 573–578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sherman SK, Tran CG, and Howe JR, ASO Author Reflections: Indolent Growth and Small Bowel Neuroendocrine Tumor Management. Ann Surg Oncol, 2021. 28(5): p. 2752–2753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Boffa DJ, et al. , Using the National Cancer Database for Outcomes Research: A Review. JAMA Oncol, 2017. 3(12): p. 1722–1728. [DOI] [PubMed] [Google Scholar]
- 25.Kaplan EL MP, Nonparametric Estimation from Incomplete Observations. J Am Stat Assoc, 1958. 53: p. 457–81. [Google Scholar]
- 26.DR C, Regression Models and Life-Tables. J R Statist Soc B., 1972. 34: p. 187–220. [Google Scholar]
- 27.Tran CG, Sherman SK, and Howe JR, The Landmark Series: Management of Small Bowel Neuroendocrine Tumors. Ann Surg Oncol, 2021. 28(5): p. 2741–2751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Dasari A, et al. , Trends in the Incidence, Prevalence, and Survival Outcomes in Patients With Neuroendocrine Tumors in the United States. JAMA Oncol, 2017. 3(10): p. 1335–1342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Garcia-Carbonero R, et al. , Incidence, patterns of care and prognostic factors for outcome of gastroenteropancreatic neuroendocrine tumors (GEP-NETs): results from the National Cancer Registry of Spain (RGETNE). Ann Oncol, 2010. 21(9): p. 1794–1803. [DOI] [PubMed] [Google Scholar]
- 30.Howe JR, Carcinoid Tumors: Past, Present, and Future. Indian J Surg Oncol, 2020. 11(2): p. 182–187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Wagner AD, et al. , Gender medicine and oncology: report and consensus of an ESMO workshop. Ann Oncol, 2019. 30(12): p. 1914–1924. [DOI] [PubMed] [Google Scholar]
- 32.Kim HI, Lim H, and Moon A, Sex Differences in Cancer: Epidemiology, Genetics and Therapy. Biomol Ther (Seoul), 2018. 26(4): p. 335–342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Keshgegian AA WJ, Estrogen receptor protein in malignant carcinoid tumor: a report of 2 cases. Cancer, 1980. 45(2): p. 293–6. [DOI] [PubMed] [Google Scholar]
- 34.Stathopoulos GP KG, Yiotis J, Tamoxifen in carcinoid syndrome. NEJM, 1981. 305(1): p. 52. [DOI] [PubMed] [Google Scholar]
- 35.Myers CF EW, Tannenbaum MA, Barth R, Tamoxifen and Carcinoid Tumor. Ann Intern Med, 1982. 96(3): p. 383. [DOI] [PubMed] [Google Scholar]
- 36.Arganini M CG, Miccoli P, Long term treatment with tamoxifen for metastatic carcinoid tumor. Acta Chir Belg, 1989. 89(4): p. 209–11. [PubMed] [Google Scholar]
- 37.Xu Z, et al. , Epidemiologic Trends of and Factors Associated With Overall Survival for Patients With Gastroenteropancreatic Neuroendocrine Tumors in the United States. JAMA Netw Open, 2021. 4(9): p. e2124750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.White BE, et al. , Sex Differences in Survival from Neuroendocrine Neoplasia in England 2012–2018: A Retrospective, Population-Based Study. Cancers (Basel), 2023. 15(6). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Greenberg JA, et al. , Sex-Based Clinicopathologic and Survival Differences Among Patients with Pancreatic Neuroendocrine Tumors. J Gastrointest Surg, 2022. 26(11): p. 2321–2329. [DOI] [PubMed] [Google Scholar]
- 40.Levin ER, Integration of the extranuclear and nuclear actions of estrogen. Mol Endocrinol, 2005. 19(8): p. 1951–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Gadkar-Sable S, et al. , Progesterone receptors: various forms and functions in reproductive tissues. Front Biosci, 2005. 10: p. 2118–30. [DOI] [PubMed] [Google Scholar]
- 42.Risbridger GP, et al. , Breast and prostate cancer: more similar than different. Nat Rev Cancer, 2010. 10(3): p. 205–12. [DOI] [PubMed] [Google Scholar]
- 43.Estrella JS, et al. , Progesterone receptor and PTEN expression predict survival in patients with low- and intermediate-grade pancreatic neuroendocrine tumors. Arch Pathol Lab Med, 2014. 138(8): p. 1027–36. [DOI] [PubMed] [Google Scholar]
- 44.Kim SJ, et al. , Loss of Progesterone Receptor Expression Is an Early Tumorigenesis Event Associated with Tumor Progression and Shorter Survival in Pancreatic Neuroendocrine Tumor Patients. J Pathol Transl Med, 2017. 51(4): p. 388–395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Arnason T, et al. , Immunohistochemical expression and prognostic value of ER, PR and HER2/neu in pancreatic and small intestinal neuroendocrine tumors. Neuroendocrinology, 2011. 93(4): p. 249–58. [DOI] [PubMed] [Google Scholar]
- 46.Su R, et al. , Comprehensive analysis of androgen receptor status in prostate cancer with neuroendocrine differentiation. Front Oncol, 2022. 12: p. 955166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Wang H, et al. , Heterogeneity in signaling pathways of gastroenteropancreatic neuroendocrine tumors: a critical look at notch signaling pathway. Mod Pathol, 2013. 26(1): p. 139–47. [DOI] [PubMed] [Google Scholar]
- 48.Blazevic A, et al. , Sexual Dimorphism in Small-intestinal Neuroendocrine Tumors: Lower Prevalence of Mesenteric Disease in Premenopausal Women. J Clin Endocrinol Metab, 2022. 107(5): p. e1969–e1975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Anaissie J, et al. , Testosterone deficiency in adults and corresponding treatment patterns across the globe. Transl Androl Urol, 2017. 6(2): p. 183–191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Mafficini A and Scarpa A, Genetics and Epigenetics of Gastroenteropancreatic Neuroendocrine Neoplasms. Endocr Rev, 2019. 40(2): p. 506–536. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Krzyszczyk P, et al. , The growing role of precision and personalized medicine for cancer treatment. Technology (Singap World Sci), 2018. 6(3–4): p. 79–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Barros M FT, de Jesus VHF et al. , HORMONET: Study of tamoxifen in well differentiated neuroendocrine tumors and hormone receptor positive expression. Annals of Oncology, 2019. 30. [Google Scholar]