Abstract
Background/Aim: Turner syndrome confers increased cancer susceptibility; however, large-scale epidemiological evidence is lacking. This study aimed to analyze the incidence and prevalence of various malignancies in patients with Turner syndrome over 20 years of age to inform screening strategies. Patients and Methods: We performed a retrospective cohort analysis of 11,502 patients with Turner syndrome from 2000 to 2020 utilizing the TriNetX research network database. The outcomes encompassed the incidence and prevalence of 20 cancers. Stratified analyses were used to evaluate variations in age, sex, and race. Results: Key findings demonstrated markedly elevated risks of breast (1.7%), colon (1.0%), renal (0.4%), gonadoblastoma (0.4%), and other cancers. Significant demographic variations were observed in the incidence of cancers, such as gonadoblastoma, renal, and colon cancer. Conclusion: This large real-world study offers novel insights into the spectrum of cancer risk across adulthood in Turner syndrome. Our findings elucidate Turner syndrome’s complex cancer phenotype to inform clinical decision-making, prognostication, and tailored screening strategies to ultimately advance patient care.
Turner syndrome is a genetic disorder occurring in women characterized by partial or complete X chromosome monosomy, affecting approximately 1 in 2,000 to 2,500 live-born girls (1, 2). Loss of X chromosome material leads to a variety of clinical manifestations, including short stature, ovarian insufficiency, cardiovascular defects, autoimmune thyroid disease, hearing impairment, neurocognitive deficits, and distinctive physical features (1, 3, 4). Additionally, §Turner syndrome confers substantially elevated risks across a spectrum of malignant and benign neoplasms compared to the general population (5-7). The underlying mechanisms driving tumor susceptibility remain unclear, although hypothesized factors include abnormal gonadal development and hormonal perturbations (4).
Current clinical guidelines recommend surveillance for certain malignancies in Turner syndrome, but evidence on lifetime cancer risks and optimal screening strategies is limited (8, 9). Most existing studies have been confined to small cohorts with short follow-up periods (10). Large-scale epidemiological analyses with long-term tracking of outcomes are needed to delineate cancer patterns across the lifespan and inform prevention efforts (11).
Therefore, we performed a retrospective observational cohort study using real-world data from over 10,000 patients with Turner syndrome from the TriNetX research network. We analyzed the 20-year incidence and prevalence of a range of cancers, including gonadoblastoma, ovarian, breast, renal, colon, endometrial, liver, melanoma, and other skin tumors. Stratification by age, sex, and race allowed assessment of demographic risk factors. This large cohort study provides novel insights into the spectrum of malignancy risks from childhood to adulthood in Turner syndrome. Elucidating lifetime incidence rates across tumor types is imperative to guide clinical decision-making and surveillance, improve prognostication, and enhance patient quality of life. Our findings aimed to elucidate Turner syndrome’s complex cancer phenotype to tailor screening and advance patient care.
Patients and Methods
We performed a retrospective observational cohort study using data from the TriNetX global federated research database. This database contains real-time electronic medical records for approximately 130 million patients across 110 healthcare organizations, primarily in the United States, with additional data from centers in Australia, Germany, the UK, Italy, Singapore, Israel, and Taiwan. Patient data within TriNetX were anonymized to protect privacy (12).
We identified all patients diagnosed with Turner syndrome through July 25, 2023 using the International Classification of Diseases, Tenth Revision (ICD-10) code Q96. We included patients based on documented phenotypic subtype, defined by the ICD-10 code Q96 combined with codes indicating Polycystic ovarian syndrome (E28.2), Short stature due to endocrine disorder (E34.3), Delayed puberty (E30.0), Hypothyroidism, unspecified (E03.9), Specific reading disorder (F81.0), Developmental disorder of scholastic skills, unspecified (F81.9), Esotropia (H50.0), Chronic serous otitis media (H65.2), Unspecified sensorineural hearing loss (H90.5), Lymphedema, not elsewhere classified (I89.0), Other specified acquired deformities of limbs (M21.8), Age-related osteoporosis without current pathological fracture (M81.0), Primary amenorrhea (N91.0), Female infertility associated with anovulation (N97.0), Misplaced ear (Q17.4), Other specified congenital malformations of face and neck (Q18.8), Lobulated, fused and horseshoe kidney (Q63.1), Other congenital deformities of chest (Q67.8), Congenital malformation of kidney, unspecified (Q63.9), Hereditary lymphedema (Q82.0), Type 2 Diabetes Mellitus (E11), Vitamin D Deficiency (E55), Hyperlipidemia, Unspecified (E78.5), Other Hyperlipidemia (E78.4), Overweight and Obesity (E66), Other Hypothyroidism (E03), Congenital Malformation Syndromes Predominantly Associated with Short Stature (Q87.1), Congenital Malformation Syndromes Predominantly Involving Limbs (Q87.2), Other Specified Congenital Malformation Syndromes, Not Elsewhere Classified (Q87.8), Other Congenital Musculoskeletal Deformities (Q68), Indeterminate Sex and Pseudohermaphroditism (Q56), Other Congenital Malformations of Heart (Q24), Congenital Malformations of Ear Causing Impairment of Hearing (Q16), Hypertensive Diseases (I10-I16), Ischemic Heart Diseases (I20-I25), Celiac Disease (K90.0), Noninfective Enteritis and Colitis (K50-K52), Liver Disease, Unspecified (K76.9), Other Disorders of Kidney and Ureter, Not Elsewhere Classified (N28), Other Symptoms and Signs Involving Cognitive Functions and Awareness (R41), Osteoporosis Without Current Pathological Fracture (M81), Vitiligo (L80), Other Disorders of Binocular Movement (H51) or Amblyopia ex Anopsia (H53.0). Searches were conducted from July 1 to September 10, 2023. During the search, 110 organizations contributed to the data. As TriNetX provides only de-identified data, this study was exempt from ethical approval. TriNetX and contributing healthcare networks maintain compliance with Health Insurance Portability and Accountability Act (HIPAA) to ensure data privacy/security (12, 13). This study was conducted in accordance with the Declaration of Helsinki ethical principles for human subject medical research.
Study outcomes. We aimed to assess long-term cancer outcomes in patients over 20 years of age with Turner syndrome by analyzing data from 2000 to 2020. The outcomes included incidence and prevalence of Gonadoblastoma (D39), Ovarian Cancer (C56), Breast Cancer (C50), Renal Cell Carcinoma (C64), Colon Cancer (C18), Endometrial Cancer (C54), Hepatocellular Carcinoma (C22), Neoplasms of unspecified behavior (D49), Melanocytic nevi (D22), Other benign neoplasms of skin (D23), Hemangioma and lymphangioma (D18), Benign neoplasm of colon, rectum, anus and anal canal (D12), Leiomyoma of uterus (D25), Benign neoplasm of other and unspecified sites (D36), Benign lipomatous neoplasm (D17), Benign neoplasm of meninges (D32), Benign neoplasm of other and unspecified endocrine glands (D35), Neoplasm of uncertain behavior of other and unspecified sites (D48), Other neoplasms of uncertain behavior of lymphoid, hematopoietic and related tissue (D47), Malignant neoplasm without specification of site (C80), Secondary malignant neoplasm of other and unspecified sites (C79), Secondary malignant neoplasm of respiratory and digestive organs (C78), Secondary and unspecified malignant neoplasm of lymph nodes (C77), Malignant melanoma of skin (C43), and Other and unspecified malignant neoplasm of skin (C44). Incidence rates were calculated per 100,000 person-years.
Stratified analyses. To assess cancer risk factors and the impact of patient characteristics on outcomes, stratified analyses of the demographic and clinical variables were performed. Specifically, we separated patients into 5-10 year age groups, with narrower 0-4, 5-9, and 10-14 year increments to closely analyze younger groups at and shortly after initial diagnosis. Male and female patients were analyzed separately to evaluate sex differences. We also categorized the patients into broad racial groups: White, Black/African American, Asian, American Indian/Alaska Native, Native Hawaiian/Pacific Islander, and of unknown race. These categories were broadly defined, given the variability in classifications across sources. These stratified analyses aimed to elucidate how key attributes influence cancer risks/outcomes over time to identify the factors contributing to health disparities and to tailor interventions.
Statistical analyses. Descriptive analyses were performed. Incidence rates were calculated as the number of patients with events during the follow-up period at risk, divided by the total follow-up time in person-years. The rates were reported per 1,000 person-years. Patients with Turner syndrome who developed phenotypes after the index date contributed to person-time.
Results
Baseline demographic characteristics. The study cohort comprised 11,502 patients diagnosed with Turner syndrome (Figure 1). The vast majority (n=10,638; 92%) were women, reflecting the sex-linked nature of this condition. A small proportion of patients (n=804; 7%) were male, likely reflecting instances of partial X chromosome loss or mosaicism. The race was predominantly White (n=7,736; 67%), with fewer Black or African Americans (n=982; 9%), Asians (n=321; 3%), American Indians/Alaska Natives (n=62; 1%), and of native Hawaiian/Pacific (n=19; <1%) descent. Ethnicity could not be determined for 2,382 (n=21%) patients. These findings reveal ethnic variations in Turner syndrome incidence, which warrants further investigation (Table I).
Flow diagram of Turner syndrome patients’ selection from TriNetX database.
Baseline characteristics of patients with Turner syndrome.
Clinical characteristics. The mean age at diagnosis was 27.8±19.9 years, indicating that many patients were not identified until early adulthood, highlighting the need for improved awareness and screening. The average body mass index was 25.9±8.1 kg/m2, in the overweight range. This aligns with the existing evidence that Turner syndrome increases the risk of obesity, likely due to hormonal and metabolic factors. Cardiovascular parameters were within normal limits, with mean systolic blood pressure of 118±18.5 mmHg, diastolic blood pressure of 71.9±12.8 mmHg, total cholesterol of 182±42.9 mg/dl, LDL of 104±42.2 mg/dl, HDL of 55.5±18.4 mg/dl, and triglycerides of 128±103 mg/dl. Metabolic parameters were also largely normal, including glucose 103±41.8 mg/dl, calcium 9.36±0.621 mg/dl, and kidney function (creatinine 0.985±6.52 mg/dl). Liver enzyme levels were within the expected range. Hematologic parameters revealed a mean hemoglobin of 12.8±1.78 g/dl, slightly below the normal range, indicating a higher prevalence of anemia compared to the general population. White blood cell and platelet counts were within normal limits (Table I).
Incidence and prevalence of cancer outcomes in patients with Turner syndrome were tracked in both 2000-2020 cohort. Our analysis revealed notable disparities in cancer incidence across multiple malignancies among patients with Turner syndrome. Several key findings warrant attention: Gonadoblastoma has an incidence rate of 0.42%, highlighting its rarity despite the need for clinical vigilance. Ovarian cancer accounted for 0.62% of the cases, underscoring its occasional occurrence. Breast cancer showed a significant incidence rate of 1.70%, necessitating dedicated surveillance strategies. Renal cell carcinoma occurred at a lower but noteworthy rate of 0.41%. Colon cancer incidence reached approximately 1.00%, signifying the importance of rigorous screening. Endometrial and hepatocellular cancers exhibited relatively low incidences of 0.26% and 0.34%, respectively, but still require monitoring given their morbidity. Neoplasms of an unspecified nature were substantially higher at 3.58%, raising questions about cancer categorization specificity in this population. In contrast, melanocytic nevi and other benign skin tumors showed incidences of 5.70% and 5.28%, respectively (Figure 2).
Incidence of cancer outcomes in Turner syndrome during the 2000-2020 follow-up period.
Prevalence rates further delineate long-term cancer risks. Gonadoblastoma and ovarian cancer maintained nearly equivalent prevalence to incidence, while breast cancer increased slightly to 1.73%, indicating not only incident cases but also survival. Renal cell carcinoma and colon cancer showed steady prevalence rates of 0.43% and 1.00%, respectively. Endometrial and hepatocellular cancers persisted at low prevalence rates of 0.27% and 0.34%, respectively, stressing the need for ongoing vigilance. Neoplasms of unspecified origin, melanocytic nevi, and benign skin neoplasms remained highly prevalent at 3.67%, 5.73%, and 5.56%, meriting further investigation (Figure 3).
Prevalence of cancer outcomes in Turner syndrome during the 2000-2020 follow-up period.
Incidence rates of the 2000-2020 cohort’s follow-up in patients with Turner syndrome. When analyzing incidence rates per 1,000 person-days, notable differences emerged across cancer types. Gonadoblastoma occurred at a rate of 0.14 cases per 1,000 person-days, indicating its rarity. Ovarian cancer followed closely with 0.21 cases per 1,000 person-days. Breast cancer showed a higher rate of 0.57 cases per 1,000 person-days, aligning with its significant incidence. Renal cell carcinoma was reported at 0.14 cases per 1,000 person-days, whereas colon cancer occurred at 0.33 cases per 1,000 person-days. Endometrial and hepatocellular cancers exhibited lower rates of 0.09 and 0.11 cases per 1,000 person-days, respectively. The incidence rate of neoplasms of unspecified origin stood much higher at 1.21 cases per 1,000 person-days. Benign tumors also had elevated rates, with melanocytic nevi in 1.93 cases and other benign skin neoplasms in 1.83 cases per 1,000 person-days (Figure 4). The relatively high incidence rates of breast cancer, neoplasms of unspecified behavior, and benign skin tumors highlight the need for preventative measures and surveillance across cancer types in Turner syndrome.
Incidence rate of cancer outcomes in Turner syndrome during the 2000-2020 follow-up period.
Stratified analyses. Gonadoblastoma showed the highest incidence in patients aged 30-34 years (0.02) and 40-44 years (0.02). Incidence was markedly higher in men (0.01) compared to women (0.004). Patients of Black/African American race had the highest incidence (0.01). For ovarian cancer, the incidence peaked at ages 25-29 years (0.02) and 30-34 years (0.02). Only women were affected. One American Indian/Alaska Native patient had a very high incidence (0.17), likely reflecting the small sample size of this group. Breast cancer incidence rose steadily from age 30 onwards, reaching 0.23 at ages 65-69 years. Women were predominantly affected, with a peak incidence of 0.017 versus 0.012 in men. Incidence was similar across racial groups. Renal cell carcinoma was rare before age 30 but increased markedly in older groups, peaking at ages 50-54 years (0.05) and 55-59 years (0.08). Incidence was higher in men than women across all age groups. Racial variation was minimal. For colon cancer, the incidence was low until the age of 35 to39 years and then increased steadily. Incidence peaked at 0.13 for ages 60 to 64 years. Men had a slightly higher incidence than women. Asian patients showed an elevated incidence of 0.03. Endometrial cancer incidence only affected women, with a peak of 0.05 at ages 50-54 years. One Asian patient had a higher incidence rate. Hepatocellular carcinoma was rare in women (peak 0.002) but reached 0.012 in men aged 25-79 years. Racial variation was minimal.
Overall, these results clearly demonstrate that age, sex, and race influence cancer risk and should inform screening practices for patients with Turner syndrome. Several malignancies showed marked disparities among the demographic groups.
Discussion
This large cohort study elucidated the spectrum of cancer risks across the lifespan in patients with Turner syndrome. Several key findings of this study have important clinical and research implications.
First, our results validate established guidelines recommending surveillance for certain malignancies, such as breast cancer and gonadoblastoma in patients with Turner syndrome. The 1.7% incidence of breast cancer underscores the need for regular mammography starting in early adulthood. Although rare, the 0.4% incidence of gonadoblastoma warrants continued vigilance and consideration of prophylactic gonadectomy, balancing this against potential impacts on bone health and the timing of procedures.
Second, our study revealed additional cancer types, such as renal, colon, and endometrial malignancies that may merit more dedicated screening efforts in this population than currently advised. Though incidence proportions were under 1%, the steady increase in rates from middle age onwards suggests value in initiating earlier routine monitoring. Colonoscopy every 5-10 years from the age of 35 to 40 years can promote early detection and prevention. Kidney imaging and urinalysis should also be considered in routine care.
Third, the disproportionately high incidence of neoplasms of unspecified origin highlights the need for more precise cancer phenotyping and categorization in Turner syndrome. Uncertainty regarding tumor histology and biomarkers may complicate prognostication and treatment decisions. Standardized diagnostic protocols should be developed to optimize categorization accuracy.
Fourth, the elevated incidence of benign skin tumors reinforces existing evidence that patients with Turner syndrome have increased dermatologic manifestations. The high rates of melanocytic nevi and other benign growths warrant counseling on sun protection and self-skin checks starting in childhood. Dermatological care should be integrated into comprehensive healthcare.
Fifth, our stratified analyses clearly demonstrated that demographic factors, such as age, sex, and race influence cancer risk, informing screening priorities. For example, older patients, men, and certain racial groups showed heightened incidence of cancers, such as gonadoblastoma and renal cell carcinoma. Overall, our findings elucidate Turner syndrome’s multifaceted cancer phenotype to guide personalized screening and surveillance.
Study limitations. The retrospective observational design was subject to inherent biases compared to prospective trials. The reliance on diagnoses coded in electronic health records may underestimate cancer incidence if documentation is incomplete. We lacked detailed information on potential confounders, such as correlation between hormone therapy and cancer incidence rates (14-16). The predominant US-based population may not be generalizable internationally. However, the key strengths of the study include the large sample size, enabling the analysis of even rare cancers, and the real-world data source, providing insights into actual clinical outcomes. The 20-year timeframe allowed assessment of long-term risks across the lifespan. The granular phenotypic data enabled narrow definitions of Turner syndrome. Furthermore, the stratified analyses elucidated the impact of age, sex, and race on cancer susceptibility. Overall, this study significantly advances the evidence based on malignancy risks in Turner syndrome despite some limitations.
Gravholt et al., (1, 4) and Schoemaker et al. (6) have provided important foundations for understanding the increased risk of various diseases, including certain cancers, in patients with Turner syndrome. These previous studies identified elevated risks for cancers, such as breast and ovarian cancer, which have since become a focus for clinical monitoring in this population (7, 17, 18). In contrast, our 20-year observational cohort study expands on these traditional risk markers by identifying novel cancer types with high incidences that have been underreported in the literature (19). Overall, our long-term cohort analysis builds on seminal research to advance knowledge of cancer susceptibility in Turner syndrome beyond conventional disease endpoints of interest.
Future research should explore the mechanisms underlying tumor susceptibility, such as gonadal hormone deficits, genomic instability, and abnormalities in cell cycle regulation and DNA repair. Elucidating genetic and biological pathways driving cancer development will enable the development of targeted prevention strategies. Long-term tracking of outcomes using large databases could further clarify risks across the lifespan and evaluate screening efficacy. Overall, this study significantly advances our understanding of the cancer burden in Turner syndrome, ultimately improving patient care.
Acknowledgements
The Authors are deeply grateful to the Center for Health Data Science, Chung Shan Medical University Hospital for their indispensable assistance with TriNetX.
Footnotes
Authors’ Contributions
YNH, SCC, JCC, KWL, and SLC analyzed the data; YNH, DTB, PHS, and CHW wrote and revised the manuscript; CHW supervised the project at China Medical University Hospital. To guarantee that any concerns about the accuracy or integrity of any portion of this study are properly examined and addressed, all Authors have approved the paper and agreed to be held responsible for all elements of the study.
Conflicts of Interest
All Authors declare no conflicts of interest associated with this study.
Funding
This research was funded by grants from China Medical University Hospital. At China Medical University Hospital, grant numbers DMR-111-063, DMR-112-049, and DMR-112-161 were awarded to CHW. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
- Received September 16, 2023.
- Revision received October 7, 2023.
- Accepted October 9, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.










