The Role of HMGA1 and Height in Breast Cancer Risk and Prognosis: Insights from UK Biobank Data

Abstract

Background/Aim: Tall women are more likely to develop breast cancer (BC). High Mobility Group AT-Hook 1 (HMGA1), an oncofetal protein, plays a role in BC progression. Variants near HMGA1 have been associated with increased height. This study examines the relationship between HMGA1, height, and BC risk and prognosis using UK Biobank data. Patients and Methods: Data from 10,527 women with invasive BC were analyzed. Subjects were grouped by height: short (<155 cm), medium (155-175 cm), and tall (>175 cm). HMGA1 SNP rs41269028, a single nucleotide intron variant, was evaluated for its influence on height, BC risk, and survival. Statistical analysis included Fisher’s exact test, regression models, and survival analysis using the log-rank test. Results: HMGA1 SNP rs41269028 carriers (CT+TT) were taller (162.88 cm) compared to homozygotes for the major allele (162.29 cm, p=0.005). Tall women with BC showed poorer survival than short women (p=0.032). However, HMGA1 genotype did not significantly affect BC risk (p=0.602) or survival (p=0.439). Multivariate analysis confirmed an independent effect of age and HMGA1 genotype on height. Conclusion: While HMGA1 influences height, no direct association with increased BC risk or poor prognosis in tall women was demonstrated. Nevertheless, tall women with BC had worse survival, suggesting height might be considered in treatment decisions. Future studies should explore mechanisms linking height to BC outcomes.

Keywords: Breast cancer, height, genetics, risk, survival

Introduction

Tall women are more likely to develop breast cancer (BC) (1). Women who are 176 cm or taller have a 20%-30% higher risk of breast cancer than women who are roughly 155 cm or lower, according to a pooled analysis that included data from 20 prospective cohort studies (2). The growth spurts tall women experienced as children have been postulated to elevate the risk of breast cancer associated with height. Increased hormone levels like IGF-1 or other growth factors can trigger growth spurts (3). Higher hormone levels and rapid cell proliferation during a growth spurt are thought to influence risk of breast cancer in later life.

High Mobility Group AT-Hook 1 (HMGA1), an oncofetal protein, plays a role in the progression of breast cancer (4). HMGA1 establishes an autocrine loop in invasive triple-negative breast cancer (TNBC) cells, which mediates the migration, invasion, and metastasis of TNBC cells and predicts the onset of metastasis in these patients. Hawkes et al. performed a whole genome sequencing association analysis for height using 333,100 individuals from three datasets: UK Biobank, TOPMed and All of Us. They identified non-coding sequences proximal to HMGA1 containing variants associated with a 4.83 cm taller height (5). In the current study, we used UK Biobank data to examine the relationship of HMGA1 to height, risk, and prognosis of women with breast cancer.

Patients and Methods

Patients. The UK Biobank is a large prospective observational study of men and women with no link to MedWatch. Participants were recruited from across 22 centers located throughout England, Wales, and Scotland between 2006 and 2010 and continue to be longitudinally followed for capture of subsequent health events (6). This methodology is like that of the ongoing Framingham Heart Study (7), with the exception that the UKB program collects postmortem samples, which Framingham did not.

UK Biobank has approval from the Northwest Multi-center Research Ethics Committee (MREC) to obtain and disseminate data and samples from the participants, and these ethical regulations cover the work in this study. Written informed consent was obtained from all participants. Details can be found at www.ukbiobank.ac.uk/ethics.

Our UK Biobank application was approved as UKB project 57245 (S.L., P.H.R.). Our analysis included all subjects with invasive BC that occurred either before or after participant enrollment and was recorded in the UK Biobank database using self-reported data and the International Classification of Diseases (ICD10, ICD9).

Methods. We divided the subjects into three previously described height groups (1): Short (<155 cm), Medium (155 cm to 175 cm), Tall (>175 cm). We analyzed the HMGA1 SNP rs41269028, a single nucleotide intron variant, C>T, minor allele frequency 0.044. SNP rs41269028 was previously evaluated in subjects with diabetes (8,9).

Statistical analysis. Descriptive statistics. Mean and standard deviations of height were calculated for different genotype groups (homozygous major allele CC vs. minor allele carriers CT+TT). A two-tailed t-test was used to assess the significance of height differences between genotype groups.

Logistic regression analysis. To evaluate the association between BC risk and independent variables (age, menopause status, height group, and HMGA1 genotype), logistic regression models were applied. Odds ratios (ORs) with 95% confidence intervals were calculated for each variable to determine its contribution to BC risk.

Survival analysis. Survival outcomes for BC patients were assessed using Kaplan-Meier curves. Survival differences between height groups (short, medium, tall) and HMGA1 genotypes (CC vs. CT or TT) were evaluated with the log-rank test. Statistical significance for survival outcomes was set at p<0.05.

Multivariate linear regression. To correct for potential confounding effects of age on height (as aging is associated with height loss), multivariate linear regression was conducted. The dependent variable was height, while independent variables included HMGA1 genotype and age. Regression coefficients (ß) and significance values were reported to determine the independent effect of genotype and age on height.

Software used. Statistical analysis was performed using SPSS (version 26, IBM, Armonk, NY, USA), and figures illustrating survival and genotype distributions were generated to visualize the results (10).

Results

Data from 273,378 women, of which 10,527 were invasive breast cancer cases, was analyzed. Breast cancer patients were aged 60±7 [mean±standard deviation (SD)]. 95% of subjects were white British. Table I shows the HMGA1 SNP rs41269028 genotype versus height group in 8,327 post-menopausal breast cancer cases. A greater proportion of tall women with breast cancer (11.9%) than short women (6.8%) were carriers or homozygotes (CT+TT) of the minor allele T (p=0.02, two tail Fisher exact test). No significant effect was present in pre-menopausal women (p=0.441).

The height of 9583 women with BC homozygous for the HMGA1 SNP rs41269028 major allele (CC) was 162.29 cm±6.18. The height of 944 women with BC who were carriers or homozygotes (CT+TT) of the minor allele T was 162.88 cm±6.001. This difference was significant (p=0.005). Figure 1 illustrates 8,327 post-menopausal breast cancer patients stratified by height group and the HMGA1 SNP rs41269028 genotype (CC versus CT or TT).

Aging is linked to a reduction in height over time (11). To account for this effect, multivariate linear regression was conducted using breast cancer cases. In this analysis, height group was the dependent variable, while HMGA1 SNP rs41269028 genotype and age were included as independent variables. The effect of genotype on height groups was significant (β=0.040, p=0.002) and independent of the effect of age (β=-0.005, p<0.001). In other words, carriers or homozygotes of the minor allele T were taller than homozygotes for the major allele (CC); while older women were shorter than younger women.

Figure 2 illustrates the survival of breast cancer subjects stratified by HMGA1 SNP rs41269028 genotype. The effect of genotype was insignificant (p=0.439, log rank test). Figure 3 illustrates the survival outcomes of breast cancer patients categorized by height group. Height group had a significant impact on survival (p=0.032, log-rank test), with tall women exhibiting the poorest survival rates. Table II presents the results of the logistic regression analysis. The dependent variable is breast cancer status (yes or no), with data from 228,611 women. The independent variables include age, menopausal status, and height group. The risk of breast cancer was increased in post-menopausal women (OR=4.315, p<0.001). The risk increased with each year of age (OR=1.034, p<0.001). Short women were at decreased risk compared to tall women (OR=0.819, p=0.026). Women of medium height were at decreased risk that was not significant compared to tall women (OR=0.929, p=0.391). HMGA1 SNP rs41269028 had no significant relationship to BC risk (p=0.602).

Discussion

Most of the genetic variation associated with complex traits, such as height, is located in non-coding regions of the genome. Since 99% of the human genome is non-coding, most inherited genetic variations are rare and occur within these regions. Identifying these rare non-coding variations associated with common traits and diseases can reveal new regulatory gene pathways and significantly enhance our understanding of human biology and disease mechanisms (5).

HMGA1 SNP rs41269028 is an intron variant in a non-coding region of the genome, chromosome 6. Rare variants such as those of HMGA1 are said to confer most of the heredity for height, about 79% (12). In other words, in a large group of people, 79% of height differences are genetic (13). HMGA1 is a protein that has been found to play a role in the progression of breast cancer. One study suggests that HMGA1 establishes an autocrine loop in invasive triple-negative breast cancer (TNBC) cells, which mediates the migration, invasion, and metastasis of TNBC cells and predicts the onset of metastasis in these patients (14).

HMGA1 has been reported to promote breast cancer angiogenesis by supporting the stability, nuclear localization, and transcriptional activity of FOXM1 (15). FOXM1 is an oncogenic transcription factor that is greatly upregulated in breast cancer and many other cancers where it promotes tumorigenesis, cancer growth and progression (16). It is expressed in all subtypes of breast cancer and is the factor most associated with risk of poor patient survival, especially in TNBC.

Study limitations. We did not have any data on tumor size, histology, grade, or hormone receptor status. We did not have the recently released UKBB whole genome sequence data for HMGA1 that Hawkes et al. used (5,17). Instead, we evaluated imputed genotypes from the UKB data field 22828 (18). We found that the HMGA1 SNP rs41269028 minor allele T carriers (CT) and homozygotes (TT) were significantly taller, but the effect size was small (0.59 cm) compared to the Hawkes et al. report (4.83 cm) (5).

Conclusion

We conclude that HMGA1 influences height, but we were unable to demonstrate that HMGA1 is related to increased incidence or poor prognosis of tall women with breast cancer. Our finding that tall women have a worse prognosis is important because it could help the oncologist decide, along with other prognostic factors, whether adjuvant therapy is warranted.

Data Availability

Data sources described in the article are publicly available or can be accessed after an approved application to the UK Biobank.

Conflicts of Interest

None.

Authors’ Contributions

SL and PHR contributed equally to the conception, writing, and data analysis of this study.

Acknowledgements

This work was supported in part through the computational and data resources and staff expertise provided by Scientific Computing and Data at the Icahn School of Medicine at Mount Sinai and the Clinical and Translational Science Awards (CTSA) grant UL1TR004419 from the National Center for Advancing Translational Sciences.