Servicios personalizados
Inglés (pdf)
Articulo en XML
Referencias del artículo
Traducción automática
Enviar articulo por email
Citado por SciELO 
Citado por Google 
Similares en
SciELO 
Similares en Google 
Permalink
INTRODUCTION
Breast cancer is one of the three most common types of cancer worldwide, alongside lung and colon cancer (1). It has a high incidence, with approximately one in every eight to ten women developing breast cancer at some point in their lifetime (1). In 2020, the United States alone reported over 276,000 new cases of invasive breast cancer and over 48,000 cases of noninvasive breast cancer, according to data from the National Cancer Foundation (2). Despite numerous studies conducted on breast cancer, its incidence rate continues to rise, making it the leading cause of disease burden among women (3). Additionally, breast cancer is the second leading cause of death among females, underscoring the importance of further studying its risk factors (4). Various factors, including physical activity, hormones, and circulating lipids, have been proven to be associated with the occurrence and progression of breast cancer (5). Recent studies have indicated a potential association between sarcopenia and breast cancer, likely stemming from age-related muscle mass reduction, which is a significant risk factor in the elderly population (6,7).
Sarcopenia is a syndrome characterized by the progressive loss of skeletal muscle volume, strength, and function, which has been demonstrated to be associated with the development of various diseases (8). There is a strong association between sarcopenia and a higher risk of experiencing several detrimental health outcomes, including more severe postoperative complications, lower overall and progression-free survival rates, and extended hospital stays (9). In cancer patients, muscle mass is generally reduced, which is considered a crucial factor in predicting adverse clinical outcomes (10). Moreover, sarcopenia is closely linked to several types of cancer, such as esophageal, gastric, pancreatic, colorectal, and breast cancer (11). These studies highlight a significant correlation between sarcopenia and human diseases, particularly cancer.
Mendelian randomization (MR) is widely recognized as a credible method for elucidating causal relationships between exposures and outcomes, effectively controlling for confounding factors and avoiding reverse causation (12). By capitalizing on the natural randomness of alleles during meiosis, MR analysis can reveal the causal associations between exposures and outcomes. Single nucleotide polymorphisms (SNPs) serve as instrumental variables (IVs) in MR studies, as they are not influenced by potential environmental confounders or disease status (13). However, there remains a lack of MR investigations into the causal effects of sarcopenia-related traits on the incidence and prognosis of breast cancer. Therefore, we employed genetic variants strongly linked to sarcopenia-related traits to estimate the effects on breast cancer using two-sample bidirectional and multivariable MR analyses.
MATERIALS AND METHODS
STUDY DESIGN
Two-sample bidirectional and multivariate MR analyses were employed in this study based on the STROBE-MR statement (12). The MR analysis is based on the following three assumptions: a) assumption 1: the genetic IVs used in the analysis are strongly associated with the sarcopenia-related traits under investigation; b) assumption 2: the genetic IVs are not influenced by confounding factors that could introduce bias into the analysis; and c) Assumption 3: the genetic IVs solely affect the outcome through the exposure being studied and do not operate through other paths (14) (Fig. 1). The Mendelian randomization-Egger (MR-Egger), weighted median, heterogeneity test, pleiotropy test, and Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) methods were performed to assess the reliability of the associations and investigate the potential presence of horizontal pleiotropy in the IVs. Pleiotropy and heterogeneity tests were conducted to assess whether the genetic IVs exhibited pleiotropy or heterogeneity. By detecting and addressing outliers, the MR-PRESSO method helped improve the robustness and accuracy of the causal estimates obtained in the MR analysis (15). To assess the influence of outlying and pleiotropic SNPs on the results, we conducted a leave-one-out sensitivity test. Since body mass index (BMI), waist circumference, and whole-body fat mass were the related confounding factors for sarcopenia-related traits and breast cancer, multivariate MR analysis was conducted to effectively control for possible confounding factors.
DATA SOURCE
Sarcopenia-related traits such as appendicular lean mass, hand grip strength (right and left), and walking pace were used as the exposures in this study. Appendicular lean mass is considered a reliable indicator of muscle mass (16). Appendicular lean mass data were calculated using bioelectrical impedance analysis with a large cohort of European individuals (n = 450,243), while also adjusting for appendicular fat mass and other covariates to account for potential confounding factors (16). Grip strength has been widely recognized as an important indicator of sarcopenia (17). The grip strength data were collected from 461,089 individuals of European descent for right-hand grip strength and 461,026 individuals for left-hand grip strength after adjusting for age and sex (18). Meanwhile, the data on walking pace were obtained from the summary-level statistic, which included 459,915 individuals of European ancestry (19). Genetic association summary statistics for breast cancer risk were obtained from two consortia: the Breast Cancer Association Consortium (BCAC), which consisted of 68 studies, and the Discovery, Biology, and Risk of Inherited Variants in Breast Cancer Consortium (DRIVE) (20). The study included data on the risk and survival time of breast cancer, including breast cancer (cases = 122,977, controls = 105,974), ER+ breast cancer (cases = 69,501, controls = 105,974), ER- breast cancer (cases = 21,468, controls = 105,974), survival time of breast cancer (cases = 2,900, controls = 35,054), ER+ breast cancer (cases = 1,333, controls = 21,726), and ER- breast cancer (cases = 920, controls = 5,961) (20,21). Detailed information about the data sources can be found in supplementary table I (https://www.nutricionhospitalaria.org/anexos/05139-01.pdf).
SELECTION OF GENETIC IVS
The IVs were selected based on rigorous criteria to ensure a strong correlation with sarcopenia-related traits. SNPs that showed a significant association (p-value < 5 × 10-8) with the traits were considered potential IVs. To avoid potential bias caused by linkage disequilibrium (LD), SNPs that might have existing LD were removed. The criteria for removal included a low possibility of LD (R2 < 0.001) and a longer physical distance between the SNPs (≥ 10,000 kb) (22). The criterion of an F value (F = Beta2 / SE2) greater than 10 was used to validate the IVs in this study and avoid potential bias caused by weak instruments. Additionally, the IVs selected for this study underwent a thorough examination on the PhenoScanner website (http://www.phenoscanner.medschl.cam.ac.uk/) to account for any pleiotropic effects. The details of the IVs used in this study are presented in supplementary table II (https://www.nutricionhospitalaria.org/anexos/05139-01.pdf).
STATISTICAL ANALYSIS
All analyses were performed using the R software (Version 4.2.1). All analyses were based on the “TwoSampleMR” (Version 0.5.6) and “MR-PRESSO” (Version 1.0) R packages. The Bonferroni-corrected significance level of p < 0.002 (0.05 / 24) was utilized to avoid bias (23). A p-value between 0.002 and 0.05 was considered a suggestive association. A p-value larger than 0.05 indicated that there was no statistical association between the corresponding exposures and outcomes.
RESULTS
TWO-SAMPLE BIDIRECTIONAL MR ANALYSIS
The IVW results revealed associations between sarcopenia-related traits and breast cancer. Specifically, appendicular lean mass was found to be associated with the occurrence of ER- breast cancer (OR = 0.873, 95 % CI: 0.817-0.933, p = 6.570 × 10-5). Additionally, the grip strength of the right hand showed an effect on the survival time of ER+ breast cancer (OR = 0.463, 95 % CI: 0.242-0.882, p = 0.019), while the grip strength of the left hand was associated with the occurrence of ER- breast cancer (OR = 0.744, 95 % CI: 0.579-0.958, p = 0.022). However, no evidence was observed for other sarcopenia-related traits and breast cancer in the MR analysis. Reverse MR analysis revealed a significant association between the occurrence of breast cancer and lower grip strength in the right hand (OR = 0.990, 95 % CI: 0.980-1.000, p = 0.043). Furthermore, the survival time of ER-negative breast cancer was found to be significantly associated with walking pace (OR = 0.998, 95 % CI: 0.996-1.000, p = 0.026). There were no significant associations of the occurrence and prognosis of breast cancer with other sarcopenia-related traits. Tables I and II, supplementary table III (https://www.nutricionhospitalaria.org/anexos/05139-01.pdf), and figures 2 and 3 present the results of the two-sample bidirectional MR analysis.
Table I. Causal association between sarcopenia-related traits and breast cancer.
95 % CI: 95 % confidence interval; IV: instrumental variables; IVW: inverse-variance weighted; OR: odds ratio; WM: weighted median.
Table II. Causal association between sarcopenia-related traits and survival time of breast cancer.
95 % CI: 95 % confidence interval; IV: instrumental variables; IVW: inverse-variance weighted; OR: odds ratio; WM: weighted median.
Figure 2. Forest plot of two-sample bidirectional Mendelian randomization estimation of the causal association between sarcopenia-related traits and breast cancer (95 % CI: 95 % confidence interval; OR: odds ratio).
MULTIVARIABLE MR ANALYSIS
After conducting multivariable MR analysis to control for BMI, waist circumference, and whole-body fat mass, we found a significant association between appendicular lean mass and the occurrence of ER- breast cancer (OR = 0.878, 95 % CI: 0.810-0.951, p = 0.001). Additionally, grip strength in the right hand was found to be related to the occurrence of breast cancer (OR = 0.740, 95 % CI: 0.581-0.943, p = 0.015), as well as ER- breast cancer (OR = 0.696, 95 % CI: 0.502-0.967, p = 0.031). Similarly, grip strength in the left hand was shown to be connected with the occurrence of breast cancer (OR = 0.695, 95 % CI: 0.547-0.883, p = 0.003), ER+ breast cancer (OR = 0.732, 95 % CI: 0.565-0.948, p = 0.018), and ER- breast cancer (OR = 0.676, 95 % CI: 0.488-0.936, p = 0.018). Furthermore, walking pace had a significant impact on the occurrence of both breast cancer (OR = 0.553, 95 % CI: 0.342-0.895, p = 0.016) and ER- breast cancer (OR = 0.491, 95 % CI: 0.250-0.965, p = 0.039). The results of the multivariable MR analysis are presented in table III.
Table III. Influence of sarcopenia-related traits on breast cancer and survival time of breast cancer after regulating BMI, waist circumference, and whole body fat mass by multivariate Mendelian randomization analysis.
95 % CI: 95 % confidence interval; IV: instrumental variables; IVW: inverse-variance weighted; OR: odds ratio; WM: weighted median.
SENSITIVITY ANALYSIS
To evaluate the credibility of the aforementioned findings, we conducted thorough sensitivity analyses. These entailed employing the heterogeneity test, pleiotropy test, MR-PRESSO test, and F statistics (Supplementary tables IV and V: https://www.nutricionhospitalaria.org/anexos/05139-01.pdf). The F values of all chosen IVs were higher than 10, indicating their effectiveness in minimizing potential bias. Additional information concerning the IVs can be found in supplementary table II. Furthermore, the scatter plot and funnel plot displayed the causal effect of appendicular lean mass on ER- breast cancer, grip strength of the left hand on ER- breast cancer, and grip strength of the right hand on survival time of ER+ breast cancer. These plots provide support for the reliability of the two-sample bidirectional MR results, as shown in supplementary figures 1-3 (https://www.nutricionhospitalaria.org/anexos/05139-01.pdf).
DISCUSSION
In this study, we evaluated the causal relationship between sarcopenia-related traits and the occurrence and prognosis of breast cancer. Appendicular lean mass was found to be connected with the occurrence of ER- breast cancer, the grip strength of the left hand was associated with ER- breast cancer, and higher grip strength of the right hand was connected with longer survival time of ER+ breast cancer in the two-sample bidirectional MR analysis. Multivariate MR analysis demonstrated that appendicular lean mass was related to ER- breast cancer; the grip strength of the right hand and walking pace were associated with the occurrence of total breast cancer and ER- breast cancer; and stronger grip strength of the left hand was connected with a lower risk of total breast cancer, ER+ breast cancer, and ER- breast cancer after adjusting for genetically predicted BMI, waist circumference, and whole-body fat mass.
Breast cancer is a prevalent malignancy among women. Its onset and progression are influenced by factors such as age, genetic factors including BRCA1 and BRCA2 gene mutations, and other significant contributors affecting women’s health (24). In recent years, some studies have found that sarcopenia and other age-related metabolic diseases are related to the occurrence and development of breast cancer (25). Sarcopenia, commonly observed in cancer patients, could significantly impact their overall outcome (26). An observational study revealed a general decrease in muscle mass among breast cancer patients, highlighting the importance of early screening for sarcopenia symptoms (such as reduced muscle mass and grip strength) to effectively mitigate the risk of complications (27). A meta-analysis uncovered that sarcopenia significantly affected a wide range of adverse health-related outcomes, particularly in patients with breast cancer (28). This observation aligns with the symptoms associated with sarcopenia observed in our study and the survival time of breast cancer patients. Similarly, myopenia plays a significant role in determining the prognosis of various types of cancer, such as resectable esophageal cancer, noninvasive bladder cancer, and pancreatic cancer (29-31). This finding aligned with the observed causal relationship between sarcopenia-related traits and survival time in breast cancer patients in our MR study. In recent years, several studies have highlighted a higher prevalence of sarcopenia in cancer patients, which significantly impacts their prognosis and quality of life (32,33). Furthermore, a study demonstrated a significant reduction in muscle mass among breast cancer patients undergoing chemotherapy (34). These findings supported the conclusion of our reverse-MR study, which revealed lower grip strength and walking speed in breast cancer patients. A randomized controlled trial assessed the associations of sarcopenia with poor performance status, increased mortality risk, and greater side effects in oncologic patients (35). Additionally, breast cancer patients exhibited markedly impaired muscle strength and joint dysfunctions both before and after anticancer treatment (35). In a 13-year cohort study by Betty Kane et al., which followed 3,241 women, it was discovered that muscular atrophy (low muscle mass) and poor muscle mass (low muscle radioactive density) were linked to higher mortality in patients with metastatic breast cancer and poor prognosis in patients with nonmetastatic breast cancer (36). Theresa Mader et al. conducted an experiment involving a physical exercise intervention on mice with breast cancer, which resulted in a significant improvement in muscle quality and the enhancement of their mitochondria and antioxidant status, suggesting a potential mechanism for myopenia affecting breast cancer (37). Moreover, a study employing deep-learning imageomics technology found that body muscle and fat content significantly impacted the distant metastasis and related prognosis of breast cancer patients (38). These findings aligned with our own observation that myopenia was closely related to the occurrence and development of breast cancer, even after adjusting for relevant confounding factors.
This study focused on identifying the causal effects of sarcopenia-related traits on the occurrence and prognosis of breast cancer. To enhance the interpretability of the results, we utilized multivariate MR analysis to eliminate the confounding effect of BMI, waist circumference, and whole-body fat mass. Additionally, to ensure accuracy, we employed Bonferroni correction to mitigate the risk of type-I error (23). Finally, the utilization of relevant phenotypic data derived from European cohorts for exposure and outcomes had the potential to significantly reduce population selection bias (39).
This study has several limitations. First, in the multivariate MR analysis, we considered only three vital risk factors for pancreatic cancer; other risk factors were not included due to data limitations. Second, the causal effects of sarcopenia-related traits on the occurrence and prognosis of breast cancer in populations of different races remain unknown due to the inclusion of predominantly European cohorts in this study. Third, future studies should explore in detail the potential mechanism of how sarcopenia-related traits affect the occurrence and prognosis of breast cancer.
CONCLUSION
In conclusion, our two-sample bidirectional and multivariable MR study revealed the genetic association between certain sarcopenia-related traits and the occurrence and prognosis of breast cancer. These findings suggested that older women should prioritize improving their muscle quality to effectively prevent the onset and progression of breast cancer.
