Periodic Reporting for period 1 - Sex Dimorphism (Molecular circuits of sex dimorphism in cardiometabolic traits and risk factors)
Période du rapport: 2023-01-01 au 2025-07-31
Cardiometabolic traits and risk factors including cardiovascular diseases, type 2 diabetes, and hypertension are the leading causes of death worldwide. These conditions exhibit some degree of sex differences, including differences in incidence or prevalence, age of onset, severity, disease progression, susceptibility, response to treatment and pharmacological adverse events. Unfortunately, the reasons behind this sex dimorphism are largely unknown, hampering the realization of an effective personalized medicine.
My project focused on understanding observational sex differences in cardiometabolic risk factors, and on the study of the genetic and molecular components that differentially impact males and females.
To reach these objectives, I first performed a systematic analysis of sex and age-dependent sex differences for 70 cardiometabolic risk factors included in the SardiNIA cohort, such as blood test parameters, anthropometric measurements, blood pressure, echocardiogram and electrocardiogram measures by fitting a generalized additive model with integrated smoothness estimation.
Secondly, I used sex-specific pQTL summary statistics from the UK Biobank Pharma Proteomics Project, along with sex-specific GWAS summary statistics of lipids from the Global Lipids Genetics Consortium. I combined these using two-sample Mendelian Randomization analyses and applying stringent multiple testing p-values correction and sensitivity analyses to identify proteins that are causally linked to lipids in a sex-specific or sex-differentiated manner. These results will lead to the discovery of new drug target to perform equally well in males and females and will contribute to the improvement of personalized and precision medicine in the near future.
My project focused on understanding observational sex differences in cardiometabolic risk factors, and on the study of the genetic and molecular components that differentially impact males and females.
To reach these objectives, I first performed a systematic analysis of sex and age-dependent sex differences for 70 cardiometabolic risk factors included in the SardiNIA cohort, such as blood test parameters, anthropometric measurements, blood pressure, echocardiogram and electrocardiogram measures by fitting a generalized additive model with integrated smoothness estimation.
Secondly, I used sex-specific pQTL summary statistics from the UK Biobank Pharma Proteomics Project, along with sex-specific GWAS summary statistics of lipids from the Global Lipids Genetics Consortium. I combined these using two-sample Mendelian Randomization analyses and applying stringent multiple testing p-values correction and sensitivity analyses to identify proteins that are causally linked to lipids in a sex-specific or sex-differentiated manner. These results will lead to the discovery of new drug target to perform equally well in males and females and will contribute to the improvement of personalized and precision medicine in the near future.
Here below the main achievements resulting for the Sex Dimorphism project:
• Exhaustive description of observational associations in males and females with cardiometabolic traits and risk factors.
Cardiometabolic diseases exhibit some degree of sex differences, including differences in incidence or prevalence, age of onset, severity, disease progression, susceptibility, response to treatment and pharmacological adverse events. Previous studies showed that for some of the related risk factors, these sex differences are not static, but rather age-dependent and thus change over life course. A better understanding of these trajectories in all populations and for all cardiometabolic risk factors is essential for the design of personalized prevention strategies. I performed a systematic analysis of sex and age-dependent sex differences for 70 cardiometabolic risk factors included in the SardiNIA cohort such as blood test parameters, anthropometric measurements, blood pressure, echocardiogram and electrocardiogram measures.
The findings of this objective can be summarized in three main messages: (1) males and females exhibit linear differences in the 84% of the phenotypes analyzed; (2) in general, sex differences are pronounced before 60 years old and tend to decrease in a proportional way with the increase of age; (3) 37% of the phenotypes analyzed showed a significant non-linear age by sex interaction.
Overall, this study highlights the importance of conducting sex-specific epidemiological research as a route to implementing sex-specific tailored preventive strategies and to aid correct identification of people at risk to develop cardiometabolic disease
• Causal inference analyses of sex specific protein quantitative trait loci (pQTL) with sex-specific GWAS summary statistics of lipids
Lipid traits are known to be sex-differential, but the underlying molecular players are largely unknown. Since protein levels are downstream products of gene expression, proteomics data can be crucial to understand etiology of sex-differences in lipids metabolism. I used sex-specific pQTL summary statistics for 2,923 circulating proteins measured in the UK Biobank Pharma Proteomics Project, along with sex-specific summary statistics of lipids from the Global Lipids Genetics Consortium. I combined these using two-sample Mendelian Randomization analyses and applied stringent multiple testing p-values correction and sensitivity analyses.
The main achievement of this section are the following: (1) I identified a total of 82 and 82 robust causal protein-lipids links that are specific to females and males; (2) a total of 38 protein–lipids links that showed sex-differentiated causal effects; (3) the identified links corresponding to 127 unique proteins, specifically 116 sex-specific, 21 acting in sex-differentiated manner on lipid levels and 10 acting as sex-specific or sex-differentiated depending on the outcome. Most sex-specific proteins within each sex were outcome-specific and not shared across the five lipid-related outcomes analyzed. I also observed an higher proportion of sex-specific relationships in males for all traits (except for LDL and non-HDL), suggesting that, in females, regulation of lipids is more strongly influenced by non-genetic factors, for example sex hormones fluctuations during life stage. Of note, the difference in discovery is unlikely to be due to differences in statistical power, since the sample size of females-only GWASs was overall similar than the males-GWAS for both proteins and lipid GWASs. Intriguingly, several of the sex-specific proteins were previously shown to be involved in inflammation and cardiometabolic disease, such as Apolipoprotein(a) and lipoprotein lipase. These results provide insight into potential genetic drivers of sex dimorphism in lipid metabolism.
The results of this objectives are part of the manuscript titled “Exploring sex-specific causal links between thousands of proteins and lipid metabolism using the UK Biobank Pharma proteomics data” available on MedrXiv at this link https://medrxiv.org/cgi/content/short/2025.09.16.25335948v1(s’ouvre dans une nouvelle fenêtre)
• Exhaustive description of observational associations in males and females with cardiometabolic traits and risk factors.
Cardiometabolic diseases exhibit some degree of sex differences, including differences in incidence or prevalence, age of onset, severity, disease progression, susceptibility, response to treatment and pharmacological adverse events. Previous studies showed that for some of the related risk factors, these sex differences are not static, but rather age-dependent and thus change over life course. A better understanding of these trajectories in all populations and for all cardiometabolic risk factors is essential for the design of personalized prevention strategies. I performed a systematic analysis of sex and age-dependent sex differences for 70 cardiometabolic risk factors included in the SardiNIA cohort such as blood test parameters, anthropometric measurements, blood pressure, echocardiogram and electrocardiogram measures.
The findings of this objective can be summarized in three main messages: (1) males and females exhibit linear differences in the 84% of the phenotypes analyzed; (2) in general, sex differences are pronounced before 60 years old and tend to decrease in a proportional way with the increase of age; (3) 37% of the phenotypes analyzed showed a significant non-linear age by sex interaction.
Overall, this study highlights the importance of conducting sex-specific epidemiological research as a route to implementing sex-specific tailored preventive strategies and to aid correct identification of people at risk to develop cardiometabolic disease
• Causal inference analyses of sex specific protein quantitative trait loci (pQTL) with sex-specific GWAS summary statistics of lipids
Lipid traits are known to be sex-differential, but the underlying molecular players are largely unknown. Since protein levels are downstream products of gene expression, proteomics data can be crucial to understand etiology of sex-differences in lipids metabolism. I used sex-specific pQTL summary statistics for 2,923 circulating proteins measured in the UK Biobank Pharma Proteomics Project, along with sex-specific summary statistics of lipids from the Global Lipids Genetics Consortium. I combined these using two-sample Mendelian Randomization analyses and applied stringent multiple testing p-values correction and sensitivity analyses.
The main achievement of this section are the following: (1) I identified a total of 82 and 82 robust causal protein-lipids links that are specific to females and males; (2) a total of 38 protein–lipids links that showed sex-differentiated causal effects; (3) the identified links corresponding to 127 unique proteins, specifically 116 sex-specific, 21 acting in sex-differentiated manner on lipid levels and 10 acting as sex-specific or sex-differentiated depending on the outcome. Most sex-specific proteins within each sex were outcome-specific and not shared across the five lipid-related outcomes analyzed. I also observed an higher proportion of sex-specific relationships in males for all traits (except for LDL and non-HDL), suggesting that, in females, regulation of lipids is more strongly influenced by non-genetic factors, for example sex hormones fluctuations during life stage. Of note, the difference in discovery is unlikely to be due to differences in statistical power, since the sample size of females-only GWASs was overall similar than the males-GWAS for both proteins and lipid GWASs. Intriguingly, several of the sex-specific proteins were previously shown to be involved in inflammation and cardiometabolic disease, such as Apolipoprotein(a) and lipoprotein lipase. These results provide insight into potential genetic drivers of sex dimorphism in lipid metabolism.
The results of this objectives are part of the manuscript titled “Exploring sex-specific causal links between thousands of proteins and lipid metabolism using the UK Biobank Pharma proteomics data” available on MedrXiv at this link https://medrxiv.org/cgi/content/short/2025.09.16.25335948v1(s’ouvre dans une nouvelle fenêtre)
The Sex Dimorphism project has successfully achieved most of its objectives and milestones, delivering significant scientific contributions, with only minor deviations from the original plan.
In line with IRGB-CNR efforts on innovation and technology transfer in biomedicine, the results of this project will be published in two high impact factor international peer-review journals.
The “Age-dependent sex differences of cardiometabolic traits in the Sardinian founder population” manuscript, resulting from Specific Objective 1, has relevant potential implications. I observed that quantitative phenotypes related to cardiovascular and metabolic disorders show differences between males and females that are not static but change over age. Thus, studies on clinical outcome aiming to identify sex differences without taking age into account can lead to misleading conclusion. Current diagnostic criteria for cardiometabolic traits are not sex specific and fail to discriminate physiological and pathological adaptation. Cardiometabolic diseases in females are frequently overlooked by routine exams and are diagnosed later and with more severe symptoms than in males. Although fewer females are diagnosed with cardiovascular disease than males, these females are more likely to die from the disease. For example, heart failure is currently diagnosed with sex-neutral ejection fraction thresholds. Females have a higher baseline ejection fraction than males. When the same ejection fraction criteria are used for both female and male patients, this may under-diagnose females whose baseline ejection fraction is higher than that of male. Ideally, for the diagnosis of heart failure with preserved ejection fraction, the critical ejection fraction threshold in female should be much higher. It was recently estimated that using sex-specific assays to diagnose type 1 myocardial infarction results in 30% more females and 4.9% more males being diagnosed, with all patients exhibiting the same symptoms associated with this pathology. Clearly, there is an urgent need to better understand the female heart and design sex-specific diagnostic criteria that will allow us to diagnose cardiac disease in female equally as early, robustly, and reliably as in male.
The manuscript titled “Exploring sex-specific causal links between thousands of proteins and lipid metabolism using the UK Biobank Pharma proteomics data”, resulting from Specific Objective 2 and 3, highlighted several proteins with sex-specific and sex-differentiated causal impact on lipid levels, results that could be obtained only with the availability of sex-specific GWASs. The discovery of these sex-differences can provide important etiological insights into the management of lipid metabolism. Such knowledge may in turn improve the predictive use of sex-specific pQTLs and point to new therapeutic gender specific strategies to prevent cardiovascular diseases. Furthermore, my approach provides a proof-of-concept of the utility of sex-stratifed pQTLs and GWAS. Thus, my work, encourages all future GWAS of any complex trait and disease to follow this strategy as route to derive relevant insights into potential genetic drivers of sex dimorphism not only in lipid metabolism.
In line with IRGB-CNR efforts on innovation and technology transfer in biomedicine, the results of this project will be published in two high impact factor international peer-review journals.
The “Age-dependent sex differences of cardiometabolic traits in the Sardinian founder population” manuscript, resulting from Specific Objective 1, has relevant potential implications. I observed that quantitative phenotypes related to cardiovascular and metabolic disorders show differences between males and females that are not static but change over age. Thus, studies on clinical outcome aiming to identify sex differences without taking age into account can lead to misleading conclusion. Current diagnostic criteria for cardiometabolic traits are not sex specific and fail to discriminate physiological and pathological adaptation. Cardiometabolic diseases in females are frequently overlooked by routine exams and are diagnosed later and with more severe symptoms than in males. Although fewer females are diagnosed with cardiovascular disease than males, these females are more likely to die from the disease. For example, heart failure is currently diagnosed with sex-neutral ejection fraction thresholds. Females have a higher baseline ejection fraction than males. When the same ejection fraction criteria are used for both female and male patients, this may under-diagnose females whose baseline ejection fraction is higher than that of male. Ideally, for the diagnosis of heart failure with preserved ejection fraction, the critical ejection fraction threshold in female should be much higher. It was recently estimated that using sex-specific assays to diagnose type 1 myocardial infarction results in 30% more females and 4.9% more males being diagnosed, with all patients exhibiting the same symptoms associated with this pathology. Clearly, there is an urgent need to better understand the female heart and design sex-specific diagnostic criteria that will allow us to diagnose cardiac disease in female equally as early, robustly, and reliably as in male.
The manuscript titled “Exploring sex-specific causal links between thousands of proteins and lipid metabolism using the UK Biobank Pharma proteomics data”, resulting from Specific Objective 2 and 3, highlighted several proteins with sex-specific and sex-differentiated causal impact on lipid levels, results that could be obtained only with the availability of sex-specific GWASs. The discovery of these sex-differences can provide important etiological insights into the management of lipid metabolism. Such knowledge may in turn improve the predictive use of sex-specific pQTLs and point to new therapeutic gender specific strategies to prevent cardiovascular diseases. Furthermore, my approach provides a proof-of-concept of the utility of sex-stratifed pQTLs and GWAS. Thus, my work, encourages all future GWAS of any complex trait and disease to follow this strategy as route to derive relevant insights into potential genetic drivers of sex dimorphism not only in lipid metabolism.