In Need of Signalling Pathway Data

This article refers to ‘Adiposity, body composition and ventricular-arterial stiffness in the elderly: the Atherosclerosis Risk in Communities Study’ by M.M. Fernandes-Silva et al., published in this issue on pages 1191–1201. As the world population gets older, clinicians can no longer deal only with children, adults and elderly since a fourth category of patients emerged: the very old adults represented by individuals over 80 years of age. Those very old adults have survived and adapted to all risk factors and adverse events they encountered during their lifespan. While this new category of patients is largely uncharacterized—mostly lacking reference data for any considered studied phenotypes—clinicians have to adapt their practice to them. Aging is a risk factor for cardiovascular diseases and clinicians have to prevent its premature onset. In order to phenotype this group of patients, clinical efforts like the one described in the study of Fernandes-Silva et al.1 in the current issue of the Journal are required to unravel the specificity of aged populations. Focusing on the characterization of the heart and large arterial stiffness—known consequences of aging—Fernandes-Silva et al. hypothesized that increased adiposity would be associated with higher left ventricular (LV) and arterial stiffness in the individuals examined during the fifth visit of the Atherosclerosis Risk in Communities (ARIC) study (aged 66–99 years). At the fifth visit, the ARIC population presented hypertension in over 73% of the participants, excessive adiposity (overweight to obese) in 73% of the individuals which was associated with higher prevalence of hypertension, diabetes mellitus and heart failure. Among markers of coronary artery disease, arterial stiffness has proven to be an important parameter for the assessment of cardiovascular risk. The carotid to femoral pulse wave velocity (PWV) used by the authors is the gold standard method2 and remains a prognostic marker in heart failure patients.3 If reference and normal values for PWV have been previously described based on a large (n = 16 867) European population,2 no stratification for adiposity was specifically analysed. The study from Scuteri et al.,4 however, has extensively investigated the relationship between total and abdominal adiposity in another large cohort of community dwelling volunteers. They demonstrated a non-linear relationship between obesity and arterial stiffness in 6148 participants aged 14 to 102 years. In this study, while age was strongly associated with PWV, the effects of body mass index (BMI) on PWV did not differ across age groups, although the association appeared weaker in subjects >65 years (not discriminating over 80 years old individuals). Interestingly, Fernandes-Silva et al.1 demonstrated that body fat percentage—another parameter assessing adiposity—showed as well non-linear relationships with PWV in the ARIC population. This non-linear relationship implies that smaller reduction in adiposity (by lifestyle intervention) will result in a dramatically greater impact on PWV and, thus, on individual cardiovascular risk profile in the elderly group of studied individuals. This comes in reinforcement of the observation made in the systematic review and meta-analysis performed by Petersen et al.5 and offers a clinical alternative to try reducing arterial stiffness in elderly patients. The protocol of the fifth visit allowed also a cardiac function evaluation of the ARIC participants. Of interest, this investigation showed that the association between obesity and high LV stiffness still holds among the elderly and that this was independent of other potentially confounding variables. This comes in line with the conclusion of Russo et al.6 who observed that in a population-based elderly cohort (mean age 71 ± 9 years) abdominal adiposity was independently associated with subclinical LV dysfunction in all BMI categories, suggesting that increased abdominal obesity might be a risk factor for LV dysfunction regardless of the presence of general obesity. Fernandes-Silva et al. have elegantly used three different morphometric parameters to evaluate fat deposition: BMI, waist circumference (WC) and body fat percentage, but yet none of them clearly locate fat deposits. Furthermore they reported stronger or weaker associations with ventricular and arterial stiffness depending on the used morphometric parameter. Thus, morphometric parameters used to clinically evaluate adiposity are not alike and this is probably because adipose tissue is a very complex organ and does not display the same characteristics depending on its location.7 In healthy individuals, white (WAT) and brown (BAT) adipose tissue deposits serve as storages and buffers for fatty acids. They serve also as attenuators of glycaemia and dyslipidaemia, and as controllers of vascular tension and inflammation. However, under abnormal or excessive fat accumulation, both types of deposits become dysfunctional. WAT hypertrophies and has a high lipolytic activity releasing free fatty acids, which are detrimental for the heart structure and function.8 It is also infiltrated by macrophages, which are interacting with the adipocytes, and contributes to trigger a low-grade inflammatory process by releasing pro-inflammatory factors (cytokines, chemokines, renin–angiotensin–aldosterone system compounds) into the circulation. In contrast, when becoming dysfunctional, BAT atrophies and becomes inactive by losing its protective anti-glycaemic/dyslipidaemic and anti-inflammatory properties. Experimental and clinical evidence have suggested that when dysfunctional adipocytes are located in close vicinity (visceral adipose tissue) with major organs, they become more harmful to the myocardium and the coronary arteries than the subcutaneous ones.9 Furthermore, there is a growing body of evidence that when compared to the subcutaneous adipocyte, visceral fat cells are shown to have a differential production of adipokines. Furthermore, epicardial adipose tissue has been described as a unique and multifaceted fat depot with local and systemic effects. This tissue shares with the heart an unobstructed microcirculation that facilitates the interaction between these two organs.10 The epicardial-fat-specific transcriptome has been shown to be down-regulated in the presence of severe and advanced coronary artery disease. If straightforward morphometric parameters could help identifying patients with increased risk of ventricular and arterial stiffness or patients who might gain greater benefit from a given treatment,11 those parameters do not inform on the altered biological functions and signalling. The clinical evidence of Fernandes-Silva et al.1 sustain the conclusion that increased adiposity is associated with increased ventricular–arterial stiffness among very old adults (i.e. >80 years), suggesting a potential mechanism by which obesity might contribute to the development of heart failure. The deleterious effects of the exocrine activity of the adipose tissue have probably harmed the ventricle and artery since several visits. This long-lasting dysregulation of many signalling pathways before the fifth visit resulting in organ stiffness could have been followed through bioprofiling at different follow-up times of the participant. Excess adiposity is associated with numerous alterations that could partly be evaluated through biomarker analysis: higher levels of inflammation, insulin resistance as evaluated by the authors, but also fibrosis and alteration of extracellular matrix protein expression.12-14 However, numerous other parameters could follow the alterations arising from excess adiposity.15 Stiffer arteries have been proposed to be associated with higher leptin and high-sensitivity C-reactive protein.16 An excessive production of aldosterone influences outcomes in obese patients, and excessive activity of the renin–angiotensin–aldosterone system has been associated with LV hypertrophy, arterial stiffness and fibrosis development, experimentally as well as clinically.17 As mentioned by the authors, a limitation of their study is the cross-sectional design of the approach. It precludes the determination of this stiffening onset—potential early vascular aging18—in the individuals presenting increased adiposity as compared to the leaner ones. As mentioned above, this limitation could have been partly compensated by the coupling of their clinical observations with molecular evidence. Indeed, information about alteration of pre-identified or novel signalling pathways would have offered additional hypothesis towards understanding of the patient's stiffness status and origin. Taking advantage of the development of high throughput technologies, assessment of hundreds of biomarkers of different kinds (proteomic, transcriptomic, metabolomic, genomic) can now be performed even in very small volumes of biosamples. Hence, avoiding the exhaustion of the precious clinical biosamples, those optimized approaches can help stratifying patients at the visit they have been fully phenotyped. Where cross-sectional design has been seen as a limitation in the current ARIC study, the use of bioprofiles could have helped identifying biomarkers within signalling pathways associated with stiffness phenotype. Even better, they could have helped following their dysregulation along aging based on retrospective bioprofiling of the individuals at earlier visits (determination of the onset of ventricular and arterial stiffness). Pushing forwards, as vascular and cardiac stiffening is not exclusively characteristic of older subjects and as it has been described in obese children cohorts, the expression of the identified biomarkers could also be evaluated in any cohort where premature aging is suspected. Of course dealing with several hundreds of supplemental measurements performed in addition to more classical phenotyping requires from clinicians to collaborate with computer scientists specialized in computational biology and larger budgets. Several huge ongoing scientific programmes are already bridging clinicians (to phenotype individual and suggest associations) together with mathematicians (to model diseases), computational biologists (to stratify patients according to signalling pathways) and basic scientists (to experimentally validate the suggested association) (HOMAGE: www.homage-hf.eu; HERMES: www.hermes-h2020.eu). Approaching disease using trans-disciplinary approaches will optimize the expected outcomes of huge investigations like the ARIC study. Joining efforts to cross validate hypothesis between consortia will also empower the data and their analysis and substantiate the conclusions. And because of this in-depth characterization of clinical phenotypes, one can expect that precision medicine is not just an utopia but will progressively be integrated in the daily practice. Conflict of interest: none declared.