Is ageing a modifiable risk factor for atrial fibrillation?

This editorial refers to ‘The role of cellular senescence in profibrillatory atrial remodeling associated with cardiac pathology’, by M. Mehdizadeh et al., https://doi.org/10.1093/cvr/cvae003. Recently, the management of risk factors and comorbidities has taken on a prominent position in addressing the growing burden of atrial fibrillation (AF). Comorbidity/risk-factor management is currently positioned as the third pillar of the integrated ABC pathway in the ESC guidelines, next to anticoagulation and better symptom control through rate/rhythm-control therapy.1 This central role of risk-factor management reflects the growing awareness that AF is the result of an atrial cardiomyopathy resulting from complex interactions between numerous dynamic risk factors and comorbidities.2 These risk factors are typically divided into modifiable and non-modifiable risk factors. Modifiable risk factors include obesity, hypertension, and other cardiovascular diseases, with several clinical studies showing, e.g. pronounced reductions in AF burden and improved outcomes with rhythm-control therapy after significant weight loss.1 By contrast, advancing age, male sex, and genetic predisposition have traditionally been considered the primary non-modifiable risk factors for AF. In this issue of Cardiovascular Research, Mehdizadeh et al.3 provide intriguing data indirectly questioning the non-modifiable nature of ageing-induced AF promotion. In particular, the authors show that old rats (20 months, i.e. ∼60 years in humans) have the expected increase in AF susceptibility compared to young rats, which is associated with an increase in markers of cellular senescence, a molecular hallmark of ageing.3 Moreover, cellular senescence and AF inducibility were also increased in young rats with myocardial infarction (MI), another major AF risk factor. Preventing cellular senescence with the accepted senolytic cocktail quercetin and dasatinib abolished AF inducibility in rats post-MI.3 As acknowledged by the authors, the study does not directly show that senolytic therapy reduces AF inducibility in old animals. However, the combination of an accepted senolytic therapy dramatically reducing AF inducibility under a pathological condition that is clinically relevant and associated with increased senescence suggests that AF-promoting effects of ageing might be similarly amenable to therapeutic intervention. Senescence represents a phenotype of stable cell cycle arrest and a unique secretome (the senescence-associated secretory phenotype; SASP) involving proinflammatory and profibrotic factors that further promote senescence and stimulate atrial remodelling through paracrine effects.4 Senescence is regulated via multiple cascades, including the p16 and p53-p21 pathways that repress the activity of cyclin-dependent kinases, preventing phosphorylation of retinoblastoma protein, which controls the cell cycle (Graphical Abstract).4 Mehdizadeh et al. identified increased levels of p16 and p21, but not p53, in atrial samples of patients above 70 years of age compared to patients under 60, in line with the increased cellular senescence observed in older animals.3 By contrast, there was fairly limited upregulation of markers of cellular senescence in atrial samples of patients with AF compared to sinus-rhythm controls of similar age.3 This may, in part, be due to the nature of these samples, originating from patients with sufficient cardiovascular disease to necessitate open-heart surgery. Moreover, even a single senescent cell among approximately 10,000 cells is sufficient to initiate a senescence-promoting feedback mechanism and trigger persistent age-related phenotypes in mice,5 suggesting that even small differences may be clinically relevant. On the other hand, these data suggest that targeting cellular senescence may provide a strong nodal point to prevent atrial remodelling by breaking the positive feedback initiated by ageing itself, as well as other AF-promoting risk factors (Graphical Abstract). This is consistent with the data from Mehdizadeh et al., indicating that senolytic therapy prevents the development of an AF-promoting substrate post-MI.3 Although studies assessing classical upstream therapy, e.g. with angiotensin-converting enzyme inhibitors or aldosterone antagonists, have shown some value in preventing new-onset AF, results for secondary prevention were generally disappointing.6 These data suggest that senolytic therapy might need to be initiated early in the disease process. In agreement, Mehdizadeh et al. started senolytic therapy immediately after MI in young rats, which limited the development of atrial structural remodelling and AF inducibility.3 However, the effectiveness of this therapy when initiated at (moderately) advanced age or after risk factors have been present for some time remains unknown. Similarly, the optimal frequency and duration of therapy is unknown. Importantly, cellular senescence may protect against tumorigenesis, suggesting that senolytic compounds could potentially promote cancer growth.4 Mehdizadeh et al. used intermittent treatment for 3 days at 2-week intervals to clear senescent cells,3 which may limit adverse effects. Intermittent application of senolytics was indeed safe in rodents in previous work.5 Nevertheless, in light of these findings, it becomes crucial to start treatment only when a true phenotype is present. To this end, reliable biomarkers for (detrimental) cellular senescence are needed to guide treatment initiation and monitor efficacy. At present, no specific biomarkers to assess senescence in humans are available.4 However, a combination of blood biomarkers and artificial intelligence-based aging clocks or ECG analyses might represent a promising approach.7,8 The data from Mehdizadeh et al. contribute to the growing preclinical literature on the therapeutic potential of senescence in ageing-associated diseases, such as AF. Nevertheless, there are several major challenges for the translation of preclinical data to clinical application.6 There are important differences in ageing-associated alterations in metabolite signatures between male and female cardiac tissues in mice,9 indicating that sex differences will have to be taken into account in future preclinical and translational studies.10 Moreover, animal models can only partially capture the complexity of the dynamic risk factors promoting AF in patients. Thus, there is a clear need for well-designed clinical trials assessing prevention of ageing-related AF through senolytic therapy. Unfortunately, such trials will intrinsically need long follow-up times, which, in combination with the strict safety end-points required for a non-life-threatening condition such as AF, will most likely make them incredibly costly. Several clinical trials with senolytics for non-cardiovascular disease conditions are currently ongoing.4 If successful, such trials may pave the way for repurposing strategies aiming to apply senolytics in patients with AF or AF-promoting comorbidities. Even if the modifiability of ageing as a risk factor for AF remains uncertain at present, these data are encouraging for the development of innovative therapeutic approaches to address the growing burden of AF. No new data were generated or analysed in support of this research. The authors did not receive direct funding for this editorial.