Treatment Targets for Right Ventricular Dysfunction in Pulmonary Arterial Hypertension

Highlights • RV dysfunction is the strongest predictor of mortality in PAH.• Unfortunately, there are no effective therapies for RV failure.• Differences between the left ventricle and RV may allow for RV-enhancing or RV-directed therapies.• Here, we highlight the known molecular mechanisms that promote RV dysfunction in PAH and the ongoing clinical trials investigating RV function as a therapeutic target.

, and the response of the RV to PAHspecific therapy determines survival (4,5). Moreover, RV dysfunction is not solely caused by increased afterload, as some PAH patients exhibit progression of RV dysfunction despite treatment with pulmonary vasodilators (6,7). Furthermore, patients with systemic sclerosis-associated PAH (SSc-PAH) have worse RV function-and consequently increased mortality-compared with patients who have idiopathic PAH (IPAH), despite having similar severity of pulmonary vascular disease (6,8). Unfortunately, no currently available PAH therapy directly targets the RV. Therefore, there is an unmet need to combat the mechanisms underlying RV dysfunction directly to improve long-term outcomes in PAH.
In addition to PAH, RV dysfunction is prognostic in the 2 leading causes of pulmonary hypertension (PH): PH caused by left heart disease (World Health Organization [WHO] Group 2) and lung disease (WHO Group 3) (9)(10)(11). RV function predicts mortality (12,13) and is affected independent of afterload in Group 2 PH (14). Despite having less severe pulmonary vascular disease than patients with PAH, Group 3 PH patients have disproportionate RV dysfunction and poor survival (15). The presence of RV dysfunction in Group 3 PH also identifies patients at high risk of poor outcomes (11,16). Unfortunately, PAH-specific therapies do not significantly improve exercise capacity or reduce symptom burden in Group 2 and 3 PH (17).
Thus, RV-directed therapy may be beneficial for these prevalent and currently untreatable causes of PH.

RIGHT AND LEFT VENTRICLES
Mechanistic dissections of RV dysfunction have lagged behind our understanding of left ventricular (LV) dysfunction. This is exemplified by the fact that there are multiple therapies with proven survival benefits for LV failure (18,19) but no approved drugs for RV dysfunction. Importantly, the use of standard therapies for LV failure, such as beta blockers and targets of the renin-angiotensin-aldosterone system (RAAS), are not indicated or potentially contraindicated in patients with PAH and RV dysfunction (20,21). Thus, understanding the differences between the RV and LV may be important to define RV-directed therapies ( Figure 1, Table 1).
There are developmental, anatomic, and functional differences between the RV and LV ( Figure 1) that may provide insight into ways to enhance ventriclespecific function ( Figure 2, Table 1). Developmentally, the LV originates first from the splanchnic mesoderm within the primary heart field (22), whereas the RV develops second from the extracardiac mesoderm within the secondary heart field (23). The prenatal RV is thick walled and generates high pressures to support fetal blood flow (24). After birth, the pulmonary circulation becomes a low pressure circuit (25), whereas the systemic circulation becomes a high pressure system, which leads to relative LV hypertrophy (26). Ultimately, the RV becomes a thin-walled, crescent-shaped chamber, whereas the LV takes a muscular bullet shape (27). Furthermore, there are dissimilarities in cardiomyocyte arrangement that lead to differences in contractility between the 2 ventricles. The adult RV has 2 layers of cardiomyocytes: a circumferential layer that brings the RV free wall toward the interventricular septum and a deeper layer of vertical fibers that result in longitudinal shortening, which accounts for 75% of the total RV contractility (28). In contrast, the LV has 3 myocardial layers. LV contraction is a function of radial fiber thickening, longitudinal fiber shortening, and oblique fiber thickening (29). Circumferential and longitudinal shortening contribute to 67% and 33% of total LV contractility, respectively (30). Moreover, there are differences in ventricular perfusion, as the RV is perfused throughout the cardiac cycle, whereas the LV is only perfused during diastole (31). Although the normal RV and LV have similar coronary flow reserves of 400% to 500% (32)(33)(34)(35), the RV has lower baseline oxygen consumption and greater oxygen extraction reserve, which lowers its risk of ischemic injury caused by decreased coronary flow (31). However, the pressure-overloaded RV is more susceptible to ischemia, with acute or chronic increases in afterload as elevated RV afterload reduces the systolic RV perfusion gradient. This curtails overall RV perfusion because the RV is normally perfused during both systole and diastole (31). Finally, there are important anatomic differences between the ventricles that alter their response to preload and afterload. The RV is thinner and has lower volume to wall surface area, which makes the RV more compliant and better adapted to initial volume increases but renders it unable to accommodate afterload increases as well as the LV (36). This is supported by a study of isolated rat hearts that shows the LV has a higher afterload reserve than the RV (37).
There are divergent molecular responses between RV and LV failure documented in animal and human studies (27,38,39). In brief, there are differences in adrenergic signaling regulation as b1and a1-adenergic and dopamine-1 receptors are downregulated in the pressure overloaded RV, which leads to a diminished inotropic response (38,39). Although there is also downregulation of b1-adenergic receptors in LV dysfunction (40), a1-adrenergic receptors are actually upregulated and increase contractility in the pressure overloaded LV (27,38,39,41). Conversely, a1-receptor signaling decreases inotropy in the overloaded RV (27,38,39).
Furthermore, there are dissimilarities in the response to catecholamines, as only the LV develops hypertrophy following norepinephrine infusion (27,38).

MECHANISMS OF RV DYSFUNCTION IN PAH
EXPERIMENTAL MODELS OF PAH. Several commonly used preclinical models of PAH are discussed in the following sections.  (79). We also showed t-tubule disarray and downregulation of junctophilin-2 in the monocrotaline RV (80). Colchicine treatment to combat pathological microtubule remodeling increases junctophilin-2, partially corrects t-tubule architecture, and improves RV function in monocrotaline rats, albeit in the setting of less severe pulmonary vascular remodeling (80). These data suggest that t-tubule remodeling promotes RV dysfunction, but both

Fibrosis
At baseline, RV has higher collagen content (75) and different expression of matrix metalloproteinases and extracellular matrix proteins (76) compared with the LV. Antifibrotic therapies that are effective in the LV (pirfenidone and eplerenone) do not reverse fibrosis in the RV (57,72).
Inflammation RV has more macrophages and dendritic cells at baseline compared with the LV (95).
Estrogen signaling In the dysfunctional RV, the beneficial effects of estrogen are predominately mediated by ER-a (98), but in LV pressure overload, stimulation of ER-b normalizes LV ejection fraction (106).
Ischemia RV is perfused during both systole and diastole, whereas the LV is only perfused during diastole (31). At baseline, RV has a reduced microvascular bed with less tissue perfusion compared with the LV (135). There may be decreased angiogenesis during RV failure (110,(121)(122)(123).
studies have the caveat of reduced PAH severity (79,80). Nonetheless, strategies to normalize junctophilin-2 levels and restore t-tubule architecture may enhance RV function.
SARCOMERIC ABNORMALITIES. There is emerging evidence that altered sarcomere function promotes RV dysfunction in PAH. In pulmonary artery-banded rats, RV myocardial stiffness is observed in both mild and severe dysfunction (62 expression, leading to more severe heart failure following pressure overload in mice (86). Importantly, this phenotype is reversed by either upregulation of regnase-1 expression or administration of an anti-IL-6 receptor antibody (86). Clearly, there is evidence that inflammation negatively affects cardiac function.
Prisco et al. Clinical trials investigating the effects of antagonizing IL-1 (anakinra), IL-6 (tocilizumab), and CD20 on B cells (rituximab) in PAH are completed (   Table 3). Larger trials may be needed to determine the utility of FAO inhibitors in RV dysfunction in PAH. Prisco et al.    Recruiting *Some of the trials that are active but not currently recruiting as of May 2020 may be due to the current COVID-19 pandemic. †Trial in patients with severe functional regurgitation.
Prisco et al.  heightens RV contractility in isolated rodent hearts (165). The beneficial effects of PDE5 inhibitors on the RV and exercise capacity are also supported by a randomized controlled trial that investigated the effects of sildenafil (a PDE5 inhibitor) therapy on exercise performance in patients with PH due to LV systolic function (166). In this trial, sildenafil treatment heightened maximal volume of oxygen consumption, exercise cardiac output, and RV ejection fraction at rest and at peak exercise (166). However, the significant reduction in pulmonary vascular resistance needs to be considered when interpreting the change in RV function with sildenafil therapy in these patients.
Certainly, further research is needed to understand the mechanisms by which exercise improves RV function in PAH and other PH subgroups. Several ongoing clinical trials are investigating the role of exercise on RV function ( Table 3); it is hoped that they will provide more insight moving forward.

TARGETING THE RV IN CLINICAL TRIALS
Multiple clinical trials have investigated or are currently investigating the targets of RV dysfunction discussed in this review (Table 3). However, majority of these clinical trials are not specifically investigating RV function as a primary outcome.
There are some encouraging preliminary findings, as carvedilol enhances RV function (167) and decreases RV glucose uptake (168). In addition, ranolazine (118) and trimetazidine show trends for improvement in RV function. It is hoped that 1 or more of these trials will identify targets for RV failure to improve outcomes in patients with PAH and, eventually, patients with PH due to other causes.

CONCLUSIONS
RV dysfunction is the strongest predictor of mortality in PAH (1-3), but none of the currently available PAH therapies directly target the failing RV. Here, we discussed the known molecular mechanisms involved in RV dysfunction in PAH to delineate potential targets for RV-enhancing or RV-directed therapies (Central Illustration). Furthermore, we highlighted the important differences between the RV and LV, which may be particularly relevant for developing RVdirected therapies. Finally, we summarized recent and ongoing clinical trials targeting the molecular mechanisms of RV dysfunction in PAH. We hope that ongoing and future studies will identify RV-targeted therapies that will enhance quality of life and improve survival in PAH.
ACKNOWLEDGMENT The authors thank Cynthia Faraday for her assistance with designing the figures.