Post-traumatic osteoarthritis: A review of pathogenic mechanisms and novel targets for mitigation

Abstract

Post-traumatic osteoarthritis (PTOA) develops secondary to a joint injury and accounts for 12 % of all osteoarthritis. These injuries, often of the lower extremity joints, occur due to trauma or accidents related to athletic or military activities. They primarily affect younger individuals although PTOA can occur across the spectrum of age. Pain and functional disability caused by PTOA confer a heavy economic toll on patients, in addition to detrimentally affecting their quality of life. Both high energy injuries that cause articular surface fracture with or without subchondral bone disruption and low-energy injuries involving joint dislocations or ligamentous injury cause PTOA, albeit through different mechanisms. Regardless, chondrocyte death, mitochondrial dysfunction, reactive oxygen species production, subchondral bone remodeling, inflammation and cytokine release in the cartilage and synovium play integral roles in the pathogenesis of PTOA. Evolving surgical methods are focused on stabilizing articular surface and joint structure congruity. However, to date there are no disease modifying medical therapies against PTOA. Increased recent understanding of the pathogenesis of the subchondral bone and synovial inflammation as well as that of chondrocyte mitochondrial dysfunction and apoptosis have led to the investigation of new therapeutics targeting these mechanisms to prevent or delay PTOA. This review discusses new advances in our understanding of cellular mechanisms underlying PTOA, and therapeutic approaches that are potentially effective in reducing the self-propagating cycle of subchondral bone alterations, inflammation, and cartilage degradation. Within this context, we focus therapeutic options involving anti-inflammatory and anti-apoptotic candidates that could prevent PTOA.

Highlights

• •There are no effective clinical therapies for preventing or mitigating PTOA.
• •PTOA is a whole joint disease involving synovium, subchondral bone, and cartilage dysregulation.
• •Cartilage impact following intraarticular fracture initiates chondrocyte apoptosis and mitochondrial dysfunction.
• •Chronic aberrant loading of the joint following injuries activate inflammatory mechanisms in chondrocytes and synovium.
• •Prevention of inflammation protects against PTOA.
• •Clinical trials are studying preclinical mitoprotective and anti-inflammatory therapies to prevent PTOA.

1. Introduction

Post-traumatic osteoarthritis (PTOA) is a major contributor to disability worldwide across all age groups (Brown et al., 2006; Cross et al., 2014; Rivera et al., 2012; Whittaker et al., 2015). Currently, there are no disease-modifying therapies to prevent or mitigate the progression of the disease. PTOA typically occurs after a direct insult in the form of an intra-articular fracture (IAF), with incidence rates from 11 to 75 % depending on the joint involved (Doornberg et al., 2007; Giannoudis et al., 2005; Lutz et al., 2011; Marsh et al., 2003; Rademakers et al., 2004; Rademakers et al., 2007; Weigel and Marsh, 2002). The initial impact to the cartilage, combined with downstream pathobiological and pathomechanical changes, leads to disease development. PTOA can also result from chronic changes in the prevailing trans articular loading environment that can result from residual instability, incongruity, or malalignment. For example, PTOA (9–12 % at 5 years; 23 % at 10 years) occurs after injuries that lead to instability of the joint, such as in an anterior cruciate ligament tear (ACL-t), joint dislocation or injury to other static and dynamic stabilizers of a given joint even after surgical repair (Bodkin et al., 2020; Rhon et al., 2019; Everhart et al., 2021). PTOA in acetabular fractures is clearly associated with residual mal-reduction leading to joint surface incongruity (Tannast et al., 2012; Ziran et al., 2019). Collectively, both acute mechanical damage from articular surface impact and chronic aberrant loading from residual changes in stability, congruity and alignment following an injury can lead to PTOA.

Current clinical practice for IAFs and major articular ligamentous injuries usually involves a surgical procedure to anatomically restore the articular surface, the ligament or tendon, respectively, and reestablish joint surface congruity, joint alignment, and joint stability to avoid chronically abnormal articular contact stresses and stress rates. The importance of articular surface congruity and joint stability is dependent on the injured joint (Canadian Orthopaedic Trauma, 2006; Carbonell-Escobar et al., 2017; Giannoudis et al., 2010; Tang et al., 2014; Thomson et al., 2008; Virkus et al., 2018). For instance, knee alignment and stability are more important than articular surface congruity, whereas in fractures of the superior weightbearing dome of the acetabulum in the hip joint, restoring articular congruity is very important in preventing PTOA (Giannoudis et al., 2005; Giannoudis et al., 2010). Residual incongruities can increase peak and mean stresses at the cartilage surface, which have been shown to lead to cartilage loss and PTOA progression in patients with fractures exceeding a specific stress threshold (3 MPa-seconds/gait cycle) (Anderson et al., 2011a; Segal et al., 2009; Li et al., 2008). Recently, weightbearing computed tomography (CT) technology has been introduced and it is likely that this method will allow much more precise analyses of changes in the residual mechanical environment in a joint (Willey et al., 2020; Day et al., 2020). However, even with these surgical treatments, it is not always possible to prevent the development of symptomatic PTOA as the initial impact damage may predominate the pathophysiology (Doornberg et al., 2007; Giannoudis et al., 2005; Lutz et al., 2011; Marsh et al., 2003; Rademakers et al., 2004; Rademakers et al., 2007; Weigel and Marsh, 2002).

In some cases, the magnitude of the original injury likely has the dominant effect on outcomes. Anderson et al. (Beardsley et al., 2005; Beardsley et al., 2002; Rao et al., 2019; Thomas et al., 2010; Anderson et al., 2016) developed a CT-guided assessment of articular fracture energy and generate a severity score capable of predicting PTOA risk in articular fracture patients. They observed a positive correlation between increased severity score associated with initial fracture dispersion with articular comminution and increased risk of moderate to severe PTOA. In this work, it has been shown that fracture energy thresholds exist that if exceeded, PTOA will likely result regardless of the residual mechanical environment.

Over the past two decades, a great increase in the understanding of the pathophysiology of PTOA has been achieved, with two overarching mechanisms emerging. The first of these mechanisms involves acute cartilage insult resulting in chondrocyte apoptosis and necrosis largely driven by dysregulated mitochondrial biogenesis and redox imbalance progressing to chronic inflammation and cartilage loss (Ayala et al., 2021; Coleman et al., 2018; D'Lima et al., 2001). The second mechanism is mediated by chronic changes in the residual mechanical environment that prevails in the joint following an injury such as ACL-t or posterior cruciate ligament tear (PCL-t), resulting from changes in congruity, alignment, and stability, and leading to subchondral bone remodeling, synovitis, cartilage degradation, and PTOA (Chen et al., 2017; Kapoor et al., 2011; Koike et al., 2015; Rai et al., 2017; Rieder et al., 2018; Wang et al., 2015; Wood et al., 2016).

This review will discuss the current understanding of the pathogenesis of PTOA after acute injury, current preclinical models, potential treatment targets, and therapeutics currently under clinical investigation. Further, we will discuss the mechanisms contributing to PTOA when chronic articular instability occurs due to traumatic joint injuries such as ACL-t or PCL-t. Joint malignment in gait and postural-alignment conditions such as knee varus or valgus deformation and flexion contractures result in degenerative OA – not PTOA (Tateuchi, 2019), will not discussed here.

2. Pathogenesis of PTOA – intra-articular fracture

2.1. Articular cartilage damage after acute trauma: mitochondrial dysregulation and chondrocyte apoptosis

It is estimated that articular fractures increase the risk of PTOA by 20-fold (Anderson et al., 2011b). A high energy impact to the articular surface, with or without subchondral bone fracture and displacement, results in localized tissue damage and significant chondrocyte death, both at the location of impact and its vicinity (Bajaj et al., 2010). Chondrocyte necrosis is predominant at higher loads and with complete osteochondral fracture (Ayala et al., 2021; Coleman et al., 2016; Milentijevic et al., 2005; Stolberg-Stolberg et al., 2013). Acute trauma to the cartilage damages chondrocyte cell integrity leading to an influx of Ca²⁺ and a leakage of intracellular material such as mitochondrial cytochrome C, triggering a wave of apoptotic responses in surrounding cells (D'Lima et al., 2001; Olson and Guilak, 2006). The current understanding of early cellular changes that occur after IAF describes an early window of increased mitochondrial electron transport chain (ETC) activity and elevated levels of reactive oxygen species (ROS) presumably generated at electron transport chain Complex I in chondrocytes, triggering the release of free oxygen radicals and metabolites within the impacted cartilage and activating intrinsic apoptotic pathways (caspases 3 and 9, and glycogen synthase kinase 3) in chondrocytes (Ayala et al., 2021; Coleman et al., 2018; D'Lima et al., 2001; Koike et al., 2015; Rieder et al., 2018; Wang et al., 2015; Wood et al., 2016; Wegner et al., 2019; Johnson et al., 2000) (Table 1).

Table 1.

Mechanisms associated with PTOA pathogenesis following intra-articular fracture.

PTOA etiology	Pathogenesis mechanisms	References
IAF	Acute cartilage trauma:
	-Chondrocyte apoptosis, mitochondrial dysfunction, and release of free oxygen radicals	(Ayala et al., 2021; Coleman et al., 2018; D'Lima et al., 2001; Koike et al., 2015; Rieder et al., 2018; Wang et al., 2015; Wood et al., 2016; Anderson et al., 2011b; Bajaj et al., 2010; Coleman et al., 2016; Milentijevic et al., 2005; Stolberg-Stolberg et al., 2013; Olson & Guilak, 2006; Wegner et al., 2019; Johnson et al., 2000; Coleman et al., 2017)
	-Synovial inflammation – CXCL10 expression, macrophage infiltration of synovium, Nox4, secretion of pro-inflammatory cytokines such as IL-1, IL-6 and TNF-α by inflamed macrophages triggering cartilage catabolism	(Wegner et al., 2019; Furman et al., 2015; Dwivedi et al., 2022; Furman et al., 2014; Olson et al., 2015; Martini et al., 2005; Furman et al., 2018)
	-Synovial inflammation and subchondral bone alterations in trauma – MRL/Mpj super healer mice	(Ward et al., 2008; Clark et al., 1998)

Given these findings, investigators have evaluated different therapies to address mitochondrial dysfunction. Coleman et al. (Coleman et al., 2018; Coleman et al., 2017) conducted a series of studies demonstrating the integral role played by ROS and mitochondrial ETC imbalance in chondrocyte apoptosis and necrosis after supraphysiological loading. In vitro studies by this group demonstrated that chondrocyte mitochondrial dysfunction and increased superoxide release resulting from repeated supraphysiological loading of cartilage explants could be suppressed by the administration of an antioxidant, N-acetylcysteine (NAC), and superoxide dismutase mimetics. Further, impacted osteochondral explants displayed increased chondrocyte viability when treated with cardiolipin, a mitochondrial-based apoptosis inhibitor. Finally, in a porcine pilon IAF model, treatment with the reversible mitochondrial complex I inhibitor amobarbital or oxidant scavenging with NAC significantly reduced PTOA at 6 and 12 months after injury (Coleman et al., 2018). Collectively, mitochondrial dysfunction plays a pathoetiologic role in chondrocyte death and dysfunction following a high energy impact (Table 1). In addition, translational level evidence has demonstrated that mitochondrial manipulation prevents IAF-induced PTOA.

2.2. Cartilage-synovium-subchondral bone crosstalk after acute cartilage trauma

Mitochondrial dysfunction and release of free oxygen radicals, resulting in chondrocyte cell death following acute trauma, also lead to elevated levels of the pro-inflammatory cytokine interleukin (IL)-6 in chondrocytes, which activates downstream mediators such as mitogen activated protein kinases (MAPK). Together with impact-mediated release of fibronectin fragments into the joint, this stimulates further chondrocyte death and matrix degradation in addition to initiating a cascade of inflammatory events in the joint capsule (Ayala et al., 2021; Coleman et al., 2018; D'Lima et al., 2001; Koike et al., 2015; Rieder et al., 2018; Wang et al., 2015; Wood et al., 2016; Wegner et al., 2019).

Inflammation of the synovium plays a role in the continued dysregulation of cartilage homeostasis and altered subchondral bone remodeling (Table 1). Analysis of human synovial tissue and fluid collected from patients with acute IAFs revealed increased synovial inflammation, as evidenced by the presence of CD68⁺ macrophages in the synovial tissue and increased inflammatory biomarkers levels in the synovial fluid (Furman et al., 2015). Human osteochondral explants co-cultured with synovium after cartilage injury expressed inflammatory markers (IL-1, IL-6, IL-8, and tumor necrosis factor α (TNF-α)), and released glycosaminoglycan (GAG) and aggrecan fragments, compared to injured osteochondral plugs in monoculture (Dwivedi et al., 2022; Furman et al., 2014; Olson et al., 2015). Murine models of IAF showed acute (within 24 h of injury) expression of the chemokine C-X-C motif chemokine ligand 10 (CXCL10) in the synovial tissue and increased synovial inflammation in areas adjacent to the fracture and other joint compartments. Enhanced CXCL10 was also observed in human OA cartilage, implying that its prolonged expression in chondrocytes promotes chemotaxis of inflammatory macrophages into the synovium (Martini et al., 2005; Furman et al., 2018).

Synovial inflammation in PTOA is correlated with altered subchondral bone density, and IL-1β plays a role in both processes. Intra-articular treatment with an IL-1β receptor antagonist in conjunction with NADPH oxidase 4 (Nox4) at the time of IAF inhibited synovial inflammation and osteochondral damage, whereas its systemic treatment did not protect against PTOA (Wegner et al., 2019; Furman et al., 2014; Olson et al., 2015). Further, closed articular fractures performed in the “super healer” MRL/Mpj mice, that naturally produce extremely low levels of IL-1β and TNF-α, did not progress to PTOA even when the fractures were not reduced or internally fixed (Ward et al., 2008). Notably, MRL/Mpj mice did not display synovitis, their synovial fluid possessed lower levels of pro-inflammatory cytokines and higher levels of anti-inflammatory cytokines and their subchondral bone was unaltered. Thus, synovial inflammation plays a key role in subchondral bone pathology following IAF (Ward et al., 2008; Clark et al., 1998).

Taken together, these studies imply that the chondrocyte and synovial inflammatory responses that lead to subchondral bone alterations and PTOA following acute cartilage injury are at least impart incited by chondrocyte mitochondrial dysfunction, ROS release and apoptosis at the point of impact and vicinity (Fig. 1). Blocking these early cellular events in the chondrocytes is protective against PTOA as demonstrated in pre-clinical translational models of IAF. On the other hand, inhibiting synovial inflammation in the impacted joint could also restrict subchondral bone pathology and PTOA as a sequela of chondrocyte dysfunction, ROS release and death.

Fig. 1.

Schematic review of cross talk between cartilage, bone, and synovial tissue in the pathogenesis of PTOA in the setting of acute chondral injury. Acute supraphysiological loading of cartilage resulting in osteochondral fracture results in chondrocyte necrosis at the direct area of impact, and mitochondrial dysregulation and chondrocyte apoptosis in adjacent chondrocytes. This results in the release of inflammatory mediators such as IL-6 which stimulates remaining chondrocytes to release MMPs. Direct injury also causes damage to the synovial tissue leading to polarization of resident macrophages that release TNFα, IL1β, and IL-6, and MMPs into the synovial fluid. This along with the MMPs released from chondrocytes leads to cartilage degradation. These inflammatory mediators also cross into the subchondral bone leading to osteoclastogenesis and activation resulting in bone remodeling, and release of MMPs into the synovial fluid leading to further cartilage catabolism. These activities promote recruitment of monocytes and polarization of macrophages in the synovium. These interplaying mechanisms lead to increased bone subchondral bone remodeling, synovial inflammation, and cartilage catabolism which are hallmarks of PTOA.

3. Pathogenesis of PTOA – chronic aberrant loading of cartilage

3.1. Chondrocyte dysregulation in chronic aberrant loading

Joint injuries including IAF, and major ligamentous injuries can result in chronic changes in the prevailing trans-articular stress environment in a joint. For example, an ACL-t can result in PTOA secondary to habitually elevated shear stresses which leads to chondrocyte apoptosis, cartilage degradation, synovial inflammation, and subchondral bone remodeling (Arunakul et al., 2013; Heard et al., 2011; Maerz et al., 2021; Mevel et al., 2022; White et al., 2022; Liu et al., 2016; Tochigi et al., 2011). Similarly, changes in knee stresses after meniscal damage leads to loss of chondrocytes as early as 3 days post-operatively, and a complete loss of chondrocytes and cartilage degradation at later time points in small animal models (David et al., 2017). Aberrant loading of the joint surface is associated with elevated levels of inflammatory markers such as IL-1β and IL-6 released by synoviocytes and chondrocytes into the joint microenvironment and their downstream mediators such as inducible nitric oxide synthase (iNOS) and NO (He et al., 2018; Abramson, 2008). This is coupled with decreased expression of markers of cartilage anabolism (type II collagen (COL2), aggrecan), and increased levels of markers of cartilage catabolism (ADAMTS-4, ADAMTS-5, MMP-13, MMP-3) in chondrocytes (Liu et al., 2016; Huang et al., 2020) (Table 2). Even after the restoration of joint biomechanics through surgical restoration of joint stability, such as ACL reconstruction, MMP-13 levels remain elevated for a prolonged period.

Table 2.

Pathogenic mechanisms associated with chronic aberrant loading-induced PTOA.

PTOA etiology	Pathogenesis Mechanisms	References
Joint instability	Chronic aberrant loading – cartilage:
	-Altered joint biomechanics, expression of inflammatory mediators by chondrocytes leading to enhanced cartilage catabolism and diminished cartilage anabolism	(Arunakul et al., 2013; Heard et al., 2011; Maerz et al., 2021; Mevel et al., 2022; White et al., 2022; Liu et al., 2016; Tochigi et al., 2011; David et al., 2017; He et al., 2018; Abramson, 2008; Huang et al., 2020; Han et al., 2018)
	-Anti-inflammatory mechanisms: PLC-γ, CaMKK2, Nox4	(Wegner et al., 2019; Mevel et al., 2022; Takeuchi et al., 2021)
	-Therapeutics used in the clinic targeting inflammation	(Sherman et al., 2015; Bajpayee et al., 2017; Heard et al., 2015; Stefani et al., 2020; Takeuchi et al., 2021)
	Chronic aberrant loading – subchondral bone:
	-Uncoupled bone remodeling in early PTOA and uncoupled bone formation in late PTOA – osteophytes and cysts	(Jiang et al., 2021; Bertuglia et al., 2016; Couchourel et al., 2009)
	-Enhanced release of TGF-β1from bone matrix by osteoclasts, recruitment of osteoblasts, neovascularization	(Hayami et al., 2004; Hayami et al., 2006; Zhang et al., 2019; Janssens et al., 2005; Zhen et al., 2013; Cui et al., 2016; Fermor et al., 2005; Kean et al., 2016; Wu et al., 2020)
	-PDGF-BB and Netrin release by osteoclasts leading to angiogenesis and sensory nerve innervation of subchondral bone and cartilage	(Su et al., 2020; Zhu et al., 2019)
	-Osteoclast secreted molecules and exosomes	(Larrouture et al., 2021; Lofvall et al., 2018; Cherifi et al., 2021; Liu et al., 2021)
	-Osteoblast-chondrocyte crosstalk in aberrant loading-induced PTOA	(Lavigne et al., 2005; Jiang et al., 2022; Sun et al., 2022; Sanchez et al., 2005a; Sanchez et al., 2005b; Prasadam et al., 2012; Lin et al., 2018; Wu et al., 2021; Wu et al., 2016; Zhang & Wen, 2021)
	-Therapeutic targeting of osteoclast activity	(Disease-modifying effects of a novel cathepsin k inhibitor in osteoarthritis, 2020; Lampropoulou-Adamidou et al., 2014; She et al., 2017; Ziemian et al., 2021; Fernandez-Martin et al., 2021; Fernandez-Martin et al., 2020; Xing et al., 2016)
	Chronic aberrant loading – synovium:
	-Mechanisms mediating synovial macrophage activation - NLRP3/Inflammasome, TRPV-1 and TRPV-4, alarmins, TLR4, CaMKK2	(Mevel et al., 2022; Blom et al., 2007; Ebata et al., 2021; Arya et al., 2021; Escolano et al., 2021; Lv et al., 2021; Maruyama et al., 2019; O'Conor et al., 2016; Wojdasiewicz et al., 2014; van den Bosch et al., 2016; Cremers et al., 2017; Schelbergen et al., 2012)
	-Mechanisms in inflamed synovial fibroblasts – IL-6, Resistin, IL-17, VCAM1.

Meniscal damage, osteophyte formation and alterations in the subchondral bone also occur despite the restoration of joint mechanics (Narez et al., 2021). Patients who underwent ACL reconstruction developed subchondral thickening and sclerosis in the injured knee compared to their uninjured knees (Bhatla et al., 2018; Kroker et al., 2018). These changes in the subchondral bone were noted prior to any discernible changes in cartilage thickness (Bhatla et al., 2018). Other investigators have noted changes in loading and chondrocyte response to loads after ACL injury. Mice with bilateral ACL-t had increased loads across the tibiofemoral joint during gait, in contrast to mice with unilateral ACL-t had decreased loads across the injured limb during gait (Kotelsky et al., 2022). In both unilateral and bilateral ACL-t groups, chondrocytes in the early post-operative period were highly mechanosensitive demonstrating high levels of chondrocyte necrosis/apoptosis. However, at 8 weeks post-injury, only the chondrocytes isolated from the bilateral ACL-t groups had elevated apoptosis/necrosis to loading compared to controls (Kotelsky et al., 2022). In contrast to these reports, Li et al. (Li et al., 2022) investigated the role of regimented loading of the knee after ACL-t in a murine model and showed that mechanical loading was chondroprotective. Mechanical loading increased expression of the mechanoreceptor Piezo1 in ACL-t knees to control levels which aided in recruitment of stem cells to areas of damaged articular cartilage as well as promoted the expression of the chondrogenic differentiation marker SOX9 (Li et al., 2022). Aberrant mechanical loading following trauma also elicits changes in the synovium, as discussed below, by triggering inflammatory responses by synoviocytes and resulting in synovial tissue hyperplasia, immune cell infiltration and joint inflammation. Thus, changes in joint mechanical loading following injury or trauma detrimentally affects the subchondral bone, cartilage and synovium during PTOA pathogenesis.

The emerging theme from preclinical and clinical studies is that inflammation is a hallmark of chronic aberrant loading-induced PTOA and may augment chondrocyte sensitivity to aberrant loading. The amount of loading experienced by inflammation-primed chondrocytes appears to influence the preservation or degradation of cartilage. Increased inflammation also correlates with enhanced catabolic and inflammatory marker expression by chondrocytes (Fig. 2) (Heard et al., 2011; Huang et al., 2020; Han et al., 2018). Corticosteroids, such as dexamethasone, have been shown to mitigate the inflammatory response to TNF-α and IL-6, while decreasing GAG loss in cartilage and increasing proteoglycan synthesis (Lu et al., 2011). Currently, dexamethasone is used in clinical management of OA to alleviate pain. However, most of the traditional injections in clinic leave the joint compartment and enter the systemic circulation within hours to days, limiting their ability to reliably target chondrocytes (Habib, 2009). In addition, the high doses delivered from clinical injections lead to reductions in chondrocyte and synoviocyte viability and metabolism (Sherman et al., 2015). Recent advances have been made with low dose delivery vehicles of dexamethasone to damaged cartilage in preclinical models of ACL-t (Bajpayee et al., 2017) and acute injury (Heard et al., 2015; Stefani et al., 2020). When dexamethasone was delivered at a low dose via micro-particles, decreased cartilage catabolism was noted in a canine model of PTOA, but no changes were observed in inflammatory mediators at 1 month compared to controls (Stefani et al., 2020). In a rabbit model of PTOA via ACL-t or a drilled defect, intra-articular injection of dexamethasone immediately after surgery reduced inflammatory mediators and MMP-3, and partially protected against PTOA (Bajpayee et al., 2017; Heard et al., 2015). However, prolonged treatment (up to 3 weeks) leads to systemic toxicity (Bajpayee et al., 2017), which may limit its clinical utility as a chronically administered agent to prevent OA. Glucocorticoids directly affect bone cells and their chronic use to treat PTOA will detrimentally affect the subchondral bone.

Fig. 2.

Schematic review of cross talk between cartilage, bone, and synovial tissue in the pathogenesis of PTOA in the setting of chronic aberrant loading. Aberrant loading leads to increased shear stresses in the cartilage leads to the release of inflammatory mediators such as IL-6 and iNOS. Aberrant loading also leads to activation of mechanoreceptors in macrophages and fibroblasts in the synovial tissue leading to recruitment of monocytes to the synovium and polarization of macrophages in the synovium. Inflammatory mediators from macrophages, synovial fibroblasts and chondrocytes lead to production of MMPs from chondrocytes, macrophages and osteoclasts into the synovial fluid which results in cartilage catabolism. In addition, chondrocytes release RANKL which stimulates osteoclasts leading to an overall catabolic bone turnover. Osteoclasts also release PDGF-BB which recruits endothelial cells to the subchondral bone. This allows neovascularization of the subchondral plate. Further osteoclast activity leads to release of TGF-B1 from the osteoid matrix which stimulates osteoblastic bone formation. Osteoblasts enhance the activities of osteoclasts by secreting RANKL. These interplaying mechanisms lead to increased bone subchondral bone remodeling, synovial inflammation, and cartilage catabolism which are hallmarks of PTOA.

Colchicine is another common anti-inflammatory agent that is utilized in clinical practice to treat PTOA (Table 2). In murine DMM models, colchicine treatment decreased chondrocyte apoptosis and catabolic marker expression, and increased chondrocyte anabolic markers (Takeuchi et al., 2021). Colchicine was also shown to decrease MMP13 expression on human chondrocytes in vitro in response to IL-1β treatment (Takeuchi et al., 2021). Aberrant mechanical stress upregulates IL-1β and TNF-α signaling in chondrocytes, resulting in the phosphorylation of phospholipase C gamma (PLC-γ), which then interacts with tubulin and promotes Ca²⁺ release from the endoplasmic reticulum. Increase in intracellular Ca²⁺ promotes chondrocyte apoptosis and cartilage catabolism via MMP-13 upregulation while decreasing cartilage anabolism (Takeuchi et al., 2021), and these processes are blocked by colchicine.

As elevated Ca²⁺ is associated with chondrocyte apoptosis and cartilage catabolism, targeting Ca²⁺-stimulated signaling pathways may prevent PTOA. One such protein involved in intracellular Ca²⁺ signaling is Ca²⁺/calmodulin (CaM)-dependent protein kinase kinase 2 (CaMKK2), a serine/threonine protein kinase (Lu and Means, 1993; Dadwal et al., 2018). Mevel et al. (Mevel et al., 2022) recently reported CaMKK2 to be elevated in the articular cartilage of mice that underwent DMM and associated with elevated levels of inflammatory and catabolic marker expression by chondrocytes. In contrast, the genetic ablation or pharmacological inhibition of CaMKK2 protected against pathologies associated with PTOA. Specifically, CaMKK2-deficient chondrocytes were protected against IL-1β induced activation of the IL-6-Stat3-MMP13 pathway that mediates inflammation and cartilage catabolism (Mevel et al., 2022; Liu-Bryan, 2015). These studies show that Ca²⁺ signaling downstream of IL-1β is an integral part of chondrocyte inflammatory and catabolic pathways in chronic aberrant loading-induced PTOA.

In addition to inflammation, there is evidence of mitochondrial dysfunction and ROS after alteration of normal joint forces. In a rabbit DMM model, mitochondrial respiration was decreased in the load bearing medial femoral condyle and tibial plateau, and there was evidence of increased proton leakage at 4 weeks after surgery that preceded histological changes indicative of PTOA in the affected joint compartment (Goetz et al., 2017). Nox4, a member of a family of enzymes that generate ROS, is associated with elevated H₂O₂ production and PTOA phenotype in a mouse model of tibial compression-induced acute ACL-t. Genetically ablating or pharmacologically inhibiting Nox4 was protective against early joint changes in this model (Wegner et al., 2019). These studies suggest that mitochondrial dysregulation and ROS may contribute to PTOA in the setting of chronic joint instability either through secondary signaling or acute cellular injury like IAF (Fig. 2). These mechanisms and inflammatory signaling in chondrocytes could be therapeutically targeted to mitigate PTOA.

3.2. Subchondral bone-cartilage crosstalk in chronic aberrant loading

Subchondral bone is integral to the development of PTOA in joints subjected to aberrant joint loading. Cartilage catabolism and degradation alter stress transfer into the adjacent subchondral bone leading to changes in its structure. In PTOA, there is evidence of subchondral bone remodeling preceding changes in hyaline cartilage (Bhatla et al., 2018; Kroker et al., 2018; Fang et al., 2018; Hayami et al., 2004; Hayami et al., 2006; Intema et al., 2010; Lavigne et al., 2005; Sulaiman et al., 2021). In early PTOA, with abnormal loading, microfractures occur in the subchondral bone resulting in the uncoupling of bone remodeling leading to increased osteoclast-mediated bone resorption, which renders the subchondral bone plate porous and thinner (Jiang et al., 2021; Bertuglia et al., 2016). This increased porosity enables increased penetration of subchondral bone and calcified cartilage by blood vessels and sensory nerves. Later stages of OA are characterized by increased osteoblast-mediated uncoupled bone formation leading to thicker subchondral bone plate and bone sclerosis (Jiang et al., 2021). The matrix secreted by OA osteoblasts possess an abnormal α1 to α2 type I collagen (COL1) ratio that impairs mineral deposition, which contributes to bone cysts and osteophytes (Couchourel et al., 2009). Targeting the early abnormal subchondral bone remodeling is actively being pursued as a therapeutic option to ultimately protect against residual changes in loading ultimately leading to PTOA (Table 2 and Fig. 2).

Transforming growth factor β1 (TGF-β1), a cytokine with roles in bone and cartilage homeostasis, is secreted and deposited into the bone matrix in a latent form by osteoblasts (Zhang et al., 2019). It is activated and released by acids and MMPs secreted by osteoclasts during bone resorption, whereupon it promotes osteoprogenitor proliferation and differentiation (Zhang et al., 2019; Janssens et al., 2005). Increased subchondral bone remodeling by osteoclasts in early PTOA releases increased levels of TGF-β1, which in turn recruits osteoblasts that promote bone formation and sclerosis. Elevated TGF-β1 is found in the subchondral bone of human OA patients and in rodent models of PTOA (Zhen et al., 2013). Targeted deletion of TGF-β1 in Nestin-positive mesenchymal stem cells attenuated whereas its activation promoted ACL-t-induced PTOA in mice (Zhen et al., 2013). Intraarticular injection of halofuginone, a TGF-β1 inhibitor, attenuated subchondral bone disease and PTOA in mice and rats that underwent ACL-t (Cui et al., 2016). Whereas localized inhibition of TGF-β1 in the subchondral bone protects against PTOA, its systemically targeting is not beneficial as the cytokine plays a crucial role in normal cartilage homeostasis. Moreover, being a potent anti-inflammatory agent, its systemic inhibition triggers massive inflammation (van der Kraan, 2018).

TGF-β1 also promotes angiogenesis in the subchondral bone during early OA (Table 2). Osteoprogenitors recruited by TGF-β1 promote neovascularization by type H blood vessels that further impair bone remodeling and alter bone mineral deposition by osteoblasts, resulting in the formation of bone cysts and osteophytes. Additionally, this increased angiogenesis of the subchondral bone leads to the blood vessels breaching the tidemark, as often observed in animal PTOA models and human OA samples. Neovascularization of the normally avascular articular cartilage alters the oxygen tension of nearby chondrocytes, causing a decrease in cartilage proliferation, while promoting their hypertrophy and enhancing their expression of inflammatory markers. This leads to decreases in COL2 and proteoglycan production and increased cartilage catabolism (Hayami et al., 2004; Hayami et al., 2006; Fermor et al., 2005; Kean et al., 2016; Wu et al., 2020). Neovascularization by H-type vessel formation has been shown to be mediated by focal adhesion kinase (FAK). Treatment with a FAK inhibitor in the subchondral bone of rats undergoing ACL-t attenuated subchondral bone remodeling, MMP-13 expression, and cartilage degradation (Wu et al., 2020).

Other studies have explored the molecular mechanisms underlying enhanced subchondral bone vascularization in early PTOA and found osteoclasts to play a major role. Mononuclear osteoclasts secrete enhanced levels of platelet-derived growth factor -BB (PDGF-BB) that activate PDGF-receptor β in pericytes to promote subchondral bone angiogenesis during initial stages of DMM-induced PTOA, and this neovascularization is visible even before cartilage degradation is apparent (Su et al., 2020). Targeted deletion of PDGF-BB in pre-osteoclasts attenuated subchondral bone angiogenesis and joint degradation in this model of PTOA (Su et al., 2020).

Along with angiogenesis, perivascular calcitonin gene-related peptide (CGRP) expressing sensory and sympathetic nerve fibers increasingly innervate the OA subchondral bone, often reaching the cartilage and causing pain in OA (Table 2). Magnetic resonance imaging studies indicate correlation between the number of subchondral bone marrow lesions and OA pain severity (O'Neill and Felson, 2018). Osteoclasts play a prominent role in the sensory nerve innervation of the subchondral bone in ACL-t-induced PTOA by secreting Netrin-1, an axon guidance molecule that promotes subchondral bone innervation. The conditional deletion of Netrin-1 or its receptor DCC (deleted in colorectal cancer) in osteoclasts attenuated sensory innervation of the subchondral bone and PTOA pain (Zhu et al., 2019). Moreover, inhibition of osteoclasts using the bisphosphonate alendronate or by conditionally deleting receptor activator of nuclear factor κ-B (NFκ-B) ligand (RANKL) in osteocytes attenuated these pathologies (Zhu et al., 2019).

Prostaglandin E2 (PGE2), produced by the cyclooxygenase 2 (COX-2) gene and secreted by osteoblasts, inflamed chondrocytes, and macrophages, is highly elevated in the OA subchondral bone (Mevel et al., 2022). Recent work by Jiang et al. (Jiang et al., 2022) shows that osteoblast-derived PGE2 acts via E prostanoid 4 (EP4) receptors in osteoclasts to promote PDGF-BB and Netrin-1 secretion by osteoclasts, type H blood vessel sprouting, and CGRP-positive sensory innervation of the subchondral bone, leading to pain hyper-sensitization and PTOA development in mice that underwent ACL-t. Oral administration of EP4 antagonist HL-43 attenuated subchondral bone angiogenesis and sensory innervation leading to reduced pain in this model (Jiang et al., 2022). Sun et al. (Sun et al., 2022) recently reported that whereas PGE2 is elevated in the OA subchondral bone, the conditional deletion of sensory nerve-specific EP4 attenuated subchondral bone sclerosis, innervation and PTOA in mice. These findings are especially significant as COX-2 inhibitors such as NSAIDs are routinely used in the clinic to alleviate PTOA pain but cause severe side-effects with long-term use in patients.

Many osteoclast-secreted molecules influence cartilage homeostasis. Osteoclasts release MMP-8 and MMP-9 when they are in contact with articular cartilage causing cartilage degradation in PTOA (Larrouture et al., 2021; Lofvall et al., 2018). They also secrete sphingosine 1-phosphate (S1P), a ceramide metabolite that increased MMP3 and MMP13 expression in chondrocytes (Cherifi et al., 2021). Conditional deletion of S1P in myeloid cells or intraarticular injection of the S1P neutralizing antibody sphingomab suppressed MMP13 expression by chondrocytes and attenuated cartilage catabolism and PTOA progression (Cherifi et al., 2021). Liu et al. (Liu et al., 2021) reported increased circulation of osteoclast-derived exosomes containing miRNAs in initial stages of ACL-t-induced PTOA in mice. Exosomal transfer of these osteoclast-derived miRNAs to chondrocytes suppressed their expression of tissue inhibitor of metalloproteases (TIMP-2 and TIMP3), which inhibit cartilage degradation. Conversely, conditional deletion of the key miRNA processing enzyme Dicer in osteoclasts, blocking exosome secretion by silencing ras-related protein 27a (Rab27a), or systemic administration of an osteoclast-targeted exosome inhibitor termed OCExoInhib attenuated PTOA progression in mouse models of surgically induced OA, revealing a novel therapeutic approach of targeting osteoclast-mediated exosomal transfer of TIMP-suppressing miRNAs (Liu et al., 2021). More recently, Zhao et al. (Zhao et al., 2022) reported that osteoclast-derived leukemia inhibitory factor (LIF) contributes to abnormal bone remodeling in early ACL-t-induced PTOA in mice.

Osteoblasts also have been shown to alter chondrocyte protein expression in models of PTOA. Increased levels of IL-6, NO, and MMPs are found in PTOA subchondral bone tissue (Lavigne et al., 2005). Chondrocytes co-cultured with osteoblasts from subchondral bone in PTOA display decreased SOX9 and anabolic protein expression. Subchondral bone osteoblasts increase expression of cartilage catabolic proteins via ERK1/2 and PI3K/AKT signaling pathways (Sanchez et al., 2005a; Sanchez et al., 2005b; Prasadam et al., 2012; Lin et al., 2018) Crosstalk may also be occurring via exosomes carrying miRNA from subchondral bone osteoblasts eliciting a decrease in cartilage anabolism and an increase in cartilage catabolism (Wu et al., 2021). Additionally, enhanced levels of sclerostin secreted by osteocytes in early-stage OA supports osteoclastogenesis that promote excessive subchondral bone remodeling, angiogenesis and innervation leading to cartilage degradation and PTOA pain (Wu et al., 2016; Zhang and Wen, 2021).

As mentioned earlier, cartilage degradation in early PTOA alters stress transfer into the subchondral bone, triggering rapid osteoclast activity, neovascularization, and innervation of the bone hastening disease progression. Therapeutically targeting subchondral bone osteoclasts could be beneficial for halting PTOA and the amelioration of associated joint pain. A randomized, double-blind, placebo-controlled phase 2a human clinical trial using the cathepsin K inhibitor MIV-711 showed significantly reduced OA subchondral bone and cartilage pathologies in patients, although it was not effective against pain more than placebo (Disease-modifying effects of a novel cathepsin k inhibitor in osteoarthritis, 2020). Hayami et al. (Hayami et al., 2004) found that treatment with the bisphosphonate alendronate in an ACL-t model reduced subchondral bone remodeling, osteoclast recruitment, and neovascularization in a murine model. Similarly, zoledronic acid treatment also decreased proteoglycan, cartilage degradation, and subchondral bone remodeling in a rabbit model with short duration of treatment (Lampropoulou-Adamidou et al., 2014; She et al., 2017). Ziemian et al. (Ziemian et al., 2021) demonstrated that the timing of bisphosphonate therapy influenced the ability to mitigate the progression of PTOA in a murine model. They found that treatment immediately after surgery with bisphosphonate therapy mitigated subchondral bone remodeling, osteophyte formation, and cartilage degradation, whereas delaying treatment for 2 weeks led to minimal protection against PTOA (Ziemian et al., 2021).

In contrast to these findings, long term systemic treatment of rabbits with bisphosphonates did not protect against subchondral bone remodeling or PTOA severity (Fernandez-Martin et al., 2021; Fernandez-Martin et al., 2020). This indicated that bisphosphonate therapy may be better utilized as a pre-emptive or early therapy against developing PTOA rather than mitigating further progression of established PTOA. A meta-analysis of randomized controlled human clinical trials indicates that bisphosphonate therapy may be effective in relieving pain and stiffness associated with OA, thus accelerating functional recovery from the disease, but not in preventing OA progression (Xing et al., 2016). Emerging evidence from mechanistic investigations in preclinical models indicate that the timing and duration bisphosphonate treatment will be critical for the alleviation of OA-associated pain and attenuating disease progression. Still, more research is needed to fully understand how the crosstalk between bone cells and chondrocytes influences PTOA.

3.3. Synovium-cartilage crosstalk in chronic aberrant loading

In addition to osteochondral alterations, substantial changes occur in the synovium during PTOA associated with chronic instability (Fig. 2). Macrophages and fibroblasts are the two major cell types in the synovium. Macrophages lining the synovial membrane serve an integral role in joint homeostasis by phagocytosing and clearing ECM components and other joint debris produced during normal activity, through their expression of scavenger receptors such as CD163 and Triggering Receptor Expressed on Myeloid Cells 2 (TREM2) (Ambarus et al., 2012; Kurowska-Stolarska and Alivernini, 2017). Synovial fibroblasts, the main stromal cells of the joint, play a major role in joint lubrication by synthesizing components of the synovial fluid such as the glycosaminoglycan hyaluronic acid and the mucous glycoprotein lubricin or proteoglycan-4. They also play important roles in maintaining synovial macrophage homeostasis and cartilage integrity during normal joint loading (Lefevre et al., 2015). However, PTOA is associated with increased inflammatory macrophage infiltration of the synovial tissue, tissue hyperplasia and inflammation (Mevel et al., 2022; Takeuchi et al., 2021; Chang et al., 2021; Lin et al., 2021; Rzeczycki et al., 2021; Thomas et al., 2017; Gilbert et al., 2018; Liao et al., 2020). Mechanisms underlying synovial pathology in PTOA are extensively reviewed elsewhere (Lieberthal et al., 2015; Evers et al., 2022; Khella et al., 2021), and will only be briefly discussed here.

Cartilage ECM components such as aggrecan released in increased numbers after chronic or acute trauma to the cartilaginous surface polarize synovial macrophages via NOD-, LRR- and pyrin domain-containing protein 3 (NLRP3)/inflammasome mechanism (Blom et al., 2007; Ebata et al., 2021). Macrophages respond to aberrant mechanical loading stimuli through transient receptor potential vanilloid type receptors (TRPV) 1 and 4, which stimulate their polarization through the NLRP3/inflammasome mechanism (Arya et al., 2021; Escolano et al., 2021; Lv et al., 2021; Maruyama et al., 2019). Modulation of TRPV-1 and TRPV-4 receptor activity attenuated OA severity in a murine model (O'Conor et al., 2016). Activated macrophages produce IL-1β, which stimulates chondrocytes to express CX3CL1 to further attract peripheral monocytes with a pro-inflammatory phenotype to the synovium (Wojdasiewicz et al., 2014). In addition, alarmins S1008 and S100A9 released from the damaged cartilage enable the mobilization of pro-inflammatory monocytes to the synovium via CCL2 signaling (van den Bosch et al., 2016; Cremers et al., 2017). S100A8 and S100A9 exert an autocrine effect on chondrocytes to enable enhanced expression of catabolic markers and diminished expression of anabolic markers via toll-like receptor 4 (TLR4) signaling (Schelbergen et al., 2012). Activated macrophages also produce MMPs and Flightless-1 (Fli-1), which promote chondrocyte hypertrophy and cartilage degradation while decreasing cartilage anabolism via the Fli-1-TLR4-ERK1/2 pathway in chondrocytes (Blom et al., 2007; Ebata et al., 2021). CaMK family members including CaMKK2 play key roles in the regulation of macrophage responsiveness to TLR4 signaling (Racioppi et al., 2012). Indeed, CaMKK2 expression is elevated in inflamed synovial macrophages whereas its pharmacological inhibition or genetic ablation mitigated synovial inflammation and hyperplasia in a murine DMM-PTOA model (Mevel et al., 2022).

Synovial fibroblasts become activated in OA and play a major role in joint inflammation and cartilage degradation through the release of proinflammatory factors such as IL-6 (Ramirez-Perez et al., 2022; Pearson et al., 2017; Yang et al., 2017). In addition, CX3CL-1 released from inflamed chondrocytes stimulates MMP production in synovial fibroblasts via NFκ-B and Wnt pathways (Hou et al., 2017; Lietman et al., 2018). Synovial fibroblasts express adhesion molecules such as vascular cell adhesion molecule type 1 (VCAM-1), which become elevated in PTOA via Resistin and IL-17 mediated pathways (Kalichman et al., 2011; Liu et al., 2013; Wu et al., 2022). Knockdown of Resistin attenuated PTOA in a murine ACL-t model through reductions in VCAM-1 positive fibroblasts and CD68⁺ monocytes in the synovium (Chen et al., 2020). Additionally, activated fibroblast-like synoviocytes contribute to synovial fibrosis in OA, as reviewed elsewhere (Zhang et al., 2021; Maglaviceanu et al., 2021).

In summary, PTOA severity is mediated, in part, by the polarization of resident macrophages to a pro-inflammatory state, recruitment of pro-inflammatory monocytes to the synovial tissue, and production of inflammatory markers and MMPs from polarized macrophages and synovial fibroblasts leading to cartilage catabolism (Table 2 and Fig. 2).

4. Clinical therapeutic trials addressing PTOA

Several therapeutic targets for OA and PTOA identified in preclinical studies have made their way to clinical trials. An IL-1 receptor antagonist was trialed in the treatment of symptomatic OA of the knee but did not alleviate pain versus placebo (Cohen et al., 2011). There are several ongoing trials investigating novel therapeutics to prevent or mitigate PTOA. One double-blind randomized controlled trial that has begun recruiting is assessing the efficacy of intra-articular administration of amobarbital to prevent PTOA in patients that sustained a tibial pilon fracture based on success in a translational model (NCT04589611) (Coleman et al., 2018). Another trial investigating the efficacy of intra-articular dexamethasone to prevent PTOA after distal radius fracture has been completed as of 2020, but no results from the trial are available to date (NCT02318433). PTOA may develop even after restoration of joint mechanics and stability secondary to a pro-inflammatory state (Heard et al., 2011). As such, the therapeutic potential of the cysteinyl leukotrienes inhibitor, montelukast, to prevent PTOA after ACL reconstruction is being investigated in clinical trials (NCT04572256). Another study is current recruiting and evaluating the effects of the anti-fibrinolytic agent, tranexamic acid (TXA). TXA binds to plasminogen and prevents the formation of plasmin. There is a current human trial evaluating if treatment with TXA reduces intra-articular hemorrhage, inflammation, neovascularization and PTOA development after ACL injury (NCT03552705).

5. Conclusion and future directions

PTOA is a multifactorial disease that occurs after an acute impact injury to the cartilage or due to adverse changes in chronic loads resulting from incongruity, instability and malalignment. Treatment options have been remarkably stagnant over decades and have essentially entirely focused on surgical restoration of joint congruity and stability. Collectively, surgical options have clearly reached an efficacy limit as rates of PTOA remain high regardless of improvements in surgical methods. Accordingly, investigators are aggressively pursuing interventions directed at treating the acute mechanical damage to the cartilage. Evidence over the past 2 decades has shown that the pathophysiology of acute injury to the chondral surface is partially mediated by mitochondrial dysfunction in response to elevated oxygen tension and ROS generation, leading to chondrocyte apoptosis and cartilage catabolism and associated progression of inflammation (Fig. 1). Recent mitoprotective interventions have shown promise in preventing PTOA in the preclinical setting, and amobarbital is under clinical investigation as a therapeutic agent to prevent PTOA. Pathophysiologic events secondary to adverse chronic changes in joint loading leading to PTOA have become better defined opening other therapeutic opportunities to prevent PTOA after chronic aberrant loading. There is substantial crosstalk between the tissues involved in PTOA (Fig. 2). Osteoblasts and osteoclasts contribute to elevated cartilage catabolism and decreased anabolism via signaling to chondrocytes in PTOA. Bisphosphonate therapies have shown promise in alleviating bone disease and OA pain in preclinical models and in the clinic. Activated synovial macrophages release inflammatory factors such as IL-6 that further exacerbate cartilage catabolism and suppress anabolism. Several clinical studies are underway investigating anti-inflammatory agents in preventing PTOA.

Much future research remains be done investigating the role of different joint tissues in the pathogenesis of PTOA and determining which one or combination of tissues, resident cells and mechanisms would be the optimal targets to prevent PTOA. Although there are similarities in the pathogenesis of PTOA resulting from acute intra-articular injury and chronic aberrant loading, the contribution of mitochondrial and inflammatory mediators appears to differ between the two etiologies. Thus, IAF results in an acute initial effect of chondrocyte apoptosis and/or necrosis at the point of impact, which triggers mitochondrial dysfunction and ROS generation, leading to further chondrocyte apoptosis and cartilage catabolism. In addition, chondrocyte apoptosis may diminish anabolic function to support regional matrix production and turnover. This then stimulates inflammation in chondrocytes and synovium, eventually also affecting subchondral bone remodeling (Fig. 1). On the other hand, PTOA originating from chronic aberrant loading of the joint following an injury such as ACL-t or PCL-t occurs primarily via an inflammatory route that leads to global joint alteration affecting subchondral bone, synovium and cartilage (Fig. 2). These differences could help identify unique targets for treatment of both PTOA etiologies.

Funding

This work was supported by DoD Peer Reviewed Medical Research Program – Investigator-Initiated Research Award W81XWH-20-1-0304 from the U.S. ARMY MEDICAL RESEARCH ACQUISITION ACTIVITY, a Comprehensive Musculoskeletal T32 Training Program from the NIH (AR065971), the Indiana Clinical and Translational Sciences Institute which is funded in part by National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Sciences Award (Award Number UL1TR002529), and the National Institutes of Health – NIAMS R01AR076477.

CRediT authorship contribution statement

Julian Emerson Dilley: Conceptualization, Methodology, Investigation, Writing- Original draft preparation, and Writing - Reviewing and Editing. Margaret Anne Bello: Writing - Original draft preparation, Writing - Reviewing and Editing, and Investigation. Todd McKinley: Writing- Reviewing and Editing. Natoli Roman: Writing- Reviewing and Editing. Uma Sankar: Conceptualization, Writing- Reviewing and Editing, Supervision, and Project administration.

Declaration of competing interest

Julian Emerson Dilley, Margaret Anne Bello and Uma Sankar have no financial disclosures or conflicts of interest in relation to this study. Roman Natoli receives consultation fees from Quince. Todd McKinley receives royalties from Innomed.

Data availability

The data that has been used is confidential.