Angiogenesis, the growth and development of new BVs from extant BVs, is a well-studied and characterized process (Carmeliet and Jain, 2011). While vasculogenesis refers to the development of the heart and basic vascular structures (Patan, 2004), angiogenesis is the process by which this network expands and matures into a fully functional system. The primary driver of angiogenesis is VEGF-A, which binds to VEGF receptors 1 and 2 (VEGFR-1/2) and induces vasodilation and cellular permeability. A detailed overview of the protein families and molecular mediators implicated in angiogenesis can be found in the review by Carmeliet and Jain (2011). Lymphangiogenic progression follows growth patterns like those observed in angiogenesis, with some key distinctions (Adams and Alitalo, 2007; Lohela et al., 2009). Similar to VEGF-A being the primary factor behind angiogenesis, VEGF-C is the driving factor for lymphangiogenesis, with VEGF-D exhibiting a lesser effect. A detailed mechanism of lymphangiogenesis is expertly reviewed by Oliver et al. (2020).

The lymphatic system plays an important role in both the vascular and immune systems. It regulates and returns fluid to the venous system while also transporting immune cells to appropriate locations. Lymph nodes are key in antigen presentation and generating humoral and cell-mediated immune responses. Although necessary for healthy tissue homeostasis, the development of lymphatic vasculature can also contribute to immunopathologies, such as chronic autoimmune disease (Kesler et al., 2013). In such chronic inflammatory states, lymphangiogenesis can create tertiary lymphatic organs (TLOs), which are very similar to normal lymph nodes. TLOs have been found in organ transplant recipients and corneal transplant recipients as well as in patients with rheumatoid arthritis and inflammatory bowel disease (Kesler et al., 2013). Pathologic lymphatic drainage is also a mechanism for tumor metastasis (Kesler et al., 2013). Tumor cells can induce lymphangiogenesis to connect to existing lymphatic vasculature, thereby increasing the risk of metastasis (Karpanen et al., 2001; Skobe et al., 2001). Although tumor-induced lymphangiogenesis may result in dysfunctional intratumoral lymphatics, Padera et al. (2002) demonstrated that VEGF-C–induced lymphangiogenesis promotes metastasis through the functional lymphatics formed at the tumor border. Unlike treatments targeting BV-mediated metastasis, anti-lymphangiogenic therapy is still underdeveloped clinically (Kesler et al., 2013; Yamakawa et al., 2018).

Blood vessel regression is a complex, multifaceted process that is still poorly understood compared to angiogenic growth (Korn and Augustin, 2015). BV regression has been studied in the context of vessel pruning, which is the process of microvessel regression during BV maturation and is known to involve a diverse set of molecular markers and pathways. Korn and Augustin described multiple mechanisms of BV regression (Figure 1), with VEGF, Wnt/Notch, angiopoietin (Ang)/tyrosine kinase with immunoglobulin-like and EGF-like domains (TIE), and other signaling pathways playing key roles (Korn and Augustin, 2015). Claxton and Fruttiger (2003) used an induced hyperoxia mouse model in which VEGF was downregulated to demonstrate that FGF2, ANG2, platelet-derived growth factor (PDGF), and Delta-like canonical Notch ligand 4 (DLL4) are involved in maintaining vessel stability independent of VEGF. A mouse model study of Parkinson’s disease focused specifically on the vascular pathologic effect, and VEGF was shown to maintain mature vessels and induce plasticity in response to external factors (Elabi et al., 2021). Korn et al. (2014) demonstrated how the Wnt/Notch signaling pathway also has a key role in BV maturation and stabilization (Korn and Augustin, 2015), with improper signaling potentially resulting in inducing spontaneous vessel regression and death. DLL4 expressed by ECs can bind with Notch as well and downregulate VEGF, a key signaling event for capillary remodeling (Lobov et al., 2011). As Notch ligand interactions are primarily involved in cell death and apoptosis (Miele and Osborne, 1999), there appear to be multiple approaches to induce cell apoptosis. These pathways occur concurrently with the Ang/TIE2 pathway, which regulates vessel plasticity, and are crucial for vessel pruning and ending the vessel maturation process (Welti et al., 2013). Other pathways include Hairy/enhancer-of-split related with YRPW motif protein (HEY)/p53, cysteine-rich transmembrane BMP regulator 1 (CRIM1), and C-X-C motif chemokine ligand 10 (CXCL10)/C-X-C motif chemokine receptor 3 (CXCR3), which increase VEGFR-1, stimulate VEGF-A signaling, and detach ECs to prepare for apoptosis, respectively, all of which are key processes in BV pruning and regression (Bodnar et al., 2009; Fan et al., 2014). Researchers have observed a unique pattern in these complex pathways, termed the “anti-angiogenic switch,” which represents a modified form of the “angiogenic switch” concept from tumor vasculature. Essentially, the mere addition or inhibition of VEGF does not necessarily change the angiogenic direction. Gosain et al. (2006) assessed the abilities of exogenous VEGF, FGF, PDGF, and a combination of VEGF and FGF to reverse vascular regression. They only observed a transient increase in vessel density, and regression continued after an observed period. Conversely, Irani et al. (2017) tested the ability of an anti-VEGF-B antibody fragment to induce BV regression and found that the antibody fragment alone could not significantly reduce BV density. These findings indicate that signaling pathways are not simply “pro” or “anti” angiogenesis; instead, a myriad of concurrent signals with a basal level of VEGF make angiogenic regression a resilient state (Elabi et al., 2021).

Physiological BV regression occurs to facilitate the maturation of developing blood vessels and to prevent excessive BV growth (Korn and Augustin, 2015). There are many instances in which blood vessels regress almost completely in healthy states. Hyaloid vessels, which are present in the eye during embryonic development, completely regress after complete eye development, primarily through pathways involving Wnt signaling (Wang et al., 2019a,b). The menstrual cycle also presents a unique BV growth and regression cycle; during luteolysis of the ovarian corpus luteum, Goede et al. (1998) observed an association between upregulated Ang2 expression (relative to Ang1) and BV regression. Another unique physiological example of BV regression is in the context of embryological finger patterning, a process involving apoptosis of embryologic interdigital webbing to form free digits (Hurle et al., 1985). Prior to physiologic apoptosis of interdigital webbing, robust interdigital angiogenesis increases the local oxygen concentration, resulting in the generation of reactive oxygen species (ROS) that impart cellular damage (Eshkar-Oren et al., 2015; Cordeiro et al., 2019). This oxidative stress culminates in interdigital webbing apoptosis, which is followed by remodeling and regression of the interdigital vasculature. Pathological BV regression has been seen in a wide variety of conditions. Retinopathy of prematurity (Keshet, 2003), pulmonary fibrosis (Ebina et al., 2004; Parra et al., 2005; Johnson and DiPietro, 2013; Barratt and Millar, 2014), chronic kidney disease (Polichnowski, 2018), kidney microvessel rarefaction (Goligorsky, 2010; Chade, 2013), Parkinson’s disease (Elabi et al., 2021), spinal cord injury (Trivedi et al., 2016), and initial tumor BV co-option (Holash et al., 1999) have all been linked to BV regression during specific pathological time frames (Table 1).

The phenomenon of LV regression is not well understood compared to BV regression. The exact mechanisms responsible for LV regression are not well defined, but TH1- and TH2-derived cytokines have been shown to play a role in negative regulation of lymphangiogenesis (Shao and Liu, 2006; Kataru et al., 2011; Savetsky et al., 2015). LV regression has been observed in experimental settings. In a study by Zhang et al. (2011) lymphatic regression was found to be transient as vessels eventually grew back, but this is not true for all LV regression states. Shi et al. (2020) observed LV regression due to aqueous humor in both healthy and inflamed states. They linked alpha-melanocyte–stimulating hormone (a-MSH), vasoactive intestinal peptide (VIP), thrombospondin-1 (TSP-1), transforming growth factor beta (TGF-β), and Fas ligand (FasL) found in aqueous humor to LV regression, but the roles of these factors in induced LV regression have yet to be confirmed (Shi et al., 2020).

Similar to LV regression mechanism knowledge, the knowledge of LV functions in both physiology and pathology is limited. LV regression in the endometrium has been linked to the menstrual cycle (Tomita and Mah, 2014), as Girling and Rogers found fewer vessels in the functionalis layer and reduced vessel density in the subepithelium in a pig uterus model (Girling and Rogers, 2005). Tomita and Mah (2014) demonstrate that LVs in the functionalis layer undergo a cycle of “proliferation and degeneration” that is synchronized with cyclic menstrual changes of endometrial arteries. Additionally, it has been noted that embryologic pulmonary LVs undergo physiological regression, with a persistence of these vessels implicated in the pathophysiology behind pulmonary lymphangiectasia (Esther and Barker, 2004). As mentioned above, promotion of LV regression was also observed in relation to eye development and aqueous humor (Shi et al., 2020). Although limited, these physiological and pathological examples of LV regression can be probed and further dissected to potentially yield context-specific information on the mechanisms of LV regression.

The basal state of BVs is one of quiescence and homeostasis. Stabilized networks are remarkably unperturbed by both pro- and anti-angiogenic factors various protective mechanisms ensure that a vascular network does not undergo inappropriate pruning or proliferation. These stabilizing factors include pericyte coverage (Bergers et al., 2003), vascular endothelial (VE)-cadherin cell–cell junctions (Murakami, 2012), and autocrine VEGF-A signaling (Lee et al., 2007). Factors promoting either homeostasis or BV regression are illustrated in Figure 2. For a more detailed summary of the factors involved in BV homeostasis and quiescence, see the review by Murakami (2012). The signaling required to maintain BV homeostasis is a result of upstream phenomena. For example, HIF-1 alpha accumulates in cells exposed to hypoxic conditions and consequently leads to the transcription of pro-angiogenic genes including VEGF-A (Fallah and Rini, 2019). Absence of this upstream requirement for homeostasis can lead to dramatic BV regression. For example, in retinopathy of prematurity, administration of high concentration oxygen to premature infants obliterates the local hypoxic signaling required to maintain the immature retinal vessels, leading to pathological BV regression (Keshet, 2003). Another upstream phenomenon that promotes BV survival signaling is shear stress imparted by laminar blood flow in a properly perfused vessel (Korn and Augustin, 2015).

Extending these findings to LVs, it is certain that these vessels similarly experience pro-survival signaling, the disruption of which can lead to LV regression (Figure 3). Although the LV network experiences less shear stress than the BV network, Stritt et al. (2021) showed that LVs have greater sensitivity to shear stress and further elucidated the mechanisms by which shear stress may lead to LEC proliferation in vitro. In addition to LEC proliferation and potential survival, shear stress has also been implicated in the survival of lymphatic valves (Iyer et al., 2020). Although not as clear as in the BV context, shear stress signaling on a perfused LV has been seen to play a role in LV proliferation, and further studies may evaluate its role in LV regression. Additionally, similar to the angiogenic context, local hypoxia and subsequent HIF signaling induces lymphangiogenesis through increased expression of VEGF-C and VEGF-D (Ji, 2014; Schito, 2019). This commonality between angiogenesis and lymphangiogenesis may allude to normoxia as a survival signal for both BVs and LVs. The murine oxygen-induced retinopathy model is a tool that allows for the study of BV regression under hyperoxic conditions. Because the retina is physiologically devoid of LVs, another model must be utilized to study the effects of hyperoxia on LV regression. As our laboratory has demonstrated previously, Prox1-green fluorescent protein (GFP)/Flt1-DsRed transgenic mice can be utilized as a model to observe the physiologic LV regression that occurs postnatally in mice (Zhong et al., 2017). Application of this research tool in conjunction with administration of high concentration oxygen may allow for elucidation of the effects of hyperoxia on LV regression.

In contrast to BVs’ dependence on autocrine VEGF-A, most LVs do not seem to require continuous VEGF-C signaling for homeostasis; intestinal and meningeal lymphatics are cited as exceptions (Stritt et al., 2021). Thus, it is likely that there are more nuances left to be unearthed in lymphatic vessel maintenance signaling. As evidenced by the example of retinopathy of prematurity in the BV context, further characterization of these LV survival signals will allow for an increased understanding of LV regression.

Inflammation-induced angiogenesis has been studied in a variety of contexts, including herpes simplex virus 1–induced corneal neovascularization, helicobacter pylori infection, cancer, and inflammatory bowel disease (Sajib et al., 2018; Filiberti et al., 2020). Sajib et al. (2018) described the intimate relationship between inflammation and angiogenesis, citing the dual role inflammatory markers such as IL-8 and cyclooxygenase-2 may play in both inflammation and promoting angiogenesis. As inflammatory markers may induce angiogenesis, it follows that inhibition of inflammation may block angiogenesis, and indeed, Folkman described this phenomenon as early as Folkman et al. (1983). Summarized by Logie et al. (2010), the literature is rich with examples of glucocorticoids, potent inflammatory inhibitors, preventing inflammation-induced angiogenesis.

Inflammation has also been implicated in lymphangiogenesis (Figure 4; Cursiefen et al., 2006; Kelley et al., 2011; Liao and von der Weid, 2014), and glucocorticoids have been shown to inhibit inflammation-induced lymphangiogenesis (Steele et al., 2011; Martínez-Corral et al., 2012). Although Steele et al. (2011) demonstrated that glucocorticoids inhibit corneal lymphangiogenesis, they found that this anti-inflammatory treatment does not induce LV regression. However, their definition of regressing vessels was determined visually and perhaps further speaks to the lack of a standardized definition for LV regression (Steele et al., 2011). Additionally, this work utilized corneal removal and fixation, therefore necessitating comparison of samples from different mice; a similar methodology using transgenic mice with fluorescent lymphatic detection would allow for continuous live-imaging and determination of the fate of lymphatic vessels, thereby allowing for verification of LV regression. Another study found that upon resolution of inflammation, newly proliferated lymph node LVs do indeed undergo regression (Mumprecht et al., 2012). Further research of the effects of both inflammation resolution and anti-inflammatory therapy on LV regression are warranted and may provide insights into novel mechanisms of LV regression.

In the context of wound healing, Gosain et al. (2006) described that after robust angiogenesis, exogenous application of pro-angiogenic stimuli, such as VEGF, FGF, and PDGF, is insufficient for preventing the BV regression characteristic of wound healing resolution. Thus, it follows that there must be a summation of anti-angiogenic stimuli that supersede the effects of pro-angiogenic molecules in physiologic BV regression. This concept, termed the anti-angiogenic switch, refers to the negative feedback mechanism by which VECs may be protected from excessive angiogenesis and VEGF stimulation. This notion was referred to as early as 1996 when Lingen et al. (1996) described the ability of retinoic acid to “switch [cultured oral squamous cell carcinomas] from an angiogenic to an anti-angiogenic phenotype”. Over time, proteins in the vasohibin and mammalian Sprouty families have emerged as negative feedback modulators of angiogenesis (Watanabe et al., 2004; Sato and Sonoda, 2007; Wietecha et al., 2011). This research provides evidence for the existence of endogenous mediators that through a negative-feedback mechanism inhibit angiogenesis and prime BVs for physiological regression.

Negative feedback mechanisms are pervasive in biology, as they prevent overstimulation and excessive action. Within the context of inflammation-associated lymphangiogenesis in regional lymph nodes, Kim et al. (2014) described some of the antilymphatic signaling pathways that occur upon inflammation resolution. The authors describe a balancing act between prolymphangiogenic molecules, derived from intranodal follicular B cells, CD11b macrophages, and fibroblast-type reticular stromal cells, and antilymphatic molecules, derived from T lymphocytes. This dynamic tug-of-war accounts for the ability of the lymphatic vasculature to adapt to the situation at hand, with prolymphangiogenic molecules dominating the early phase of inflammation and antilymphatic molecules dominating the resolution of inflammation.

As described by Kim et al. (2014) several antilymphatic molecules derived from T lymphocytes have been identified. Shao and Liu (2006) utilized a pig thoracic duct assay to demonstrate that interferon-α and interferon-γ treatment cause a dose-dependent decline in LEC proliferation in vitro; the authors hypothesize this effect may be through the induction of apoptosis (Shao and Liu, 2006). A 2011 study similarly implicated interferon-γ signaling as a main driver in lymph node LV regression (Kataru et al., 2011); through the transfer of T cells to athymic nude mice, demonstrated the role of TH1 cell signaling as a negative feedback regulator of lymph node LV formation. In addition to the effects of TH1 interferon signaling, studies published in 2015 asserted the anti-lymphangiogenic function of TH2 cytokine signaling, with IL-4 and IL-13 as the TH2-derived cytokines implicated in negative regulation of lymphangiogenesis (Savetsky et al., 2015; Shin et al., 2015). Two other studies in 2008 confirmed that TGF-β also serves as a negative regulator of lymphangiogenesis (Clavin et al., 2008; Oka et al., 2008). While the story of negative regulation of lymphangiogenesis is likely incomplete, TH1-derived interferon signaling, TH2-derived IL-4 and IL-13 signaling, and TGF-β signaling all play roles in inhibiting lymphangiogenesis.

The role of apoptosis in regressing BVs seems to be context dependent. Based on high-resolution imaging of murine cornea and zebrafish models, Franco et al. (2015) concluded that BV regression in these models is apoptosis-independent and, instead, reliant upon VEC migration into neighboring vessels. In contrast, the physiological regression of the hyaloid vessels and pupillary membrane occurs in a cell-death dependent manner (Ito and Yoshioka, 1999; Zhu et al., 2000; Kim et al., 2010; Schafer et al., 2020). In consideration of these diametrically opposite means of vascular regression, one involving VEC migration and the other necessitating cell death, apoptosis plays a nuanced role in BV regression. Korn and Augustin (2015) extended the apoptotic mechanism a step further, contending that apoptosis may be the primary driver of BV regression in some cases or a consequence of depleted survival signaling. Through this nuanced understanding of apoptosis in BV regression and much ingenuity, Schafer et al. (2020) were able to repurpose YK-4-279, a small molecule inhibitor of E-26 transformation-specific transcription factors, to further promote apoptosis and enhance hyaloid vessel regression in vivo. These findings in a hyaloid vessel regression murine model successfully translated to human in vitro culture, demonstrating YK-4-279’s efficacy in regressing human umbilical vein endothelial cell tubes. Their discovery, facilitated by an understanding of apoptosis in BV regression, may prove fruitful in therapeutic applications for pathological neovascularization (Schafer et al., 2020). For an expert review on EC apoptosis in BV regression, readers are referred to the review by Watson et al. (2017).

We have arrived at this nuanced understanding of apoptosis in BV regression as a result of much research, and to our knowledge, there have been few investigations of the role of apoptosis in LV regression. In a study, Hou et al. (2021) utilized ultraviolet A light crosslinking to induce regression of corneal LVs and BVs in an apoptosis-dependent manner. Perhaps in an example of migration-dependent temporary LV regression, Gur-Cohen et al. (2019) demonstrated a transient dissociation of LVs from the hair follicle stem cell niche after the onset of the hair cycle and noted the absence of apoptosis in this context. In Mäkinen et al. (2001) demonstrated that expression of soluble VEGFR-3 in the skin of transgenic mice starting at embryonic day (E)15 lead to regression of dermal LVs with noted apoptosis of LECs in the embryo. Apart from these studies, there has been little research on the role of apoptosis in LV regression. Extending what is known from the BV regression literature, it is similarly possible that LV regression may proceed through migration of LECs into neighboring vessels or through apoptosis of pruned LV segments. Further studies are warranted to unpack the context-specific role of apoptosis in LV regression.

Prior to the discovery of meningeal LVs by Louveau et al., 2015, it had long been accepted that the central nervous system lacked a lymphatic vascular system (Louveau et al., 2015; Petrova and Koh, 2020). On the heels of this initial discovery, Antila et al. (2017) utilized Prox1-eGFP transgenic mice to modulate and observe meningeal LV development. Through their work, they elucidated that meningeal LVs develop postnatally through VEGF-C and VEGFR-3 signaling. Lymphangiogenesis from meningeal LVs was observed upon injection of AAV-VEGF-C, demonstrating that meningeal LVs are capable of dynamically regressing or undergoing lymphangiogenesis dependent on VEGF-C concentrations. This ability to modulate meningeal LV density may have additional therapeutic implications in neurodegenerative diseases such as Alzheimer’s disease and Parkinson’s disease. In Ma et al. (2017) demonstrated that LVs are the primary conduit for CSF drainage and also observed an age-dependent decline in CSF lymphatic outflow. Considering an age-related decline in CSF outflow along with the important role of CSF in waste removal, Da Mesquita et al. (2018) posited a role for lymphatic flow impairment in Alzheimer’s disease pathology. Their studies of photodynamic meningeal lymphatic ablation in murine models showed that LV impairment translates into impaired efflux of interstitial macromolecules as well as impaired cognitive function (Da Mesquita et al., 2018). Similar studies utilizing mouse models overexpressing α-synuclein demonstrated that LV ligation and the subsequent decrease in CSF clearance increase α-synuclein accumulation, increase neuronal loss, and impair motor function, suggesting that age-related LV dysfunction may also contribute to the pathogenesis of Parkinson’s disease (Zou et al., 2019). In support of this hypothesis, Ding et al. (2021) used magnetic resonance imaging to visualize lymphatic outflow in patients with Parkinson’s disease and found significantly reduced meningeal lymphatic flow in these patients as compared to controls. The current literature suggests that age-related regression of LVs may underpin and exacerbate both Alzheimer’s disease and Parkinson’s disease, and thus, an improved understanding of LV regression could support the development of therapeutics that can modulate this pathological lymphatic impairment.

Lymphatic vessel regression has been hypothesized to exacerbate atherosclerotic cardiovascular disease, the leading cause of death worldwide (Csányi and Singla, 2019; Vourakis et al., 2021). In health, the lymphatic system serves as a conduit for reverse cholesterol transport, the process by which peripheral cholesterol is retrieved and shuttled to the liver by high-density lipoproteins (HDLs) (Huang et al., 2015). Some evidence suggests that reverse cholesterol transport may hamper the pathogenesis of atherosclerosis, although Huang et al. (2015) highlighted that clinical studies have yet to corroborate the athero-protective role of HDL-mediated cholesterol transport (Rosenson et al., 2012). By serving as a channel for cholesterol retrieval, LVs may play a role in reducing the cholesterol burden in atherosclerosis (Fernández-Hernando, 2013; Csányi and Singla, 2019). Studies have suggested that adventitial lymphatics of atherosclerotic vessels undergo regression, leading to potentially impaired drainage of inflammatory cytokines and pro-atherosclerotic lipids, thereby exacerbating atherosclerosis (Taher et al., 2016; Csányi and Singla, 2019). While a component of LV regression may exacerbate atherosclerotic disease, therapeutic lymphangiogenesis is seen to promote recovery of cardiac tissue post-myocardial infarction in mice (Houssari et al., 2020). Using an adeno-associated viral gene delivery of VEGF-C, Houssari et al. (2020) demonstrate that induced-lymphangiogenesis accelerates the resolution of cardiac inflammation following experimental myocardial infarction. A separate work by Liu et al. (2020) demonstrates that LECs play a cardioprotective role through the release of a molecule called Reelin, corroborating the role of lymphatic vessels in cardiac homeostasis. Overall, LVs are hypothesized to have an athero-protective role while LECs are seen to have an important role in injury resolution following myocardial infarction.

A comprehensive understanding of LV regression will require robust research tools, including appropriate animal models. Zhang et al. (2011), Shi et al. (2020) employed a murine cornea model to observe lymphatic regression. Zhang et al. (2011) proposed that the neovascular privilege of the cornea allows for a wide range of lymphatic growth and plasticity, while Shi et al. (2020) concluded that the cornea has strong potential for studying the effect of aqueous humor on lymphatic vessels (Zhang et al., 2011). The mouse tail has also been identified as a useful tool in lymphatic visualization (Weiler et al., 2019). Researchers have used this model in the context of lymphedema (Bramos et al., 2016; Zhou et al., 2020) and general lymphatic trafficking (Hassanein et al., 2021), but Weiler et al. (2019) were optimistic about its use in many other lymphatic conditions. In the context of Gorham Stout disease, Monroy et al. (2020) studied the extent of lymphatic growth due to pre-existing vessels in mouse bones. Like the cornea, bone is also devoid of lymphatic vessels in a healthy state, and using a model for lymphatic research similar to the cornea could lead to valuable insight into conditions such as Gorham Stout disease. Similarly, Yao et al. (2014) investigated VEGF-C overexpression mice to determine whether this model is useful to study LV development in the respiratory tract but found a condition resembling pulmonary lymphangiectasia, a serious condition in the newborn. Hence, models do not necessarily need to be devoid of LV originally, as this example demonstrates the lung as a valid option to better understand pulmonary lymphatic pathology. A valid model is needed also for future LV regression research in the context of cancer, and Frenkel et al. (2021) proposed the use of long-lived human LECs to better understand lymphangiogenesis and tumor–LV interactions. These cells represent a novel method for studying specifically human tumor lymphangiogenesis and can be further employed to study LV regression. In contrast, sentinel lymph node lymphangiogenesis is a well-documented research model that has been utilized in the study of cancer metastasis (Hirakawa et al., 2005). The lymph node lymphangiogenesis model also has been employed to study LV regression (Mumprecht et al., 2012). Table 3 outlines research models that can facilitate increased understanding of LV regression. Most of these research models are in vivo tools that allow for the perturbation and observation of LV dynamics in transgenic mice. The noted LEC microfluidic device represents an emerging area of research with microfluidics; in this example, the microfluidic device enables researchers to exert precise control over interactions between the tumor and LV.

In Section “Pathologies Involving Lymphangiogenic Regression and Dysfunction,” we explored pathologies with a component of LV regression. Accordingly, potential avenues exist for the therapeutic use of LV regression, further emphasizing the importance of characterizing LV regression. An understanding of the nuanced mechanisms by which LV regression occurs may facilitate methods to either therapeutically block these mechanisms in cases of pathological LV or therapeutically induce these mechanisms in cases of pathological lymphangiogenesis. Table 4 lists anti-lymphangiogenic agents that have been employed in pre-clinical studies, and Table 5 lists anti-lymphangiogenic modulatory agents that have been tested in clinical trials for their therapeutic potential for a select few pathologies.