A cancer amidst us: the plexiform lesion in pulmonary arterial hypertension

Despite decades of active research and the development of targeted medical therapy, pulmonary arterial hypertension (PAH) remains largely idiopathic and is fatal (10). Current therapies are vasodilators, which reduce vasoconstriction but do not reverse fixed arterial occlusive lesions, including plexiform lesions (6). These plexiform lesions are common in World Health Organization group 1 PAH patients, but they are not necessarily pathognomonic for group 1 disease. Pulmonary hypertension of other groups, although less common, can manifest with plexiform lesions when their disease is advanced and complicated by precapillary involvement (1, 3, 17). The presence of precapillary disease portends poor patient prognosis among all groups of pulmonary hypertension, and therefore, it is advisable to identify these subgroups of patients with precapillary arteriopathy (16). Since current medical therapy does not impair the growth of plexiform lesions (13), a better understanding of the mechanisms involved in lesion formation may lead to new therapeutic strategies. In the current issue of the American Journal of Physiology-Lung Cellular and Molecular Physiology, Cool et al. (5) reexamine whether the PAH precapillary arteriopathy is a “cancer.” They make a strong case that the plexiform lesions of PAH exhibit features of cancer, which by extension, may offer new therapeutic options for patients with this devastating disorder.

While cancer is a disease of the cell cycle, defining “cancer” is a difficult proposition. The hallmarks of cancer encompass a spectrum of molecular changes to cells that are observed over years, but these disease-defining molecular changes have evolved as the field has developed a better understanding of cell cycle control (8). Genetic instability may arise secondary to some random mutation (i.e., mutator phenotype), multiple environmental and genetic interactions (i.e., two-hit hypothesis), or through genetic instability occurring along with cell division (i.e., accumulation of random mutations during replication). Nonetheless, there are exceptions seen in each of these cases, which sparks debate over causation (11). In the context of PAH (Table 1), using cancer hallmarks might create a framework for developing new targeted therapies, but it might be limiting, too. PAH lesions may represent early-stage tumor development, where high replicative and metastatic potential has not developed. Lesions may be an initiation stage where one-hit has occurred, but the second (and possibly additional) hit(s) is/are lacking. Or multiple hits may have occurred, but the tipping point of the necessary driver mutations has not yet been achieved (14). A more interesting cancer descriptor, not included by Cool et al. (5), is tumor dormancy (2). PAH lesions may represent a dormant cancer state, characterized by altered cell growth and development of adaptive mechanisms, or stem-like characteristics, in response to their extracellular matrix environment, hypoxia, or another stress, like shear stress. Dormancy would make cancer analogies difficult to convert into therapies. Applying cancer descriptors to PAH may stem new lines of investigation, but cancer and our understanding of it are always evolving.

Table 1.

Hypotheses relevant for developing targets in PAH “cancer”

Model	Features	PAH Features
Two-hit (or multiple-hit) hypothesis	Loss of function for tumor suppressors requires both alleles to be inactivated, either through mutations or through epigenetic silencing, to cause a phenotypic change	Dysregulation of angiogenic signaling, turbulent flow/shear stress, hypervasocontractility
Mutator phenotype	Increase in mutation rate results from early sporadic mutations in gene responsible for maintaining genomic fidelity, which results in an increase in mutation rate from normal spontaneous or replicative errors	Mutations and chromosomal abnormalities, loss of p53, genomic instability
Replicative errors	Repeated cell division over the life of a cell results in replicative errors to genes and fragile DNA sites that cause cancer	Telomerase expression, mutations in genes not known to be oncogenic or associated with tumor suppression, increase proliferation signaling for these tissues
Cancer dormancy	Residual cancer or abnormal cells that lack proliferative and apoptotic markers and maintain in a state of quiescence without a continuous growth	Bone marrow-derived stem cells, immune involvement, TGF-β, SOX, hypoxia or hypoxic response

PAH, pulmonary arterial hypertension; TGF-β, transforming growth factor-β.

Plexiform lesions look like cancer. They are occlusive, glomeruloid-like vascular lesions that are widely considered a “hallmark” of severe disease—due to both their appearance as an indicator of disease severity and their apparent role in its progression through inflammatory vascular remodeling (12). Although most reports indicate that plexiform lesions are commonly found in distal, muscularized pulmonary artery segments and branch points, recent findings highlight their presence in the perialveolar segment and in both bronchiolar and supernumerary arteries (13). The current hypothesis of lesion formation posits that vascular injury initiates endothelial apoptosis that leads to the selection of a dysfunctional apoptosis-resistant and hyperproliferative endothelial population (13, 15). Along with impaired smooth muscle cell apoptosis and inflammatory cell infiltration, this process results in plexiform lesions that are morphologically complex. Indeed, plexiform lesions are characterized by slit-like “channels” of endothelial cells within the occluded vessel lumen that are surrounded by smooth muscle cells, myofibroblasts, connective tissue matrix, platelet aggregates, and inflammatory cells including CD3-positive T cells and CD68-positive macrophages (12). Interestingly, CD44, which mediates lymphocyte adhesion and T cell extravasation and contributes to malignancies, is localized in the endothelial cells within plexiform lesions and T cells proximal to the lesions (12).

The origins of plexiform lesions remain a conundrum. Much of the evidence supporting the cancer hypothesis is directly derived from work on the endothelium within the lesion (9). Monoclonality, hyperproliferation, loss of tumor suppressor protein, and dysregulation of the VEGF/VEGF-R signaling paradigm clearly indicate an abnormal vascular cell state (9). However, one distinction between plexiform lesions and cancer is that angiogenesis in solid tumors establishes tortuous, leaky vessels. In a direct histological comparison between plexiform lesions and glomeruloid-like lesions from glioblastoma, one immediately obvious difference is the formation of structurally sound hemichannels lined by endothelium in the plexiform lesions. These are markedly absent from not only tumors of this type, but solid tumors in general. The stabilization of hemichannels in plexiform lesions may provide an advantage over traditional antineoplastic treatment because there is the potential for improved drug delivery to the stable plexiform lesion, provided the lesions retain blood flow. Even with good drug delivery, however, the complexity of cell phenotypes within the stable lesion will represent a challenge for medical therapy.

While it is important to explore the molecular and cellular basis of plexiform lesions, it is also important to recognize that their pathophysiological relevance remains in question; while precapillary arteriopathy is associated with poor outcomes, it is uncertain how, or whether, reversing these lesions will improve prognosis. Plexiform lesions undoubtedly contribute to fixed vascular lesions, but are they present in sufficient numbers to increase pulmonary vascular resistance either alone or in combination with the noncompliant yet open feeder arteries/arterioles (18)? Maybe so, but it is likely that 3D pathology will be required to gain a more rigorous estimate of their frequency and distribution within the circulation. Plexiform lesions are found in small precapillary arterioles and in intrapulmonary bronchopulmonary anastomotic pathways (7). They undoubtedly impair blood flow through the microcirculation (7, 19), but how does this change in flow impact vascular remodeling that is sensed by the right ventricle, and can any corrective measure bring the blood flow distribution back to within normal limits? Plexiform lesions represent an overgrowth of “apoptosis-resistant” endothelial cells into the lumen, yet along with this, there is vascular dropout downstream from the occlusion (4). So, is this a disease of hyperproliferation, apoptosis, or both, and is it possible that antiproliferative therapy increases the risk for pulmonary hemorrhage and hemoptysis? Answers to these questions can only be empirically derived. The cancer paradigm of PAH is hypothesis generating, and it offers an innovative approach to consider new treatment options that themselves will help answer vexing concerns regarding the pathophysiology of this unremitting disease.

GRANTS

This work was supported by NIH National Heart, Lung, and Blood Institute Grants HL66299 (T.S.), HL60024 (T.S.), HL140182 (T.S.), HL133066 (N.N.B.), HL136869 (C.M.F.), American Heart Association Grant 18CDA34080151 (J.Y.L.), and the Breast Cancer Research Fund of Alabama Innovation Award (N.R.G.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.Y.L., C.M.F., N.N.B., N.R.G., and T.S. drafted manuscript; edited and revised manuscript; and approved final version of manuscript.