Assessing the cancer hypothesis of pulmonary arterial hypertension: the devil is in the detail

Many pathogenic attributes of pulmonary arterial hypertension (PAH), including proliferative, antiapoptotic, and angiogenic features, together with metabolic dysregulation and proinflammatory phenomena, are frequently observed in malignant tissues and are the basis for the “cancer theory” of PAH. This concept is revisited in this issue of AJP Lung (2) after being first proposed in 2008 by Voelkel and colleagues (10). The authors compare pathogenic features of PAH to the updated hallmarks of cancer put forward by Hanahan and Weinberg (5). The original hallmarks of cancer are 1) sustaining proliferative signaling, 2) evading growth suppressors, 3) activating invasion and metastasis, 4) enabling replicative immortality, 5) inducing angiogenesis, and 6) resisting cell death. In their update based on conceptual progress over the past decade, Hanahan and Weinberg identify two emerging hallmarks and two enabling characteristics (the latter are described as cellular characteristics crucial to the acquisition of the cancer hallmarks, but not as hallmarks themselves). The emerging hallmarks include 1) deregulating cellular energetics and 2) avoiding immune detection, while the enabling characteristics include 1) genome instability and mutation and 2) tumor-promoting inflammation.

The overarching goal of the Hanahan and Weinberg paper was to create a conceptual framework describing the acquired behaviors of neoplastic cells and to provide a foundation for understanding the underlying biology of human cancers, encompassing as many types and stages as possible (5). The current review by Cool et al. (2) applies a similar conceptual framework to severe PAH. The authors emphasize key areas shared between PAH and cancer, such as apoptosis resistance, cell-cell interactions, and the role of the microenvironment, autoimmunity and inflammation, metabolic reprogramming, DNA damage, disordered angiogenesis, endothelial-to-mesenchymal transition and phenotype switching, contributions of bone marrow cells, as well as systemic disease components. Of note, some of these concepts are more fully recognized manifestations in PAH than others. Importantly, studies of cancer mechanisms have led to significant progress in the field resulting in several clinical trials (9). The authors also identified several knowledge gaps, areas of future investigation, and propose new ideas moving forward.

Despite the many cancer-like pathogenic features and signaling pathways shared between cancer and PAH (10), there are also distinct and nuanced differences. For example, cultured PAH-derived cells do not completely behave like cancer cells. Specifically, pulmonary artery endothelial and smooth muscle cells isolated from human PAH patients may have a higher proliferative index or rate compared with healthy controls, but they have a limited proliferative capacity and are sensitive to cell-cell contact inhibition in culture (8, 11). This differs from cancer cells, which exhibit sustained proliferative capacity in culture and insensitivity to contact inhibition (one of the most fundamental characteristics of cancer cells) (1, 5, 8, 11).

Furthermore, the cancer hallmark of “inducing angiogenesis” describes the tumor inducing angiogenesis to supply itself with nutrients and oxygen and to eliminate waste. Meanwhile, the role of angiogenesis in PAH is still unclear and may be a response to a vascular injury followed by improper resolution of a repair pathway rather than an actual pathogenic induction as by a tumor. In addition, inhibition of angiogenesis is used to induce experimental PAH (12). This may be an important difference between cancer and PAH and a distinct characteristic to PAH.

Another similar but distinct characteristic of PAH includes the emerging hallmark of reprogramming energy metabolism. While this mechanism clearly occurs in PAH cells, PAH patients may also have systemic metabolic abnormalities (7). In cancer, on the other hand, this appears to be primarily confined to the tumor, tumor microenvironment, or sites of metastasis (5).

Finally, responses to therapeutic strategies used in cancer may differ between cancer and PAH and may cause rather than alleviate PAH. For example, while the tyrosine kinase inhibitor imatinib [successfully used to treat chronic myeloid leukemia (CML)] significantly improved hemodynamics and exercise capacity in PAH (6), the tyrosine kinase inhibitor dasatinib (also used to treat CML) has been linked to the development of PAH (4). Similarly, while inhibiting vascular endothelial growth factor signaling is a therapeutic strategy to block the induction of angiogenesis in some cancers (5), this strategy is also used to induce experimental PAH in rodents (12).

Taken together, we agree that pathways underlying cancer pathobiology are likely at play in PAH. However, there are also several distinct and significant differences that we are only beginning to understand. Maybe the question is not so much whether PAH is like cancer or not, but rather which pathways active in cancer are similarly contributing to disease development and progression in PAH, and which of these pathways can be harnessed therapeutically. A pathway active in cancer may also be active in PAH, but may engage different targets and have different effects. In this framework, it is important to understand that PAH is not a homogenous disease, but rather a group of diseases with differences in pathogenesis, progression, and prognosis. We posit that some forms of PAH may be more cancer-like than others. Similarly, cancer-like features may differ between various stages of the disease (early vs. late) and may even differ between specific cell types or compartments of the pulmonary vasculature (e.g., cells in plexiform lesions or bronchial artery-pulmonary intersections vs. cells of unaffected vascular sections). To dissect and understand these nuances, tissue sampling and deep phenotyping will be critical. Personalized medicine approaches may tell us if certain patients exhibit a more cancer-like phenotype than others and if therapies used for cancer may be indicated. Similarly, several clinical trials of cancer-derived therapies are currently underway (e.g., mTOR inhibition, PARP inhibition, FoxO activation) (9) and may identify responders and nonresponders, providing us with information on which types of patients may benefit from such therapies.

Along those lines, when discussing the cancer hypothesis of PAH, we should also think about innovative novel methods for drug delivery, immune cell modification, and gene editing being used in the cancer field (3). Homing-based regimens, CAR-T therapy, and CRISPR are examples of highly promising novel and cutting-edge strategies that may ultimately be applicable to PAH as well.

Cool et al. provide us with “food for thought” on how to further test the cancer hypothesis of PAH. Some hypotheses may hold up whereas others may be refuted. Rigorous experiments will provide the answer. The devil is in the detail. Having said this, a better understanding of similarities and differences between PAH and cancer hopefully will lead to novel diagnostic and therapeutic strategies that benefit patients with either disease.

GRANTS

This work was supported in part by American Heart Association Grant 19CDA34660173 (A.L.F.), American Lung Association Award CA-629145 (A.L.F.), Actelion Pharmaceuticals US, Inc. Actelion Entelligence Young Investigator Award (A.L.F.), NIH National Heart, Lung, and Blood Institute (NHLBI) Grant R01 HL142638 (Y.-C.L.), Veterans Affairs Merit Review Award 2 I01 BX002042-05 (T.L.), and NIH NHLBI Grant HL144727-01A1 (T.L.).

DISCLOSURES

A.L.F.’s institution received research funding from Actelion. Y.-C.L. received research funding from United Therapeutics Corp. T.L. received consulting fees from Bayer and research reagents from Eli Lilly and Company.

AUTHOR CONTRIBUTIONS

A.L.F., Y.-C.L. and T.L. drafted manuscript; A.L.F. and T.L. edited and revised manuscript; A.L.F., Y.-C.L., and T.L. approved final version of manuscript.