Hereditary hemorrhagic telangiectasia: from signaling insights to therapeutic advances

Abstract

Hereditary hemorrhagic telangiectsia (HHT) is an inherited vascular disorder with highly variable expressivity, affecting up to 1 in 5,000 individuals. This disease is characterized by small arteriovenous malformations (AVMs) in mucocutaneous areas (telangiectases) and larger visceral AVMs in the lungs, liver, and brain. HHT is caused by loss-of-function mutations in the BMP9-10/ENG/ALK1/SMAD4 signaling pathway. This Review presents up-to-date insights on this mutated signaling pathway and its crosstalk with proangiogenic pathways, in particular the VEGF pathway, that has allowed the repurposing of new drugs for HHT treatment. However, despite the substantial benefits of these new treatments in terms of alleviating symptom severity, this not-so-uncommon bleeding disorder still currently lacks any FDA- or European Medicines Agency-approved (EMA-approved) therapies.

Figure 1. AVMs: The hallmark of HHT and its clinical consequences.

In HHT, focal dilations of the postcapillary venules (indicated on images by V) progressively encompass the normal capillaries and establish a direct connection with dilated arterioles (indicated by A), leading to AVM formation, muscularization of large vessels, and dilation of capillary beds, which are often surrounded by inflammatory cells. At the microvascular level, AVMs appear as telangiectases on specific mucocutaneous areas (finger pads, lips, tongue, nasal and digestive mucosa). They are responsible for spontaneous and recurrent epistaxis/bleeding, leading to bleeding iron-deficiency anemia. At the macrovascular level, large AVMs mainly affect the liver, leading to high-output cardiac failure, and more rarely biliary ischemia and portal hypertension; lung AVMs can provoke ischemic stroke and cerebral abscess and, more rarely, hypoxemia and hemoptysis. Central nervous system AVM is rarely complicated by hemorrhage. Skin and liver sections were stained for H&E. Photo of the liver section was provided by J.Y. Scoazec (Institut Gustave Roussy, University Paris-Saclay, Paris, France).

Figure 2. BMP9-10/ENG/ALK1/SMAD4 signaling pathway maintains vascular quiescence by repressing angiogenic pathways.

HHT occurs due to LOF mutations in ENG, ALK1, SMAD4, and, more rarely, BMP9 (respective proteins indicated with red asterisks), which are all in the same signaling pathway. On endothelial cells, BMP9 or BMP10 recruits a heterocomplex composed of two type II receptors (BMPRII or ActRIIA, which are the main type II receptors expressed on ECs, and two similar type I receptors (ALK1), and the coreceptor ENG (endoglin). Upon BMP binding, the type II receptor phosphorylates ALK1, which subsequently phosphorylates the transcription factors SMAD1/5. SMAD1/5 bind SMAD4, which is shared with the TGF-β pathway, to regulate transcription of many genes (in association with other transcription factors). BMP9 and BMP10 maintain vascular quiescence (middle panel) via repression of angiogenesis pathways (right panel). VEGF-A (red) binds to VEGFR2, which activates the ERK1/2 and P38 MAPK pathways and the PI3K/AKT/mTORC1 pathway. In turn, the PI3K/AKT/mTORC1 pathway activates the signaling cascade P70S6K/S6. VEGF can also activate the calcineurin phosphatase, which activates, via dephosphorylation, the NFAT transcription factor family. The PI3K/AKT/mTOR pathway is negatively regulated by the phosphatase PTEN, which is active when unphosphorylated. VEGF-A can also bind to VEGFR1, but this will not generate a signal. Two other members of the VEGF family, VEGF-B (yellow) and PlGF (blue), also bind to VEGFR1. BMP9 induces the expression of VEGFR1, thus inhibiting VEGF signaling. BMP9 also induces PTEN expression and phosphorylation, which inhibit AKT activity as well as the expression of SGK1 kinase, which can activate the mTORC1/P70S6K/S6 pathway. Moreover, BMP9 inhibits ERK activation and CDK4/6 kinases through not-yet-characterized mechanisms. Ang1 activates the TIE2 receptor to maintain vascular quiescence, and this pathway can be antagonized by Ang2.

Figure 3. Therapeutic targets of antiangiogenic drugs tested in preclinical models and in HHT patients.

Figure shows the targets of drugs tested in completed clinical trials (red), drugs currently under testing or with a case report (orange), and drugs tested only in preclinical models (blue). For further details, see Table 1 and Table 2. Drugs have been developed to block VEGF-A signaling using neutralizing anti–VEGF-A mAbs (bevacizumab) or soluble trap/decoy receptors that bind VEGF-A (red), VEGF-B (yellow), and PlGF (blue) (aflibercept), or neutralizing anti-VEGFR2 antibodies (D5B1 and DC101). Drugs developed to block intracellular signaling, such as tyrosine kinase receptor inhibitors that block VEGFR2 activity but also other receptors, are currently undergoing testing: pazopanib (VEGFR, PDGFR, c-KIT, and FGFR) and nintedanib (VEGFR, PDGFR, and FGFR). Drugs have also been developed to block PI3K and AKT (VAD044), as well as mTORC1 (sirolimus) and calcineurin and FKBP12 (tacrolimus). Immunomodulatory imide drugs (IMIDS), such as thalidomide and pomalidomide, have been tested in HHT patients. Other drugs have been tested so far only in preclinical models, such as the neutralizing anti-Ang2 monoclonal antibodies (LC10) and inhibitors of CDK4/6 (palbociclib and ribociclib).