Discussion

Pulmonary arterial hypertension in type 1 neurofibromatosis has not received enough attention. Recent classification of pulmonary hypertension12 places type 1 neurofibromatosis associated pulmonary hypertension in category 5.2, that is, systemic disorders with unknown or multifactorial aetiology. Initially pulmonary arterial hypertension in type 1 neurofibromatosis was ascribed to severe interstitial fibrosis.2 3 Later, many cases were described where no such associations could be defined and pulmonary arterial hypertension was hypothesised to have resulted from plexiform vasculopathy.4–11 In fact, characteristic mosaic pattern on tomographic scan of chest with histopathological changes of pulmonary vasculopathy with an intimal proliferative plexiform lesions characteristic of type 1 neurofibromatosis arterial involvement4–7 were described and in other cases indirect evidence from imaging and pulmonary function testing was used.8–11 The pathologic picture is remarkably similar to idiopathic and heritable pulmonary arterial hypertension which are also characterised by pathological lesions affecting the pulmonary arteries, distal in particular. They are characterised by medial hypertrophy, intimal proliferative and fibrotic changes (concentric, eccentric), adventitial thickening with moderate perivascular inflammatory infiltrates, complex lesions (plexiform, dilated lesions) and thrombotic lesions. Pulmonary veins are classically unaffected.13 The basic molecular mechanisms by which deficiency of neurofibrin, which has guanosinetriphosphatase activating domain, results in vascular proliferative response is well studied (figure 3). Neurofibrin negatively regulates the Rat sarcoma system of proto-oncogene output; its deficiency causes uninhibited proto-oncogene activity and downstream activation of cell growth. Several pathways, including the mitogen-activated protein kinase pathway and tuberous sclerosis protein 1–2 complex,14 are involved resulting in downstream activation of mammalian target of rapamycin. This in turn increases vascular endothelial growth factor and hypoxia induced factor 1 production.15 16 Neurofibromin also modulates adenylate-cyclase activity through a Rat sarcoma system of proto-oncogene independent mechanism.17 In type II neurofibromatosis manifestations are usually limited to nervous system tissue and no vasculopathy is described. This in turn might be related to the Merlin, the defective gene product which acts through entirely different pathway and target genes might not be expressed in pulmonary vessels. However, the exact mechanism still remains elusive. Table 2 summarises all available case reports and series in type 1 neurofibromatosis patients with pulmonary hypertension.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Signalling pathways modulated by NF1. NF1 leads to decreased activity of Ras thus leading to downstream activation of the vascular growth pathway in NF1-mutated cells. mTOR, mammalian target of rapamycin; NF1, neurofibromin type 1; PDGFR, platelet-derived growth factor receptor; PI3K, phosphatidylinositol 3-kinase; Rheb, Ras homologue enriched in brain; TSC, tuberous sclerosis protein. Pointed arrowheads indicate positive regulation while blunt arrowheads indicate negative regulation.

Analysis of all these studies reveals few characteristic features of type 1 neurofibromatosis associated pulmonary hypertension. There is female preponderance unlike type 1 neurofibromatosis itself which affects male and female subjects equally, suggesting a role of oestrogen as reported in heritable pulmonary arterial hypertension.18 In contrast to heritable pulmonary arterial hypertension, the age of onset is very late, with median age being more than 60; also, the median time from diagnosis of type 1 neurofibromatosis to diagnosis of pulmonary arterial hypertension is more than 30 years. This further suggests that arteriopathy of pulmonary vasculature is late phenomenon in natural history of type 1 neurofibromatosis and develops in the long standing disease. Most reported patients have presented with advanced disease before diagnosis and this may be due to lack of awareness of this association and consequent delay in the diagnosis. As previously discussed, the lung abnormalities are variable. Most patients have an impaired diffusion capacity but other parameters are normal. Impaired diffusion capacity in setting of normal lung volumes suggests vascular involvement; it is quite remarkable because type 1 neurofibromatosis patients with lung disease resulting in pulmonary hypertension will invariably show some obstructive, restrictive or mixed patterns and a male preponderance.19 Ventilation perfusion scans and high resolution CT scans of chest might show bilateral filling defects and mosaic pattern (representing irregular perfusion) which again favours vascular involvement as a primary cause of pulmonary arterial hypertension. In all studies where genetic analysis was done, bone morphogenic protein receptor 2 and activin A receptor type II-like kinase-1 mutations were absent, which are known mutations in idiopathic and heritable pulmonary arterial hypertension. In contrast, heterozygous germ line mutations of the type 1 neurofibromatosis gene were present in all cases and were of different types, including short deletions, nonsense and missense mutations, or complete deletion (rare, gene negative cases). No correlation exists between type of mutation and type 1 neurofibromatosis phenotype or development of pulmonary arterial hypertension. The most definitive evidence comes from the cases where biopsy was done and shows the classical plexiform lesions. Undoubtedly, type 1 neurofibromatosis and pulmonary arterial hypertension are rare but it is very important to recognise this association since it implicates mutations in the tumour suppressor gene type 1 neurofibromatosis with the pathogenesis of a pulmonary vasculopathy. Furthermore, early recognition is important because rapid progression and poor prognosis are reported.6 7 11 Both of our patients had long standing neurofibromatosis with advanced disease at presentation and no secondary causes and evidence of primary vascular involvement could be found as described.