INTRODUCTION

TOP ABSTRACT INTRODUCTION CASE REPORTS METHODS RESULTS DISCUSSION REFERENCES

Anorexic drugs such as aminorex fumarate and dexfenfluramine are associated with the development of severe pulmonary hypertension (PH) (1). The PH associated with the use of anorexigens has morphological and clinical features considered indistinguishable from those in idiopathic or primary pulmonary hypertension (PPH) (9). In both conditions, the involved pulmonary vessels exhibit medial smooth muscle hypertrophy and plexiform lesions (10), which represent an abnormal growth of endothelial cells within pulmonary arteries (11).

We have recently shown that the endothelial cells in the plexiform lesions in PPH are predominantly monoclonal, whereas those in secondary PH caused by congenital heart abnormalities or the CREST (calcinosis, Raynaud's phenomena, esophageal dysmotility, sclerodactyly, and telangiectasia) syndrome demonstrate polyclonal growth (12). To date, the clonality status of proliferating endothelial cells in plexiform lesions is the only finding, irrespective of the clinical history, to discriminate between the vascular lesions of patients with PPH and patients with secondary PH. Clonal expansion, a hallmark of neoplastic processes such as Kaposi's sarcoma (13), colonic carcinomas (14), and desmoid tumors (15) result from a somatic genetic event that allows growth advantage of the altered cells.

For the current study, we asked whether appetite suppressant-associated PH is characterized by monoclonal pulmonary endothelial cell proliferation or, alternatively, is associated with a polyclonal endothelial cell proliferation as found in secondary PH (12). We describe herein the clonality studies in plexiform lesions of two patients with appetite suppressant- associated PH by analysis of the pattern of X-chromosome inactivation, targeting the highly polymorphic trinucleotide repeat of the human androgen receptor.

	CASE REPORTS

TOP ABSTRACT INTRODUCTION CASE REPORTS METHODS RESULTS DISCUSSION REFERENCES

Case 1

A 50-yr-old woman weighing 166 lbs. who was treated in April 1995 with fenfluramine and phentermine to induce weight loss. In December 1995, she reported fatigue, occasional dizziness, and dyspnea with minimal exercise. An echocardiogram showed flattening of the interventricular septum and thickening of the right ventricular free wall. A two-dimensional pulse Doppler examination revealed tricuspid regurgitation, and the estimated systolic pulmonary artery pressure was 96 mm Hg. In March 1996, a right heart catheterization revealed a mean right atrial pressure of 17 mm Hg, pulmonary artery pressure of 107 mm Hg systolic and 47 mm Hg diastolic. The cardiac index was markedly reduced to 1.4 L/min/m². A coronary angiogram revealed normal coronary arteries, a ventilation/perfusion lung scan ruled out pulmonary embolic disease. Antinuclear antibodies were negative. The diagnosis of PPH was made. In June 1996, a central line was placed for the continuous delivery of prostacyclin (Flolan). Subsequently, the patient developed several episodes of line sepsis and died from right heart failure in October 1997. An autopsy was performed, which demonstrated severe right ventricular hypertrophy and plexiform pulmonary arteriopathy.

Case 2

A 28-yr-old woman who had been receiving anorexigen fenfluramine for several months in 1973. She first developed dyspnea on exertion in 1976. At 50 yr of age she noted increasing dyspnea. In March 1995, she experienced a dizzy spell while undergoing a phlebotomy for pseudopolycythemia, and over the next month she developed leg edema. The electrocardiogram revealed evidence of right ventricular hypertrophy, and the echocardiogram showed a dilated, poorly contracting right ventricle with tricuspid valve regurgitation and an estimated systolic pulmonary artery of > 100 mm Hg. A ventilation/perfusion scan of her lungs showed no evidence of pulmonary embolic disease. A right heart catheter study demonstrated a pulmonary artery pressure of 133 mm Hg systolic and 44 mm Hg diastolic. The cardiac index was 1.6 L/min/m². A coronary angiogram revealed normal coronary arteries. The patient was initially started on coumadin and treated with a calcium channel blocker. However, there was a gradual deterioration of the patient's dyspnea. In July 1995, the patient received a heart/ lung transplant. The histologic examination of the explanted lung revealed plexiform pulmonary arteriopathy, consistent with PPH (9, 10). The patient is alive and active as a physician and medical researcher.

	METHODS

TOP ABSTRACT INTRODUCTION CASE REPORTS METHODS RESULTS DISCUSSION REFERENCES

Patient Material

Eleven and five paraffin blocks with lung tissue were examined in Patients 1 and 2, respectively. From selected blocks containing multiple plexiform lesions, 50 serial tissue sections 10 µm thick were obtained. Plexiform lesions were mapped on hematoxylin-eosin-stained sections from every sixth slide. Only plexiform lesions of more than 120 µm in length were selected for microdissection. Proliferating endothelial cells in each plexiform lesion were pooled from all eosin-stained slides with dissectable lesions. Control tissue consisted of microdissected lung parenchyma from each of the patients.

We analyzed the pattern of X-chromosome inactivation using the methylation status of the human androgen receptor gene (human androgen receptor assay, HUMARA) as a marker of clonality, as recently described (12, 16). For each sample, the allele inactivation ratio was calculated by dividing the allele amplification ratio of the HhaI digested sample (allele 1/allele 2) by the allele amplification ratio of the HhaI non-digested sample (allele 1/allele 2) to correct for preferential amplification of one allele versus the other, as might occur if the two alleles differed markedly in the length of their repeats (17). A clonality ratio for each lesion was determined by dividing the allele inactivation ratio of the lesion by that of the lung parenchyma of each patient, thereby normalizing values for possible unequal lyonization (17). In cases with a balanced pattern of X-chromosome methylation, suggestive of polyclonal composition, the clonality ratio should range between 0.25 and 1. When a cell population derives from a single progenitor, i.e., monoclonality, the clonality ratio should range between 0 and 0.25 (13, 18).

	RESULTS

TOP ABSTRACT INTRODUCTION CASE REPORTS METHODS RESULTS DISCUSSION REFERENCES

The lungs of the two patients with anorexigen-related pulmonary hypertension exhibited plexiform lesions morphologically similar to those observed in PPH (9, 11).

Endothelial cells obtained by microdissection from four plexiform lesions from Patient 1 and six lesions from Patient 2 were submitted to determine clonality by HUMARA. All 10 lesions exhibited inactivation of either one of the human androgen receptor alleles, consistent with a monoclonal expansion of pulmonary endothelial cells. The 10 endothelial cell samples exhibited a mean clonality ratio of 0.03 ± 0.01 SE (Figure 1). The clonality ratios for the lesions from Patient 1 were 0.02, 0.03, 0, and 0 and in Patient 2 they were 0.14, 0.06, 0.004, 0.016, 0.001, 0.02, 0.03, 0, and 0. The lung parenchyma of both patients, because it is composed of a mixed cellular population with respect to the X-chromosome inactivation of the human androgen receptor alleles, exhibited a polyclonal pattern by HUMARA.

View larger version (90K): [in this window] [in a new window]   Figure 1.   Clonality analysis based on HUMARA of microdissected endothelial cells from four plexiform lesions of Patient 1 (A) and six lesions of Patient 2 (B). In the monoclonal endothelial cell proliferation, the electrophorectic band representative of one parental allele for the human androgen receptor gene is either markedly reduced in intensity or absent after HhaI digestion (+). However, the polyclonal lung cell population of both lung DNA samples exhibits two bands before () and after (+) HhaI digestion, indicative of balanced methylation pattern in the polyclonal population of cells.

	DISCUSSION

TOP ABSTRACT INTRODUCTION CASE REPORTS METHODS RESULTS DISCUSSION REFERENCES

The finding of endothelial cell monoclonality in plexiform lesions lends further support to the interpretation that patients with appetite suppressant-related PH indeed have PPH. By contrast, in severe secondary PH such as seen associated with congenital heart malformations or CREST, the endothelial cells within plexiform lesions are instead polyclonal (12).

Monoclonal expansion of the pulmonary endothelial cells in appetite suppressant-related PH and in PPH could arise by a somatic gene mutation that allows for a selective growth advantage of a single or a cluster of genetically similar cells. By blocking serotonin re-uptake or via an alternative mechanism (i.e., increased expression of endothelial growth factors), it is possible that dexfenfluramine and its metabolite might trigger or accelerate the growth rate of pulmonary endothelial cells, thereby causing PPH in genetically susceptible patients (20). These patients, in the absence of anorexigen drug intake, may not have developed PH.

Because of the prolonged latency period (between 3 and 23 yr in the second patient), one may doubt the relationship between the precipitating cause (fenfluramine intake) and the resulting effect (PPH). However, latency periods of greater than 6 yr have been reported for aminorex-related PPH and, in addition, the natural history of anorexigen-related PH may be more variable than in pure PPH (19, 20).

In conclusion, our results indicate that appetite suppressant-associated PH is identical to PPH not only in clinical and histopathologic features but also in terms of the monoclonal nature of the endothelial cell proliferation. The intake of anorexigens may trigger the disease in a patient with the predisposition to develop PPH.