Lymphangioleiomyomatosis (LAM), which affects

Transcription

1 Frequent Estrogen and Progesterone Receptor Immunoreactivity in Renal Angiomyolipomas From Women With Pulmonary Lymphangioleiomyomatosis* Helen Logginidou, MD; Xiang Ao, MD; Irma Russo, MD; and Elizabeth Petri Henske, MD Objective: To determine whether renal angiomyolipomas from women with pulmonary lymphangioleiomyomatosis (LAM) express estrogen receptor (ER) and progesterone receptor (PR). Design: Retrospective study of archival tissue. Patients: Twelve women with LAM and angiomyolipomas. Setting: Fox Chase Cancer Center. Interventions: ER and PR expression was studied using immunohistochemistry. The hormonal status of the patients at the time of resection of the angiomyolipoma was determined. Results: Ten of the angiomyolipomas had ER immunoreactivity (83%), and all 12 had PR immunoreactivity (100%). The ER and PR positivity was in the smooth muscle component of the angiomyolipomas only. For five women, pulmonary LAM specimens were also available; two were ER positive (40%), and all five were PR positive (100%). All four angiomyolipomas from women receiving progesterone therapy were ER and PR positive. One tumor from a woman receiving tamoxifen was ER negative and strongly PR positive. One woman was pregnant; her tumor was ER and PR positive. Conclusions: ER and PR expression is frequent in renal angiomyolipoma cells from women with LAM. PR was more consistently present than ER in angiomyolipomas and in LAM. Our data suggest that angiomyolipoma growth could be affected by hormonal factors. If the growth of LAM-associated angiomyolipomas slows during hormonal therapy, there are two potential implications for LAM patients: first, angiomyolipoma size could serve as a measurable indication of response to hormonal therapy; and second, surgical removal of angiomyolipomas might be avoided in some cases. (CHEST 2000; 117:25 30) Key words: angiomyolipoma; estrogen receptor; lymphangioleiomyomatosis; progesterone receptor; smooth muscle proliferation; tuberous sclerosis complex Abbreviations: ER estrogen receptor; LAM lymphangioleiomyomatosis; MoAb monoclonal antibody; PR progesterone receptor; TSC tuberous sclerosis complex Lymphangioleiomyomatosis (LAM), which affects women almost exclusively, is a rare disease of unknown etiology that was first described 60 years ago. 1 5 The average age at onset of symptoms, which include shortness of breath, pneumothorax, cough, and chest pain, is 33 years. 6,7 Chest radiographs typically reveal a diffuse interstitial infiltrate. Although most LAM is pulmonary, retroperitoneal and pelvic lymph *From the Departments of Medical Oncology (Drs. Logginidou and Henske) and Pathology (Drs. Ao and Russo), Fox Chase Cancer Center, Philadelphia, PA. This work was supported by the LAM Foundation (Cincinnati, OH) and the National Institutes of Health (HL60746). Manuscript received April 29, 1999; revision accepted August 9, Correspondence to: Elizabeth Petri Henske, MD, Fox Chase Cancer Center, 7701 Burholme Ave, Philadelphia, PA 19111; EP_Henske@fccc.edu node involvement can also occur. 8 Most patients have a slowly declining clinical course. 6 Lung transplantation is the only effective therapy for end-stage disease. LAM can occur as an isolated disorder, which we have referred to as sporadic LAM, or in association with tuberous sclerosis (TSC). LAM affects 2.3% of individuals (or 4.6% of women) with TSC. 9 TSC is an autosomal dominant disorder characterized by seizures, mental retardation, and hamartomatous tumors of the brain, heart, kidney, lung, and skin. These tumors include cerebral cortical tubers, subependymal giant cell astrocytomas, retinal hamartomas, cardiac rhabdomyomas, renal angiomyolipomas, and facial angiofibromas. Angiomyolipomas are benign tumors composed of fat, smooth muscle, and dysmorphic vessels. Renal angiomyolipomas occur in CHEST / 117 / 1/ JANUARY,

2 70% of TSC patients and in 33 to 63% of women with sporadic LAM. 5,10 12 We suspect that the occurrence of angiomyolipomas in both TSC-associated and sporadic LAM reflects a common underlying genetic basis for both diseases. 13 LAM consists of a diffuse proliferation of smooth muscle cells around lymphatic vessels, blood vessels, and airways. In later stages, the smooth muscle cells form nodular aggregates. Pathologically, sporadic LAM and TSC-associated LAM are indistinguishable. 9 By electron microscopy, both types of LAM cells contain characteristic microvesicles These microvesicles often stain positively with HMB- 45, a monoclonal antibody (MoAb) named for its immunogen, human melanoma black. HMB-45 binds to melanosomes in the cytoplasm of normal fetal melanocytes. 14,15 The pulmonary smooth muscle cells of LAM can also be immunoreactive for estrogen and/or progesterone receptors The role of these receptors in the pathogenesis of LAM is not known. The smooth muscle proliferation in the lungs of LAM patients is likely to be driven in part by hormonal stimuli. The evidence for this includes the onset of LAM predominately in premenopausal women, the reports of exacerbation of LAM during pregnancy, and the therapeutic benefit of hormonal therapy. Whether these hormonal stimuli contribute to the growth of angiomyolipomas in LAM is not yet understood. However, because the smooth muscle cells of angiomyolipomas appear to be closely related to those of LAM, 18 it seems likely that the mechanisms of growth are also related. We previously found that 55% of angiomyolipomas from women with TSC were immunoreactive for progesterone receptors (PR). 25 In the study reported here, we examined the expression of estrogen receptors (ER) and PR in 12 angiomyolipomas and 5 lung biopsy specimens from women with sporadic LAM. Patients Materials and Methods This study was approved by the Institutional Review Board of Fox Chase Cancer Center. None of the 12 women in this study had any clinical signs or symptoms of TSC, or a family history of either TSC or LAM. Ten of the patients had brain CT or MRI scans to exclude TSC, all 12 had dermatologic examinations, and 6 had ophthalmologic examinations. The degree to which individual patients were screened for TSC has been previously reported, 13 with the exception of patient 482. Patient 482 had dermatologic and ophthalmologic examinations and brain MRI, none of which showed evidence of TSC. The patients ranged in age from 19 to 49 years, with a mean of 33 years, at the time of resection of the angiomyolipoma. Seven of the 12 patients had the angiomyolipoma diagnosed first; 5 patients had the LAM diagnosed first. All underwent resection of the entire angiomyolipoma. All patients had lung biopsies and/or high-resolution CT scans to document the presence of pulmonary LAM. Paraffin blocks from the lung biopsy specimens were available for immunohistochemistry for five of the patients. Angiomyolipomas from four patients in the current study (patients 432, 436, 437, and 492) were previously found to have TSC2 gene loss of heterozygosity. 13 Antibodies A MoAb to ER, mouse IgG1 clone ER88, raised against human recombinant estrogen receptor protein 26 ; and a MoAb to PR, mouse IgG1 clone PR88, raised against purified human progesterone receptor protein (both from Biogenex; San Ramon, CA) were used in this study. Immunohistochemistry Five-micrometer paraffin-embedded tissue sections were deparaffinized, rehydrated, and incubated in 3% hydrogen peroxide (Fisher Scientific; Pittsburgh, PA) for 15 min to quench endogenous peroxidase activity. The sections were then incubated in Citra Solution (Biogenex), ph 6.2, at 98 C for 7 min. Slides used to analyze PR immunohistochemistry were also incubated in 0.02% trypsin for 10 min. All of the tissue sections were incubated for 20 min in 10% diluted normal horse serum (GibcoBRL; Gaithersburg, MD). Excess serum was blotted from the slides, and the sections were incubated with prediluted mouse monoclonal antihuman ER88 antibody or prediluted mouse monoclonal antihuman PR88 antibody. Phosphate-buffered saline solution was used as a control. The PR88 antibody was incubated with the slides overnight at 4 C. The ER88 antibody was incubated with the slides for 90 min at room temperature. The sections were rinsed in Optimax Solution (Biogenex) and then incubated for 30 min with the MultiLink biotinylated goat secondary antibody (Biogenex) at room temperature, followed by a rinse in phosphate buffered saline solution and incubation for 30 min with the peroxidase-conjugated streptavidin (Biogenex). After rinsing in phosphate-buffered saline solution, the slides were incubated in peroxidase substrate solution containing hydrogen peroxide and 3,3 -diaminobenzidine (Biogenex) for 2 min. Breast cancer specimens previously determined to be ER and PR positive were included as positive controls with each reaction. The intensity of the staining was graded as follows: negative ( ), weak ( ), moderate ( ), strong ( ), or very strong ( ). The staining intensity of each slide was determined by comparing it to the intensity of an ER- and PR-positive breast cancer specimen that was assayed simultaneously. The staining intensity of the most positive nuclei in the angiomyolipoma was compared with the intensity of the most positive nuclei in the breast cancer. The intensity of the breast cancer nuclei was designated as strong ( ) for the purpose of this comparison. Two angiomyolipoma specimens (443 and 482) had nuclear staining for PR that was more intense than the breast cancer control and were designated as very strong ( ). The percentage of positive nuclei was also estimated for each specimen. Results As shown in Table 1, all 12 of the angiomyolipomas were PR positive (100%) and 10 were ER 26 Clinical Investigations

3 Table 1 ER and PR Immunoreactivity in Angiomyolipomas and Pulmonary LAM* AML LAM Patient Hormonal Status ER PR ER PR 432 (30) (30) 436 (30) (30) 437 Progesterone (20) (50) 443 Pregnant 10 wk (50) (50) 480 Progesterone (50) (50) 481 Progesterone (30) (80) 482 Tamoxifen (40) 487 (30) (30) (20) 489 (30) (20) (30) 490 Progesterone (50) (50) (20) 491 (30) (30) (20) (20) 492 (10) (20) (10) N (83) 12 (100) 2 (40) 5 (100) *Data are presented as degree of positivity (percentage of positive cells), unless otherwise indicated; weak; moderate; strong; very strong; negative; AML angiomyolipoma. Hormonal status at the time of resection of the renal angiomyolipoma. A dash indicates that the patient was not pregnant and not receiving hormonal therapy. None of the patients whose lung tissue was studied were receiving hormonal therapy at the time of the lung biopsy. Number (percentage) of the AML and LAM specimens that stained positively for ER or PR. positive (83%). The degree of positivity for ER, relative to the breast cancer specimen used as a positive control, was moderate in four cases (33%), weak in six cases (50%), and absent in two cases (17%). The PR positivity was strong in five cases (42%), moderate in six cases (50%), and weak in one case (8%). The ER positivity was seen in between 0% and 50% of the nuclei. The PR positivity was seen in 20 to 80% of the smooth muscle cell nuclei. Of the five patients for whom lung specimens were also available, two patients (40%) were ER positive and five patients (100%) were PR positive. For 10 of the 12 angiomyolipoma specimens and for all 5 of the lung specimens, the degree of positivity and/or the percentage of positive cells was greater for PR than for ER. A representative example of ER and PR immunoreactivity in the pulmonary LAM is shown in Figure 1. An example of immunoreactivity in an angiomyolipoma is shown in Figure 2. Because LAM has been reported to clinically worsen during pregnancy, and is often treated with hormonal agents, we determined the hormonal status of the women in this study at the time of the removal of the angiomyolipoma (Table 1). At the time of lung biopsy, none of the five women from whom lung biopsy specimens were obtained was receiving hormonal therapy. At the time of the angiomyolipoma surgery, one patient was pregnant, four were receiving progesterone therapy, and one was receiving tamoxifen therapy. The angiomyolipoma from the patient who was receiving tamoxifen therapy (patient 482) was ER negative and very strongly PR positive. PR and ER positivity were seen in the angiomyolipomas from all four women receiving progesterone therapy (patients 437, 480, 481, and 490), as well as in the tumor from the woman who was pregnant (patient 443). Discussion LAM occurs almost exclusively in women of childbearing age. The underlying reasons for this are not known. A better understanding of the hormone receptor status in LAM could contribute toward the rational selection of appropriate hormonal therapy. 4 Currently, many women with LAM are treated with high-dose progesterone, oophorectomy, or tamoxifen. The clinical efficacy of these therapies has never been assessed in a controlled study. In this study, we examined ER and PR expression in angiomyolipomas from 12 women with sporadic LAM and in pulmonary LAM specimens from 5 women. We found PR immunoreactivity in the smooth muscle component of all 12 of the angiomyolipomas (100%). ER immunoreactivity was present in 10 of the tumors (83%). The smooth muscle cells of all five of the pulmonary LAM specimens had PR expression. Two specimens (40%) also had ER expression. The strongest PR immunoreactivity in an angiomyolipoma was seen in patients 443 and 482. This strong positivity may be related to hormonal factors: patient 443 was in the first trimester of pregnancy, CHEST / 117 / 1/ JANUARY,

4 Figure 1. Pulmonary LAM from patient 491 showing weak ER positivity (top, A) and moderate PR (bottom, B) positivity. Figure 2. Renal angiomyolipoma from patient 443, who was pregnant at the time of resection of the tumor, showing moderate ER positivity (top, A) and very strong PR (bottom, B) positivity. and patient 482 had been on tamoxifen therapy for 10 years at the time that the angiomyolipoma was removed. The relationship between tamoxifen therapy and changes in ER and PR in breast cancer is not well defined. In a recent immunohistochemical study of primary breast tumors during tamoxifen therapy, a decrease in ER was seen in 3 of 15 tumors, and an increase in PR was seen in 7 of 17 tumors.27 Tumors that responded to the tamoxifen therapy were more likely than nonresponding tumors to have a decrease in ER and an increase in PR. The relationship between progesterone therapy and changes in ER and PR in primary tumors is not known. Four patients in our study were receiving progesterone therapy at the time the angiomyolipoma was removed. All four had weak or moderate ER positivity and moderate or strong PR positivity in the angiomyolipomas. This is, to our knowledge, the first series of LAM-associated angiomyolipomas that has been studied for ER and PR expression. There is one other study by Tawfik et al21 that examined a single angiomyolipoma from a woman with sporadic LAM for ER and PR immunoreactivity. This patient was 25 weeks pregnant at the time of angiomyolipoma resection. The angiomyolipoma was negative for both ER and PR (Table 2). Pulmonary LAM samples were available for immunohistochemistry for five of the patients in our study, making this one of the largest series of pulmonary LAM specimens studied for ER and PR immunoreactivity. All five of our pulmonary specimens were immunoreactive for PR in the pulmonary LAM, and two were immunoreactive for ER. The results of our study and previous studies in which ER and PR immunoreactivity has been studied in LAM are summarized in Table 2. Ohori et al28 reported five cases of pulmonary LAM, none of which had ER or PR immunoreactivity. Kinoshita et al29 reported two cases of pulmonary LAM, both of which were ER positive and PR negative. Berger et al22 reported another two cases: one was ER and PR positive, and one was ER positive and PR negative. Other reports are of single patients, including one who was ER negative, with 15% of the nuclei PR positive,7 and one with strong staining for PR and weak staining for ER in. 80% of the smooth muscle cell nuclei.30 There is only one previous report in which both lung and angiomyolipoma cells from the same patient were examined: the pregnant patient studied by 28 Clinical Investigations

5 Table 2 Literature Review of ER and PR Immunoreactivity in LAM-Associated Angiomyolipomas and Pulmonary LAM* AML LAM First Author Year ER PR ER PR Colley /1 1/1 Berger /2 1/2 Ohori /5 0/5 Johnson /1 1/1 Kinoshita /2 0/2 Tawfik /1 0/1 1/1 0/1 Logginidou This report 10/12 12/12 2/5 5/5 Total 10/13 (77) 12/13 (92) 8/17 (47) 8/17 (47) *Data are presented as No. of positive specimens/no. tested, unless otherwise indicated; see Table 1 for abbreviation. Data are presented as No. of positive specimens/no. tested (percent of total tested). Tawfik et al. 21 Pulmonary LAM cells from this patient had weak immunoreactivity for ER and no immunoreactivity for PR. 21 The clinical and genetic distinctions between LAM and TSC have been debated. 31 Chromosome 16p13 loss of heterozygosity in the region of the TSC2 gene has been found in angiomyolipomas from women with LAM, 13 suggesting that these angiomyolipomas may result from mutations in both copies of the TSC2 gene. We have previously found that 55% of TSC-associated angiomyolipomas are PR positive. 25 Among angiomyolipomas from patients who do not have either TSC or LAM, PR immunoreactivity was present in only 7%. 25 The PR positivity in the LAM-associated angiomyolipomas could be a further indication of common genetic and biological mechanisms underlying TSC and sporadic LAM. However, in our previous study of TSC angiomyolipomas, none were ER positive, 25 while 83% of the LAM angiomyolipomas were ER positive. This could indicate a true biological difference between the expression of ER and PR in angiomyolipomas from women with TSC vs those with sporadic LAM. Alternatively, it is possible that angiomyolipomas from women who develop LAM (whether or not they have TSC) are more likely to be ER immunoreactive than angiomyolipomas from women who do not develop LAM. The TSC patients in our previous study were not ascertained for the presence or absence of LAM. However, it is important to note that these two studies used different ER antibodies (a Biogenex MoAb in the current study and a Dako [Carpinteria, CA] MoAb in the previous study). There are currently no in vitro or in vivo data to support the hypothesis that the proliferation of LAM cells is dependent on hormonal stimuli. There is, however, considerable indirect evidence that smooth muscle cell growth in pulmonary LAM is influenced by steroid hormones. This evidence includes the following: the occurrence of LAM almost exclusively in women; the presence of ER and PR in pulmonary LAM cells 19,22 24,28 ; the reports of exacerbation of LAM during pregnancy 21,32,33 and during estrogen therapy 34 ; and the apparent therapeutic response to exogenous hormonal agents in some patients, as reviewed by Kalassian et al 2 and Sullivan. 3 Our data indicate that the smooth muscle cells in angiomyolipomas may also be capable of responding to hormonal stimuli. Clinical improvement or stabilization of pulmonary symptoms during hormonal therapy for LAM has occurred in many cases and has been recently reviewed by Sullivan. 3 There are no randomized or controlled studies of LAM progression during hormonal therapy. Such studies are likely to be challenging to perform because of the difficulty in quantitating disease progression. Our data suggest that angiomyolipomas, like pulmonary LAM, may be responsive to hormonal therapy. It is possible, therefore, that in women with sporadic LAM who have angiomyolipomas, the size of the angiomyolipoma could be used as a marker of clinical response to hormonal therapy. The size of angiomyolipomas can be objectively determined, and many angiomyolipomas have measurable growth in 1 year There are currently no reports of the growth rates of angiomyolipomas during hormonal therapy for LAM. In summary, we studied 12 angiomyolipomas from women with sporadic LAM and found that 10 were ER immunoreactive and all 12 were PR immunoreactive. Of five specimens of pulmonary LAM, we found that two were ER immunoreactive and all five were PR immunoreactive. It is important to note that the antibody used in this study was raised against the estrogen receptor. Whether estrogen receptor is present in either pulmonary LAM cells or angiomyolipoma cells is not known. Elucidation of the role of steroid hormone receptors in the pathogenesis of LAM could contribute to both our understanding of CHEST / 117 / 1/ JANUARY,

6 the hormonal factors that contribute to smooth muscle proliferation in LAM, and also to the selection of appropriate hormonal therapy for women with LAM. ACKNOWLEDGMENT: We are grateful to Dr. Alfred Knudson, Dr. Frank McCormack, Ms. Sue Byrnes, and the LAM Foundation for their ongoing support; to William A. Petri and Drs. Ann Petri, Rebecca Raftogianis, and Leena Khare for critical review of the manuscript; and to Ms. Catherine Thompson for expert secretarial assistance. References 1 Van Stossel E. Uber muskulare cirrhose der lunge. Beitr Klin Tuberk 1937; 90: Kalassian KG, Doyle R, Kao P, et al. Lymphangioleiomyomatosis: new insights. Am J Respir Crit Care Med 1997; 155: Sullivan EJ. Lymphangioleiomyomatosis: a review. Chest 1998; 114: Workshop NHLBI. Report of Workshop on Lymphangioleiomyomatosis. Am J Respir Crit Care Med 1999; 159: Chu SC, Horiba K, Usuki J, et al. Comprehensive evaluation of 35 patients with lymphangioleiomyomatosis. Chest 1999; 115: Taylor JR, Ryu J, Colby TV, et al. Lymphangioleiomyomatosis: clinical course in 32 patients. N Engl J Med 1990; 323: Johnson SE, Davey DD, Cibull ML, et al. Lymphangioleiomyomatosis. Am Surg 1993; 59: Torres VE, Bjornsson J, King BF, et al. Extrapulmonary lymphangioleiomyomatosis and lymphangiomatous cysts in tuberous sclerosis complex. Mayo Clin Proc 1995; 70: Castro M, Shepherd CW, Gomez MR, et al. Pulmonary tuberous sclerosis. Chest 1995; 107: Bernstein SM, Newell JD, Adamczyk D, et al. How common are renal angiomyolipomas in patients with pulmonary lymphangiomyomatosis? Am J Respir Crit Care Med 1995; 152: Kerr LA, Blute ML, Ryu JH, et al. Renal angiomyolipoma in association with pulmonary lymphangioleiomyomatosis: forme fruste of tuberous sclerosis. Urology 1993; 41: Maziak DE, Kesten S, Rappaport DC, et al. Extrathoracic angiomyolipomas in lymphangioleiomyomatosis. Eur Respir J 1996; 9: Smolarek TA, Wessner LL, McCormack FX, et al. Evidence that lymphangiomyomatosis is caused by TSC2 mutations: chromosome 16p13 loss of heterozygosity in angiomyolipomas and lymph nodes from women with lymphangiomyomatosis. Am J Hum Genet 1998; 62: Kapur RP, Bigler SA, Skelly M, et al. Anti-melanoma monoclonal antibody HMB-45 identifies an oncofetal glycoconjugate associated with immature melanosomes. J Histochem Cytochem 1992; 40: Bacchi CE, Bonetti F, Pea M, et al. HMB-45: a review. Appl Immunohistochem 1996; 4: Pea M, Bonetti F, Zamboni G, et al. Melanocyte-marker- HMB-45 is regularly expressed in angiomyolipoma of the kidney. Pathology 1991; 23: Hoon V, Thung SN, Kaneko M, et al. HMB-45 reactivity in renal angiomyolipoma and lymphangioleiomyomatosis. Arch Pathol Lab Med 1994; 118: Chan JK, Tsang WY, Pau MY, et al. Lymphangiomyomatosis and angiomyolipoma: closely related entities characterized by hamartomatous proliferation of HMB-45-positive smooth muscle. Histopathology 1993; 22: Bonetti F, Chiodera PL, Pea M, et al. Transbronchial biopsy in lymphangiomyomatosis of the lung: HMB45 for diagnosis. Am J Surg Pathol 1993; 17: Kaiserling E, Krober S, Xiao JC, et al. Angiomyolipoma of the kidney: immunoreactivity with Hmb-45; light- and electronmicroscopic findings. Histopathology 1994; 25: Tawfik O, Austenfeld M, Persons D. Multicentric renal angiomyolipoma associated with pulmonary lymphangioleiomyomatosis: case report, with histologic, immunohistochemical, and DNA content analysis. Urology 1996; 48: Berger U, Khaghani A, Pomerance A, et al. Pulmonary lymphangioleiomyomatosis and steroid receptors. Am J Clin Pathol 1990; 93: Graham ML, Spelsberg TC, Dines DE, et al. Pulmonary lymphangiomyomatosis: with particular reference to steroid receptor assay studies and pathologic correlation. Mayo Clin Proc 1984; 59: Brentani MM, Carvalho CR, Saldiva PH, et al. Steroid receptors in pulmonary lymphangiomyomatosis. Chest 1984; 85: Henske EP, Ao X, Short MP, et al. Frequent progesterone receptor immunoreactivity in tuberous sclerosis-associated renal angiomyolipomas. Mod Pathol 1998; 11: Maiorana A, Cavallari V, Bagni A, et al. Nuclear areas in breast cancer: relationship with estrogen and progesterone receptor expression. Anal Cell Pathol 1996; 11: Makris A, Powles TJ, Allred DC, et al. Changes in hormone receptors and proliferation markers in tamoxifen-treated breast cancer patients and the relationship with response. Breast Cancer Res Treat 1998; 48: Ohori NP, Yousem SA, Sonmez-Alpan E, et al. Estrogen and progesterone receptors in lymphangioleiomyomatosis, epithelioid hemangioendothelioma, and sclerosing hemangioma of the lung. Am J Clin Pathol 1991; 96: Kinoshita M, Yokoyama T, Higuchi E, et al. Hormone receptors in pulmonary lymphangiomyomatosis. Kurume Med J 1995; 42: Colley MH, Geppert E, Franklin WA. Immunohistochemical detection of steroid receptors in a case of pulmonary lymphangioleiomyomatosis. Am J Surg Pathol 1989; 13: Bonetti F Chiodera P. Lymphangioleiomyomatosis and tuberous sclerosis: where is the border? Eur Respir J 1996; 9: Hughes E, Hodder RV. Pulmonary lymphangioleiomyomatosis complicated by pregnancy. J Reprod Med 1987; 32: Yockey CC, Riepe RE, Ryan K. Pulmonary lymphangioleiomyomatosis complicated by pregnancy. Kans Med 1986; 87: , Shen A, Iseman MD, Waldron JA, et al. Exacerbation of pulmonary lymphangioleiomyomatosis by exogenous estrogens. Chest 1987; 91: Lemaitre L, Robert Y, Dubrulle F, et al. Renal angiomyolipoma: growth followed up with CT and/or US. Radiology 1995; 197: van Baal JG, Smits NJ, Keeman N, et al. The evolution of renal angiomyolipomas in patients with tuberous sclerosis. J Urol 1994; 152: Ewalt DH, Sheffield E, Sparagana SP, et al. Renal lesion growth in children with tuberous sclerosis complex. J Urol 1998; 160: Clinical Investigations