Bronchopulmonary carcinoid in multiple endocrine neoplasia type 1

Abstract

BACKGROUND

Multiple endocrine neoplasia type 1 (MEN 1) is an autosomal-dominant syndrome associated with neoplasia of pituitary, pancreas, parathyroid, and foregut lineage neuroendocrine tissue. Although enteropancreatic carcinoid has been well described in patients with MEN 1, it was believed that bronchopulmonary carcinoid was relatively uncommon, occurring in approximately 5% of patients. It is unclear whether the increased screening of asymptomatic patients with MEN 1 will facilitate early diagnosis of this tumor and improve patient prognosis.

METHODS

The authors reviewed the patient records and, when available, thoracic computed tomographic (CT) images of 129 MEN 1-affected adult members of a single family to determine the prevalence and prognosis of bronchopulmonary nodules and carcinoid.

RESULTS

Among 129 patients, a diagnosis of bronchopulmonary carcinoid was noted in the records for 6 individuals (1 male and 5 females; 5%). Thoracic CT scans also were available for review from 32 of those patients. Twelve patients (38%) had pulmonary nodules evident on CT scans. Only hypergastrinemia was significantly more common in patients with pulmonary nodules; otherwise, the spectrum of neoplasia was similar between individuals with and without pulmonary lesions. Histologic diagnoses were available in four patients (three female) with abnormal CT images, and carcinoid was confirmed in each patient. No deaths or distant metastases occurred among the patients despite long-term follow-up (mean, 127 months).

CONCLUSIONS

The findings suggested that bronchopulmonary carcinoid is more prevalent in patients with MEN 1 than was recognized previously. Furthermore, the diagnosis did not appear to portend a poor prognosis in the majority of affected patients. Cancer 2005. © 2004 American Cancer Society.

Multiple endocrine neoplasia type 1 (MEN 1) is an autosomal-dominant tumor syndrome arising from inactivating mutations of a putative tumor suppressor gene (MEN1) situated on chromosome 11q13.1 The MEN1 gene comprises 10 exons (only exons 2–9 are transcribed) and encodes a 610-amino-acid protein product termed menin.2 Menin appears to behave as a transcription regulator, although its exact mechanism of action and its function in cellular biology remain to be elucidated.3, 4 Mutation sequencing in MEN 1-affected families has revealed in excess of 250 different causative mutations with no particular genotype-phenotype correlation.5 However, there are reports of intrafamilial clustering in the expression of some MEN 1 phenotypes (prolactinoma and enteropancreatic lesions) as well as variations in the frequency of individual tumor syndromes in different MEN 1 series.6–10

Although MEN 1 is characterized by parathyroid hyperplasia (primary hyperparathyroidism) and neoplasia of enteropancreatic neuroendocrine, pituitary, and adrenal tissue,5, 11, 12 it also is recognized that other tumors, such as collagenomas, lipomas, and angiofibromas of the skin, as well as foregut carcinoids (thymic, gastric, and bronchial) also occur.13–17 Unlike thymic carcinoids, which, in the setting of MEN 1, develop predominantly in males (> 95%) age > 40 years (> 70%) with a history of tobacco smoking (70%) and carry a poor prognosis, it is believed that bronchopulmonary carcinoids occur infrequently (< 5% of patients with MEN 1) and predominantly in females.1, 15, 16, 18 However, the prevalence, natural history, and role of computed tomographic (CT) screening for pulmonary lesions on the outcome of bronchopulmonary carcinoid in the setting of MEN 1 have not been evaluated systematically and remain unclear.

patients with well differentiated, non-MEN 1, sporadic pulmonary carcinoid generally have an excellent prognosis, with 5-year survival rates in excess of 90%. In contrast, patients with less well differentiated, atypical pulmonary carcinoids have 5-year survival rates between 40% and 60%.19 Atypical carcinoids have an aggressive clinical course, metastasizing to local lymph nodes in 30–50% patients cases, and some authors have classified these as well differentiated neuroendocrine carcinomas.19 It is believed that chromosomal losses in the region of chromosome 11q13 play a significant part in the pathogenesis of both types of pulmonary carcinoids: One recent study showed a inactivated copies of the MEN1 gene in 4 of 11 patients with sporadic pulmonary carcinoid.20 Unlike midgut carcinoids, foregut carcinoids are not associated with serotonin hypersecretion or with the carcinoid syndrome.21

The Tasman 1 MEN 1 kindred is a well characterized MEN 1 family with > 160 affected members identified to date.6, 7, 10, 22 It has been shown that a novel splice site (IVS2–3 [C-G]) mutation of the MEN1 gene is responsible for the clinical syndrome in this kindred.22 Unlike most other published series describing the phenotypic features of MEN 1 (which used data from institutional case series, representing patients from multiple families with heterogeneous underlying MEN1 mutations and a relatively high proband-to-asymptomatic patient ratio), the Tasman 1 kindred provides a unique opportunity to assess the different phenotypic manifestations of MEN 1 in a very large kindred with a common founder. Active screening of asymptomatic family members has been offered to kindred members for > 20 years.10, 22 In the current retrospective review, we assessed the prevalence and natural history of pulmonary nodules identified during MEN 1 screening and evaluated their relation to histopathologically confirmed bronchopulmonary carcinoid in this very large MEN 1 kindred.

MATERIALS AND METHODS

Members of the Tasman 1 MEN 1 kindred are descendants of a common English ancestor who immigrated to Tasmania in the early 19th century. After the identification of 2 index cases in 1982, the overall pedigree was traced.10 A comprehensive biochemical and radiologic surveillance program has been available for members of this family since that time. Audit of clinical records is undertaken periodically to determine the natural history of MEN 1 and gain insights into the optimal management of affected individuals. This process and the publication of results from these studies has been approved by the Royal Hobart Hospital Research Ethics Committee.

In the current study, we retrospectively reviewed the records of all (living and deceased) adult Tasman 1, MEN 1 patients with a confirmed diagnosis of MEN 1. Individuals were selected based on the following criteria: 1) adults (age ≥ 18 years) with a confirmed genetic diagnosis of MEN 1, 2) previously performed chest CT imaging, and 3) the availability of CT images for review. In total, 129 patients with MEN 1 fulfilled Criteria 1; and, among those 129 patients, 36 patients satisfied Criteria 2, and 32 patients satisfied Criteria 3. CT radiographs were sought for all patients who had an history of chest imaging and, when available, were reviewed by a single experienced radiologist who was blinded to the patients' clinical and biochemical profile.

Individuals either had the Tasman 1 genotype confirmed by MEN1 gene sequencing or were obligate carriers of the mutation, as described previously by Burgess et al.22 Screening has been offered routinely to gene carriers that includes annual biochemical assessment of serum ionized calcium, parathyroid hormone, prolactin, and gastrin. The analytical methods and biochemical definitions for hyperparathyroidism and hypergastrinemia have been described previously.7 Radiologic assessment comprised of ultrasound examination of the pancreas, CT imaging of the chest and abdomen, and magnetic resonance imaging (MRI) of the pituitary are under taken annually, every 3–5 years, and every 5 years, respectively. Most CT scans were obtained with a Hi Speed Advantage Plus CT scanner (General Electric, Milwaukee, WI), most MRI studies were obtained with a Picker HPQ 1.5-T scanner (Picker, OH), and pancreatic ultrasound studies were obtained with an Acuson Sequoia (Acuson Corp., Mountain View, CA).

Data were analyzed using Student t tests for normally distributed variables and chi-square tests for nonparametric data. Where appropriate, results are expressed as the mean ± standard error of mean.

RESULTS

Overall Eligible Adults

Among 129 adults with a confirmed MEN 1 genotype, thoracic CT images were obtained from 36 patients (28%), and radiographs were available for review in 32 patients (25%) (Table 1). A review of records and death certificates among the 93 individuals who were not screened by thoracic CT revealed 2 patients with pulmonary carcinoid. One patient had a histologically confirmed tumor diagnosed at age 45 years that was treated surgically and developed a subsequent lesion, which was diagnosed at age 65 years and was managed with radiotherapy. The patient died at age 71 years of unrelated causes. The second historic case was a female patient who died at age 50 years in 1981, possibly as a result of complications related to bronchopulmonary carcinoid. However, a review of available records did not permit full confirmation that pulmonary carcinoid was the primary cause of death in this patient, because enteropancreatic malignancy with lymph node metastases was coexistent.13 For both of those patients, the original histopathologic material and CT images were not available (patients deceased > 20 years) for confirmation of the reported diagnosis of carcinoid, so these two patients were not included in the more detailed evaluation described below.

Table 1. Overview of Tasman 1 Participants Suitable for Study Inclusion

Characteristic	Total	Pulmonary nodules on CT scan	Confirmed histology of pulmonary carcinoid	Confirmed death due to pulmonary carcinoid
MEN 1: multiple endocrine neoplasia type 1; CT: computed tomography; NA: not applicable. a Individuals age ≥ 18 years with MEN 1 confirmed by gene sequencing or obligate carrier based on pedigree. b Death occurred prior to recognition of MEN 1 in the family. Available information indicates death at age 50 years, possibly from pulmonary carcinoid. c Chest CT scans were reviewed by a radiologist. d Chest CT scans were unavailable for review by a radiologist, and only the original reports were available for review.
Total adults with MEN 1a	129	13	6	0b
No history of chest CT	93	NA	2	0b
Currently living	53	NA	0	0
Deceased	40	NA	2	0b
CT scans reviewedc	32	12	4	0
Currently living	27	9	4	0
Deceased	5	3	0	0
CT report not reviewedd	4	1	0	0
Currently living	4	1	0	0
Deceased	0	0	0	0

Radiograph Review Subgroup

Among the patients who had radiographs available for review (n = 32 patients), pulmonary nodules were identified in 12 patients (38%) (Table 2). The indication for CT imaging was asymptomatic screening for intrathoracic neoplasia in 29 patients (9 of whom exhibited lesions, including 2 patients with histologically confirmed carcinoid), follow-up of preexisting bronchopulmonary carcinoid in 2 patients, and evaluation of new chest symptoms in 1 patient. Among the patients who had nodules, 83% were female compared with 55% females in the group without nodules (Table 2). Nodule size ranged from 2 mm to 3 cm, and, in 9 patients (75%), the lesions were multicentric. The average age of patients who had pulmonary nodules was 43.3 years ± 4.5 years compared with 39.7 years ± 3.3 years (nonsignificant) among patients who were without pulmonary nodules (Table 2). Of the 12 patients who had pulmonary nodules, a tissue diagnosis was obtained in 6 patients (4 patients had bronchopulmonary carcinoid confirmed, 1 patient was diagnosed with metastatic disease arising from an enteropancreatic malignancy, and the sixth patient was diagnosed with primary nonsmall cell lung carcinoma). In the remaining six patients, the pulmonary nodules remained small, of stable size, and asymptomatic. For these reasons, those patients have not undergone resection or biopsy. Aside from gastrinoma (which was more common in patients who had pulmonary nodules; 50% vs. 15%; P = 0.03), there were no significant differences in expression of the MEN 1 phenotype between groups (Table 2).

Table 2. Clinical and Biochemical Characteristics of 32 Patients with Multiple Endocrine Neoplasia Type 1 and Reviewable Thoracic Computed Tomographic Images

Variable	Nodules (n = 12)	No nodules (n = 20)	P value
Scan neg: age at most recent scan for individuals with scans that were negative for pulmonary lesions; PHPT: primary hyperparathyroidism.
Female gender	10 (83.0%)	11 (55.0%)	0.10
Age (yrs) (diagnosis/scan neg)	43.45 ± 4.5	39.7 ± 3.3	0.51
PHPT	12 (100.0%)	18 (90.0%)	0.26
Hypergastrinemia	6 (50.0%)	3 (15.0%)	0.03
Pituitary adenoma	6 (54.5%)	5 (25.0%)	0.15
Pancreatic adenoma	11 (92.0%)	13 (65.0%)	0.09
Adrenal adenoma	4 (33.0%)	5 (25.0%)	0.61
Thymic carcinoid	0 (0.0%)	2 (10.0%)	0.26
Gastroid carcinoid	4 of 8 tested	2 of 8 tested	0.30

Original radiographs were not available for four additional patients (two males and two females) who had a history of prior thoracic CT imaging. Among these patients, the written CT scan report described pulmonary nodules in a female patient age 67 years.

Of the 4 patients who required surgery, tumor resection was undertaken either due to bronchial obstructive symptoms or because of a significant increase in lesion size (tumor size > 3 cm) during radiologic follow-up (Table 3). Of those four patients, one patient presented with symptoms suggestive of bronchial obstruction, requiring prompt resection of the pulmonary lesion; 1 patient had a single, large (> 3 cm), asymptomatic lesion on initial CT chest screening that required resection; and the 2 remaining patients underwent surgery, because their bronchopulmonary lesions showed substantial size increases between interval radiologic imaging. In each patient, bronchopulmonary carcinoid was confirmed histopathologically. Somatostatin receptor scintigraphy was performed in one patient and demonstrated strong receptor binding for radiolabeled octreotide.

Table 3. Clinical, Biochemical, and Histopathologic Characteristics of Four Patients with Histologically Confirmed Bronchopulmonary Carcinoida

Characteristic	Patient no.
	1	2	3	4
XRT: radiation therapy; PHPT: primary hyperparathyroidism; MEN 1: multiple endocrine neoplasia type 1. a Not including 1 patient who died prior to recognition of MEN 1 in the family. Available information indicates death at age 50 years, possibly from pulmonary carcinoid (female patient).
Gender	Female	Female	Female	Female
Age at carcinoid diagnosis (yrs)	20	25	52	35
Clinical presentation	Symptoms	No symptoms	No symptoms	No symptoms
Treatment	Surgery	Surgery, XRT	Surgery	Surgery
Metastases	None	Lymph node	None	None
Multiple lesions
At diagnosis	No	Yes	No	No
During follow-up	Yes	Yes	No	Yes
Survival since diagnosis (mos)	192	420	87	36
MEN 1-related neoplasia
PHPT	Yes	Yes	Yes	Yes
Hypergastremia	Yes	No	Yes	Yes
Adenoma
Pancreas	Yes	Yes	Yes	Yes
Pituitary	Yes	Yes	No	Yes
Adrenal	No	No	Yes	Yes
Thymic carcinoid	No	No	No	No
Gastric carcinoid	Yes	No	No	Yes

One of the four surgical patients exhibited local lymph node involvement at diagnosis. This patient also had a second asymptomatic pulmonary nodule at the time of diagnosis that was resected during the initial surgery and was confirmed as a bronchopulmonary carcinoid. This patient received radiotherapy for local metastatic disease; however, none of the other patients required adjuvant chemotherapy, biotherapy (somatostatin analogues, interferon α), or radiotherapy—although two patients currently are being treated with somatostatin analogues for gastrinoma (Table 3).

The age of patients at the time they were diagnosed with pulmonary carcinoid varied widely between 20 years and 55 years (mean, 37.3 years). Parathyroid hyperplasia (hyperparathyroidism) and gastroenteropancreatic neoplasia were present uniformly in all 4 patients with bronchopulmonary carcinoid (Table 3), adrenal neoplasia was present in 2 patients, and pituitary neoplasia was present in 3 patients. Hypergastrinemia was detected in two patients, and gastric carcinoid was present in two patients. Thymic carcinoid was not found in any of the four patients who had bronchopulmonary carcinoid. Two patients developed additional, new, asymptomatic pulmonary nodules during follow-up (mean, 127 months; range, 22–264 months). One patient developed multiple pulmonary nodules; and, at bronchoscopy, several endobronchial lesions were biopsied and were confirmed as pulmonary carcinoid. No deaths or distant metastases in relation to bronchopulmonary carcinoid have been observed in this group to date (Table 3).

DISCUSSION

Recent advances have provided extensive insights into the pathogenesis and natural history of MEN 1-associated neoplasia. This has facilitated improved strategies for the management of these disorders and appears to have reduced the morbidity as well as the mortality associated with MEN 1.13, 16, 23 However there is a paucity of knowledge regarding the natural history of MEN 1-related foregut carcinoids, in particular, bronchial carcinoids. Prior studies have suggested that foregut carcinoids may be an important determinant of long-term survival in patients with MEN 1.13

Based on historic records, it was believed that all intrathoracic carcinoids in the Tasman 1 kindred carried a poor prognosis.13 Although this remains true for thymic carcinoid, the current study suggests that bronchopulmonary carcinoid has a much more indolent natural history.10, 13, 24 Results of the current study suggest that bronchopulmonary carcinoid is a common phenotypic manifestation of MEN 1 with a minimum histologically confirmed prevalence of 5% and a possible maximum of 31% (10 of 32 patients) based on radiographic evidence of pulmonary nodules (although a proportion of the nodules evident on CT scans, but not biopsied, are likely to result from processes unrelated to carcinoid).

However, limitations of the current study include retrospectivity and the potential for patient enrichment by referral bias—individuals with symptomatic disease are more likely to be referred to the MEN 1 clinic than individuals who are asymptomatic. Although thoracic CT imaging is offered every 5 years during adulthood, it is not possible to exclude preferential participation in the screening program of individuals with symptoms. Similarly, the prevalence of pulmonary nodules in the general population is unclear and has not been evaluated in this study. It is probable that a proportion of the small nodules identified on CT imaging but not biopsied in our study represented nonneoplastic abnormalities unrelated to MEN 1.

The high prevalence of pulmonary nodules in our series, in conjunction with their small size on presentation (< 1 cm) and their multifocal and peripheral distribution, obliged us to use surgical therapy judiciously. Furthermore, other forms of investigation, including bronchoscopy and CT-guided fine-needle biopsy, have only limited use in these situations (many nodules are too peripheral for visualization at bronchoscopy and are too small for biopsy under CT guidance). Given the apparently indolent natural history of MEN 1-related bronchopulmonary carcinoid observed in the current study, radiologic surveillance may be sufficient for a proportion of patients who have pulmonary nodules identified during MEN 1 screening. By incorporating a pragmatic approach utilizing clinical symptoms and a size criteria, (e.g., 3 cm or rapid interval growth), surgical intervention in patients with MEN 1 who are undergoing thoracic CT screening may be minimized.

Like enteropancreatic neuroendocrine lesions and gastric carcinoids in patients with MEN 1, bronchopulmonary carcinoids are multicentric and develop both synchronously and metasynchronously in the same patient.25, 26 Three of the 4 patients (75%) with pulmonary carcinoid in our series have developed additional new pulmonary lesions on follow-up, with carcinoid confirmed at biopsy in 1 patient. Moreover, in the group of patients who had pulmonary nodules evident on CT scans, 75% of patients had multiple nodules. The multicentric nature of bronchopulmonary carcinoid in MEN 1 underscores the need for judicious use of surgical intervention. We did not incorporate somatostatin receptor scintigraphy or ¹⁸F-fluorodeoxyglucose positron emission tomography scanning in further stratifying disease extent in our patients; however, these techniques also may be useful in this regard.27, 28

Previous reports of MEN 1 kindreds have suggested that, unlike hyperparathyroidism and gastroenteropancreatic disease, which occur with equal frequency in both genders, bronchial carcinoids occur more frequently in female MEN 1 patients.12 In agreement with prior reports, results from the current study also indicated that a significant preponderance of females among those who were diagnosed with bronchopulmonary carcinoid. This parallels the predilection of thymic carcinoid for males. In both instances, the basis for gender dichotomy remains unclear, but elucidation of the underlying mechanism may offer important insights into the genesis of both MEN 1 and sporadic foregut neuroendocrine tumors.

It is known that hypergastrinemia, which is a potent stimulus for neuroendocrine cell proliferation, is associated with disseminated gastroenteropancreatic tumors.25, 26 The enterochromaffin-like cells of the gastric oxyntic mucosa are prone to hyperplasia when they are exposed to persistently elevated levels of gastrin. Transformation through hyperplasia, dysplasia, and (potentially) progression to malignant gastric carcinoid tumors may occur.7, 29–31 In the current study, hypergastrinemia was more common in the subgroup of patients with pulmonary nodules. Although this may reflect the older age of patients with nodules (the likelihood of hypergastrinemia in MEN 1 increases with age) or common antecedent determinants, it is possible that elevated levels of gastrin may play a role in the transformation of pulmonary neuroendocrine precursor cells, which are believed to be the progenitor of bronchial carcinoid. The potential impact of lowering serum gastrin levels with antisecretory agents, such as the somatostatin analogues, is unclear but is worthy of further study.32

In conclusion, pulmonary nodules are a frequent finding on thoracic CT imaging in patients with MEN 1, with bronchopulmonary carcinoid an important underlying diagnosis. Based on our observations, bronchopulmonary carcinoid may occur in at least 5% of adults with MEN 1 and up to a possible maximum of 31% of adults with MEN 1, with the gender-specific prevalence higher in females and lower in males. Tumors often are multicentric, exhibiting both synchronous and metasynchronous occurrence. Unlike thymic carcinoid, which has an aggressive clinical course and carries a poor prognosis, bronchopulmonary carcinoid that occurs in the majority of MEN 1 patients behaves indolently, albeit with the potential for local mass effect, metastasis, and recurrence after resection. An empirical approach incorporating the presence of symptoms, the rate of tumor expansion, and absolute lesion size may prove helpful in selecting those patients with pulmonary lesions identified during MEN 1 screening who merit surgical resection. However, prospective, long-term studies are needed to clarify further the natural history, pathogenesis, and optimal management of this not uncommon MEN 1-related tumor.

Acknowledgements

The authors acknowledge the substantial contribution made by the late Professor J. J. Shepherd to multiple endocrine neoplasia type 1 (MEN 1) research, particularly in relation to the identification and characterization of Tasmanian MEN 1 kindreds.