This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Granberg, D. Articles by Skogseid, B. Search for Related Content PubMed PubMed Citation Articles by Granberg, D. Articles by Skogseid, B.

Prognostic Markers in Patients with Typical Bronchial Carcinoid Tumors¹

Departments of Medicine (D.G., K.Ö., B.S.) and Pathology (E.W.), University Hospital, S-751 85 Uppsala, Sweden

Address all correspondence and requests for reprints to: Dr. Dan Granberg, Department of Endocrine Oncology, Division of Internal Medicine, University Hospital, S-751 85 Uppsala, Sweden. E-mail: dan.granberg{at}medicin.uu.se

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Typical bronchial carcinoids are usually considered fairly benign tumors. Metastases do however occur, and up to 10% of the patients ultimately die from their disease. To identify prognostic markers, we immunostained 43 typical bronchial carcinoids with antibodies against 8 possibly relevant hormones, oncogenes, tumor suppressor genes, adhesion molecules, and proliferation markers. Altogether 12 patients (28%) had metastatic disease, of whom 10 had regional lymph node metastases at diagnosis. Distant metastases have occurred in 5 patients (12%); all of these have died from their disease. Patients with high expression of Ki-67 had shorter survival time (P < 0.01). None of the immunostained hormones correlated to distant metastases or shorter survival time, but gastrin-releasing peptide correlated to metastatic disease (P < 0.05). All patients who died had CD44-negative tumors (P < 0.001). Nuclear nm23 staining correlated to decreased risk for metastatic disease and distant metastases per se (P < 0.01). Bcl-2 and p53 were associated with increased risk for distant metastases (P < 0.05 and P < 0.01, respectively). We conclude that some patients with typical bronchial carcinoids die from their disease and that gastrin-releasing peptide, Bcl-2, and p53 may be of importance for the malignant transformation of the tumor. Moreover, CD44, nm23, and Ki-67 may give valuable prognostic information and help identify the patients at risk of disease-related death.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
LUNG NEUROENDOCRINE tumors have traditionally been classified according to their histological appearance and divided into typical carcinoids, atypical carcinoids (1), and small cell lung carcinomas. Recently, new criteria were proposed for atypical carcinoids, including a mitotic range of 2–10 mitoses/2 mm² (10 high power fields) (2). Bronchial carcinoid tumors frequently produce various combinations of hormones and neuropeptides (3, 4, 5), which, more rarely, can result in different clinical syndromes, such as Cushing’s syndrome, acromegaly, or the carcinoid syndrome.

Typical bronchial carcinoids are generally more benign than atypical carcinoids (1, 2, 6, 7), but both types are able to metastasize to regional lymph nodes or distantly to the liver, bones, or brain. The reported 5-yr survival is 87–94% for typical and 56% for atypical carcinoid patients, and 10-yr survival rates are 87% and 35%, respectively (2, 8).

Little is known about prognostic factors in bronchial carcinoids, except that high mitotic count is unfavorable (2). Lymph node metastases at diagnosis (6) and intensive staining with Ki-67 (9) have also been associated with worsened outcome in patients with bronchial neuroendocrine tumors.

The adhesion molecule CD44 is highly expressed in normal bronchial epithelium and squamous cell lung cancers, whereas small cell lung cancers have only weak or absent expression (10). Bcl-2 is often expressed in small cell lung cancer (11), and decreased survival has been reported in patients with bronchial carcinoids, with overexpression of Bcl-2 and down-regulation of its antagonist, Bax (12).

The more malignant variants of neuroendocrine tumors accumulate mutated p53 protein, which can be detected by positive immunostaining (5, 12, 13). Inactivation of the retinoblastoma gene may have pathogenetic implications for the malignant phenotype and neuroendocrine differentiation in lung tumors (14). In a recent study of neuroendocrine lung tumors no correlation was found between the metastasis suppressor gene nm23 and the occurrence of metastases (15).

To learn more about the tumor biology of typical bronchial carcinoids and identify potential prognostic markers, we immunostained tumor tissue with a panel of antibodies against hormones, oncogenes, tumor suppressor genes, growth factors, Ki-67, and CD44 and compared the results with clinical data.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Patients

All patients admitted to any of the departments at our hospital between 1985 and 1995 with tumors fulfilling the criteria for typical bronchial carcinoid (2) were included in the study. Altogether 43 cases were collected (16 men and 27 women). Their median age was 56 yr (range, 18–75), and median follow-up comprised 74 months (range, 14–330).

In 32 patients the tumor was diagnosed because of respiratory symptoms. In addition, 2 of these patients had carcinoid syndrome. Two more patients developed this syndrome when liver metastases were detectable 12 and 20 yr after primary surgery. Nine cases were diagnosed by routine chest x-ray, and 1 patient presented with ectopic Cushing’s syndrome. Finally, 1 patient, suffering from mental confusion, was operated on for a brain metastasis. Histological examination showed metastasis from a typical bronchial carcinoid.

Altogether 12 patients (28%) displayed metastatic disease. Ten patients (23%) had lymph node involvement at diagnosis, and distant metastases have occurred in 5 patients (12%) up to 20 yr after diagnosis. All 5 patients with distant metastases have died from their disease during the observation period; in addition, 1 patient died from unrelated cause, free from carcinoid recurrence. Five-year survival was 95%, and 10-yr survival was 91%.

Surgery with removal of the primary tumor and lymph node dissection was performed in 42 patients (lobectomy in 30, bilobectomy in 6, pulmectomy in 4, and segmentectomy or wedge resection in the remaining 2), but was denied in the patient presenting with a brain lesion because of extensive intrathoracic tumor growth. One patient subjected to pulmectomy had undergone bronchoscopic extirpation of a carcinoid 26 yr earlier. She is still free from recurrence 9 yr after the last operation. One patient received postoperative radiotherapy. In 3 additional patients, radiotherapy was given against brain or bone metastases. One patient was subjected to liver embolization with gel-foam. Interferon treatment, alone or in combination with octreotide, was given to 7 patients. One patient who showed progressive disease after interferon administration received chemotherapy.

Morphology and immunohistochemistry

Diagnostic criteria for typical bronchial carcinoids according to Travis (2) were applied, mitosis rates were calculated, and all specimens showed less than 2 mitoses/10 high power fields. All specimens were routinely stained with hematoxylin and eosin (Fig. 1A). Immunohistochemical stainings were also performed with the antibodies shown in Table 1 according to the following procedure. Formalin-fixed and paraffin-embedded blocks were cut in 4-µm sections. The sections were deparaffinized and rehydrated, after which endogenous peroxidase was blocked with 0.3% hydrogen peroxide for 30 min. For some of the antibodies, antigen unmasking was made by pretreatment, as described in Table 1. After repeated washing in phosphate saline buffer solution, pH 7.4 (PBS), horse or goat serum diluted 1:5 in PBS was employed to avoid nonspecific secondary antibody binding. Primary antibody diluted in PBS was applied for 1.5 h at room temperature or overnight at 4 C. Secondary antimouse or antirabbit antibody diluted 1:200 was applied for 45 min. The reaction was visualized with an ELITE kit (Vector Laboratories, Inc., Burlingame, CA) with 0.006% hydrogen peroxide as substrate and 3-amino-9-ethyl-carbazol (Sigma, St. Louis, MO) in dimethylsulfoxide as chromogen. The sections were finally counterstained with hematoxylin. A positive control was included in each run. The slides were assessed by one of the authors (E.W.) without knowledge of the clinical parameters. For c-erbB-2 only membranous staining, which correlates to gene amplification (16), was considered positive. The stainings of Bax and CD44 were only graded as positive or negative. The Ki-67 results were expressed as the percentage of positive cells. The remaining stainings were graded as follows: -, negative; +, less than 10% positive cells; ++, 10–50% positive cells; and +++, more than 50% positive cells.

View larger version (160K): [in this window] [in a new window]   Figure 1. A, Typical bronchial carcinoid. Hematoxylin-eosin stain. Magnification, x200. B, Typical bronchial carcinoid. Ki-67 stain; 1.1% cells staining positive. Magnification, x400. C, Typical bronchial carcinoid. Strong positive GRP stain. Magnification, x200. D, Typical bronchial carcinoid. Strong nuclear and weak cytoplasmic nm23 stain. Magnification, x400.

When calculating the statistics, only positive vs. negative staining was compared, except for Ki-67. The ² test or (for Ki-67) the Mann-Whitney U test were used for correlation between antibody expression and metastasizing. Survival time was analyzed by Kaplan-Meier cumulative survival plot and Mantel-Cox log-rank test, except for Ki-67, where the Cox proportional hazards test was used. P < 0.05 was considered significant. The patient who died from an unrelated cause has been omitted from the survival calculations.

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Clinical parameters, morphology, and proliferation

All tumors contained 0–1 mitoses/2 mm² (10 high power fields) and were chromogranin A positive; 40 stained positively for cytokeratin and 10 for S-100 protein. S-100 did not correlate to metastases or mortality. Patients with lymph node metastases found at primary surgery had increased risk for distant metastases (P = 0.04) in addition to shorter survival time (P < 0.001). There was no correlation between sex or age and frequency of metastases or mortality. The patients who were diagnosed on routine chest x-ray had the same prognosis as those presenting with symptoms. Endocrine symptoms at diagnosis did not correlate to prognosis; however, if all patients displaying endocrine symptoms were included, a correlation was found to metastatic disease and mortality. The Ki-67 results are shown in Table 2 and Fig. 1B. Ki-67 did not correlate to lymph node metastases at diagnosis (P = 0.4), distant metastases (P = 0.2), liver metastases (P = 0.6), or metastatic disease (P = 0.4). However, patients with high Ki-67 expression had significantly shorter survival time (P < 0.01). Patients presenting with endocrine symptoms had higher Ki-67 index (P = 0.02) than patients without endocrine symptoms at diagnosis; however, when all patients with endocrine symptoms were compared to those without, the difference was not significant (P = 0.075).

The results are summarized in Tables 2 and 3. Positive staining for gastrin-releasing peptide (GRP; Fig. 1C) was associated with increased risk for metastases (P = 0.04), but did not correlate specifically to lymph node metastases (P = 0.09) or distant metastases per se or to shorter survival (P = 0.25). The tumors of patients displaying endocrine symptoms were more often GRP positive than those of patients without endocrine symptoms (P = 0.02). None of the patients with tumors positive for pancreatic polypeptide has died, but significance was not reached (P = 0.08). Only one patient with distant metastases stained positively for hCG, and a tendency toward protection against distant metastases (P = 0.09) as well as metastatic disease (P = 0.06) was noted. Positive serotonin staining showed a tendency to increased risk for metastatic disease (P = 0.08).

The results of the CD44 stainings are presented in Fig. 2. All tumors from patients with distant metastases stained negatively for CD44. Positive staining for CD44 correlated to decreased risk for distant metastases and mortality (P < 0.001).

View larger version (16K): [in this window] [in a new window]   Figure 2. CD44 vs. mortality. , Positive staining; , negative staining.

The results of the oncogene and tumor suppressor gene protein stainings are summarized in Tables 2 and 4 and Fig. 3. Positive nuclear nm23 staining (Fig. 1D) correlated to decreased risk for distant metastases and metastatic disease (P < 0.01), but not to lymph node metastases (P = 0.09) or survival time (P = 0.16). In addition, positive nuclear nm23 staining correlated to the absence of endocrine symptoms (P < 0.01). Cytoplasmic nm23 staining did not correlate to any of the clinical parameters.

View larger version (16K): [in this window] [in a new window]   Figure 3. Nuclear nm23 staining vs. mortality. , Positive staining; , negative staining.

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
The comparably high mortality rate (12%) for our patients with typical bronchial carcinoids has only been approached in one other large study (17), whereas some previous studies report no disease-related deaths despite sufficient follow-up time (7, 18). However, samples comprising only referrals to surgical clinics could represent selection of patients with less advanced disease. Moreover, in one study reporting surprisingly low mortality (1.1%), several patients with distant metastases at diagnosis were excluded from the survival calculations (19). As recurrences and metastases may develop decades after initial diagnosis, these patients should be followed for a long time. Special attention should be paid if lymph node metastases are present at diagnosis or the proliferative activity of the tumor assessed by Ki-67 expression is high, as these patients have decreased survival time. A prognostic implication for Ki-67 has previously been found in pancreatic endocrine tumors (20).

In accordance with other studies (3, 4) many of our tumors stained positive for multiple hormones, although endocrine symptoms were infrequent. In our study hCG showed near-significant inverse correlation to distant metastases and metastatic disease. GRP, which stains positively in a considerable proportion of bronchial carcinoids, may act as a growth factor for normal bronchial epithelial cells (21) and small cell lung cancer (22). In our study, GRP correlated to metastatic disease.

None of the patients with distant metastases showed positive CD44 staining, which correlated strongly to decreased risk for distant metastases and death. If this means that lack of CD44 expression has a causative role for malignant behavior in bronchial neuroendocrine tumors or merely accompanies poor differentiation is unclear, but immunostaining for CD44 may be a valuable prognostic marker in typical bronchial carcinoids. Recurrences are uncommon in patients with CD44-positive tumors. We previously reported that splice variants of CD44 show individual correlations to distant metastases and survival in patients with typical bronchial carcinoids (23).

The findings of increased incidence of distant metastases and higher mortality in patients with Bcl-2-positive tumors are consistent with the conclusion of Brambilla et al. (12) that apoptosis may have prognostic importance in these tumors. We were able to detect positive p53 staining in a few tumors, and positive staining correlated strongly to distant metastases and decreased survival. However, we have chosen to consider the tumor positive even if a limited number of cells were immunoreactive. Weak positive staining for p53 has been described previously in 21% of the tumors (13).

The protective effect against metastases suggested for nm23 is usually attributed to the H1 form. In a recent study of lung tumors, eight of nine bronchial carcinoids showed positive cytoplasmic staining for nm23-H1, whereas nuclei were negative, and no correlation to clinical parameters could be recorded (15). In contrast, we found that positive nuclear staining (86%) correlated inversely to distant metastases and death, and that cytoplasmic staining showed no correlation to the clinical parameters. Despite the fact that our antibody recognized both nm23-H1 and H2, the indicated antimetastatic effect of nm23 in our typical bronchial carcinoids is consistent with the observations in other cancer types (24, 25, 26). Although the mechanism of action is still unclear, our results demonstrate that positive nuclear staining with nm23 may give valuable prognostic information in bronchial carcinoids.

In summary, the subdivision of bronchial carcinoids into typical and atypical has historically formed the basis for assessment of prognosis in these patients. Our data, however, indicate that a small proportion of typical carcinoids are indeed malignant and may alter the life expectancy of the patient. The results from this study suggest that analysis of CD44, nm23, Ki-67, Bcl-2, and p53 might provide information about the biology of the individual typical bronchial carcinoid tumor and contribute to the prognostic evaluation of the patient. As for the nearly significant markers GRP, pancreatic polypeptide, and hCG, further studies of even more comprehensive patient materials are necessary to assess their potential prognostic implications.

	Footnotes

 
¹ This work was supported by the Lions Cancer Foundation, the Swedish Medical Council, the Swedish Cancer Society, the Torsten and Ragnar Söderbergs Foundation, the Swedish Society for Medical Research, and the Uppsala County Association against Heart and Lung Diseases.

Received January 8, 2000.

Revised May 11, 2000.

Accepted May 16, 2000.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Arrigoni MG, Woolner LB, Bernatz PE. 1972 Atypical carcinoid tumors of the lung. J Thorac Cardiovasc Surg. 64:413–421.[Medline]
2. Travis WD, Rush W, Flieder DB, et al. 1998 Survival analysis of 200 pulmonary neuroendocrine tumors with clarification of criteria for atypical carcinoid and its separation from typical carcinoid. Am J Surg Pathol. 22:934–944.[CrossRef][Medline]
3. Warren WH, Gould VE, Faber LP, Kittle CF, Memoli VA. 1985 Neuroendocrine neoplasms of the bronchopulmonary tract. A classification of the spectrum of carcinoid to small cell carcinoma and intervening variants. J Thorac Cardiovasc Surg. 89:819–825.[Abstract]
4. Travis WD, Linnoila RI, Tsokos MG, et al. 1991 Neuroendocrine tumors of the lung with proposed criteria for large-cell neuroendocrine carcinoma. An ultrastructural, immunohistochemical, and flow cytometric study of 35 cases. Am J Surg Pathol. 15:529–553.[Medline]
5. Barbareschi M, Girlando S, Mauri FA, et al. 1992 Tumour suppressor gene products, proliferation, and differentiation markers in lung neuroendocrine neoplasms. J Pathol. 166:343–350.[CrossRef][Medline]
6. McCaughan BC, Martini N, Bains MS. 1985 Bronchial carcinoids—review of 124 cases. J Thorac Cardiovasc Surg. 89:8–17.[Abstract]
7. Padberg B-C, Woenckhaus J, Hilger G, et al. 1996 DNA Cytophotometry and prognosis in typical and atypical bronchopulmonary carcinoids. Am J Surg Pathol. 20:815–822.[CrossRef][Medline]
8. Okike N, Bernatz PE, Woolner LB. 1976 Carcinoid tumors of the lung. Ann Thorac Surg. 22:270–277.[Abstract]
9. Böhm J, Koch S, Gais P, Jütting U, Präuer HW, Höfler H. 1996 Prognostic value of Mib-1 in neuroendocrine tumours of the lung. J Pathol. 178:402–409.[CrossRef][Medline]
10. Givehchian M, Woerner SM, Lacroix J, et al. 1996 Expression of CD44 splice variants in normal respiratory epithelium and bronchial carcinomas: no evidence for altered CD44 splicing in metastasis. Oncogene. 12:1137–1144.[Medline]
11. Jiang S-X, Kameya T, Sato Y, Yanase N, Yoshimura H, Kodama T. 1996 Bcl-2 protein expression in lung cancer and close correlation with neuroendocrine differentiation. Am J Pathol. 148:837–846.[Abstract]
12. Brambilla E, Negoescu A, Gazzeri S, Lantuejoul S, Moro D, Brambilla C, Coll J-L. 1996 Apoptosis-related factors p53, Bcl₂, and Bax in neuroendocrine lung tumors. Am J Pathol. 149:1941–1952.[Abstract]
13. Coppola D, Clarke M, Landreneau R, Weyant RJ, Cooper D, Yousem SA. 1996 Bcl-2, p53, CD44, and CD44v6 isoform expression in neuroendocrine tumors of the lung. Mod Pathol. 9:484–490.[Medline]
14. Gouyer V, Gazzeri S, Brambilla E, et al. 1994 Loss of heterozygosity at the Rb locus correlates with loss of Rb protein in primary malignant neuro-endocrine lung carcinomas. Int J Cancer. 58:818–824.[Medline]
15. Gazzeri S, Brambilla E, Negoescu A, et al. 1996 Overexpression of nucleoside diphosphate/kinase A/nm23–H1 protein in human lung tumors: association with tumor progression in squamous carcinoma. Lab Invest. 74:158–167.
16. Gusterson BA, Gullick WJ, Venter DJ, et al. 1988 Immunohistochemical localization of c-erbB-2 in human breast carcinoma. Mol Cell Probes. 2:383–391.
17. Vadasz P, Palffy G, Egervary M, Schaff Z. 1993 Diagnosis and treatment of bronchial carcinoid tumors: clinical and pathological review of 120 operated patients. Eur J Cardiothorac Surg. 7:8–11.[Abstract]
18. Warren WH, Gould VE. 1990 Long-term follow-up of classical bronchial carcinoid tumors. Clinicopathologic observations. Scand J Thorac Cardiovasc Surg. 24:125–130.[Medline]
19. Schreurs AJM, Westermann CJJ, van den Bosch JMM, Vanderschueren RGJRA, de la Rivière AB, Knaepen PJ. 1992 A twenty-five-year follow-up of ninety-three resected typical carcinoid tumors of the lung. J Thorac Cardiovasc Surg. 104:1470–1475.[Abstract]
20. Pelosi G, Bresaola E, Bogina G, et al. 1996 Endocrine tumors of the pancreas: Ki-67 immunoreactivity on paraffin sections is an independent predictor for malignancy: a comparative study with proliferating-cell nuclear antigen and progesterone receptor protein immunostaining, mitotic index, and other clinicopathologic variables. Hum Pathol. 27:1124–1134.[CrossRef][Medline]
21. Willey JC, Lechner JF, Harris CC. 1984 Bombesin and the C-terminal tetradecapeptide of gastrin-releasing peptide are growth factors for normal human bronchial epithelial cells. Exp Cell Res. 153:245–248.[CrossRef][Medline]
22. Cuttitta F, Carney DN, Mulshine J, et al. 1985 Bombesin-like peptides can function as autocrine growth factors in human small-cell lung cancer. Nature. 316:823–826.[CrossRef][Medline]
23. Granberg D, Wilander E, Öberg K, Skogseid B. 1999 Decreased survival in patients with CD44-negative typical bronchial carcinoid tumors. Int J Cancer. 84:484–488.[CrossRef][Medline]
24. Tokunaga Y, Urano T, Furukawa K, Kondo H, Kanematsu T, Shiku H. 1993 Reduced expression of nm23–H1, but not of nm23–H2, is concordant with the frequency of lymph-node metastasis of human breast cancer. Int J Cancer. 55:66–71.[Medline]
25. Yamaguchi A, Urano T, Fushida S, et al. 1993 Inverse association of nm23–H1 expression by colorectal cancer with liver metastasis. Br J Cancer. 68:1020–1024.[Medline]
26. Yamaguchi A, Urano T, Goi T, et al. 1994 Expression of human nm23–H1 and nm23–H2 proteins in hepatocellular carcinoma. Cancer. 73:2280–2284.[CrossRef][Medline]

This article has been cited by other articles:

				K. Oberg, S. Jelic, and On behalf of the ESMO Guidelines Working Group Neuroendocrine bronchial and thymic tumors: ESMO Clinical Recommendation for diagnosis, treatment and follow-up Ann. Onc., May 1, 2009; 20(suppl_4): iv147 - iv149. [Full Text] [PDF]

				J. E S Ardill Circulating markers for endocrine tumours of the gastroenteropancreatic tract Ann Clin Biochem, November 1, 2008; 45(6): 539 - 559. [Abstract] [Full Text] [PDF]

				M. Rugge, M. Fassan, R. Clemente, G. Rizzardi, L. Giacomelli, G. Pennelli, C. Mescoli, D. Segat, and F. Rea Bronchopulmonary Carcinoid: Phenotype and Long-term Outcome in a Single-Institution Series of Italian Patients Clin. Cancer Res., January 1, 2008; 14(1): 149 - 154. [Abstract] [Full Text] [PDF]

				J K Ramage, A H G Davies, J Ardill, N Bax, M Caplin, A Grossman, R Hawkins, A M McNicol, N Reed, R Sutton, et al. Guidelines for the management of gastroenteropancreatic neuroendocrine (including carcinoid) tumours Gut, June 1, 2005; 54(suppl_4): iv1 - iv16. [Full Text] [PDF]

				G. A. Kaltsas, G. M. Besser, and A. B. Grossman The Diagnosis and Medical Management of Advanced Neuroendocrine Tumors Endocr. Rev., June 1, 2004; 25(3): 458 - 511. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Granberg, D. Articles by Skogseid, B. Search for Related Content PubMed PubMed Citation Articles by Granberg, D. Articles by Skogseid, B.