Pancreatic tumors in patients with lung malignancies: a spectrum of clinicopathologic considerations.

Objectives: Lung cancer and pancreatic cancer are the most lethal tobacco-associated malignancies. To elucidate possible clinical interrelationships, the authors reviewed the clinicopathologic characteristics of patients treated for both pulmonary and pancreatic neoplasms.

Methods: Patients presenting with a potentially resectable pancreatic mass and a diagnosis of metachronous malignant neoplasm of the lung were studied by retrospective chart audit and review of histopathologic material.

Results: Seven patients were identified over 6 years, representing five different clinical entities: metachronous presence of lung cancer and pancreatic cancer (n = 3), lung cancer metastatic to the pancreas (n = 1), lung cancer with a benign pancreatic neoplasm (n = 1), periampullary cancer metastatic to the lung (n = 1), and malignant melanoma metastatic to both lung and pancreas (n = 1). A tobacco history was present in all patients but one. Primary treatment modality was complete resection of isolated sites whenever feasible (lung resection, n = 6; pancreatic resection, n = 5). In four cases, a differential diagnosis of adenocarcinomas of both lung and pancreas was obtained after cytokeratin (CK) 7 and CK 20 immunohistochemistry. All patients with evidence of nodal or visceral metastasis from either primary site (n = 4) died within 5 to 9 months after the last operation. Three of four patients who had undergone resection of both pulmonary and pancreatic tumors were alive between 17 and 67 months after the last operation. All three survivors had presented with early disease stages and/or a protracted course (diagnostic interval, 16-66 months).

Conclusions: Our experience with neoplastic conditions that can involve lungs and pancreas metachronously may be useful to the clinician who is confronted with a similar situation. If therapeutic decision-making depends on differential diagnostic analysis, examination of CK 20 expression appears to be helpful. Although biologically favorable circumstances are rarely present, long-term survival seems possible after complete operative treatment in selected patients with early-stage disease.

**********

Lung cancer and pancreatic cancer are tobacco-associated malignancies that are among the four leading causes of cancer mortality in both males and females. (1-3) The relative risk of developing pancreatic cancer in patients with a history of lung cancer is significantly elevated, ranging from 1.3 times the norm in males to 2.5 in females. (4) Although occupational radon exposure or other air pollutants have not been conclusively proven to influence the risk for pancreatic cancer development, (5,6) other environmental or genetic factors may be shared among these neoplasms. (7) The clinician may be faced with a situation in which pulmonary and pancreatic masses coincide. We have recently encountered several patients with pancreatic masses who had a history of a malignant lung neoplasm, and in whom consideration of a relationship between lung and pancreatic cancers became relevant. We were therefore interested in exploring the clinicopathologic characteristics of individuals presenting with both pulmonary and pancreatic tumors, including the use of immunohistochemical techniques in the evaluation of these neoplasms.

Materials and Methods

The subjects of this analysis were patients with neoplasms of both lung and pancreas presenting for a pancreatectomy evaluation. Of 62 patients seen over the course of 6 years by a single surgeon, and who presented with an isolated, potentially resectable pancreatic mass, 7 were identified as carrying a diagnosis of a lung malignancy as well. Clinical presentation information, treatment aspects, findings of pathologic evaluation, and outcomes were all charted retrospectively. All available tissue sections were reexamined and, where indicated, subjected to immunohistochemical staining with antibodies against cytokeratin (CK) 7 and CK 20. Relevant information was tabulated to descriptively present various underlying clinical entities.

Results

Seven patients had been diagnosed with a pulmonary neoplasm and a pancreatic mass. There were four females and three males, with a median age of 67 years (range, 47-86 years). Relevant clinicopathologic information is listed in Table 1. In five cases, pancreatic masses were identified through routine imaging or laboratory studies as part of a lung cancer follow-up. Only four patients were mildly symptomatic, complaining of epigastric discomfort (n = 3) or back pain (n = 1). A tobacco history was present in all patients but one.

All patients appeared to have a potentially resectable pancreatic mass based on examination using appropriate computed tomography techniques. Three patients had undergone positron emission tomography scanning that confirmed uptake isolated to the pancreatic area only. Five individuals (patients 1-5) had a proven history of lung cancer that had been treated by resection with curative intent. The median interval between lung resection and pancreas mass diagnosis was 24 months (range, 2-76 months). All these patients underwent operative treatment intended to result in a pancreatectomy, but only three were found to be resectable. Patient 1 was diagnosed with liver metastases on laparoscopy and did not seek any further treatment. Patient 3 was found to have an unresectable recurrence of a previously resected periampullary cancer. All other patients underwent complete (R0) resections. Patient 6 was diagnosed with an isolated pancreatic mass and a long disease-free interval after left lower lobectomy for melanoma metastatic to the lung; his pancreatic mass was confirmed to be of similar histopathologic origin. Patient 7 had a history of early-stage endometrial and colon cancers, both completely resected, and had undergone pancreatoduodenectomy for a stage 3 periampullary adenocarcinoma involving the pancreatic head; after more than 2 years, she presented with an advanced lung mass with mediastinal involvement, which was palliated with radiotherapy.

Differential diagnostic evaluations included comparative histopathologic examination and immunohistochemical analysis. Simple comparison of lung and pancreas tissues confirmed the presence of two separate neoplastic processes in patients 1 and 2, and a similar metastatic process in patient 6. Tissue from the remaining four cases was examined with CK 7 and CK 20 stains. In patient 3, CK 7-positive and CK 20-negative adenocarcinoma cells in the lung confirmed the presence of a second primary tumor; the subsequent pancreatic mass was consistent with a recurrence from the original periampullary cancer. In patient 4, both lung and pancreatic tissues were CK 7-positive and CK 20-negative. This pattern and the clinical impression were consistent with lung cancer metastatic to the pancreas, despite thyroid transcription factor-1 negativity. Patient 5 presented with CK 7-positive and CK 20-positive adenocarcinoma in the lung; although a pancreatic primary was possible, the resected pancreatic mass was benign. Finally, patient 7 is included to underscore the benefit of cytokeratin profile comparison. In this case, the clinical presentation of a large lung mass with bulky mediastinal lymphadenopathy had originally been diagnosed and treated as a new lung primary tumor. However, the cytokeratin profile (CK 7-negative/CK 20-positive) (Fig. 1) obtained during the histopathologic reexamination for this review, and the history of node-positive pancreatic head adenocarcinoma, make a diagnosis of a pulmonary relapse more plausible in retrospect.

[ILLUSTRATION OMITTED]

Four patients subsequently died as a result of recurrent cancer, and three patients were alive and free of disease between 17 and 67 months after their last operation. All three had presented with early disease stages and/or a protracted course (diagnostic interval, 16-66 months), and were able to undergo complete resections of both pulmonary and pancreatic sites. All patients who died had adenocarcinomas of either lung or pancreatic origin, with either lymphatic or visceral metastases; their remaining lifetime ranged from 5 to 9 months after diagnosis of the neoplasm that presented last.

Discussion

Although the cases presented here are representative of an entire spectrum of neoplastic conditions involving both lungs and pancreas, we are reminded of a possible clinical association between lung cancer and pancreatic cancer de novo, as demonstrated by three of the five patients who had a primary lung cancer. Despite the fairly high prevalence of both lung and pancreatic cancers, clinical information on patients with neoplastic diseases of lung and pancreas combined is limited. The reasons are multiple: few lung cancer patients undergo a potentially curative lung resection and develop a de novo pancreatic cancer at a stage that is still amenable to pancreatectomy; few lung cancer patients undergo a potentially curative lung resection and relapse at an isolated intrapancreatic site; (8,9) and even fewer patients with pancreatic cancer develop an isolated lung metastasis during the subsequent course, (10) or survive long enough to develop a new lung primary cancer. In this context, our patients represent a highly selected group by definition.

Aside from the rarity of these events, clinical parameters are often confusing. As demonstrated by patient 7, the distinction between lung cancer and pancreatic cancer is not infrequently wrong. (11-13) Diagnostic findings characteristic for pancreatic cancer, such as painless jaundice, have also been described in lung cancer patients with or without periampullary metastases. (14,15) Involvement of peripancreatic lymph nodes by a malignancy may be from metastatic lung carcinoma. (16) In fact, even the possibility of pancreatic metastasis resulting from a pulmonary adenocarcinoma metastasis has been raised. (17)

Immunohistochemical techniques allow for better distinction and delineation of the underlying neoplastic mechanism. Although carcinoembryonic antigen staining remains common in both lung and pancreatic adenocarcinomas, (18,19) CA 19-9 immunostaining has been reported in 90% of pancreatic and 25% of lung cancers. (20) Among cytokeratins, CK 20 appears valuable to provide some delineation; the majority of pancreatic cancers stain positive, (21,22) and most lung adenocarcinomas stain negative. (23) CK 7 is frequently reactive in both conditions. (22) Thyroid transcription factor-1 reactivity is present in up to 76% of lung adenocarcinomas and appears to be the best marker for delineating pulmonary adenocarcinoma, despite our experience with patient 4. (23-25) Another marker for metastatic lung cancer, PE-10, has not been widely used. (26)

What is the therapeutic impact of this distinction? In the properly selected patient with a lung cancer history and an isolated pancreatic mass, it is probably of lesser importance to define the operative indication. Despite generally dismal survival statistics for both diseases, isolated lung cancer metastasis, isolated metastases to the pancreas, and isolated pancreatic cancer primaries have been treated in properly selected patients by complete resection with some achievable long-term survival. (26-30) Although isolated pancreatic masses are frequently not subjected to preoperative biopsy, one may have to reconsider this in cases of a node-positive lung cancer history. After pancreatectomy, the direction of any postoperative treatment effort would also depend on the underlying oncologic mechanism. Finally, in the nonoperative patient with nonisolated, diffuse disease, the differential diagnosis may again have some value for the selection of a systemic treatment regimen, although outcomes are expected to remain predominantly grim.

Given that there is an increased risk of developing pancreatic cancer among long-term survivors of lung cancer, (31) should these patients be screened to diagnose de novo pancreatic cancer earlier? Routine surveillance and imaging may have limited value in the postoperative follow-up of lung cancer patients, and follow-up strategies remain heterogeneous. (32,33) The greatest potential benefit in these "screening" efforts may be found among cases in which metachronous lung cancers or other malignancies of early, treatable stages are diagnosed in those patients with greater survival probability. There is good evidence that death from second malignant neoplasms caused by tobacco products is a major cause of both early and late mortality after surgical resection of non-small-cell lung cancer (NSCLC). (34,35) Over a 5-year period, Lamont et al (36) followed 124 patients after resection for NSCLC with annual computerized tomograms of the chest and detected 19 (15.3%) second primary NSCLCs. The median diameter of these neoplasms was 14 mm (range, 8-28 mm); 84% were stage IA cancers. Although screening for pancreatic cancer is a complex proposal (37) and certainly cannot be advocated on the basis of our very limited experience, extension of the computed tomography (CT) scan to include the pancreas in surveillance CT scans after lung resections merits further discussion. All three surviving patients in this series had their pancreatic tumor identified in an asymptomatic stage through surveillance imaging, two of them for primary lung cancer. An extended CT scan would provide little additional radiation exposure, but may allow early detection of some pancreatic or renal neoplasms. Already, chest CT scans for this purpose include the liver and adrenal glands, and even nonenhanced scans have proven beneficial. (38) Although contrast-enhanced, fine-cut CT scans are most sensitive in the diagnosis of pancreatic cancer, many lesions would be visible with conventional techniques, provided only the pancreas is included in the scan. This follow-up imaging approach should perhaps be debated for intermediate-term survivors of node-negative lung cancer. In contrast, new onset of clinical symptoms such as jaundice, back pain, diarrhea, diabetes, or weight loss should certainly prompt the association with possible pancreatic cancer involvement, and mandate specific diagnostic tests in every affected patient.

Conclusion

The review of our experience with neoplastic conditions that can involve the lung and pancreas metachronously may be useful to the clinician who is confronted with a similar situation. This clinical scenario represents primarily de novo pancreatic cancer, or the less likely event of pancreatic metastasis, among other conditions. If therapeutic decision-making depends on differential diagnostic analysis, examination of CK 20 expression appears helpful. Although biologically favorable circumstances are rarely present, long-term survival seems possible after aggressive operative treatment in carefully selected patients with early-stage disease.

Do, or do not. There is no 'try'.
--Yoda

Table. Clinicopathologic patient information (a)

                                      Patient No.
                             1                          2

Clinical entity    Two separate cancers        Two separate cancers
Final              Adenosquamous lung cancer;  SCCa of lung, moderate
  histopathologic    well-differentiated         differentiation;
  diagnosis          adenocarcinoma of           well-differentiated
                     pancreas                    adenocarcinoma of
                                                 pancreas
Age (yr)           75                          47
Gender             Female                      Female
Tobacco history    80                          30
  (pack-yr)
Interval lung to   76                          16
  pancreas (mo)
Stage, lung        TIN0M0                      T2N0M0
Stage, pancreas    T2NxM1 (4)                  yT1N0M0
Treatment, lung    Lobectomy, RUL              Lobectomy, RUL
Treatment,         None                        CRT, followed
  pancreas                                       by PD
Current status     DOD                         NED
Follow-up time,    82                          83
  lung (mo)
Follow-up time,     5                          67
  pancreas (mo)

                                      Patient No.
                             3                           4

Clinical entity    Two separate cancers         Lung cancer metastatic
                                                  to pancreas
Final              Poorly differentiated        Poorly differentiated
  histopathologic    periampullary cancer,        mucin-producing
  diagnosis          T3N0M0, treated with PD      adenocarcinoma in both
                     16 mo prior; moderately      sites; CK 7-positive,
                     differentiated               CK 20-negative, but
                     adenocarcinoma of lung,      TTF-1--negative
                     CK 7-positive, CK 20-
                     negative; pancreatic
                     adenocarcinoma recurrence
Age (yr)           59                           58
Gender             Male                         Male
Tobacco history    0                            35
  (pack-yr)
Interval lung to   2                            26
  pancreas (mo)
Stage, lung        T1N0M0                       T2N2M0 (3A)
Stage, pancreas    Recurrence                   Metastasis, 2 LN-
                                                  positive
Treatment, lung    LUL wedge resection          Lobectomy, RUL; CRT
Treatment,         Biopsy, CTX                  DP
  pancreas
Current status     DOD                          DOD
Follow-up time,    7                            35
  lung (mo)
Follow-up time,    5 (21 from original          9
  pancreas (mo)      diagnosis)

                                       Patient No.
                              5                           6

Clinical entity    Lung cancer, benign          Metastases to lung and
                     pancreatic tumor             to pancreas
Final              Moderately differentiated    Well-differentiated
  histopathologic    lung adenocarcinoma, CK      spindle cell tumor,
  diagnosis          7- and CK 20-positive;       consistent with
                     pancreatic IPMT with         malignant melanoma;
                     malignant potential          primary site unknown
Age (yr)           73                           86
Gender             Female                       Male
Tobacco history    25                           15
  (pack-yr)
Interval lung to   24                           66
  pancreas (mo)
Stage, lung        T1N0M0                       Metastatic
Stage, pancreas    Benign                       Metastatic
Treatment, lung    Lobectomy, RUL               Lobectomy, LLL
Treatment,         PD                           DP
  pancreas
Current status     NED                          NED
Follow-up time,    55                           83
  lung (mo)
Follow-up time,    31                           17
  pancreas (mo)

                               Patient No.
                                   7

Clinical entity      Pancreatic cancer, metastatic to lung
Final                Poorly differentiated periampullary
  histopathologic      adenocarcinoma; poorly differentiated
  diagnosis            lung adenocarcinoma, CK 7-negative,
                       CK 20-positive
Age (yr)             67
Gender               Female
Tobacco history      50
  (pack-yr)
Interval lung to     -29
  pancreas (mo)
Stage, lung          Metastatic
Stage, pancreas      T3N1M0 (3)
Treatment, lung      RT
Treatment,           PD
  pancreas
Current status       DOD
Follow-up time,      5
  lung (mo)
Follow-up time,      34
  pancreas (mo)

(a) SCCa, squamous cell cancer; PD, pancreatoduodenectomy; CK 7,
cytokeratin 7; CK 20, cytokeratin 20; TTF-1, thyroid transcription
factor-1; LN, lymph node; RUL, right upper lobe; LUL, left upper lobe;
LLL, left lower lobe; CRT, chemoradiation therapy; RT, radiation
therapy; CTX, chemotherapy; DP, distal pancreatectomy; DOD, died of
disease: NED, no evidence of disease; IPMT, intraductal pancreatic
mucinous tumor.

Accepted December 8, 2003.

References

1. Engeland A, Andersen A, Haldorsen T, et al. Smoking habits and risk of cancers other than lung cancer: 28 years' follow-up of 26,000 Norwegian men and women. Cancer Causes Control 1996;7:497-506.

2. Gold EB. Epidemiology of and risk factors for pancreatic cancer. Surg Clin North Am 1995;75:819-843.

3. Jemal A, Murray T, Samuels A, et al. Cancer statistics, 2003. CA Cancer J Clin 2003;53:5-26.

4. Neugut AI, Ahsan H, Robinson E. Pancreas cancer as a second primary malignancy: a population-based study. Cancer 1995;76:589-592.

5. Epstein LM, Katz L, Tamir A, et al. Incidence of lung cancer in five towns in Israel, 1960-74. Isr J Med Sci 1984;20:27-32.

6. Darby SC, Whitley E, Howe GR, et al. Radon and cancers other than lung cancer in underground miners: a collaborative analysis of 11 studies. J Natl Cancer Inst 1995;87:378-384.

7. Price TF, Payne RL, Oberleitner MG. Familial pancreatic cancer in south Louisiana. Cancer Nurs 1996;19:275-282.

8. Maeno T, Satoh H, Ishikawa H, et al. Patterns of pancreatic metastasis from lung cancer. Anticancer Res 1998;18:2881-2884.

9. Barz H, Barz D, Klemm P. Distribution of lung cancer metastases: I. combination and frequency of organ metastases [in German]. Arch Geschwulstforsch 1982;52:551-560.

10. Matsuno S, Kato S, Kobari M, et al. Clinicopathological study on hematogenous metastasis of pancreatic cancer. Jpn J Surg 1986;16:406-411.

11. Mattsson B, Rutqvist LE. Some aspects on validity of breast cancer, pancreatic cancer and lung cancer registration in Swedish official statistics. Radiother Oncol 1985;4:63-70.

12. Doring C, Lindlar F. Clinically a primary lung carcinoma: during autopsy metastasis of a pancreatic cancer [in German]. Med Welt 1969;8:407-411.

13. Mishriki YY. Pancreatic carcinoma presenting as primary lung cancer. Postgrad Med 1990;87:107-108.

14. Johnson DH, Hainsworth JD, Greco FA. Extrahepatic biliary obstruction caused by small-cell lung cancer. Ann Intern Med 1985;102:487-490.

15. Yapp RG, Siegel JH. Unexplained biliary tract dilatation in lung cancer patients. Endoscopy 1992;24:593-595.

16. Onuigbo WI. Lung cancer metastasis to the pancreas and its surrounding lymph nodes. Br J Dis Chest 1966;60:152-155.

17. Inagaki H. Nakao A, Ando N, et al. A case of solitary metastatic pancreatic cancer from rectal carcinoma: a case report. Hepatogastroenterology 1998;45:2413-2417.

18. Ghoneim AH, Brisson ML, Fuks A, et al. Monoclonal anti-CEA antibodies in the discrimination between primary pulmonary adenocarcinoma and colon carcinoma metastatic to the lung. Mod Pathol 1990;3:613-618.

19. Sheahan K, O'Brien MJ, Burke B, et al. Differential reactivities of carcinoembryonic antigen (CEA) and CEA-related monoclonal and polyclonal antibodies in common epithelial malignancies. Am J Clin Pathol 1990;94:157-164.

20. Loy TS, Sharp SC, Andershock CJ, et al. Distribution of CA 19-9 in adenocarcinomas and transitional cell carcinomas: an immunohistochemical study of 527 cases. Am J Clin Pathol 1993;99:726-728.

21. Tot T. Adenocarcinomas metastatic to the liver: the value of cytokeratins 20 and 7 in the search for unknown primary tumors. Cancer 1999;85:171-177.

22. Chu PG, Weiss LM. Keratin expression in human tissues and neoplasms. Histopathology 2002;40:403-439.

23. Harlamert HA, Mira J, Bejarano PA, et al. Thyroid transcription factor-1 and cytokeratins 7 and 20 in pulmonary and breast carcinoma. Acta Cytol 1998;42:1382-1388.

24. Khoor A, Whitsett JA, Stahlman MT, et al. Utility of surfactant protein B precursor and thyroid transcription factor 1 in differentiating adenocarcinoma of the lung from malignant mesothelioma. Hum Pathol 1999;30:695-700.

25. Lau SK, Luthringer DJ, Eisen RN. Thyroid transcription factor-1: a review. Appl Immunohistochem Mol Morphol 2002;10:97-102.

26. Seki M, Tsuchiya E, Hori M, et al. Pancreatic metastasis from a lung cancer: preoperative diagnosis and management. Int J Pancreatol 1998;24:55-59.

27. Luketich JD, Martini N, Ginsberg RJ, et al. Successful treatment of solitary extracranial metastases from non-small cell lung cancer. Ann Thorac Surg 1995;60:1609-1611.

28. Kim SH, Brennan MF, Russo P, et al. The role of surgery in the treatment of clinically isolated adrenal metastasis. Cancer 1998;82:389-394.

29. Harrison LE, Merchant N, Cohen AM, et al. Pancreaticoduodenectomy for nonperiampullary primary tumors. Am J Surg 1997;174:393-395.

30. Janes R Jr, Niederhuber JE, Chmiel JS, et al. National patterns of care for pancreatic cancer: results of a survey by the Commission on Cancer. Ann Surg 1996;223:261-272.

31. Velie EM, Schatzkin, A. Common environmental risk factors, in Neugut AI, Meadows AT, Robinson E (eds): Multiple Primary Cancers. Philadelphia, Lippincott Williams & Wilkins, 1999, pp 213-223.

32. Naunheim KS, Virgo KS, Coplin MA, et al. Clinical surveillance testing after lung cancer operations. Ann Thorac Surg 1995;60:1612-1616.

33. Younes RN, Gross JL, Deheinzelin D. Follow-up in lung cancer: how often and for what purpose? Chest 1999;115:1494-1499.

34. Martini N, Bains MS, Burt ME, et al. Incidence of local recurrence and second primary tumors in resected stage I lung cancer. J Thorac Cardiovasc Surg 1995;109:120-129.

35. Martini N, Rusch VW, Bains MS, et al. Factors influencing ten-year survival in resected stages I to IIIa non-small cell lung cancer. J Thorac Cardiovasc Surg 1999;117:32-38.

36. Lamont JP, Kakuda JT, Smith D, et al. Systematic postoperative radiologic follow-up in patients with non-small cell lung cancer for detecting second primary lung cancer in stage IA. Arch Surg 2002;137:935-940.

37. Okada S, Yoshimori M, Kakizoe T. Pancreatic cancer: medical aspects. Pancreas 1998;16:349-354.

38. Macari M, Rofsky NM, Naidich DP, et al. Non-small cell lung carcinoma: usefulness of unenhanced helical CT of the adrenal glands in an unmonitored environment. Radiology 1998;209:807-812.

RELATED ARTICLE: Key Points

* A spectrum of different disease mechanisms has been encountered in patients with metachronous lung and pancreas neoplasms.

* Analysis of cytokeratin 20 expression appears helpful in the differential diagnosis of lung and pancreatic cancer.

* After therapy of early-stage lung cancer, treatable pancreatic neoplasms may be identified.

* There is a risk of additional tobacco-associated malignancies in survivors of lung (or another tobacco-related) cancer.

Roderich E. Schwarz, MD, Peiguo G. Chu, MD, and Frederic W. Grannis, Jr, MD

From the Departments of General Oncologic Surgery and Anatomic Pathology, City of Hope National Medical Center, Duarte, CA; and the Division of Surgical Oncology, Cancer Institute of New Jersey, New Brunswick, NJ.

Reprint requests to Roderich E, Schwarz, MD, PhD, Department of Surgery, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, The Cancer Institute of New Jersey, 195 Little Albany Street, New Brunswick, NJ 08901. Email: [e-mail]