Secondary Logo

Journal Logo

Metastasis to the Lung From Carcinomas of Nonpulmonary Sites After Prolonged Disease-free Intervals

Mukhopadhyay, Sanjay MD*; Patil, Pradnya D. MD; Roma, Andres A. MD*

Journal of Bronchology & Interventional Pulmonology: July 2017 - Volume 24 - Issue 3 - p 216–224
doi: 10.1097/LBR.0000000000000382
Original Investigations
Free

Background: Most carcinomas of nonpulmonary sites that metastasize to the lung do so within 5 years of diagnosis. Although examples of late metastasis to the lung after prolonged disease-free intervals (>5 y) have been reported sporadically, this phenomenon has not been systematically analyzed. The aim of this study was to describe the clinical and pathologic features of metastases to the lung from carcinomas of nonpulmonary origin after prolonged disease-free intervals.

Methods: We searched our pathology archives to identify lung biopsies/resections containing metastases from carcinomas of nonpulmonary origin. Medical records were reviewed to determine the interval between resection of the nonpulmonary primary and subsequent detection of lung metastasis. Cases were included if the disease-free interval between initial diagnosis and lung metastasis exceeded 5 years and the site of origin could be verified by pathologic examination.

Results: Of 195 consecutive lung metastases from carcinomas of nonpulmonary sites, the recurrence-free interval before lung metastasis was >5 years in 20 (10.3%). Primary sites (number of cases, recurrence-free interval) included kidney (5, 6 to 33 y), endometrium (5, 8 to 10 y), colon (3, 6 to 13 y), breast (2, 8 y, 12 y), esophagus (1, 8 y), thyroid (1, 10 y), epiglottis (1, 12 y), prostate (1, 12 y), and ovary (1, 15 y). At diagnosis of lung metastasis, lung nodules/masses were multiple in 12 and solitary in 8.

Conclusions: Carcinomas of nonpulmonary sites can metastasize to the lung after prolonged disease-free intervals and present as a solitary lung mass. The most common culprits are carcinomas of the kidney and endometrium.

Departments of *Pathology

Hospital Medicine, Cleveland Clinic, Cleveland, OH

S.M. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis, including and especially any adverse effects. P.D.P. and A.A.R. contributed substantially to the study design, data analysis and interpretation, and the writing of the manuscript.

Disclosure: There is no conflict of interest or other disclosures.

Reprints: Sanjay Mukhopadhyay, MD, Department of Pathology, Cleveland Clinic, 9500 Euclid Ave., Cleveland, OH 44195 (e-mail: mukhops@ccf.org).

Received September 18, 2016

Accepted January 16, 2017

The finding of a new lung mass in a patient with a remote history of carcinoma of a nonpulmonary site/organ can be a diagnostic challenge. It is well known that most carcinomas that recur and/or metastasize do so within 5 years of initial diagnosis.1 Tumors that do not recur within this time-frame are generally considered cured, and surveillance is often discontinued or decreased. The discovery of a lung mass in such patients raises the differential diagnosis between a late metastasis from the original tumor and a new primary lung carcinoma. Factors that commonly influence clinicians’ judgment in this situation include the length of the recurrence-free interval, the size of the lung mass, the number of masses, and the history of cigarette smoking. For example, a large solitary lung mass in a current smoker with a remote history of carcinoma elsewhere is likely to be considered a new lung primary.

Over the past few years, we have encountered cases in which pathologically confirmed metastases to the lung from carcinomas of nonpulmonary sites occurred after surprisingly long disease-free intervals in patients whose primary tumor had been resected several years previously. In some of these cases, oncologists questioned the validity of the pathologic diagnosis based on the length of the disease-free interval and radiologic features of the lung mass. Upon reviewing the literature, we found that the phenomenon of late metastases to the lung after prolonged disease-free intervals had not been systematically examined. The primary aim of this study was therefore to determine whether metastases to the lung from nonpulmonary carcinomas can occur after prolonged disease-free intervals. We also sought to ascertain whether such metastases can show clinical and/or radiologic features typically associated with primary lung carcinoma.

Back to Top | Article Outline

METHODS

Case Selection

The study was approved by the Cleveland Clinic Institutional Review Board (IRB14-501). We searched our surgical pathology archives to identify all lung samples (including endobronchial biopsies, transbronchial biopsies, core needle biopsies, wedge resections, lobectomies, and pneumonectomies) containing metastatic carcinomas from nonpulmonary sites over a period of 7 years (January 2008 to December 2014). Cytology samples were excluded because at our institution many of these are accompanied by a biopsy sample. As pathologic examination and immunohistochemistry are most helpful in distinguishing primary lung carcinoma from metastases to the lung from carcinomas of other sites, only carcinomas were included in the study. The following were excluded: (1) malignancies in which pathologic features do not definitively establish site of origin (melanoma, sarcoma, and lymphoma), (2) tumors in which local recurrence or distant metastases occurred within 5 years after diagnosis, (3) tumors in which site of origin was unclear even after pathologic examination (mainly adenocarcinomas with equivocal immunohistochemical profiles), and (4) tumors in which differentiation between primary lung carcinoma and metastatic carcinoma from a nonpulmonary site was not possible on pathologic grounds (mainly squamous cell carcinomas).

In each case, clinical and pathologic records were reviewed retrospectively to identify the date of resection of the primary tumor and the date that a lung nodule (subsequently proven to be metastatic carcinoma) was first detected. The interval between these dates was considered the recurrence-free (disease-free) interval.

Back to Top | Article Outline

Review of Clinical and Radiologic Findings

In cases with late lung metastases, the following parameters were documented: (1) whether surveillance chest imaging had been performed after resection of the primary nonpulmonary tumor, (2) clinical presentation leading to detection of lung metastasis, (3) number of lung nodules at the time of lung metastasis, (4) treatment after diagnosis of lung metastasis, and (5) clinical outcome.

Back to Top | Article Outline

Review of Pathologic Diagnosis

In each case, the pathologic diagnosis was reverified by 2 surgical pathologists, 1 with expertise in pulmonary pathology (S.M.) and 1 with expertise in gynecologic and genitourinary pathology (A.R.). Both pathologists reviewed H&E-stained slides from the lung metastasis, compared them with slides from the resected primary nonpulmonary carcinoma (if available) and reevaluated pertinent immunohistochemical findings. In cases with unusual histologic features and in those where confirmatory immunohistochemical stains were not available or not performed at the time of the original diagnosis, additional immunohistochemical stains were performed specifically for the purpose of this study.

Back to Top | Article Outline

RESULTS

One hundred ninety-five consecutive cases of lung metastasis from nonpulmonary carcinomas were identified in the Cleveland Clinic pathology database over a 7-year period. The recurrence-free interval between resection of the primary tumor and discovery of lung metastases was >5 years in 20/195 cases (10.3%, Table 1). The specimen types in which these were diagnosed included bronchoscopic biopsies (6), core needle biopsies (2), wedge resections (7), segmentectomy (1), lobectomies (3), and a pneumonectomy (1). The site of origin, duration of the recurrence-free interval before lung metastasis, and data regarding treatment and follow-up after lung metastasis are shown in Table 2. Of the 20 patients with late lung metastases, only 8 received adjuvant therapy. Of these, 5 received adjuvant radiation, 2 received adjuvant chemotherapy, and 1 received adjuvant chemoradiation. The longest duration of adjuvant therapy was 7 months. Hence, adjuvant therapy did not significantly alter our calculation of the recurrence-free interval.

TABLE 1

TABLE 1

TABLE 2

TABLE 2

The most common carcinomas metastasizing to the lung after a recurrence-free interval >5 years were carcinomas of the kidney and endometrium (5 cases each), followed by adenocarcinoma of the colon (3 cases) and breast (2 cases). There was 1 case each of carcinoma of the esophagus, thyroid, larynx, prostate, and ovary. In 7 of 20 cases, the disease-free interval before lung metastasis was >10 years (Table 1).

In 14 patients, late metastases were isolated to the lung, whereas in 6 concurrent metastases were also found in other sites. In the latter group, bone metastases were most common. They were found at the same time as lung metastases in 3 patients (cases 14, 17, and 19) and subsequent to lung metastases in 1 (case 7). Other metastatic sites were lymph nodes (cases 1 and 20), base of tongue (case 1) and pericardium (case 1).

In most patients (13/20), surveillance after resection of the primary tumor was performed purely by clinical follow-up, without radiologic surveillance in the years before detection of lung metastases. Of the 7 patients who underwent chest imaging for surveillance, 6 had chest radiographs and 1 had a chest computed tomography, all of which were negative for lung nodules during the surveillance period. In 5 patients (cases 4, 10, 11, 16, and 18), multiple chest imaging studies performed over a span of several years during the recurrence-free interval had been negative.

Back to Top | Article Outline

Clinical Presentation

At the time of diagnosis of late lung metastases, 12 patients were symptomatic and 8 were asymptomatic. Presenting symptoms in the former group were cough (5), dyspnea (4, including superior vena cava syndrome in 1), hemoptysis (2), and chest pain (1). In the asymptomatic individuals, lung nodules were discovered incidentally on imaging performed for unrelated reasons (5 cases) or during work-up for elevated tumor markers (3 cases—carcinoembryonic antigen in 2, thyroglobulin in 1). Fourteen patients were never-smokers; 6 were current or ex-smokers (cases 5, 6, 11, 12, 16, and 19; pack-years 2.5 to 20, mean 7.9, median 5).

Back to Top | Article Outline

Radiologic Findings

At the time of lung metastasis, 12 patients had multiple lung nodules/masses and 8 had a solitary nodule/mass. In the former group, nodule size ranged from a few millimeters (cases 2, 4, 7, and 17) to 13.9 cm (case 8; this patient also had a 2.1 cm nodule in a different lobe on the same side). In those who presented with a solitary lung nodule/mass, the lesions ranged from 1.3 cm (case 16) to 7 cm (case 9). Three patients had endobronchial extension of tumor.

With respect to radiologic distribution of nodules, 10 cases had a hematogenous pattern favoring metastases, and 3 had a bronchocentric pattern, favoring a lung primary. In the remaining 7 cases, the radiology report did not state where a primary or metastasis was favored.

Back to Top | Article Outline

Follow-up

Of the 20 patients with late lung metastasis, 14 are alive after follow-up intervals subsequent to lung metastasis ranging from 1 to 13 years (Table 2). Of the remaining 6, 2 were lost to follow-up and 4 died after follow-up intervals ranging from 3 months to 3 years after lung metastasis. Of the latter, 2 patients died of progressive disease. The cause of death in the other 2 was likely unrelated to cancer. Six patients were followed without therapy after pulmonary metastasectomy; only 2 of these developed progressive disease. Overall, of the 14 individuals who are alive after late lung metastasis, 4 are in remission with no evidence of disease, 6 are alive with stable disease and 4 have progressive disease. In this group (n=14), the overall duration of survival since initial diagnosis of the nonpulmonary primary carcinoma ranges from 7 to 35 years (median, 14.75 y; mean, 16 y).

Back to Top | Article Outline

Pathologic Findings

Details of the pathologic findings and immunohistochemical profiles are provided in Table 2. The case with the longest recurrence-free interval before lung metastasis in this series is illustrated in Fig. 1. The patient was a 75-year-old man who presented with cough and was found to have a 3.1-cm solitary lung mass (Fig. 1). He had undergone radical nephrectomy for renal cell carcinoma 33 years ago and had been disease-free. The lung mass was resected and found to be a clear cell carcinoma with an immunoprofile consistent with a renal primary.

FIGURE 1

FIGURE 1

Back to Top | Article Outline

DISCUSSION

This retrospective review shows that metastases to the lung from carcinomas of nonpulmonary sites can occur even after prolonged disease-free intervals following resection of the primary tumor. In contrast to prior reports, this study focuses exclusively on the phenomenon of late lung metastasis and demonstrates that this uncommon phenomenon can occur in carcinomas of a variety of primary sites. In some cases in our series, the length of the disease-free interval caused clinicians to fail to consider the possibility of a metastasis and 1 patient was misdiagnosed with primary lung cancer. In others, a pathologically proven diagnosis of metastatic carcinoma was questioned based at least in part on the length of the disease-free interval. Although it is true that most patients with carcinoma who have been recurrence-free for >5 years are unlikely to develop lung metastasis, our findings demonstrate that pathologically proven exceptions to this dictum do exist. In addition, our clinical and radiologic data show that late metastases can occur in patients with clinical risk factors and radiologic features strongly suggestive of primary lung cancer.

There are only sporadic reports of late lung metastasis from carcinomas of nonpulmonary sites in the literature. A study of 15 pulmonary resections for metastases from gynecologic malignancies included 2 cases of endometrial carcinoma that metastasized to the lung as a solitary lung nodule after recurrence-free intervals of 81 months and 110 months.2 In another series of 100 patients with lung metastases from uterine malignancies, 6 were found to have metastases >10 years after diagnosis of their uterine tumor, including 2 in whom lung metastases were discovered at autopsy 15 and 26 years following diagnosis of endometrial adenocarcinoma.3 In reports where pathologic features have been adequately documented, the longest recurrence-free interval between hysterectomy for endometrial carcinoma and subsequent lung metastasis is 17 years.4,5 The phenomenon of delayed lung metastasis from renal cell carcinoma is more widely recognized, although most relapses occur within the first 3 years after diagnosis.6 Miyao et al7 followed 470 patients with renal cell carcinoma who were recurrence-free in the first 10 years after diagnosis. Of these, 30 (6.4%) were found to have a late recurrence after a 10-year disease-free interval, and the lungs were the most common site of metastasis. The longest recurrence-free interval between nephrectomy and lung metastasis was 20 years. Recently, Watanabe et al8 reported a patient who presented with a solitary lung mass 31 years after nephrectomy. To our knowledge, case 5 of our series documents the longest recurrence-free interval ever reported between resection of renal cell carcinoma and a pathologically confirmed metastasis to the lung (33 y). Of the tumors encountered infrequently in this study, adenoid cystic carcinoma is known to be associated with late recurrences and delayed lung metastasis. Late recurrences after a 5-year disease-free interval occur in 26% of cases and delayed lung metastases have been reported after disease-free intervals of as long as 136 months.9,10 In adenoid cystic carcinoma of the upper airway, the longest reported interval before development of lung metastasis is 200 months.11 Late lung metastases from colorectal carcinoma after prolonged disease-free intervals are quite infrequent (0.9%), and there are only a few reports of this phenomenon in carcinomas of the esophagus and ovary.12–14

An interesting observation is that the majority of patients in our series had a favorable outcome even after developing lung metastasis. Overall, 14 of 20 patients are still alive 1 to 13 years after lung metastasis, attesting to the relatively indolent behavior of carcinomas that are detected in the lungs after prolonged disease-free intervals. The association between long recurrence-free intervals, slow tumor doubling time, and favorable outcome has been previously noted,2 but the biological basis of this phenomenon remains unclear. Our morphologic observations failed to provide a compelling explanation other than the low grade of some of the tumors. It is likely that these neoplasms have an unusually low growth rate that might allow microscopic foci of tumor in the lungs at the time of initial diagnosis to remain occult for several years until they grow to a size that can be detected radiologically. Alternatively metastatic tumor foci may remain dormant for many years and subsequently “reawaken,” as has been proposed recently.15

In summary, late metastases to the lung from carcinomas of nonpulmonary sites can occur after prolonged disease-free intervals and may occasionally present as a solitary lung mass mimicking primary lung cancer. Renal cell carcinomas and carcinomas of the endometrium are the most common culprits.

Back to Top | Article Outline

REFERENCES

1. Gloeckler Ries LA, Reichman ME, Lewis DR, et al. Cancer survival and incidence from the Surveillance, Epidemiology, and End Results (SEER) Program. Oncologist. 2003;8:541–552.
2. Fuller AF Jr, Scannell JG, Wilkins EW Jr.. Pulmonary resection for metastases from gynecologic cancers: Massachusetts General Hospital experience, 1943-1982. Gynecol Oncol. 1985;22:174–180.
3. Bouros D, Papadakis K, Siafakas N, et al. Natural history of patients with pulmonary metastases from uterine cancer. Cancer. 1996;78:441–447.
4. Ito H, Nakayama H, Noda K, et al. A case of lung metastasis from endometrial adenoacanthoma 17 years after initial treatment. Jpn J Clin Oncol. 2001;31:337–340.
5. Miyamoto H, Jones CE, Raymond DP, et al. Pulmonary metastases from uterine neoplasms after long tumour-free interval: four cases and review of the literature. Pathology. 2009;41:234–241.
6. Motzer RJ, Bander NH, Nanus DM. Renal-cell carcinoma. N Engl J Med. 1996;335:865–875.
7. Miyao N, Naito S, Ozono S, et al. Late recurrence of renal cell carcinoma: retrospective and collaborative study of the Japanese Society of Renal Cancer. Urology. 2011;77:379–384.
8. Watanabe Y, Harada A, Aoki M, et al. Pulmonary metastasectomy 31 years after surgery for renal cell carcinoma. Ann Thorac Surg. 2015;99:2195–2197.
9. Chen AM, Garcia J, Granchi PJ, et al. Late recurrence from salivary gland cancer: when does “cure” mean cure? Cancer. 2008;112:340–344.
10. Sung MW, Kim KH, Kim JW, et al. Clinicopathologic predictors and impact of distant metastasis from adenoid cystic carcinoma of the head and neck. Arch Otolaryngol Head Neck Surg. 2003;129:1193–1197.
11. Maziak DE, Todd TR, Keshavjee SH, et al. Adenoid cystic carcinoma of the airway: thirty-two-year experience. J Thorac Cardiovasc Surg. 1996;112:1522–1531.
12. Seo SI, Lim SB, Yoon YS, et al. Comparison of recurrence patterns between ≤5 years and >5 years after curative operations in colorectal cancer patients. J Surg Oncol. 2013;108:9–13.
13. Hulscher JB, van Sandick JW, Tijssen JG, et al. The recurrence pattern of esophageal carcinoma after transhiatal resection. J Am Coll Surg. 2000;191:143–148.
14. Cormio G, Rossi C, Cazzolla A, et al. Distant metastases in ovarian carcinoma. Int J Gynecol Cancer. 2003;13:125–129.
15. Pollard JW. Defining metastatic cell latency. N Engl J Med. 2016;375:280–282.
Keywords:

lung; metastatic; carcinoma; adenocarcinoma; nodules; late recurrence

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.