A 79-Year-Old Man With Progressive Dyspnea and Multiple Pulmonary Nodules

A 79-year-old man was referred to our department because of an approximate six-month history of breathlessness and abnormality on chest radiograph. The patient had type 2 diabetes mellitus and prostatic hyperplasia but no history of pulmonary or autoimmune disease. He was a lifelong nonsmoker and denied any specific history of inhalational injury. He denied a history of bodyweight loss. Naftopidil (alpha-1 blocker) and mirabegron (beta-3 agonist), prescribed for the treatment of prostatic hyperplasia two months prior, were discontinued on referral due to the suspicion of drug-induced pneumonitis. Physical examination revealed fine crackles from posterior lung base. The patient did not exhibit clubbed fingers or abdominal/back pain and had no other specific findings. He was afebrile with a pulse rate of 81 beats/min, a respiratory rate of 14 breaths/min, BP of 104/54 mm Hg, and 96% oxygen saturation under ambient air. Routine blood tests showed mild anemia (hemoglobin, 12.9 g/dL) and renal dysfunction (BUN, 27.3 mg/dL; creatinine, 1.07 mg/dL). Chest radiograph showed interstitial shadows and bilateral multiple small nodules that had progressed over the past 14 months (Fig 1). Chest high-resolution CT (HRCT) scans revealed poorly defined, clustered, predominantly centrilobular nodules with some areas of coalescence and interstitial shadows with traction bronchiectasis in the right lower lobe (Fig 2A-C). Further laboratory examinations revealed the following abnormalities and normal studies: Klebs von den Lungen-6, 594 U/mL (normal range, 0 to 500); surfactant protein D, 223.4 ng/mL (normal range, 0 to 110); carcinoembryonic antigen, 14.8 ng/mL (normal range, 0 to 5.0); carbohydrate antigen 19-9, 382 U/mL (normal range, 0 to 38.0); sialyl Lewis X-i antigen, 201.3 U/mL (normal range, 0 to 38.0); soluble IL-2 receptor, 906 U/mL (normal range, 127 to 582); and angiotensin converting enzyme, 8.7 U/L (normal range, 8.3 to 21.4). No autoimmune antibodies were detected. Bronchoscopic analyses produced no diagnostic results (BAL fluid from the right middle lobe medial segment: total cell counts, 0.57 × 105/mL; macrophages, 70%; neutrophils, 11%; lymphocytes, 12%; eosinophils, 7%; cytology showed no atypical cells; transbronchial lung biopsy showed only nonspecific inflammation; and cultures were negative). Because the tumor marker tests (carcinoembryonic antigen, carbohydrate antigen 19-9, sialyl Lewis X-i antigen) suggested the existence of adenocarcinoma, a whole-body screening was performed. However, abdominal ultrasonography and CT scanning did not detect any specific findings. 18F-fluorodeoxyglucose (FDG)-PET/CT scanning showed diffuse bilateral uptake of FDG in peripheral lungs. No significant uptake was detected in mediastinal or hilar lymph nodes or any other intraabdominal/retroperitoneal organs (Fig 3).Figure 2A, B, Chest CT scan on admission shows bilateral multiple nodules; fibrotic changes with bronchiectasis existed predominantly in the right lower lobe. High-resolution CT scan shows that centrilobular nodules on admission (C) got worsened by day 20 (D), and day 27 (E), which resulted in panlobular to diffuse consolidations.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 318F-fluorodeoxyglucose-PET/CT images at the levels of tracheal bifurcation (A), lung base (B), and pancreas (C) show diffuse 18F-fluorodeoxyglucose uptake in peripheral lung and lung bases. No significant uptake was observed elsewhere.View Large Image Figure ViewerDownload Hi-res image Download (PPT) His respiratory failure progressed severely; CT scanning showed enlargement and fusion of multiple nodules over a course of 20 days (Fig 2D) and panlobar consolidations by 27 days (Fig 2E). Although sputum samples were obtained repeatedly for cytology, no diagnostic malignant cells were identified. Antibiotic and systemic corticosteroid therapy targeting infectious disease and acute exacerbation of interstitial lung disease was provided, despite the lack of sufficient diagnostic evidence. He showed partial relief from respiratory symptoms but died of respiratory failure on day 40 after admission. Autopsy was performed to investigate the pathology of the disease. What is the diagnosis? Diagnosis: Diffuse pulmonary metastasis of occult pancreatic carcinoma Pancreatic carcinoma is highly lethal, with the worst 5-year relative survival rate (9%) of all cancers in the United States,1Siegel R.L. Miller K.D. Jemal A. Cancer statistics, 2019.CA Cancer J Clin. 2019; 69: 7-34Crossref PubMed Scopus (10226) Google Scholar likely a result of low diagnostic frequency at early stages and limited treatment response.2Kamisawa T. Wood L.D. Itoi T. Takaori K. Pancreatic cancer.Lancet. 2016; 388: 73-85Abstract Full Text Full Text PDF PubMed Scopus (1127) Google Scholar Lung is one of the frequent metastatic sites for pancreatic carcinoma, along with liver, bone, and kidney.3Kamisawa T. Isawa T. Koike M. Tsuruta K. Okamoto A. Hematogenous metastases of pancreatic ductal carcinoma.Pancreas. 1995; 11: 345-349Crossref PubMed Scopus (98) Google Scholar In the metastatic stage, a mass lesion at the primary pancreatic site usually is detected. Surgical resection of the primary lesions occurred in 2055 of the 28,655 patients who were registered to the Japanese pancreatic cancer database between 2000 and 2004. Among these, only 13 cases (0.45%) exhibited a primary lesion size ≤10 mm, and only one of these cases had metastasis.4Matsuno S. Egawa S. Fukuyama S. et al.Pancreatic cancer registry in Japan: 20 years of experience.Pancreas. 2004; 28: 219-230Crossref PubMed Scopus (273) Google Scholar,5Egawa S. Pancreatic cancer less than 1 cm in nation-wide pancreatic cancer registry [in Japanese].Tan to Sui. 2009; 30: 311-316Google Scholar To the best of our knowledge, this is the first report of occult pancreatic carcinoma to show distant metastasis with no visibly identifiable primary lesion. In the present case, based on the elevation of several serum tumor markers and chest imaging, the most likely diagnosis was pulmonary metastasis of an abdominal or peritoneal organ cancer, including pancreatic carcinoma possibly accompanied by tumor emboli in pulmonary arteries and/or lymphangitic carcinomatosis. However, we could not obtain sufficient diagnostic evidences despite intensive investigations that included FDG-PET/CT, abdominal ultrasonography, and cytology/histology of sputum and bronchoscopic samples. Video-assisted thoracoscopic lung biopsy, which could have been diagnostic, could not be performed because of the rapid, progressive deterioration of the patient’s condition. The differential diagnoses included acute exacerbation of interstitial pneumonia, undiagnosed lymphoproliferative disease, sarcoidosis, bronchopneumonia, and pulmonary septic embolism, which we widely covered by treatment with antibiotics and corticosteroids. Metastatic pulmonary cancer typically shows well-defined solid nodule(s) or tumor(s). However, especially when the metastasis shows lepidic growth in the lung, the radiologic appearance can vary, with nodules, ground-glass attenuations, and consolidations.6Seo J.B. Im J.G. Goo J.M. Chung M.J. Kim M.Y. Atypical pulmonary metastases: spectrum of radiologic findings.Radiographics. 2001; 21: 403-417Crossref PubMed Scopus (229) Google Scholar Rosenblatt et al7Rosenblatt M.B. Lisa J.R. Collier F. Primary and metastatic bronchiolo-alveolar carcinoma.Dis Chest. 1967; 52: 147-152Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar reported that, of 215 cases of metastatic cancers to the lung, 34 cases (15.8%) showed lepidic growth. Among these, pancreatic cancer was the most common (24%) followed by colon (18%), breast (18%), stomach (12%), and kidney (9%) cancers. Such metastases often mimic invasive mucinous adenocarcinoma (IMA) of the lung (formerly called mucinous bronchioloalveolar carcinoma [BAC]).8Foster C.S. Mucus-secreting ‘alveolar-cell’ tumour of the lung: a histochemical comparison of tumours arising within and outside the lung.Histopathology. 1980; 4: 567-577Crossref PubMed Scopus (13) Google Scholar In the present case, the chest HRCT initially showed peripheral (mainly centrilobular) multiple small nodules, which progressed to be pan-lobular and then to diffuse consolidations. This transformation was similar to that observed in the progression of IMA (ie, mucous obstruction initially occurs in the periphery and then, as the disease progresses, more proximal airways are filled with mucus). In this case, however, the multiple small nodular lesions were not exclusively centrilobular, which suggested various potential diagnoses, such as respiratory infections (eg, mycobacterial and atypical pneumonia), lymphoproliferative diseases, metastatic cancer, tumor or septic emboli, lymphangitic carcinomatosis, and so on. Moreover, the existence of interstitial lung disease made the situation more complex. In the current case, FDG-PET/CT was performed to investigate for malignancies. Though it did not identify the primary lesion, it provided several important findings. First, mucus-plugged peripheral lungs showed diffuse FDG uptake. Although data for metastatic lung cancers that show lepidic growth are limited, there have been several studies of FDG-PET images for IMA that report similar radiographic and morphologic characteristics with the present case. Heyneman and Patz9Heyneman L.E. Patz E.F. PET imaging in patients with bronchioloalveolar cell carcinoma.Lung cancer. 2002; 38: 261-266Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar reported that local BAC were frequently false-negative for FDG uptake, but six of seven patients (86%) with multifocal BAC had diffusely positive FDG uptake. Suárez-Piñera et al10Suarez-Pinera M. Belda-Sanchis J. Taus A. et al.FDG PET-CT SUVmax and IASLC/ATS/ERS histologic classification: a new profile of lung adenocarcinoma with prognostic value.Am J Nucl Med Mol Imaging. 2018; 8: 100-109PubMed Google Scholar reported lower FDG uptake in IMA; Nakamura et al11Nakamura H. Saji H. Shinmyo T. et al.Close association of IASLC/ATS/ERS lung adenocarcinoma subtypes with glucose-uptake in positron emission tomography.Lung Cancer. 2015; 87: 28-33Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar showed comparable FDG uptake in IMA with in solid predominant tumors. Second, the fibrotic lung lesion in the right lower lobe showed high FDG uptake. Idiopathic pulmonary fibrosis and other interstitial pneumonia generally show positive FDG uptake, presumably because of inflammation-mediated up-regulation of glucose metabolism.12Meissner H.H. Soo Hoo G.W. Khonsary S.A. Mandelkern M. Brown C.V. Santiago S.M. Idiopathic pulmonary fibrosis: evaluation with positron emission tomography.Respiration. 2006; 73: 197-202Crossref PubMed Scopus (46) Google Scholar,13Groves A.M. Win T. Screaton N.J. et al.Idiopathic pulmonary fibrosis and diffuse parenchymal lung disease: implications from initial experience with 18F-FDG PET/CT.J Nucl Med. 2009; 50: 538-545Crossref PubMed Scopus (99) Google Scholar Therefore, when malignancy and interstitial lung disease coexist, evaluation of FDG-PET scans in the lung may not be diagnostic. Last, this case highlights one pitfall of FDG-PET scans, which is poor sensitivity to detect lymph node metastasis. Although autopsy (described later) revealed that lymphatic metastases were extensive, each node had only small foci of metastatic cancer cells (diameters, <5 mm), and these small foci were not resolved by FDG-PET scan. It is reported that sensitivity of FDG-PET/CT scans to detect local lymph node metastasis of pancreatic cancer is poor.14Kauhanen S.P. Komar G. Seppanen M.P. et al.A prospective diagnostic accuracy study of 18F-fluorodeoxyglucose positron emission tomography/computed tomography, multidetector row computed tomography, and magnetic resonance imaging in primary diagnosis and staging of pancreatic cancer.Ann Surg. 2009; 250: 957-963Crossref PubMed Scopus (217) Google Scholar Similarly, in lung cancer, Nomori et al15Nomori H. Watanabe K. Ohtsuka T. Naruke T. Suemasu K. Uno K. The size of metastatic foci and lymph nodes yielding false-negative and false-positive lymph node staging with positron emission tomography in patients with lung cancer.J Thorac Cardiovasc Surg. 2004; 127: 1087-1092Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar reported that none of the lymph node metastasis of lung cancer with the foci size of <4 mm could be detected by FDG-PET scanning. Autopsy revealed that both lungs were involved diffusely, with accumulated whitish mucus produced by cancer cells, especially in the subpleural parenchyma. Honeycomb changes at the right lung base reflected the preexisting chronic interstitial pneumonitis/fibrosis (Fig 4A). Microscopically, lepidic growth of tall columnar cells with intracytoplasmic mucin along the alveolar wall was observed throughout both lungs (Fig 4B and C). Cytologic atypia was mild, which suggested the reason that malignant cells were not detected in BAL fluid or induced sputa. These characteristics are very similar to those of IMA.16Travis W.D. Brambilla E. Noguchi M. et al.International association for the study of lung cancer/American thoracic society/European respiratory society international multidisciplinary classification of lung adenocarcinoma.J Thorac Oncol. 2011; 6: 244-285Abstract Full Text Full Text PDF PubMed Scopus (3200) Google Scholar Lymphangitic carcinomatosis and tumor emboli in the pulmonary arteries were also detected (Fig 4D), but these were minor components of the pulmonary involvement of the tumor. Therefore, acute deterioration of respiratory symptoms was considered to have been caused by extensive lepidic spread of the cancer. Despite thorough macroscopic and microscopic examination, no primary lesion was identified in the lungs or other organs, except pancreas in which pancreatic intraepithelial neoplasia I-III (precancerous lesions) that are associated with multiple foci of invasive carcinoma was observed (Fig 5A). Although each invasive focus was small and macroscopically undetectable, the cancer was associated with marked lymphatic invasion and extensive lymph node involvement in peripancreatic, paraaortic, mediastinal, and hilar nodes (Fig 5B). As mentioned earlier, none of these metastatic foci exceeded 5 mm in size. Both pulmonary and pancreatic lesions showed similar histologic condition, and immunohistochemical characteristics were identical: negative for thyroid transcription factor 1, NapsinA, keratin 20, mucin 2 and p16; positive for keratin 7 and keratin 19; partially positive for mucin 5AC and mucin 6; and wild type for p53. Finally, we concluded that the pulmonary lesions that showed extensive lepidic growth mimicking IMA were metastases from occult pancreatic carcinoma and that the multiple centrilobular nodules, which progressed to be diffuse consolidations in the HRCT images, reflected the mucous accumulation spread from the peripheral lungs to distal airways.Figure 5Autopsy findings of pancreas. A, Minimal stromal invasion (arrowheads) of pancreatic carcinoma associated with pancreatic intraepithelial neoplasia (hematoxylin-eosin staining; original magnification, ×100). B, Extensive lymphatic invasion in the adipose tissue around pancreas (hematoxylin-eosin staining; original magnification, ×40).View Large Image Figure ViewerDownload Hi-res image Download (PPT) To the best of our knowledge, this is the first report of occult pancreatic carcinoma accompanying pulmonary metastases. This case showed lepidic growth of cancer cells and mucous accumulation in the lung, which mimicked IMA of the lung. The combination of progressing centrilobular opacities in the chest HRCT and the elevation of several tumor markers were clues pointing to the diagnosis. Furthermore, this case highlighted the failure of FDG-PET scanning to detect lymph node metastases when the metastatic loci are small. Finally, pancreatic carcinoma should be considered in the differential diagnosis of multiple centrilobular lung nodules, even without clinical detection of the primary lesion.