Discussion
According to several autopsy studies, gastrointestinal metastasis of primary lung cancer occurs in 4.7–14.0% of cases (3–5). However, in past clinical studies, the incidence of gastrointestinal metastasis has been reported to be as low as 0.2–1.7% (5–8), and in the current study, the clinical prevalence of gastrointestinal metastasis of lung cancer was ~0.33% (7/2,066). The method underlying the spread of metastasis to the intra-abdominal region is believed to involve hematogenous and lymphatic routes (12).
In the present study, the gastrointestinal metastases in 2 cases were formed as submucosal tumors with a normal overlying mucosa. In such cases, metastatic tumors appear in submucosal locations in the gastrointestinal tract with normal overlying mucosa as a result of lymphatic or bloodstream dissemination to the gastrointestinal tract (13). Therefore, it may be difficult to obtain an accurate diagnosis using a biopsy due to the presence of the overlying mucosa.
The most common histological type in the current study was adenocarcinoma. Although certain clinical studies and autopsy series (14–16) have shown adenocarcinoma to be prominent, as observed in the present study, other clinical studies (3,4) have demonstrated that squamous cell carcinoma, large cell carcinoma and pleomorphic carcinoma are more frequent (5,17–21). Therefore, the histological type predominantly associated with gastrointestinal metastasis remains unclear.
The left upper lobe was the predominant primary location of lung cancer in the present study. In a study by Yang et al (22), the left and right upper lobes were predominant. By contrast, there were no significant differences in the primary site of lung cancer in an autopsy study by Yoshimoto et al (5). Furthermore, in past case studies (16,19–21,23), the primary site of lung cancer included a range of lobes. To the best of our knowledge, no previous studies have clarified the predominant primary lobe. In general, lung cancer commonly affects the upper lobes more frequently than the lower lobes (24) and the right lung more often than the left (25). Therefore, the primary lobe associated with gastrointestinal metastasis may more frequently be the upper lobes than the lower lobes.
Gastric metastasis arising from lung cancer is extremely rare, and only a few studies have been published on the subject (26–28). In the current study, the 2 patients with gastric metastasis presented with symptoms of tarry stools and anemia due to chronic bleeding, and the cause of mortality was uncontrolled bleeding in each case. Surgical resection of gastric metastasis for palliation may result in prolonged survival. Regarding chemotherapy, chemotherapy-induced gastric perforation as a complication of lung cancer treatment has been reported in a previous study. According to this study (27), chemotherapy-induced necrosis of metastatic tumors may lead to gastric perforation. In the present cases, the administration of chemotherapy may have worsened the bleeding of the gastric metastasis due to necrosis. Therefore, we suggest that surgical resection of gastric metastasis is a beneficial option for palliation in correctly selected patients.
Previous studies have also reported that the small bowel is the most common gastrointestinal metastatic site of lung cancer (5,22,29). The prognosis of such lesions is worse than that of other locations (22,29), and emergent intervention is often required, as small bowel involvement regularly leads to perforation, which is the most common presentation, as well as obstruction or bleeding (22,29,30). In the current study, 3 out of the 7 patients presented with small bowel involvement. Although all of these patients underwent emergent surgery, they exhibited a relatively shorter survival time than the subjects with other gastrointestinal metastatic sites. This may be since small bowel metastasis is typically associated with widespread disease. Stenbygaard et al reported that gastrointestinal metastases usually occur as a component of otherwise widespread metastatic diseases (14). In addition, within the study by McNeill et al, it was reported that 46 patients with small bowel metastases had at least one other site of metastatic disease, with a mean of 4.8 sites (3). In the present study, multiple metastatic nodules were detected in the mesentery based on the intraoperative findings, and the cause of mortality was carcinomatous peritonitis in each case.
Large bowel metastasis of lung cancer is also extremely rare. In the current study, while 1 patient with metastasis of the appendix developed acute appendicitis and succumbed as a result of carcinomatous peritonitis 4 months later, another patient with metastasis of the ascending colon remains alive 8 years after a right hemicolectomy. In the case of metastasis of the appendix, perforation of the appendix may have led to dissemination to the intra-abdominal region or the metastasis may have been a result of widespread intra-abdominal dissemination with respect to the cause of mortality due to carcinomatous peritonitis.
To the best of our knowledge, only a few studies of long-term survival in cases of large bowel metastasis of lung cancer have been published (17,31). In the current study, we hypothesize that the long-term survival in 1 case was due to the presence of solitary metastasis with resectable lesions.
There was no case of esophageal metastasis in the present study. Esophageal metastasis is also extremely rare, with only a few published clinical studies on lesions of esophageal metastasis (32,33), although the incidence of this condition is 6.4% according to an autopsy report by Antler et al (4). Therefore, more thorough scientific documentation is required to discuss the characteristics of esophageal metastasis.
The interval period between the diagnosis of the lung tumors and the detection of gastrointestinal metastasis was within 1 year in the present study, with the exception of the patient with metastasis to the appendix. It is noteworthy that the interval in all patients, including those with stage I disease who underwent complete resection, was so short (≤7 months). According to the study by Yoshino et al (34), even patients with recurrence in distant organs can expect a long survival time if they receive treatment in the early pathological stage of primary cancer or have resectable recurrent disease. However, there were no cases of gastrointestinal metastasis in this study. The prognosis of patients with recurrence in distant organs, including the gastrointestinal tract, may be worse than that of patients with recurrence in distant organs, excluding the gastrointestinal tract, particularly those with symptomatic gastrointestinal metastasis (22), despite being in the early pathological stage. However, as aforementioned, long-term survival is possible if the gastrointestinal lesions are solitary and resectable. Therefore, physicians should perform comprehensive evaluations to assess asymptomatic gastrointestinal metastasis during follow-up. In the current case of large bowel solitary metastasis, FDG-PET images were effective for detecting the metastases. According to the findings of the study by Lardinois et al, solitary extrapulmonary lesions are observed on PET-CT imaging in 21% of patients with non-small cell lung cancer (35). Furthermore, PET scanning may reveal a higher incidence of gastrointestinal metastasis than previously suspected (36), although additional studies are required to identify better methods for recognizing and treating gastrointestinal metastasis.
In conclusion, the presence of clinical gastrointestinal metastasis may be life threatening, and comprehensive evaluations are required to detect and monitor gastrointestinal metastasis during follow-up. The limitations of the study were that it was a single center, retrospective study with a small sample size. Therefore, prospective, randomized trials are required to verify the frequency and clinical characteristics of metastases to the gastrointestinal tract.