With Institutional Review Board approval (IRB no. 2015-06-104), we reviewed medical records to identify uterine cancer patients diagnosed with pulmonary metastases who underwent curative resection via thoracotomy or video-assisted thoracic surgery (VATS) between June 1994 and December 2011. At our institution, the following selection criteria were used to identify candidates for pulmonary metastasectomy: (1) controlled primary tumor, (2) no extrapulmonary lesions at the time of metastasectomy, (3) pulmonary lesions amenable to surgical resection based on chest computed tomography (CT) or positron emission tomography (PET)-CT scan, (4) clinical status and pulmonary function compatible with planned operation, and (5) more effective treatment options unavailable. All patients underwent a chest CT or PET-CT prior to operation, and the resectability of pulmonary metastatic lesions was discussed with thoracic surgeons. All procedures were performed by thoracic surgeons. Pulmonary biopsies for the confirmation of metastasis were excluded.

Demographic, clinicopathologic, surgical, and survival data were retrospectively collected from medical records. Potential prognostic variables included in this study were age at pulmonary metastasectomy, initial stage (following the 2009 International Federation of Gynecology and Obstetrics [FIGO] staging system), symptoms related to lung metastasis, laterality, number and largest size of metastatic foci, disease-free interval (DFI), post-metastasectomy chemotherapy and recurrence after metastasectomy. The number and largest size of pulmonary lesions were documented based on final pathologic reports. Symptoms related metastases were limited to the symptoms of the respiratory systems in medical record at the time of metastatic diagnosis. For investigation of performance status, we used Eastern Cooperative Oncology Group (ECOG) performance status. DFI was defined as the time from diagnosis of the primary tumor to the detection of pulmonary metastases. Survival from initial metastasectomy was defined as the interval between pulmonary metastasectomy and death or last follow-up.

Survival curves were generated according to the Kaplan-Meier method. To evaluate the prognostic significance of the variables, the log-rank test and Cox proportional hazards model were used for univariate and multivariate analysis, respectively. All statistical analyses were performed using IBM SPSS ver. 21.0 (IBM Co., Armonk, NY, USA). A p<0.05 was considered statistically significant and all p-values were two-sided.

Demographics of patients with pulmonary metastasis are shown in Table 1. The median age at diagnosis of the primary tumor was 53 years (range, 34 to 72 years). Twelve patients (41.4%) had stage I, two (6.9%) stage II, six (20.7%) stage III, and nine (31.0%) stage IV disease according to the 2009 FIGO staging system. Histopathologic findings were endometrial adenocarcinoma in 15 patients (41.4%) and leiomyosarcoma in 8 patients (27.6%). All patients underwent primary surgical treatment. A staging operation, including hysterectomy, bilateral salpingoophorectomy and lymphadenectomy and/or omentectomy, was performed in 19 patients (65.6%). Five patients (17.2%) had only a hysterectomy and five (17.2%) had a hysterectomy with bilateral salpingoophorectomy. For adjuvant therapy after primary treatment, thirteen patients (44.8%) did not receive any adjuvant chemotherapy, nine (31.0%) received radiation therapy, one (3.4%) received concurrent chemoradiotherapy, and six (20.7%) had chemotherapy.

The characteristics of pulmonary metastases are shown in Table 2. The median age at pulmonary metastasectomy was 54 years (range, 35 to 74 years). Most patients had no symptoms suggestive of lung lesions (21/29, 72.4%). Eight patients showed symptoms related to lung lesions including cough (4), dyspnea with cough (3) and hemoptysis (1). Three patients (10.3%) had lung metastasis at the time of primary tumor diagnosis and one of them underwent pulmonary metastasectomy concomitant with surgical resection of the primary tumor. Of the 26 remaining patients who subsequently developed pulmonary metastasis, 23 patients (23/26, 88.4%) showed initial disease recurrence in the lung. Seventeen patients (58.6%) had unilateral lung lesions as determined by chest CT or PET-CT. According to the final pathology, 17 patients (58.6%) had less than three metastatic lesions and 18 patients (62.1%) patients had lesions larger than 1cm. The median DFI was 23.4 months (range, 3.0 to 112.3 months). When the patients were subdivided by DFI, 21 patients (72.4%) had greater than 12 months of DFI. The majority of patients (69%) received post-metastasectomy chemotherapy.

A total of 41 surgical procedures for the resection of pulmonary lesions were performed in these 29 patients via thoracotomy (4/41, 9.8%) and VATS (37/41, 90.2%); 22 patients had a single pulmonary operation, 7 patients had multiple operations (4 patients had two, 2 patients had three, and 1 patient had five) (Table 3). The patient who underwent five pulmonary metastasectomies was diagnosed with primary leiomyosarcoma and showed no evidence of lung lesions after the fifth operation. The 41 total pulmonary resections consisted of 32 wedge resections (78.0%), 6 lobectomies with mediastinal lymphadenectomy (14.6%), 1 lobectomy (2.4%), 1 lobectomy with partial pericardiectomy (2.4%), and 1 segmentectomy (2.4%). The pathologic assessment revealed metastasis from uterine cancer in 39 (95.1%) and primary lung cancer in 2 cases (4.9%). There was no mortality within the first 30 days after operation. One patient developed a hemothorax postoperatively.

Univariate and multivariate analyses were performed to identify relationships between survival and prognostic variables. Table 4 shows the 5-year survival rate and results of the univariate and multivariate analyses according to the potential prognostic variables. The 5-year survival rate was 48.2%. The median overall survival of the study population was 26 months (range, 2 to 117 months) and 12 patients (41.3%) were dead at the time of censor. Significant prognostic variables for survival from metastasectomy on univariate analysis were symptoms related to lung metastasis (hazard ratio [HR], 4.013; 95% confidence interval [CI], 1.272 to 12.655; p=0.011) and number of pulmonary metastatic lesions (>3, HR, 3.822; 95% CI, 1.119 to 13.059; p=0.023). On multivariate analysis, which included statistically significant variables from the univariate analysis, pulmonary symptoms (HR, 4.011; 95% CI, 1.219 to 13.200; p=0.022) and more than three metastatic lesions (HR, 3.885; 95% CI, 1.076 to 14.029; p=0.038) were significant prognostic factors associated with survival following metastasectomy. Figs. 1, 2 show survival curves from the time of metastasectomy influenced by significant prognostic factors, pulmonary symptoms and number of metastatic lesions.