Long-term Survival Outcomes of Surgical Resection for Lung Adenocarcinoma with Intraoperatively Diagnosed Pleural Metastasis: Target Treatment Era
Article information
Abstract
Purpose
This study aimed to evaluate the clinical impact of main tumor resection on long-term survival compared with pleural biopsy alone in patients with lung adenocarcinoma who were intraoperatively diagnosed with pleural metastasis.
Materials and Methods
A total of 176 patients with adenocarcinoma who had unexpected pleural metastasis detected during surgery from 2002 to 2021 were retrospectively analyzed. Each surgeon decided whether to perform main tumor resection or pleural biopsy alone.
Results
The patients were grouped based on the surgical approaches: main tumor resection (resection group; n=83) and pleural biopsy only (O&C group; n=93). The resection group had better overall survival (OS; 10-year survival, 27.9% vs. 9.4%; median survival, 68.3 vs. 36.6 months; p < 0.01) and locoregional progression-free survival (10-year survival, 12.5% vs. 7.1%; median survival, 19.6 vs. 10.6 months; p < 0.01) than the O&C group. Similar results were found for OS in patients who received tyrosine kinase inhibitors (TKIs) as first-line therapy (10-year survival, 49.2% vs. 15.0%; median survival, 72.2 vs. 45.4 months; p=0.03), patients who did not undergo TKIs treatment (10-year survival, 29.4% vs. 9.2%; median survival, 82.4 vs. 23.8 months; p < 0.01), and patients with positive target gene mutation (10-year survival, 31.7% vs. 10.1%; median survival, 72.2 vs. 33.7 months; p < 0.01). In multivariate analysis, pleural biopsy only (hazard ratio, 1.73; p=0.04) was a significant predictor of OS.
Conclusion
Main tumor resection can improve survival in patients with lung adenocarcinoma who had unexpected pleural metastasis during operation.
Introduction
The 8th TNM staging system classifies non–small cell lung cancer (NSCLC) with pleural metastasis as stage IV-M1a, which has a 2- and 5-year survival of 23% and 10%, respectively [1]. In this stage, patients are considered to have systemic disease, and surgical resection is not recommended. Despite advancements in diagnostic accuracy, unexpected pleural seeding during surgery poses a challenge, which often leads to the termination of the procedure with pleural biopsy alone. In such instances, the clinical implications of primary tumor resection, particularly in cases of localized pleural seeding, remain unknown.
Recently, several studies have reported the clinical outcomes of patients with NSCLC who were intraoperatively diagnosed with pleural seeding [2-4]. Patients with unforeseen pleural metastasis presented better survival than those with clinical pleural metastasis. Moreover, main tumor resection could improve survival. However, most studies presented short-term or intermediate outcomes, which may not accurately reflect long-term survival.
Therefore, the present study aimed to investigate the clinical impact of main tumor resection compared with pleural biopsy alone in patients with lung adenocarcinoma who were intraoperatively diagnosed with pleural metastasis, focusing on long-term survival.
Materials and Methods
1. Patients
This study included patients who were intraoperatively diagnosed with pleural metastasis of lung adenocarcinoma between July 2002 and October 2021. Before operation, the patients were clinical stage M0. All data were obtained from the medical records.
2. Preoperative evaluation and operation
The diagnostic, staging, and surgical resection procedures employed in this study adhered to well-established protocols, as detailed elsewhere [5]. Preoperative staging assessments encompassed a comprehensive array of modalities, including chest radiograph, blood laboratory analysis, brain computed tomography (CT) or magnetic resonance imaging (MRI), chest and abdominal CT, positron emission tomography (PET) scan, and pulmonary function tests. In cases where clinical N2 disease was suspected, mediastinal lymph node biopsy was conducted under the guidance of endobronchial ultrasonography or endoscopic ultrasonography. Moreover, chest MRI was used to assess resectability for cases with suspected invasion of adjacent structures. All patients included in this study cohort were confirmed to be resectable.
Pleural seeding was pathologically confirmed by intraoperative frozen section biopsy. After confirmation of pleural seeding, the surgeon decided whether to terminate the surgery (O&C group) or to resect the main lesion (resection group). Generally, pleural biopsy only was performed in cases where pleural seeding presented as diffuse (which is considered as unresectable). However, if pleural metastasis was limited to a few nodules and pulmonary function was satisfactory, primary tumor resection with clear margins was undertaken.
3. Tyrosine kinase inhibitor treatment
All patients were recommended to undergo testing for epidermal growth factor receptor (EGFR) mutation or anaplastic lymphoma kinase (ALK) rearrangement.
Since 2008, tyrosine kinase inhibitors (TKIs) have been used in patients. However, they were primarily administered upon recurrence following first-line platinum-based chemotherapy. TKIs were selectively used as first-line treatment in a subset of patients enrolled in clinical trials in 2005-2006. Since 2010, they have been available as first-line treatment for stage IV NSCLC patients with EGFR mutations or ALK rearrangement.
If patients had positive EGFR mutation or ALK rearrangement, the corresponding targeted drugs were recommended for first-line treatment. However, platinum-based chemotherapy was recommended for patients who did not harbor these gene mutations or did not opt for TKI treatment because of various reasons, including cost, allergies, adverse effects, or other factors.
4. Follow-up
Postoperative surveillance was performed with a chest CT scan every 6 months for 5 years after operation. In case of any suspicious lesions or symptoms, a PET-CT scan, brain MRI, or bone scan was performed.
Based on previous studies, “local progression” was defined as the enlargement of the primary lesion or recurrence at the resection site. “Regional progression” was defined as increasing pleural effusion, pleural nodules, lung lesions, or ipsilateral lymph node recurrence and enlargement. “Distant metastasis” was defined as new lesions in the contralateral lung or any other organ other than the lung (brain, bone, etc.). Vital status was checked through the institutional medical records and the National Population Registry of the Korea National Statistical Office.
5. Statistical analysis
Variables are presented as mean±standard deviation or number (percentage). The characteristics of patients between the two groups were compared using Student’s t test and Pearson’s chi-square tests. Statistical significance was considered at p < 0.05. Overall survival (OS) was defined as the time from surgery to death or the last follow-up. OS was tested using Kaplan-Meier survival plots and analyzed with log-rank tests. The Cox proportional hazards model was fitted and adjusted for several parameters associated with survival. The results are presented as the hazard ratio (HR) and 95% confidence intervals (CIs). All tests were two-sided, and all analyses were conducted using R software ver. 4.2.2 (R Foundation for Statistical Computing).
Results
1. Patient
Our institution performed surgery for lung cancer on 12,772 consecutive patients between July 2002 and October 2021. Among them, we identified 176 patients (mean age, 61.0±10.6 years, 84 females) who were intraoperatively diagnosed with pleural metastasis of lung adenocarcinoma. These patients were divided into two groups based on the extent of surgical intervention: resection group (n=83), which underwent main tumor resection, and O&C group (n=93), which underwent pleural biopsy only. Table 1 shows the baseline characteristics of patients, including age, sex, comorbidities, and pulmonary function.
Baseline characteristics of patients
No significant differences were observed in age, sex, comorbidities, pulmonary function, or malignant pleural effusion between the two groups. The clinical T category was higher in the O&C group (p < 0.01) than in the resection group. Moreover, the clinical N state appeared to be higher in the O&C group; however, the difference was not statistically significant (p=0.06).
2. Treatments
Patients’ treatment details data are summarized in Table 2. In the resection group, 27 patients (32.5%) underwent sublobar resection (segmentectomy, n=3; wedge resection, n=24), 38 patients (45.8%) underwent lobectomy, two patients (2.4%) underwent bilobectomy, and three patients (3.6%) underwent pneumonectomy. Macroscopic complete resection (R0/R1) of the primary tumor was possible in 81 patients (81/83, 97.6%).
Patients’ treatment data
EGFR or ALK TKIs were used to treat 57 (68.7%) and 60 (64.5%) patients in the resection and O&C groups, respectively (p=0.67). Among them, 34 (41.0%) and 28 (30.1%) patients in the resection and O&C groups received EGFR or ALK TKIs as first-line treatment, respectively. Meanwhile, 23 (27.7%) and 32 (34.4%) patients in the resection and O&C groups received EGFR or ALK TKIs as salvage treatment, respectively, when platinum-based chemotherapy failed.
Eight patients underwent neoadjuvant chemotherapy: six (7.2%) patients in the resection group and two patients (2.2%) in the O&C group (p=0.21). Postoperative chemotherapy was performed in 65 (78.3%) patients in the resection group and 76 (81.7%) patients in the O&C group. The remaining patients did not undergo postoperative chemotherapy because of old age and poor general condition with patient refusal. Twenty patients received postoperative immunotherapy: nine (10.8%) patients in the resection group and 11 patients (11.8%) in the O&C group. Ten patients received adjuvant radiotherapy for control of local progression or distant metastasis. The treatment plans were established by a multidisciplinary lung cancer team.
3. Clinical outcomes
The median follow-up was 44.4 months. The 5- and 10-year OS of the study population was 40.7% (34.0-49.0) and 17.5% (11.7-26.2), respectively. The resection group had better OS (10-year survival, 27.9% vs. 9.4%; median survival, 68.3 vs. 36.6 months; p < 0.01) (Fig. 1) and locoregional progression-free survival (PFS; 10-year survival, 12.5% vs. 7.1%; median survival, 19.6 vs. 10.6 months; p < 0.01) (Fig. 2A) and tended to have a higher distant PFS compared with the O&C group (10-year survival, 14.2% vs. 5.2%; median survival, 36.2 vs. 27.5 months; p=0.06) (Fig. 2B).
Ten-year overall survival of the study group patients. O&C group, pleural biopsy only; resection group, main tumor resection.
Progression-free survival of the study group patients. Locoregional (A) and distant metastasis (B) progression-free survival of the Resection group vs. O&C group. O&C group, pleural biopsy only; resection group, main tumor resection.
Patients who received TKI treatment as first-line therapy presented better OS than those who received TKIs as salvage treatment or those who did not undergo TKI treatment (median survival, 56.3 vs. 41.8 vs. 40.3 months) (Fig. 3). When analyzed according to TKI treatment, similar results were found for OS in patients who received TKIs as first-line therapy (10-year survival, 49.2% vs. 15.0%; median survival, 72.2 vs. 45.4 months; p=0.03) (Fig. 4A) and in patients who did not undergo TKI treatment (10-year survival, 29.4% vs. 9.2%; median survival, 82.4 vs. 23.8 months; p < 0.01) (Fig. 4C), but not in patients who received TKIs as salvage treatment (10-year survival: 6.8% vs. 7.5%; median survival: 45.0 vs. 36.6 months; p=0.90) (Fig. 4B). In patients with positive EGFR or ALK gene mutation, the resection group showed higher OS than the O&C group (10-year survival, 31.7% vs. 10.1%; median survival, 72.2 vs. 33.7 months; p < 0.01) (Fig. 4D). In patients who underwent TKI treatment regardless timing, there was no statistically significant survival difference (median survival, 4.64 [3.75-8.28] vs. 3.76 [2.79-5.00] years; p=0.10) (S1 Fig.).
Overall survival depending on tyrosine kinase inhibitor treatment.
Subgroup analysis according to tyrosine kinase inhibitor (TKI) treatment: patients who received TKI treatment as first-line therapy (A), patients who received TKIs as salvage treatment (B), patients who did not undergo TKI treatment (C), and patients who harbored epidermal growth factor receptor/anaplastic lymphoma kinase mutation regardless of TKI treatment (D). O&C group, pleural biopsy only; resection group, main tumor resection.
No significant differences were observed for OS regarding the clinical N status (p=0.88) (S2 Fig.).
4. Prognostic factors
In the univariable analysis, age (HR, 1.02; 95% CI, 1.01 to 1.04; p=0.01) and pleural biopsy only (HR, 1.95; 95% CI, 1.33 to 2.86; p < 0.01) were significant predictors of OS (Table 3). In the multivariate analysis, the two factors remained statistically significant (age: HR, 1.02; 95% CI, 1.00 to 1.04; p=0.02; pleural biopsy only: HR, 1.73; 95% CI, 1.01 to 2.96; p=0.04).
Prognostic factors associated with patient characteristics
Discussion
In this study cohort, patients with lung adenocarcinoma who were intraoperatively diagnosed with pleural metastasis during operation presented better OS than those with IVA NSCLC TNM stage group from International Association for the Study of Lung Cancer 8th (5-year OS, 40.7% vs. 10%) [1]. Patients who underwent main tumor resection presented better OS and PFS, which was consistent with the findings of previous studies, [3,4,6] despite the National Comprehensive Cancer Network (NCCN) guideline recommending only systemic treatments for NSCLC with pleural metastasis [7]. In their meta-analysis, Deng et al. [3] evaluated whether surgical resection of the primary tumor was superior to exploratory thoracotomy in patients who were intraoperatively diagnosed with unexpected pleural metastasis. Most of the included studies found that primary tumor resection yielded better outcomes over exploration only [3]. Given that most of these studies reported the 3-year (28.8%-82.9% vs. 10.9%-38.5%) or 5-year OS rate (14.9%-42.7% vs. 0%-19.5%, surgical resection vs. pleural biopsy only), the 10-year OS of the present study showed long-term outcomes with favorable results (10-year survival, 27.9% vs. 9.4%). Although surgical resection could not achieve complete resection of malignancy in patients with pleural metastasis, the cytoreductive effect of primary tumor resection, along with the rapid development of systemic therapy, would contribute to the outcomes.
In this study, pleural biopsy alone was generally performed in cases where pleural seeding presented as diffuse, which is considered as unresectable. Consequently, the O&C group might have included patients with more advanced pleural metastasis, although the prevalence of malignant pleural effusion did not differ significantly between the groups (p=0.83). Several studies have categorized pleural metastasis as either localized or diffuse [4,8]. These studies reported that the extent of pleural metastasis was not a significant factor for survival. However, Li et al. [8] found that malignant pleural effusion was a significant factor in univariate survival analysis, a finding that contrasts with our results (p=0.07). Notably, all patients with localized pleural metastasis in these studies underwent primary tumor resection [4,8]. TKI treatment has improved the survival outcomes of patients with lung adenocarcinoma and has become the first-line treatment recommendation according to the NCCN guidelines for advanced NSCLC with EGFR mutation or ALK rearrangement [7]. Several studies have reported that primary tumor resection did not improve survival in patients with pleural metastasis who received targeted therapy [6,8,9]. Similar findings were obtained in the present study in patients who underwent TKI treatment regardless timing (S1 Fig.). There were no significant intergroup baseline differences (S3 Table). However, when patients who received TKI treatment were subdivided into two groups (receiving TKIs as first-line therapy or salvage treatment), there were significant differences between them (Fig. 3). Patients who received TKI treatment after platinum-based chemotherapy failed had higher risk of worse prognosis. Therefore, the mixed oncologic characteristics of the population might have confounded the results. Moreover, this study provided the OS depending on driver gene mutation regardless of the use of TKI treatment, which might offset the confounding effects of different timing of TKI treatment. In NSCLC, salvage surgery may be a viable option for selected patients with locally persistent, progressive tumors or locoregional recurrence after definitive nonoperative therapy. The advent of TKI treatment and immune checkpoint inhibitors has significantly expanded the possibilities of survival improvement in patients with advanced stage and unresectable tumors. Antonoff et al. [10] have reported the feasibility of surgical resection after TKI therapy, even in stage IV NSCLC, marking a substantial advancement in the approach to advanced disease. This underscores the potential role of salvage operations, particularly after PACIFIC-type regimens, which may become more prevalent than after concurrent chemoradiation [11]. This evolving trend suggests a shift toward actively addressing the primary tumor lesion in patients diagnosed with pleural metastasis intraoperatively.
Previous studies exploring similar subjects identified surgical resection of the main tumor as a prognostic factor for PFS and OS [2,4,6,8,9]. In the present study, multivariate analysis also showed that surgical resection was a significant prognostic factor, suggesting the importance of main tumor resection even on long-term outcomes in patients with intraoperatively diagnosed stage IV-M1a lung adenocarcinoma.
The present study has several limitations. Although the data used in this study were collected prospectively, this study was a retrospective study, with inherent shortcomings. Because there was no established protocol to decide surgical extent in intraoperatively diagnosed pleural seeding, primary tumor resection was decided by the surgeon, which could lead to selection bias. A higher clinical T and N category might be associated with a surgeon’s tendency to select pleural biopsy alone. Moreover, patients were included from 2002 covering the beginning era of TKI therapy to maximize the number of the study population, which could increase the risk of time-trend bias. Further studies showing the effects of new-generation TKIs or immunotherapy are needed to validate the results.
In patients with lung adenocarcinoma who were intraoperatively diagnosed with unexpected pleural metastasis, surgical resection of the main tumor may improve long-term survival compared with pleural biopsy alone.
Electronic Supplementary Material
Supplementary materials are available at Cancer Research and Treatment website (https://www.e-crt.org).
Notes
Ethical Statement
This study was approved by the Asan Medical Center Ethics Committee and Review Board (No. 2025-0656), which waived the requirement for informed patient consent because of the retrospective nature of the study.
Author Contributions
Conceived and designed the analysis: Kim HR.
Collected the data: Kwon Y.
Contributed data or analysis tools: Kwon Y, Yun JK, Lee GD, Choi SH, Kim YH, Kim HR.
Performed the analysis: Kwon Y.
Wrote the paper: Kwon Y.
Conflict of Interest
Conflict of interest relevant to this article was not reported.