A prognostic nomogram for oligometastatic non-small cell lung cancer incorporating extent of primary tumor resection as a key predictor Authors: Zixuan Fei, Yang Pan, Xuanhong Jin, Leilei Wu, Junnan Ru, Yan Zhang, Yuqi Lin, Jian Zeng, Feng Li Published: 2026 ABSTRACT AbstractBackgroundThe optimal extent of surgical resection for oligometastatic non-small cell lung cancer (NSCLC) remains controversial. This study aimed to evaluate the impact of lobectomy versus sublobar resection on survival and to develop a prognostic nomogram for this patient population.MethodsThis study retrospectively analyzed data from the Surveillance, Epidemiology, and End Results (SEER) database [2010–2019] for patients with stage IV NSCLC who had a single metastatic site. We used propensity score matching (PSM) to balance covariates. The Kaplan-Meier method was employed to estimate survival, with comparisons made via the log-rank test. A prognostic nomogram was built using a Cox model and its performance was checked with area under the receiver operating characteristic (AUC) curves and calibration plots.ResultsFollowing PSM, undergoing surgery on the primary tumor was linked to better overall survival (OS) and cancer-specific survival (CSS). In the surgical group, lobectomy was associated with a significantly longer median OS (28 vs. 13 months, P<0.001) and CSS (32 vs. 14 months, P<0.001) compared to sublobar resection. In a multivariate model, both lobectomy and receiving systemic therapy before surgery were independently linked to better prognoses. The resulting nomogram demonstrated strong predictive ability, with 1-, 2-, and 3-year OS AUC values ranging from 0.74 to 0.75 in the training set and 0.72–0.73 in the validation set, showing good calibration.ConclusionsIn our large, population-based analysis, lobectomy showed a clear survival advantage over sublobar resection for individuals with oligometastatic NSCLC. Furthermore, our validated nomogram serves as a dependable instrument for estimating individual patient outcomes and guiding treatment choices.Keywords: Nomogram; oligometastatic non-small cell lung cancer (oligometastatic NSCLC); local consolidative therapy (LCT); lobectomy; Surveillance, Epidemiology, and End Results (SEER) Highlight box Key findings • Primary tumor resection, specifically lobectomy, significantly improves overall survival (OS) and cancer-specific survival (CSS) compared to sublobar resection in patients with oligometastatic non-small cell lung cancer (NSCLC). Neoadjuvant systemic therapy is also an independent protective factor. A prognostic nomogram was successfully developed and validated to predict 1-, 2-, and 3-year OS. What is known and what is new? • Local consolidative therapy, including pulmonary resection, provides survival benefits for patients with oligometastatic NSCLC, but the optimal extent of surgical resection remains controversial. • This study demonstrates that lobectomy offers superior overall and CSS compared to sublobar resection in this patient population, and introduces a validated prognostic nomogram to predict individual patient outcomes. What is the implication, and what should change now? • Lobectomy should be prioritized over sublobar resection for eligible patients with oligometastatic NSCLC. Clinicians can utilize the proposed nomogram as a reliable tool for individual risk assessment and to guide tailored treatment decisions. Introduction The application of local consolidative therapy (LCT) for patients with oligometastatic non-small cell lung cancer (NSCLC) is a quickly developing field. A growing body of evidence supports its effectiveness, particularly when pulmonary resection is used in suitable patients (1-4). A key unresolved issue, however, is determining the most effective extent of surgical resection for these individuals. Existing research that compares lobectomy and sublobar resection presents contradictory findings on the most effective surgical method. For instance, Hao and colleagues utilized the Surveillance, Epidemiology, and End Results (SEER) database to study patients with metastatic NSCLC who had their primary tumor removed (5). Their results indicated that lobectomy provided a greater survival advantage than sublobar resection. A key weakness of their work, however, was the failure to separate oligometastatic cases from the larger group of metastatic patients. Conversely, a different multi-center retrospective study focusing solely on oligometastatic NSCLC found no survival difference between the two surgical approaches (6). That study’s conclusions were weakened by a small sample size, which prevented robust comparisons. Given these discrepancies, we used the SEER database to assemble a cohort of individuals with oligometastatic lung cancer. After verifying that primary tumor resection improved prognosis in this group, we proceeded to investigate how the extent of surgery influenced patient outcomes. We present this article in accordance with the TRIPOD reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2661/rc). Methods Study design We used data from the SEER Database from 2010 to 2019. A cohort of NSCLC patients with stage IV NSCLC and a single organ metastasis was extracted. The inclusion criteria were: (I) diagnosis of at least one of the following metastases—brain, liver, or bone; (II) histological confirmation of NSCLC; (III) lung as the primary cancer site, confirmed by international standards. We excluded cases with ambiguous or indeterminate data related to tumor (T) and node (N) staging, primary cancer site, race, treatment, or survival time, as well as patients under 20 years of age. A meta-analysis by Rim and colleagues (7) found that patients with oligometastatic NSCLC benefit from local therapy, regardless of the number of metastatic deposits. Therefore, in our study, we defined patients with NSCLC involving only one metastatic organ as having oligometastatic disease. The endpoint of this study is overall survival (OS) and cancer-specific survival (CSS). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Covariates A comprehensive set of baseline characteristics was collected for all patients to serve as covariates in the analysis. Demographic data included age at diagnosis (categorized as <70 or ≥70 years), sex (male or female), and race (Asian, Black, or White). Clinicopathological features consisted of primary tumor site (upper, middle, or lower lobe), laterality (left or right), tumor size (dichotomized at ≤30 and >30 mm), and histological subtype [lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), or others]. For patients in the surgical cohort, we additionally collected data on pathological grade (I–IV) and pathologic nodal status (pN0 vs. pN+). Data regarding metastatic burden and treatment history were also obtained. This included the specific sites of distant metastasis (bone, brain, or liver) and whether surgery for distant metastases was performed. Other treatment-related variables included the administration of radiation therapy, the use of systemic therapy prior to surgery, and whether lymph node (LN) dissection was performed during the surgical procedure. Statistical analysis We used Chi-squared or Fisher’s exact tests to assess variables as needed. To balance covariates such as age, sex, race, tumor characteristics, and treatment details, we applied propensity score matching (PSM) with a 3:1 ratio. The Kaplan-Meier method was used to generate survival estimates, which were then compared using the log-rank test. To understand the relationship between patient variables and survival, we applied Cox proportional hazard models, presenting the findings as hazard ratios (HRs) with 95% confidence intervals (CIs). The final Cox model was used to create a prognostic nomogram. We evaluated the nomogram’s performance with the area under the receiver operating characteristic (ROC) curve (AUC) and calibration plots. All analyses were performed using R (version 4.3.1) and SPSS (version 29.0), and we considered a two-tailed P value below 0.05 to be statistically significant. Results Population characteristics Initially, baseline characteristics were compared between patients who received primary site surgery and those who did not (Table S1). Before matching, no significant differences were observed between the two groups regarding age, sex, race, tumor site, tumor laterality, histologic type, radiotherapy, chemotherapy, metastatic site, or surgery to metastatic site. However, patients with tumor size ≤30 mm (P<0.001) and node-negative disease (N0) (P<0.001) were more prevalent in the surgery group. After 3:1 matching for the covariates, 2,688 NSCLC patients treated with or without primary site surgery were enrolled in the following survival analysis. Within the cohort of patients treated with primary site surgery, we compared the characteristics of those who underwent lobectomy versus those who had a sublobar resection (Table 1). The sublobar resection group had a greater proportion of smaller tumors (≤30 mm: 68.2% vs. 42.0%; P<0.001) and node-negative disease (N0: 77.8% vs. 51.3%; P<0.001). In contrast, the lobectomy group received more frequent neoadjuvant systemic therapy (25.5% vs. 8.1%; P<0.001) and underwent near-universal LN dissection (92.8% vs. 33.8%; P<0.001). Radiotherapy was more common in the lobectomy cohort (72.8% vs. 56.1%; P<0.001). Furthermore, the lobectomy group had a higher incidence of brain metastases (66.2% vs. 38.9%; P<0.001), while the sublobar group had a higher prevalence of bone metastases (47.5% vs. 26.4%; P<0.001). Decision curve analysis (DCA) confirmed the clinical utility of the nomogram, demonstrating a positive net benefit for predicting 3-year OS (Figure 1). Table 1Baseline clinical, tumor, and treatment characteristics of the study cohortCharacteristicsSublobar resection (n=198)Lobectomy (n=474)PAge0.52 <70 years142 (71.7)353 (74.5) ≥70 years56 (28.3)121 (25.5)Sex0.42 Male99 (50.0)219 (46.2) Female99 (50.0)255 (53.8)Race0.18 Asian19 (9.6)35 (7.4) Black21 (10.6)34 (7.2) White158 (79.8)405 (85.4)Site0.36 Lower lobe60 (30.3)148 (31.2) Middle lobe9 (4.5)35 (7.4) Upper lobe129 (65.2)291 (61.4)Laterality0.32 Left95 (48.0)206 (43.5) Right103 (52.0)268 (56.5)Tumor size<0.001 ≤30 mm135 (68.2)199 (42.0) >30 mm63 (31.8)275 (58.0)pN<0.001 N0154 (77.8)243 (51.3) N+44 (22.2)231 (48.7)Histological type0.004 LUAD155 (78.3)309 (65.2) LUSC16 (8.1)62 (13.1) Others27 (13.6)103 (21.7)Grade0.56 I7 (3.5)16 (3.4) II70 (35.4)145 (30.6) III68 (34.3)193 (40.7) IV3 (1.5)10 (2.1) Unknown50 (25.3)110 (23.2)Neoadjuvant systemic therapy<0.001 No182 (91.9)353 (74.5) Yes16 (8.1)121 (25.5)Neoadjuvant radiotherapy<0.001 No172 (86.9)337 (71.1) Yes26 (13.1)137 (28.9)Surgery<0.001 Sublobar resection198 (100.0)0 (0.0) Lobectomy0 (0.0)474 (100.0)LN dissection<0.001 No131 (66.2)34 (7.2) Yes67 (33.8)440 (92.8)Radiotherapy<0.001 No87 (43.9)129 (27.2) Yes111 (56.1)345 (72.8)Metastatic organ<0.001 Bone94 (47.5)125 (26.4) Brain77 (38.9)314 (66.2) Liver27 (13.6)35 (7.4)Surgery for distant metastases<0.001 No174 (87.9)330 (69.6) Yes24 (12.1)144 (30.4)Data are presented as n (%). LN, lymph node; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma. Baseline clinical, tumor, and treatment characteristics of the study cohort Data are presented as n (%). LN, lymph node; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma. DCA of the prognostic nomogram for predicting 3-year OS. DCA, decision curve analysis; OS, overall survival. Impact of primary tumor resection on survival After PSM, patients who received primary site surgery had longer OS and CSS compared with patients who did not. The median OS time was 28 [24–33] months for patients who underwent lobectomy and 13 [12–13] months for patients who underwent sublobar resection (P<0.001, Figure S1A). The median CSS was 32 [27–37] and 14 [13–14] months in the lobectomy and sublobar resection groups, respectively (P<0.001, Figure S1B). Primary tumor lobectomy as independent prognostic factors for survival in oligometastatic NSCLC patients In the multivariate Cox analysis of the patients treated with primary site surgery, lobectomy and systemic therapy before surgery was found to be independently associated with better OS (HR: 0.62, 95% CI: 0.48–0.79, P<0.001, Table 2) and CSS (HR: 0.61, 95% CI: 0.46–0.80, P<0.001, Table 3). Age, sex, race, histologic type, systemic therapy before surgery and metastatic site were all independent factors for survival of oligometastatic NSCLC patients. A sensitivity analysis of patients who underwent LN dissection (n=507) confirmed that lobectomy remained a significant protective factor for OS (HR: 0.63, 95% CI: 0.46–0.86, P=0.004; Table S2). Table 2Univariate and multivariate Cox analyses of OSVariablesUnivariateMultivariateHR (95% CI)P valueHR (95% CI)P valueAge ≥70 years (vs. <70 years)1.62 (1.32–2.00)<0.0011.51 (1.22–1.86)<0.001Sex Female (vs. male)0.57 (0.48–0.69)<0.0010.62 (0.51–0.75)<0.001Race Black (vs. Asian)1.61 (1.01–2.57)0.041.53 (0.96–2.46)0.07 White (vs. Asian)1.39 (0.96–2.00)0.081.55 (1.07–2.25)0.02Site Middle lobe (vs. lower lobe)0.72 (0.47–1.10)0.13 Upper lobe (vs. lower lobe)0.87 (0.71–1.06)0.17Laterality Right (vs. left)0.95 (0.79–1.14)0.61Tumor size >30 mm (vs. ≤30 mm)1.01 (0.84–1.21)0.92pN N+ (vs. N0)0.95 (0.78–1.14)0.56Histological type LUSC (vs. LUAD)1.64 (1.24–2.16)<0.0011.5 (1.12–2.01)0.007 Others (vs. LUAD)1.09 (0.86–1.38)0.491.19 (0.93–1.52)0.17Grade II (vs. I)1.18 (0.70–1.98)0.53 III (vs. I)1.48 (0.89–2.46)0.14 IV (vs. I)1.86 (0.84–4.10)0.12 Unknown (vs. I)1.16 (0.68–1.97)0.59Neoadjuvant systemic therapy Yes (vs. no)0.55 (0.43–0.70)<0.0010.64 (0.49–0.85)0.002Neoadjuvant radiotherapy Yes (vs. no)0.61 (0.48–0.76)<0.0010.94 (0.72–1.24)0.67Surgery Lobectomy (vs. sublobar resection)0.51 (0.42–0.61)<0.0010.62 (0.48–0.79)<0.001LN dissection Yes (vs. no)0.54 (0.44–0.66)<0.0010.83 (0.63–1.08)0.16Radiotherapy Yes (vs. no)0.73 (0.6–0.88)0.0010.95 (0.74–1.21)0.67Metastatic organ Brain (vs. bone)0.65 (0.53–0.79)<0.0010.81 (0.65–1.02)0.07 Liver (vs. bone)0.86 (0.62–1.2)0.370.68 (0.48–0.96)0.03Surgery for distant metastases Yes (vs. no)0.73 (0.59–0.9)0.0040.83 (0.66–1.05)0.12CI, confidence interval; HR, hazard ratio; LN, lymph node; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; OS, overall survival. Univariate and multivariate Cox analyses of OS CI, confidence interval; HR, hazard ratio; LN, lymph node; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; OS, overall survival. Table 3Univariate and multivariate Cox analyses of CSSVariablesUnivariateMultivariateHR (95% CI)P valueHR (95% CI)P valueAge ≥70 years (vs. <70 years)1.47 (1.18–1.84)<0.0011.37 (1.10–1.72)0.006Sex Female (vs. male)0.57 (0.47–0.69)<0.0010.61 (0.50–0.75)<0.001Race Black (vs. Asian)1.35 (0.82–2.23)0.24 White (vs. Asian)1.32 (0.90–1.92)0.16Site Middle lobe (vs. lower lobe)0.69 (0.44–1.09)0.12 Upper lobe (vs. lower lobe)0.87 (0.70–1.07)0.19Laterality Right (vs. left)1.01 (0.83–1.23)0.90Tumor size >30 mm (vs. ≤30 mm)0.98 (0.81–1.19)0.86pN N+ (vs. N0)0.97 (0.80–1.18)0.78Histological type LUSC (vs. LUAD)1.68 (1.25–2.24)<0.0011.61 (1.19–2.19)0.002 Others (vs. LUAD)1.02 (0.79–1.32)0.891.14 (0.87–1.49)0.33Grade II (vs. I)1.17 (0.67–2.03)0.58 III (vs. I)1.46 (0.85–2.52)0.17 IV (vs. I)2.11 (0.94–4.75)0.07 Unknown (vs. I)1.09 (0.62–1.93)0.76Neoadjuvant systemic therapy Yes (vs. no)0.54 (0.41–0.70)<0.0010.63 (0.47–0.85)0.002Neoadjuvant radiotherapy Yes (vs. no)0.64 (0.50–0.81)<0.0010.96 (0.72–1.28)0.79Surgery Lobectomy (vs. sublobar resection)0.50 (0.41–0.62)<0.0010.61 (0.46–0.80)<0.001LN dissection Yes (vs. no)0.53 (0.43–0.66)<0.0010.83 (0.63–1.10)0.19Radiotherapy Yes (vs. no)0.79 (0.64–0.97)0.020.99 (0.76–1.29)0.97Metastatic organ Brain (vs. bone)0.69 (0.56–0.85)<0.0010.85 (0.67–1.08)0.18 Liver (vs. bone)0.82 (0.57–1.18)0.280.66 (0.45–0.97)0.03Surgery for distant metastases Yes (vs. no)0.73 (0.58–0.93)0.0090.83 (0.65–1.07)0.15CI, confidence interval; CSS, cancer-specific survival; HR, hazard ratio; LN, lymph node; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma. Univariate and multivariate Cox analyses of CSS CI, confidence interval; CSS, cancer-specific survival; HR, hazard ratio; LN, lymph node; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma. Development and validation of a prognostic nomogram We constructed a nomogram based on the multivariable Cox regression model (Figure 2). This tool assigns a score to each variable, and the total score corresponds to a predicted probability of 1-, 2-, and 3-year OS. A total of 672 patients were randomly split into training and validation cohorts at a 1:1 ratio. In the training cohort, the nomogram demonstrated good predictive accuracy, with 1-, 2-, and 3-year AUC values of 0.75, 0.74, and 0.74, respectively (Figure 3A). The performance in the validation cohort was also strong, with corresponding AUC values of 0.73, 0.72, and 0.73 (Figure 3B). Calibration curves showed a high level of agreement between the nomogram’s predictions and actual survival outcomes in both the training and validation cohorts (Figure 3C,3D), yielding a predictive tool based on specific regression parameters (see Table S3 for full model coefficients). Nomogram predicting 1-, 2-, and 3-year survival based on extent of primary tumor resection and other clinical parameters. LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma. Validation of the nomogram. ROC curves for 1-, 2-, 3-year OS in the training cohort (A) and the validation cohort (B). The calibration curves of nomogram between predicted and actual 1-, 2- and 3-year OS in the training cohort (C) and the validation cohort (D). AUC, area under the ROC curve; OS, overall survival; ROC, receiver operating characteristic. Discussion Both prospective and retrospective data confirm a better prognosis for oligometastatic lung cancer patients who undergo primary tumor surgery (2-4). This finding was substantiated by our analysis, which showed that both OS and CSS were improved in patients with oligometastatic NSCLC who underwent primary tumor surgery versus those who did not. Subsequent analysis of patients who received primary tumor surgery found that age, sex, race, histologic type, preoperative systemic therapy, extent of surgical resection, and metastatic site were all independent predictors of survival in this oligometastatic NSCLC cohort. Evidence comparing the extent of pulmonary resection in patients with oligometastatic lung cancer is limited. Our research demonstrated that lobectomy correlates with more favorable outcomes. This is consistent with the findings of Hao and colleagues (5), who analyzed metastatic NSCLC patients who underwent primary tumor resection and observed that lobectomy resulted in longer OS compared with sublobar resection. One possible explanation is that a more comprehensive surgical resection serves to cytoreduce resistant tumor cells that cannot be eradicated by systemic therapy, ultimately leading to improved OS. However, conflicting evidence exists. A study by Kaba and colleagues (8) in an oligometastatic cohort with only brain metastases reported that sublobar resection, in fact, led to a better survival outcome than lobectomy. The authors hypothesized that parenchymal-sparing lung resections improve survival by leaving patients with the greater physiological reserve needed to better tolerate systemic therapies. Two broad strategies exist regarding the timing of local consolidation (9). One approach advocates for deferring local therapy until there is evidence of disease progression, thereby maximizing the duration of effective systemic treatment. The alternative strategy promotes early local intervention at the point of maximal response to systemic therapy. Contributing to this discussion, our analysis revealed that preoperative systemic therapy is an independent prognostic factor. We acknowledge several limitations in our study. First, in accordance with previous literature, we defined oligometastatic disease as NSCLC involving a single metastatic organ, regardless of the number of individual lesions. It should be noted that all patients in this study were diagnosed with stage IV disease at their initial presentation, thus representing a cohort of synchronous oligometastatic NSCLC. This operational definition was adopted due to the inherent constraints of the SEER database, which consistently records the site of metastasis but lacks granular, longitudinal data on the exact number of metastatic deposits for the entire study period. And the SEER database lacks information on patient comorbidities, which could lead to selection bias in treatment choices. Second, detailed data on systemic therapies, such as specific regimens or duration, were unavailable. The inherent limitations of the SEER database meant that specific data on surgical margins (R status), molecular profiles, and immunotherapy regimens were unavailable, which should be considered when interpreting our findings. Despite these drawbacks, we believe our approach is justified. Data from single-center studies are often limited by small sample sizes and lack the generalizability required to evaluate this clinical question. Therefore, the SEER database, as a comprehensive, population-based registry, represents an ideal resource for studying patient survival in this setting. Conclusions In conclusion, our study demonstrates that primary tumor resection is associated with a significant survival benefit for patients with oligometastatic NSCLC. Crucially, our analysis reveals that the extent of surgery is a critical prognostic factor, with lobectomy providing superior overall and CSS compared to sublobar resection. This finding, along with the identification of preoperative systemic therapy as another independent predictor of improved outcomes, underscores the value of a comprehensive, multimodal treatment strategy. To further aid clinical decision-making, we developed and validated a novel prognostic nomogram that accurately predicts 1-, 2-, and 3-year survival, offering a personalized risk assessment tool for clinicians. While these results from a large, population-based cohort strongly support lobectomy as the preferred surgical approach, the inherent limitations of a retrospective analysis suggest that prospective trials are ultimately needed to establish the optimal treatment paradigm. Supplementary The article’s supplementary files as Acknowledgments None. Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Footnotes Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2661/rc Funding: This work was supported by the National Natural Science Foundation of China (No. 82503714), the Natural Science Foundation of Zhejiang Province (No. LQN26H160021), the Zhejiang Traditional Chinese Medicine Co-construction Project (No. GZY-ZJ-KJ–23004), the National Key Scientific Program of China (No. 2022YFA1304500), the China Postdoctoral Science Foundation (No. 2024M763330), the Zhejiang Provincial Medical and Health Science and Technology Plan (No. 2025KY731), and the Zhejiang Provincial Traditional Chinese Medicine Science and Technology Project (No. 2026ZL0221). Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2661/coif). The authors have no conflicts of interest to declare. References Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2661/prf Supplementary Materials The article’s supplementary files as