Survival of patients with node-negative lung cancer and chest wall invasion

Introduction

Evidence-based recommendations for the Tumor, Node, Metastasis (TNM) classification of lung cancer are provided by the International Association for the Study of Lung Cancer (IASLC) Staging Project.

New T descriptor definitions were published in 2016 as part of the 8th edition of the staging system (1). Although changes to be implemented in the 9^th TNM edition were recently presented at the World Lung Conference in Singapore in September 2023, the T descriptors for lung cancer in the 9^th edition remain the same as those used for classification in the 8^th edition (2). The T3 classification encompasses tumors with the following characteristics: largest tumor diameter >5 but ≤7 cm, the presence of a satellite nodule (SN) in the same lobe, parietal pleura infiltration (PPI), chest wall invasion (CWI), parietal pericardium infiltration, or phrenic nerve infiltration. Tumors >7 cm in diameter and tumors with diaphragmatic involvement were reclassified as T4 tumors in the 8th edition because of the lower overall survival (OS) rates associated with these tumors. Tumors with 2 or more satellite nodes in different lobes on the same side of the lung and tumors with invasion into mediastinum, heart, large blood vessels near the heart, trachea, or esophagus are also classified as T4 non-small cell lung cancer (NSCLC) (1,3).

Various studies have aimed to validate the revisions that were released in the 8^th edition TNM staging system. One study performed in the Netherlands found differing survival for patients with NSCLC classified by different T3 descriptors, based on the 7^th TNM edition. Patients with SNs had the highest 5-year OS (62.8%), and patients with multiple T3 descriptors had the worst survival (28.7%) (4). An analysis of patients in the National Cancer Database (NCDB) compared OS rates among patients with T3N0 NSCLC defined by different T3 descriptors, based on the 8^th TNM edition, and found that patients with CWI had the lowest 5-year survival (52.2%) of any single-T3 subtype (5). Similarly, a single-institution retrospective review of 280 consecutive patients with pT3N0M0 NSCLC revealed that CWI, incomplete resection, and age >65 years were independently associated with worse OS, and that a patient prognostic group defined by T3N0M0 NSCLC with CWI had poorer 5-year OS as compared with patient prognostic groups defined by other single T3 NSCLC descriptors (6). These findings suggest that CWI confers a different prognostic profile than other T3 descriptors, and that placing T3 tumors with CWI in the T4 category may improve the accuracy of future editions of the TNM classification system. This, in turn, would improve the accuracy of staging and provide a better understanding of overall prognosis. The treatment strategies that are currently offered for T3 with CWI tumors may or may not change for T4 CWI tumors, since the treatment options that are ultimately pursued depend on numerous factors such as overall patient health. However, this reclassification would influence ongoing discussions about treatment options for this subset of patients.

Despite previous work on the implications of CWI and data highlighting poor OS in patients with CWI, it is currently unknown how survival of patients with T3 NSCLC with CWI compares with that of patients with T4 NSCLC. Given the different prognostic profile CWI has compared to that of other T3 descriptors, it is possible that CWI might be better incorporated as a T4 category. Therefore, our objective was to compare outcomes patients with NSCLC and CWI classified as T3N0M0 with patients with tumors classified as T4N0M0 NSCLC. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2039/rc).

Methods

Data source

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. We queried the 2018 participant user file of the NCDB (https://training.seer.cancer.gov/operations/standards/setters/ncdb.html), which includes clinical and pathologic staging utilizing the 7^th and 8^th editions of the American Joint Committee on Cancer (AJCC) TNM staging system. For the purposes of this study, all patients in the dataset were staged according to the 8^th edition TNM staging system.

Cohort selection

Figure 1 details the selection of patients for this study. We queried the NCDB for patients with NSCLC diagnosed from 2010 to 2015. From this population, we identified patients with cT3N0M0 disease with CWI (defined as invasion into the chest wall, brachial plexus, superior sulcus, or parietal pleura) and patients with cT4N0M0 disease (tumor size at least 7 cm). We excluded patients with node-positive disease, patients treated with wedge resections, and those with T4 tumors measuring <7 cm.

Figure 1 Flow diagram of cohort selection. CWI, chest wall invasion; NSCLC, non-small cell lung cancer.

Statistical analysis

Demographics, comorbidities, tumor characteristics, treatment factors, and outcomes were compared between the patients with T3 NSCLC with CWI and patients with T4 NSCLC in the unmatched cohort. Chi-squared tests were used to compare categorical variables, and Wilcoxon rank-sum tests were used to compare continuous variables. Patients were then 1:1 propensity-score matched. Covariates included in the matching algorithm were determined a priori and included age, sex, race, insurance status, income, education, Charlson-Deyo comorbidity score, tumor location, histology, tumor grade, neoadjuvant treatment, stereotactic body radiation therapy (SBRT), surgery type, year of diagnosis, hospital type, and distance of residence from hospital. Demographics, comorbidities, tumor characteristics, treatment factors, and outcomes were then re-examined in the matched cohort. Kaplan-Meier models were used to compare 5-year OS in matched patients. All statistical analyses were completed utilizing Stata Statistical Software, release 13.0 (StataCorp). A P value <0.05 was considered statistically significant.

Results

Analysis of the unmatched cohort

A total of 38,795 patients were included in the initial analysis, of which 3,391 (8.7%) had T3N0M0 disease with CWI and 35,404 (91.3%) had T4N0M0 disease (Table 1). Patients with T3 disease with CWI tended to have more tumors in the right upper lobe (46.7%), while patients with T4 disease tended to have tumors in the right and left upper lobes (27.3% and 24.4%, respectively). Most patients with T3 tumors with CWI had moderately- or poorly differentiated tumors (28.3% and 46.5%, respectively), while the T4 tumors were primarily moderately differentiated (22.8%) (P<0.001). Additionally, significantly more patients in the T3 group received neoadjuvant chemotherapy (14.6% vs. 5.1%) or surgery (64.1% vs. 33.9%) as compared with their T4 counterparts (P<0.001).

Analysis of the matched cohort

After propensity-score matching, the cohort consisted of 1,734 patients: 867 with T3 NSCLC with CWI and 867 with T4 NSCLC. The two groups were not significantly different with regards to socio-demographics, Charlson-Deyo comorbidity score, hospital characteristics, tumor location, histology, treatment, and surgical approach (all P>0.05) (Table 2).

There were significantly more T4 patients with nodal upstaging as compared with the T3 patients with CWI (20.8% vs. 14.8%, P<0.001). The median size of the T4 tumors [8 cm, interquartile range (IQR), 7.1–9 cm] was significantly larger than the median size of the T3 tumors (5.1 cm, IQR, 3.9–6 cm, P<0.001).

Survival analysis in the propensity-matched cohort

There were no significant differences in 30-day (P=0.27) or 90-day (P=0.33) mortality between the propensity-matched groups. Additionally, there was no significant difference in 5-year OS between the patients with T3 NSCLC with CWI [44.0%, 95% confidence interval (CI): 41–47%] and the patients with T4 NSCLC (43.8%, 95% CI: 40–47%; P=0.50) in the propensity-matched groups (Figure 2). Median survival in both groups was 40 months (range 0.2–123 months for the T3 cohort and 0.3–120 months for the T4 cohort).

Figure 2 Kaplan-Meier 5-year OS estimates of patients with T3 and chest wall invasion vs. patients with T4 disease. There were no differences in 5-year OS between the two groups. CI, confidence interval; OS, overall survival.

Discussion

This analysis of patients diagnosed with NSCLC demonstrated that patients with node-negative T3 tumors with CWI have similar 5-year OS as patients with node-negative T4 NSCLC, suggesting that CWI might better serve as a T4 descriptor rather than a T3 descriptor.

Several studies have validated the T stage descriptors that were published in 2016 comprising the 8^th edition TNM lung cancer staging classification, and some have focused specifically on the prognostic significance of CWI as a T3 descriptor (4,5). CWI and pleural invasion are associated with poor prognosis in patients with T3 or T4 tumors, and survival is dependent on several additional key factors including completeness of resection, nodal involvement, and receipt of adjuvant chemotherapy (7-10). In a recent study, published in 2024, Ugalde Figueroa and colleagues analyzed 1,448 patients with pT3N0M0 in the IASLC database (1999–2010). The authors found that T3 tumors with CWI or parietal pleural infiltration had the highest rates of incomplete resection. Additionally, these patients had worse 5-year OS as compared with patients with T3 NSCLC defined by other T3 descriptors, such as size or the presence of a SN, independent of completeness of resection (11).

Importantly, we observed equivalent 30-day and 90-day mortality and 5-year OS in patients with T3N0M0 NSCLC with CWI and patients with T4 NSCLC. Our finding of 5-year OS of approximately 44% in both groups is consistent with the median survival of patients with node-negative NSCLC with CWI reported in other studies, which ranges from 30–70% (6,11-13). Thus, CWI is associated with worse outcomes and worse OS as compared with other T3 descriptors but is associated with similar survival and mortality as T4 NSCLC, supporting the argument to reclassify CWI as a T4 descriptor.

Our propensity-matched analysis demonstrated that patients with T4 disease had higher rates of nodal upstaging as compared with patients with T3 disease and CWI. As expected, the median tumor size in the T4 group (8 cm) was larger than that in the T3 group (5.1 cm). A larger median tumor size in the T4 patients might explain the increased incidence of nodal upstaging in the T4 group, as several studies have demonstrated an increased risk of upstaging with larger tumors (14,15).

There are several limitations to acknowledge. First, this was a retrospective analysis and is therefore prone to confounding or misclassification bias. Second, the accuracy of the data present in the NCDB is dependent on accurate reporting from the participating institutions. Therefore, the data is subject to reporting bias and inaccuracies. Third, the NCDB can have significant amounts of missing information depending on the variable of interest. For instance, information regarding surgical margins or outcomes after adjuvant therapy was not provided. This lack of granularity and the degree of missing data present in such large datasets can make the interpretation and validity of the findings from these datasets challenging. Therefore, because we did not know how this might influence our findings, we based our conclusions on propensity-matched analyses, which mitigate these external limitations. Additionally, the NCDB is one of the largest databases available in the United States, encompassing data from ~1,500 facilities, so our work sets a strong foundation for further investigation. For example, future research can include additional subgroup analyses by tumor histology or surgical approach of CWI vs. other T4 tumors to better understand potential differences in outcomes.

Conclusions

This study provides further evidence that CWI might serve better as a T4 descriptor, rather than as a T3 descriptor, in future iterations of the TNM staging system. The TNM staging system is continually revised as new evidence is obtained, which is paramount for continued progress in the staging of lung cancer and, subsequently, continued improvements in the care of our patients.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2039/rc

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2039/prf

Funding: This work was supported by the John D. Mitchell Thoracic Oncology Fellowship (to A.S. and A.L.D.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2039/coif). A.S. and A.L.D. state that their salaries were funded by the John D. Mitchell Thoracic Oncology Fellowship. C.F.J.Y. is on the advisory board for AstraZeneca and Genentech and also reports an honorarium from AstraZeneca. P.U.F. is on the advisory board for AstraZeneca and Medtronic and also reports honoraria from AstraZeneca, Medtronic, Johnson & Johnson and Roche. The other authors have no conflicts of interest to declare.

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.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Goldstraw P, Chansky K, Crowley J, et al. The IASLC Lung Cancer Staging Project: Proposals for Revision of the TNM Stage Groupings in the Forthcoming (Eighth) Edition of the TNM Classification for Lung Cancer. J Thorac Oncol 2016;11:39-51. [Crossref] [PubMed]
2. Lewis D. WCLC Attendees Hear Preview of Proposed Changes for the 9th Edition of the TNM Staging Classification for Thoracic Cancers. ILCN WCLC News. Published October 10, 2023. Accessed March 21, 2024. Available online: https://www.ilcn.org/wclc-attendees-hear-preview-of-proposed-changes-for-the-9th-edition-of-the-tnm-staging-classification-for-thoracic-cancers/
3. Rami-Porta R, Bolejack V, Crowley J, et al. The IASLC Lung Cancer Staging Project: Proposals for the Revisions of the T Descriptors in the Forthcoming Eighth Edition of the TNM Classification for Lung Cancer. J Thorac Oncol 2015;10:990-1003.
4. Blaauwgeers H, Damhuis R, Lissenberg-Witte BI, et al. A Population-Based Study of Outcomes in Surgically Resected T3N0 Non-Small Cell Lung Cancer in The Netherlands, Defined Using TNM-7 and TNM-8; Justification of Changes and an Argument to Incorporate Histology in the Staging Algorithm. J Thorac Oncol 2019;14:459-67. [Crossref] [PubMed]
5. Peng T, Wightman SC, Ding L, et al. Exploring the Potential Different Outcomes Associated With the Different Phenotypes Under the Shared Pathologic T3N0 Designation. JTO Clin Res Rep 2021;2:100186. [Crossref] [PubMed]
6. Marques E, Kennedy KF, Giroux DJ, et al. Oncologic Outcomes of Patients With Resected T3N0M0 Non-small Cell Lung Cancer. Semin Thorac Cardiovasc Surg 2023;35:796-804. [Crossref] [PubMed]
7. Yabuki H, Sakurada A, Eba S, et al. Chest wall/parietal pleural invasions worsen prognosis in T4 non-small cell lung cancer patients after resection. Gen Thorac Cardiovasc Surg 2019;67:788-93. [Crossref] [PubMed]
8. Yutaka Y, Terada K, Tanaka S, et al. Prognostic impact of pathologically confirmed rib invasion in patients with lung cancer requiring chest wall resection. J Thorac Dis 2022;14:4660-8. [Crossref] [PubMed]
9. Mazzella A, Loi M, Alifano M. Prognostic factors of resected lung cancer with chest wall involvement. Curr Chall Thorac Surg 2020;2:6.
10. Drake JA, Sullivan JL, Weksler B. Adjuvant chemotherapy improves survival in patients with completely resected T3N0 non-small cell lung cancer invading the chest wall. J Thorac Cardiovasc Surg 2018;155:1794-802. [Crossref] [PubMed]
11. Ugalde Figueroa PA, Marques E, Cilento VJ, et al. Completeness of Resection and Long-Term Survival of Patients Undergoing Resection for Pathologic T3 NSCLC: An International Association for the Study of Lung Cancer Analysis. J Thorac Oncol 2024;19:141-52. [Crossref] [PubMed]
12. Kumar N, Malik PS, Bharati SJ, et al. Primary lung cancer with chest wall involvement: Outcomes of a multimodality management approach. Lung India 2021;38:338-42. [Crossref] [PubMed]
13. Facciolo F, Cardillo G, Lopergolo M, et al. Chest wall invasion in non-small cell lung carcinoma: a rationale for en bloc resection. J Thorac Cardiovasc Surg 2001;121:649-56. [Crossref] [PubMed]
14. Bott MJ, Patel AP, Crabtree TD, et al. Pathologic Upstaging in Patients Undergoing Resection for Stage I Non-Small Cell Lung Cancer: Are There Modifiable Predictors? Ann Thorac Surg 2015;100:2048-53. [Crossref] [PubMed]
15. Tsai TM, Liu CY, Lin MW, et al. Factors Associated with Nodal Upstaging in Clinical T1a-bN0M0 Non-Small Cell Lung Cancers. Cancers (Basel) 2022;14:1277. [Crossref] [PubMed]