Study of non-surgical therapeutic patterns and prognosis in elderly patients with stage II non-small-cell lung cancer: an investigational study based on the SEER database

Introduction

Lung cancer is the most common cause of cancer-related death (1). Approximately 654,620 people have a history of lung cancer, and by the year 2022, there are an estimated 236,740 new cases in the United States. Around 80% to 85% of lung cancers are classified as non-small-cell lung cancer (NSCLC) (2,3). With the growing use of low-dose computed tomography (CT) for lung cancer screenings, early-stage NSCLC is more often detected (4). Sufficient evidence has shown (5,6) that over the last few decades, the typical lung cancer patient is an older patient, with 60–70% of people diagnosed with lung cancer at 65 years and over, and about 30–40% at 70 years and above.

Surgery remains the first-line treatment option for early-stage NSCLC, but surgery is not suitable for all patients, especially older patients, where surgery is not possible or tolerated. The most common reasons for surgical ineligibility include patient refusal and medical ineligibility due to complications such as reduced pulmonary reserve (7). Stereotactic body radiation therapy (SBRT) remains the first non-surgical option for these cases (stage I). However, stage II lung cancer has a variety of therapeutic options (8). As an example, SBRT may not always be the best option for a patient when chemoradiotherapy (CRT) remains a curative option. Above all, there is no consensus on the best treatment for elderly patients, and there has been limited research on patients aged 65 years and older. Because of the clinical heterogeneity and lack of high-quality evidence informing treatment choice for older patients, we applied propensity score matching (PSM) methods to compare the effectiveness of CRT vs. RT alone among patients aged 65 years and older with stage II NSCLC. We present this article in accordance with the STROBE reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-639/rc).

Methods

Study population

We used the Surveillance, Epidemiology, and End Results (SEER) database at the National Cancer Institute (http://seer.cancer.gov/data/options.html) for the study and analyzed the data in a retrospective fashion under the auspices of the Exemption Agency Review Committee. The study population comprised 2,646 patients with stage IIA and IIB NSCLC who received CRT or radiotherapy (RT) from 2010 to 2017. Specifically, inclusion criteria was as follows: (I) histologic classification by coding of International Classification of Diseases for Oncology, 3rd edition [ICD-O-3; e.g., 8012, 8046, 8070, 8071, 8072, 8140, 8250, 8255, 8260, 8480, 8481, 8490, 8570, 8550, and 8560 (9)]; (II) stage IIA or IIB based on the 7th edition of the AJCC; (III) age ≥65 years at diagnosis; (IV) diagnosis between 2010 and 2017; and (V) patients were not treated surgically and were treated with CRT or RT, only.

Exclusion criteria were: (I) without recorded survival months; and (II) having multiple primary tumors. Ultimately, patients were separated into CRT or RT categories.

Demographic and clinical variables concerning analysis and PSM consisted of age, sex, race, marital status, year of diagnosis, tumor grade (I/II, III/IV, unknown), histology [adenocarcinoma (AC), squamous cell carcinoma (SCC)], laterality (left/right), primary tumor site (main bronchus, upper/middle/lower lobes, other), tumor stage (T stage; T1–T3), and node stage (N stage; N0–N1) (9).

The survival outcomes included vital status, survival months, and cancer-specific death. Overall survival (OS) was defined as time from treatment to last follow-up or death by any cause. Cancer-specific survival (CSS) was defined as time from treatment to last follow-up or death by lung cancer. We did not have data to assess comorbidities, e.g., Charlson Comorbidity Index, in the SEER database; this is considered a limitation for the study.

The SEER database is a public-use database, and this study did not use individual patient identifiers; therefore, it would not require any additional ethical approvals.

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Statistical methods

Frequencies and percentages were used to report categorical variables, while medians with interquartile ranges were used to report continuous variables. The Chi-squared test was used to compare differences in categorical variables between groups.

A 1:1 PSM analysis was then completed to mitigate baseline biases between CRT and RT cohorts. PSM was completed by applying a logistic regression model and the nearest-neighbor matching algorithm (without replacement). A caliper width of 0.05 was applied. The PSM analysis considered variables of interest, including age, sex, race, marital status, year of diagnosis, tumor grade, histology, laterality, primary site, T stage, and N stage.

Kaplan-Meier (KM) survival curves and log-rank test were used to evaluate differences in OS and CSS between treatment groups pre- and post-matching.

Univariate Cox proportional hazards regression analysis was used to determine predictors of OS and CSS. Multivariate analyses were then completed with the variables that had P values <0.05 from the univariate analysis. Predictor variables included age, sex, race, marital status, tumor stage, tumor grade, histology, and treatment modality.

The final multivariate Cox model was created using a stepwise process (forward selection and backward elimination) based on the Akaike Information Criterion (AIC). For continuous variables, if needed, the continuous variable was transformed (i.e., log transformation) before analyses to meet proportional hazards assumptions.

Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated for each association in the study. All statistical analyses were completed using R software (version 2022; R Core Team, Vienna, Austria; https://www.R-project.org). A two-sided P value <0.05 was considered statistically significant.

Results

Patient characteristics

Using different treatment modalities, 2,646 eligible patients were selected from the SEER database and then divided into two groups: CRT group (n=1,307) and RT group (n=1,339). In the unmatched cohorts for age, sex, marital status, grade, histology, N stage, primary site, survival months, and vital status, there were meaningful differences between the CRT and RT groups. There were no statistically significant differences between the CRT and RT groups for race, year of diagnosis, laterality, T stage, or CSS.

As a result of SEER data limitations, comorbidity burden (e.g., Charlson Comorbidity Index) could not able to be adjusted for in either PSM or regression models, which may have affected the survival outcomes, especially because of the older population.

After doing a 1:1 PSM with 11 variables, we maintained a total of 1,746 patients. After matching the baseline characteristics between the two groups, vital status and survival months were compared. However, there were no statistically significant differences between the CRT and RT groups for their age, sex, race, marital status, grade, histology, laterality, T stage, N stage, primary site, and CSS. Table 1 includes all clinical characteristics before and after PSM.

Survival analysis in OS and CSS

As of December 31, 2017, the study’s follow-up period had a median length of 17 months and a range of 0 to 119 months. OS was superior in the CRT group compared to the RT group (P<0.001) (Figure 1A). The CRT group had 1,061 deaths, a median OS of 20 months (95% CI: 19–21), and calculated OS rates of 3 and 5 years of 29.3% and 16.8%, respectively. In addition, the RT group had 1,142 deaths and a median OS of 16 months (95% CI: 15–17), with matching mortality rates of 23.4% and 11.2%. (Figure 1B). The estimated CSS 3- and 5-year rates for the CRT group were 35.3% and 25.0%, respectively, while in the RT group, the equivalent frequency was 33.3% and 22.4%, with a median duration of 20 months (95% CI: 18–21). The outcomes of the Cox analysis are shown in Table 2. In the multivariate Cox analysis, the treatment pattern served as a standalone predictive factor for OS and CSS. Different therapeutic approaches were independent predictive variables for OS and CSS, based on multivariate Cox analysis. In comparison to the RT group, the OS (HR =0.8, 95% CI: 0.73–0.87) and CSS (HR =0.89, 95% CI: 0.8–0.98) CRT groups performed better.

Figure 1 KM survival curves before PSM. (A) OS for patients treated with CRT vs. RT before PSM. (B) CSS for patients treated with CRT vs. RT before PSM. CRT, chemoradiotherapy; CSS, cancer-specific survival; KM, Kaplan-Meier; OS, overall survival; PSM, propensity score matching; RT, radiotherapy.

It should be noted that the difference in CSS was statistically significant but numerically modest, suggesting that competing non-cancer-related mortality likely contributed to overall differences in OS.

Survival analysis in OS and CSS after PSM

KM curves of patients in different treatment groups were obtained by 1:1 matching (Figure 2). After PSM, the OS of CRT patients was still better than RT patients (P<0.001) (Figure 2A). In the RT group, the OS rates of 3- and 5-year were 23.9% and 11.6%, respectively, with a median OS of 16 months (95% CI: 15–17); in the CRT group, these rates were 29.7% and 17.0%, respectively. Also, the CRT group performed better than the RT group when it comes to CSS (P=0.04) (Figure 2B). In the CRT group, the estimated CSS rates were 35.8% and 25.0%, respectively, with a median CSS of 23 months (95% CI: 21–26), while the corresponding rates were 33.1% and 23.0%, and the median CSS was 20 months (95% CI: 18–21) in the RT group. Table 3 shows the Cox analysis results of OS and CSS for the two groups after PSM. Patients who received CRT had better OS (HR =0.79, 95% CI: 0.71–0.88) and CSS (HR =0.87, 95% CI: 0.77–0.97). Covariates associated with OS were age, sex, race, year of diagnosis, histology, and therapy. Variables such as age, sex, year of diagnosis, histology, and therapy were associated with CSS.

Figure 2 KM survival curves after PSM. (A) OS for CRT vs. RT groups after PSM. (B) CSS for CRT vs. RT groups after PSM. CRT, chemoradiotherapy; CSS, cancer-specific survival; KM, Kaplan-Meier; OS, overall survival; PSM, propensity score matching; RT, radiotherapy.

Because of limitations in SEER, data on gross tumor volume (GTV) were not available and, therefore, could not be adjusted for. This limits how we interpret differences in possible tumor burden across the groups.

Further, SEER does not make it clear whether CRT regimens are sequential or concurrent. As such, we did not stratify outcomes by treatment sequencing, and thus we would interpret our findings with caution. We also could not separately exclude patients treated with SBRT due to limited RT technique coding with the SEER database. This may skew the outcomes for the RT group modestly but highlights the variability in treatment modes in real-world settings.

Discussion

As per National Comprehensive Cancer Network (NCCN) guidelines, elderly patients are defined as any patient 65 years and older (10). Older patients are clinically heterogeneous and frequently have multiple comorbidities, which most likely contribute to higher perioperative mortality (11). Age itself is an independent risk factor for postoperative complications and cancer-related mortality in patients with NSCLC (7). In patients aged ≥65 years who are not surgical candidates, there is no clear best option between CRT and RT.

In this study, we analyzed data from 2,646 patients aged 65 years and older with stage II NSCLC from the SEER database. After conducting PSM to reduce baseline differences between groups, there were 1,746 matched cases available for the survival analysis. Our results demonstrated that CRT resulted in significantly improved OS and numerically improved CSS vs. RT alone. However, it should be noted that while the difference in OS between groups was statistically and clinically relevant, the difference in CSS was modest, suggesting that non-cancer-related mortality due to age-related comorbidities likely played a significant role in the OS outcome.

RT remains the most available and commonly utilized treatment for early-stage NSCLC in patients who are not medically operable or who decline surgery. SBRT provides an excellent local control method for stage I NSCLC (12), and may not be as indicated in stage II patients, particularly if there are positive ipsilateral nodes (13). For those patients, CRT may be a more suitable non-surgical alternative.

Dudani et al. (14) completed a retrospective review that demonstrated significantly improved OS in stage II NSCLC patients who underwent CRT vs. RT (39.1 vs. 20.5 months; P=0.0019). Of note, they reported on a limited number of patients (n=158) and did not evaluate the naturally occurring elderly population of patients. Other trials have also suggested significantly improved efficacy with concurrent CRT vs. sequential treatment (15,16); we could not differentiate concurrent vs. sequential CRT in this analysis due to limitations in the SEER database, which limits our conclusions on treatment sequencing efficacy in our data.

Sekine et al. (17) conducted a retrospective population-based registry study in Japan, and noted that medically treated NSCLC cases comprised only 7.2% of stage I–II and that radiation alone reported a 25% lung CSS. Similarly, Wisnivesky et al. (18) analyzed survival outcomes from a SEER-based sample of patients receiving non-surgical intervention and found that the presence of comorbidity had a substantial impact on OS. In our study, SEER provided no comorbidity data, for example, the Charlson Comorbidity Index; thus, during our analysis, we were unable to control for baseline patient health. This is a key confounding variable, particularly in the elderly population, where treatment choice and treatment outcomes are often influenced by pre-existing health conditions.

Recent studies reported a 5-year survival after CRT of 16% to 20% (19,20), similar to our results. This study adds to the world literature and strengthens real-world evidence of treatment outcomes for elderly stage II NSCLC patients through PSM and multivariate Cox modelling, as well as KM survival analysis.

The current analysis suggests CRT may offer improved survival than RT alone, particularly in patients with good performance status and comorbidity burden. Nevertheless, clinicians must be cautious in interpreting these results due to no GTV data, designated sequenced treatment, and not measuring the comorbidity burden. There is a need to interpret treatment decisions with an individualized approach in this patient population.

We recommend that prospective studies develop measures of performance status and comorbidity indices to better stratify patients and compare CRT vs. RT and their impacts separately. Next, treatment protocols must clearly distinguish between concurrent and sequential CRT protocols to better inform clinical recommendations. Lastly, long-term association with acute and chronic toxicity and outcomes related to quality-of-life following CRT exposure in elderly patients will aid clinicians in evaluating the balance of treatment benefit vs. potential harm.

Limitations

Our study has a number of limitations. First, we were unable to adjust for the comorbidity burden due to the absence of this data in SEER. Comorbidity burden could have played a role in treatment selection and subsequent survival. Second, we were only able to look at CRT and did not differentiate between concurrent or sequential CRT. Therefore, we cannot investigate the impact of treatment timing. Third, GTV and radiation technique (e.g., SBRT) were not coded in SEER, and therefore, we were unable to eliminate the SBRT cases, nor adjust for tumor burden. Finally, while RT-only patients were older and likely had more comorbidities, these factors were not controlled and could bias our results in favor of CRT.

Nevertheless, these limitations reflect real-world treatment data and serve to illustrate the need for additional, further refined, prospective research.

Conclusions

In summary, there is still not enough high-quality prospective evidence available to dictate the most appropriate non-surgical treatment options for elderly patients with stage II NSCLC. Our study, performing a retrospective analysis utilizing a large cohort from the SEER database showed that CRT was associated with better OS mileage and a small improvement in CSS compared to RT as standalone treatment in patients of 65 or more years old. Although our findings were confounded by limitations of this retrospective analysis, and because we could not control for other variables such as comorbidities, tumor burden, treatment sequences, and other cofounders, we would advocate some caution in interpreting the results. More prospective, randomized clinical trials will be needed for verification of survival benefit, and to identify patient sub-groups to more appropriately apply this treatment modality.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-639/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-639/prf

Funding: This study was funded by the Shandong Province Medical and Health Technology Development Plan Project (No. 2018WS119).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-639/coif). The 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. This 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. Wu F, Wang L, Zhou C. Lung cancer in China: current and prospect. Curr Opin Oncol 2021;33:40-6. [Crossref] [PubMed]
2. Miller KD, Nogueira L, Devasia T, et al. Cancer treatment and survivorship statistics, 2022. CA Cancer J Clin 2022;72:409-36. [Crossref] [PubMed]
3. Wu SR, Lin CH, Shih HP, et al. HAI-2 as a novel inhibitor of plasmin represses lung cancer cell invasion and metastasis. Br J Cancer 2019;120:499-511. [Crossref] [PubMed]
4. Hida T, Hata A, Lu J, et al. Interstitial lung abnormalities in patients with stage I non-small cell lung cancer are associated with shorter overall survival: the Boston lung cancer study. Cancer Imaging 2021;21:14. [Crossref] [PubMed]
5. Sacco PC, Maione P, Palazzolo G, et al. Treatment of advanced non-small cell lung cancer in the elderly. Expert Rev Respir Med 2018;12:783-92. [Crossref] [PubMed]
6. Shinde A, Vazquez J, Novak J, et al. The role of comprehensive geriatric assessment in radiation oncology. J Geriatr Oncol 2020;11:194-6. [Crossref] [PubMed]
7. Bei Y, Murakami N, Nakayama Y, et al. Stereotactic body radiation therapy for early-stage non-small-cell lung cancer in octogenarians and older: an alternative treatment. J Radiat Res 2020;61:586-93. [Crossref] [PubMed]
8. Moore S, Leung B, Wu J, et al. Population-based analysis of curative therapies in stage II non-small cell lung cancer: the role of radiotherapy in medically inoperable patients. Radiat Oncol 2020;15:23. [Crossref] [PubMed]
9. Hansen RN, Zhang Y, Seal B, et al. Long-term survival trends in patients with unresectable stage III non-small cell lung cancer receiving chemotherapy and radiation therapy: a SEER cancer registry analysis. BMC Cancer 2020;20:276. [Crossref] [PubMed]
10. Dotan E, Walter LC, Browner IS, et al. NCCN Guidelines® Insights: Older Adult Oncology, Version 1.2021. J Natl Compr Canc Netw 2021;19:1006-19. [Crossref] [PubMed]
11. Hori M, Matsuda T, Shibata A, et al. Cancer incidence and incidence rates in Japan in 2009: a study of 32 population-based cancer registries for the Monitoring of Cancer Incidence in Japan (MCIJ) project. Jpn J Clin Oncol 2015;45:884-91. [Crossref] [PubMed]
12. Temming S, Kocher M, Stoelben E, et al. Risk-adapted robotic stereotactic body radiation therapy for inoperable early-stage non-small-cell lung cancer. Strahlenther Onkol 2018;194:91-7. [Crossref] [PubMed]
13. Rodrigues G, Choy H, Bradley J, et al. Adjuvant radiation therapy in locally advanced non-small cell lung cancer: Executive summary of an American Society for Radiation Oncology (ASTRO) evidence-based clinical practice guideline. Pract Radiat Oncol 2015;5:149-55. [Crossref] [PubMed]
14. Dudani S, Zhu X, Yokom DW, et al. Radical Treatment of Stage II Non-small-cell Lung Cancer With Nonsurgical Approaches: A Multi-institution Report of Outcomes. Clin Lung Cancer 2018;19:e11-8. [Crossref] [PubMed]
15. Dudani S, Leighl NB, Ho C, et al. Approach to the non-operative management of patients with stage II non-small cell lung cancer (NSCLC): A survey of Canadian medical and radiation oncologists. Lung Cancer 2016;94:74-80. [Crossref] [PubMed]
16. Logtenberg ME, Tomlow B, Kastelijn EA, et al. Benefit of Concurrent Versus Sequential Chemoradiotherapy in Elderly Patients With Stage III Non-Small Cell Lung Cancer. Clin Lung Cancer 2022;23:116-21. [Crossref] [PubMed]
17. Sekine I, Shintani Y, Shukuya T, et al. A Japanese lung cancer registry study on demographics and treatment modalities in medically treated patients. Cancer Sci 2020;111:1685-91. [Crossref] [PubMed]
18. Wisnivesky JP, Bonomi M, Henschke C, et al. Radiation therapy for the treatment of unresected stage I-II non-small cell lung cancer. Chest 2005;128:1461-7. [Crossref] [PubMed]
19. Yan SX, Qureshi MM, Suzuki K, et al. Definitive treatment patterns and survival in stage II non-small cell lung cancer. Lung Cancer 2018;124:135-42. [Crossref] [PubMed]
20. Sampath S, Hall M, Schultheiss TE. Definitive chemotherapy and radiotherapy in patients with stage II non-small cell lung cancer: A population-based outcomes study. Lung Cancer 2015;90:61-4. [Crossref] [PubMed]