Abstract
Background
Stage IIIA non-small cell lung cancer (NSCLC) is a heterogeneous disease, and the role of surgery within multimodality treatment remains controversial. With the emergence of immunotherapy, there is a need to re-evaluate surgical outcomes in real-world populations. This study compares overall survival (OS) between surgical and non-surgical approaches for stage IIIA NSCLC from a multicenter cohort.
Methods
We conducted a retrospective cohort study of 598 stage IIIA NSCLC patients diagnosed between January 1, 2010, and June 30, 2021, across 21 Kaiser Permanente Northern California (KPNC) hospitals. Patients were categorized into surgical and non-surgical groups; the surgical group was further divided into surgery alone, surgery first (adjuvant therapy), and surgery later (neoadjuvant or perioperative therapy).
Results
The surgical group had higher 5-year OS (52.6% vs. 30.8%, P<0.001) and longer median survival (5.77 vs. 2.17 years). All surgical subgroups were independently associated with reduced mortality after adjusting for age, sex, body mass index (BMI), smoking status, Charlson Comorbidity Score (CCS), and race: surgery later [adjusted hazard ratio (AHR) =0.50; 95% confidence interval (CI): 0.34–0.74], surgery alone (AHR =0.58; 95% CI: 0.39–0.89), and surgery first (AHR =0.48; 95% CI: 0.37–0.62). Age >80 years (AHR =1.65; 95% CI: 1.07–2.55; P=0.02) and male sex (AHR =1.64; 95% CI: 1.30–2.06; P<0.001) were independently associated with increased mortality.
Conclusions
Surgery, when integrated into multimodal treatment, is associated with significantly improved survival in stage IIIA NSCLC. These findings support continued consideration of surgery in appropriately selected patients and highlight the need for prospective studies to clarify its role in NSCLC management.
Keywords: Stage IIIA non-small cell lung cancer (stage IIIA NSCLC), Kaiser Permanente Northern California (KPNC), surgical resection, survival outcome
Highlight box.
Key findings
• In a retrospective cohort of 598 patients with stage IIIA non-small cell lung cancer (NSCLC) across Kaiser Permanente Northern California, surgical treatment was associated with significantly improved 5-year overall survival compared to non-surgical treatment (52.6% vs. 30.8%; P<0.001).
• Median survival was nearly three times longer in the surgical group than the non-surgical group (5.77 vs. 2.17 years).
• After adjusting for age, sex, body mass index, smoking status, and comorbidity burden, surgery remained independently associated with a 50% reduction in mortality risk (hazard ratio: 0.498; 95% confidence interval: 0.397–0.624).
What is known and what is new?
• The role of surgery in stage IIIA NSCLC remains controversial, and real-world survival data comparing surgical and non-surgical multimodal approaches in large, diverse cohorts are limited.
• This study provides real-world evidence from a large multicenter cohort demonstrating a robust and durable survival benefit associated with surgical resection as part of multimodal treatment, with findings consistent across sensitivity analyses and key subgroups.
What is the implication, and what should change now?
• Surgery should remain a central consideration in multidisciplinary treatment planning for appropriately selected patients with resectable stage IIIA NSCLC.
• Prospective randomized trials are needed to further define surgery’s role in the evolving landscape of perioperative immunotherapy.
Introduction
Background
Lung cancer is the leading cause of cancer death worldwide (1). It is broadly divided into two main types: non-small cell lung cancer (NSCLC), which makes up 85% of cases, and small cell lung cancer, which accounts for the rest (2). Most patients with NSCLC are diagnosed at advanced stages, such as stage III or IV (3). Stage IIIA NSCLC is highly heterogeneous, resulting in variable multimodality treatments across clinicians and institutions (4). Lung cancer staging is defined by the American Joint Committee on Cancer (AJCC) tumor, node and metastasis (TNM) system. In the 9th AJCC edition, stage IIIA includes the following classifications: T1N2bM0; T2–T3N2aM0; T3N1M0; and T4N0-1M0. The new addition in the ninth edition is the subdivision of N2 into single-station (N2a) and multi-station (N2b) ipsilateral/subcarinal lymph node involvement (5).
Rationale and knowledge gap
Because of its heterogeneity, the role of surgery in treating stage IIIA remains highly controversial. While established guidelines such as the National Comprehensive Cancer Network (NCCN) (6) and the American Society of Clinical Oncology (ASCO) (4) recommend a range of multimodality treatment options including surgery, chemoradiation, and neoadjuvant and adjuvant immunotherapies, the strength of these recommendations is generally weak and ultimately rely on the clinical judgment and expertise of the multidisciplinary team.
Objective
This study aims to compare the overall survival (OS) outcomes of stage IIIA NSCLC patients who received non-surgical versus surgical multimodal treatment plans in a real-world diverse population. By analyzing data from a large-volume, multicenter cohort, we seek to provide updated insights into the efficacy of surgical treatment approaches, particularly in the context of immunotherapy. This research will help inform clinical decision-making and potentially refine surgical treatment guidelines for stage IIIA NSCLC, emphasizing the importance of a multidisciplinary approach to optimize patient outcomes. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1715/rc).
Methods
Study population
This is a retrospective cohort study across 21 Kaiser Permanente Northern California (KPNC) hospitals that includes patients diagnosed with stage IIIA NSCLC from January 1, 2010, to June 30, 2021, with follow-up extending to June 30, 2023. Eligible patients were identified from hospital records at Commission on Cancer (CoC) American College of Surgeons (ACS)-accredited institutions. Inclusion criteria were members of KPNC, confirmed diagnosis of stage IIIA NSCLC, treatment initiated at one of the participating institutions, and complete follow-up data available until the study end date or loss to follow-up. Patients were excluded if they had previous lung cancer in the prior 10 years, no KPNC membership in the year prior to cancer diagnosis, incomplete treatment records, concurrent malignancies, or metastatic disease at the time of diagnosis. Patients were categorized as non-surgical (chemotherapy, radiation, immunotherapy, or a combination) and surgical (with or without perioperative therapy). The surgical treatment group received lobectomies, pneumonectomies, or sublobar resections with lymph node dissection and was further subdivided into three groups: surgery alone, surgery first (adjuvant therapy), and surgery later (neoadjuvant and/or perioperative therapies). The choice of treatment was made by a multidisciplinary team based on tumor staging, patient comorbidities, and other relevant clinical factors. Demographic and clinical characteristics such as age, body mass index (BMI), Charlson Comorbidity Score (CCS), race, sex, smoking status, and date of diagnosis were collected and analyzed. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Institutional Review Board of Kaiser Permanente Northern California (KPNC). Informed consent was waived in this retrospective study.
Measurements
The primary outcome of interest was OS, defined as the time from the initiation of stage IIIA NSCLC diagnosis to death from any cause. Patients who were alive at the study end date or lost to follow-up were censored at the time of their last known follow-up. Secondary outcomes included 1-, 3-, and 5-year survival rates. Data on age, BMI, CCS, race, gender, smoking status, and diagnosis date were collected from electronic medical records. These variables were chosen for their potential impact on OS and were included as covariates in the statistical analysis. Data were carefully verified for accuracy, and any discrepancies were resolved by cross-referencing with the original medical records.
Statistical analysis
Descriptive statistics were used to describe the demographic and clinical characteristics of the study participants. Kaplan-Meier survival curves were generated to compare OS between the surgical and non-surgical treatment groups, and to compare survival within each cohort between the pre-PACIFIC (2010–2016) and post-PACIFIC (2017–2021) eras. The log-rank test was used to assess the statistical significance in survival curves. To identify independent predictors of OS, multivariable Cox proportional hazards regression models were constructed, adjusting for age, gender, race, BMI, smoking status, and CCS. Hazard ratios (HRs) with 95% confidence intervals (CIs) were reported for each variable included in the model. All tests were 2-sided, and a P value less than 0.05 was considered statistically significant. Statistical analyses were performed using SAS 9.4 (Cary, NC).
Results
Study population characteristics
A total of 598 patients diagnosed with stage IIIA NSCLC between January 1, 2010 and June 30, 2021, were included in the study. The median diagnosis age was 70 years [interquartile range (IQR), 63–76 years], 64.4% White, 13.0% Asian, and 55.5% female. Two hundred eighty-one patients (47.0%) received non-surgical treatment, consisting of chemoradiotherapy with or without immunotherapy. The remaining 317 patients (53.0%) received surgical treatment, consisting of surgery alone or surgery combined with neoadjuvant, adjuvant, or perioperative therapies (a combination of chemotherapy, radiation, and immunotherapy). Baseline characteristics such as CCS, ethnicity, BMI, and age, were comparable between groups. However, smoking status and sex were statistically significant—the non-surgical group included a higher proportion of male patients and patients with current or former tobacco use. The baseline characteristics of the study population are summarized in Table 1.
Table 1. Baseline demographics of patients with stage IIIA NSCLC, stratified by treatment groups.
| Variables | Total (N=598) | Non-surgical (N=281) | Surgical (N=317) | P value |
|---|---|---|---|---|
| Age (years) | 0.38 | |||
| <60 | 99 (16.6) | 43 (15.3) | 56 (17.7) | |
| 60–69 | 192 (32.1) | 92 (32.7) | 100 (31.5) | |
| 70–79 | 241 (40.3) | 109 (38.8) | 132 (41.6) | |
| ≥80 | 66 (11.0) | 37 (13.2) | 29 (9.1) | |
| Sex | 0.005* | |||
| Female | 332 (55.5) | 139 (49.5) | 193 (60.9) | |
| Male | 266 (44.5) | 142 (50.5) | 124 (39.1) | |
| BMI (kg/m2) | 0.15 | |||
| <25 | 244 (40.8) | 126 (44.8) | 118 (37.2) | |
| 25–<30 | 208 (34.8) | 89 (31.7) | 119 (37.5) | |
| ≥30 | 146 (24.4) | 66 (23.5) | 80 (25.2) | |
| Tobacco use | <0.001* | |||
| Current | 125 (20.9) | 75 (26.7) | 50 (15.8) | |
| Former | 359 (60.0) | 173 (61.6) | 186 (58.7) | |
| Never | 109 (18.2) | 30 (10.7) | 79 (24.9) | |
| Unknown | 5 (0.8) | 3 (1.1) | 2 (0.6) | |
| CCS | 0.14 | |||
| 0 | 135 (22.6) | 56 (19.9) | 79 (24.9) | |
| 1–2 | 280 (46.8) | 129 (45.9) | 151 (47.6) | |
| ≥3 | 183 (30.6) | 96 (34.2) | 87 (27.4) | |
| Race | 0.09 | |||
| Asian | 78 (13.0) | 26 (9.3) | 52 (16.4) | |
| Black | 47 (7.9) | 24 (8.5) | 23 (7.3) | |
| Hispanic | 59 (9.9) | 25 (8.9) | 34 (10.7) | |
| Other | 29 (4.8) | 14 (5.0) | 15 (4.7) | |
| White | 385 (64.4) | 192 (68.3) | 193 (60.9) | |
| Diagnosis age (years) | 0.16 | |||
| Mean ± SD | 68.9±9.3 | 69.6±8.9 | 68.2±9.6 | |
| Min–Max | 18.0–89.0 | 42.0–89.0 | 18.0–89.0 | |
| Median (IQR) | 70.0 (63.0–76.0) | 70.0 (64.0–76.0) | 70.0 (62.0–75.0) |
Data are presented as n (%) unless otherwise specified. *, P<0.05. BMI, body mass index; CCS, Charlson Comorbidity Score; IQR, interquartile range; NSCLC, non-small cell lung cancer; SD, standard deviation.
Among patients in the surgical treatment group, the majority received adjuvant therapy alone (62.1%), most commonly chemotherapy only (35.3%). This was followed by surgery alone (18.0%) and then neoadjuvant therapy alone (12.9%). Treatment combinations not listed were not observed in this cohort (Table 2).
Table 2. Frequency of surgical treatment modalities.
| Modality | N=317 (%) |
|---|---|
| Surgery only | 57 (18.0) |
| Neoadjuvant | 41 (12.9) |
| Chemotherapy | 33 (10.4) |
| Chemoradiation | 6 (1.9) |
| Chemotherapy + immunotherapy | 1 (0.3) |
| Immunotherapy | 1 (0.3) |
| Adjuvant | 197 (62.1) |
| Chemotherapy | 112 (35.3) |
| Chemotherapy + immunotherapy | 8 (2.5) |
| Chemotherapy + immunotherapy + radiation | 8 (2.5) |
| Chemoradiation | 59 (18.6) |
| Immunotherapy | 4 (1.3) |
| Radiation | 6 (1.9) |
| Perioperative | 22 (6.9) |
| Neoadjuvant chemotherapy, adjuvant immunotherapy | 4 (1.3) |
| Neoadjuvant chemotherapy, adjuvant radiation | 16 (5.0) |
| Neoadjuvant chemotherapy, adjuvant immunotherapy + radiation | 2 (0.6) |
Survival outcomes
Kaplan-Meier analysis demonstrated significantly higher OS in patients who underwent surgery at any point during their treatment course compared with those managed non-surgically (5-year OS: 52.6% vs. 30.8%; P<0.001). Median survival after diagnosis was also significantly longer in the surgical group (5.77 vs. 2.17 years; P<0.001) (Table 3). To assess the impact of evolving standards of care after the PACIFIC trial (7), we additionally stratified survival outcomes by diagnosis era (pre-PACIFIC: 2010–2016; post-PACIFIC: 2017–2021) within each treatment cohort. Among non-surgical patients, OS improved significantly in the post-PACIFIC era compared to the pre-PACIFIC era (P=0.03). In contrast, among surgical patients, survival was statistically unchanged between the two eras (P=0.57) (Figure 1).
Table 3. Comparison of survival outcomes between surgical and non-surgical treatment groups.
| Treatment group | Overall survival rate (%) (95% CI) | Median (years) (95% CI) | P value | ||
|---|---|---|---|---|---|
| 1-year | 3-year | 5-year | |||
| Surgical treatment | 89.8 (85.8–92.7) | 69.5 (64.0–74.4) | 52.6 (46.3–58.6) | 5.77 (4.44–7.71) | – |
| Non-surgical treatment | 77.1 (71.7–81.6) | 39.7 (33.8–45.5) | 30.8 (25.1–36.7) | 2.17 (1.68–2.64) | <0.001 |
CI, confidence interval.
Figure 1.
Kaplan-Meier overall survival comparing pre-PACIFIC era (2010–2016) and post-PACIFIC era (2017–2021). (A) Non-surgical cohort (P=0.03). (B) Surgical cohort (P=0.57). Shaded bands represent 95% confidence limits; tick marks denote censored observations.
Multivariable Cox proportional hazards regression analysis was conducted to identify independent predictors of OS, adjusting for age, gender, race, BMI, smoking status, and CCS. We also further stratified the surgical cohort into three treatment categories to assess whether any specific surgical approach conferred a survival advantage: surgery first (adjuvant therapy), surgery alone, and surgery later (neoadjuvant or perioperative therapies). The analysis revealed that being older than or equal to 80 years [adjusted hazard ratio (AHR) =1.65; 95% CI: 1.07–2.55; P=0.02] and being male (AHR =1.64; 95% CI: 1.30–2.06; P<0.001) were independently associated with increased mortality. Accordingly, younger age and female sex were significant demographic predictors of improved survival. Current smoking [unadjusted hazard ratio (UHR) =1.53; 95% CI: 1.06–2.20; P=0.02] and former smoking (UHR =1.62; 95% CI: 1.19–2.21; P=0.002) was initially associated with a higher risk of death, but the statistical significance disappeared after adjustment for age, gender, race, BMI, and CCS. In the adjusted model, no statistically significant associations were observed for BMI or CCS, while a significant association was observed for Hispanic relative to White (AHR =1.45; 95% CI: 1.01–2.07; P=0.04).
When comparing the four treatment groups (non-surgical, surgery later, surgery alone, and surgery first), and using the non-surgical group as the reference, all surgical approaches were associated with a statistically significant reduction of approximately half in the hazard of death. Specifically, surgery later (AHR =0.50; 95% CI: 0.34–0.74; P<0.001), surgery alone (AHR =0.58; 95% CI: 0.39–0.89; P=0.01), and surgery first (AHR =0.48; 95% CI: 0.37–0.62; P<0.001) were each significantly associated with improved survival compared to no surgery (Table 4).
Table 4. Association between treatment and overall survival controlled for patient’s age, gender, race, BMI, smoking status, and CCS.
| Characteristic | UHR | 95% CI | P value | AHR* | 95% CI | P value |
|---|---|---|---|---|---|---|
| Age group (years) | ||||||
| <60 (Ref.) | 1.0 | 1.0 | ||||
| 60–69 | 1.11 | 0.79–1.57 | 0.54 | 1.02 | 0.71–1.47 | 0.91 |
| 70–79 | 1.30 | 0.94–1.80 | 0.11 | 1.26 | 0.88–1.79 | 0.21 |
| ≥80 | 1.77 | 1.19–2.63 | 0.005* | 1.65 | 1.07–2.55 | 0.02* |
| Sex | ||||||
| Female (Ref.) | 1.0 | 1.0 | ||||
| Male | 1.62 | 1.31–2.00 | <0.001* | 1.64 | 1.30–2.06 | <0.001* |
| BMI (kg/m2) | ||||||
| <25 (Ref.) | 1.0 | 1.0 | ||||
| 25–<30 | 1.04 | 0.82–1.33 | 0.74 | 0.98 | 0.76–1.27 | 0.90 |
| ≥30 | 1.14 | 0.87–1.50 | 0.33 | 1.14 | 0.85–1.51 | 0.38 |
| Smoking status | ||||||
| Never (Ref.) | 1.0 | 1.0 | ||||
| Current | 1.53 | 1.06–2.20 | 0.02* | 1.19 | 0.79–1.79 | 0.40 |
| Former | 1.62 | 1.19–2.21 | 0.002* | 1.22 | 0.86–1.73 | 0.27 |
| Unknown | 1.65 | 0.60–4.58 | 0.33 | 2.05 | 0.72–5.86 | 0.18 |
| Race/Ethnicity | ||||||
| White (Ref.) | 1.0 | 1.0 | ||||
| Asian | 0.71 | 0.50–1.01 | 0.058 | 0.87 | 0.58–1.29 | 0.48 |
| Black | 0.95 | 0.62–1.44 | 0.80 | 0.96 | 0.63–1.48 | 0.87 |
| Hispanic | 1.30 | 0.92–1.83 | 0.14 | 1.45 | 1.01–2.07 | 0.04* |
| Other | 1.27 | 0.80–2.00 | 0.31 | 1.36 | 0.86–2.16 | 0.19 |
| CCS | ||||||
| 0 (Ref.) | 1.0 | 1.0 | ||||
| 1–2 | 1.16 | 0.87–1.55 | 0.32 | 1.04 | 0.77–1.41 | 0.81 |
| ≥3 | 1.61 | 1.19–2.18 | 0.002* | 1.29 | 0.93–1.78 | 0.13 |
| Treatment group | ||||||
| Guideline (Ref.) | 1.0 | 1.0 | ||||
| Surgery later | 0.46 | 0.31–0.67 | <0.001* | 0.50 | 0.34–0.74 | <0.001* |
| Surgery alone | 0.54 | 0.36–0.81 | 0.003* | 0.58 | 0.39–0.89 | 0.01* |
| Surgery first | 0.46 | 0.36–0.59 | <0.001* | 0.48 | 0.37–0.62 | <0.001* |
*, P<0.05. AHR, adjusted hazard ratio; BMI, body mass index; CCS, Charlson Comorbidity Score; CI, confidence interval; Ref., reference; UHR, unadjusted hazard ratio.
To further investigate whether any of the three surgical modalities were associated with improved survival, pairwise comparisons among the three surgical treatment groups were conducted using AHRs from the Cox regression model. The results are summarized in Table 5. No statistically significant differences in survival were observed between any of the surgical treatment groups. However, these direct pairwise comparisons are constrained by limited statistical power due to small sample sizes within specific cohorts, particularly the perioperative group (N=22).
Table 5. Hazards ratio pairwise comparison between three surgical treatment groups.
| Treatment group comparison | AHR | 95% Wald confidence limits |
|---|---|---|
| Surgery later vs. surgery alone | 0.850 | 0.499–1.445 |
| Surgery later vs. surgery first | 1.042 | 0.684–1.586 |
| Surgery alone vs. surgery first | 1.226 | 0.788–1.906 |
AHR, adjusted hazard ratio.
Discussion
Key findings
The role of surgery in the treatment of stage IIIA NSCLC remains controversial. The 2017 PACIFIC trial established consolidation durvalumab following definitive chemoradiotherapy as the standard of care for unresectable stage III NSCLC (8), marking a watershed moment that led some to question the necessity of surgery (9). However, as Adachi and colleagues note, PACIFIC trial’s results apply only to unresectable disease and do not guide management in surgical candidates (9). Our era-stratified analysis reflects this distinction: non-surgical patients experienced significant OS improvement in the post-PACIFIC era (2017–2021) compared to the pre-PACIFIC era (2010–2016), likely driven by real-world adoption of consolidation durvalumab, while survival among surgical patients remained stable across the same period. These findings suggest that the benefit of surgical resection is durable and largely independent of evolving systemic therapy paradigms, supporting its continued role in appropriately selected stage IIIA patients even as non-surgical outcomes improve.
Strengths and limitations
Several potential limitations of our study warrant consideration. The retrospective nature of the study introduces inherent limitations, including the potential for selection bias and residual confounding. Although we adjusted for several key covariates, unmeasured factors such as socioeconomic status, access to care, detailed TNM subgroups (particularly invasive N2 disease), pulmonary function tests, patient performance status, biomarker mutation status, and different treatment regimens within the surgical and non-surgical groups may have influenced outcomes. Another potential limitation is the possibility of stage migration, where improvements in diagnostic imaging and staging techniques over the study period may have led to more accurate identification of patients with stage IIIA disease. This could result in a more favorable prognosis for patients diagnosed in the later years of the study. However, the decade-long study duration help mitigate this effect, providing a more comprehensive evaluation of treatment outcomes over time.
Our study period of 2010–2021 also has several implications. First, it predates the practice-changing neoadjuvant and perioperative chemoimmunotherapy trials (CheckMate 816, CheckMate 77T, KEYNOTE-671, and AEGEAN) that have since been established as the preferred treatment regimens for resectable stage IIIA disease (10-13). As a result, the relatively high proportions of upfront surgery and adjuvant chemotherapy in our cohort reflect our institution’s practice patterns of the era rather than the current multimodal standard. Second, molecular biomarker data [epidermal growth factor receptor (EGFR), anaplastic lymphoma kinase (ALK), programmed death ligand 1 (PD-L1)] were unavailable, as testing was largely performed as a send-out during this period, and the biomarkers that were available played a limited role in management decisions at the time.
Moreover, while our primary multi-variable Cox model robustly evaluates surgical pathways against non-surgical care, certain surgical subgroups are relatively small (surgery alone: N=57; surgery later: N=63), and the surgery later group contains a very small subset of perioperative patients (N=22). Consequently, our direct pairwise comparisons lack the statistical power required to detect subtle survival variations between the different surgical approaches. The result should be therefore interpreted with caution as descriptive trends rather than definitive proof of therapeutic equivalence. Despite these limitations, the consistency of our findings after adjusting for multiple variables supports the robustness of the observed survival benefit.
Comparison with similar research
For resectable stage III NSCLC, several recent immunotherapy trials have revitalized interest in surgical approaches, even in patients with N2 disease. The CheckMate trials, 816 and 77T, both evaluated the efficacy of the immunotherapy agent nivolumab. CheckMate 816 (10) demonstrated that neoadjuvant chemoimmunotherapy (nivolumab plus chemotherapy) followed by surgery improved event-free survival (HR for death, 0.72; 95% CI: 0.523–0.998; P=0.048) and resulted in a 5-year OS of 65%, compared to 55% with chemotherapy alone. CheckMate 77T showed that perioperative nivolumab (administered as both neoadjuvant and adjuvant chemoimmunotherapy) further enhanced survival over perioperative chemotherapy alone, with an 18-month event-free survival of 70% in the nivolumab group and 50% in the chemotherapy group (HR for disease progression, abandoned surgery, or death, 0.58; 97.36% CI: 0.42–0.81; P<0.001) (11).
The KEYNOTE-671 (14) and AEGEAN (13) trials evaluated perioperative regimens—consisting of neoadjuvant immunotherapy plus chemotherapy, surgery, and then adjuvant immunotherapy—with different immunotherapy agents (pembrolizumab and durvalumab, respectively). KEYNOTE-671 showed 36-month OS estimates of 71% (95% CI: 66–76%) in the pembrolizumab group and 64% (95% CI: 58–69%) in the placebo group (HR, 0.72; 95% CI: 0.56–0.93; one-sided P=0.005) (14), while AEGEAN showed a 12-month event-free survival of 73.4% (95% CI: 67.9–78.1%) of the durvalumab group as compared with 64.5% (95% CI: 58.8–69.6%). of the placebo group (stratified HR for disease progression, recurrence, or death, 0.68; 95% CI: 0.53–0.88; P=0.004) (13).
Among adjuvant-only clinical trials, the IMpower010 (15) and PEARLS/KEYNOTE-091 (16) trials evaluated adjuvant-only immunotherapy with atezolizumab and pembrolizumab, respectively. IMpower010 demonstrated that after a median follow-up of 32.2 months (IQR, 27.4–38.3 months) in the stage II–IIIA population, atezolizumab treatment improved disease-free survival (HR, 0.79; 95% CI: 0.64–0.96; P=0.02) (15), while PEARLS/KEYNOTE-091 showed a median disease-free survival of 53.6 months (95% CI: 39.2–not reached) in the pembrolizumab group compared with 42.0 months (95% CI: 31.3–not reached) in the placebo group (HR, 0.76; 95% CI: 0.63–0.91; P=0.001) (16). Despite the different immunotherapy agents and timing of treatments, all trials indicated that surgery remains an essential component of managing resectable stage III NSCLC, even in patients with N2 disease.
Explanations of findings
We found strong evidence that a multimodal surgical approach was associated with improved survival in patients with stage IIIA NSCLC compared to non-surgical treatment strategies. Hu et al. showed that among Taiwanese patients, surgery alone showed better survival in T1N2 patients (P<0.001), and that the perioperative approach (neoadjuvant therapy, surgery, and then adjuvant therapy) improved survival in stage IIIA/N2 NSCLC patients (17). For long-term outcomes of stage IIIA-N2 NSCLC patients treated with upfront surgery followed by adjuvant therapy, Fu et al. showed that the 5- and 10-year OS rates were 42.2% and 27.7%, respectively, with better outcomes for patients with R0 resection (P=0.04 for progression-free survival) and N2a (P<0.001 for progression-free survival) (18). Similarly, our study found that all patients with stage IIIA NSCLC, regardless of TNM stage, had a markedly 5-year OS rate (52.6%) compared to the non-surgical group (30.8%), and the median survival after diagnosis was several years longer in the surgical group than the non-surgical group (5.77 vs. 2.17 years; P<0.0001).
While the survival benefit associated with surgery is evident, it is important to acknowledge that patient selection plays a key role in treatment decisions. Suitable candidates for surgery are often younger, have better performance status, and fewer comorbidities, all of which contribute to improved outcomes independent of the surgical intervention itself (19). This is consistent with findings in a recent epidemiological study by Casal-Mouriño et al., who reported higher survival rates among patients who were younger, female, White, had better performance status, adenocarcinoma without driver mutations, and received multidisciplinary-team-based diagnoses (3). Our findings similarly identified that being older than 80 years (AHR =1.65; 95% CI: 1.07–2.55; P=0.02) and being male (AHR =1.64; 95% CI: 1.30–2.06; P<0.001) were independently associated with increased mortality. In our patient population, the highest proportion of surgical cases (56/99=56.5%) was seen in the youngest age group (<60 years), although the difference across age groups was not statistically significant. However, a greater proportion of female patients underwent surgery compared to males (58.1% vs. 46.6%). Thus, to reinforce that the observed survival benefit is likely attributable to surgical treatment itself rather than selection bias, our multivariable analysis adjusted for key confounders such as age, gender, BMI, and comorbidities.
Our study also included additional variables—the surgical modality, specifically surgery alone, surgery later, and surgery first—to determine whether the specific modality of surgery affected survival outcomes. Overall, all surgery treatment groups had better outcomes than the non-surgical group. However, there were no statistically significant differences in outcomes between the surgical groups themselves, indicating that no single surgical approach was superior. Additionally, although current guidelines recommend neoadjuvant chemotherapy and immunotherapy with or without radiation for patients with operable stage IIIA NSCLC (20), our results show that the majority instead received adjuvant chemotherapy (35.3%). This discrepancy highlights the need for further research to determine the optimal treatment regimen in clinical practice.
Implications and actions needed
The findings of this study are most applicable to patients treated at CoC ACS-accredited hospitals, where multidisciplinary teams and high-volume thoracic surgery programs are likely to provide optimal care. However, the generalizability of these results to other healthcare settings, particularly those with limited resources or less specialized surgical expertise, may be limited. Future studies should explore the applicability of these findings in diverse clinical settings to better inform global treatment guidelines for stage IIIA NSCLC.
The significant survival benefit associated with surgical intervention in this study suggests that surgery should still be considered as a viable treatment option for selected patients with stage IIIA NSCLC, particularly when combined with neoadjuvant or adjuvant therapies. These findings support the inclusion of surgery in standard of care guidelines for stage IIIA NSCLC and highlight the importance of a multidisciplinary approach in treatment planning. Thoracic surgeons, pulmonologists, radiologists, and medical and radiation oncologists should collaborate closely to identify patients who may benefit most from a surgical approach, considering individual patient factors and tumor characteristics.
Our study underscores the need for prospective, randomized controlled trials between surgical and non-surgical therapies to further clarify the role of surgery in stage IIIA NSCLC, particularly in the context of emerging therapies such as immunotherapy. Additionally, beyond identifying predictive biomarkers, future prospective studies should leverage emerging translational approaches—including multi-omics integration and spatial transcriptomics—to better characterize the tumor microenvironment and refine selection of patients most likely to benefit from surgical intervention.
Conclusions
Our findings suggest that a multimodal surgical approach may improve survival in patients with stage IIIA NSCLC compared to non-surgical approaches. Despite the inherent limitations associated with a retrospective design, our findings across multiple settings support integrating surgery into treatment plans for appropriately selected patients. Future research evaluating the role of biomarker mutation status is essential to further refine treatment strategies and ensure that all patients with stage IIIA NSCLC receive the most effective and personalized care possible.
Supplementary
The article’s supplementary files as
Acknowledgments
The authors would like to thank members of the Biostatistical Consulting Unit (BSCU) and the Kaiser Permanente Division of Research for their support and feedback on this project.
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. This study was approved by the Institutional Review Board of Kaiser Permanente Northern California (KPNC). Informed consent was waived in this retrospective study.
Footnotes
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1715/rc
Funding: This project was funded by the Kaiser Permanente Division of Research Graduate Medical Education Grant.
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1715/coif). J.B.V. serves as an unpaid editorial board member of Journal of Thoracic Disease from September 2026 to December 2028. The other authors have no conflicts of interest to declare.
Data Sharing Statement
Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1715/dss
References
- 1. Leiter A, Veluswamy RR, Wisnivesky JP. The global burden of lung cancer: current status and future trends. Nat Rev Clin Oncol 2023;20:624-39.. 10.1038/s41571-023-00798-3 [DOI] [PubMed] [Google Scholar]
- 2. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin 2020;70:7-30.. 10.3322/caac.21590 [DOI] [PubMed] [Google Scholar]
- 3. Casal-Mouriño A, Ruano-Ravina A, Lorenzo-González M, et al. Epidemiology of stage III lung cancer: frequency, diagnostic characteristics, and survival. Transl Lung Cancer Res 2021;10:506-18.. 10.21037/tlcr.2020.03.40 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Daly ME, Singh N, Ismaila N, et al. Management of Stage III Non-Small-Cell Lung Cancer: ASCO Guideline. J Clin Oncol 2022;40:1356-84.. 10.1200/JCO.21.02528 [DOI] [PubMed] [Google Scholar]
- 5. Detterbeck FC, Woodard GA, Bader AS, et al. The Proposed Ninth Edition TNM Classification of Lung Cancer. Chest 2024;166:882-95.. 10.1016/j.chest.2024.05.026 [DOI] [PubMed] [Google Scholar]
- 6. Riely GJ, Wood DE, Ettinger DS, et al. Non-Small Cell Lung Cancer, Version 4.2024, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2024;22:249-74.. 10.6004/jnccn.2204.0023 [DOI] [PubMed] [Google Scholar]
- 7. Antonia SJ, Villegas A, Daniel D, et al. Overall Survival with Durvalumab after Chemoradiotherapy in Stage III NSCLC. N Engl J Med 2018;379:2342-50.. 10.1056/NEJMoa1809697 [DOI] [PubMed] [Google Scholar]
- 8. Spigel DR, Faivre-Finn C, Gray JE, et al. Five-Year Survival Outcomes From the PACIFIC Trial: Durvalumab After Chemoradiotherapy in Stage III Non-Small-Cell Lung Cancer. J Clin Oncol 2022;40:1301-11.. 10.1200/JCO.21.01308 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Effect of Surgical Treatment for N2-Positive c-stage III Non-Small Cell Lung Carcinoma in the “PACIFIC” Era - ClinicalKey. Accessed July 27, 2025. Available online: https://www-clinicalkey-com.kaiserpermanente.idm.oclc.org/#!/content/playContent/1-s2.0-S1525730423001444?scrollTo=%23bib0004
- 10. Forde PM, Spicer JD, Provencio M, et al. Overall Survival with Neoadjuvant Nivolumab plus Chemotherapy in Lung Cancer. N Engl J Med 2025;393:741-52.. 10.1056/NEJMoa2502931 [DOI] [PubMed] [Google Scholar]
- 11. Cascone T, Awad MM, Spicer JD, et al. Perioperative Nivolumab in Resectable Lung Cancer. N Engl J Med 2024;390:1756-69.. 10.1056/NEJMoa2311926 [DOI] [PubMed] [Google Scholar]
- 12. Wakelee H, Liberman M, Kato T, et al. Perioperative Pembrolizumab for Early-Stage Non-Small-Cell Lung Cancer. N Engl J Med 2023;389:491-503.. 10.1056/NEJMoa2302983 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Heymach JV, Harpole D, Mitsudomi T, et al. Perioperative Durvalumab for Resectable Non-Small-Cell Lung Cancer. N Engl J Med 2023;389:1672-84.. 10.1056/NEJMoa2304875 [DOI] [PubMed] [Google Scholar]
- 14. Spicer JD, Garassino MC, Wakelee H, et al. Neoadjuvant pembrolizumab plus chemotherapy followed by adjuvant pembrolizumab compared with neoadjuvant chemotherapy alone in patients with early-stage non-small-cell lung cancer (KEYNOTE-671): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet 2024;404:1240-52.. 10.1016/S0140-6736(24)01756-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Felip E, Altorki N, Zhou C, et al. Adjuvant atezolizumab after adjuvant chemotherapy in resected stage IB-IIIA non-small-cell lung cancer (IMpower010): a randomised, multicentre, open-label, phase 3 trial. Lancet 2021;398:1344-57.. 10.1016/S0140-6736(21)02098-5 [DOI] [PubMed] [Google Scholar]
- 16. O'Brien M, Paz-Ares L, Marreaud S, et al. Pembrolizumab versus placebo as adjuvant therapy for completely resected stage IB-IIIA non-small-cell lung cancer (PEARLS/KEYNOTE-091): an interim analysis of a randomised, triple-blind, phase 3 trial. Lancet Oncol 2022;23:1274-86.. 10.1016/S1470-2045(22)00518-6 [DOI] [PubMed] [Google Scholar]
- 17. Hu PC, Huang JY, Cheng YF, et al. Trimodal therapy and surgical approaches in stage IIIA/N2 non-small cell lung cancer. Sci Rep 2024;14:29441.. 10.1038/s41598-024-79158-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Fu F, Sun W, Bai J, et al. Long-Term Outcomes of Selected Patients with IIIA-N2 Non-small Cell Lung Cancer Receiving Upfront Surgical Resection. Ann Surg Oncol 2023;30:8261-70.. 10.1245/s10434-023-14072-4 [DOI] [PubMed] [Google Scholar]
- 19. Bradley JD, Paulus R, Komaki R, et al. Standard-dose versus high-dose conformal radiotherapy with concurrent and consolidation carboplatin plus paclitaxel with or without cetuximab for patients with stage IIIA or IIIB non-small-cell lung cancer (RTOG 0617): a randomised, two-by-two factorial phase 3 study. Lancet Oncol 2015;16:187-99.. 10.1016/S1470-2045(14)71207-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Rice JD, Heidel J, Trivedi JR, et al. Optimal Surgical Timing After Neoadjuvant Therapy for Stage IIIa Non-Small Cell Lung Cancer. Ann Thorac Surg 2020;109:842-7.. 10.1016/j.athoracsur.2019.09.076 [DOI] [PubMed] [Google Scholar]