Neoadjuvant chemoimmunotherapy in resected stage III non-small-cell lung cancer: real-world data from a nationwide registry

Introduction

Outcomes for locally advanced non-small-cell lung cancer (NSCLC) remain suboptimal despite multimodality treatment (1). Preoperative platinum-based chemotherapy (ChT) provides only a modest survival benefit of around 5% (2). In contrast, neoadjuvant or perioperative chemoimmunotherapy (CIT) has consistently increased pathological complete and major pathological response rates and improved event-free survival (EFS) and overall survival (OS) in resected disease within phase III trials (3-5).

However, trial populations, centralised testing and optimised pathways may not fully reflect routine oncological and surgical practice (6,7). Real-world observational studies address the complementary question of effectiveness, helping to assess external validity and implementation-related outcomes such as surgical feasibility, operative complexity, perioperative safety and oncological radicality after CIT (8).

To address this lack of evidence, a study was conducted using real-world data from the multicentre Registry of the Spanish Society of Thoracic Surgery (ReSECT), comparing CIT versus ChT in resected stage III NSCLC, with the aim of assessing oncological effectiveness, surgical complexity and perioperative safety in routine clinical practice. We present this article in accordance with the STROBE reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0388/rc) (9).

Methods

Study design and data source

A retrospective cohort study was designed using prospectively collected data from ReSECT.

ReSECT was created in 2022, approved by the Research Ethics Committee of Aragón (CEICA; PI22/367), and registered at ClinicalTrials.gov (NCT05600569) (10). It is a prospective, voluntary, multicentre registry of anatomical lung resections from 44 thoracic surgery departments in Spain.

Ethical statement

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Research Ethics Committee of Alicante (CEIm; 2025-036) and no informed consent was required, as it involved secondary analyses of pseudonymised data collected within the approved registry framework.

Patients

Consecutive patients with clinical stage III NSCLC who received neoadjuvant ChT or CIT followed by anatomical resection between January 2023 and April 2025 were included. Cases were staged according to the 8th edition of the tumour-node-metastasis (TNM) classification (11) by imaging and minimally invasive or invasive techniques when indicated. Exclusion criteria were N3 nodal involvement (stage IIIC), neoadjuvant regimens other than ChT or CIT, absence of systematic or lobe-specific lymph-node dissection, and incomplete clinical or pathological data. After applying the exclusion criteria, 114 patients from 22 centres remained available for analyses (CIT 77; ChT 37) (Figure 1).

Figure 1 Flow chart of patient selection and matching process. ChT, chemotherapy; CIT, chemoimmunotherapy; NSCLC, non-small cell lung cancer; PSM, propensity-score matching.

Variables and definitions

Primary endpoints were the rate of pathological complete response (pCR), downstaging and complete resection status. pCR was defined as the absence of viable tumour in both the primary tumour and regional lymph nodes (12). Downstaging was defined as any reduction in pathological stage from pretreatment clinical T, N and stage descriptors. Resection status was classified as complete (R0), uncertain [R(un)] or incomplete (R1/2) according to the International Association for the Study of Lung Cancer (IASLC) classification (13-15). Based on registry-available data, R0 required negative margins, absence of extracapsular nodal extension, removal of at least six lymph nodes (≥3 N1 and ≥3 N2), and a negative highest mediastinal lymph node station. In addition, a composite non-incomplete resection endpoint [R0/R(un)] was also analysed.

Secondary endpoints included surgical variables: approach, operative time, conversions, pleural adhesions, fissure completeness, extended resections involving extrapulmonary structures, bronchial or vascular reconstruction and intraoperative mortality. Lymphadenectomy technique (systematic vs. lobe-specific) and nodal yield were recorded as well.

Additionally, postoperative complications according to the Clavien-Dindo classification (CD) (16), reinterventions, intermediate/intensive care admission and length of stay, overall hospital stay, presence of a chest drain at discharge, readmission within 30 days, and in-hospital, 30-day and 90-day mortality were registered.

Statistical analysis

Categorical variables are reported as counts (percentages) and continuous variables as median (interquartile range). Baseline characteristics were summarised descriptively, and balance was assessed using standardised mean differences (SMD), with absolute values <0.10 considered optimal.

The propensity score was estimated by multivariable logistic regression including sex, age, smoking status, comorbidities, malignancy status, histology, invasive staging, and clinical T and N categories. One-to-one nearest-neighbour matching without replacement used a maximum absolute propensity-score difference of 0.05, primarily estimating the average treatment effect in the matched cohort. Common support was assessed by visual inspection of propensity-score distributions before and after matching (Figure 2), and post-matching balance was reassessed using SMD (Table 1).

Figure 2 Mirror histograms of the propensity score. (A) Before matching; (B) after matching. Blue bars represent chemotherapy; green bars represent chemoimmunotherapy.

In the unmatched cohort, categorical outcomes were compared using χ² or Fisher’s exact tests and continuous outcomes using the Mann-Whitney U test; in the matched cohort, McNemar’s test and the Wilcoxon signed-rank test were used, respectively. Effect sizes are reported as relative risks (RRs) with 95% confidence intervals (CIs). Missing data were handled using available-case analysis without multiple imputation. All tests were two-sided, with P<0.05 considered statistically significant. Analyses were performed using IBM SPSS Statistics v22.0.

Results

The study included 114 patients. After 1:1 propensity-score matching (PSM), 68 patients (34 pairs) were analysed, improving measured covariate balance and baseline comparability between groups (Table 1, Figure 2).

Primary outcomes

Before matching, pCR was significantly higher after CIT in both the tumour and nodal stations (Table S1). In the matched cohort, the difference was larger (38% vs. 12%; RR 3.3; 95% CI: 1.2–9.0; P=0.02), remained increased for the primary tumour but missed statistical significance for nodal status (Table 2, Figure 3A).

Figure 3 pCR (A), downstaging (B), and complete resection rate (C) after matching. ChT, chemotherapy; CIT, chemoimmunotherapy; IASLC, International Association for the Study of Lung Cancer; pCR, pathological complete response.

Downstaging was also greater with CIT in the overall cohort, largely driven by T-downstaging (Table S1). After PSM, downstaging remained numerically higher in the CIT group (79% vs. 56%), but the difference did not reach statistical significance (RR 1.4; 95% CI: 1.0–2.0; P=0.057) (Table 2, Figure 3B).

Rates of non-incomplete resection (R0/R(un)) were high and comparable between groups both before (Table S1) and after matching. Nevertheless, R0 was significantly more frequent after CIT in the overall cohort and the matched analysis (74% vs. 41%; RR 1.8, 95% CI: 1.1–3.1; P=0.049) (Table 2, Figure 3C).

Surgical outcomes

After PSM, the minimally invasive approach predominated in both groups (Table 3). Operative time was longer after CIT (median 240 vs. 180 min; P=0.048). There were no statistically significant differences in other surgical variables, but there were clinically higher rates of extended resections and bronchoplasties, a greater number of N2 lymph nodes retrieved, and more frequent systematic lymphadenectomy in the CIT arm. There were no intraoperative deaths. See additional surgical outcomes in Tables S2,S3.

Postoperative outcomes

In the matched cohort, overall postoperative complications were more common after CIT (41% vs. 21%; RR 2.0; 95% CI: 0.9–4.3; P=0.12) as well as major complications (CD ≥ III) (15% vs. 6%; RR 2.5; 95% CI: 0.5–12.0; P=0.38), although differences were not statistically significant. Median length of stay, reinterventions, intensive care and hospital readmissions were similar between groups. One in-hospital death occurred in each group, with no additional 30- or 90-day mortality (Table 3). See additional postoperative outcomes in Table S4.

Discussion

pCR has been suggested as a good surrogate for EFS in the neoadjuvant setting for resected lung cancer, but the strength of correlation for OS still deserves further research (17). Phase II and III trials consistently show that adding immunotherapy to platinum-based ChT significantly increases pCR rates up to around 28–42% (18-20), far exceeding the 5–8% achieved with ChT (2,3). Real-world series have reported similar results (7). Our findings align with and extend this evidence in a multicentre study obtained from a prospective national registry: after PSM, CIT achieved a statistically and clinically significant increase in pCR (38% vs. 12%), consistent with trial data and slightly higher than rates reported most real-world studies (7,21,22). Beyond statistical significance, this represents an absolute difference of 26 percentage points and an approximately 3.3-fold relative increase in pCR. This magnitude of the ratio is clinically very relevant, although it should be interpreted cautiously given the limited matched sample size and the width of the confidence interval. Importantly, pCR was defined as the absence of viable tumour cells in both the primary tumour and regional lymph nodes, capturing double complete responders (23), a dimension frequently underreported in randomized trials and indicative of deeper pathological response.

Unlike pCR, pathological downstaging has been less studied, as trials typically rely on radiological (RECIST) criteria without systematic pathological confirmation (24). Real-world data suggest limited concordance between radiological nodal response and true pathological clearance after neoadjuvant treatment, particularly in mediastinal nodes (25). Nonetheless, real-world studies have reported substantial clinical stage reduction with CIT compared with ChT (approximately 73% vs. 45%) (7). In our series, CIT achieved higher pathological downstaging than ChT after PSM (79% vs. 56%), mainly driven by T-component reduction. Assessment of N-downstaging was limited by incomplete pretreatment invasive staging in a minority of patients, warranting cautious interpretation. Overall, these findings suggest that neoadjuvant immunotherapy can induce true stage migration mainly at the expense of primary tumour response, while nodal response is still controversial due to the lack of agreement on staging and restaging requirements and techniques.

Most trials and real-world series have reported R0 rates above 90% after CIT and significantly higher than with ChT in meta-analyses (3,5,8). However, these definitions are usually based on the absence of micro/macroscopic disease in the margins (R1, R2) and do not fully fulfil IASLC criteria (13-15). Under the broad definition of non-incomplete resection (R0/R(un)), our results (94% vs. 82%) are similar to published data. When IASLC criteria were applied, R0 reached 74% after CIT versus 41% with ChT, corresponding to an absolute difference of 33 percentage points and an approximately 1.8-fold relative increase. This suggests that strict complete resection was about 80% more frequent after CIT and that its benefit may extend beyond tumour shrinkage to more oncologically complete resections. This is supported by the quality of lymphadenectomy in our matched cohort, with ≥3 N1 nodes retrieved in approximately 70–82% of cases and ≥3 N2 nodes in 82–96%, figures rarely reported in phase III CIT trials or the IASLC datasets underpinning the current R0/R(un) redefinition, where detailed nodal information was available in <12% of cases (26). This granular nodal information illustrates how structured surgical registries can complement trial evidence.

Surgery after CIT often entails a more demanding anatomical scenario, with perihilar fibrosis, mediastinal adhesions and challenging dissection planes that may increase technical difficulty (27). In our cohort, CIT was associated with longer operative times and a higher proportion of complex procedures—including extended resections and bronchoplasties—yet rates of minimally invasive approaches were high when compared to clinical trials (28). Conversion, pneumonectomy and major complications rates remained comparable to ChT. Any in-hospital complication was numerically more frequent with CIT, mainly due to persistent air leak and discharge with a chest drain, and three patients required reoperation. This increase in postoperative air leak might be related to higher rate of fibrosis and difficult dissection secondary to tumour response. A trend towards a higher rate of major complications that didn’t reach statistical significance may be highlighting the more challenging nature of surgery following CIT. Nevertheless, there were no intraoperative deaths and in-hospital, 30- and 90-day mortality were low and similar between strategies (one per group). These findings are consistent with contemporary meta-analyses and series reporting postoperative-complication rates of 22–40% after CIT and very low early mortality (29,30).

Limitations

First, despite PSM, its observational design cannot exclude selection bias and residual confounding. Matching improved measured covariate balance, although some residual imbalance remained after PSM. Matched analyses were also restricted to patients with sufficient overlap, reducing the effective sample size. Second, centre-level clustering could not be modelled because centre-specific identifiers were not available in the analyzable dataset to preserve pseudonymisation and registry governance; therefore, unmeasured centre-level effects cannot be excluded. Third, the IASLC-defined R0 analysis was sensitive to missing data because it requires complete surgical and pathological information, including nodal-yield variables. Although registry data were rechecked and additional evaluable matched pairs were recovered, some residual risk of information bias cannot be excluded. Finally, follow-up was not mature enough to assess EFS and OS; therefore, conclusions are restricted to pathological, surgical and perioperative outcomes.

Clinical and research implications

For selected patients with resectable stage III NSCLC, neoadjuvant CIT appears to be a clinically relevant strategy when multidisciplinary assessment anticipates a reasonable likelihood of complete resection (31). In this cohort, it was associated with higher pCR, greater downstaging and improved IASLC-defined R0 rates, although surgery appeared more demanding and these findings should be interpreted in light of the observational design and limited matched sample size. Implementation should be accompanied by: (I) proactive planning (selective mediastinal re-evaluation, setup for bronchovascular reconstruction); (II) standardisation of lymphadenectomy and pathology reporting; and (III) surgical quality monitoring. Research priorities include larger cohorts with mature follow-up (EFS/OS), validation of response predictors (clinical, imaging, biomarkers), and prospective designs confirming safety and effectiveness in more complex scenarios (multi-station N2, T4).

Conclusions

In this multicentre real-world cohort, neoadjuvant CIT for resected stage III NSCLC achieved greater pCR, downstaging and complete resection rates than ChT alone. Operations were longer and overall morbidity slightly higher, but major complications and early mortality remained comparable between strategies.

Acknowledgments

The authors thank all participating centres for their contribution to the database, the Aragonese Institute of Health Sciences (IACS) as the registry’s data management hub, and the Alicante Institute for Health and Biomedical Research (ISABIAL) for methodological and statistical assistance.

Preliminary results of this study were presented at the World Conference on Lung Cancer (IASLC 2025, Barcelona, Spain), the National Congress of the Spanish Society of Thoracic Surgery (SECT 2026, Cádiz, Spain), and the Annual Meeting of the European Society of Thoracic Surgeons (ESTS 2026, Athens, Greece).

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0388/rc

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0388/prf

Funding: This study was supported by the Spanish Society of Thoracic Surgery (SECT) and its Foundation as the promoter of the ReSECT registry. Open access publication costs were supported by the Alicante Institute for Health and Biomedical Research (ISABIAL). No other specific funding was received for the conduct of this study.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0388/coif). C.G. reports receiving minor royalties from Springer, consulting fees from BMS, honoraria from BMS, Firma Rivolution, Johnson & Johnson and Meril; and meeting support from Medilevel and Johnson & Johnson. U.J. reports lecture fees from BMS and serves as the unpaid Vice President of SECT (Spanish Society of Thoracic Surgery). 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. The study was approved by the Research Ethics Committee of Alicante (CEIm; 2025-036) and no informed consent was required, as it involved secondary analyses of pseudonymised data collected within the approved registry framework.

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. Rami-Porta R, Nishimura KK, Giroux DJ, et al. The International Association for the Study of Lung Cancer Lung Cancer Staging Project: Proposals for Revision of the TNM Stage Groups in the Forthcoming (Ninth) Edition of the TNM Classification for Lung Cancer. J Thorac Oncol 2024;19:1007-27. [Crossref] [PubMed]
2. NSCLC Meta-analysis Collaborative Group. Preoperative chemotherapy for non-small-cell lung cancer: a systematic review and meta-analysis of individual participant data. Lancet 2014;383:1561-71. [Crossref] [PubMed]
3. Sorin M, Prosty C, Ghaleb L, et al. Neoadjuvant Chemoimmunotherapy for NSCLC: A Systematic Review and Meta-Analysis. JAMA Oncol 2024;10:621-33. [Crossref] [PubMed]
4. Marinelli D, Nuccio A, Di Federico A, et al. Improved Event-Free Survival After Complete or Major Pathologic Response in Patients With Resectable NSCLC Treated With Neoadjuvant Chemoimmunotherapy Regardless of Adjuvant Treatment: A Systematic Review and Individual Patient Data Meta-Analysis. J Thorac Oncol 2025;20:285-95. [Crossref] [PubMed]
5. 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. [Crossref] [PubMed]
6. Hawkins A, Martin LW. Operative Challenges After Induction Immunotherapy for Lung Cancer. Thorac Surg Clin 2025;35:285-98. [Crossref] [PubMed]
7. Bertoglio P, Gallina FT, Balzani E, et al. Surgical outcomes after neoadjuvant chemoimmunotherapy for resectable NSCLC: a systematic review and meta-analysis. Eur J Surg Oncol 2026;52:111805. [Crossref] [PubMed]
8. Saesen R, Van Hemelrijck M, Bogaerts J, et al. Defining the role of real-world data in cancer clinical research: The position of the European Organisation for Research and Treatment of Cancer. Eur J Cancer 2023;186:52-61. [Crossref] [PubMed]
9. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008;61:344-9. [Crossref] [PubMed]
10. Registry of the Spanish Society of Thoracic Surgery. Summary protocol. Available online: https://cdn.clinicaltrials.gov/large-docs/69/NCT05600569/Prot_SAP_ICF_000.pdf
11. 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]
12. Hellmann MD, Chaft JE, William WN Jr, et al. Pathological response after neoadjuvant chemotherapy in resectable non-small-cell lung cancers: proposal for the use of major pathological response as a surrogate endpoint. Lancet Oncol 2014;15:e42-50. [Crossref] [PubMed]
13. Rami-Porta R, Wittekind C, Goldstraw P, et al. Complete resection in lung cancer surgery: proposed definition. Lung Cancer 2005;49:25-33. [Crossref] [PubMed]
14. Detterbeck FC, Ostrowski M, Hoffmann H, et al. The International Association for the Study of Lung Cancer Lung Cancer Staging Project: Proposals for Revision of the Classification of Residual Tumor After Resection for the Forthcoming (Ninth) Edition of the TNM Classification of Lung Cancer. J Thorac Oncol 2024;19:1052-72. [Crossref] [PubMed]
15. Asamura H, Yotsukura M, Rami-Porta R, et al. Updates on the Version 9 American Joint Committee on Cancer Staging System for Lung Cancer. Ann Surg Oncol 2025;32:4569-71. [Crossref] [PubMed]
16. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 2009;250:187-96. [Crossref] [PubMed]
17. Hines JB, Cameron RB, Esposito A, et al. Evaluation of Major Pathologic Response and Pathologic Complete Response as Surrogate End Points for Survival in Randomized Controlled Trials of Neoadjuvant Immune Checkpoint Blockade in Resectable in NSCLC. J Thorac Oncol 2024;19:1108-16. [Crossref] [PubMed]
18. Cao C, Le A, Bott M, et al. Meta-Analysis of Neoadjuvant Immunotherapy for Patients with Resectable Non-Small Cell Lung Cancer. Curr Oncol 2021;28:4686-701. [Crossref] [PubMed]
19. Provencio M, Nadal E, González-Larriba JL, et al. Perioperative Nivolumab and Chemotherapy in Stage III Non-Small-Cell Lung Cancer. N Engl J Med 2023;389:504-13. [Crossref] [PubMed]
20. Nuccio A, Viscardi G, Salomone F, et al. Systematic review and meta-analysis of immune checkpoint inhibitors as single agent or in combination with chemotherapy in early-stage non-small cell lung cancer: Impact of clinicopathological factors and indirect comparison between treatment strategies. Eur J Cancer 2023;195:113404. [Crossref] [PubMed]
21. Huang X, Pang G, Mao Z, et al. Clinical outcomes associated with neoadjuvant therapy for the treatment of resectable non-small cell lung cancer in real-world practice. Clin Respir J 2024;18:e13761. [Crossref] [PubMed]
22. Xu Y, Ma D, Qin Y, et al. Prognostic significance of pathological response and lymph node status in neoadjuvant immunotherapy for potentially resectable non-small cell lung cancer. Ann Med 2025;57:2453825. [Crossref] [PubMed]
23. Huang S, Wu J, Li S, et al. Evaluation of combined pathological responses in primary tumor and lymph nodes following neoadjuvant chemoimmunotherapy in non-small cell lung cancer. Lung Cancer 2023;186:107401. [Crossref] [PubMed]
24. Schwartz LH, Litière S, de Vries E, et al. RECIST 1.1-Update and clarification: From the RECIST committee. Eur J Cancer 2016;62:132-7. [Crossref] [PubMed]
25. Han T, Cheng S, Wang X, et al. Prognostic Significance, Radiological, and Metabolic Characteristics of Metastatic Lymph Nodes in Resectable Non-Small Cell Lung Cancer Following Neoadjuvant Chemoimmunotherapy. Thorac Cancer 2025;16:e70073. [Crossref] [PubMed]
26. Edwards JG, Chansky K, Van Schil P, et al. The IASLC Lung Cancer Staging Project: Analysis of Resection Margin Status and Proposals for Residual Tumor Descriptors for Non-Small Cell Lung Cancer. J Thorac Oncol 2020;15:344-59. [Crossref] [PubMed]
27. Trabalza Marinucci B, Mancini M, Siciliani A, et al. Surgical Techniques for Non-Small-Cell Lung Cancer After Neoadjuvant Chemo-Immunotherapy: State of Art and Review of the Literature. Cancers (Basel) 2025;17:638. [Crossref] [PubMed]
28. Bardoni C, Chiari M, Bertolaccini L, et al. Surgical Outcomes After Neoadjuvant Chemo-Immunotherapy for Stage III NSCLC: A Systematic Review and Meta-Analysis. Cancers (Basel) 2025;17:1426. [Crossref] [PubMed]
29. Kim SS, Cooke DT, Kidane B, et al. The Society of Thoracic Surgeons Expert Consensus on the Multidisciplinary Management and Resectability of Locally Advanced Non-small Cell Lung Cancer. Ann Thorac Surg 2025;119:16-33. [Crossref] [PubMed]
30. Li Y, Hu X, Zhang R, et al. Analysis of treatment-related adverse events and wound complications of surgical resection after neoadjuvant chemoimmunotherapy for non-small cell lung cancer. Int Wound J 2024;21:e14831. [Crossref] [PubMed]
31. Houda I, Bahce I, Dickhoff C, et al. An international and multidisciplinary EORTC survey on resectability of stage III non-small cell lung cancer. Lung Cancer 2025;199:108061. [Crossref] [PubMed]