To the Editor: The management of early-stage lung adenocarcinoma (LUAD) is fundamentally surgical, while systemic therapy has evolved from chemotherapy to targeted and immunotherapeutic approaches.1 This shift reflects not only therapeutic advances but also a deeper understanding of the disease biology. Neoadjuvant immunotherapy aims to achieve long-term clinical benefit by inducing systemic anti-tumor immune responses. Currently, immune checkpoint inhibitors targeting the programmed death-1 (PD-1)/programmed death-ligand 1 (PD-L1) pathway play a central role. By blocking PD-1/PD-L1 interaction, these agents reactivate suppressed T cells, restore their cytotoxic function, and promote cytokine-mediated tumor elimination.2 The neoadjuvant chemoimmunotherapy (NCI) is designed to harness synergistic effects: chemotherapy induces immunogenic cell death and enhances tumor antigen presentation, thereby facilitating the effector functions of immunotherapy-activated T cells.2 Neoadjuvant targeted therapy (NTT), administered before surgery, aims to reduce tumor burden and eradicate micrometastatic disease, thereby improving surgical outcomes and long-term survival.3 Currently, NCI or NTT is increasingly becoming a first-line strategy for locally advanced LUAD. Studies comparing perioperative outcomes in patients with LUAD treated with NTT or NCI are limited. This study aimed to analyze the perioperative outcomes in clinical stage II–III LUAD patients undergoing NCI or NTT. We retrospectively collected the data from patients who underwent curative surgery after NTT or NCI at the National Cancer Center of China between January 2017 and May 2024. This study was approved by the Medical Ethics Committee of Cancer Hospital, Chinese Academy of Medical Sciences (22/492–3694), and informed consent for this retrospective analysis was waived. The detailed information on the study design is presented in Supplementary Methods, https://links.lww.com/CM9/C789. In this study, 700 patients with NSCLC undergoing neoadjuvant immunotherapy or targeted therapy were identified, with 265 LUAD patients (164 received NTT and 101 received NCI) meeting the inclusion criteria for analysis Supplementary Figure 1, https://links.lww.com/CM9/C789. To balance the baseline difference, a 2:1 nearest propensity score matching (PSM) was conducted between the NTT group and the NCI group. Ultimately, a well-balanced cohort was established, comprising 106 patients in the NTT group and 74 patients in the NCI group. The clinical and pathological characteristics of the included patients before and after PSM are summarized Supplementary Tables 1 and 2, https://links.lww.com/CM9/C789. Compared with the NTT group, the NCI group had significantly better pathological responses before PSM, with 5.5% (9/165) of patients in the NTT group achieving major pathological response (MPR) and 16.8% (17/101) in the NCI group achieving MPR. Furthermore, compared to the NTT group, more frequent T-stage downstaging was observed in the NCI group. Although the pathological responses after PSM showed no statistically significant difference between the two groups, the MPR in the NCI group (17.6%, 13/74) remained higher than that in the NTT group (7.5%, 8/106). Regarding radiological responses, the NTT group had a higher proportion of patients achieving partial response (PR)/completed response (CR) than the NCI group, both before and after PSM Supplementary Figure 2 and Supplementary Table 3, https://links.lww.com/CM9/C789. Furthermore, most patients (79.2%, 210/265) underwent uniportal video-assisted thoracoscopic surgery (81.1% 133/164 in the NTT group and 76.2% 77/101 in the NCI group). The conversion rate for patients receiving NTT was significantly higher than that for those receiving NCI before PSM (P = 0.041), while it became comparable between the two groups after PSM (P = 0.097). No significant differences were observed in the postoperative complications between the NTT and NCI groups, either before or after PSM Supplementary Table 4, https://links.lww.com/CM9/C789. Logistic regression analysis was conducted to assess the potential risk factors for postoperative complications. Multivariate logistic regression analysis revealed that neoadjuvant therapy method (odds ratio OR, 0.31; 95% confidence interval CI, 0.07–1.47; P = 0.141] and the surgical approach (OR, 0.81; 95% CI, 0.17–3.82; P = 0.785) were not risk factors for postoperative complications Supplementary Table 5, https://links.lww.com/CM9/C789. Univariate logistic regression analysis revealed that vascular invasion (OR, 5.32; 95% CI, 1.24–22.74; P = 0.024), nerve invasion (OR, 4.62; 95% CI, 1.09–19.62; P = 0.038), and tumor thrombosis (OR, 4.68; 95% CI, 1.21–18.01; P = 0.025) were risk factors for conversion to thoracotomy. Supplementary Table 6, https://links.lww.com/CM9/C789. Given that first- and third-generation NTT are currently the most potent tyrosine kinase inhibitors (TKI) available, a subgroup analysis was conducted to compare the perioperative outcomes between first- or third-generation targeted therapies and neoadjuvant immunotherapy Supplementary Table 7, https://links.lww.com/CM9/C789. After PSM, the pathological responses showed no statistically significant difference between the first-/third-generation NTT and NCI. Regarding radiological responses, the first-generation NTT group exhibited a higher proportion of patients achieving PR/CR in comparison to the NCI group (P <0.001). Similarly, a greater proportion of patients in the third-generation NTT group achieved PR/CR compared to the NCI group (P = 0.005) Supplementary Figure 3, https://links.lww.com/CM9/C789. A subgroup analysis was conducted to compare the perioperative outcomes of neoadjuvant single-target therapy or targeted combined with chemotherapy (TC) versus NCI Supplementary Table 8, https://links.lww.com/CM9/C789. After PSM, the results showed that there were no significant differences observed in pathological response (P = 0.074) between the single-target therapy group and the NCI group. The single-target therapy group had a higher rate of conversion from thoracoscopy to thoracotomy (P = 0.020), whereas the NCI group had a greater T-stage downstaging (P = 0.044). Similarly, the results also indicated that the TC group had a superior radiological response rate than the NCI group (P <0.001) Supplementary Figure 4, https://links.lww.com/CM9/C789. Similarly, no significant differences were observed in the postoperative complications in the subgroups. In our cohort, 31 patients with EGFR mutation received NCI treatment. We further compared the perioperative outcomes of patients with EGFR mutations undergoing NCI treatment with those undergoing NTT treatment. The results indicate that patients receiving NCI had higher rates of MPR/pCR (P = 0.025) than those receiving NTT for EGFR mutations. However, the NTT exhibited a higher rate of PR/CR (P = 0.027) in radiological response. The NCI group exhibited a higher incidence of T-stage down-staging (64.5% vs. 35.3%, P = 0.003) and N-stage down-staging (61.3% vs. 33.1%, P = 0.003). There were no significant differences in the postoperative complications between the two groups Supplementary Table 9, https://links.lww.com/CM9/C789. As key indicators for neoadjuvant interventions, patients diagnosed with pathological stage III LUAD (85 patients 81.0% diagnosed with stage IIIA NSCLC and 20 patients 19.0% with stage IIIB NSCLC) exhibited no significant surgical and pathological outcomes after the neoadjuvant treatments, no matter NCI or NTT Supplementary Table 10, https://links.lww.com/CM9/C789. Our findings indicated that in patients diagnosed with LUAD, the pathological responses after PSM showed no statistically significant difference between the NCI and NTT groups. Conversely, the NTT group exhibited a higher radiological PR rate. A subgroup analysis revealed that the single-target therapy group had a higher conversion rate to thoracotomy due to increased surgical difficulty than the NCI group; however, no significant differences were noted in postoperative complications. Mathey-Andrews et al4 conducted a study involving 4229 patients in a neoadjuvant cohort to compare the outcomes of neoadjuvant immunotherapy and chemotherapy. They found that neoadjuvant immunotherapy did not increase the rate of open lobectomy (60.8% vs. 51.6%, P = 0.11) or result in worse perioperative outcomes. However, a study by Bott et al5 involving 20 patients who underwent surgical resection after nivolumab chemotherapy reported that conversion to thoracotomy occurred in 25% of stage I, 50% of stage IIA, and 71% of stage IIB or IIIA NSCLC patients. Our study found that the pathological responses showed no statistically significant difference between the first-/third-generation NTT and NCI. Previous studies on neoadjuvant treatment with first-generation NTT reported low rates of pCR, with no instances of pCR observed in phase II neoadjuvant trials of erlotinib.6,7 However, with the introduction of third-generation NTT in neoadjuvant therapy, phase II studies have demonstrated that osimertinib achieved a higher overall response rate of 71.1%, surpassing the results of first-generation EGFR-TKIs in neoadjuvant studies.8 Similarly, the results also showed that there were no significant differences observed in pathological response among the single-target therapy group, or TC group and the NCI group. Therefore, monotherapy or combination therapy with third-generation TKIs may be a promising option for neoadjuvant treatment. Notably, within the EGFR-mutation subgroup, NCI was associated with significantly higher MPR/pCR than NTT. This finding suggested that for a subset of EGFR-mutant LUAD, chemotherapy-containing regimens may be particularly effective in achieving deep pathological remission. The underlying mechanism likely involves the cytotoxic chemotherapy component of NCI, which induced extensive tumor cell death and fibrosis, contrasting with the predominantly cytostatic effect of TKIs that may not fully eradicate tumor cells despite radiological shrinkage.9–11 This study had several limitations. First, its retrospective design introduces inherent limitations such as potential selection bias and incomplete data collection. Some data exhibited random missing owing to incomplete retrospective medical record documentation, which may have introduced a certain degree of selection bias. Secondly, although we applied PSM to control for known confounding factors, unknown or unmeasured confounding factors may still exist and cannot be completely ruled out. Third, since this study was a retrospective analysis of previous clinical diagnoses and treatment data, pathologists could not completely separate clinical background information (such as records of treatment history in imaging reports and endoscopic reports) when evaluating specimens. Therefore, there are inherent difficulties in practice in achieving complete “blinding” in the true sense. In conclusion, this study provides a comprehensive analysis of the perioperative outcomes of patients undergoing NCI and NTT for clinical stage II–III LUAD, offering guidance for clinicians in assessing surgical feasibility and treatment options. These findings contribute to the ongoing discussion of the efficacy of neoadjuvant therapies and underscore the importance of considering patient-specific factors when making treatment decisions. Funding This work was supported by the National Key R&D Program of China (No. 2021YFC2500900), National Natural Science Foundation of China (No. 82273129), Central Health Research Key Projects (No. 2022ZD17), CAMS Innovation Fund for Medical Sciences (CIFMS) (Nos. 2021-I2M-1-015, 2024-I2M-C&T-C-008, and 2024-I2M-ZH-005), and National High Level Hospital Clinical Research Funding (No. 2025-LYZX-C-A02). Acknowledgments The authors thank the National Cancer Center, Cancer Hospital, Chinese Academy of Medical Sciences (CHCAMS) for providing the research environment. We appreciate the statistical assistance provided by Zhicheng Du, Department of Medical Statistics, School of Public Health, Sun Yat-sen University, Guangzhou, China. Conflicts of interest None.
Wu et al. (Tue,) studied this question.