Perioperative anxiety and depression predict long-term survival after neoadjuvant therapy in non-small cell lung cancer: a multicenter analysis

Introduction

Neoadjuvant therapy has become a key component of multidisciplinary management for non-small cell lung cancer (NSCLC), improving resectability and survival in selected patients (1,2). However, protracted treatment courses, repeated hospitalizations, and treatment-related symptoms (pain, fatigue, sleep disturbance, dyspnea) impose a substantial psychological burden during the perioperative period (3-6). Anxiety and depression are common yet not timely valued in thoracic oncology and have been associated with impaired adherence, delayed recovery, and adverse outcomes (7,8). While numerous studies have reported associations between baseline psychological distress and survival (9-11), the specific context of neoadjuvant-treated NSCLC and perioperative trajectories of anxiety and depression remains insufficiently characterized.

The Hospital Anxiety and Depression Scale (HADS) is widely used to screen psychological distress in cancer care and has shown acceptable performance in lung cancer populations across clinical settings (12-14). Nevertheless, optimal thresholds for clinically meaningful risk remain debated and may differ from conventional diagnostic cut-offs in surgical cohorts, where even subthreshold symptoms can be prognostically relevant (15,16). Critically, the prognostic value of perioperative trajectories of anxiety and depression—captured at baseline, immediately preoperatively, and during early postoperative recovery—has not been well quantified in neoadjuvant-treated NSCLC, and it is unclear whether these measures add information beyond pathological staging.

To address this gap, we conducted a multicenter retrospective study of patients with NSCLC who received neoadjuvant therapy followed by curative-intent resection. HADS-Anxiety (HADS-A) and HADS-Depression (HADS-D) were assessed at baseline, preoperatively, and three months postoperatively; data-driven thresholds were derived; and associations with overall survival (OS) were evaluated using time-to-event methods. We further integrated psychological indices with post-neoadjuvant pathologic tumor lymph nodes metastasis (ypTNM) to develop and internally validate a pragmatic nomogram, hypothesizing that perioperative anxiety—particularly elevations at baseline and early postoperative follow-up—would independently predict worse survival and improve risk discrimination beyond staging alone. We present this article in accordance with the TRIPOD reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0006/rc).

Methods

Study participants

We conducted a multicenter retrospective cohort study of consecutive patients with NSCLC who underwent surgery after neoadjuvant therapy. A total of 252 patients were screened; 167 from three tertiary centers met the eligibility criteria and were included in the analysis (Figure 1). Eligibility required a pathologic diagnosis of NSCLC, receipt of neoadjuvant systemic therapy followed by curative-intent resection, and availability of prespecified clinical and psychological variables. Reasons for exclusion (e.g., not meeting oncologic/surgical criteria or missing key assessments) are detailed in the flow diagram (Figure 1).

Figure 1 Outcomes of participants. AUC, area under the curve; C-index, concordance index; HADS, Hospital Anxiety and Depression Scale; NSCLC, non-small cell lung cancer; ROC, receiver operating characteristic.

Participant outcomes at follow-up are summarized in Figure 1, with survival defined from the date of surgery to death from any cause or last contact (OS definition described in the Follow-up and Statistics sections). These outcomes form the analytic population for subsequent time-to-event models. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of Fujian Medical University Union Hospital (No. 2025KY667). Informed consent was waived in this retrospective study. All participating centers were informed and agreed to the study.

Tumor remission grade

Postoperative pathological tumor regression was assessed by dedicated thoracic pathologists who were blinded to the patients’ psychological evaluation results. The evaluation followed standardized histopathologic criteria based on the proportion of residual viable tumor cells in the primary lesion and sampled lymph nodes after neoadjuvant therapy (17).

The degree of remission was classified as follows:

• Pathologic complete response (pCR): no residual viable tumor cells in either the resected primary lesion or regional lymph nodes.
• Major pathologic response (MPR): ≤10% residual viable tumor cells.
• Partial response: >10% but <90% viable tumor cells.
• No response/stable disease: ≥90% viable tumor cells.

These categories were used for descriptive analyses and subgroup comparisons of treatment efficacy and psychological characteristics. The ypTNM stage was determined according to the 8th edition of the TNM Classification for Lung Cancer and used in subsequent survival analyses.

Data collection

Clinical data

Clinicopathologic and perioperative data were abstracted from institutional records according to a prespecified template. Collected variables included age at diagnosis, sex, smoking history, tumor location, pathological type, neoadjuvant regimen (chemotherapy vs chemo-immunotherapy), surgical approach, pathologic stage, tumor remission grade, body mass index (BMI), postoperative complications (Clavien-Dindo grade), treatment costs, length of postoperative hospitalization, time to first chest-tube removal, and postoperative drainage volume. For subsequent analyses, BMI was dichotomized at 24 kg/m² based on a clinical threshold. Treatment costs, postoperative length of stay, time to first chest-tube removal, and cumulative postoperative drainage volume were converted to binary variables using receiver-operating characteristic (ROC) analysis with Youden’s index to determine optimal cut-offs.

HADS

The HADS was administered to evaluate patients’ psychological status at three prespecified timepoints: baseline (before neoadjuvant therapy), preoperatively (within one week before surgery), and three months postoperatively. The scale comprises 14 items, with 7 assessing anxiety (HADS-A) and 7 assessing depression (HADS-D). Each item is rated on a 4-point Likert scale (0–3), yielding subscale scores ranging from 0–21. Higher scores indicate greater symptom severity.

The Chinese-validated version of HADS was used for all assessments (18-20). Patients self-completed the questionnaires in a quiet clinical setting, under the supervision of trained oncology nurses to ensure comprehension and completeness. Missing items were handled according to standardized HADS scoring guidelines.

For primary analyses, established literature thresholds of ≥8 points were initially used to classify patients as having clinically significant anxiety or depression (21,22). In addition, ROC analysis was performed within this cohort to determine data-driven cut-offs specific to perioperative NSCLC patients, using Youden’s index to optimize sensitivity and specificity. These refined thresholds were subsequently applied to survival and subgroup analyses.

Follow-up

Postoperative surveillance included chest CT, abdominal ultrasonography, and cranial magnetic resonance imaging (MRI) every 3-month during the first postoperative year, followed by every 6–12 months thereafter. Annual follow-up was conducted via outpatient visits or telephone contact. For patients initially unreachable, telephone interviews were used to ascertain late postoperative outcomes. When death was confirmed by telephone but the exact date was unavailable, the date of the last outpatient visit was recorded as the final follow-up date. OS was defined as the interval from surgery to death from any cause or last follow-up.

Statistical analysis

Continuous variables were summarized as mean ± standard deviation (if approximately normally distributed) and compared using independent t-tests; categorical variables were expressed as counts (percentages) and compared using χ² or Fisher’s exact tests. All tests were two-sided, with statistical significance set at P<0.05. Analyses were performed in SPSS v21 and R v4.2.3.

ROC analysis with Youden’s index was used to derive data-driven cut-offs for HADS-A and HADS-D at each timepoint. Patients were then dichotomized into low- and high-risk groups using these thresholds. OS distributions were estimated by the Kaplan-Meier method and compared with log-rank tests. Univariable and multivariable Cox proportional-hazards models were fitted to evaluate associations between candidate predictors and OS; variables with univariable P<0.05 entered the multivariable model. Hazard ratios (HRs) with 95% confidence intervals (CIs) are reported.

For model performance and clinical utility, we quantified discrimination using time-dependent ROC curves and the concordance index (C-index), and assessed calibration with bootstrap internal validation (800 resamples). Decision-curve analysis (DCA) was used to estimate net clinical benefit across clinically relevant threshold probabilities.

Results

Clinical characteristics

Of 252 screened neoadjuvant-treated NSCLC patients, 167 were analyzed (155 men, 92.8%) without severe respiratory diseases or other clinical symptoms. Histology: squamous 121 (72.5%), adenocarcinoma 43 (25.7%), other 3 (1.8%). c-TNM: I 25 (15.0%), II 45 (27.0%), III 93 (55.7%), IV 4 (2.4%). Neoadjuvant regimens: chemo-immunotherapy 93 (55.7%) vs. chemotherapy 74 (44.3%). Surgery: video-assisted thoracoscopic surgery (VATS) 138 (82.6%) and thoracotomy 29 (17.4%). Pathology after resection: pCR 44 (26.3%), MPR 39 (23.4%), grade IIA 84 (50.3%). Post-neoadjuvant pathologic tumor lymph nodes metastasis ypTNM: I 82 (49.1%), II 44 (26.3%), III 36 (21.6%), IV 5 (3.0%). Mean BMI 23.10±2.81 kg/m²; mean neutrophil-to-lymphocyte ratio 2.61±2.36. Details are in Table 1 and Figures S1,S2.

Follow-up

During follow-up, 45 patients experienced tumor recurrence and 27 died; all deaths were cancer-related. The estimated 1-, 3-, and 5-year OS rates were 95.81% (160/167), 86.23% (144/167), and 83.83% (140/167), respectively. The mean follow-up was 38.65 months (range, 6–60 months), and no patients were lost to follow-up. OS was calculated from the date of surgery to death from any cause or last contact. These data provide a stable outcome framework for subsequent survival and prognostic analyses.

HADS scores and group classification

HADS-A and HADS-D were assessed at three key timepoints: baseline, preoperatively, and 3-month postoperatively. Compared with preoperative values, both HADS-A and HADS-D increased significantly at 3-month after surgery (HADS-A: P=0.005; HADS-D: P=0.02), suggesting persistent psychological distress during early postoperative recovery (Figure 2A,2B). In contrast, no significant differences were observed between baseline and preoperative measurements (HADS-A: P=0.10; HADS-D: P=0.77), as illustrated in Figure S3.

Figure 2 Changes in HADS scores and grouping over time. HADS-A and HADS-D scores 3-months postoperatively (A) and preoperatively (B). ROC curves of HADS-A (C) and HADS-D (D) scores at the three specified time points based on OS. Numbers and proportions of patients in the two risk groups according to ROC HADS-anxiety and depression scores at the three specified time points (E,F). AUC, area under the curve; HADS, Hospital Anxiety and Depression Scale; HADS-A, HADS-Anxiety; HADS-D, HADS-Depression; OS, overall survival; ROC, receiver operating characteristic.

ROC analyses were used to identify optimal thresholds for high-risk classification at each timepoint. For HADS-A, the best cut-off values were 5.5 (baseline), 4.5 (preoperative), and 7.5 (3 months postoperative) [all P<0.001; area under the curve (AUC) =0.70, 0.74, and 0.84, respectively; Figure 2C]. For HADS-D, the optimal thresholds were 3.5, 5.5, and 7.5 (all P<0.001; AUC =0.70, 0.72, and 0.82; Figure 2D).

Using these data-driven cut-offs, all patients were stratified into low- and high-risk groups at each timepoint. The corresponding distributions and proportions are displayed in Figure 2E,2F. These risk categories were subsequently employed for downstream survival analyses and multivariable modeling.

Physical and psychological predictors of survival after neoadjuvant therapy

On univariable Cox analysis, both higher pathological stage and elevated anxiety/depression scores at all perioperative time points were significantly associated with worse OS. Notably, ypTNM stage III–IV (stage III: HR =3.73, 95% CI: 1.54–9.02, P=0.003; stage IV: HR =4.92, 95% CI: 1.30–18.61, P=0.02), HADS-A (baseline HR =13.71, 95% CI: 5.47–34.36; preoperative HR =4.89, 95% CI: 2.12–11.27; 3-month postoperative HR =10.10, 95% CI: 4.49–22.73; all P<0.001), and HADS-D (baseline HR =16.74, 95% CI: 5.23–53.59; preoperative HR =18.47, 95% CI: 5.82–58.57; 3-month postoperative HR =7.84, 95% CI: 2.84–21.65; all P<0.001) emerged as significant predictors.

In multivariable models, ypTNM stage III (HR =5.72, 95% CI: 2.05–15.97, P<0.001), baseline HADS-A >5.5 (HR =4.59, 95% CI: 1.01–20.77, P=0.048), and 3-month postoperative HADS-A >7.5 (HR =7.50, 95% CI: 2.85–19.71, P<0.001) independently predicted reduced OS. Preoperative HADS-A lost significance (P=0.16) and all HADS-D associations attenuated to non-significance (P>0.90). These findings are detailed in Figure 3 and Table S1.

Figure 3 Results of multivariable Cox proportional-hazards model of factors related to anxiety. A, Hospital Anxiety and Depression Scale-Anxiety; CI, confidence interval; D, Hospital Anxiety and Depression Scale-Depression; HR, hazard ratio; ypTNM, post-neoadjuvant pathologic tumor lymph nodes metastasis.

Physical and psychological predictors of survival after neoadjuvant therapy

Kaplan-Meier analyses demonstrated clear survival separation between high- and low-risk groups defined by both anxiety and depression thresholds at each time point (two-sided log-rank P<0.001; Figure 4). For HADS-A, 5-year OS was 75.92% vs. 35.43% (baseline low- vs. high-risk), 77.02% vs. 15.77% (preoperative), and 82.88% vs. 33.33% (3-month postoperative). For HADS-D, corresponding 5-year OS rates were 73.18% vs. 23.80% (baseline), 73.19% vs. 20.20% (preoperative), and 73.80% vs. 35.02% (3-month postoperative).

Figure 4 Kaplan-Meier curves stratified by HADS-defined risk at baseline, preoperative, and 3-month postoperative timepoints (two-sided log-rank tests). The relationship between preoperative neoadjuvant for survival benefit and risk groups according to HADS scores (risk low and high). P values were calculated using a two-sided log-rank test. The HR and corresponding 95% CI were calculated using a Cox proportional-hazards regression. CI, confidence interval; HADS, Hospital Anxiety and Depression Scale; HADS-A, HADS-Anxiety; HADS-D, HADS-Depression; HR, hazard ratio.

Subgroup analyses showed that the prognostic effect of 3-month postoperative HADS-A persisted across histology [lung squamous cell carcinoma (LUSC) vs. non-LUSC], age (<65 vs. ≥65 years), pathologic response (pCR/MPR vs. non-OR), and treatment regimen (chemo-immunotherapy vs. chemotherapy)]. Stronger hazard ratios were observed in patients with LUSC histology, <65 years, achieving pCR/MPR, and receiving chemo-immunotherapy (Figure S4).

Collectively, these findings indicate that perioperative anxiety—particularly baseline and early postoperative elevations—provides independent prognostic information beyond ypTNM stage, whereas depression, though correlated with survival in univariable analyses, loses significance after adjustment for confounders.

Association between anxiety/depression and long-term survival

Kaplan-Meier analyses showed clear separation of survival curves between high- and low-risk groups defined by HADS thresholds at baseline, preoperative, and 3-month postoperative time points (two-sided log-rank P<0.001; Figure 4). For HADS-A, 5-year OS was 75.92% vs. 35.43% at baseline (low vs. high risk), 77.02% vs. 15.77% preoperatively, and 82.88% vs. 33.33% at 3-month postoperatively. For HADS-D, 5-year OS was 73.18% vs. 23.80% at baseline, 73.19% vs. 20.20% preoperatively, and 73.80% vs. 35.02% at 3-month postoperatively.

These Kaplan-Meier findings are concordant with Cox results reported above, underscoring that higher perioperative anxiety and depression burdens are associated with materially worse long-term survival, with the strongest and most consistent gradients observed for HADS-A across time points.

Nomogram construction and performance comparison

Using the multivariable Cox regression results for OS, a prognostic nomogram was developed integrating baseline HADS-A, 3-month postoperative HADS-A, and pathologic ypTNM stage to estimate long-term survival in patients with neoadjuvant-treated NSCLC (Figure 5A). The model exhibited excellent discrimination, with AUCs of 0.89, 0.90, and 0.89 for predicting 1-, 2-, and 3-year OS, respectively (Figure 5B). Compared with the traditional ypTNM system, the nomogram achieved a substantially higher C-index (0.85 vs. 0.64; 95% CI: 0.75–0.95 vs. 0.54–0.75; P<0.001), confirming its superior predictive accuracy. Time-dependent ROC curves across the 10–60-month follow-up interval further verified its robust and consistent discriminatory ability (Figure 5C).

Figure 5 Model incorporating HADS scores and prediction of survival. Prediction of overall survival of patients undergoing neoadjuvant NSCLC 1, 2 and 3 years after surgery based on constructed nomogram of HADS scores (A). The 1-, 2- and 3-year AUC values and C-indices of the nomogram (B); time-dependent ROC (10–60 months) comparing nomogram vs. ypTNM (C); 3-year calibration (D); decision-curve analysis (E). Gray lines: the horizontal line means that all samples are negative (individual predicted probability < threshold probability), no one has intervened, and the net benefit is 0. The slanted bar indicates that all samples were positive and all participants received the intervention. AUC, area under the curve; C-index, concordance index; HADS, Hospital Anxiety and Depression Scale; HADS-A, HADS-Anxiety; NSCLC, non-small cell lung cancer; ROC, receiver operating characteristic; ypTNM, post-neoadjuvant pathologic tumor lymph nodes metastasis.

Model calibration was evaluated through bootstrap internal validation (800 resamples). The 3-year calibration curve closely approximated the 45° reference line, demonstrating excellent agreement between predicted and observed OS (Figure 5D). DCA showed that the nomogram provided a consistently greater net clinical benefit than ypTNM alone across a clinically relevant threshold probability range of 0.20–0.80, underscoring its enhanced practical applicability and clinical utility (Figure 5E).

Analysis of factors influencing HADS scores

ROC analyses were performed to identify perioperative indicators associated with elevated HADS scores. For HADS-A, longer postoperative hospital stay and time to first chest-tube removal were linked to higher scores, with optimal cut-offs of 5.5 days (P=0.004; AUC =0.63) and 2.5 days (P=0.004; AUC =0.63), respectively. For HADS-D, time to first chest-tube removal ≥5.5 days was associated with higher scores (P=0.03; AUC =0.60). Among patients with postoperative drainage retained >2.5 days (n=130), greater cumulative drainage volume correlated with higher HADS-D, with an optimal threshold of 1,382.5 mL (P=0.046; AUC =0.60).

Discussion

In this multicenter cohort of neoadjuvant-treated NSCLC, perioperative anxiety—particularly elevations at baseline and three months after surgery—emerged as an independent predictor of inferior OS after adjustment for ypTNM stage. Depressive symptoms correlated with survival in univariable analyses but did not retain significance after multivariable adjustment, suggesting shared variance with anxiety or differences in prognostic salience across perioperative timepoints. The nomogram that combined baseline and 3-month HADS-A with ypTNM showed strong discrimination (AUCs for 1-, 2-, and 3-year OS ≈0.89, 0.90, and 0.89), a higher C-index than ypTNM alone, and favorable calibration and decision-curve profiles, indicating meaningful incremental clinical utility for individualized risk stratification.

Clinically, these findings support two pragmatic checkpoints for psychological screening: before treatment initiation and during early postoperative recovery. The data-driven thresholds were lower than conventional diagnostic cut-offs, implying that subthreshold anxiety still carries prognostic information in surgical oncology (15,23). Embedding brief screening at these timepoints—paired with stepped, evidence-based responses [psychoeducation, cognitive behavior therapy (CBT)-informed strategies, pain and sleep optimization, and referral to psycho-oncology when indicated]—is feasible within thoracic enhanced recovery after surgery (ERAS) pathways and may improve adherence, symptom control, and longer-term outcomes (24-26). The perioperative anxiety in the multivariate model is not sufficient to serve as the basis for causal logic in this observational study. Anxiety may be a surrogate marker of disease burden, recovery capacity, or unmeasured clinical and social factors. Besides, during the treatment process, patients’ awareness of the cancer may lead to varying degrees of anxiety risk, resulting in potential residual confounding. Therefore, clinicians shall judge the necessity and timings of psychological intervention based on the actual situation of patients cautiously, while avoiding neglecting other relevant factors.

Several mechanisms may plausibly link anxiety with poorer survival, including stress-related neuroendocrine regulation (27), increased pro-inflammatory markers (28), and impacts on the tumor microenvironment (29,30) that impair antitumor immunity. Furthermore, psychological distress can impede T cells and facilitate immune escape by regulating HPA-axis and inflammatory signaling pathway (31,32). While the present study was not designed to test causality or biologic mediators, the consistency of the anxiety signal across subgroups and the superior performance of the anxiety-augmented nomogram are compatible with a biobehavioral pathway that warrants prospective investigation.

This study has notable strengths: a multicenter design; standardized HADS assessments at baseline, preoperative, and early postoperative timepoints; and rigorous time-to-event modeling with internal validation, calibration analysis, time-dependent ROC, and decision-curve analysis. Limitations include the retrospective design, potential residual confounding (e.g., analgesic strategies, sleep quality, social support) (33-35), and a modest number of events that may constrain model complexity and risk overfitting despite bootstrap validation. Thresholds were derived within this cohort and require external replication before routine adoption which was not conducted in the current study due to insufficient case numbers; the attenuation of depression in adjusted models may reflect collinearity with anxiety, measurement timing, or limited power. The cases in study mainly come from male populations in Asian cities and the HADS scores were assessed using a Chinese-validated version, so the results may vary in other regions, cultures or ethnic backgrounds.

In summary, perioperative anxiety measured at baseline and early postoperative follow-up provides independent, clinically meaningful prognostic information in neoadjuvant-treated NSCLC. Incorporating routine anxiety screening at these two checkpoints and applying the validated nomogram can refine postoperative risk stratification beyond staging alone, offering a practical avenue to target supportive interventions and inform shared decision-making.

Conclusions

In this multicenter cohort of neoadjuvant-treated NSCLC, perioperative anxiety—measured at baseline and again three months after surgery—independently predicted long-term OS beyond pathological staging. Depression showed univariable associations but did not retain significance after multivariable adjustment. A nomogram integrating baseline and 3-month HADS-A with ypTNM provided superior discrimination, sound calibration, and higher net clinical benefit compared with staging alone. These findings support routine, brief anxiety screening at treatment initiation and during early postoperative recovery, paired with targeted psycho-oncology interventions, to refine postoperative risk stratification and inform shared decision-making (36,37).

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0006/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0006/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0006/prf

Funding: This work was supported by the National Natural Science Foundation of China (No. 82203307), the Fujian Provincial Health Technology Project (No. 2022GGA021) and the Talent Fund Project of Fujian Medical University Union Hospital (No. 2021XH029).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0006/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. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of Fujian Medical University Union Hospital (Approval No. 2025KY667). Informed consent was waived in this retrospective study. All participating centers were informed and agreed to the study.

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. Desai A, Schwed K, Kalesinskas L, et al. Clinical Outcomes of Perioperative Immunotherapy in Resectable Non-Small Cell Lung Cancer. JAMA Netw Open 2025;8:e2517953. [Crossref] [PubMed]
2. Ay L, Steiner D, Fabikan H, et al. Neoadjuvant therapy in early-stage non-small cell lung cancer: A real-world analysis. Lung Cancer 2024;198:107997. [Crossref] [PubMed]
3. Merlo A, Carlson R, Espey J 3rd, et al. Postoperative Symptom Burden in Patients Undergoing Lung Cancer Surgery. J Pain Symptom Manage 2022;64:254-67. [Crossref] [PubMed]
4. Said NB, Alsayed A. Cancer-Related Fatigue, Psychological Symptoms, and Sleep Quality and Influencing Factors in Cancer Patients Receiving Chemotherapy. Inquiry 2026;63:469580261422596. [Crossref] [PubMed]
5. Hou T, Peng W, Ho MH, et al. Factors associated with cancer-related cognitive impairment in patients with lung cancer: A systematic review. Eur J Oncol Nurs 2024;71:102657. [Crossref] [PubMed]
6. Linares-Moya M, Rodríguez-Torres J, Heredia-Ciuró A, et al. Psychological distress prior to surgery is related to symptom burden and health status in lung cancer survivors. Support Care Cancer 2022;30:1579-86. [Crossref] [PubMed]
7. Velasco-Durantez V, Cruz-Castellanos P, Hernandez R, et al. Prospective study of predictors for anxiety, depression, and somatization in a sample of 1807 cancer patients. Sci Rep 2024;14:3188. [Crossref] [PubMed]
8. Berggreen C, Schrøder JH, Christensen T, et al. TO INFORM OR NOT TO INFORM about venous thromboembolisms - A qualitative study on communication between healthcare professionals and patients with lung cancer. Thromb Res 2024;243:109132. [Crossref] [PubMed]
9. Cruzado JA, Martínez-García V, González IP, et al. Preoperative detection and evaluation of the need for psychological intervention in cancer and non-cancer patients referred to thoracic surgery. Cir Esp (Engl Ed) 2022;100:359-66. [Crossref] [PubMed]
10. Li J, Ma C. Anxiety and depression during 3-year follow-up period in postoperative gastrointestinal cancer patients: prevalence, vertical change, risk factors, and prognostic value. Ir J Med Sci 2023;192:2621-9. [Crossref] [PubMed]
11. Huang WC, Kao TN, Liao SC, et al. Psychological features and quality of life in low-dose CT lung cancer screening: a comparative analysis with general population data. Lung Cancer 2026;213:109307. [Crossref] [PubMed]
12. Naser AY, Hameed AN, Mustafa N, et al. Depression and Anxiety in Patients With Cancer: A Cross-Sectional Study. Front Psychol 2021;12:585534. [Crossref] [PubMed]
13. Yan F, Yuan LH, He X, et al. Correlation between pre-anesthesia anxiety and emergence agitation in non-small cell lung cancer surgery patients. World J Psychiatry 2024;14:930-7. [Crossref] [PubMed]
14. Zeng Y, Hu CH, Li YZ, et al. Association between pretreatment emotional distress and immune checkpoint inhibitor response in non-small-cell lung cancer. Nat Med 2024;30:1680-8. [Crossref] [PubMed]
15. Stylianou C, Konstantinidis T, Kalemikerakis I, et al. Anxiety and Depression in Non-Small Cell Lung Cancer Patients Receiving Chemotherapy Compared to Those Receiving Immunotherapy. Maedica (Bucur) 2024;19:453-60. [Crossref] [PubMed]
16. Carey M, Noble N, Sanson-Fisher R, et al. Identifying psychological morbidity among people with cancer using the Hospital Anxiety and Depression Scale: time to revisit first principles? Psychooncology 2012;21:229-38. [Crossref] [PubMed]
17. Liang W, Cai K, Chen C, et al. Expert consensus on neoadjuvant immunotherapy for non-small cell lung cancer. Transl Lung Cancer Res 2020;9:2696-715. [Crossref] [PubMed]
18. Guo C, Huang X. Hospital anxiety and depression scale exhibits good consistency but shorter assessment time than Zung self-rating anxiety/depression scale for evaluating anxiety/depression in non-small cell lung cancer. Medicine (Baltimore) 2021;100:e24428. [Crossref] [PubMed]
19. Deng L, Chen B. Two-scale assessment of anxiety and depression in postoperative non-small cell lung cancer patients: their prevalence, risk factors, and prognostic potency. Ir J Med Sci 2023;192:2613-9. [Crossref] [PubMed]
20. Jiang Y, Yang JP, Liu Q. Effectiveness of Remote Symptom Monitoring and Progressive Muscle Relaxation in Enhancing Well-Being and Quality of Life in Lung Cancer Patients Undergoing Chemotherapy. Tohoku J Exp Med 2025;267:225-33. [Crossref] [PubMed]
21. Reiner AS, Alici Y, Correa DD, et al. Anxiety and depression in patients with histiocytic neoplasms and their associated clinical features. Blood Adv 2025;9:1376-86. [Crossref] [PubMed]
22. Chau YF, Zhou H, Chen B, et al. Screening for depression and anxiety in lung cancer patients: A real-world study using GAD-7 and HADS. Thorac Cancer 2024;15:1041-9. [Crossref] [PubMed]
23. Osborne RH, Elsworth GR, Sprangers MA, et al. The value of the Hospital Anxiety and Depression Scale (HADS) for comparing women with early onset breast cancer with population-based reference women. Qual Life Res 2004;13:191-206. [Crossref] [PubMed]
24. Wang F, Zhang S, Song B, et al. Anxiety, depression, and quality of life in postoperative non-small cell lung cancer patients under the intervention of cognitive-behavioral stress management. Front Psychol 2023;14:1138070. [Crossref] [PubMed]
25. Yuan Z, Gao L, Zheng M, et al. Effect of Multimodal Health Education Combined with the Feedback Method in Perioperative Patients with Lung Cancer: A Randomised Controlled Study. Patient Prefer Adherence 2023;17:413-20. [Crossref] [PubMed]
26. Zhao L, Ma L, Chen X, et al. Psychological nursing intervention improve the mental health status of young patients with lung cancer surgery during the perioperative period. Medicine (Baltimore) 2021;100:e26736. [Crossref] [PubMed]
27. Zhang S, Yu F, Che A, et al. Neuroendocrine Regulation of Stress-Induced T Cell Dysfunction during Lung Cancer Immunosurveillance via the Kisspeptin/GPR54 Signaling Pathway. Adv Sci (Weinh) 2022;9:e2104132. [Crossref] [PubMed]
28. Cohen M, Shamay Y, Czamanski-Cohen J, et al. Linkage between Psychological Factors and Response to Immune Checkpoint Inhibitor Therapy: A Preliminary Study. Cells 2023;12:2471. [Crossref] [PubMed]
29. Tian W, Liu Y, Cao C, et al. Chronic Stress: Impacts on Tumor Microenvironment and Implications for Anti-Cancer Treatments. Front Cell Dev Biol 2021;9:777018. [Crossref] [PubMed]
30. Li Y, Yu H, Li ZM, et al. Colorectal cancer cells hijack a brain-gut polysynaptic circuit from the lateral septum to enteric neurons to sustain tumor growth. Nat Cancer 2025;6:1800-20. [Crossref] [PubMed]
31. Bucsek MJ, Qiao G, MacDonald CR, et al. β-Adrenergic Signaling in Mice Housed at Standard Temperatures Suppresses an Effector Phenotype in CD8(+) T Cells and Undermines Checkpoint Inhibitor Therapy. Cancer Res 2017;77:5639-51. [Crossref] [PubMed]
32. Yang H, Xia L, Chen J, et al. Stress-glucocorticoid-TSC22D3 axis compromises therapy-induced antitumor immunity. Nat Med 2019;25:1428-41. [Crossref] [PubMed]
33. Lee S, Xue Y, Petricca J, et al. The impact of pre-operative depression on pain outcomes after major surgery: a systematic review and meta-analysis. Anaesthesia 2024;79:423-34. [Crossref] [PubMed]
34. Chen M, Huang Y, Zhang J, et al. Impact of preoperative anxiety on postoperative outcomes in patients undergoing minimally invasive thoracoscopic surgery: A prospective cohort study. Eur J Surg Oncol 2024;50:108605. [Crossref] [PubMed]
35. Zhu J, Chen SH, Guo JY, et al. Effect of digital storytelling intervention on resilience, self-efficacy and quality of life among patients with non-small cell lung cancer (NSCLC): A randomized controlled trial. Eur J Oncol Nurs 2024;69:102535. [Crossref] [PubMed]
36. Jansen F, Lissenberg-Witte BI, Krebber AMH, et al. Stepped care targeting psychological distress in head and neck cancer and lung cancer patients: which groups specifically benefit? Secondary analyses of a randomized controlled trial. Support Care Cancer 2019;27:4543-53. [Crossref] [PubMed]
37. Li S, Ding X, Zhao Y, et al. Intravenous patient-controlled analgesia plus psychoeducational intervention for acute postoperative pain in patients with pulmonary nodules after thoracoscopic surgery: a retrospective cohort study. BMC Anesthesiol 2021;21:281. [Crossref] [PubMed]