The value of positron emission tomography scan in predicting pathologic response of non-small cell lung cancer managed by neoadjuvant chemo-immunotherapy

Introduction

Non-small cell lung cancer (NSCLC) remains the leading cause of cancer-related mortality worldwide (1). For patients with resectable, locally advanced NSCLC, neoadjuvant chemo-immunotherapy has emerged as a standard therapeutic approach alone or in combination with adjuvant immunotherapy, significantly improving overall survival compared to chemotherapy alone (2-5). However, the accurate assessment of tumor response following induction chemo-immunotherapy is critical for surgical planning and to predict patient outcome. Tumor response assessment may also tailor adjuvant approaches in the future (6).

Radiological re-staging after neoadjuvant therapy is typically performed by contrast-enhanced computed tomography (CT). This tumor response assessment is mostly based on the changes in size and morphology of the tumor and lymph node following induction therapy. However, this imaging modality provides no information on their metabolic activity. It has been demonstrated that patients experiencing pathologic complete response (pCR) with induction chemo-immunotherapy (i.e. no residual tumor in the tumor and lymph node beds on anatomopathological assessment of the resected specimen) had the best event free and overall survival (7). Therefore, the use of ¹⁸F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) may be superior to standard imaging in predicting tumor response to chemo-immunotherapy and potentially patient survival (8,9).

Lymph nodes are center in the immune response to the tumor (10,11). Metabolic tumor or lymph node activity is typically measured as a standardized fluorodeoxyglucose uptake value (SUV). However, it is hard to discriminate between the glucose uptake from the tumor or the immune cells located in the lymph nodes. Thus, a better understanding of the glucose metabolism and consumption happening in the lymph nodes in the context of neo-adjuvant immunotherapy may be relevant. While prior studies have explored the role of PET/CT in NSCLC, the relative predictive significance of variation in SUV (ΔSUV) in the primary tumor and lymph node following chemo-immunotherapy remains unclear (8,9).

In this study, we evaluated the prognostic value of PET/CT metabolic changes in the primary tumor and involved lymph nodes of NSCLC patients undergoing neoadjuvant chemo-immunotherapy followed by surgery. Specifically, we assessed whether the mean and max ¹⁸F-FDG PET/CT SUV changes between pre and post chemo-immunotherapy induction of tumors and lymph nodes predicted pCR and patient overall survival. We report a potential role for PET/CT at the level of the NSCLC lymph nodes. After confirmation, these findings could refine post-induction assessment strategies and improve decision-making in NSCLC management. We present this article in accordance with the STARD reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1686/rc).

Methods

Patients and data collection

This retrospective study included all stage II or III [following tumor-node-metastasis (TNM) stage 8^th edition] node positive proven by endobronchial ultrasound (EBUS) NSCLC patients treated with neoadjuvant chemo-immunotherapy followed by surgery in the Lausanne University Hospital (CHUV) between 2017 and 2023. Patients were included from our prospectively collected database if they presented a NSCLC managed by perioperative chemo-immunotherapy and surgery and underwent a ¹⁸F-FDG PET/CT scan before and after the induction therapy prior to their surgery. The study was approved by the Ethics Committee of Vaud (No. CER-VD 2022-01883) and general consent was obtained from all participants. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Surgical data were extracted from our prospectively collected electronic database and included: patient demographics, comorbidities, oncological TNM stage (8^th edition), surgical characteristics, pathological findings [major pathologic response (MPR), i.e., less than 10% of viable tumor in the initial tumor bed; pCR, i.e., no viable tumor tissue on the specimen; complete resection], postoperative complications patient follow-up and survival on the December 31^st, 2023. Overall survival was the time between the initiation of a therapy and death. Patients were censored from the moment of their last follow-up. Last inclusion was on December 31^st, 2023.

¹⁸F-FDG PET/CT acquisition and image analysis

All examinations were performed on a Siemens Biograph Vision 600 (Siemens Healthineers, Erlangen, Germany) or a Discovery D690 TOF (GE Healthcare, Boston, USA) system and served for primary lung cancer staging or for early therapy-response assessment. The details of ¹⁸F-FDG PET/CT acquisition and patient management are provided in Appendix 1. All quantitative measurements were performed by experienced members of the Nuclear Medicine Department and verified by a senior nuclear physician (>20 years experience), blinded to clinical outcomes. Discordances in image interpretation or measurements were resolved by consensus. Volumes of interest (VOIs) were drawn semi-automatically around lesions that exhibited visually abnormal FDG avidity. Within each VOI, the maximum and mean standardized uptake values (SUVmax, SUVmean) were recorded. To avoid differences in the absolute SUV values, each patient was imaged by the same machine at baseline and at the post-induction period. Moreover, we mostly focused on changes in SUV with therapy (ΔSUV) with each patient being its own control. Finally, following the recommendations from European Association of Nuclear Medicine (EANM) standardized reconstructions have been used to assure inter-scanner comparison (12).

For nodal assessment, the mediastinal lymph node with abnormal uptake on either side was delineated with a three-dimensional contour large enough to encompass the entire node. In case of multiple positive lymph nodes, a mean of the measurement across all the nodes was performed. To compensate for inter-individual variability in systemic tracer distribution, SUVmax of the target node was normalized to SUVmax of the spleen and of the fifth lumbar vertebral body.

Statistical analysis

Results are presented as numbers with percentages for binary variables, means with standard deviation for normally distributed continuous variables or medians with interquartile range for non-normally distributed continuous variables or nominal variables with large number categories. Age and follow-up time were considered not normally distributed. Numerical variables were compared between surgery and radiotherapy groups using the unpaired Student’s t-test or the Mann-Whitney U test according to the distribution. Categorical variables were compared using the Chi-squared test. Overall survival was calculated using the Kaplan-Meier formula and compared with a log-rank test. A P value <0.05 was defined as the threshold for statistical significance. Missing data were handled using listwise deletion. Receiver operating characteristic (ROC) curve analysis was applied to estimate the prediction accuracy of parameters in pathological and survival evaluation. The area under the ROC curve (AUC) and 95% confidential interval (CI) were calculated to evaluate the efficacy in pCR and survival prediction. The optimal cut-off values and corresponding sensitivity and specificity for pCR and survival prediction were calculated via ROC analyses. The statistics provided are descriptive in nature. All statistical analyses were performed using GraphPad Prism software.

Results

Patient characteristics

Among stage II–III node positive NSCLC patients who underwent induction chemo-immunotherapy followed by resection between 2017 and 2023, we identified 37 that had pre- and post-induction ¹⁸F-FDG PET/CT scans. All completed 3 to 4 cycles of chemo-immunotherapy and could undergo surgery. The male/female ratio was 20/17 (54% men). Mean patient age was 64±7 years old. Cancer histologies were adenocarcinoma (n=26, 70%), squamous cell carcinoma (n=9, 24%) and not otherwise specified (n=2, 5%). PD-L1 status was <1% in 12 (32%) patients, between 1–50% in 9 (24%) patients, over 50% in 8 patients (22%) and not reported in 8 patients (22%). All patients had a node positive disease (N1 or N2) proven by EBUS. Cancer stages were IIB (n=1, 3%), IIIA (n=27, 73%) and IIIB (n=9, 24%). Patient demographic and comorbidity characteristics are reported in Table 1. Among the most frequent comorbidities, there was high blood pressure (n=17, 46%) and chronic obstructive pulmonary disease (n=13, 35%). Most surgical procedures were open thoracotomies (n=22, 59%). Anatomical resections were lobectomy (n=34, 92%), bilobectomies (n=2, 5%) and one segmentectomy (n=1, 3%). There were no pneumonectomies. Among these procedures, there were 4 bronchial sleeves (10%) and 2 bronchial and vascular double sleeves (5%) performed. Postoperative morbidity was frequent (59% of patients) but manageable with the most frequent complication being pneumonia (32%). There was no 90-day mortality. The surgical characteristics of the cohort are reported in Table 2. Anatomopathological assessment of the resected tumor specimen showed complete tumor resection in 34 of 37 patients (92%) with three R1 resections due to lymph node capsular effraction (8%). MPR and pCR occurred in 19 and 11 patients, respectively (51% and 30%, respectively). All patients underwent adjuvant immunotherapy for one year (4). Mean PFS was 19.8 months across the cohort and tumor recurrences occurred in 14 patients. Of these 14 recurrences, the majority were distant (n=7, 50%), local and distant (n=6, 43%) and local (n=1, 7%). The cohort 5-year overall survival was 81%.

Value of primary tumor and involved lymph node ¹⁸F-FDG SUV changes to predict pCR

We first assessed max (ΔSUVTmax) and mean (ΔSUVTmean) tumor SUV changes and how they correlated with pCR. We found that ΔSUVTmax and ΔSUVTmean had a predictive value of AUC =0.8086 (P=0.005, 95% CI: 0.6555–0.9637) and AUC =0.8135 (P=0.004, 95% CI: 0.6667–0.9602) for pCR following chemo-immunotherapy respectively (Figure 1A,1B). When including both MPR and pCR, ΔSUVTmax was slightly better (AUC =0.8465, P<0.001, 95% CI: 0.7147–0.9783) than ΔSUVTmean (AUC =0.8421, P<0.001, 95% CI: 0.7083–0.9759) to predict either MPR or pCR following induction (Figure S1). Furthermore, both ΔSUVTmax and ΔSUVTmean predicted downstaging from Tx to T0 was comparable (AUC =0.8600, P=0.005, 95% CI: 0.7290–0.9910 and AUC =0.8600, P=0.005, 95% CI: 0.7354–0.9846, respectively) after induction therapy (Figure S1).

Figure 1 ROC curve analysis for ΔSUVT Max (A), ΔSUVT Mean (B), ΔSUVN Max (C) and ΔSUVN Mean (D) to predict pCR following induction therapy. AUC, area under the curve; pCR, complete pathologic response; ROC, receiver operating characteristic; SUV, standard uptake value; ΔSUVN Max, variation in the maximum SUV of lymph nodes; ΔSUVN Mean, variation in the mean SUV of the lymph nodes; ΔSUVT Max, variation in the maximum SUV of the primary tumor; ΔSUVT Mean, variation in the mean SUV of the primary tumor.

We then determined how lymph node SUV mean (ΔSUVNmean) and max (ΔSUVNmax) variations predicted tumor pathologic response. Interestingly, ΔSUVNmax was the best predictor of pCR (AUC =0.9044, P<0.001, 95% CI: 0.8027–1) followed by ΔSUVNmean (AUC =0.8990, P<0.001, 95% CI: 0.7934–1, Figure 1C,1D). Of interest, PD-L1 status was not predictive of the drop in SUVNmax (Table 3). When predicting both MPR and pCR, ΔSUVNmean and max were comparable to ΔSUVT (AUC =0.7843, P=0.004; 95% CI: 0.6284–0.9402 and AUC =0.7595, P=0.01, 95% CI: 0.5893–0.9298), respectively (Figure S1). To predict a downstaging from Nx to N0, ΔSUVNmax was a better predictor (AUC =0.7978, P=0.004, 95% CI: 0.6475–0.9481) than ΔSUVNmean (AUC =0.7578, P=0.01, 95% CI: 0.5888–0.9268; Figure S1). Thus, lymph node ΔSUV seemed to be a better predictor of tumor response than primary tumor ΔSUV.

Survival outcomes following SUV changes in primary tumor and lymph nodes

Because ΔSUVNmax was the best parameter to predict the occurrence of pCR following neoadjuvant chemo-immunotherapy in our cohort, we next investigated if this parameter could predict patient survival. Based on the ROC sensitivity and specificity analysis (Table 4), we found that a 70% drop of the ΔSUVNmax predicted pCR with a sensitivity of 88% and a specificity of 78%. We then used this parameter to determine if it correlated with overall survival. We found that patients with a more than 70% drop in ΔSUVNmax had a median survival of 57 months compared to the 32 months median overall survival for patients with less ΔSUVNmax reduction (P=0.06, Figure 2). Furthermore, a Pearson correlation did not find an association between tumor SUVmean and max (P=0.78) or between lymph node SUVmean and max at baseline (P=0.36) and patient survival (Table 5). Overall, ΔSUVNmax between the pre- and post-chemo-immunotherapy induction imaging was the best predictor of patient survival.

Figure 2 Kaplan-Meier analysis comparing the survival of patients presenting a reduction of more than 70% vs. less than 70% in the maximum SUV of the associated lymph nodes. SUV, standard uptake value; ΔSUVN Max, variation in the maximum SUV of lymph nodes.

Discussion

Perioperative chemo-immunotherapy has improved the management of locally advanced NSCLC (2-5). These therapeutic approaches have led to important tumor responses with pCRs in up to 25% of patients and better survival outcomes (13). There is a great interest in being able to predict pCR after induction therapy to optimize subsequent patient management. Here, we used ¹⁸F-FDG PET/CT SUV variation in tumor and lymph nodes between pre- and post-induction therapy to determine how it predicted pCR and overall survival in locally advanced NSCLC patients. We found that both tumor and lymph node ΔSUV could predict pCR but that lymph node ΔSUV was the more precise and correlated with patient overall survival.

CT scans have shown very limited predictive value for assessing local disease response after induction chemo-immunotherapy, particularly regarding nodal involvement, as morphological changes do not reliably correlate with pathologic response. The use of PET/CT has been tested in the context of stage II–III NSCLC patients receiving pre-operative immunotherapy (9). It was found that tumor SUVmax drop following induction therapy correlated with MPR (9,14,15). While we did find a correlation between tumor ΔSUVmax drop and pCR, our study indicated that lymph node ΔSUVmax was a better predictor of pCR. This is in line with Liu et al. who established a machine learning model using ¹⁸F-FDG PET radiomics parameters and found that lymph node ΔSUV was the best predictor of pCR (16).

PET/CT was also investigated in predicting patient survival in the context of immunotherapy. Dall’Olio et al. defined the total metabolic tumor volume (tMTV) as a predictor of tumor response to immunotherapy and survival (17). A later study by Vekens et al. concluded that the tumor SUVmax was the only ¹⁸F-FDG PET/CT parameter associated with progression free but not overall survival (18). Here, we observed that patients with a ΔSUVmax drop of 70% in the lymph nodes following induction chemo-immunotherapy had a better overall survival. Nevertheless, false positive lymph node findings on PET/CT imaging after neoadjuvant chemoimmunotherapy in resectable early-stage NSCLC are a clinical challenge. These findings can complicate restaging and surgical decision-making, as radiological assessments may not always accurately reflect pathologic outcomes (19,20). A notable phenomenon contributing to false-positive results is the “nodal flaring”, characterized by radiologically abnormal lymph nodes that, upon pathological examination, are cancer-free (19). In our study, only one patient experienced a nodal flaring (with a ΔSUVmax increase of 159%) while presenting negative nodes at the pathologic examination. Further studies have highlighted the limitations of PET/CT in accurately staging lymph nodes post-neoadjuvant therapy. For instance, a multicenter study reported a sensitivity of 66.7% and a specificity of 83.5% for PET/CT in detecting ypN2 disease after neoadjuvant chemoimmunotherapy, indicating a significant rate of false positives. Finally, conditions such as sarcoidosis or sarcoid-like reactions induced by immunotherapy can lead to increased FDG uptake in the lymph nodes. Thus, caution in the imaging interpretation of PET/CT is granted to avoid unnecessary alterations in treatment plans (21,22).

A better understanding of the contribution of lymph nodes in the context of neoadjuvant chemo-immunotherapy may be needed. Indeed, the role of lymph nodes in tumor immune responses remains unclear. While some have suggested that lymph node sparing during surgery could be associated to better tumor prognosis because of CD8⁺ memory T-cells repertoire preservation (23,24) the current guidelines require complete lymph node removal for best survival (25). In our study, the lymph node metabolism changes were more strongly correlated to pCR than the tumor metabolism. However, given ¹⁸F-FDG PET/CT, SUV measurement is not specific for tumor or immune cells, further studies are required to understand which cells specifically change their metabolism in the context of pCR.

Being able to predict pCR is a great area of interest in the context of locally advanced NSCLC managed by chemo-immunotherapy. Studies have looked into the predictive value of circulating tumor DNA (ctDNA) in regards to pCR or MPR (26). However, the sensitivity/specificity of this parameter is still insufficient for decision making in this context. Here, we show a nice correlation of the lymph node SUVmax variation with relatively good sensitivity and specificity. While these findings require further confirmation on a wider prospective cohort, the combination of different parameters such as PET/CT + ctDNA and others could potentially be promising elements to predict pCR in the future.

Our study presents several limitations including missing information (retrospective trial) and a low sample size. This could impact on the variability and induce sample bias in the ROC analysis. In addition, two different PET/CT scanners were used to perform the measurements which could have introduced a bias in our comparisons. However, we made sure that each post induction patient was assessed by the same machine used for pre-induction. In addition, because each patient was its own control, the ΔSUV used in this study seems reliable. This increases the sensitivity of what can be measured even with a small sample size. Furthermore, even in the case of imaging with different machines, the EANM standardized reconstructions allow safe and reliable inter-matching assessments (12). Larger, prospective trials are required to confirm the predictive value of SUV change in lymph nodes described in this study.

Conclusions

Our results suggest ¹⁸F-FDG PET/CT and particularly the SUV change in the lymph nodes following chemo-immunotherapy seems to be a promising parameter to predict tumor response following chemo-immunotherapy. In the era of personalized care, SUV change in lymph nodes could be an additional parameters to combine with others (such as ctDNA) in order to refine the management proposed to NSCLC patients in multidisciplinary tumor board.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1686/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1686/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1686/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1686/coif). N.M. reports presentation’s fees to institution from AstraZeneca; travel support to institution from Rohe and PharmaMar; fees to institution for participation on advisory board for PharmaMar, AstraZeneca, MSD, BMS and Takeda. S.P. reports grants from Amgen, Arcus, AstraZeneca, Beigene, Boehringer Ingelheim, Bristol-Myers Squibb, Eli Lilly, GSK, iTeos, Merck Sharp and Dohme, Mirati, Pharma Mar, Pfizer, Promontory Therapeutics, Roche/Genentech, Seattle Genetics; consulting fees from AbbVie, Amgen, Arcus, AstraZeneca, Bayer, Beigene, BioNTech, BerGenBio, Bicycle Therapeutics, Biocartis, BioInvent, Blueprint Medicines, Boehringer Ingelheim, Bristol-Myers Squibb, Clovis, Daiichi Sankyo, Debiopharm, Eli Lilly, F-Star, Foundation Medicine, Genmab, Genzyme, Gilead, GSK, Hutchmed, Illumina, Incyte, Ipsen, iTeos, Janssen, Qlucore, Merck Sharp and Dohme, Merck Serono, Merrimack, Mirati, Nuvation Bio, Nykode Therapeutics, Novartis, Novocure, Pharma Mar, Promontory Therapeutics, Pfizer, Regeneron, Roche/Genentech, Sanofi, Seattle Genetics, Takeda, Zymeworks; honoraria from AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Eli Lilly, Foundation Medicine, GSK, Illumina, Ipsen, Merck Sharp and Dohme, Mirati, Novartis, Pfizer, Roche/Genentech, Sanofi, Seattle Genetics, Takeda; support for travel from AstraZeneca, Bristol-Myers Squibb, Daiichi Sankyo, Eli Lilly, Merck Sharp and Dohme, Novartis, Pfizer, Roche/Genentech, Takeda; is present at the safety or advisory board of AbbVie, Amgen, Arcus, AstraZeneca, Bayer, Beigene, BioNTech, BerGenBio, Bicycle Therapeutics, Biocartis, BioInvent, Blueprint Medicines, Boehringer Ingelheim, Bristol-Myers Squibb, Clovis, Daiichi Sankyo, Debiopharm, Eli Lilly, F-Star, Foundation Medicine, Genmab, Genzyme, Gilead, GSK, Hutchmed, Illumina, Incyte, Ipsen, iTeos, Janssen, Qlucore, Merck Sharp and Dohme, Merck Serono, Merrimack, Mirati, Nuvation Bio, Nykode Therapeutics, Novartis, Novocure, Pharma Mar, Promontory Therapeutics, Pfizer, Regeneron, Roche/Genentech, Sanofi, Seattle Genetics, Takeda, Zymeworks. J.Y.P. reports industry-supported symposia from AstraZeneca, Merck Sharp and Dohme, Bristol-Myers Squibb, Johnson and Johnson and Medtronic. 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 approved by the Ethics Committee of Vaud (No. CER-VD 2022-01883) and general consent was obtained from all participants. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Siegel RL, Miller KD, Fuchs HE, et al. Cancer statistics, 2022. CA Cancer J Clin 2022;72:7-33. [Crossref] [PubMed]
2. Forde PM, Spicer J, Lu S, et al. Neoadjuvant Nivolumab plus Chemotherapy in Resectable Lung Cancer. N Engl J Med 2022;386:1973-85. [Crossref] [PubMed]
3. Provencio M, Nadal E, Insa A, et al. Perioperative chemotherapy and nivolumab in non-small-cell lung cancer (NADIM): 5-year clinical outcomes from a multicentre, single-arm, phase 2 trial. Lancet Oncol 2024;25:1453-64. [Crossref] [PubMed]
4. 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. [Crossref] [PubMed]
5. Cascone T, Awad MM, Spicer JD, et al. Perioperative Nivolumab in Resectable Lung Cancer. N Engl J Med 2024;390:1756-69. [Crossref] [PubMed]
6. Reck M, Nadal E, Girard N, et al. MDT-BRIDGE: Neoadjuvant Durvalumab Plus Chemotherapy Followed by Either Surgery and Adjuvant Durvalumab or Chemoradiotherapy and Consolidation Durvalumab in Resectable or Borderline-resectable Stage IIB-IIIB NSCLC. Clin Lung Cancer 2024;25:587-593.e3.
7. Travis WD, Dacic S, Wistuba I, et al. IASLC Multidisciplinary Recommendations for Pathologic Assessment of Lung Cancer Resection Specimens After Neoadjuvant Therapy. J Thorac Oncol 2020;15:709-40. [Crossref] [PubMed]
8. Chen X, Bai G, Zang R, et al. Utility of 18F-FDG uptake in predicting major pathological response to neoadjuvant immunotherapy in patients with resectable non‑small cell lung cancer. Transl Oncol 2023;35:101725. [Crossref] [PubMed]
9. Chen ZY, Fu R, Tan XY, et al. Dynamic 18 F-FDG PET/CT can predict the major pathological response to neoadjuvant immunotherapy in non-small cell lung cancer. Thorac Cancer 2022;13:2524-31. [Crossref] [PubMed]
10. Zhong NN, Liu B, Bu LL. Neoadjuvant immunotherapy: new horizon for lymph node preservation. MedComm (2020) 2024;5:e577.
11. Jones D, Villena-Vargas J. The Evolving Role of Lymph Node Resection in the Era of Immunotherapy for Non-Small Cell Lung Cancer. Semin Thorac Cardiovasc Surg 2025;37:441-6. [Crossref] [PubMed]
12. Aide N, Lasnon C, Veit-Haibach P, et al. EANM/EARL harmonization strategies in PET quantification: from daily practice to multicentre oncological studies. Eur J Nucl Med Mol Imaging 2017;44:17-31. [Crossref] [PubMed]
13. 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]
14. Yoo J, Lee J, Cheon M, et al. Predictive Value of (18)F-FDG PET/CT Using Machine Learning for Pathological Response to Neoadjuvant Concurrent Chemoradiotherapy in Patients with Stage III Non-Small Cell Lung Cancer. Cancers (Basel) 2022;14:1987. [Crossref] [PubMed]
15. Liu J, Sui C, Bian H, et al. Radiomics based on (18)F-FDG PET/CT for prediction of pathological complete response to neoadjuvant therapy in non-small cell lung cancer. Front Oncol 2024;14:1425837. [Crossref] [PubMed]
16. Liu X, Ji Z, Zhang L, et al. Prediction of pathological complete response to neoadjuvant chemoimmunotherapy in non-small cell lung cancer using (18)F-FDG PET radiomics features of primary tumour and lymph nodes. BMC Cancer 2025;25:520. [Crossref] [PubMed]
17. Dall'Olio FG, Zrafi W, Roelants V, et al. Metabolic Tumor Volume Assessed by 18F FDG-PET CT Scan as a Predictive Biomarker for Immune Checkpoint Blockers in Advanced NSCLC and Its Biological Correlates. Clin Cancer Res 2025;31:352-64. [Crossref] [PubMed]
18. Vekens K, Everaert H, Neyns B, et al. The Value of (18)F-FDG PET/CT in Predicting the Response to PD-1 Blocking Immunotherapy in Advanced NSCLC Patients with High-Level PD-L1 Expression. Clin Lung Cancer 2021;22:432-40. [Crossref] [PubMed]
19. Cascone T, Weissferdt A, Godoy MCB, et al. Nodal immune flare mimics nodal disease progression following neoadjuvant immune checkpoint inhibitors in non-small cell lung cancer. Nat Commun 2021;12:5045. [Crossref] [PubMed]
20. Zhang L, E H, Huang J, et al. Clinical utility of [18F]FDG PET/CT in the assessment of mediastinal lymph node disease after neoadjuvant chemoimmunotherapy for nonsmall cell lung cancer. Eur Radiol 2023;33:8564-72. [Crossref] [PubMed]
21. Zhuang F, Haoran E, Huang J, et al. Utility of (18)F-FDG PET/CT uptake values in predicting response to neoadjuvant chemoimmunotherapy in resectable non-small cell lung cancer. Lung Cancer 2023;178:20-7. [Crossref] [PubMed]
22. Guo R, Yan W, Wang F, et al. The utility of (18)F-FDG PET/CT for predicting the pathological response and prognosis to neoadjuvant immunochemotherapy in resectable non-small-cell lung cancer. Cancer Imaging 2024;24:120. [Crossref] [PubMed]
23. Zhou Y, Zhai W, Sun W, et al. Safety and necessity of omitting mediastinal lymph node dissection in cN0/N1 non-small cell lung cancer after neoadjuvant immunotherapy. Front Immunol 2025;16:1587658. [Crossref] [PubMed]
24. Saddawi-Konefka R, Schokrpur S, Gutkind JS. Let it be: Preserving tumor-draining lymph nodes in the era of immuno-oncology. Cancer Cell 2024;42:930-3. [Crossref] [PubMed]
25. Huang Y, Lin X, Fang C, et al. Impact of lymph node dissection on survival after neoadjuvant immunochemotherapy for esophageal squamous cell cancer: a double-center real-world retrospective study. Cancer Immunol Immunother 2025;74:303. [Crossref] [PubMed]
26. Reck M, Gale D, Zhu Z, et al. Associations of ctDNA clearance (CL) during neoadjuvant Tx with pathological response and event-free survival (EFS) in pts with resectable NSCLC (R-NSCLC): Expanded analyses from AEGEAN. Ann Oncol 2024;35:S1239.