Early-stage Lung Cancer Research Articles

90 Effect of Treatment Delivery Schedule for Early-Stage Non-Small Cell Lung Cancer Patients Treated with Stereotactic Ablative Radiation Therapy: A Population-Based Analysis

90 Effect of Treatment Delivery Schedule for Early-Stage Non-Small Cell Lung Cancer Patients Treated with Stereotactic Ablative Radiation Therapy: A Population-Based Analysis

103 Predictive Models and Biomarkers for Post-Treatment Local Control Assessment in Early-Stage and Oligometastatic Lung Cancer Treated with Stereotactic Body Radiation Therapy, Using Radiomics and Machine Learning

103 Predictive Models and Biomarkers for Post-Treatment Local Control Assessment in Early-Stage and Oligometastatic Lung Cancer Treated with Stereotactic Body Radiation Therapy, Using Radiomics and Machine Learning

Effect of Treatment Delivery Schedule for Patients With Early-Stage Non-Small Cell Lung Cancer Treated With Stereotactic Ablative Radiation Therapy: A Population-Based Analysis

PurposeThe optimal stereotactic ablative body radiotherapy (SABR) delivery schedule in stage I non-small cell lung cancer (NSCLC) remains unclear. This population-based study investigated grade ≥ 2 toxicity rates, local failure (LF), and overall survival (OS) in patients treated with 48 Gy in 4 fractions scheduled every other day (QOD) versus daily with weekends (QDW) and consecutive daily without weekends (QD). Materials and MethodsBetween January 2019 and June 2022, treatment records using 48 Gy in 4 fractions were extracted from a provincial cancer registry and grouped by delivery as QOD, QDW, or QD. Toxicity events were recorded using CTCAE v5.0. The Kaplan-Meier method was used to compute OS while LF was calculated using cumulative incidence methods with death as a competing risk. Cox regression analyses and Fine-Gray modelling was used to assess for variables associated with OS and LF, respectively. ResultsOf 404 patients meeting study criteria, 190, 111, and 103 received QOD, QDW, and QD SABR, respectively. More patients receiving QDW SABR were medically inoperable and more patients receiving QD SABR had tumors abutting the chest wall. Median follow-up was 29.5 months (interquartile range [IQR] 19.2-38.4). Overall toxicity was low, with crude rates of acute and late grade ≥ 2 toxicity not being statistically different among the groups. No grade 4 or 5 toxicities were recorded. LF rates at 24 months were not different at 7.5% (95% confidence interval [CI] 3.7-11.3), 9.5% (95% CI 3.9-15.1), and 11.0% (95% CI 4.9-17.2) for the QOD, QDW, and QD groups, respectively (p = 0.60). QDW and QD schedules were not associated with LR. Similarly, no significant differences in median OS were found among the QOD, QDW, and QD groups at 47.5 months (95% CI 39.26-55,74), 52.7 (95% CI 34.7-70.7), and 49.0 months (95% CI 31.6-66.4), respectively. QDW and QD schedules were not associated with OS. ConclusionsThis population-based study demonstrated no statistically significant differences in grade ≥ 2 toxicity rates, LF, and OS for stage I NSCLC patients treated with lung SABR using 48 Gy in 4 fractions delivered QOD, QDW, and QD. Patient convenience and optimization of resources may be considered when choosing a lung SABR treatment delivery schedule.

A Phase II Study of Neoadjuvant Opnurasib KRAS G12C Inhibitor in Patients With Surgically Resectable Non-Small Cell Lung Cancer (CCTG IND.242A): A Substudy of the IND.242 Platform Master Protocol

Molecularly targeted agents are increasingly being studied in the treatment of early-stage non-small cell lung cancer (NSCLC) to try and improve cure. However, phase 3 data on neoadjuvant therapy have largely been conducted in a biomarker agnostic manner with inconsistent exclusion of EGFR and ALK alterations. Our objective was to develop IND.242 as a large-scale neoadjuvant platform trial to introduce novel agents into the preoperative window for molecularly-defined NSCLC patient populations. Given that KRAS G12C mutations are common in the overall NSCLC patient population, ranging from 9.4% to 13% of cases across different cohorts, and may be associated to worse prognosis, the initial IND.242A Substudy was designed to test neoadjuvant JDQ443 (opnurasib), a selective KRAS G12C inhibitor. This current trial report describes the multicenter, Canadian Cancer Trials Group (CCTG)-led IND.242 neoadjuvant phase 2 platform master protocol and its first Substudy (IND.242A) of neoadjuvant opnurasib KRAS G12C inhibitor for patients with surgically resectable NSCLC (AJCC 8th edition stage IA2 to IIIA). In IND.242A, a maximum of 27 patients will be accrued in participating Canadian sites. The primary objective is the rate of major pathological response (MPR) following neoadjuvant opnurasib. Secondary objectives include safety and tolerability of the treatment regimen, objective response rate (ORR) by RECIST 1.1 for the neoadjuvant treatment period, pathological complete response (pCR) rate, event-free survival (EFS) at 2 years, and surgical outcomes. Exploratory objectives are to explore patient related outcomes (PROs) and identify potential predictive biomarkers of response and mechanisms of resistance on tissue and peripheral blood samples.

4 Treatment Toxicity and Outcomes Following Definitive Radiotherapy for Patients with Early-Stage Non-Small Cell Lung Cancers and Pre-Existing Interstitial Lung Disease: A Systematic Review

4 Treatment Toxicity and Outcomes Following Definitive Radiotherapy for Patients with Early-Stage Non-Small Cell Lung Cancers and Pre-Existing Interstitial Lung Disease: A Systematic Review

1214P Imaging AI prognosis of early stage lung cancer using CT radiomics

1214P Imaging AI prognosis of early stage lung cancer using CT radiomics

103P Epigenetic regulated genes enhanced fragmentomics-based model for early-stage lung cancer detection

103P Epigenetic regulated genes enhanced fragmentomics-based model for early-stage lung cancer detection

A lung SBRT treatment planning technique to focus high dose on gross disease

This study investigated a straightforward treatment planning technique for definitive stereotactic body radiation therapy (SBRT) for patients with early-stage lung cancer aimed at increasing dose to gross disease by strategically penalizing the normal tissue objective (NTO) in the EclipseTM treatment planning system. Twenty-five SBRT cases were replanned to 50 Gy in 5 fractions using static and dynamic NTO methods (50 plans total). The NTO had a start dose of 100% at the target border, end dose of 20%, fall-off rate of 0.4/mm, and a priority of 150. For the static NTO plans, a lower planning target volume (PTV) objective was placed at 52 Gy with a priority of 100. Maximum dose was not penalized. Optimization was performed without user interaction. In contrast, the planner incrementally increased the priority of the NTO on the dynamic NTO plans until 95% of the target volume was covered by the prescription dose. Further, the dynamic NTO plans used both PTV lower and upper objectives at 63-64 Gy with priorities of 50. Maximum dose was penalized to ensure that the hot spot was within ± 2% of the static NTO global maximum dose. Following optimization, all plans were normalized so that the prescription dose covered 95% of the PTV. Plans were scored based on RTOG 0813 criteria, and dose to the internal target volume (ITV) and PTV was evaluated. The Wilcoxon signed-rank test (threshold = 0.05) was used to evaluate differences between the static and dynamic NTO plans. All plans met RTOG 0813 planning guidelines. In comparison to the static NTO plans, the dynamic NTO plans exhibited statistically significant increases in PTV mean dose, ITV mean dose, and PTV-ITV mean dose. Notably, the dynamic NTO plans more effectively concentrated the high dose on gross disease at the center of the PTV. As compared to the static NTO plans, the mean dose was 4.6 Gy higher in the ITV while only 1.3 Gy higher in the PTV-ITV rind of the dynamic NTO plans. Global maximum doses were similar. There were some small yet statistically significant differences in dose conformity between plan types. Furthermore, the dynamic NTO plans demonstrated a significant reduction in total monitor units (MU). This study demonstrated an efficient optimization strategy for lung SBRT plans that concentrates the highest dose in the gross disease, which may improve local control.

Impact of imaging features on selecting limited lymph node resection for cT1N0M0 lung cancer.

Controversy still exists in the medical community regarding the performance of limited mediastinal lymphadenectomy (LML) in early-stage lung cancer. The objective of this study was to identify predictors of mediastinal lymph node (mLN) status and analyze its role in guiding surgical strategy. A retrospective cohort study was conducted on 2,834 surgical patients with peripheral cT1N0M0 non-small cell lung cancer between 2016 and 2018. Logistic regression was employed to identify predictors of N2 metastasis. Prognosis was compared between groups and independent prognostic factors were identified using Kaplan-Meier and multivariate Cox analysis. There were 2,126 patients with systematic mLN dissection and 708 with LML. The multivariate analysis showed that N2 metastasis were associated with tumor size and consolidation tumor ratio (CTR). Patients in group A, with CTR >0.5 and tumor size ≤1 cm or CTR ≤0.5, had a significantly lower rate of N2 metastasis compared to those in group B, with CTR >0.5 and tumor size >1 cm (14.2% vs. 0.2%, P<0.001). Additionally, LML demonstrated comparable recurrence-free survival (RFS) and overall survival (OS) in group A, but a worse prognosis in group B compared to systematic lymph node dissection (SND). Furthermore, multivariate Cox regression analysis indicated that SND (vs. LML) was a favorable prognostic predictor for patients in group B [RFS: hazard ratio (HR) =0.71, P=0.005; OS: HR =0.66, P=0.01]. But univariate analysis in group A showed no significant difference in prognosis between SND and LML (RFS: P=0.24; OS: P=0.10). The combination of CTR and tumor size can predict mLN metastasis and procedure-specific outcome (SND vs. LML). This information may assist surgeons in identifying suitable candidates for LML.

The links between symptom burden, illness perception, psychological resilience, social support, coping modes, and cancer-related worry in Chinese early-stage lung cancer patients after surgery: a cross-sectional study

ObjectivesThis study aims to investigate the links between the clinical, demographic, and psychosocial factors and cancer-related worry in patients with early-stage lung cancer after surgery.MethodsThe study utilized a descriptive cross-sectional design. Questionnaires, including assessments of cancer-related worry, symptom burden, illness perception, psychological resilience, coping modes, social support and participant characteristics, were distributed to 302 individuals in early-stage lung cancer patients after surgery. The data collection period spanned from January and October 2023. Analytical procedures encompassed descriptive statistics, independent Wilcoxon Rank Sum test, Kruskal-Wallis- H- test, Spearman correlation analysis, and hierarchical multiple regression.ResultsAfter surgery, 89.07% had cancer-related worries, with a median (interquartile range, IQR) CRW score of 380.00 (130.00, 720.00). The most frequently cited concern was the cancer itself (80.46%), while sexual issues were the least worrisome (44.37%). Regression analyses controlling for demographic variables showed that higher levels of cancer-related worry (CRW) were associated with increased symptom burden, illness perceptions, and acceptance-rejection coping modes, whereas they had lower levels of psychological resilience, social support and confrontation coping modes, and were more willing to obtain information about the disease from the Internet or applications. Among these factors, the greatest explanatory power in the regression was observed for symptom burden, illness perceptions, social support, and sources of illness information (from the Internet or applications), which collectively explained 52.00% of the variance.ConclusionsHealthcare providers should be aware that worry is a common issue for early stage lung cancer survivors with a favorable prognosis. During post-operative recovery, physicians should identify patient concerns and address unmet needs to improve patients’ emotional state and quality of life through psychological support and disease education.

Contribution of fluorescence imaging to thoracoscopic anatomical segmentectomy: a multicenter propensity matching analysis.

Thoracoscopic anatomical segmentectomy is increasingly recognized for managing early-stage lung cancer. However accurately identifying intersegmental planes (ISPs), especially in complex lung segments, remains challenging. In comparison to conventional methods, fluorescence imaging represents a novel solution. This study aimed to examine the potential benefits of fluorescence imaging in single-port thoracoscopic anatomical segmentectomy. A multicenter (three regional hospitals), retrospective, comparative analysis was conducted using data from 402 consecutive patients who underwent single-port thoracoscopic anatomical segmentectomy from June 2020 to December 2022. The cohort included 191 patients treated with the fluorescence method and 211 patients treated with the modified inflation-deflation method. Among the cohort, 130 patients were placed in the simple segmentectomy group and 272 in the complex segmentectomy group. Propensity score matching (PSM) was used to adjust for baseline differences between the fluorescence and modified inflation-deflation subgroups in the complex segmentectomy group. Perioperative outcomes were compared between the groups. In the simple segmentectomy group, no significant differences were observed between the fluorescence and modified inflation-deflation methods regarding segmental resection time, intraoperative blood loss, postoperative chest tube drainage and duration, postoperative pain, length of hospital stay, complication rate, or hospital costs. In the complex segmentectomy group, however, fluorescence imaging significantly shortened segmental resection time (69.37±28.22 vs. 78.80±34.66 min; P=0.03), while reducing intraoperative blood loss (P=0.046); and improving visual analogue scale (VAS) pain scores on the first postoperative day (P=0.006). Both methods demonstrated comparable safety and oncologic effectiveness. Fluorescence-guided single-port thoracoscopic anatomical segmentectomy demonstrated comparable perioperative safety and effectiveness to the modified inflation-deflation technique while offering advantages, such as shorter segmental resection time, for complex segmentectomies.

Stereotactic body radiotherapy in ultracentrally located lung tumors

Stereotactic body radiotherapy (SBRT) is a highly conformal radiotherapy technique that delivers high doses of radiation per fraction, typically in 1 &amp;ndash; 5 fractions. It is a standard treatment option for patients with early-stage non-small cell lung cancer that is peripherally located, for those who are medically inoperable, or for patients who do not opt for surgery. Similarly, SBRT is considered an effective treatment option for oligometastatic lung tumors. However, in the case of ultracentrally located lung tumors&amp;mdash;tumors situated close to the central airway and mediastinal region&amp;mdash;there is an increased risk of severe toxicities (grade 4 &amp;ndash; 5) due to the dose tolerances of the surrounding organs at risk. The definition of ultracentrally located lung tumors remains unclear. Some studies define an ultracentral lung tumor as one in which the gross tumor volume directly involves the proximal bronchial tree or trachea, while others define it as the planning target volume (PTV) that includes the trachea or main bronchial system. In addition, some studies consider tumors whose PTV touches or overlaps with the central bronchial tree, esophagus, pulmonary vein, or pulmonary artery as ultracentral. In addition, the optimal SBRT regime and organ-at-risk dose limitations for ultracentral tumors have yet to be clearly established. The aim of this review is to explore and discuss the role of SBRT in the treatment of ultracentrally located lung tumors.

Live-Cell Invasive Phenotyping Uncovers ALK2 as a Therapeutic Target in LKB1-Mutant Lung Cancer.

The acquisition of invasive properties is a prerequisite for tumor progression and metastasis. Molecular subtypes of KRAS-driven lung cancer exhibit distinct modes of invasion that contribute to unique growth properties and therapeutic susceptibilities. Despite this, preclinical strategies designed to exploit growth within the context of invasion are lacking. To address this, we designed an experimental system to screen for targetable signaling pathways linked to active early 3D invasion phenotypes in different molecular subtypes of KRAS-driven lung adenocarcinoma. Combined live-cell imaging of human bronchial epithelial cells in a 3D invasion matrix and transcriptomic profiling identified mutant LKB1-specific upregulation of BMP6. LKB1 loss increased BMP6 signaling, which induced the canonical iron regulatory hormone hepcidin. Intact LKB1 was necessary to maintain BMP6 signaling homeostasis and restrict ALK2/BMP6-fueled growth. Preclinical studies in a Kras/Lkb1-mutant syngeneic mouse model and in a xenograft model showed potent growth suppression by inhibiting the ALK2/BMP6 signaling axis with single-agent inhibitors that are currently in clinical trials. Lastly, BMP6 expression was elevated in tumors of patients with LKB1-mutant early-stage lung cancer. These results are consistent with those of a model in which LKB1 acts as a "brake" to iron-regulated growth and suggest that ALK2 inhibition can be used for patients with LKB1-mutant tumors. Significance: Three-dimensional invasion-linked gene expression analysis reveals a therapeutic vulnerability to inhibition of ALK2/BMP6 signaling in LKB1-mutant lung cancer that can be rapidly translated to the clinic.

Stereotactic Ablative Radiotherapy for Early-stage Lung Cancer in Patients With Left Ventricular Assist Device: A Case Series.

Radiotherapy-induced malfunction of pacemakers and cardiac defibrillators has been reported, and corresponding guidelines have been developed in various countries. Although several studies have reported the effects of radiotherapy in patients with implantable left ventricular assist device (LVAD), its safety remains unclear. Herein, we report three cases of stereotactic ablative radiotherapy (SABR) using CyberKnife for early-stage lung cancer in patients with implantable LVAD. Three patients in their 50s or 60s, including two women and one man, who had LVADs due to dilated or ischemic cardiomyopathy and performance status of 0 or 1, were diagnosed with stage IA2 lung cancer (cT1bN0M0) by imaging only. All three patients were deemed inoperable due to cardiac comorbidity and underwent SABR at the Osaka University Hospital. The total radiation dose was 42-52 Gy, administered in four fractions. All treatment plans were designed to keep the LVAD dose below 2 Gy. In all patients, SABR was completed without acute adverse events or LVAD malfunction. During the follow-up period of 3-29 months, no disease progression or chronic adverse events were observed in any of the patients. This case series indicated that SABR using CyberKnife is a safe treatment option for early-stage lung cancer in patients with LVAD by reducing the dose to the LVAD.

Early-Stage Lung Cancer in the Era of Lung Cancer Screening: Veterans Health Administration Outperforms Other Insurance Models

BackgroundLung cancer screening guidelines were introduced in the United States in 2013, with variable implementation. This study evaluated temporal diagnostic trends in non-small cell lung cancer (NSCLC) diagnosis since the introduction of these guidelines. MethodsThis retrospective cohort analysis used data from the Veterans Administration Corporate Data Warehouse and the National Cancer Database. We evaluated temporal trends in the distribution of NSCLC stage at the time of diagnosis along with differences based on insurance coverage type, including uninsured, privately insured, Medicare, Medicaid, and Veterans Affairs (VA) coverage, with adjustment for clinically relevant variables. ResultsAmong 1,450,965 patients diagnosed from 2006 to 2020, the proportion of NSCLC cases diagnosed at stage I increased in all insurance groups by 12.74%, 2%, 0.25%, and 2.57% for the VA, Medicare, private insurance, and Medicaid, respectively. If all insurance systems achieved the unadjusted stage distribution seen in the Veterans Health Administration, an additional 45,684 patients would be diagnosed with stage I NSCLC and 65,933 fewer patients would be diagnosed with stage IV disease. ConclusionsFor patients with any form of insurance, there has been an increase in the proportion of early-stage NSCLC (stage I and II) and a corresponding decrease in the proportion of stage III and IV since the introduction of national lung cancer screening guidelines. As the largest integrated single-payer health care system in the United States, the VA dramatically outperforms other insurance types, perhaps attributable to universal coverage and robust lung cancer screening programs.

Circulating cell-free and extracellular vesicles-derived microRNA as prognostic biomarkers in patients with early-stage NSCLC: results from RESTING study

BackgroundFactors to accurately stratify patients with early-stage non-small cell lung cancer (NSCLC) in different prognostic groups are still needed. This study aims to investigate 1) the prognostic potential of circulating cell-free (CF) and extracellular vesicles (EVs)-derived microRNA (miRNAs), and 2) their added value with respect to known prognostic factors (PFs).MethodsThe RESTING study is a multicentre prospective observational cohort study on resected stage IA-IIIA patients with NSCLC. The primary end-point was disease-free survival (DFS), and the main analyses were carried out separately for CF- and EV-miRNAs. CF- and EV-miRNAs were isolated from plasma, and miRNA-specific libraries were prepared and sequenced. To reach the study aims, three statistical models were specified: one using the miRNA data only (Model 1); one using both miRNAs and known PFs (age, gender, and pathological stage) (Model 2), and one using the PFs alone (Model 3). Five-fold cross-validation (CV) was used to assess the predictive performance of each. Standard Cox regression and elastic net regularized Cox regression were used.ResultsA total of 222 patients were enrolled. The median follow-up time was 26.3 (95% CI 25.4–27.6) months. From Model 1, three CF-miRNAs and 21 EV-miRNAs were associated with DFS. In Model 2, two CF-miRNAs (miR-29c-3p and miR-877-3p) and five EV-miRNAs (miR-181a-2-3p, miR-182-5p, miR-192-5p, miR-532-3p and miR-589-5p) remained associated with DFS. From pathway enrichment analysis, TGF-beta and NOTCH were the most involved pathways.ConclusionThis study identified promising prognostic CF- and EV-miRNAs that could be used as a non-invasive, cost-effective tool to aid clinical decision-making. However, further evaluation of the obtained miRNAs in an external cohort of patients is warranted.

Retrospective evaluation of plasma protein tumour markers for early lung cancer detection

BackgroundBlood-based biomarkers might help lung cancer diagnosis. A panel of serum tumour markers (TM) has been validated for hospital referrals due to clinical suspicion of lung cancer. We have compared plasma from a cohort enriched for early-stage lung cancer, including controls from a healthy population cohort.MethodsTM assays for CEA, CYFRA 21-1, CA15.3, ProGRP and SCC were run on a Roche Elecsys 2010 Immunoassay Analyser for a retrospective, nested case-control cohort from the Liverpool Lung Project. The primary endpoints were the sensitivity and specificity of a pre-defined TM panel using published thresholds.ResultsExcept for ProGRP, TM levels were significantly higher in cases and ROC AUC values demonstrated significant discriminant power. Accuracy and levels were higher for late-stage cancers, except for ProGRP which was highest in stage 1 cancers. Although providing similar sensitivity (82.4% vs 88.5%), TMs performed worse for specificity (39.3% vs 82%) and overall (Youden’s Index 0.22 vs 0.77) and this was not improved by threshold optimisation or binary logistic regression.ConclusionsAlthough the TMs were associated with lung cancer status and discriminatory with a high sensitivity when combined, performance was compromised in early-stage disease, which casts some doubt on utility in the screening setting.

A new vessel filled and heart-beating human corpse model for VATS lobectomy training

BackgroundNowadays, video-assisted thoracic surgery (VATS) lobectomy represents the treatment of choice for early-stage lung cancer. Over the years, different methods for VATS training have evolved. The aim of this study is to present an innovative beating-heart filled-vessel cadaveric model to simulate VATS lobectomies.MethodsVia selective cannulation of the cadaver heart, the pulmonary vessels were filled with a gel to improve their haptic feedback. An endotracheal tube with a balloon on its tip then allowed movement of the heart chambers, transmitting a minimum of flow to the pulmonary vessels. A simulated OR was created, using all instrumentation normally available during surgery on living patients, with trainees constantly mentored by experienced surgeons. At the end of each simulation, the participants were asked 5 questions on a scale of 1 to 10 to evaluate the effectiveness of the training method (“1” being ineffective and “10” being highly effective).ResultsEight models were set up, each with a median time of 108 min and a cost of €1500. Overall, 50 surgeons were involved, of which 39 (78%) were consultants and 11 (22%) were residents (PGY 3–5). The median scores for the 5 questions were 8.5 (Q1; IQR1–3 8-9), 8 (Q2; IQR1–3 7–9), 9 (Q3; IQR1–3 8–10), 9 (Q4; IQR1–3 8–10), and 9 (Q5; IQR1–3 8–10). Overall, the model was most appreciated by young trainees even though positive responses were also provided by senior surgeons.ConclusionsWe introduce a new beating-heart filled-vessel cadaveric model to simulate VATS lobectomies. From this initial experience, the model is cost effective, smooth to develop, and realistic for VATS simulation.

Preoperative early-stage lung cancer patients and local brain area changes: a cross-sectional observational descriptive study.

Lung cancer is a major global health concern. Patients undergo a substantial process of emotional transformation following a lung cancer diagnosis, during which subtle changes in brain function and/or structure may occur. As such, the present study aimed to investigate the neuroplastic changes induced by negative emotions in patients with early-stage lung cancer. This cross-sectional study recruited 35 patients with early-stage lung cancer and 33 age- and sex-matched healthy control patients. All participants completed the Hamilton Anxiety Rating Scale (HAMA), Hamilton Depression Rating Scale (HAMD), and functional magnetic resonance imaging (fMRI). Amplitude of low-frequency fluctuations (ALFF) and regional homogeneity (ReHo) were used as the fMRI indices. Correlations between the clinical assessments and ALFF and ReHo values were calculated. Our analysis revealed no significant differences in HAMD and HAMA scores between patients and control patients (p > 0.05). However, significant alterations in ALFF and ReHo were observed in multiple brain regions in patients with early-stage lung cancer compared to healthy controls (P FalseDiscoveryRate < 0.05). Specifically, ALFF values were decreased in the right postcentral gyrus, calcarine, and left middle cingulate, while ReHo values increased in the right angular gyrus and decreased in the bilateral postcentral gyrus, insula, left calcarine, putamen, superior temporal gyrus, middle cingulate, and right Rolandic gyrus. The HAMD score was significantly correlated with the ALFF value in the right postcentral gyrus (P = 0.007). This study provides valuable insights into the adaptive responses of the brain following the early diagnosis of lung cancer, revealing potential disturbances in negative emotional processing. Harnessing neuroplasticity may open new avenues for the establishment of personalized treatment strategies and targeted interventions to support the emotional and mental health of patients with lung cancer.

A dual-radiomics model for overall survival prediction in early-stage NSCLC patient using pre-treatment CT images.

Radiation therapy (RT) is one of the primary treatment options for early-stage non-small cell lung cancer (ES-NSCLC). Therefore, accurately predicting the overall survival (OS) rate following radiotherapy is crucial for implementing personalized treatment strategies. This work aims to develop a dual-radiomics (DR) model to (1) predict 3-year OS in ES-NSCLC patients receiving RT using pre-treatment CT images, and (2) provide explanations between feature importanceand model prediction performance. The publicly available TCIA Lung1 dataset with 132 ES-NSCLC patients received RT were studied: 89/43 patients in the under/over 3-year OS group. For each patient, two types of radiomic features were examined: 56 handcrafted radiomic features (HRFs) extracted within gross tumor volume, and 512 image deep features (IDFs) extracted using a pre-trained U-Net encoder. They were combined as inputs to an explainable boosting machine (EBM) model for OS prediction. The EBM's mean absolute scores for HRFs and IDFs were used as feature importance explanations. To evaluate identified feature importance, the DR model was compared with EBM using either (1) key or (2) non-key feature type only. Comparison studies with other models, including supporting vector machine (SVM) and random forest (RF), were also included. The performance was evaluated by the area under the receiver operating characteristic curve (AUCROC), accuracy, sensitivity, and specificity with a 100-fold Monte Carlo cross-validation. The DR model showed highestperformance in predicting 3-year OS (AUCROC=0.81 ± 0.04), and EBM scores suggested that IDFs showed significantly greater importance (normalized mean score=0.0019) than HRFs (score=0.0008). The comparison studies showed that EBM with key feature type (IDFs-only demonstrated comparable AUCROC results (0.81 ± 0.04), while EBM with non-key feature type (HRFs-only) showed limited AUCROC (0.64 ± 0.10). The results suggested that feature importance score identified by EBM is highly correlated with OS prediction performance. Both SVM and RF models were unable to explain key feature type while showing limited overall AUCROC=0.66 ± 0.07 and 0.77 ± 0.06, respectively. Accuracy, sensitivity, and specificity showed a similar trend. In conclusion, a DR model was successfully developed to predict ES-NSCLC OS based on pre-treatment CT images. The results suggested that the feature importance from DR model is highly correlated to the model prediction power.