Clinical Utility of a CT-based AI Prognostic Model for Segmentectomy in Non-Small Cell Lung Cancer.

Background Currently, no tool exists for risk stratification in patients undergoing segmentectomy for non-small cell lung cancer (NSCLC). Purpose To develop and validate a deep learning (DL) prognostic model using preoperative CT scans and clinical and radiologic information for risk stratification in patients with clinical stage IA NSCLC undergoing segmentectomy. Materials and Methods In this single-center retrospective study, transfer learning of a pretrained model was performed for survival prediction in patients with clinical stage IA NSCLC who underwent lobectomy from January 2008 to March 2017. The internal set was divided into training, validation, and testing sets based on the assignments from the pretraining set. The model was tested on an independent test set of patients with clinical stage IA NSCLC who underwent segmentectomy from January 2010 to December 2017. Its prognostic performance was analyzed using the time-dependent area under the receiver operating characteristic curve (AUC), sensitivity, and specificity for freedom from recurrence (FFR) at 2 and 4 years and lung cancer-specific survival and overall survival at 4 and 6 years. The model sensitivity and specificity were compared with those of the Japan Clinical Oncology Group (JCOG) eligibility criteria for sublobar resection. Results The pretraining set included 1756 patients. Transfer learning was performed in an internal set of 730 patients (median age, 63 years [IQR, 56-70 years]; 366 male), and the segmentectomy test set included 222 patients (median age, 65 years [IQR, 58-71 years]; 114 male). The model performance for 2-year FFR was as follows: AUC, 0.86 (95% CI: 0.76, 0.96); sensitivity, 87.4% (7.17 of 8.21 patients; 95% CI: 59.4, 100); and specificity, 66.7% (136 of 204 patients; 95% CI: 60.2, 72.8). The model showed higher sensitivity for FFR than the JCOG criteria (87.4% vs 37.6% [3.08 of 8.21 patients], P = .02), with similar specificity. Conclusion The CT-based DL model identified patients at high risk among those with clinical stage IA NSCLC who underwent segmentectomy, outperforming the JCOG criteria. © RSNA, 2024 Supplemental material is available for this article.

Enhancing NSCLC recurrence prediction with PET/CT habitat imaging, ctDNA, and integrative radiogenomics-blood insights.

While we recognize the prognostic importance of clinicopathological measures and circulating tumor DNA (ctDNA), the independent contribution of quantitative image markers to prognosis in non-small cell lung cancer (NSCLC) remains underexplored. In our multi-institutional study of 394 NSCLC patients, we utilize pre-treatment computed tomography (CT) and 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) to establish a habitat imaging framework for assessing regional heterogeneity within individual tumors. This framework identifies three PET/CT subtypes, which maintain prognostic value after adjusting for clinicopathologic risk factors including tumor volume. Additionally, these subtypes complement ctDNA in predicting disease recurrence. Radiogenomics analysis unveil the molecular underpinnings of these imaging subtypes, highlighting downregulation in interferon alpha and gamma pathways in the high-risk subtype. In summary, our study demonstrates that these habitat imaging subtypes effectively stratify NSCLC patients based on their risk levels for disease recurrence after initial curative surgery or radiotherapy, providing valuable insights for personalized treatment approaches.

Preclinical development of novel PD-L1 tracers and first-in-human study of [68Ga]Ga-NOTA-RW102 in patients with lung cancers.

BACKGROUND: The programmed cell death protein-1 (PD-1)/programmed death receptor ligand 1 (PD-L1) axis critically facilitates cancer cells' immune evasion. Antibody therapeutics targeting the PD-1/PD-L1 axis have shown remarkable efficacy in various tumors. Immuno-positron emission tomography (ImmunoPET) imaging of PD-L1 expression may help reshape solid tumors' immunotherapy landscape. METHODS: By immunizing an alpaca with recombinant human PD-L1, three clones of the variable domain of the heavy chain of heavy-chain only antibody (VHH) were screened, and RW102 with high binding affinity was selected for further studies. ABDRW102, a VHH derivative, was further engineered by fusing RW102 with the albumin binder ABD035. Based on the two targeting vectors, four PD-L1-specific tracers ([68Ga]Ga-NOTA-RW102, [68Ga]Ga-NOTA-ABDRW102, [64Cu]Cu-NOTA-ABDRW102, and [89Zr]Zr-DFO-ABDRW102) with different circulation times were developed. The diagnostic efficacies were thoroughly evaluated in preclinical solid tumor models, followed by a first-in-human translational investigation of [68Ga]Ga-NOTA-RW102 in patients with non-small cell lung cancer (NSCLC). RESULTS: While RW102 has a high binding affinity to PD-L1 with an excellent KD value of 15.29 pM, ABDRW102 simultaneously binds to human PD-L1 and human serum albumin with an excellent KD value of 3.71 pM and 3.38 pM, respectively. Radiotracers derived from RW102 and ABDRW102 have different in vivo circulation times. In preclinical studies, [68Ga]Ga-NOTA-RW102 immunoPET imaging allowed same-day annotation of differential PD-L1 expression with specificity, while [64Cu]Cu-NOTA-ABDRW102 and [89Zr]Zr-DFO-ABDRW102 enabled longitudinal visualization of PD-L1. More importantly, a pilot clinical trial shows the safety and diagnostic value of [68Ga]Ga-NOTA-RW102 immunoPET imaging in patients with NSCLCs and its potential to predict immune-related adverse effects following PD-L1-targeted immunotherapies. CONCLUSIONS: We developed and validated a series of PD-L1-targeted tracers. Initial preclinical and clinical evidence indicates that immunoPET imaging with [68Ga]Ga-NOTA-RW102 holds promise in visualizing differential PD-L1 expression, selecting patients for PD-L1-targeted immunotherapies, and monitoring immune-related adverse effects in patients receiving PD-L1-targeted treatments. TRIAL REGISTRATION NUMBER: NCT06165874.

Preoperative nomogram for predicting spread through air spaces in clinical-stage IA non-small cell lung cancer using 18F-fluorodeoxyglucose positron emission tomography/computed tomography.

PURPOSE: This study aims to assess the predictive value of 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) radiological features and the maximum standardized uptake value (SUVmax) in determining the presence of spread through air spaces (STAS) in clinical-stage IA non-small cell lung cancer (NSCLC). METHODS: A retrospective analysis was conducted on 180 cases of NSCLC with postoperative pathological assessment of STAS status, spanning from September 2019 to September 2023. Of these, 116 cases from hospital one comprised the training set, while 64 cases from hospital two formed the testing set. The clinical information, tumor SUVmax, and 13 related CT features were analyzed. Subgroup analysis was carried out based on tumor density type. In the training set, univariable and multivariable logistic regression analyses were employed to identify the most significant variables. A multivariable logistic regression model was constructed and the corresponding nomogram was developed to predict STAS in NSCLC, and its diagnostic efficacy was evaluated in the testing set. RESULTS: SUVmax, consolidation-to-tumor ratio (CTR), and lobulation sign emerged as the best combination of variables for predicting STAS in NSCLC. Among these, SUVmax and CTR were identified as independent predictors for STAS prediction. The constructed prediction model demonstrated area under the curve (AUC) values of 0.796 and 0.821 in the training and testing sets, respectively. Subgroup analysis revealed a 2.69 times higher STAS-positive rate in solid nodules compared to part-solid nodules. SUVmax was an independent predictor for predicting STAS in solid nodular NSCLC, while CTR and an emphysema background were independent predictors for STAS in part-solid nodular NSCLC. CONCLUSION: Our nomogram based on preoperative 18F-FDG PET/CT radiological features and SUVmax effectively predicts STAS status in clinical-stage IA NSCLC. Furthermore, our study highlights that metabolic parameters and CT variables associated with STAS differ between solid and part-solid nodular NSCLC.

Non-invasive prediction for pathologic complete response to neoadjuvant chemoimmunotherapy in lung cancer using CT-based deep learning: a multicenter study.

Neoadjuvant chemoimmunotherapy has revolutionized the therapeutic strategy for non-small cell lung cancer (NSCLC), and identifying candidates likely responding to this advanced treatment is of important clinical significance. The current multi-institutional study aims to develop a deep learning model to predict pathologic complete response (pCR) to neoadjuvant immunotherapy in NSCLC based on computed tomography (CT) imaging and further prob the biologic foundation of the proposed deep learning signature. A total of 248 participants administrated with neoadjuvant immunotherapy followed by surgery for NSCLC at Ruijin Hospital, Ningbo Hwamei Hospital, and Affiliated Hospital of Zunyi Medical University from January 2019 to September 2023 were enrolled. The imaging data within 2 weeks prior to neoadjuvant chemoimmunotherapy were retrospectively extracted. Patients from Ruijin Hospital were grouped as the training set (n = 104) and the validation set (n = 69) at the 6:4 ratio, and other participants from Ningbo Hwamei Hospital and Affiliated Hospital of Zunyi Medical University served as an external cohort (n = 75). For the entire population, pCR was obtained in 29.4% (n = 73) of cases. The areas under the curve (AUCs) of our deep learning signature for pCR prediction were 0.775 (95% confidence interval [CI]: 0.649 - 0.901) and 0.743 (95% CI: 0.618 - 0.869) in the validation set and the external cohort, significantly superior than 0.579 (95% CI: 0.468 - 0.689) and 0.569 (95% CI: 0.454 - 0.683) of the clinical model. Furthermore, higher deep learning scores correlated to the upregulation for pathways of cell metabolism and more antitumor immune infiltration in microenvironment. Our developed deep learning model is capable of predicting pCR to neoadjuvant chemoimmunotherapy in patients with NSCLC.

Machine learning for differentiating lung squamous cell cancer from adenocarcinoma using Clinical-Metabolic characteristics and 18F-FDG PET/CT radiomics.

Noninvasive differentiation between the squamous cell carcinoma (SCC) and adenocarcinoma (ADC) subtypes of non-small cell lung cancer (NSCLC) could benefit patients who are unsuitable for invasive diagnostic procedures. Therefore, this study evaluates the predictive performance of a PET/CT-based radiomics model. It aims to distinguish between the histological subtypes of lung adenocarcinoma and squamous cell carcinoma, employing four different machine learning techniques. A total of 255 Non-Small Cell Lung Cancer (NSCLC) patients were retrospectively analyzed and randomly divided into the training (n = 177) and validation (n = 78) sets, respectively. Radiomics features were extracted, and the Least Absolute Shrinkage and Selection Operator (LASSO) method was employed for feature selection. Subsequently, models were constructed using four distinct machine learning techniques, with the top-performing algorithm determined by evaluating metrics such as accuracy, sensitivity, specificity, and the area under the curve (AUC). The efficacy of the various models was appraised and compared using the DeLong test. A nomogram was developed based on the model with the best predictive efficiency and clinical utility, and it was validated using calibration curves. Results indicated that the logistic regression classifier had better predictive power in the validation cohort of the radiomic model. The combined model (AUC 0.870) exhibited superior predictive power compared to the clinical model (AUC 0.848) and the radiomics model (AUC 0.774). In this study, we discovered that the combined model, refined by the logistic regression classifier, exhibited the most effective performance in classifying the histological subtypes of NSCLC.

Identification of CT radiomic features robust to acquisition and segmentation variations for improved prediction of radiotherapy-treated lung cancer patient recurrence.

The primary objective of the present study was to identify a subset of radiomic features extracted from primary tumor imaged by computed tomography of early-stage non-small cell lung cancer patients, which remain unaffected by variations in segmentation quality and in computed tomography image acquisition protocol. The robustness of these features to segmentation variations was assessed by analyzing the correlation of feature values extracted from lesion volumes delineated by two annotators. The robustness to variations in acquisition protocol was evaluated by examining the correlation of features extracted from high-dose and low-dose computed tomography scans, both of which were acquired for each patient as part of the stereotactic body radiotherapy planning process. Among 106 radiomic features considered, 21 were identified as robust. An analysis including univariate and multivariate assessments was subsequently conducted to estimate the predictive performance of these robust features on the outcome of early-stage non-small cell lung cancer patients treated with stereotactic body radiation therapy. The univariate predictive analysis revealed that robust features demonstrated superior predictive potential compared to non-robust features. The multivariate analysis indicated that linear regression models built with robust features displayed greater generalization capabilities by outperforming other models in predicting the outcomes of an external validation dataset.

A distributed feature selection pipeline for survival analysis using radiomics in non-small cell lung cancer patients.

Predictive modelling of cancer outcomes using radiomics faces dimensionality problems and data limitations, as radiomics features often number in the hundreds, and multi-institutional data sharing is ()often unfeasible. Federated learning (FL) and feature selection (FS) techniques combined can help overcome these issues, as one provides the means of training models without exchanging sensitive data, while the other identifies the most informative features, reduces overfitting, and improves model interpretability. Our proposed FS pipeline based on FL principles targets data-driven radiomics FS in a multivariate survival study of non-small cell lung cancer patients. The pipeline was run across datasets from three institutions without patient-level data exchange. It includes two FS techniques, Correlation-based Feature Selection and LASSO regularization, and Cox Proportional-Hazard regression with Overall Survival as endpoint. Trained and validated on 828 patients overall, our pipeline yielded a radiomic signature comprising "intensity-based energy" and "mean discretised intensity". Validation resulted in a mean Harrell C-index of 0.59, showcasing fair efficacy in risk stratification. In conclusion, we suggest a distributed radiomics approach that incorporates preliminary feature selection to systematically decrease the feature set based on data-driven considerations. This aims to address dimensionality challenges beyond those associated with data constraints and interpretability concerns.

A comparison between centralized and asynchronous federated learning approaches for survival outcome prediction using clinical and PET data from non-small cell lung cancer patients.

BACKGROUND AND OBJECTIVE: Survival analysis plays an essential role in the medical field for optimal treatment decision-making. Recently, survival analysis based on the deep learning (DL) approach has been proposed and is demonstrating promising results. However, developing an ideal prediction model requires integrating large datasets across multiple institutions, which poses challenges concerning medical data privacy. METHODS: In this paper, we propose FedSurv, an asynchronous federated learning (FL) framework designed to predict survival time using clinical information and positron emission tomography (PET)-based features. This study used two datasets: a public radiogenic dataset of non-small cell lung cancer (NSCLC) from the Cancer Imaging Archive (RNSCLC), and an in-house dataset from the Chonnam National University Hwasun Hospital (CNUHH) in South Korea, consisting of clinical risk factors and F-18 fluorodeoxyglucose (FDG) PET images in NSCLC patients. Initially, each dataset was divided into multiple clients according to histological attributes, and each client was trained using the proposed DL model to predict individual survival time. The FL framework collected weights and parameters from the clients, which were then incorporated into the global model. Finally, the global model aggregated all weights and parameters and redistributed the updated model weights to each client. We evaluated different frameworks including single-client-based approach, centralized learning and FL. RESULTS: We evaluated our method on two independent datasets. First, on the RNSCLC dataset, the mean absolute error (MAE) was 490.80±22.95 d and the C-Index was 0.69±0.01. Second, on the CNUHH dataset, the MAE was 494.25±40.16 d and the C-Index was 0.71±0.01. The FL approach achieved centralized method performance in PET-based survival time prediction and outperformed single-client-based approaches. CONCLUSIONS: Our results demonstrated the feasibility and effectiveness of employing FL for individual survival prediction in NSCLC patients, using clinical information and PET-based features.

68Ga-FAPI-04 PET/CT in Non-Small Cell Lung Cancer: Accurate Evaluation of Lymph Node Metastasis and Correlation with Fibroblast Activation Protein Expression.

Fibroblast activation protein (FAP) is a promising diagnostic and therapeutic target in various solid tumors. This study aimed to assess the diagnostic efficiency of 68Ga-labeled FAP inhibitor (FAPI)-04 PET/CT for detecting lymph node metastasis in non-small cell lung cancer (NSCLC) and to investigate the correlation between tumor 68Ga-FAPI-04 uptake and FAP expression. Methods: We retrospectively enrolled 136 participants with suspected or biopsy-confirmed NSCLC who underwent 68Ga-FAPI-04 PET/CT for initial staging. The diagnostic performance of 68Ga-FAPI-04 for the detection of NSCLC was evaluated. The final histopathology or typical imaging features were used as the reference standard. The SUVmax and SUVmean, 68Ga-FAPI-avid tumor volume (FTV), and total lesion FAP expression (TLF) were measured and calculated. FAP immunostaining of tissue specimens was performed. The correlation between 68Ga-FAPI-04 uptake and FAP expression was assessed using the Spearman correlation coefficient. Results: Ninety-one participants (median age, 65 y [interquartile range, 58-70 y]; 69 men) with NSCLC were finally analyzed. In lesion-based analysis, the diagnostic sensitivity and positive predictive value of 68Ga-FAPI-04 PET/CT for detection of the primary tumor were 96.70% (88/91) and 100% (88/88), respectively. In station-based analysis, the diagnostic sensitivity, specificity, and accuracy for the detection of lymph node metastasis were 72.00% (18/25), 93.10% (108/116), and 89.36% (126/141), respectively. Tumor 68Ga-FAPI-04 uptake (SUVmax, SUVmean, FTV, and TLF) correlated positively with FAP expression (r = 0.470, 0.477, 0.582, and 0.608, respectively; all P ≤ 0.001). The volume parameters FTV and TLF correlated strongly with FAP expression in 31 surgical specimens (r = 0.700 and 0.770, respectively; both P < 0.001). Conclusion: 68Ga-FAPI-04 PET/CT had excellent diagnostic efficiency for detecting lymph node metastasis, and 68Ga-FAPI-04 uptake showed a close association with FAP expression in participants with NSCLC.

Integrating image and gene-data with a semi-supervised attention model for prediction of KRAS gene mutation status in non-small cell lung cancer.

KRAS is a pathogenic gene frequently implicated in non-small cell lung cancer (NSCLC). However, biopsy as a diagnostic method has practical limitations. Therefore, it is important to accurately determine the mutation status of the KRAS gene non-invasively by combining NSCLC CT images and genetic data for early diagnosis and subsequent targeted therapy of patients. This paper proposes a Semi-supervised Multimodal Multiscale Attention Model (S2MMAM). S2MMAM comprises a Supervised Multilevel Fusion Segmentation Network (SMF-SN) and a Semi-supervised Multimodal Fusion Classification Network (S2MF-CN). S2MMAM facilitates the execution of the classification task by transferring the useful information captured in SMF-SN to the S2MF-CN to improve the model prediction accuracy. In SMF-SN, we propose a Triple Attention-guided Feature Aggregation module for obtaining segmentation features that incorporate high-level semantic abstract features and low-level semantic detail features. Segmentation features provide pre-guidance and key information expansion for S2MF-CN. S2MF-CN shares the encoder and decoder parameters of SMF-SN, which enables S2MF-CN to obtain rich classification features. S2MF-CN uses the proposed Intra and Inter Mutual Guidance Attention Fusion (I2MGAF) module to first guide segmentation and classification feature fusion to extract hidden multi-scale contextual information. I2MGAF then guides the multidimensional fusion of genetic data and CT image data to compensate for the lack of information in single modality data. S2MMAM achieved 83.27% AUC and 81.67% accuracy in predicting KRAS gene mutation status in NSCLC. This method uses medical image CT and genetic data to effectively improve the accuracy of predicting KRAS gene mutation status in NSCLC.

Validation of a multiomic model of plasma extracellular vesicle PD-L1 and radiomics for prediction of response to immunotherapy in NSCLC.

BACKGROUND: Immune-checkpoint inhibitors (ICIs) have showed unprecedent efficacy in the treatment of patients with advanced non-small cell lung cancer (NSCLC). However, not all patients manifest clinical benefit due to the lack of reliable predictive biomarkers. We showed preliminary data on the predictive role of the combination of radiomics and plasma extracellular vesicle (EV) PD-L1 to predict durable response to ICIs. MAIN BODY: Here, we validated this model in a prospective cohort of patients receiving ICIs plus chemotherapy and compared it with patients undergoing chemotherapy alone. This multiparametric model showed high sensitivity and specificity at identifying non-responders to ICIs and outperformed tissue PD-L1, being directly correlated with tumor change. SHORT CONCLUSION: These findings indicate that the combination of radiomics and EV PD-L1 dynamics is a minimally invasive and promising biomarker for the stratification of patients to receive ICIs.

Promising Candidate Prognostic Biomarkers in [18F]FDG PET Images: Evaluation in Independent Cohorts of Non-Small Cell Lung Cancer Patients.

The normalized distances from the hot spot of radiotracer uptake (SUVmax) to the tumor centroid (NHOC) and to the tumor perimeter (NHOP) have recently been suggested as novel PET features reflecting tumor aggressiveness. These biomarkers characterizing the shift of SUVmax toward the lesion edge during tumor progression have been shown to be prognostic factors in breast and non-small cell lung cancer (NSCLC) patients. We assessed the impact of imaging parameters on NHOC and NHOP, their complementarity to conventional PET features, and their prognostic value for advanced-NSCLC patients. Methods: This retrospective study investigated baseline [18F]FDG PET scans: cohort 1 included 99 NSCLC patients with no treatment-related inclusion criteria (robustness study); cohort 2 included 244 NSCLC patients (survival analysis) treated with targeted therapy (93), immunotherapy (63), or immunochemotherapy (88). Although 98% of patients had metastases, radiomic features including SUVs were extracted from the primary tumor only. NHOCs and NHOPs were computed using 2 approaches: the normalized distance from the localization of SUVmax or SUVpeak to the tumor centroid or perimeter. Bland-Altman analyses were performed to investigate the impact of both spatial resolution (comparing PET images with and without gaussian postfiltering) and image sampling (comparing 2 voxel sizes) on feature values. The correlation of NHOCs and NHOPs with other features was studied using Spearman correlation coefficients (r). The ability of NHOCs and NHOPs to predict overall survival (OS) was estimated using the Kaplan-Meier method. Results: In cohort 1, NHOC and NHOP features were more robust to image filtering and to resampling than were SUVs. The correlations were weak between NHOCs and NHOPs (r ≤ 0.45) and between NHOCs or NHOPs and any other radiomic features (r ≤ 0.60). In cohort 2, the patients with short OS demonstrated higher NHOCs and lower NHOPs than those with long OS. NHOCs significantly distinguished 2 survival profiles in patients treated with immunotherapy (log-rank test, P < 0.01), whereas NHOPs stratified patients regarding OS in the targeted therapy (P = 0.02) and immunotherapy (P < 0.01) subcohorts. Conclusion: Our findings suggest that even in advanced NSCLC patients, NHOC and NHOP features pertaining to the primary tumor have prognostic potential. Moreover, these features appeared to be robust with respect to imaging protocol parameters and complementary to other radiomic features and are now available in LIFEx software to be independently tested by others.

Factors affecting accuracy of clinical staging in resectable non-small cell lung cancer in a real-world study.

BACKGROUND: The clinical staging of non-small cell lung cancer (NSCLC) is well known to be related to their prognosis. However, there is usually a discrepancy between clinical staging and pathological staging. There are few analyses of clinical staging accuracy in patients with NSCLC. We compared the concordance rate between clinical and pathological staging of NSCLC and evaluated factors affecting the accuracy in real-world data. METHODS: Altogether, 811 patients with primary NSCLC who had undergone curative lung resection surgery in Severance Hospital from January 2019 to December 2020 were retrospectively reviewed. We used the eighth edition of the American Joint Committee on Cancer TNM staging. RESULTS: Among 811 patients, endobronchial ultrasound (EBUS) and positron emission tomography (PET-CT) were performed in 31.6% and 96.7%, respectively. The concordance rates between clinical and pathological TNM staging, T factor, and N factor, were 68.7%, 77.7%, and 85.8%, respectively. With multivariable logistic regression analysis, current smokers (OR 0.49; 95% CI: 0.32-0.76, p = 0.001) and a higher clinical stage (p < 0.001) contributed to the clinical staging inaccuracy. Additionally, the presence of a bronchoscopy specialist was significantly associated with clinical staging accuracy (OR 1.53; 95% CI: 1.10-2.13, p = 0.011). CONCLUSION: Clinical staging accuracy in NSCLC improved compared to before the widespread use of PET-CT and EBUS in clinical staging work-up. Smoking history and absence of expert bronchoscopy specialists showed a meaningful correlation with the inaccuracy of clinical staging. Thus, training more bronchoscopy experts would improve the staging accuracy of NSCLC, which could positively affect the prognosis of NSCLC.

2021 WHO Classification of Lung Cancer: Molecular Biology Research and Radiologic-Pathologic Correlation.

The 2021 World Health Organization (WHO) classification system for thoracic tumors (including lung cancer) contains several updates to the 2015 edition. Revisions for lung cancer include a new grading system for invasive nonmucinous adenocarcinoma that better reflects prognosis, reorganization of squamous cell carcinomas and neuroendocrine neoplasms, and description of some new entities. Moreover, remarkable advancements in our knowledge of genetic mutations and targeted therapies have led to a much greater emphasis on genetic testing than that in 2015. In 2015, guidelines recommended evaluation of only two driver mutations, ie, epidermal growth factor receptor (EGFR) mutations and anaplastic lymphoma kinase (ALK) fusions, in patients with nonsquamous non-small cell lung cancer. The 2021 guidelines recommend testing for numerous additional gene mutations for which targeted therapies are now available including ROS1, RET, NTRK1-3, KRAS, BRAF, and MET. The correlation of imaging features and genetic mutations is being studied. Testing for the immune biomarker programmed death ligand 1 is now recommended before starting first-line therapy in patients with metastatic non-small cell lung cancer. Because 70% of lung cancers are unresectable at patient presentation, diagnosis of lung cancer is usually based on small diagnostic samples (ie, biopsy specimens) rather than surgical resection specimens. The 2021 version emphasizes differences in the histopathologic interpretation of small diagnostic samples and resection specimens. Radiologists play a key role not only in evaluation of tumor and metastatic disease but also in identification of optimal biopsy targets. ©RSNA, 2024 Test Your Knowledge questions in the supplemental material and the slide presentation from the RSNA Annual Meeting are available for this article.

Improved outcome models with denoising diffusion.

PURPOSE: Radiotherapy outcome modelling often suffers from class imbalance in the modelled endpoints. One of the main options to address this issue is by introducing new synthetically generated datapoints, using generative models, such as Denoising Diffusion Probabilistic Models (DDPM). In this study, we implemented DDPM to improve performance of a tumor local control model, trained on imbalanced dataset, and compare this approach with other common techniques. METHODS: A dataset of 535 NSCLC patients treated with SBRT (50 Gy/5 fractions) was used to train a deep learning outcome model for tumor local control prediction. The dataset included complete treatment planning data (planning CT images, 3D planning dose distribution and patient demographics) with sparsely distributed endpoints (6-7 % experiencing local failure). Consequently, we trained a novel conditional 3D DDPM model to generate synthetic treatment planning data. Synthetically generated treatment planning datapoints were used to supplement the real training dataset and the improvement in the model's performance was studied. Obtained results were also compared to other common techniques for class imbalanced training, such as Oversampling, Undersampling, Augmentation, Class Weights, SMOTE and ADASYN. RESULTS: Synthetic DDPM-generated data were visually trustworthy, with Fréchet inception distance (FID) below 50. Extending the training dataset with the synthetic data improved the model's performance by more than 10%, while other techniques exhibited only about 4% improvement. CONCLUSIONS: DDPM introduces a novel approach to class-imbalanced outcome modelling problems. The model generates realistic synthetic radiotherapy planning data, with a strong potential to increase performance and robustness of outcome models.

A deep learning-based framework (Co-ReTr) for auto-segmentation of non-small cell-lung cancer in computed tomography images.

PURPOSE: Deep learning-based auto-segmentation algorithms can improve clinical workflow by defining accurate regions of interest while reducing manual labor. Over the past decade, convolutional neural networks (CNNs) have become prominent in medical image segmentation applications. However, CNNs have limitations in learning long-range spatial dependencies due to the locality of the convolutional layers. Transformers were introduced to address this challenge. In transformers with self-attention mechanism, even the first layer of information processing makes connections between distant image locations. Our paper presents a novel framework that bridges these two unique techniques, CNNs and transformers, to segment the gross tumor volume (GTV) accurately and efficiently in computed tomography (CT) images of non-small cell-lung cancer (NSCLC) patients. METHODS: Under this framework, input of multiple resolution images was used with multi-depth backbones to retain the benefits of high-resolution and low-resolution images in the deep learning architecture. Furthermore, a deformable transformer was utilized to learn the long-range dependency on the extracted features. To reduce computational complexity and to efficiently process multi-scale, multi-depth, high-resolution 3D images, this transformer pays attention to small key positions, which were identified by a self-attention mechanism. We evaluated the performance of the proposed framework on a NSCLC dataset which contains 563 training images and 113 test images. Our novel deep learning algorithm was benchmarked against five other similar deep learning models. RESULTS: The experimental results indicate that our proposed framework outperforms other CNN-based, transformer-based, and hybrid methods in terms of Dice score (0.92) and Hausdorff Distance (1.33). Therefore, our proposed model could potentially improve the efficiency of auto-segmentation of early-stage NSCLC during the clinical workflow. This type of framework may potentially facilitate online adaptive radiotherapy, where an efficient auto-segmentation workflow is required. CONCLUSIONS: Our deep learning framework, based on CNN and transformer, performs auto-segmentation efficiently and could potentially assist clinical radiotherapy workflow.

METnet: A novel deep learning model predicting MET dysregulation in non-small-cell lung cancer on computed tomography images.

BACKGROUND: Mesenchymal epithelial transformation (MET) is a key molecular target for diagnosis and treatment of non-small cell lung cancer (NSCLC). The corresponding molecularly targeted therapeutics have been approved by Food and Drug Administration (FDA), achieving promising results. However, current detection of MET dysregulation requires biopsy and gene sequencing, which is invasive, time-consuming and difficult to obtain tumor samples. METHODS: To address the above problems, we developed a noninvasive and convenient deep learning (DL) model based on Computed tomography (CT) imaging data for prediction of MET dysregulation. We introduced the unsupervised algorithm RK-net for automated image processing and utilized the MedSAM large model to achieve automated tissue segmentation. Based on the processed CT images, we developed a DL model (METnet). The model based on the grouped convolutional block. We evaluated the performance of the model over the internal test dataset using the area under the receiver operating characteristic curve (AUROC) and accuracy. We conducted subgroup analysis on the basis of clinical data of the lung cancer patients and compared the performance of the model in different subgroups. RESULTS: The model demonstrated a good discriminative ability over the internal test dataset. The accuracy of METnet was 0.746 with an AUC value of 0.793 (95% CI 0.714-0.871). The subgroup analysis revealed that the model exhibited similar performance across different subgroups. CONCLUSIONS: METnet realizes prediction of MET dysregulation in NSCLC, holding promise for guiding precise tumor diagnosis and treatment at the molecular level.

Integration of on-treatment modified Glasgow prognostic score (mGPS) to improve imaging-based prediction of outcomes in patients with non-small cell lung cancer on immune checkpoint inhibition.

INTRODUCTION: A large number of patients with non-small cell lung cancer (NSCLC) on immune checkpoint inhibition (ICI) achieve stable disease (SD) as the best overall response, which is associated with heterogeneous outcomes. In this context, complementary biomarkers that improve outcome prediction are needed. We have recently demonstrated that measuring the on-treatment modified Glasgow prognostic score (mGPS), which is based on the two serum markers C-reactive protein (CRP) and albumin, can improve outcome prediction complementary to radiological staging in metastatic renal cell carcinoma. However, this concept has not been assessed for patients with NSCLC on ICI. METHODS: We assessed the prognostic and predictive value of on-treatment mGPS at week six in patients with NSCLC treated with atezolizumab or docetaxel in the phase 3 OAK trial (NCT02008227) comprising n = 750 patients and validated the findings in the phase 2 BIRCH (NCT02031458, n = 560). RESULTS: On-treatment mGPS assessed at week six demonstrated valuable prognostic information (Hazard Ratio (HR) for mGPS low-risk vs intermediate risk 2.34 (95 % CI 1.76-3.11, p < 0.001) and vs high risk 3.56, (95 % CI 2.57-4.91, p < 0.001) in the atezolizumab-treated subgroup. On-treatment mGPS predicted overall survival more accurately than imaging using RECIST criteria (concordance index: on-treatment mGPS 0.646 (95 % CI 0.615-0.677) vs RECIST 0.606 (95 % CI 0.575-0.637)). On-treatment mGPS provides additional prognostic information to imaging-assessed treatment response at first staging, especially for the patient subgroup with SD. These findings were validated in the BIRCH trial. CONCLUSIONS: We highlight the novel concept of integrating on-treatment mGPS for improved outcome prediction in conjunction with radiological imaging for patients with NSCLC on ICI.

First exploration of the on-treatment changes in tumor and organ uptake of a radiolabeled anti PD-L1 antibody during chemoradiotherapy in patients with non-small cell lung cancer using whole body PET.

BACKGROUND: In patients with locally advanced unresectable non-small cell lung cancer (NSCLC), durvalumab, an anti-programmed cell death ligand-1 (PD-L1) antibody, has shown improved overall survival when used as consolidation therapy following concurrent chemoradiotherapy (CRT). However, it is unclear whether CRT itself upregulates PD-L1 expression. Therefore, this study aimed to explore the changes in the uptake of the anti PD-L1 antibody [89Zr]Zr-durvalumab in tumors and healthy organs during CRT in patients with NSCLC. METHODS: Patients with NSCLC scheduled to undergo CRT were scanned 7±1 days after administration of 37±1 MBq [89Zr]Zr-durvalumab at baseline, 1-week on-treatment and 1 week after finishing 6 weeks of CRT. First, [89Zr]Zr-durvalumab uptake was visually assessed in a low dose cohort with a mass dose of 2 mg durvalumab (0.13% of therapeutic dose) and subsequently, quantification was done in a high dose cohort with a mass dose of 22.5 mg durvalumab (1.5% of therapeutic dose). Tracer pharmacokinetics between injections were compared using venous blood samples drawn in the 22.5 mg cohort. Visual assessment included suspected lesion detectability. Positron emission tomography (PET) uptake in tumoral and healthy tissues was quantified using tumor to plasma ratio (TPR) and organ to plasma ratio, respectively. RESULTS: In the 2 mg dose cohort, 88% of the 17 identified tumor lesions were positive at baseline, compared with 69% (9/13) for the 22.5 mg cohort. Although the absolute plasma concentrations between patients varied, the intrapatient variability was low. The ten quantitatively assessed lesions in the 22.5 mg cohort had a median TPR at baseline of 1.3 (IQR 0.7-1.5), on-treatment of 1.0 (IQR 0.7-1.4) and at the end of treatment of 0.7 (IQR 0.6-0.7). On-treatment, an increased uptake in bone marrow was seen in three out of five patients together with a decreased uptake in the spleen in four out of five patients. CONCLUSIONS: This study successfully imaged patients with NSCLC with [89Zr]Zr-durvalumab PET before and during CRT. Our data did not show any increase in [89Zr]Zr-durvalumab uptake in the tumor 1-week on-treatment and at the end of treatment. The changes observed in bone marrow and spleen may be due to an CRT-induced effect on immune cells. TRIAL REGISTRATION NUMBER: EudraCT number: 2019-004284-51.