Computed tomography (CT) images are formally taken as an assistance of early diagnosis in lung nodule analysis. Thus the accurate lung nodule segmentation is in great need for image-driven tasks. However, as heterogeneity exists between different types of lung nodules, the similar visual appearance between the pixels of nodules and pixels of non-nodule area make it difficult for automatic lung nodule segmentation. In this article, we propose a fast end-to-end framework, called Fast Multi-crop Guided Attention (FMGA) network, to accurately segment lung nodules in CT images. Our method utilizes multi-crop nodule slices as input to aggregate contextual information (2D context from current image slice and 3D context from adjacent axial slices), and exploits a global convolutional layer for nodule pixel embedding matching. To further make use of the information from border pixels near the nodule margin for better segmentation, we develop a weighted loss function to facilitate the model training by considering a balanced class samples of pixels around the nodule margin. Moreover, we utilize a central pooling layer to facilitate the contexts feature propagation in pixel neighbors. We evaluate our method on the largest public lung CT dataset LIDC and the collected lung CT data from Wuhan local hospital, respectively. Experimental results show that FMGA achieves superior performance among the state-of-the-arts. In addition, we give an ablation study and visualization results to illustrate how each component works for accurate lung nodule segmentation.

Liver cancer is one of the leading causes of cancer deaths worldwide. Among all primary liver cancers, hepatocellular carcinoma (HCC) is the most common type, representing 75%‐85% of all primary liver cancer cases. Median survival following diagnosis of HCC is approximately 6 to 20 months due to late diagnosis in its course and few effective treatment options. Interventional therapy with minimal invasiveness is recognized as a promising treatment for HCC. However, due to the heterogeneity of HCC and the complexity of the tumor microenvironment, the long‐term efficacy of treatment for HCC remains a challenge in the clinic. Tumor microenvironment, including factors such as hypoxia, angiogenesis, low extracellular pH, interstitial fluid pressure, aerobic glycolysis, and various immune responses, has emerged as a key contributor to tumor residual and progression after locoregional treatment for HCC. New approaches to noninvasively assess the treatment response and assist in the clinical decision‐making process are therefore urgently needed. Molecular imaging tools enabling such an assessment may significantly advance clinical practice by allowing real‐time optimization of treatment protocols for the individual patient. This review discusses recent advances in the application of molecular imaging technologies for noninvasively assessing changes occurring in the microenvironment of HCC after locoregional treatment.

Setting. Thoracic surgery department of a tertiary hospital. Background. Fear of pain (FOP) has been recognized as an influential moderator and determinant of the perception and disability of chronic pain. However, studies on FOP in postoperative acute pain are few and inconsistent. Objective. To explore whether FOP is related to pain intensity after thoracic surgery and provide a reference for FOP study in postoperative pain. Methods. From February to March 2022, 89 patients completed Chinese Version of Fear of Pain-9 Items (FOP-9), Chinese version of the Brief Pain Inventory (BPI, including least, worst, and average pain) and Simplified Chinese version of the Pain Catastrophizing Scale (PCS). Correlation analyses and mediation analyses were used for exploring the relationship between factors. Results. Mediation analyses showed that the total effects of FOP-9 on BPI all were significant (least pain: effect = 0.085, , 95% CI = 0.019∼0.151; worst pain: effect = 0.116, , 95% CI = 0.037∼0.196; average pain: effect = 0.102, , 95% CI = 0.031∼0.174) indicating that FOP-9 was a predictor to BPI. The 95% bias-corrected bootstrap confidence interval of estimate of indirect effect between FOP-9 and least pain/average pain through PCS was −0.036∼0.024 and −0.003∼0.069 (all contain zero), which indicated that PCS is not a mediator between FOP-9 and least pain/average pain. However, the estimate of indirect effect between FOP-9 and worst pain through PCS were 0.048 (95% CI = 0.095∼0.088), and direct effect was not statistically significant (95% CI = −0.017∼0.153), indicating that PCS acted as a complete intermediary between FOP-9 and worst pain. FOP-9 and PCS showed significant positive prediction effect on worst pain. Conclusions. Both trait FOP and state FOP were associated with higher postoperative pain reports after thoracic surgery. Trait FOP influences postoperative pain through the mediating effect of state FOP.

Objective: In this study, we aim to establish a non-invasive tool to predict epidermal growth factor receptor (EGFR) mutation status and subtypes based on radiomic features of computed tomography (CT). Methods: A total of 233 lung adenocarcinoma patients were investigated and randomly divided into the training and test cohorts. In this study, 2300 radiomic features were extracted from original and filtered (Exponential, Laplacian of Gaussian, Logarithm, Gabor, Wavelet) CT images. The radiomic features were divided into four categories, including histogram, volumetric, morphologic, and texture features. An RF-BFE algorithm was developed to select the features for building the prediction models. Clinicopathological features (including age, gender, smoking status, TNM staging, maximum diameter, location, and growth pattern) were combined to establish an integrated model with radiomic features. ROC curve and AUC quantified the effectiveness of the predictor of EGFR mutation status and subtypes. Results: A set of 10 features were selected to predict EGFR mutation status between EGFR mutant and wild type, while 9 selected features were used to predict mutation subtypes between exon 19 deletion and exon 21 L858R mutation. To predict the EGFR mutation status, the AUC of the training cohort was 0.778 and the AUC of the test cohort was 0.765. To predict the EGFR mutation subtypes, the AUC of training cohort was 0.725 and the AUC of test cohort was 0.657. The integrated model showed the most optimal predictive performance with EGFR mutation status (AUC = 0.870 and 0.759) and subtypes (AUC = 0.797 and 0.554) in the training and test cohorts. Conclusion: CT-based radiomic features can extract information on tumor heterogeneity in lung adenocarcinoma. In addition, we have established a radiomic model and an integrated model to non-invasively predict the EGFR mutation status and subtypes of lung adenocarcinoma, which is conducive to saving clinical costs and guiding targeted therapy.

Background: The low response rate of anti-PD-1 monoclonal antibodies (mAbs) in hepatocellular carcinoma (HCC) requires the development of combination immunotherapy strategies to improve their efficacy. This study aimed to use LAG-3-targeted PET imaging to monitor the efficacy of anti-PD-1 mAb, a stimulator of interferon genes (STING) agonist, and anti-LAG-3 mAb, both individually and in combination. Furthermore, we evaluated the potential of a triple immunotherapy regimen (anti-PD-1 mAb, STING agonist, and anti-LAG-3 mAb) to improve HCC treatment. Methods: The LAG-3 inhibitor C25 based on a cyclic peptide was chelated with NOTA, radiolabeled with [⁶⁸Ga]GaCl₃. The resulting [⁶⁸Ga]Ga-NOTA-C25 underwent in vivo PET imaging and ex vivo biodistribution examination in Hepa1-6 tumor-bearing mice. [⁶⁸Ga]Ga-NOTA-C25 PET was used to monitor the efficacy of monotherapy and dual immunotherapy with anti-PD-1 monoclonal antibody (mAb) and STING agonists. The tumor uptake of [⁶⁸Ga]Ga-NOTA-C25, tumor response, and survival rates were measured following different treatments. The therapeutic efficacy, molecular mechanisms, and safety of triple immunotherapy were validated using histopathological analysis and flow cytometry. Results: [⁶⁸Ga]Ga-NOTA-C25 PET imaging effectively and noninvasively detected LAG-3⁺ tumor-infiltrating lymphocytes (TILs) in Hepa1-6 tumor-bearing mice. In mice treated with anti-PD-1 mAb, STING agonist, or a combination immunotherapy, [⁶⁸Ga]Ga-NOTA-C25 PET revealed significantly increased LAG-3⁺ TIL levels. At the treatment endpoint, the combination of the STING agonist with the anti-PD-1 mAb resulted in a significantly higher uptake (1.35 ± 0.191%ID/g) compared to the control group (0.402 ± 0.017%ID/g), the anti-PD-1 mAb group (0.647 ± 0.037%ID/g), and the STING agonist group (0.874 ± 0.089%ID/g). Uptake of [⁶⁸Ga]Ga-NOTA-C25 was positively correlated with tumor therapeutic effects and survival rates. Triple immunotherapy with anti-PD-1 mAb, a STING agonist, and anti-LAG-3 mAb further enhanced efficacy compared to any dual immunotherapy regimen, and treatment efficacy was linearly associated with [⁶⁸Ga]Ga-NOTA-C25 tumor uptake. Conclusions: Anti-PD-1 mAb and STING agonists have shown notable synergy in upregulating LAG-3 expression on TILs in HCC, which can be successfully tracked by [⁶⁸Ga]Ga-NOTA-C25 PET imaging. Furthermore, integration of a triple immunotherapy regimen comprising an anti-PD-1 mAb, STING agonist, and anti-LAG-3 mAb demonstrated a significant improvement in therapeutic efficacy over dual immunotherapy approaches.

Purpose: The spleen serves as an important immune organ which influences the anti-tumor immune response by modulating the immune microenvironment. This study investigated the prognostic impact of spleen volume (SV) on the survival in hepatocellular carcinoma (HCC) patients receiving immune checkpoint inhibitors (ICIs). Patients and Methods: This retrospective study included 224 HCC patients treated with ICIs, categorized into Higher and Lower SV groups by median SV and further into SV increased and Non-SV increased groups based on changes in SV at 3 months after ICIs. Kaplan-Meier curves and Cox regression models were used to evaluate the influence of SV and clinical indicators on progression-free survival (PFS) and overall survival (OS). Independent prognostic factors identified via multivariate analysis were incorporated into nomograms, with their accuracy assessed using concordance index (C-index), time-dependent receiver operating characteristic (ROC) and calibration curves. Restricted cubic spline (RCS) analysis was conducted to assess the relationship between baseline SV and survival. Results: The Higher SV and SV increased groups demonstrated shorter PFS and OS compared to the Lower SV and Non-SV increased groups, respectively. These results were consistent with different regimens in the Child A. The C-index of nomogram for PFS were 0.700 (0.678– 0.721) and OS 0.733(0.709– 0.757). The ROC and calibration curves confirmed robust discrimination and predictive accuracy of models. RCS analysis revealed a nonlinear association between baseline SV and survival risk, providing a more comprehensive overview of SV in relation to survival in HCC patients treated with ICIs. Conclusion: The baseline SV and its relative change at three months after treatment are expected to become routine imaging makers for predicting survival in HCC patients receiving ICIs, which consequently contributes to their clinical management.