Modified spatial architecture of regulatory T cells after neoadjuvant chemotherapy in non-small cell lung cancer patients.

It is crucial to decipher the modulation of regulatory T cells (Tregs) in tumor microenvironment (TME) induced by chemotherapy, which may contribute to improving the efficacy of neoadjuvant chemoimmunotherapy in resectable non-small cell lung cancer (NSCLC). We retrospectively collected specimens from patients with II-III NSCLC, constituting two cohorts: a neoadjuvant chemotherapy (NAC) cohort (N = 141) with biopsy (N = 58) and postoperative specimens (N = 141), and a surgery-only cohort (N = 122) as the control group. Then, the cell density (Dens), infiltration score (InS), and Treg-cell proximity score (TrPS) were conducted using a panel of multiplex fluorescence staining (Foxp3, CD4, CD8, CK, CD31, ɑSMA). Subsequently, the association of Tregs with cancer microvessels (CMVs) and cancer-associated fibroblasts (CAFs) was analyzed. Patients with NAC treatment have a higher density of Tregs in both paired (P < 0.001) and unpaired analysis (P = 0.022). Additionally, patients with NAC treatment showed higher infiltration score (paired, P < 0.001; unpaired, P = 0.014) and more CD8+T cells around Tregs (paired/unpaired, both P < 0.001). Subgroup analysis indicated that tumors with a diameter of ≤ 5 cm exhibited increase in both Dens(Treg) and InS(Treg), and gemcitabine, pemetrexed and taxel enhanced Dens(Treg) and TrPS(CD8) following NAC. Multivariate analysis identified that the Dens(Tregs), InS(Tregs) and TrPS(CD8) were significantly associated with better chemotherapy response [OR = 8.54, 95%CI (1.69, 43.14), P = 0.009; OR = 7.14, 95%CI (1.70, 30.08), P = 0.024; OR = 5.50, 95%CI (1.09, 27.75), P = 0.039, respectively] and positive recurrence-free survival [HR = 3.23, 95%CI (1.47, 7.10), P = 0.004; HR = 2.70; 95%CI (1.27, 5.72); P = 0.010; HR = 2.55, 95%CI (1.21, 5.39), P = 0.014, respectively]. Moreover, TrPS(CD8) and TrPS(CD4) were negatively correlated with the CMVs and CAFs. These discoveries have deepened our comprehension of the immune-modulating impact of chemotherapy and underscored that the modified spatial landscape of Tregs after chemotherapy should be taken into account for personalized immunotherapy, aiming to ultimately improve clinical outcomes in patients with NSCLC.

Dysfunction of CD8+ T cells around tumor cells leads to occult lymph node metastasis in NSCLC patients.

Occult lymph node metastasis (OLNM) is one of the main causes of regional recurrence in inoperable N0 non-small cell lung cancer (NSCLC) patients following stereotactic ablation body radiotherapy (SABR) treatment. The integration of immunotherapy and SABR (I-SABR) has shown preliminary efficiency in mitigating this recurrence. Therefore, it is necessary to explore the functional dynamics of critical immune effectors, particularly CD8+ T cells in the development of OLNM. In this study, tissue microarrays (TMAs) and multiplex immunofluorescence (mIF) were used to identify CD8+ T cells and functional subsets (cytotoxic CD8+ T cells/predysfunctional CD8+ T cells (CD8+ Tpredys)/dysfunctional CD8+ T cells (CD8+ Tdys)/other CD8+ T cells) among the no lymph node metastasis, OLNM, and clinically evident lymph node metastasis (CLNM) groups. As the degree of lymph node metastasis escalated, the density of total CD8+ T cells and CD8+ Tdys cells, as well as their proximity to tumor cells, increased progressively and remarkably in the invasive margin (IM). In the tumor center (TC), both the density and proximity of CD8+ Tpredys cells to tumor cells notably decreased in the OLNM group compared with the group without metastasis. Furthermore, positive correlations were found between the dysfunction of CD8+ T cells and HIF-1α+CD8 and cancer microvessels (CMVs). In conclusion, the deterioration in CD8+ T cell function and interactive dynamics between CD8+ T cells and tumor cells play a vital role in the development of OLNM in NSCLC. Strategies aimed at improving hypoxia or targeting CMVs could potentially enhance the efficacy of I-SABR.

Spatial features of specific CD103+CD8+ tissue-resident memory T cell subsets define the prognosis in patients with non-small cell lung cancer.

BACKGROUND: Tissue-resident memory T (TRM) cells can reside in the tumor microenvironment and are considered the primary response cells to immunotherapy. Heterogeneity in functional status and spatial distribution may contribute to the controversial role of TRM cells but we know little about it. METHODS: Through multiplex immunofluorescence (mIF) (CD8, CD103, PD-1, Tim-3, GZMB, CK), the quantity and spatial location of TRM cell subsets were recognized in the tissue from 274 patients with NSCLC after radical surgery. By integrating multiple machine learning methods, we constructed a TRM-based spatial immune signature (TRM-SIS) to predict the prognosis. Furthermore, we conducted a CD103-related gene set enrichment analysis (GSEA) and verified its finding by another mIF panel (CD8, CD103, CK, CD31, Hif-1α). RESULTS: The density of TRM cells was significantly correlated with the expression of PD-1, Tim-3 and GZMB. Four types of TRM cell subsets was defined, including TRM1 (PD-1-Tim-3-TRM), TRM2 (PD-1+Tim-3-TRM), TRM3 (PD-1-Tim-3+TRM) and TRM4 (PD-1+Tim-3+TRM). The cytotoxicity of TRM2 was the strongest while that of TRM4 was the weakest. Compare with TRM1 and TRM2, TRM3 and TRM4 had better infiltration and stronger interaction with cancer cells. The TRM-SIS was an independent prognostic factor for disease-free survival [HR = 2.43, 95%CI (1.63-3.60), P < 0.001] and showed a better performance than the TNM staging system for recurrence prediction. Furthermore, by CD103-related GSEA and mIF validation, we found a negative association between tumor angiogenesis and infiltration of TRM cells. CONCLUSIONS: These findings reveal a significant heterogeneity in the functional status and spatial distribution of TRM cells, and support it as a biomarker for the prognosis of NSCLC patients. Regulating TRM cells by targeting tumor angiogenesis may be a potential strategy to improve current immunotherapy.

Growing Like a Tree: Finding Trunks From Graph Skeleton Trees.

The message-passing paradigm has served as the foundation of graph neural networks (GNNs) for years, making them achieve great success in a wide range of applications. Despite its elegance, this paradigm presents several unexpected challenges for graph-level tasks, such as the long-range problem, information bottleneck, over-squashing phenomenon, and limited expressivity. In this study, we aim to overcome these major challenges and break the conventional "node- and edge-centric" mindset in graph-level tasks. To this end, we provide an in-depth theoretical analysis of the causes of the information bottleneck from the perspective of information influence. Building on the theoretical results, we offer unique insights to break this bottleneck and suggest extracting a skeleton tree from the original graph, followed by propagating information in a distinctive manner on this tree. Drawing inspiration from natural trees, we further propose to find trunks from graph skeleton trees to create powerful graph representations and develop the corresponding framework for graph-level tasks. Extensive experiments on multiple real-world datasets demonstrate the superiority of our model. Comprehensive experimental analyses further highlight its capability of capturing long-range dependencies and alleviating the over-squashing problem, thereby providing novel insights into graph-level tasks.

Explainable survival analysis with uncertainty using convolution-involved vision transformer.

Image-based precision medicine research is able to help doctors make better decisions on treatments. Among all kinds of medical images, a special form is called Whole Slide Image (WSI), which is used for diagnosing patients with cancer, aiming to enable more accurate survival prediction with its high resolution. However, One unique challenge of the WSI-based prediction models is processing the gigabyte-size or even terabyte-size WSIs, which would make most models computationally infeasible. Although existing models mostly use a pre-selected subset of key patches or patch clusters as input, they might discard some important morphology information, making the prediction inferior. Another challenge is improving the prediction models' explainability, which is crucial to help doctors understand the predictions given by the models and make faithful decisions with high confidence. To address the above two challenges, in this work, we propose a novel explainable survival prediction model based on Vision Transformer. Specifically, we adopt dual-channel convolutional layers to utilize the complete WSIs for more accurate predictions. We also introduce the aleatoric uncertainty into our model to understand its limitation and avoid overconfidence in using the prediction results. Additionally, we present a post-hoc explainable method to identify the most salient patches and distinct morphology features as supporting evidence for predictions. Evaluations of two large cancer datasets show that our proposed model is able to make survival predictions more effectively and has better explainability for cancer diagnosis.

Efficacy and safety of immunoradiotherapy for advanced non-small cell lung cancer: a retrospective comparative cohort study.

Background: Treatment of radiotherapy (RT) combined with immune checkpoint inhibitor (ICI) may remarkably improve the prognosis in patients with metastatic non-small cell lung cancer (NSCLC). However, the treatment time of RT, irradiated lesion and the optimum combined scheme, have not been fully determined. Methods: Data regarding overall survival (OS), progression-free survival (PFS), treatment response, and adverse events of 357 patients with advanced NSCLC treated with ICI alone or in combination with RT prior to/during ICI treatment were retrospectively collected. Additionally, subgroup analyses for radiation dose, time interval between RT and immunotherapy, and number of irradiated lesions were performed. Results: Median PFS of the ICI alone and ICI + RT groups was 6 and 12 months, respectively (P<0.0001). The objective response rate (ORR) and disease control rate (DCR) were significantly higher in the ICI + RT group than in the ICI alone group (P=0.014; P=0.015, respectively). However, OS, the distant response rate (DRR), and the distant control rate (DCRt) did not differ significantly between the groups. Out-of-field DRR and DCRt were defined in unirradiated lesions only. Compared with RT application prior to ICI, its application concomitant with ICI led to higher DRR (P=0.018) and DCRt (P=0.002). Subgroup analyses revealed that single-site, high biologically effective dose (BED) (≥72 Gy), planning target volume (PTV) size (<213.7 mL) RT groups had better PFS. In multivariate analysis, PTV volume [≥213.7 vs. <213.7 mL: hazard ratio (HR), 1.89; 95% confidence interval (CI): 1.04-3.42; P=0.035] was an independent predictor of immunotherapy PFS. Additionally, radio-immunotherapy increased the incidence rate of grade 1-2 immune-related pneumonitis compared with ICI alone. Conclusions: Combination therapy using ICIs and radiation may improve PFS and tumor response rates in advanced NSCLC patients regardless of programmed cell death 1 ligand 1 (PD-L1) level or previous treatments. However, it may increase the incidence of immune-related pneumonitis.

Functional status and spatial architecture of tumor-infiltrating CD8+ T cells are associated with lymph node metastases in non-small cell lung cancer.

BACKGROUND: Anti-PD-(L)1 immunotherapy has been recommended for non-small cell lung cancer (NSCLC) patients with lymph node metastases (LNM). However, the exact functional feature and spatial architecture of tumor-infiltrating CD8 + T cells remain unclear in these patients. METHODS: Tissue microarrays (TMAs) from 279 IA-IIIB NSCLC samples were stained by multiplex immunofluorescence (mIF) for 11 markers (CD8, CD103, PD-1, Tim3, GZMB, CD4, Foxp3, CD31, αSMA, Hif-1α, pan-CK). We evaluated the density of CD8 + T-cell functional subsets, the mean nearest neighbor distance (mNND) between CD8 + T cells and neighboring cells, and the cancer-cell proximity score (CCPS) in invasive margin (IM) as well as tumor center (TC) to investigate their relationships with LNM and prognosis. RESULTS: The densities of CD8 + T-cell functional subsets, including predysfunctional CD8 + T cells (Tpredys) and dysfunctional CD8 + T cells (Tdys), in IM predominated over those in TC (P < 0.001). Multivariate analysis identified that the densities of CD8 + Tpredys cells in TC and CD8 + Tdys cells in IM were significantly associated with LNM [OR = 0.51, 95%CI (0.29-0.88), P = 0.015; OR = 5.80, 95%CI (3.19-10.54), P < 0.001; respectively] and recurrence-free survival (RFS) [HR = 0.55, 95%CI (0.34-0.89), P = 0.014; HR = 2.49, 95%CI (1.60-4.13), P = 0.012; respectively], independent of clinicopathological factors. Additionally, shorter mNND between CD8 + T cells and their neighboring immunoregulatory cells indicated a stronger interplay network in the microenvironment of NSCLC patients with LNM and was associated with worse prognosis. Furthermore, analysis of CCPS suggested that cancer microvessels (CMVs) and cancer-associated fibroblasts (CAFs) selectively hindered CD8 + T cells from contacting with cancer cells, and were associated with the dysfunction of CD8 + T cells. CONCLUSION: Tumor-infiltrating CD8 + T cells were in a more dysfunctional status and in a more immunosuppressive microenvironment in patients with LNM compared with those without LNM.

Sinkhorn Distance Minimization for Adaptive Semi-Supervised Social Network Alignment.

Social network alignment, aiming at linking identical identities across different social platforms, is a fundamental task in social graph mining. Most existing approaches are supervised models and require a large number of manually labeled data, which are infeasible in practice considering the yawning gap between social platforms. Recently, isomorphism across social networks is incorporated as complementary to link identities from the distribution level, which contributes to alleviating the dependency on sample-level annotations. Adversarial learning is adopted to learn a shared projection function by minimizing the distance between two social distributions. However, the hypothesis of isomorphism might not always hold true as social user behaviors are generally unpredictable, and thus a shared projection function is insufficient to handle the sophisticated cross-platform correlations. In addition, adversarial learning suffers from training instability and uncertainty, which may hinder model performance. In this article, we propose a novel meta-learning-based social network alignment model Meta-SNA to effectively capture the isomorphism and the unique characteristics of each identity. Our motivation lies in learning a shared meta-model to preserve the global cross-platform knowledge and an adaptor to learn a specific projection function for each identity. Sinkhorn distance is further introduced as the distribution closeness measurement to tackle the limitations of adversarial learning, which owns an explicitly optimal solution and can be efficiently computed by the matrix scaling algorithm. Empirically, we evaluate the proposed model over multiple datasets, and the experimental results demonstrate the superiority of Meta-SNA.

Efficacy and safety of immune checkpoint inhibitors plus anlotinib in small cell lung cancer with brain metastases: a retrospective, multicentre study.

BACKGROUND: Previous studies showed that both immune checkpoint inhibitors (ICIs) and anlotinib have central nervous system (CNS) efficacy. This study aimed to evaluate the efficacy and safety of ICIs combined with anlotinib in small cell lung cancer (SCLC) patients with brain metastases (BMs). METHODS: We retrospectively reviewed SCLC patients with CNS metastases confirmed by brain magnetic resonance imaging (MRI). Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST 1.1) and Response assessment in neuro-oncology brain metastases (RANO-BM) were used to evaluate treatment response to ICIs plus anlotinib. Kaplan-Meier analysis and Cox regression analysis were performed to determine patient's prognosis. RESULTS: Sixty-six patients with baseline BMs were included. For patients with measurable intracranial lesions, the intracranial objective response rate (ORR) was 29.2% based on RECIST 1.1 criteria and was 27.1% based on RANO-BM criteria. The median intracranial progression-free survival (PFS) was 9.0 months (95% confidence interval [CI], 6.5-11.5 months). The median overall survival (OS) was 13.4 months (95% CI, 10.7-20.5 months). Multivariate Cox regression analysis showed that high disease burden (hazard ratio [HR] = 4.83, P < 0.001), multiple BMs (HR = 2.71, P = 0.036), and more than or equal to 3 prior lines of therapy (HR = 2.56, P = 0.023) were independent negative predictors of OS. The overall incidence of treatment-related adverse events (TRAEs) was 75.8%, and grade 3-4 TRAEs were reported in 19.7% of patients. CONCLUSIONS: Our results suggested that ICIs plus anlotinib had potent CNS efficacy with tolerable toxicity and could be a promising treatment option for SCLC patients with BMs.

Visual Sentiment Analysis With Social Relations-Guided Multiattention Networks.

These days, social media users tend to express their feelings through sharing images online. Capturing the emotions embedded in these social images involves great research challenges and practical values. Most existing works concentrate on extracting the visual feature from a global view, while ignoring the fact that visual objects are also rich in emotion. How to leverage the multilevel visual features to improve the sentiment analysis performance is important yet challenging. Besides, existing works view each social image as an independent sample while ignoring the rich correlations among social images, which may be helpful in detecting visual emotion. In this article, we propose a novel model called social relations-guided multiattention networks (SRGMANs) to incorporate both the multilevel (region-level and object-level) visual features of a single image and the correlations among multiple social images to conduct visual sentiment analysis. Specifically, we first construct a heterogeneous network consisting of various types of social relations and introduce a heterogeneous network embedding method to learn the network representation for each image. Then, two visual attention branches (region attention network and object attention network) are devised to extract emotional and discriminative visual features. For each branch, we design a self-attention module to capture the emotional dependencies among visual parts. Besides, a network-guided attention module is also designed in each branch to focus on more network-related emotional visual parts with the guidance of the topology information. Finally, the attended visual features from the two attention models, together with network representation features, are combined within a holistic framework to predict the sentiment of social images. Extensive experiments demonstrate the superiority of our model on three benchmark datasets.

Distribution and Determinants of Unmet Need for Supportive Care Among Women with Breast Cancer in China.

BACKGROUND The aim of this study was to determine the need for supportive care among women suffering from breast cancer in China and to identify its potential determinants to inform the development of effective and efficient healthcare services across different settings. MATERIAL AND METHODS In a tertiary-care hospital in Weifang, China, between July 2015 and January 2016, all women attending the Breast Cancer Clinic for regular physical examinations after treatment for breast cancer were consecutively recruited. The 34-item Supportive Care Needs Survey tool (Chinese version) (SCNS-SF34-C) was used to assess the unmet needs among participants. RESULTS Among 264 recruited patients, based on at least single-item endorsement, 60.2% had moderate to high level of need for supportive care, while only 13.3% expressed no need. Lack of information regarding health systems was the most common domain with moderate to high unmet needs, more so among rural patients (8 vs. 5 out of 10). In each information-related domain, huge unmet need was observed among all patients irrespective of urban or rural residence. Both overall and individual information-related domain-specific unmet needs were significantly higher among rural patients as opposed to their urban counterparts. Multiple regression analyses revealed a significant rural-urban variation of unmet needs. Moreover, education and post-diagnosis time duration were negatively associated with unmet needs while stage of cancer was positively associated with these unmet needs. CONCLUSIONS There is a huge burden of unmet needs for information on the healthcare system among breast cancer survivors in China. Rural residence, less education, advanced stage of cancer, and shorter duration since diagnosis were the identified determinants requiring targeted intervention.