During the diagnostic process, clinicians leverage multimodal information, such as the chief complaint, medical images and laboratory test results. Deep-learning models for aiding diagnosis have yet to meet this requirement of leveraging multimodal information. Here we report a transformer-based representation-learning model as a clinical diagnostic aid that processes multimodal input in a unified manner. Rather than learning modality-specific features, the model leverages embedding layers to convert images and unstructured and structured text into visual tokens and text tokens, and uses bidirectional blocks with intramodal and intermodal attention to learn holistic representations of radiographs, the unstructured chief complaint and clinical history, and structured clinical information such as laboratory test results and patient demographic information. The unified model outperformed an image-only model and non-unified multimodal diagnosis models in the identification of pulmonary disease (by 12% and 9%, respectively) and in the prediction of adverse clinical outcomes in patients with COVID-19 (by 29% and 7%, respectively). Unified multimodal transformer-based models may help streamline the triaging of patients and facilitate the clinical decision-making process.

A thorough interrogation of the immune landscape is crucial for immunotherapy strategy selection and prediction of clinical responses in non-small-cell lung cancer (NSCLC) patients. Single-cell RNA sequencing (scRNA-seq) techniques have prompted the opportunity to dissect the distinct immune signatures between lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC), the two major subtypes of NSCLC. Here, we performed scRNA-seq on 72,475 immune cells from 40 samples of tumor and matched adjacent normal tissues spanning 19 NSCLC patients, and drew a systematic immune cell transcriptome atlas. Joint analyses of the distinct cellular compositions, differentially expressed genes (DEGs), cell–cell interactions, pseudotime trajectory, transcriptomic factors and prognostic factors based on The Cancer Genome Atlas (TCGA), revealed the central roles of cytotoxic and effector T and NK cells and the distinct functional macrophages (Mφ) subtypes in the immune microenvironment heterogeneity between LUAD and LUSC. The dominant subtype of Mφ was FABP4-Mφ in LUAD and SPP1-Mφ in LUSC. Importantly, we identified a novel lymphocyte-related Mφ cluster, which we named SELENOP-Mφ, and further established its antitumor role in both types, especially in LUAD. Our comprehensive depiction of the immune heterogeneity and definition of Mφ clusters could help design personalized treatment for lung cancer patients in clinical practice.

There have been hundreds of millions of cases of coronavirus disease 2019 (COVID-19), which is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). With the growing population of recovered patients, it is crucial to understand the long-term consequences of the disease and management strategies. Although COVID-19 was initially considered an acute respiratory illness, recent evidence suggests that manifestations including but not limited to those of the cardiovascular, respiratory, neuropsychiatric, gastrointestinal, reproductive, and musculoskeletal systems may persist long after the acute phase. These persistent manifestations, also referred to as long COVID, could impact all patients with COVID-19 across the full spectrum of illness severity. Herein, we comprehensively review the current literature on long COVID, highlighting its epidemiological understanding, the impact of vaccinations, organ-specific sequelae, pathophysiological mechanisms, and multidisciplinary management strategies. In addition, the impact of psychological and psychosomatic factors is also underscored. Despite these crucial findings on long COVID, the current diagnostic and therapeutic strategies based on previous experience and pilot studies remain inadequate, and well-designed clinical trials should be prioritized to validate existing hypotheses. Thus, we propose the primary challenges concerning biological knowledge gaps and efficient remedies as well as discuss the corresponding recommendations.

The widespread implementation of low-dose computed tomography (LDCT) in lung cancer screening has led to the increasing detection of pulmonary nodules. However, precisely evaluating the malignancy risk of pulmonary nodules remains a formidable challenge. Here we propose a triage-driven Chinese Lung Nodules Reporting and Data System (C-Lung-RADS) utilizing a medical checkup cohort of 45,064 cases. The system was operated in a stepwise fashion, initially distinguishing low-, mid-, high- and extremely high-risk nodules based on their size and density. Subsequently, it progressively integrated imaging information, demographic characteristics and follow-up data to pinpoint suspicious malignant nodules and refine the risk scale. The multidimensional system achieved a state-of-the-art performance with an area under the curve (AUC) of 0.918 (95% confidence interval (CI) 0.918–0.919) on the internal testing dataset, outperforming the single-dimensional approach (AUC of 0.881, 95% CI 0.880–0.882). Moreover, C-Lung-RADS exhibited a superior sensitivity compared with Lung-RADS v2022 (87.1% versus 63.3%) in an independent cohort, which was screened using mobile computed tomography scanners to broaden screening accessibility in resource-constrained settings. With its foundation in precise risk stratification and tailored management, this system has minimized unnecessary invasive procedures for low-risk cases and recommended prompt intervention for extremely high-risk nodules to avert diagnostic delays. This approach has the potential to enhance the decision-making paradigm and facilitate a more efficient diagnosis of lung cancer during routine checkups as well as screening scenarios.

Background: To investigate clinicopathological variables influencing overall survival, overall recurrence, and post-recurrence survival (PRS) in patients who experienced curative-intent surgical resection of stage I non-small-cell lung cancer (NSCLC). Methods: We investigated a series of 1387 patients with stage I NSCLC who underwent surgical resection from 2008 to 2015. The effect clinicopathological factors on death, recurrence, and PRS were evaluated by Kaplan-Meier estimates and cox regression analysis. Results: Among the 1387 stage I patients, 301 (21.7%) experienced recurrence. The 5-year cumulative incidence of recurrence (CIR) for all patients was 20.2% and median PRS was 25.5 months. The older age (P = 0.036), p-stage IB (P = 0.001), sublobar resection(P<0.001), histology subtype (P<0.001), and lymphovascular invasion (LVI) (P = 0.042) were significantly associated with worse overall survival. Among 301 recurrent patients, univariable analysis indicated that p-stage IB (versus IA) (P<0.001), LVI (P<0.001) and visceral pleural invasion (VPI) (P<0.001) were remarkably correlated with the higher incidence of recurrence. Taking the effect of clinicopathological variables on PRS into consideration, smoking history (P = 0.043), non-adenocarcinoma (P = 0.013), high architectural grade of LUAD (P = 0.019), EGFR wild status (P = 0.002), bone metastasis (P =0.040) and brain metastasis (P = 0.042) were substantially related with poorer PRS. Multivariate analysis demonstrated that high architectural grade of LUAD (P = 0.008), brain metastasis (P = 0.010) and bone metastasis (P = 0.043) were independently associated with PRS. Conclusion: In patients with resected stage I NSCLC, the older age, p-stage IB (versus IA), sublobar resection, histology subtype, and LVI were significantly associated with worse overall survival. P-stage IB (versus IA), LVI, and VPI were significantly correlated with the higher incidence of recurrence. High architectural grade of LUAD, brain metastasis and bone metastasis were independent risk factors with PRS.

Background: Circular RNAs (circRNAs) are differentially expressed between normal and cancerous tissues, contributing to tumor initiation and progression. However, comprehensive landscape of dysregulated circRNAs across cancer types remains unclear. Methods: In this study, we conducted Ribo-Zero transcriptome sequencing on tumor tissues and their adjacent normal samples including glioblastoma, esophageal squamous cell carcinoma, lung adenocarcinoma, thyroid cancer, colorectal cancer, gastric cancer and hepatocellular carcinoma. CIRCexplorer2 was employed to identify circRNAs and dysregulated circRNAs and genes were determined by DESeq2 package. The expression of hsa_circ_0072309 (circLIFR) was measured by reverse transcription and quantitative real-time PCR, and its effect on cell migration was examined by Transwell and wound healing assays. The role of circLIFR in tumor metastasis was evaluated via mouse models of tail-vein injection and spleen injection for lung and liver metastasis, respectively. Results: Distinct circRNA expression signatures were identified among seven types of solid tumors, and the dysregulated circRNAs exhibited cancer-specific expression or shared common expression signatures across cancers. Bioinformatics analyses indicated that aberrant expression of host genes and/or RNA-binding proteins contributed to circRNA dysregulation in cancer. Finally, circLIFR was experimentally validated to be downregulated in six solid tumors and to significantly inhibit cell migration in vitro and tumor metastasis in vivo. Conclusions: Our results provide a comprehensive landscape of differentially expressed circRNAs in solid tumors and highlight that circRNAs are extensively involved in cancer pathogenesis.