SlHB8 Is a Novel Factor in Enhancing Cold Resistance in Tomato Anthers by Modulating Tapetal Cell Death.

Tomato plants favor warmth, making them particularly susceptible to cold conditions, especially their reproductive development. Therefore, understanding how pollen reacts to cold stress is vital for selecting and improving cold-resistant tomato varieties. The programmed cell death (PCD) in the tapetum is particularly susceptible to cold temperatures which could hinder the degradation of the tapetal layer in the anthers, thus affecting pollen development. However, it is not clear yet how genes integral to tapetal degradation respond to cold stress. Here, we report that SlHB8, working upstream of the conserved genetic module DYT1-TDF1-AMS-MYB80, is crucial for regulating cold tolerance in tomato anthers. SlHB8 expression increases in the tapetum when exposed to low temperatures. CRISPR/Cas9-generated SlHB8-knockout mutants exhibit improved pollen cold tolerance due to the reduced temperature sensitivity of the tapetum. SlHB8 directly upregulates SlDYT1 and SlMYB80 by binding to their promoters. In normal anthers, cold treatment boosts SlHB8 levels, which then elevates the expression of genes like SlDYT1, SlTDF1, SlAMS, and SlMYB80; however, slhb8 mutants do not show this gene activation during cold stress, leading to a complete blockage of delayed tapetal programmed cell death (PCD). Furthermore, we found that SlHB8 can interact with both SlTDF1 and SlMYB80, suggesting the possibility that SlHB8 might regulate tapetal PCD at the protein level. This study sheds light on molecular mechanisms of anther adaptation to temperature fluctuations.

Genome-Wide Identification of GH17s Family Genes and Biological Function Analysis of SlA6 in Tomato.

Glycoside hydrolases (GHs), enzymes that break down glycosidic bonds in carbohydrates and between carbohydrates and non-carbohydrates, are prevalent in plants, animals, microorganisms, and other organisms. The tomato is a significant crop that contains the GH17 gene family. However, its role in tomatoes has yet to be fully investigated. In this study, we identified 43 GH17 genes from the tomato genome, distributed unevenly across 12 chromosomes. We further analyzed their gene structure, phylogenetic relationships, promoter elements, and expression patterns. The promoter element analysis indicated their potential roles in response to biotic and abiotic stresses as well as phytohormone effects on growth and development. The expression studies across different tomato tissues revealed that 10 genes were specifically expressed in floral organs, with SlA6 prominently expressed early during bud formation. By using CRISPR/Cas9 gene-editing technology, SlA6 knockout plants were generated. Phenotypic characterization showed that pollen viability, pollen tube germination, fruit weight, and seed number were significantly reduced in the Sla6 mutant, but the soluble solids content (TSS) was significantly higher in the Sla6 mutant, suggesting that SlA6 affects pollen development and fruit quality.

The hormone regulatory mechanism underlying parthenocarpic fruit formation in tomato.

Parthenocarpic fruits, known for their superior taste and reliable yields in adverse conditions, develop without the need for fertilization or pollination. Exploring the physiological and molecular mechanisms behind parthenocarpic fruit development holds both theoretical and practical significance, making it a crucial area of study. This review examines how plant hormones and MADS-box transcription factors control parthenocarpic fruit formation. It delves into various aspects of plant hormones-including auxin, gibberellic acid, cytokinins, ethylene, and abscisic acid-ranging from external application to biosynthesis, metabolism, signaling pathways, and their interplay in influencing parthenocarpic fruit development. The review also explores the involvement of MADS family gene functions in these processes. Lastly, we highlight existing knowledge gaps and propose directions for future research on parthenocarpy.

A copper missile-triggered power coalescence and death vortex within tumor cell mitochondria for synergistic cuproptosis/phototherapy/chemotherapy.

The importance of copper homeostasis in mitochondria and copper-triggered modality of mitochondrial cell death have been confirmed. However, the existing copper-based nanoplatforms are focused on synergistic therapies while the intracellular therapeutic targets are relatively scattered. Effective integration of all targets within mitochondria to generate power coalescence remains a challenge. Herein, we developed a novel copper-based delivery system to trigger power coalescence and death vortex within tumor cell mitochondria. Specifically, a mitochondrial targeting "copper missile" loaded with curcumin (termed as Cur@CuS-TPP-HA, CCTH) was designed for cuproptosis/phototherapy/chemotherapy synergistic anti-tumor therapy. Once the CCTH NPs are shuttled to the mitochondria, near-infrared (NIR) irradiation initiates the release of copper ions and curcumin for in situ drug accumulation in cancer cell mitochondria. An excess of copper ions and curcumin can activate cuproptosis and mitochondrial apoptosis pathways, respectively. When combined, they can cause an increase in reactive oxygen species (ROS), damage to mitochondrial DNA (mt-DNA), and a decrease in energy supply, thereby leading to a "vicious circle" of mitochondrial damage that further enhances the tumor-killing efficacy. As a consequence, this "copper missile" exhibits advanced anti-tumor effects as verified through in vitro assessments and in vivo evaluations using the 4T1 breast tumor model, providing a promising approach for cuproptosis-based synergistic anti-tumor therapy.

Recent research advances in the development of Dabie Banda virus vaccines.

Severe fever with thrombocytopenia syndrome (SFTS) is a newly identified tick-borne viral hemorrhagic fever caused by Dabie Banda virus (DBV). The virus was first discovered in eastern China in 2009 and is now considered an infectious disease with a mortality rate ranging from 6.3% to 30%. The best strategy for controlling SFTS is to develop effective vaccines. However, no approved vaccines are currently available to prevent this disease, despite the number of extensive and in-depth studies conducted on DBV in the past few years. This review focuses on the structure of DBV and the induced host immune responses which are the fundamental factors in vaccine development, and thoroughly summarizes the current research progress on DBV vaccines. The developing DBV vaccines include protein subunit vaccines, live attenuated vaccines, recombinant virus vector vaccines, and DNA vaccines. At present, almost all candidate vaccines for DBV are in the laboratory development or preclinical stages. There remain challenges in successfully developing clinically approved DBV vaccines.

MAZ-mediated up-regulation of BCKDK reprograms glucose metabolism and promotes growth by regulating glucose-6-phosphate dehydrogenase stability in triple-negative breast cancer.

Tumour metabolic reprogramming is pivotal for tumour survival and proliferation. Investigating potential molecular mechanisms within the heterogeneous and clinically aggressive triple-negative breast cancer (TNBC) subtype is essential to identifying novel therapeutic targets. Accordingly, we investigated the role of branched-chain α-keto acid dehydrogenase kinase (BCKDK) in promoting tumorigenesis in TNBC. We analysed The Cancer Genome Atlas dataset and immunohistochemically stained surgical specimens to investigate BCKDK expression and its prognostic implications in TNBC. The effects of BCKDK on tumorigenesis were assessed using cell viability, colony formation, apoptosis, and cell cycle assays, and subsequently validated in vivo. Metabolomic screening was performed via isotope tracer studies. The downstream target was confirmed using mass spectrometry and a co-immunoprecipitation experiment coupled with immunofluorescence analysis. Upstream transcription factors were also examined using chromatin immunoprecipitation and luciferase assays. BCKDK was upregulated in TNBC tumour tissues and associated with poor prognosis. BCKDK depletion led to reduced cell proliferation both in vitro and vivo. MYC-associated zinc finger protein (MAZ) was confirmed as the major transcription factor directly regulating BCKDK expression in TNBC. Mechanistically, BCKDK interacted with glucose-6-phosphate dehydrogenase (G6PD), leading to increased flux in the pentose phosphate pathway for macromolecule synthesis and detoxification of reactive oxygen species. Forced expression of G6PD rescued the growth defect in BCKDK-deficient cells. Notably, the small-molecule inhibitor of BCKDK, 3,6-dichlorobenzo(b)thiophene-2-carboxylic acid, exhibited anti-tumour effects in a patient-derived tumour xenograft model. Our findings hold significant promise for developing targeted therapies aimed at disrupting the MAZ/BCKDK/G6PD signalling pathway, offering potential advancements in treating TNBC through metabolic reprogramming.

Spectrum and characteristics of germline PALB2 pathogenic variants in 1556 early-onset breast cancer patients in China.

PURPOSE: Limited data are available regarding the partner and localizer of BRCA2 (PALB2) in Chinese patients with early breast cancer. This study aimed to assess the spectrum and characteristics of germline PALB2 pathogenic variants in this population. METHODS: Peripheral blood samples were collected from 1556 patients diagnosed with BRCA1/2-negative early-onset breast cancer. All coding regions and exon‒intron boundaries of the PALB2 genes were screened through next-generation sequencing. RESULTS: The prevalence of PALB2 pathogenic variants was approximately 0.77% in the cohort. Eleven PALB2 pathogenic variants were identified in twelve participants, including five frameshift mutations and six nonsense mutations. All other variants were detected once, except for PALB2 c.1056_1057del (detected twice). Two PALB2 carriers (2/12, 16.7%) have documented family history of breast cancer and/or ovarian cancer. Patients with a positive family history exhibited a threefold higher possibility of being identified as PALB2 carriers than those without a family history (2% vs. 0.69%), although the difference was not statistically significant (p = 0.178). Compared to non-carriers, PALB2 carriers has a tendency to appear in younger age (≤ 30 years) (25% vs 14.4%), human epidermal growth factor receptor-2 (HER2)-negative status (83.3% vs. 70.2%), and diagnosed with invasive micropapillary carcinoma (16.7% vs 3.1%). CONCLUSION: The prevalence of the germline PALB2 pathogenic variants was approximately 0.77% in Chinese patients with BRCA1/2-negative early-onset breast cancer. Our findings is crucial for understanding population-specific genetic risks and offering insights that can enhance genetic counseling and genetic testing strategies in this population.

YY1 mediated DCUN1D5 transcriptional activation promotes triple-negative breast cancer progression by targeting FN1/PI3K/AKT pathway.

Triple-negative breast cancer (TNBC) is more aggressive and has a higher metastasis rate compared with other subtypes of breast cancer. Due to the lack of drug-targetable receptors, chemotherapy is now the only available systemic treatment for TNBC. However, some patients might still develop drug resistance and have poor prognosis. Therefore, novel molecular biomarkers and new treatment targets are urgently needed for patients with TNBC. To provide molecular insights into TNBC progression, we investigated the function and the underlying mechanism of Defective in cullin neddylation 1 domain containing 5 (DCUN1D5) in the regulation of TNBC. By TCGA dataset and surgical specimens with immunohistochemical (IHC) staining method, DCUN1D5 was identified to be significantly upregulated in TNBC tumor tissues and negatively associated with prognosis. A series of in vitro and in vivo experiments were performed to confirm the oncogenic role of DCUN1D5 in TNBC. Overexpression of FN1 or PI3K/AKT activator IGF-1 could restore the proliferative and invasive ability induced by DCUN1D5 knockdown and DCUN1D5 could act as a novel transcriptional target of transcription factor Yin Yang 1 (YY1). In conclusion, YY1-enhanced DCUN1D5 expression could promote TNBC progression by FN1/PI3K/AKT pathway and DCUN1D5 might be a potential prognostic biomarker and therapeutic target for TNBC treatment.

Linearization of wavelength sweeping lasers for the construction of 4-D FMCW LiDAR images of slow-moving objects using baseband beat note signals.

A FMCW LiDAR system of both the distributed feedback laser and external cavity laser is established in baseband beat notes, rather than up-conversion to an intermediate frequency to exclude flicker noise. Meanwhile, utilizing fast-scanning MEMS mirrors, high-quality real-time (1 fps) 4-D images of the slow-moving object (10 mm/s) can be directly constructed at the baseband with a central frequency as low as 100 kHz and a small Doppler shift. The proposed LiDAR architecture based on such a low-frequency baseband significantly improves the optical power budget on the transmitter side and eliminates the costly high-speed sampling circuits on the receiver side.

Tracing unknown tumor origins with a biological-pathway-based transformer model.

Cancer of unknown primary (CUP) represents metastatic cancer where the primary site remains unidentified despite standard diagnostic procedures. To determine the tumor origin in such cases, we developed BPformer, a deep learning method integrating the transformer model with prior knowledge of biological pathways. Trained on transcriptomes from 10,410 primary tumors across 32 cancer types, BPformer achieved remarkable accuracy rates of 94%, 92%, and 89% in primary tumors and primary and metastatic sites of metastatic tumors, respectively, surpassing existing methods. Additionally, BPformer was validated in a retrospective study, demonstrating consistency with tumor sites diagnosed through immunohistochemistry and histopathology. Furthermore, BPformer was able to rank pathways based on their contribution to tumor origin identification, which helped to classify oncogenic signaling pathways into those that are highly conservative among different cancers versus those that are highly variable depending on their origins.

Non-invasive screening and subtyping for breast cancer by serum SERS combined with LGB-DNN algorithms.

Early detection of breast cancer and its molecular subtyping is crucial for guiding clinical treatment and improving survival rate. Current diagnostic methods for breast cancer are invasive, time consuming and complicated. In this work, an optical detection method integrating surface-enhanced Raman spectroscopy (SERS) technology with feature selection and deep learning algorithm was developed for identifying serum components and building diagnostic model, with the aim of efficient and accurate noninvasive screening of breast cancer. First, the high quality of serum SERS spectra from breast cancer (BC), breast benign disease (BBD) patients and healthy controls (HC) were obtained. Chi-square tests were conducted to exclude confounding factors, enhancing the reliability of the study. Then, LightGBM (LGB) algorithm was used as the base model to retain useful features to significantly improve classification performance. The DNN algorithm was trained through backpropagation, adjusting the weights and biases between neurons to improve the network's predictive ability. In comparison to traditional machine learning algorithms, this method provided more accurate information for breast cancer classification, with classification accuracies of 91.38 % for BC and BBD, and 96.40 % for BC, BBD, and HC. Furthermore, the accuracies of 90.11 % for HR+/HR- and 88.89 % for HER2+/HER2- can be reached when evaluating BC patients' molecular subtypes. These results demonstrate that serum SERS combined with powerful LGB-DNN algorithm would provide a supplementary method for clinical breast cancer screening.

Additional prognostic value of polymorphisms within the 3'-untranslated region of programmed cell death pathway genes in early-stage breast cancer.

Introduction: The programmed cell death (PCD) pathway plays an important role in restricting cancer cell survival and proliferation. However, limited studies have investigated the association between genetic variants in the 3'-untranslated region of the PCD pathway genes and breast cancer outcomes. Methods: In this study, we genotyped 28 potentially functional single nucleotide polymorphisms (SNPs) in 23 PCD pathway genes in 1,177 patients with early-stage breast cancer (EBC) from a Han Chinese population. The median follow-up period was 174 months. Results: Among all the candidate SNPs, four independent SNPs (rs4900321 and rs7150025 in ATG2B, rs6753785 in BCL2L11, and rs2213181 in c-Kit) were associated with invasive disease-free survival (iDFS), distant disease-free survival (DDFS), breast cancer-specific survival (BCSS) and overall survival (OS), respectively. Further combined genotypes of these four SNPs revealed that the survival decreased as the number of unfavorable genotypes increased (Ptrend = 1.0 × 10-6, 8.5 × 10-8, 3.6 × 10-4, and 1.3 × 10-4 for iDFS, DDFS, BCSS, and OS, respectively). Receiver operating characteristic curve analysis demonstrated that incorporating unfavorable genotypes and clinicopathological variables improved the ability to predict EBC survival (P = 0.006, 0.004, 0.029, and 0.019 for iDFS, DDFS, BCSS, and OS, respectively). Additionally, rs6753785 and rs2213181 were associated with BCL2L11 and c-Kit mRNA expression, respectively. Conclusions: Our results suggest that these four SNPs may act as novel biomarkers for EBC survival, possibly by modulating the expression of the corresponding genes.

Incretin-based drugs decrease the incidence of prostate cancer in type 2 diabetics: A pooling-up analysis.

Incretin-based drugs, a class of Antidiabetic medications (ADMs) used in the treatment of type 2 diabetes, may affect the incidence of prostate cancer (PCa). But real-world evidence for this possible effect is lacking. Therefore, the aim of this study is to assess the effect of incretin-based drugs on the incidence of PCa, including glucagon-like peptide-1 (GLP-1) receptor agonists and dipeptidyl peptidase-4 (DPP-4) inhibitors. We searched PubMed, Embase, and Cochrane Library databases for eligible studies through September 2023. Two independent reviewers performed screening and data extraction. We used the Cochrane Handbook for Systematic Reviews and the Newcastle-Ottawa Scale (NOS) to assess the quality of included randomized controlled trials (RCTs) and cohort studies. We did a meta-analysis of available trial data to calculate overall risk ratios (RRs) for PCa. A total of 1238 articles were identified in our search. After screening for eligibility, 7 high-quality studies met the criteria for meta-analysis, including 2 RCTs and 5 cohort studies, with a total of 1165,738 patients. Compared with the control group, we found that incretin-based drugs reduced the relative risk of PCa by 35% (95% confidence interval (CI), 0.17-0.49; P = .0006). In subgroup analysis, the RR values for GLP-1 receptor agonists and DPP-4 inhibitors were 62% (95% CI, 0.45-0.85; P = .003) and 72% (95% CI, 0.46-1.12; P = .14), respectively. Incretin-based drugs are associated with lower incidence of prostate cancer and may have a preventive effect on prostate cancer in patients with type 2 diabetes.

Clinical outcomes of coronavirus disease in patients with breast cancer treated with granulocyte colony-stimulating factor following chemotherapy: Triangulation of evidence using population-based cohort and Mendelian randomization analyses.

Recombinant human granulocyte colony-stimulating factor (G-CSF) administration in patients with cancer and coronavirus disease (COVID-19) remains controversial. Concerns exist that it may worsen COVID-19 outcomes by triggering an inflammatory cytokine storm, despite its common use for managing chemotherapy-induced neutropenia (CIN) or febrile neutropenia post-chemotherapy. Here, we determined whether prophylactic or therapeutic G-CSF administration following chemotherapy exacerbates COVID-19 progression to severe/critical conditions in breast cancer patients with COVID-19. Between December 2022 and February 2023, all 503 enrolled breast cancer patients had concurrent COVID-19 and received G-CSF post-chemotherapy, with most being vaccinated pre-chemotherapy. We prospectively observed COVID-19-related adverse outcomes, conducted association analyses, and subsequently performed Mendelian randomization (MR) analyses to validate the causal effect of genetically predicted G-CSF or its associated granulocyte traits on COVID-19 adverse outcomes. Only 0.99% (5/503) of breast cancer patients experienced COVID-19-related hospitalization following prophylactic or therapeutic G-CSF administration after chemotherapy. No mortality or progression to severe/critical COVID-19 occurred after G-CSF administration. Notably, no significant associations were observed between the application, dosage, or response to G-CSF and COVID-19-related hospitalization (all p >.05). Similarly, the MR analyses showed no evidence of causality of genetically predicted G-CSF or related granulocyte traits on COVID-19-related hospitalization or COVID-19 severity (all p >.05). There is insufficient evidence to substantiate the notion that the prophylactic or therapeutic administration of G-CSF after chemotherapy for managing CIN in patients with breast cancer and COVID-19 would worsen COVID-19 outcomes, leading to severe or critical conditions, or even death, especially considering the context of COVID-19 vaccination.

PhosMap: An ensemble bioinformatic platform to empower interactive analysis of quantitative phosphoproteomics.

BACKGROUND: Liquid chromatography-mass spectrometry (LC-MS)-based quantitative phosphoproteomics has been widely used to detect thousands of protein phosphorylation modifications simultaneously from the biological specimens. However, the complicated procedures for analyzing phosphoproteomics data has become a bottleneck to widening its application. METHODS: Here, we develop PhosMap, a versatile and scalable tool to accomplish phosphoproteomics data analysis. A standardized phosphorylation data format was created for data analyses, from data preprocessing to downstream bioinformatic analyses such as dimension reduction, differential phosphorylation analysis, kinase activity, survival analysis, and so on. For better usability, we distribute PhosMap as a Docker image for easy local deployment upon any of Windows, Linux, and Mac system. RESULTS: The source code is deposited at https://github.com/BADD-XMU/PhosMap. A free PhosMap webserver (https://huggingface.co/spaces/Bio-Add/PhosMap), with easy-to-follow fashion of dashboards, is curated for interactive data analysis. CONCLUSIONS: PhosMap fills the technical gap of large-scale phosphorylation research by empowering researchers to process their own phosphoproteomics data expediently and efficiently, and facilitates better data interpretation.

An Energy-Efficient FD-fNIRS Readout Circuit Employing a Mixer-First Analog Frontend and a Σ-Δ Phase-to-Digital Converter.

This paper presents a low-power frequency-domain functional near-infrared spectroscopy (FD-fNIRS) readout circuit for the absolute value measurement of tissue optical characteristics. The paper proposes a mixer-first analog front-end (AFE) structure and a 1-bit Σ-Δ phase-to-digital converter (PDC) to reduce the required circuit bandwidth and the laser modulation frequency, thereby saving power while maintaining high resolution. The proposed chip achieves sub-0.01° phase resolution and consumes 6.8 mW of power. Nine optical solid phantoms are produced to evaluate the chip. Compared to a self-built high-precision measurement platform that combines a network analyzer with an avalanche photodiode (APD) module, the maximum measuring errors of the absorption coefficient and reduced scattering coefficient are 10.6% and 12.3%, respectively.

Healthy lifestyles, systemic inflammation and breast cancer risk: a mediation analysis.

BACKGROUND: Despite the known association between healthy lifestyles and reduced risk of breast cancer, it remains unclear whether systemic inflammation, as a consequence of unhealthy lifestyles, may mediate the association. METHODS: A cohort study of 259,435 female participants in the UK Biobank was conducted to estimate hazard ratio (HR) for breast cancer according to 9 inflammation markers using Cox regression models. We further estimated the percentage of total association between healthy lifestyle index (HLI) and breast cancer that is mediated by these inflammation markers. RESULTS: During 2,738,705 person-years of follow-up, 8,889 cases of breast cancer were diagnosed among 259,435 women in the UK Biobank cohort. Higher level of C-reactive protein (CRP), systemic immune-inflammation index (SII), CRP-to-albumin Ratio (CAR), CRP-to-lymphocyte Ratio (CLR), monocyte-to-HDL-c ratio (MHR), and neutrophil-to-HDL-c ratio (NHR) were associated with increased breast cancer risk, while a higher lymphocyte-to-monocyte ratio (LMR) was associated with a lower risk. The inverse association between HLI and breast cancer was weakly mediated by CRP (8.5%), SII (1.71%), CAR (8.66%), CLR (6.91%), MHR (6.27%), and NHR (7.33%). When considering individual lifestyle factors, CRP and CAR each mediated 16.58% and 17.20%, respectively, of the associations between diet score and breast cancer risk, while the proportion mediated for physical activity and breast cancer were 12.13% and 11.48%, respectively. Furthermore, MHR was found to mediate 13.84% and 12.01% of the associations between BMI, waist circumference, and breast cancer. CONCLUSION: The association of HLI and breast cancer is weakly mediated by the level of inflammation, particularly by CRP and CAR. Systemic inflammatory status may be an intermediate in the biological pathway of breast cancer development.

SIMarker: Cellular similarity detection and its application to diagnosis and prognosis of liver cancer.

BACKGROUND: The emergence of single-cell technology offers a unique opportunity to explore cellular similarity and heterogeneity between precancerous diseases and solid tumors. However, there is lacking a systematic study for identifying and characterizing similarities at single-cell resolution. METHODS: We developed SIMarker, a computational framework to detect cellular similarities between precancerous diseases and solid tumors based on gene expression at single-cell resolution. Taking hepatocellular carcinoma (HCC) as a case study, we quantified the cellular and molecular connections between HCC and cirrhosis. Core analysis modules of SIMarker is publicly available at https://github.com/xmuhuanglab/SIMarker ("SIM" means "similarity" and "Marker" means "biomarkers). RESULTS: We found PGA5+ hepatocytes in HCC showed cirrhosis-like characteristics, including similar transcriptional programs and gene regulatory networks. Consequently, the genes constituting the gene expression program of these cirrhosis-like subpopulations were designated as cirrhosis-like signatures (CLS). Strikingly, our utilization of CLS enabled the development of diagnosis and prognosis biomarkers based on within-sample relative expression orderings of gene pairs. These biomarkers achieved high precision and concordance compared with previous studies. CONCLUSIONS: Our work provides a systematic method to investigate the clinical translational significance of cellular similarities between HCC and cirrhosis, which opens avenues for identifying similar paradigms in other categories of cancers and diseases.

Comparative efficacy and safety of extended adjuvant endocrine therapy for hormone receptor-positive early breast cancer: a Bayesian network meta-analysis.

PURPOSE: Optimal extended adjuvant endocrine therapy (ET) duration and strategy for hormone receptor-positive (HR +) early breast cancer remain unclear. In this network meta-analysis (NMA), the efficacy and safety of all available extended adjuvant ETs were compared and ranked. METHODS: PubMed, Embase, and Cochrane Library and abstracts presented at ASCO, SABCS, and ESMO were searched on March 5, 2022. Fourteen randomized controlled trials (RCTs) comprising eight extended adjuvant ETs for HR + breast cancer and 38,070 patients were analyzed. Main outcomes were disease-free survival (DFS), overall survival (OS), grade ≥ 3 adverse events (AEs), and contralateral breast cancer (CBC). Direct and indirect comparisons were integrated via Bayesian NMA. Hierarchical cluster analysis was performed to jointly rank efficacy and safety outcomes. RESULTS: Compared with that of 5 year ET, extended 10 year aromatase inhibitor (AI) treatment provided the greatest DFS benefit (HR = 0.45, 95%CrI 0.23-0.83), whereas no strategy differed significantly in terms of the other main outcomes. Extended 10 year AI treatment was the preferred strategy for DFS improvement and CBC prevention (surface under the cumulative ranking curve: 93.51% and 91.29% probability, respectively). All strategies had comparable safeties (grade ≥ 3 AEs). Compared with that of 5 year ET, 10 year extended AI significantly increased arthralgia (OR = 1.65, 95%CrI 1.02-2.93) and osteoporosis (OR = 3.33, 95%CrI 1.19-9.68). CONCLUSION: Extended 10 year AI therapy may be optimal for HR + early breast cancer given its relatively high efficacy and safety.

Union With Recursive Feature Elimination: A Feature Selection Framework to Improve the Classification Performance of Multicategory Causes of Death in Colorectal Cancer.

Despite the use of machine learning tools, it is challenging to properly model cause-specific deaths in colorectal cancer (CRC) patients and choose appropriate treatments. Here, we propose an interesting feature selection framework, namely union with recursive feature elimination (U-RFE), to select the union feature sets that are crucial in CRC progression-specific mortality using The Cancer Genome Atlas (TCGA) dataset. Based on the union feature sets, we compared the performance of 5 classification algorithms, including logistic regression (LR), support vector machines (SVM), random forest (RF), eXtreme gradient boosting (XGBoost), and Stacking, to identify the best model for classifying 4-category deaths. In the first stage of U-RFE, LR, SVM, and RF were used as base estimators to obtain subsets containing the same number of features but not exactly the same specific features. Union analysis of the subsets was then performed to determine the final union feature set, effectively combining the advantages of different algorithms. We found that the U-RFE framework could improve various models' performance. Stacking outperformed LR, SVM, RF, and XGBoost in most scenarios. When the target feature number of the RFE was set to 50 and the union feature set contained 298 deterministic features, the Stacking model achieved F1_weighted, Recall_weighted, Precision_weighted, Accuracy, and Matthews correlation coefficient of 0.851, 0.864, 0.854, 0.864, and 0.717, respectively. The performance of the minority categories was also significantly improved. Therefore, this recursive feature elimination-based approach of feature selection improves performances of classifying CRC deaths using clinical and omics data or those using other data with high feature redundancy and imbalance.