Prediction of non-suicidal self-injury in adolescents at the family level using regression methods and machine learning.

BACKGROUND: Adolescent non-suicidal self-injury (NSSI) is a major public health issue. Family factors are significantly associated with NSSI in adolescents, while studies on forecasting NSSI at the family level are still limited. In addition to regression methods, machine learning (ML) techniques have been recommended to improve the accuracy of family-level risk prediction for NSSI. METHODS: Using a dataset of 7967 students and their primary caregivers from a cross-sectional study, logistic regression model and random forest model were used to test the forecasting accuracy of NSSI predictions at the family level. Cross-validation was used to assess model prediction performance, including the area under the receiver operator curve (AUC), precision, Brier score, accuracy, sensitivity, specificity, positive predictive value and negative predictive value. RESULTS: The top three important family-related predictors within the random forest algorithm included family function (importance:42.66), family conflict (importance:42.18), and parental depression (importance:27.21). The most significant family-related risk predictors and protective predictors identified by the logistic regression model were family history of mental illness (OR:2.25) and help-seeking behaviors of mental distress from parents (OR:0.65), respectively. The AUCs of the two models, logistic regression and random forest, were 0.852 and 0.835, respectively. LIMITATIONS: The key limitation is that this cross-sectional survey only enabled the authors to examine predictors that were considered to be proximal rather than distal. CONCLUSIONS: These findings highlight the significance of family-related factors in forecasting NSSI in adolescents. Combining both conventional statistical methods and ML methods to improve risk assessment of NSSI at the family level deserves attention.

Mutations involving TGFB and MAPK may be associated with malignancy in granular cell tumors.

Granular cell tumors (GrCTs) are mesenchymal neoplasms of presumed schwannian differentiation that may present as solitary or multifocal lesions with excision usually being curative. A minority of cases, however, show histological features associated with an increased risk for metastasis and are highly aggressive leading to death in about a third of cases. While benign and malignant cases have been shown to harbor mutations in the H + ATPase genes, there is only limited data examining molecular aberrations associated with malignancy. The departmental archives were searched for cases of atypical/malignant GrCTs. Clinical and histopathological features were noted. Whole-exome sequencing was performed. Three cases of malignant GrCTs and one case of atypical GrCTs were included. All three malignant tumors metastasized to distant sites with a median disease-free survival of 16 months and an overall follow-up time of 35 months. Whole-exome sequencing showed mutations involving TGFß and MAPK pathways in all four tumors. Although the cohort size is small, our preliminary findings suggest that mutations involving the TGFß and MAPK pathways may be associated with tumor progression or malignant transformation in GrCT pathogenesis.

Machine Learning-Based Prediction Models for Depression Symptoms Among Chinese Healthcare Workers During the Early COVID-19 Outbreak in 2020: A Cross-Sectional Study.

Background: The 2019 novel coronavirus (COVID-19)-related depression symptoms of healthcare workers have received worldwide recognition. Although many studies identified risk exposures associated with depression symptoms among healthcare workers, few have focused on a predictive model using machine learning methods. As a society, governments, and organizations are concerned about the need for immediate interventions and alert systems for healthcare workers who are mentally at-risk. This study aims to develop and validate machine learning-based models for predicting depression symptoms using survey data collected during the COVID-19 outbreak in China. Method: Surveys were conducted of 2,574 healthcare workers in hospitals designated to care for COVID-19 patients between 20 January and 11 February 2020. The patient health questionnaire (PHQ)-9 was used to measure the depression symptoms and quantify the severity, a score of ≥5 on the PHQ-9 represented depression symptoms positive, respectively. Four machine learning approaches were trained (75% of data) and tested (25% of data). Cross-validation with 100 repetitions was applied to the training dataset for hyperparameter tuning. Finally, all models were compared to evaluate their predictive performances and screening utility: decision tree, logistics regression with least absolute shrinkage and selection operator (LASSO), random forest, and gradient-boosting tree. Results: Important risk predictors identified and ranked by the machine learning models were highly consistent: self-perceived health status factors always occupied the top five most important predictors, followed by worried about infection, working on the frontline, a very high level of uncertainty, having received any form of psychological support material and having COVID-19-like symptoms. The area under the curve [95% CI] of machine learning models were as follows: LASSO model, 0.824 [0.792-0.856]; random forest, 0.828 [0.797-0.859]; gradient-boosting tree, 0.829 [0.798-0.861]; and decision tree, 0.785 [0.752-0.819]. The calibration plot indicated that the LASSO model, random forest, and gradient-boosting tree fit the data well. Decision curve analysis showed that all models obtained net benefits for predicting depression symptoms. Conclusions: This study shows that machine learning prediction models are suitable for making predictions about mentally at-risk healthcare workers predictions in a public health emergency setting. The application of multidimensional machine learning models could support hospitals' and healthcare workers' decision-making on possible psychological interventions and proper mental health management.

Chromosome 8 gain is associated with high-grade transformation in MPNST.

One of the most common malignancies affecting adults with Neurofibromatosis type 1 (NF1) is the malignant peripheral nerve sheath tumor (MPNST), an aggressive and often fatal sarcoma that commonly arises from benign plexiform neurofibromas. Despite advances in our understanding of MPNST pathobiology, there are few effective therapeutic options, and no investigational agents have proven successful in clinical trials. To further understand the genomic heterogeneity of MPNST, and to generate a preclinical platform that encompasses this heterogeneity, we developed a collection of NF1-MPNST patient-derived xenografts (PDX). These PDX were compared with the primary tumors from which they were derived using copy number analysis, whole exome sequencing, and RNA sequencing. We identified chromosome 8 gain as a recurrent genomic event in MPNST and validated its occurrence by FISH in the PDX and parental tumors, in a validation cohort, and by single-cell sequencing in the PDX. Finally, we show that chromosome 8 gain is associated with inferior overall survival in soft-tissue sarcomas. These data suggest that chromosome 8 gain is a critical event in MPNST pathogenesis and may account for the aggressive nature and poor outcomes in this sarcoma subtype.

Spermine synthase and MYC cooperate to maintain colorectal cancer cell survival by repressing Bim expression.

Dysregulation of polyamine metabolism has been linked to the development of colorectal cancer (CRC), but the underlying mechanism is incompletely characterized. Here, we report that spermine synthase (SMS), a polyamine biosynthetic enzyme, is overexpressed in CRC. Targeted disruption of SMS in CRC cells results in spermidine accumulation, which inhibits FOXO3a acetylation and allows subsequent translocation to the nucleus to transcriptionally induce expression of the proapoptotic protein Bim. However, this induction is blunted by MYC-driven expression of miR-19a and miR-19b that repress Bim production. Pharmacological or genetic inhibition of MYC activity in SMS-depleted CRC cells dramatically induces Bim expression and apoptosis and causes tumor regression, but these effects are profoundly attenuated by silencing Bim. These findings uncover a key survival signal in CRC through convergent repression of Bim expression by distinct SMS- and MYC-mediated signaling pathways. Thus, combined inhibition of SMS and MYC signaling may be an effective therapy for CRC.

Predictive and targeting value of IGFBP-3 in therapeutically resistant prostate cancer.

BACKGROUND: Our previous studies demonstrated that a novel quinazoline derivative, DZ-50, inhibited prostate cancer epithelial cell invasion and survival by targeting insulin-like-growth factor binding protein-3 (IGFBP-3) and mediating epithelial-mesenchymal transition (EMT) conversion to mesenchymal-epithelial transition (MET). This study investigated the therapeutic value of DZ-50 agent in in vitro and in vivo models of advanced prostate cancer and the ability of the compound to overcome resistance to antiandrogen (enzalutamide) in prostate tumors. APPROACH: LNCaP and LNCaP-enzalutamide resistant human prostate cancer (LNCaP-ER) cells, as well as 22Rv1 and enzalutamide resistant, 22Rv1-ER were used as cell models. The effects of DZ-50 and the antiandrogen, enzalutamide (as single agents or in combination) on cell death, EMT-MET interconversion, and expression of IGFBP3 and the androgen receptor (AR), were examined. The TRAMP mouse model of prostate cancer progression was used as a pre-clinical model. Transgenic mice (20-wks of age) were treated with DZ-50 (100 mg/kg for 2 wks, oral gavage daily) and prostate tumors were subjected to immunohistochemical assessment of apoptosis, cell proliferation, markers of EMT and differentiation and IGFBP-3 and AR expression. A tissue microarray (TMA) was analyzed for expression of IGBP-3, the target of DZ-50 and its association with tumor progression and biochemical recurrence. RESULTS: We found that treatment with DZ-50 enhanced the anti-tumor response to the antiandrogen via promoting EMT to MET interconversion, in vitro. This DZ-50-mediated phenotypic reversal to MET leads to prostate tumor re-differentiation in vivo, by targeting nuclear IGFBP-3 expression (without affecting AR). Analysis of human prostate cancer specimens and TCGA patient cohorts revealed that overexpression of IGBP-3 protein correlated with tumor recurrence and poor patient survival. CONCLUSIONS: These findings provide significant new insights into (a) the predictive value of IGFBP-3 in prostate cancer progression and (b) the antitumor action of DZ-50, [in combination or sequencing with enzalutamide] as a novel approach for the treatment of therapeutically resistant prostate cancer.