OMICS in Chronic Kidney Disease: Focus on Prognosis and Prediction

Chronic kidney disease (CKD) patients are characterized by a high residual risk for cardiovascular (CV) events and CKD progression. This has prompted the implementation of new prognostic and predictive biomarkers with the aim of mitigating this risk. The ‘omics’ techniques, namely genomics, proteomics, metabolomics, and transcriptomics, are excellent candidates to provide a better understanding of pathophysiologic mechanisms of disease in CKD, to improve risk stratification of patients with respect to future cardiovascular events, and to identify CKD patients who are likely to respond to a treatment. Following such a strategy, a reliable risk of future events for a particular patient may be calculated and consequently the patient would also benefit from the best available treatment based on their risk profile. Moreover, a further step forward can be represented by the aggregation of multiple omics information by combining different techniques and/or different biological samples. This has already been shown to yield additional information by revealing with more accuracy the exact individual pathway of disease.

Biological heterogeneity in idiopathic pulmonary arterial hypertension identified through unsupervised transcriptomic profiling of whole blood

Idiopathic pulmonary arterial hypertension (IPAH) is a rare but fatal disease diagnosed by right heart catheterisation and the exclusion of other forms of pulmonary arterial hypertension, producing a heterogeneous population with varied treatment response. Here we show unsupervised machine learning identification of three major patient subgroups that account for 92% of the cohort, each with unique whole blood transcriptomic and clinical feature signatures. These subgroups are associated with poor, moderate, and good prognosis. The poor prognosis subgroup is associated with upregulation of the ALAS2 and downregulation of several immunoglobulin genes, while the good prognosis subgroup is defined by upregulation of the bone morphogenetic protein signalling regulator NOG, and the C/C variant of HLA-DPA1/DPB1 (independently associated with survival). These findings independently validated provide evidence for the existence of 3 major subgroups (endophenotypes) within the IPAH classification, could improve risk stratification and provide molecular insights into the pathogenesis of IPAH.

Multiple Approaches at Admission Based on Lung Ultrasound and Biomarkers Improves Risk Identification in COVID-19 Patients

Background: Risk stratification of COVID-19 patients is fundamental to improving prognosis and selecting the right treatment. We hypothesized that a combination of lung ultrasound (LUZ-score), biomarkers (sST2), and clinical models (PANDEMYC score) could be useful to improve risk stratification. Methods: This was a prospective cohort study designed to analyze the prognostic value of lung ultrasound, sST2, and PANDEMYC score in COVID-19 patients. The primary endpoint was in-hospital death and/or admission to the intensive care unit. The total length of hospital stay, increase of oxygen flow, or escalated medical treatment during the first 72 h were secondary endpoints. Results: a total of 144 patients were included; the mean age was 57.5 ± 12.78 years. The median PANDEMYC score was 243 (52), the median LUZ-score was 21 (10), and the median sST2 was 53.1 ng/mL (30.9). Soluble ST2 showed the best predictive capacity for the primary endpoint (AUC = 0.764 (0.658–0.871); p = 0.001), towards the PANDEMYC score (AUC = 0.762 (0.655–0.870); p = 0.001) and LUZ-score (AUC = 0.749 (0.596–0.901); p = 0.002). Taken together, these three tools significantly improved the risk capacity (AUC = 0.840 (0.727–0.953); p ≤ 0.001). Conclusions: The PANDEMYC score, lung ultrasound, and sST2 concentrations upon admission for COVID-19 are independent predictors of intra-hospital death and/or the need for admission to the ICU for mechanical ventilation. The combination of these predictive tools improves the predictive power compared to each one separately. The use of decision trees, based on multivariate models, could be useful in clinical practice.

Establishment and Validation of a Gene Signature-Based Prognostic Model to Improve Survival Prediction in Adrenocortical Carcinoma Patients

Background. The current guideline for the management of adrenocortical carcinoma (ACC) is insufficient for accurate risk prediction to guide adjuvant therapy. Given frequent and severe therapeutic side effects, a better estimate of survival is warranted for risk-specific assignment to adjuvant treatment. We attempted to construct an integrated model based on a prognostic gene signature and clinicopathological features to improve risk stratification and survival prediction in ACC. Methods. Using a series of bioinformatic and statistical approaches, a gene-expression signature was established and validated in two independent cohorts. By combining the signature with clinicopathological features, a decision tree was generated to improve risk stratification, and a nomogram was constructed to personalize risk prediction. Time-dependent receiver operating characteristic (tROC) and calibration analysis were performed to evaluate the predictive power and accuracy. Results. A three-gene signature could discriminate high-risk patients well in both training and validation cohorts. Multivariate regression analysis demonstrated the signature to be an independent predictor of overall survival. The decision tree could identify risk subgroups powerfully, and the nomogram showed high accuracy of survival prediction. Particularly, expression of a gene hitherto unknown to be dysregulated in ACC, TIGD1, was shown to be prognostically relevant. Conclusion. We propose a novel gene signature to guide decision-making about adjuvant therapy in ACC. The score shows unprecedented survival prediction and hence constitutes a huge step towards personalized management. As a secondary important finding, we report the discovery and validation of a new oncogene, TIGD1, which was consistently overexpressed in ACC. TIGD1 might shed further light on the biology of ACC and might give rise to targeted therapies that not only apply to ACC but potentially also to other malignancies.

Whole-exome analysis of adolescents with low VWF and heavy menstrual bleeding identifies novel genetic associations

Adolescents with low von Willebrand factor (VWF) levels and heavy menstrual bleeding (HMB) experience significant morbidity. There is a need to better genetically characterize these patients and improve our understanding of the pathophysiology of bleeding. We performed whole-exome sequencing on 86 post-menarchal patients diagnosed with low-VWF levels (30-50 IU/dL) and HMB and compared them to 660 in-house controls. We compared the number of rare stop-gain/stop-loss and rare ClinVar pathogenic variants between cases and controls, as well as performed gene-burden and gene-set burden analyses. We found an enrichment in cases of rare stop-gain/stop-loss variants in genes involved in bleeding disorders, and an enrichment of rare ClinVar pathogenic variants in genes involved in anemias. The two most significant genes in the gene-burden analysis, CFB and DNASE2, are associated with atypical hemolytic uremia (aHUS) and severe anemia, respectively. VWF also surpassed exome-wide significance in the gene-burden analysis (p=7.31x10-6). Gene-set burden analysis revealed an enrichment of rare nonsynonymous variants in cases in several hematologically relevant pathways. Further, common variants in FERMT2, a gene involved in regulation of hemostasis and angiogenesis surpassed genome-wide significance. We demonstrate that adolescents with HMB and low-VWF have an excess of rare nonsynonymous and pathogenic variants in genes involved in disorders of bleeding and anemia. Variants of variable penetrance in these genes may contribute to the spectrum of phenotypes observed in HMB patients, and could partially explain the bleeding phenotype. By identifying the HMB patients who possess these variants, we may be able to improve risk stratification and patient outcomes.

The ReIMAGINE Multimodal Warehouse: Using Artificial Intelligence for Accurate Risk Stratification of Prostate Cancer

Introduction. Prostate cancer (PCa) is the most frequent cancer diagnosis in men worldwide. Our ability to identify those men whose cancer will decrease their lifespan and/or quality of life remains poor. The ReIMAGINE Consortium has been established to improve PCa diagnosis.Materials and methods. MRI will likely become the future cornerstone of the risk-stratification process for men at risk of early prostate cancer. We will, for the first time, be able to combine the underlying molecular changes in PCa with the state-of-the-art imaging. ReIMAGINE Screening invites men for MRI and PSA evaluation. ReIMAGINE Risk includes men at risk of prostate cancer based on MRI, and includes biomarker testing.Results. Baseline clinical information, genomics, blood, urine, fresh prostate tissue samples, digital pathology and radiomics data will be analysed. Data will be de-identified, stored with correlated mpMRI disease endotypes and linked with long term follow-up outcomes in an instance of the Philips Clinical Data Lake, consisting of cloud-based software. The ReIMAGINE platform includes application programming interfaces and a user interface that allows users to browse data, select cohorts, manage users and access rights, query data, and more. Connection to analytics tools such as Python allows statistical and stratification method pipelines to run profiling regression analyses. Discussion. The ReIMAGINE Multimodal Warehouse comprises a unique data source for PCa research, to improve risk stratification for PCa and inform clinical practice. The de-identified dataset characterized by clinical, imaging, genomics and digital pathology PCa patient phenotypes will be a valuable resource for the scientific and medical community.

Multi-ethnic polygenic risk modifies the association between APOL1 high risk genotypes and chronic kidney disease

The burden of advanced chronic kidney disease (CKD) falls disproportionately on minorities including African Americans (AAs) and Hispanic Americans (HAs) with admixed ancestry. Even though APOL1 high-risk genotypes increase risk of kidney disease, their penetrance is incomplete, indicating that the modification of APOL1 high risk may be polygenic. For this study, we used three multi-ethnic cohorts with APOL1 high risk genotypes and calculated a multi-ethnic PRS using publicly available summary statistics. We show that CKD risk is significantly modified by a multi-ethnic polygenic risk score. Standardizing population screening for CKD by including APOL1 high-risk genotypes and polygenic risk score may improve risk stratification and outcomes.

Development and Validation of the Pyroptosis-related genes Signature for Predicting the Prognosis of Colorectal Cancer

Background: Pyroptosis is an important component of the tumor microenvironment, associated with the occurrence and progression of cancer. However, the expression of pyroptosis-related genes and its impact on the prognosis of colon cancer (CC) remains unclear. Here, we constructed and validated a pyroptosis-related genes signature to predict the prognosis of patients with CC.Methods: Public data source was obtained to screen out candidate genes for further analysis. Various methods were combined to construct a robust pyroptosis-related genes signature for predicting the prognosis of patients with CC. Based on the gene signature and clinical features, a decision tree and nomogram were developed to improve risk stratification and quantify risk assessment for individual patients.Results: The pyroptosis-related genes signature successfully discriminated CC patients with high-risk in the training cohorts. The prognostic value of this signature was further confirmed in independent validation cohort. Multivariable Cox regression and stratified survival analysis revealed this signature was an independent prognostic factor for CC patients. The decision tree identified risk subgroups powerfully, and the nomogram incorporating the gene signature and clinical risk factors performed well in the calibration plots.Conclusions: Pyroptosis-related genes signature was an independent prognostic factor, and can be used to predict the prognosis of patients with CC.

Prospective cohort study of the predictive value of inflammatory biomarkers over clinical variables in children and young people with cancer presenting with fever and neutropenia

Introduction Fever during chemotherapy induced neutropenia is a common and potentially life-threatening complication of the treatment of childhood cancer. Predictions of poor outcome could be enhanced by incorporating serum biomarkers of inflammation at presentation and reassessment. Methods A prospective cohort study was conducted of children under 18 years old, being treated for cancer or a cancer-like condition, who presented with fever (≥ 38.0°C) and neutropenia (neutrophil count < 0.5*109/L). Clinical features were recorded, along with three experimental inflammatory biomarkers: procalcitonin (PCT), interleukin-6 (IL-6) and interleukin-8 (IL-8). Outcomes included serious medical complications (SMC): any infection related mortality, critical care and organ support, severe sepsis, septic shock, significant microbiologically defined infection, or radiologically confirmed pneumonia. Results Biomarker assessments were undertaken in 43 episodes of fever and neutropenia, from 31 patients aged between four months and 17 years old (median six years): 20 were female and 22 had acute leukaemia. Five episodes of SMC were noted. PCT, IL-6 and IL-8 had poor individual discriminatory ability (C-statistic 0.48 to 0.60) and did not add to the value of clinical risk stratification tools. Insufficient data were collected to formally assess the value of repeated assessments. Conclusions Incorporating serum biomarkers of inflammation at presentation of episodes of fever with neutropenia in childhood does not clearly improve risk stratification. Repeated assessments over time may be of value.

Machine learning approaches improve risk stratification for secondary cardiovascular disease prevention in multiethnic patients

ObjectivesIdentifying high-risk patients is crucial for effective cardiovascular disease (CVD) prevention. It is not known whether electronic health record (EHR)-based machine-learning (ML) models can improve CVD risk stratification compared with a secondary prevention risk score developed from randomised clinical trials (Thrombolysis in Myocardial Infarction Risk Score for Secondary Prevention, TRS 2°P).MethodsWe identified patients with CVD in a large health system, including atherosclerotic CVD (ASCVD), split into 80% training and 20% test sets. A rich set of EHR patient features was extracted. ML models were trained to estimate 5-year CVD event risk (random forests (RF), gradient-boosted machines (GBM), extreme gradient-boosted models (XGBoost), logistic regression with an L2 penalty and L1 penalty (Lasso)). ML models and TRS 2°P were evaluated by the area under the receiver operating characteristic curve (AUC).ResultsThe cohort included 32 192 patients (median age 74 years, with 46% female, 63% non-Hispanic white and 12% Asian patients and 23 475 patients with ASCVD). There were 4010 events over 5 years of follow-up. ML models demonstrated good overall performance; XGBoost demonstrated AUC 0.70 (95% CI 0.68 to 0.71) in the full CVD cohort and AUC 0.71 (95% CI 0.69 to 0.73) in patients with ASCVD, with comparable performance by GBM, RF and Lasso. TRS 2°P performed poorly in all CVD (AUC 0.51, 95% CI 0.50 to 0.53) and ASCVD (AUC 0.50, 95% CI 0.48 to 0.52) patients. ML identified nontraditional predictive variables including education level and primary care visits.ConclusionsIn a multiethnic real-world population, EHR-based ML approaches significantly improved CVD risk stratification for secondary prevention.