Novel data-mining approach identifies biomarkers for diagnosis of Kawasaki disease

From molecular mechanisms of prostate cancer to translational applications: based on multi-omics fusion analysis and intelligent medicine

Prostate cancer is the most common cancer in men worldwide and has a high mortality rate. The complex and heterogeneous development of prostate cancer has become a core obstacle in the treatment of prostate cancer. Simultaneously, the issues of overtreatment in early-stage diagnosis, oligometastasis and dormant tumor recognition, as well as personalized drug utilization, are also specific concerns that require attention in the clinical management of prostate cancer. Some typical genetic mutations have been proved to be associated with prostate cancer's initiation and progression. However, single-omic studies usually are not able to explain the causal relationship between molecular alterations and clinical phenotypes. Exploration from a systems genetics perspective is also lacking in this field, that is, the impact of gene network, the environmental factors, and even lifestyle behaviors on disease progression. At the meantime, current trend emphasizes the utilization of artificial intelligence (AI) and machine learning techniques to process extensive multidimensional data, including multi-omics. These technologies unveil the potential patterns, correlations, and insights related to diseases, thereby aiding the interpretable clinical decision making and applications, namely intelligent medicine. Therefore, there is a pressing need to integrate multidimensional data for identification of molecular subtypes, prediction of cancer progression and aggressiveness, along with perosonalized treatment performing. In this review, we systematically elaborated the landscape from molecular mechanism discovery of prostate cancer to clinical translational applications. We discussed the molecular profiles and clinical manifestations of prostate cancer heterogeneity, the identification of different states of prostate cancer, as well as corresponding precision medicine practices. Taking multi-omics fusion, systems genetics, and intelligence medicine as the main perspectives, the current research results and knowledge-driven research path of prostate cancer were summarized.

A machine learning model for distinguishing Kawasaki disease from sepsis

KD is an acute systemic vasculitis that most commonly affects children under 5 years old. Sepsis is a systemic inflammatory response syndrome caused by infection. The main clinical manifestations of both are fever, and laboratory tests include elevated WBC count, C-reactive protein, and procalcitonin. However, the two treatments are very different. Therefore, it is necessary to establish a dynamic nomogram based on clinical data to help clinicians make timely diagnoses and decision-making. In this study, we analyzed 299 KD patients and 309 sepsis patients. We collected patients' age, sex, height, weight, BMI, and 33 biological parameters of a routine blood test. After dividing the patients into a training set and validation set, the least absolute shrinkage and selection operator method, support vector machine and receiver operating characteristic curve were used to select significant factors and construct the nomogram. The performance of the nomogram was evaluated by discrimination and calibration. The decision curve analysis was used to assess the clinical usefulness of the nomogram. This nomogram shows that height, WBC, monocyte, eosinophil, lymphocyte to monocyte count ratio (LMR), PA, GGT and platelet are independent predictors of the KD diagnostic model. The c-index of the nomogram in the training set and validation is 0.926 and 0.878, which describes good discrimination. The nomogram is well calibrated. The decision curve analysis showed that the nomogram has better clinical application value and decision-making assistance ability. The nomogram has good performance of distinguishing KD from sepsis and is helpful for clinical pediatricians to make early clinical decisions.

An Artificial Intelligence Derived Blood Test to Diagnose Kawasaki Disease

OBJECTIVE

To develop a highly sensitive and specific blood biomarker panel that identifies febrile children with Kawasaki disease (KD).

METHODS

We tested blood samples from a single-center cohort of KD (n = 50) and control febrile children (n = 100) to develop a biomarker panel from 11 candidates selected by their assay clinical availability. We used machine learning with least absolute shrinkage and selection operator regression to identify 11 blood markers with values incorporated into a model, which provided a binary predictive risk score for KD determined with Youden's index. We further reduced the model using least angle regression.

RESULTS

Using 10-fold cross-validation with least absolute shrinkage and selection operator regression on these 11 readouts plus patient age resulted in an area under the receiver operating characteristic curve of 0.94 (95% confidence interval [CI]: 0.90-0.98; P <.01). Using Youden's index, which provided an optimal cut off for a binary predictive risk score, 88 of 97 KD-negative patients were diagnosed negative, and 47 of 50 KD-positive patients were positive, yielding a sensitivity of 0.94 (95% CI: 0.87-1.0) and specificity of 0.91 (95% CI: 0.85-0.96). Least angle regression reduced the final panel to 3 biomarkers: C-reactive protein, NT-proB-type natriuretic peptide, and thyroid hormone uptake. The predictive model then provided an area under the receiver operating characteristic curve of 0.92 (95% CI: 0.87-0.96; P <.001) along with sensitivity and specificity at 86% each.

CONCLUSIONS

Machine learning identified a highly accurate diagnostic model for KD. The reduced model employs 3 biomarkers currently approved by regulatory bodies and performed on platforms commonly used by certified diagnostic laboratories.

Phenomapping approach to interpreting coronary dimensions in febrile children

Importance

Coronary artery dilation may occur in febrile children with and without Kawasaki disease (KD).

Objective

We explored the application of unsupervised learning algorithms in the detection of novel patterns of coronary artery phenotypes in febrile children with and without KD.

Methods

A total of 239 febrile children (59 non-KD and 180 KD patients), were recruited. Unsupervised hierarchical clustering analysis of phenotypic data including age, hemoglobin, white cell count, platelet count, C-reactive protein, erythrocyte sedimentation rate, albumin, alanine aminotransferase, aspartate aminotransferase, and coronary artery z scores were performed.

Results

Using a cutoff z score of 2.5, the specificity was 98.3% and the sensitivity was 22.1% for differentiating non-KD from KD patients. Clustering analysis identified three phenogroups that differed in a clinical, laboratory, and echocardiographic parameters. Compared with phenogroup I, phenogroup III had the highest prevalence of KD (91%), worse inflammatory markers, more deranged liver function, higher coronary artery z scores, and lower hematocrit and albumin levels. Abnormal blood parameters in febrile children with z scores of coronary artery segments <0.5 and 0.5-1.5 was associated with increased risks of having KD to 8.7 (P = 0.003) and 4.4 (P = 0.002), respectively.

Interpretation

Phenomapping of febrile children with and without KD identified useful laboratory parameters that aid the diagnosis of KD in febrile children with relatively normal-sized coronary arteries.

Single center blind testing of a US multi-center validated diagnostic algorithm for Kawasaki disease in Taiwan

Background

Kawasaki disease (KD) is the leading cause of acquired heart disease in children. The major challenge in KD diagnosis is that it shares clinical signs with other childhood febrile control (FC) subjects. We sought to determine if our algorithmic approach applied to a Taiwan cohort.

Methods

A single center (Chang Gung Memorial Hospital in Taiwan) cohort of patients suspected with acute KD were prospectively enrolled by local KD specialists for KD analysis. Our previously single-center developed computer-based two-step algorithm was further tested by a five-center validation in US. This first blinded multi-center trial validated our approach, with sufficient sensitivity and positive predictive value, to identify most patients with KD diagnosed at centers across the US. This study involved 418 KDs and 259 FCs from the Chang Gung Memorial Hospital in Taiwan.

Findings

Our diagnostic algorithm retained sensitivity (379 of 418; 90.7%), specificity (223 of 259; 86.1%), PPV (379 of 409; 92.7%), and NPV (223 of 247; 90.3%) comparable to previous US 2016 single center and US 2020 fiver center results. Only 4.7% (15 of 418) of KD and 2.3% (6 of 259) of FC patients were identified as indeterminate. The algorithm identified 18 of 50 (36%) KD patients who presented 2 or 3 principal criteria. Of 418 KD patients, 157 were infants younger than one year and 89.2% (140 of 157) were classified correctly. Of the 44 patients with KD who had coronary artery abnormalities, our diagnostic algorithm correctly identified 43 (97.7%) including all patients with dilated coronary artery but one who found to resolve in 8 weeks.

Interpretation

This work demonstrates the applicability of our algorithmic approach and diagnostic portability in Taiwan.

Future Biomarkers for Infection and Inflammation in Febrile Children

Febrile patients, suffering from an infection, inflammatory disease or autoimmunity may present with similar or overlapping clinical symptoms, which makes early diagnosis difficult. Therefore, biomarkers are needed to help physicians form a correct diagnosis and initiate the right treatment to improve patient outcomes following first presentation or admittance to hospital. Here, we review the landscape of novel biomarkers and approaches of biomarker discovery. We first discuss the use of current plasma parameters and whole blood biomarkers, including results obtained by RNA profiling and mass spectrometry, to discriminate between bacterial and viral infections. Next we expand upon the use of biomarkers to distinguish between infectious and non-infectious disease. Finally, we discuss the strengths as well as the potential pitfalls of current developments. We conclude that the use of combination tests, using either protein markers or transcriptomic analysis, have advanced considerably and should be further explored to improve current diagnostics regarding febrile infections and inflammation. If proven effective when combined, these biomarker signatures will greatly accelerate early and tailored treatment decisions.

Computer-aided biomarker discovery for precision medicine: data resources, models and applications

Biomarkers are a class of measurable and evaluable indicators with the potential to predict disease initiation and progression. In contrast to disease-associated factors, biomarkers hold the promise to capture the changeable signatures of biological states. With methodological advances, computer-aided biomarker discovery has now become a burgeoning paradigm in the field of biomedical science. In recent years, the 'big data' term has accumulated for the systematical investigation of complex biological phenomena and promoted the flourishing of computational methods for systems-level biomarker screening. Compared with routine wet-lab experiments, bioinformatics approaches are more efficient to decode disease pathogenesis under a holistic framework, which is propitious to identify biomarkers ranging from single molecules to molecular networks for disease diagnosis, prognosis and therapy. In this review, the concept and characteristics of typical biomarker types, e.g. single molecular biomarkers, module/network biomarkers, cross-level biomarkers, etc., are explicated on the guidance of systems biology. Then, publicly available data resources together with some well-constructed biomarker databases and knowledge bases are introduced. Biomarker identification models using mathematical, network and machine learning theories are sequentially discussed. Based on network substructural and functional evidences, a novel bioinformatics model is particularly highlighted for microRNA biomarker discovery. This article aims to give deep insights into the advantages and challenges of current computational approaches for biomarker detection, and to light up the future wisdom toward precision medicine and nation-wide healthcare.

Biomarkers for the Discrimination of Acute Kawasaki Disease From Infections in Childhood

Background: Kawasaki disease (KD) is a vasculitis of early childhood mimicking several infectious diseases. Differentiation between KD and infectious diseases is essential as KD's most important complication-the development of coronary artery aneurysms (CAA)-can be largely avoided by timely treatment with intravenous immunoglobulins (IVIG). Currently, KD diagnosis is only based on clinical criteria. The aim of this study was to evaluate whether routine C-reactive protein (CRP) and additional inflammatory parameters myeloid-related protein 8/14 (MRP8/14 or S100A8/9) and human neutrophil-derived elastase (HNE) could distinguish KD from infectious diseases. Methods and Results: The cross-sectional study included KD patients and children with proven infections as well as febrile controls. Patients were recruited between July 2006 and December 2018 in Europe and USA. MRP8/14, CRP, and HNE were assessed for their discriminatory ability by multiple logistic regression analysis with backward selection and receiver operator characteristic (ROC) curves. In the discovery cohort, the combination of MRP8/14+CRP discriminated KD patients (n = 48) from patients with infection (n = 105), with area under the ROC curve (AUC) of 0.88. The HNE values did not improve discrimination. The first validation cohort confirmed the predictive value of MRP8/14+CRP to discriminate acute KD patients (n = 26) from those with infections (n = 150), with an AUC of 0.78. The second validation cohort of acute KD patients (n = 25) and febrile controls (n = 50) showed an AUC of 0.72, which improved to 0.84 when HNE was included. Conclusion: When used in combination, the plasma markers MRP8/14, CRP, and HNE may assist in the discrimination of KD from both proven and suspected infection.

A Random Forests Quantile Classifier for Class Imbalanced Data

Extending previous work on quantile classifiers (q-classifiers) we propose the q*-classifier for the class imbalance problem. The classifier assigns a sample to the minority class if the minority class conditional probability exceeds 0 < q* < 1, where q* equals the unconditional probability of observing a minority class sample. The motivation for q*-classification stems from a density-based approach and leads to the useful property that the q*-classifier maximizes the sum of the true positive and true negative rates. Moreover, because the procedure can be equivalently expressed as a cost-weighted Bayes classifier, it also minimizes weighted risk. Because of this dual optimization, the q*-classifier can achieve near zero risk in imbalance problems, while simultaneously optimizing true positive and true negative rates. We use random forests to apply q*-classification. This new method which we call RFQ is shown to outperform or is competitive with existing techniques with respect to tt-mean performance and variable selection. Extensions to the multiclass imbalanced setting are also considered.

Proteomic profiles by multiplex microsphere suspension array

Advances in high-throughput proteomic approaches have provided substantial momentum to novel disease-biomarker discovery research and have augmented the quality of clinical studies. Applications based on multiplexed microsphere suspension array technology are making strong in-roads into the clinical diagnostic/prognostic practice. Conventional proteomic approaches are designed to discover a broad set of proteins that are associated with a specific medical condition. In comparison, multiplex microsphere immunoassays use quantitative measurements of selected set(s) of specific/particular molecular markers such as cytokines, chemokines, pathway signaling or disease-specific markers for detection, metabolic disorders, cancer, and infectious agents causing human, plant and animal diseases. This article provides a foundation to the multiplexed microsphere suspension array technology, with an emphasis on the improvements in the technology, data analysis approaches, and applications to translational and clinical research with implications for personalized and precision medicine.

Relationship between mean platelet volume-to-lymphocyte ratio and coronary artery abnormalities in Kawasaki disease

OBJECTIVES

Recently, mean platelet volume-to-lymphocyte ratio has emerged as a novel parameter of inflammation. No study has investigated the role of mean platelet volume-to-lymphocyte ratio in children with Kawasaki disease. We aimed to evaluate the relationship between mean platelet volume-to-lymphocyte ratio and coronary artery abnormalities in Kawasaki disease.

METHODS

Between January 2008 and January 2017, a total of 58 children with Kawasaki disease and 42 healthy subjects matched for sex and age were enrolled. Before the treatment, transthoracic echocardiography for all children was performed. Clinical and laboratory results including mean platelet volume, platelet distribution width, red blood cell distribution width, and counts of platelets, neutrophils, lymphocytes, and white blood cells, erythrocyte sedimentation rate, and C-reactive protein levels were measured. Mean platelet volume-to-lymphocyte ratio was calculated as mean platelet volume divided by lymphocyte count.

RESULTS

Compared with healthy controls, mean platelet volume-to-lymphocyte ratio was significantly lower in the children with Kawasaki disease (p<0.01). A total of 14 patients (24.1%) had incomplete Kawasaki disease and 15 (25.8%) children with Kawasaki disease had coronary involvement. Mean platelet volume-to-lymphocyte ratio was significantly lower in patients with coronary artery abnormalities (p<0.01). According to receiver operating characteristic curve analysis performed for the prediction of coronary artery abnormalities, the best cut-off point for mean platelet volume-to-lymphocyte ratio was 2.5 (area under curve=0.593, sensitivity 53.3%, specificity 51.1%).

CONCLUSION

It was first shown that the children with Kawasaki disease have lower mean platelet volume-to-lymphocyte ratio compared with control subjects. Mean platelet volume-to-lymphocyte ratio may be helpful in predicting coronary artery lesions in patients with Kawasaki disease.

Novel Risk Assessment Tool for Immunoglobulin Resistance in Kawasaki Disease: Application Using a Random Forest Classifier

BACKGROUND

Resistance to intravenous immunoglobulin (IVIG) therapy is a risk factor for coronary lesions in patients with Kawasaki disease (KD). Risk-adjusted initial therapy may improve coronary outcome in KD, but identification of high risk patients remains a challenge. This study aimed to develop a new risk assessment tool for IVIG resistance using advanced statistical techniques.

METHODS

Data were retrospectively collected from KD patients receiving IVIG therapy, including demographic characteristics, signs and symptoms of KD and laboratory results. A random forest (RF) classifier, a tree-based machine learning technique, was applied to these data. The correlation between each variable and risk of IVIG resistance was estimated.

RESULTS

Data were obtained from 767 patients with KD, including 170 (22.1%) who were refractory to initial IVIG therapy. The predictive tool based on the RF algorithm had an area under the receiver operating characteristic curve of 0.916, a sensitivity of 79.7% and a specificity of 87.3%. Its misclassification rate in the general patient population was estimated to be 15.5%. RF also identified markers related to IVIG resistance such as abnormal liver markers and percentage neutrophils, displaying relationships between these markers and predicted risk.

CONCLUSIONS

The RF classifier reliably identified KD patients at high risk for IVIG resistance, presenting clinical markers relevant to treatment failure. Evaluation in other patient populations is required to determine whether this risk assessment tool relying on RF has clinical value.

Risk factors and implications of progressive coronary dilatation in children with Kawasaki disease

Background

Kawasaki disease (KD) is an acute systemic vasculitis that occurs in children and may lead to cardiovascular morbidity and mortality. Progressive coronary dilatation for at least 2 months is associated with worse late coronary outcomes in patients with KD having medium or giant aneurysms. However, the risk factors and occurrence of progressive coronary dilatation in patients with KD but without medium or giant aneurysms have been insufficiently explored.

Methods

We retrospectively enrolled 169 patients with KD from a tertiary medical center in Taiwan during 2009–2013. Medical records of all patients were reviewed. Echocardiography was performed during the acute KD phase and at 3–4 weeks, 6–8 weeks, 6 months, and 12 months after KD onset. Progressive coronary dilatation was defined as the progressive enlargement of coronary arteries on three consecutive echocardiograms. Logistic regression analysis was conducted to evaluate the potential risk factors for coronary aneurysms and progressive coronary dilatation.

Results

Of a total of 169 patients with KD, 31 (18.3%) had maximal coronary Z-scores of ≥ + 2.5 during the acute KD phase, 16 (9.5%; male/female: 9/7) had coronary aneurysms at 1 month after KD onset, and 5 (3.0%) satisfied the definition of progressive coronary dilatation. Multivariate logistic regression analysis revealed that an initial maximal coronary Z-score of ≥ + 2.5 [odds ratio (OR): 5.24, 95% confidence interval (CI): 1.31–21.3, P = 0.020] and hypoalbuminemia (OR: 4.83, 95% CI: 1.11–20.9, P = 0.035) were independent risk factors for coronary aneurysms and were significantly associated with progressive coronary dilatation. However, the association between intravenous immunoglobulin unresponsiveness and the development of coronary aneurysms at 1 month after KD onset didn’t reach the level of significance (P = 0.058).

Conclusions

In the present study, 3% (5/169) of patients with KD had progressive coronary dilatation, which was associated with persistent coronary aneurysms at 1 year after KD onset. Initial coronary dilatation and hypoalbuminemia were independently associated with the occurrence of progressive coronary dilatation. Therefore, such patients may require intensive cardiac monitoring and adjuvant therapies apart from immunoglobulin therapies.

Revisiting the role of environmental and climate factors on the epidemiology of Kawasaki disease

Can environmental factors, such as air-transported preformed toxins, be of key relevance to the health outcomes of poorly understood human ailments (e.g., rheumatic diseases such as vasculitides, some inflammatory diseases, or even severe childhood acquired heart diseases)? Can the physical, chemical, or biological features of air masses be linked to the emergence of diseases such as Kawasaki disease (KD), Henoch-Schönlein purpura, Takayasu's aortitis, and ANCA-associated vasculitis? These diseases surprisingly share some common epidemiological features. For example, they tend to appear as clusters of cases grouped geographically and temporarily progress in nonrandom sequences that repeat every year in a similar way. They also show concurrent trend changes within regions in countries and among different world regions. In this paper, we revisit transdisciplinary research on the role of environmental and climate factors in the epidemiology of KD as a paradigmatic example of this group of diseases. Early-warning systems based on environmental alerts, if successful, could be implemented as a way to better inform patients who are predisposed to, or at risk for, developing KD. Further research on the etiology of KD could facilitate the development of vaccines and specific medical therapies.

Genome-Wide Expression Profiling Reveals S100B as Biomarker for Invasive Aspergillosis

Invasive aspergillosis (IA) is a devastating opportunistic infection and its treatment constitutes a considerable burden for the health care system. Immunocompromised patients are at an increased risk for IA, which is mainly caused by the species Aspergillus fumigatus. An early and reliable diagnosis is required to initiate the appropriate antifungal therapy. However, diagnostic sensitivity and accuracy still needs to be improved, which can be achieved at least partly by the definition of new biomarkers. Besides the direct detection of the pathogen by the current diagnostic methods, the analysis of the host response is a promising strategy toward this aim. Following this approach, we sought to identify new biomarkers for IA. For this purpose, we analyzed gene expression profiles of hematological patients and compared profiles of patients suffering from IA with non-IA patients. Based on microarray data, we applied a comprehensive feature selection using a random forest classifier. We identified the transcript coding for the S100 calcium-binding protein B (S100B) as a potential new biomarker for the diagnosis of IA. Considering the expression of this gene, we were able to classify samples from patients with IA with 82.3% sensitivity and 74.6% specificity. Moreover, we validated the expression of S100B in a real-time reverse transcription polymerase chain reaction (RT-PCR) assay and we also found a down-regulation of S100B in A. fumigatus stimulated DCs. An influence on the IL1B and CXCL1 downstream levels was demonstrated by this S100B knockdown. In conclusion, this study covers an effective feature selection revealing a key regulator of the human immune response during IA. S100B may represent an additional diagnostic marker that in combination with the established techniques may improve the accuracy of IA diagnosis.