Variable selection methods for developing a biomarker panel for prediction of dengue hemorrhagic fever

Endothelial Dysfunction, HMGB1, and Dengue: An Enigma to Solve

Dengue is a viral infection caused by dengue virus (DENV), which has a significant impact on public health worldwide. Although most infections are asymptomatic, a series of severe clinical manifestations such as hemorrhage and plasma leakage can occur during the severe presentation of the disease. This suggests that the virus or host immune response may affect the protective function of endothelial barriers, ultimately being considered the most relevant event in severe and fatal dengue pathogenesis. The mechanisms that induce these alterations are diverse. It has been suggested that the high mobility group box 1 protein (HMGB1) may be involved in endothelial dysfunction. This non-histone nuclear protein has different immunomodulatory activities and belongs to the alarmin group. High concentrations of HMGB1 have been detected in patients with several infectious diseases, including dengue, and it could be considered as a biomarker for the early diagnosis of dengue and a predictor of complications of the disease. This review summarizes the main features of dengue infection and describes the known causes associated with endothelial dysfunction, highlighting the involvement and possible relationship between HMGB1 and DENV.

Combination of inflammatory and vascular markers in the febrile phase of dengue is associated with more severe outcomes

Background

Early identification of severe dengue patients is important regarding patient management and resource allocation. We investigated the association of 10 biomarkers (VCAM-1, SDC-1, Ang-2, IL-8, IP-10, IL-1RA, sCD163, sTREM-1, ferritin, CRP) with the development of severe/moderate dengue (S/MD).

Methods

We performed a nested case-control study from a multi-country study. A total of 281 S/MD and 556 uncomplicated dengue cases were included.

Results

On days 1-3 from symptom onset, higher levels of any biomarker increased the risk of developing S/MD. When assessing together, SDC-1 and IL-1RA were stable, while IP-10 changed the association from positive to negative; others showed weaker associations. The best combinations associated with S/MD comprised IL-1RA, Ang-2, IL-8, ferritin, IP-10, and SDC-1 for children, and SDC-1, IL-8, ferritin, sTREM-1, IL-1RA, IP-10, and sCD163 for adults.

Conclusions

Our findings assist the development of biomarker panels for clinical use and could improve triage and risk prediction in dengue patients.

Funding

This study was supported by the EU's Seventh Framework Programme (FP7-281803 IDAMS), the WHO, and the Bill and Melinda Gates Foundation.

Age-dependent co-dependency structure of biomarkers in the general population of the United States

Phenotypic biomarkers (e.g. cholesterol, weight, and glucose) are important to diagnose and treat diseases associated with aging. However, while many biomarkers are co-dependent (e.g. glycohemoglobin and glucose), it is generally unknown how age influences their co-dependence. In the following, we analyzed 50 biomarkers in 27,508 National Health and Nutrition Examination Survey (NHANES) participants (age range: 20 to 80, mean age: 46.3 years old, sexes: 48.9% males, 51.1% females, ethnicities: 46.0% Whites, 27.8% Hispanics, 20.0% non-Hispanic Blacks, 6.1% others) to investigate how the co-dependency structure of common biomarkers evolves with age and whether differences exist between sexes and ethnicities. First, we associated the change in correlations between biomarkers with chronological age. We identified six trends and replicated our top finding (height vs. systolic blood pressure) in participants of the UK Biobank (N=470,895). We found that, on average, correlations tend to decrease with age. Secondly, we examined how biomarkers predict other biomarkers in participants of different age groups. We found 17 (34%) biomarkers whose predictability decreases with age and 5 (10%) biomarkers whose predictability increases with age. A limitation of this study is that it cannot distinguish between biological changes related to aging and generational effects. Our results can be interactively explored here: http://apps.chiragjpgroup.org/Aging_Biomarkers_Co-Dependencies/.

Development of a prognostic composite cytokine signature based on the correlation with nivolumab clearance: translational PK/PD analysis in patients with renal cell carcinoma

BACKGROUND

Although several therapeutic options for patients with renal cell carcinoma (RCC) have been approved over recent years, including immune checkpoint inhibitors, considerable need remains for molecular biomarkers to assess disease prognosis. The higher pharmacokinetic (PK) clearance of checkpoint inhibitors, such as the anti-programmed death-1 (PD-1) therapies nivolumab and pembrolizumab, has been shown to be associated with poor overall survival (OS) across several tumor types. However, determination of PK clearance requires the collection and analysis of post-treatment serum samples, limiting its utility as a prognostic biomarker. This report outlines a translational PK-pharmacodynamic (PD) methodology used to derive a baseline composite cytokine signature correlated with nivolumab clearance using data from three clinical trials in which nivolumab or everolimus was administered.

METHODS

Peripheral serum cytokine (PD) and nivolumab clearance (PK) data from patients with RCC were analyzed using a PK-PD machine-learning model. Nivolumab studies CheckMate 009 (NCT01358721) and CheckMate 025 (NCT01668784) (n = 480) were used for PK-PD analysis model development and cytokine feature selection (training dataset). Validation of the model and assessment of the prognostic value of the cytokine signature was performed using data from CheckMate 010 (NCT01354431) and the everolimus comparator arm of CheckMate 025 (test dataset; n = 453).

RESULTS

The PK-PD analysis found a robust association between the eight top-ranking model-selected baseline inflammatory cytokines and nivolumab clearance (area under the receiver operating characteristic curve = 0.7). The predicted clearance (high vs low) based on the cytokine signature was significantly associated with long-term OS (p < 0.01) across all three studies (training and test datasets). Furthermore, cytokines selected from the model development trials also correlated with OS of the everolimus comparator arm (p < 0.01), suggesting the prognostic nature of the composite cytokine signature for RCC.

CONCLUSIONS

Here, we report a PK-PD translational approach to identify a molecular prognostic biomarker signature based on the correlation with nivolumab clearance in patients with RCC. This composite biomarker signature may provide improved prognostic accuracy of long-term clinical outcome compared with individual cytokine features and could be used to ensure the balance of patient randomization in RCC clinical trials.

Diagnostic accuracy and predictive value in differentiating the severity of dengue infection

OBJECTIVE

To review the diagnostic test accuracy and predictive value of statistical models in differentiating the severity of dengue infection.

METHODS

Electronic searches were conducted in the Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, MEDLINE (complete), PubMed and Scopus. Eligible studies to be included in this review were cohort studies with participants confirmed by laboratory test for dengue infection and comparison among the different severity of dengue infection by using statistical models. The methodological quality of the paper was assessed by independent reviewers using QUADAS-2.

RESULTS

Twenty-six studies published from 1994 to 2017 were included. Most diagnostic models produced an accuracy of 75% to 80% except one with 86%. Two models predicting severe dengue according to the WHO 2009 classification have 86% accuracy. Both of these logistic regression models were applied during the first three days of illness, and their sensitivity and specificity were 91-100% and 79.3-86%, respectively. Another model which evaluated the 30-day mortality of dengue infection had an accuracy of 98.5%.

CONCLUSION

Although there are several potential predictive or diagnostic models for dengue infection, their limitations could affect their validity. It is recommended that these models be revalidated in other clinical settings and their methods be improved and standardised in future.

Variable selection in omics data: A practical evaluation of small sample sizes

In omics experiments, variable selection involves a large number of metabolites/ genes and a small number of samples (the n < p problem). The ultimate goal is often the identification of one, or a few features that are different among conditions- a biomarker. Complicating biomarker identification, the p variables often contain a correlation structure due to the biology of the experiment making identifying causal compounds from correlated compounds difficult. Additionally, there may be elements in the experimental design (blocks, batches) that introduce structure in the data. While this problem has been discussed in the literature and various strategies proposed, the over fitting problems concomitant with such approaches are rarely acknowledged. Instead of viewing a single omics experiment as a definitive test for a biomarker, an unrealistic analytical goal, we propose to view such studies as screening studies where the goal of the study is to reduce the number of features present in the second round of testing, and to limit the Type II error. Using this perspective, the performance of LASSO, ridge regression and Elastic Net was compared with the performance of an ANOVA via a simulation study and two real data comparisons. Interestingly, a dramatic increase in the number of features had no effect on Type I error for the ANOVA approach. ANOVA, even without multiple test correction, has a low false positive rates in the scenarios tested. The Elastic Net has an inflated Type I error (from 10 to 50%) for small numbers of features which increases with sample size. The Type II error rate for the ANOVA is comparable or lower than that for the Elastic Net leading us to conclude that an ANOVA is an effective analytical tool for the initial screening of features in omics experiments.

Fc receptors in antibody-dependent enhancement of viral infections

Sensitization of the humoral immune response to invading viruses and production of antiviral antibodies forms part of the host antiviral repertoire. Paradoxically, for a number of viral pathogens, under certain conditions, antibodies provide an attractive means of enhanced virus entry and replication in a number of cell types. Known as antibody‐dependent enhancement (ADE) of infection, the phenomenon occurs when virus‐antibody immunocomplexes interact with cells bearing complement or Fc receptors, promoting internalization of the virus and increasing infection. Frequently associated with exacerbation of viral disease, ADE of infection presents a major obstacle to the prevention of viral disease by vaccination and is thought to be partly responsible for the adverse effects of novel antiviral therapeutics such as intravenous immunoglobulins. There is a growing body of work examining the intracellular signaling pathways and epitopes responsible for mediating ADE, with a view to aiding rational design of antiviral strategies. With in vitro studies also confirming ADE as a feature of infection for a growing number of viruses, challenges remain in understanding the multilayered molecular mechanisms of ADE and its effect on viral pathogenesis.

Targeted proteomics identifies liquid-biopsy signatures for extracapsular prostate cancer

Biomarkers are rapidly gaining importance in personalized medicine. Although numerous molecular signatures have been developed over the past decade, there is a lack of overlap and many biomarkers fail to validate in independent patient cohorts and hence are not useful for clinical application. For these reasons, identification of novel and robust biomarkers remains a formidable challenge. We combine targeted proteomics with computational biology to discover robust proteomic signatures for prostate cancer. Quantitative proteomics conducted in expressed prostatic secretions from men with extraprostatic and organ-confined prostate cancers identified 133 differentially expressed proteins. Using synthetic peptides, we evaluate them by targeted proteomics in a 74-patient cohort of expressed prostatic secretions in urine. We quantify a panel of 34 candidates in an independent 207-patient cohort. We apply machine-learning approaches to develop clinical predictive models for prostate cancer diagnosis and prognosis. Our results demonstrate that computationally guided proteomics can discover highly accurate non-invasive biomarkers.

Proteomic technologies are capable of identifying thousands of proteins in biological samples, but biomarker applications are lagging. Here the authors use Multiple Reaction Monitoring Mass Spectrometry to delineate peptide signatures that accurately distinguish between defined prostate cancer patient risk groups.

Dengue NS1 antigen contributes to disease severity by inducing interleukin (IL)-10 by monocytes

Both dengue NS1 antigen and serum interleukin (IL)-10 levels have been shown to associate with severe clinical disease in acute dengue infection, and IL-10 has also been shown to suppress dengue-specific T cell responses. Therefore, we proceeded to investigate the mechanisms by which dengue NS1 contributes to disease pathogenesis and if it is associated with altered IL-10 production. Serum IL-10 and dengue NS1 antigen levels were assessed serially in 36 adult Sri Lankan individuals with acute dengue infection. We found that the serum IL-10 levels correlated positively with dengue NS1 antigen levels (Spearman's r = 0·47, P < 0·0001), and NS1 also correlated with annexin V expression by T cells in acute dengue (Spearman's r = 0·63, P = 0·001). However, NS1 levels did not associate with the functionality of T cell responses or with expression of co-stimulatory molecules. Therefore, we further assessed the effect of dengue NS1 on monocytes and T cells by co-culturing primary monocytes and peripheral blood mononuclear cells (PBMC), with varying concentrations of NS1 for up to 96 h. Monocytes co-cultured with NS1 produced high levels of IL-10, with the highest levels seen at 24 h, and then declined gradually. Therefore, our data show that dengue NS1 appears to contribute to pathogenesis of dengue infection by inducing IL-10 production by monocytes.

Improved Detection of Invasive Pulmonary Aspergillosis Arising during Leukemia Treatment Using a Panel of Host Response Proteins and Fungal Antigens

Invasive pulmonary aspergillosis (IPA) is an opportunistic fungal infection in patients undergoing chemotherapy for hematological malignancy, hematopoietic stem cell transplant, or other forms of immunosuppression. In this group, Aspergillus infections account for the majority of deaths due to mold pathogens. Although early detection is associated with improved outcomes, current diagnostic regimens lack sensitivity and specificity. Patients undergoing chemotherapy, stem cell transplantation and lung transplantation were enrolled in a multi-site prospective observational trial. Proven and probable IPA cases and matched controls were subjected to discovery proteomics analyses using a biofluid analysis platform, fractionating plasma into reproducible protein and peptide pools. From 556 spots identified by 2D gel electrophoresis, 66 differentially expressed post-translationally modified plasma proteins were identified in the leukemic subgroup only. This protein group was rich in complement components, acute-phase reactants and coagulation factors. Low molecular weight peptides corresponding to abundant plasma proteins were identified. A candidate marker panel of host response (9 plasma proteins, 4 peptides), fungal polysaccharides (galactomannan), and cell wall components (β-D glucan) were selected by statistical filtering for patients with leukemia as a primary underlying diagnosis. Quantitative measurements were developed to qualify the differential expression of the candidate host response proteins using selective reaction monitoring mass spectrometry assays, and then applied to a separate cohort of 57 patients with leukemia. In this verification cohort, a machine learning ensemble-based algorithm, generalized pathseeker (GPS) produced a greater case classification accuracy than galactomannan (GM) or host proteins alone. In conclusion, Integration of host response proteins with GM improves the diagnostic detection of probable IPA in patients undergoing treatment for hematologic malignancy. Upon further validation, early detection of probable IPA in leukemia treatment will provide opportunities for earlier interventions and interventional clinical trials.

Host biomarkers are associated with progression to dengue haemorrhagic fever: a nested case-control study

OBJECTIVES

Dengue represents the most important arboviral infection worldwide. Onset of circulatory collapse can be unpredictable. Biomarkers that can identify individuals at risk of plasma leakage may facilitate better triage and clinical management.

DESIGN

Using a nested case-control design, we randomly selected subjects from a prospective cohort study of dengue in Colombia (n=1582). Using serum collected within 96 hours of fever onset, we tested 19 biomarkers by ELISA in cases (developed dengue hemorrhagic fever or dengue shock syndrome (DHF/DSS); n=46), and controls (uncomplicated dengue fever (DF); n=65) and healthy controls (HC); n=15.

RESULTS

Ang-1 levels were lower and angptl3, sKDR, sEng, sICAM-1, CRP, CXCL10/IP-10, IL-18 binding protein, CHI3L1, C5a and Factor D levels were increased in dengue compared to HC. sICAM-1, sEng and CXCL10/IP-10 were further elevated in subjects who subsequently developed DHF/DSS (p=0.008, p=0.028 and p=0.025, respectively). In a logistic regression model, age (odds ratio (OR) (95% CI): 0.95 (0.92-0.98), p=0.001), hyperesthesia/hyperalgesia (OR; 3.8 (1.4-10.4), p=0.008) and elevated sICAM-1 (>298ng/mL: OR; 6.3 (1.5-25.7), p=0.011) at presentation were independently associated with progression to DHF/DSS.

CONCLUSIONS

These results suggest that inflammation and endothelial activation are important pathways in the pathogenesis of dengue and sICAM-1 levels may identify individuals at risk of plasma leakage.

Molecular classification of outcomes from dengue virus -3 infections

OBJECTIVES

Dengue virus (DENV) infection is a significant risk to over a third of the human population that causes a wide spectrum of illness, ranging from sub-clinical disease to intermediate syndrome of vascular complications called dengue fever complicated (DFC) and severe, dengue hemorrhagic fever (DHF). Methods for discriminating outcomes will impact clinical trials and understanding disease pathophysiology.

STUDY DESIGN

We integrated a proteomics discovery pipeline with a heuristics approach to develop a molecular classifier to identify an intermediate phenotype of DENV-3 infectious outcome.

RESULTS

121 differentially expressed proteins were identified in plasma from DHF vs dengue fever (DF), and informative candidates were selected using nonparametric statistics. These were combined with markers that measure complement activation, acute phase response, cellular leak, granulocyte differentiation and viral load. From this, we applied quantitative proteomics to select a 15 member panel of proteins that accurately predicted DF, DHF, and DFC using a random forest classifier. The classifier primarily relied on acute phase (A2M), complement (CFD), platelet counts and cellular leak (TPM4) to produce an 86% accuracy of prediction with an area under the receiver operating curve of >0.9 for DHF and DFC vs DF.

CONCLUSIONS

Integrating discovery and heuristic approaches to sample distinct pathophysiological processes is a powerful approach in infectious disease. Early detection of intermediate outcomes of DENV-3 will speed clinical trials evaluating vaccines or drug interventions.

Dengue NS1 antigen as a marker of severe clinical disease

BACKGROUND

Early detection of complications significantly reduces dengue associated mortality and morbidity. We set out to determine if the NS1 rapid antigen detection test could be used as a point of care test to predict severe disease.

METHODS

186 adult patients with confirmed dengue were enrolled during day 3-8 of illness. Clinical and laboratory parameters were recorded during the course of the illness and NS1 antigen levels were determined using both the Panbio dengue early ELISA (Panbio, Australia) and a NS1 rapid antigen detection kit (SD Bioline, South Korea).

RESULTS

59.1% of patients presented to hospital on day 5-6 of illness when NS1 antigen positivity was significantly (p = 0.008) associated with severe dengue (odds ratio 3.0, 95% CI 1.39 to 6.47) and the NS1 antigen levels were significantly higher (p = 0.03) in those who went on to develop shock. Serum NS1 antigen levels significantly (p < 0.0001) and inversely correlated with the total white cell counts and lymphocyte counts. The bedside NS1 test showed comparable sensitivity (97.4%) and specificity (93.7%) to the laboratory NS1 test in our setting and cohort.

CONCLUSION

NS1 antigen positivity is associated with a higher risk of developing severe dengue especially when positive beyond day 5 of illness in our cohort, and while further validation studies are required, the test can therefore potentially be used as a bedside point of care test as a warning sign of severe dengue.