A 17-gene assay to predict prostate cancer aggressiveness in the context of Gleason grade heterogeneity, tumor multifocality, and biopsy undersampling

BACKGROUND

Prostate tumor heterogeneity and biopsy undersampling pose challenges to accurate, individualized risk assessment for men with localized disease.

OBJECTIVE

To identify and validate a biopsy-based gene expression signature that predicts clinical recurrence, prostate cancer (PCa) death, and adverse pathology.

DESIGN, SETTING, AND PARTICIPANTS

Gene expression was quantified by reverse transcription-polymerase chain reaction for three studies-a discovery prostatectomy study (n=441), a biopsy study (n=167), and a prospectively designed, independent clinical validation study (n=395)-testing retrospectively collected needle biopsies from contemporary (1997-2011) patients with low to intermediate clinical risk who were candidates for active surveillance (AS).

OUTCOME MEASURES AND STATISTICAL ANALYSIS

The main outcome measures defining aggressive PCa were clinical recurrence, PCa death, and adverse pathology at prostatectomy. Cox proportional hazards regression models were used to evaluate the association between gene expression and time to event end points. Results from the prostatectomy and biopsy studies were used to develop and lock a multigene-expression-based signature, called the Genomic Prostate Score (GPS); in the validation study, logistic regression was used to test the association between the GPS and pathologic stage and grade at prostatectomy. Decision-curve analysis and risk profiles were used together with clinical and pathologic characteristics to evaluate clinical utility.

RESULTS AND LIMITATIONS

Of the 732 candidate genes analyzed, 288 (39%) were found to predict clinical recurrence despite heterogeneity and multifocality, and 198 (27%) were predictive of aggressive disease after adjustment for prostate-specific antigen, Gleason score, and clinical stage. Further analysis identified 17 genes representing multiple biological pathways that were combined into the GPS algorithm. In the validation study, GPS predicted high-grade (odds ratio [OR] per 20 GPS units: 2.3; 95% confidence interval [CI], 1.5-3.7; p<0.001) and high-stage (OR per 20 GPS units: 1.9; 95% CI, 1.3-3.0; p=0.003) at surgical pathology. GPS predicted high-grade and/or high-stage disease after controlling for established clinical factors (p<0.005) such as an OR of 2.1 (95% CI, 1.4-3.2) when adjusting for Cancer of the Prostate Risk Assessment score. A limitation of the validation study was the inclusion of men with low-volume intermediate-risk PCa (Gleason score 3+4), for whom some providers would not consider AS.

CONCLUSIONS

Genes representing multiple biological pathways discriminate PCa aggressiveness in biopsy tissue despite tumor heterogeneity, multifocality, and limited sampling at time of biopsy. The biopsy-based 17-gene GPS improves prediction of the presence or absence of adverse pathology and may help men with PCa make more informed decisions between AS and immediate treatment.

PATIENT SUMMARY

Prostate cancer (PCa) is often present in multiple locations within the prostate and has variable characteristics. We identified genes with expression associated with aggressive PCa to develop a biopsy-based, multigene signature, the Genomic Prostate Score (GPS). GPS was validated for its ability to predict men who have high-grade or high-stage PCa at diagnosis and may help men diagnosed with PCa decide between active surveillance and immediate definitive treatment.

Design Characteristics of Studies Reporting the Performance of Artificial Intelligence Algorithms for Diagnostic Analysis of Medical Images: Results from Recently Published Papers

Objective

To evaluate the design characteristics of studies that evaluated the performance of artificial intelligence (AI) algorithms for the diagnostic analysis of medical images.

Materials and Methods

PubMed MEDLINE and Embase databases were searched to identify original research articles published between January 1, 2018 and August 17, 2018 that investigated the performance of AI algorithms that analyze medical images to provide diagnostic decisions. Eligible articles were evaluated to determine 1) whether the study used external validation rather than internal validation, and in case of external validation, whether the data for validation were collected, 2) with diagnostic cohort design instead of diagnostic case-control design, 3) from multiple institutions, and 4) in a prospective manner. These are fundamental methodologic features recommended for clinical validation of AI performance in real-world practice. The studies that fulfilled the above criteria were identified. We classified the publishing journals into medical vs. non-medical journal groups. Then, the results were compared between medical and non-medical journals.

Results

Of 516 eligible published studies, only 6% (31 studies) performed external validation. None of the 31 studies adopted all three design features: diagnostic cohort design, the inclusion of multiple institutions, and prospective data collection for external validation. No significant difference was found between medical and non-medical journals.

Conclusion

Nearly all of the studies published in the study period that evaluated the performance of AI algorithms for diagnostic analysis of medical images were designed as proof-of-concept technical feasibility studies and did not have the design features that are recommended for robust validation of the real-world clinical performance of AI algorithms.

A Biopsy-based 17-gene Genomic Prostate Score Predicts Recurrence After Radical Prostatectomy and Adverse Surgical Pathology in a Racially Diverse Population of Men with Clinically Low- and Intermediate-risk Prostate Cancer

BACKGROUND

Biomarkers that are validated in independent cohorts are needed to improve risk assessment for prostate cancer (PCa).

OBJECTIVE

A racially diverse cohort of men (20% African American [AA]) was used to evaluate the association of the clinically validated 17-gene Genomic Prostate Score (GPS) with recurrence after radical prostatectomy and adverse pathology (AP) at surgery.

DESIGN, SETTING, AND PARTICIPANTS

Biopsies from 431 men treated for National Comprehensive Cancer Network (NCCN) very low-, low-, or intermediate-risk PCa between 1990 and 2011 at two US military medical centers were tested to validate the association between GPS and biochemical recurrence (BCR) and to confirm the association with AP. Metastatic recurrence (MR) was also evaluated.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Cox proportional hazards models were used for BCR and MR, and logistic regression was used for AP. Central pathology review was performed by one uropathologist. AP was defined as primary Gleason pattern 4 or any pattern 5 and/or pT3 disease.

RESULTS AND LIMITATIONS

GPS results (scale: 0-100) were obtained in 402 cases (93%); 62 men (15%) experienced BCR, 5 developed metastases, and 163 had AP. Median follow-up was 5.2 yr. GPS predicted time to BCR in univariable analysis (hazard ratio per 20 GPS units [HR/20 units]: 2.9; p<0.001) and after adjusting for NCCN risk group (HR/20 units: 2.7; p<0.001). GPS also predicted time to metastases (HR/20 units: 3.8; p=0.032), although the event rate was low (n=5). GPS was strongly associated with AP (odds ratio per 20 GPS units: 3.3; p<0.001), adjusted for NCCN risk group. In AA and Caucasian men, the median GPS was 30.3 for both, the distributions of GPS results were similar, and GPS was similarly predictive of outcome.

CONCLUSIONS

The association of GPS with near- and long-term clinical end points establishes the assay as a strong independent measure of PCa aggressiveness. Tumor aggressiveness, as measured by GPS, and outcomes were similar in AA and Caucasian men in this equal-access health care system.

PATIENT SUMMARY

Predicting outcomes in men with newly diagnosed prostate cancer is challenging. This study demonstrates that a new molecular test, the Genomic Prostate Score, which can be performed on a patient's original prostate needle biopsy, can predict the aggressiveness of the cancer and help men make decisions regarding the need for immediate treatment of their cancer.

A genomic classifier improves prediction of metastatic disease within 5 years after surgery in node-negative high-risk prostate cancer patients managed by radical prostatectomy without adjuvant therapy

BACKGROUND

Surgery is a standard first-line therapy for men with intermediate- or high-risk prostate cancer. Clinical factors such as tumor grade, stage, and prostate-specific antigen (PSA) are currently used to identify those who are at risk of recurrence and who may benefit from adjuvant therapy, but novel biomarkers that improve risk stratification and that distinguish local from systemic recurrence are needed.

OBJECTIVE

To determine whether adding the Decipher genomic classifier, a validated metastasis risk-prediction model, to standard risk-stratification tools (CAPRA-S and Stephenson nomogram) improves accuracy in predicting metastatic disease within 5 yr after surgery (rapid metastasis [RM]) in an independent cohort of men with adverse pathologic features after radical prostatectomy (RP).

DESIGN, SETTING, AND PARTICIPANTS

The study population consisted of 169 patients selected from 2641 men who underwent RP at the Cleveland Clinic between 1987 and 2008 who met the following criteria: (1) preoperative PSA>20 ng/ml, stage pT3 or margin positive, or Gleason score≥8; (2) pathologic node negative; (3) undetectable post-RP PSA; (4) no neoadjuvant or adjuvant therapy; and (5) minimum of 5-yr follow-up for controls. The final study cohort consisted of 15 RM patients and 154 patients as non-RM controls.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

The performance of Decipher was evaluated individually and in combination with clinical risk factors using concordance index (c-index), decision curve analysis, and logistic regression for prediction of RM.

RESULTS AND LIMITATIONS

RM patients developed metastasis at a median of 2.3 yr (interquartile range: 1.7-3.3). In multivariable analysis, Decipher was a significant predictor of RM (odds ratio: 1.48; p=0.018) after adjusting for clinical risk factors. Decipher had the highest c-index, 0.77, compared with the Stephenson model (c-index: 0.75) and CAPRA-S (c-index: 0.72) as well as with a panel of previously reported prostate cancer biomarkers unrelated to Decipher. Integration of Decipher into the Stephenson nomogram increased the c-index from 0.75 (95% confidence interval [CI], 0.65-0.85) to 0.79 (95% CI, 0.68-0.89).

CONCLUSIONS

Decipher was independently validated as a genomic metastasis signature for predicting metastatic disease within 5 yr after surgery in a cohort of high-risk men treated with RP and managed conservatively without any adjuvant therapy. Integration of Decipher into clinical nomograms increased prediction of RM. Decipher may allow identification of men most at risk for metastatic progression who should be considered for multimodal therapy or inclusion in clinical trials.

PATIENT SUMMARY

Use of Decipher in addition to standard clinical information more accurately identified men who developed metastatic disease within 5 yr after surgery. The results suggest that Decipher allows improved identification of the men who should consider secondary therapy from among the majority that may be managed conservatively after surgery.

Dissecting psychiatric spectrum disorders by generative embedding

This proof-of-concept study examines the feasibility of defining subgroups in psychiatric spectrum disorders by generative embedding, using dynamical system models which infer neuronal circuit mechanisms from neuroimaging data. To this end, we re-analysed an fMRI dataset of 41 patients diagnosed with schizophrenia and 42 healthy controls performing a numerical n-back working-memory task. In our generative-embedding approach, we used parameter estimates from a dynamic causal model (DCM) of a visual-parietal-prefrontal network to define a model-based feature space for the subsequent application of supervised and unsupervised learning techniques. First, using a linear support vector machine for classification, we were able to predict individual diagnostic labels significantly more accurately (78%) from DCM-based effective connectivity estimates than from functional connectivity between (62%) or local activity within the same regions (55%). Second, an unsupervised approach based on variational Bayesian Gaussian mixture modelling provided evidence for two clusters which mapped onto patients and controls with nearly the same accuracy (71%) as the supervised approach. Finally, when restricting the analysis only to the patients, Gaussian mixture modelling suggested the existence of three patient subgroups, each of which was characterised by a different architecture of the visual-parietal-prefrontal working-memory network. Critically, even though this analysis did not have access to information about the patients' clinical symptoms, the three neurophysiologically defined subgroups mapped onto three clinically distinct subgroups, distinguished by significant differences in negative symptom severity, as assessed on the Positive and Negative Syndrome Scale (PANSS). In summary, this study provides a concrete example of how psychiatric spectrum diseases may be split into subgroups that are defined in terms of neurophysiological mechanisms specified by a generative model of network dynamics such as DCM. The results corroborate our previous findings in stroke patients that generative embedding, compared to analyses of more conventional measures such as functional connectivity or regional activity, can significantly enhance both the interpretability and performance of computational approaches to clinical classification.

Clinical and analytical comparison of six Simoa assays for plasma P-tau isoforms P-tau181, P-tau217, and P-tau231

INTRODUCTION

Studies using different assays and technologies showed highly promising diagnostic value of plasma phosphorylated (P-)tau levels for Alzheimer's disease (AD). We aimed to compare six P-tau Simoa assays, including three P-tau181 (Eli Lilly, ADx, Quanterix), one P-tau217 (Eli Lilly), and two P-tau231 (ADx, Gothenburg).

METHODS

We studied the analytical (sensitivity, precision, parallelism, dilution linearity, and recovery) and clinical (40 AD dementia patients, age 66±8years, 50%F; 40 age- and sex-matched controls) performance of the assays.

RESULTS

All assays showed robust analytical performance, and particularly P-tau217 Eli Lilly; P-tau231 Gothenburg and all P-tau181 assays showed robust clinical performance to differentiate AD from controls, with AUCs 0.936-0.995 (P-tau231 ADx: AUC = 0.719). Results obtained with all P-tau181 assays, P-tau217 Eli Lilly assay, and P-tau231 Gothenburg assay strongly correlated (Spearman's rho > 0.86), while correlations with P-tau231 ADx results were moderate (rho < 0.65).

DISCUSSION

P-tau isoforms can be measured robustly by several novel high-sensitive Simoa assays.

Validation of the Oncomine™ focus panel for next-generation sequencing of clinical tumour samples

The clinical utility of next-generation sequencing (NGS) for a diverse range of targets is expanding, increasing the need for multiplexed analysis of both DNA and RNA. However, translation into daily use requires a rigorous and comprehensive validation strategy. The aim of this clinical validation was to assess the performance of the Ion Torrent Personal Genome Machine (IonPGM^™) and validate the Oncomine^™ Focus DNA and RNA Fusion panels for clinical application in solid tumour testing of formalin-fixed, paraffin-embedded (FFPE) tissue. Using a mixture of routine FFPE and reference material across a variety of tissue and specimen types, we sequenced 86 and 31 samples on the Oncomine^™ Focus DNA and RNA Fusion assays, respectively. This validation considered a number of parameters including the clinical robustness of the bioinformatics pipeline for variant detection and interpretation. The Oncomine^™ Focus DNA assay had a sample and variant-based sensitivity of 99.1 and 97.1%, respectively, and an assay specificity of 100%. The Oncomine^™ Focus Fusion panel had a good sensitivity and specificity based upon the samples assessed, however requires further validation to confirm findings due to limited sample numbers. We observed a good sequencing performance based upon amplicon, gene (hotspot variants within gene) and sample specific analysis with 92% of clinical samples obtaining an average amplicon coverage above 500X. Detection of some indels was challenging for the routine IonReporter^™ workflow; however, the addition of NextGENe® software improved indel identification demonstrating the importance of both bench and bioinformatic validation. With an increasing number of clinically actionable targets requiring a variety of methodologies, NGS provides a cost-effective and time-saving methodology to assess multiple targets across different modalities. We suggest the use of multiple analysis software to ensure identification of clinically applicable variants.

Multiplexed cytokine detection using electrochemical point-of-care sensing device towards rapid sepsis endotyping

The implementation of endotype-driven effective intervention strategies is now considered as an essential component for sepsis management. Rapid screening and frequent monitoring of immune responses are critical for evidence-based informed decisions in the early hours of patient arrival. Current technologies focus on pathogen identification that lack rapid testing of the patient immune response, impeding clinicians from providing appropriate sepsis treatment. Herein, we demonstrate a first-of-its-kind novel point-of-care device that uses a unique approach by directly monitoring a panel of five cytokine biomarkers (IL-6, IL-8, IL-10, TRAIL & IP-10), that is attributed as a sign of the body's host immune response to sepsis. The developed point-of-care device encompasses a disposable sensor cartridge attached to an electrochemical reader. High sensitivity is achieved owing to the unique sensor design with an array of nanofilm semiconducting/metal electrode interface, functionalized with specific capture probes to measure target biomarkers simultaneously using non-faradaic electrochemical impedance spectroscopy. The sensor has a detection limit of ~1 pg/mL and provides results in less than five minutes from a single drop of undiluted plasma sample. Furthermore, the sensor demonstrates an excellent correlation (Pearson's r > 0.90) with the reference method for a total n = 40 clinical samples, and the sensor's performance is ~30 times faster compared to the standard reference technique. We have demonstrated the sensor's effectiveness to enhance diagnosis with a mechanistic biomarker-guided approach that can help disease endotypying for effective clinical management of sepsis at the patient bedside.

Clinical and analytical evaluation of the RealTime High Risk HPV assay in Colli-Pee collected first-void urine using the VALHUDES protocol

OBJECTIVE

Urine self-sampling has gained increasing interest for cervical cancer screening. In contrast to analytical performance, little information is available regarding the clinical accuracy for high-risk Human Papillomavirus (hrHPV) testing on urine.

METHODS

VALHUDES is a diagnostic test accuracy study comparing clinical accuracy to detect high-grade cervical precancer (CIN2+) of HPV testing on self-collected compared to clinician-collected samples (NCT03064087). Disease outcome was assessed by colposcopy and histology. The Abbott RealTime High Risk HPV assay performance was evaluated on Colli-Pee collected first-void urine with cervical outcomes as comparator.

RESULTS

As no assay cut-off for urine has been clinically validated, we used the predefined cut-off for cervical samples (CN ≤ 32). Using this cut-off, hrHPV testing was similarly sensitive (relative sensitivity 0.95; 95% CI: 0.88-1.01) and specific (relative specificity 1.03; 95% CI: 0.95-1.13) for detection of CIN2+ compared to testing cervical samples. In the subgroup of women of 30 years and older, similar relative sensitivity (0.97; 95% CI: 0.89-1.05) and specificity (1.02; 95% CI: 0.93-1.12) was found. Additionally, an exploratory cut-off (CN ≤ 33.86) was defined which further improved sensitivity and analytical test performance.

CONCLUSION

HrHPV-DNA based PCR testing on home-collected first-void urine has similar accuracy for detecting CIN2+ compared to cervical samples taken by a clinician.

Clinical Evaluation of a Novel and Mobile Autism Risk Assessment

The Mobile Autism Risk Assessment (MARA) is a new, electronically administered, 7-question autism spectrum disorder (ASD) screen to triage those at highest risk for ASD. Children 16 months–17 years (N = 222) were screened during their first visit in a developmental-behavioral pediatric clinic. MARA scores were compared to diagnosis from the clinical encounter. Participant median age was 5.8 years, 76.1 % were male, and most participants had an intelligence/developmental quotient score >85; 69 of the participants (31 %) received a clinical diagnosis of ASD. The sensitivity of the MARA in detecting ASD was 89.9 % [95 % CI = 82.7–97]; the specificity was 79.7 % [95 % CI = 73.4–86.1]. In a high-risk clinical setting, the MARA shows promise as a screen to distinguish ASD from other developmental/behavioral disorders.

Considerations for development of an evidence dossier to support the use of mobile sensor technology for clinical outcome assessments in clinical trials

BACKGROUND

Mobile sensors offer enormous potential for the collection of informative clinical endpoints in clinical trials to support regulatory decision making and product labelling. There are currently no specific guidelines on the information needed to enable regulators to review and accept proposed endpoints derived from mobile sensors for use in drug development trials.

OBJECTIVE

The purpose of this working group report is to recommend the structure and content of an evidence dossier intended to support whether a clinical endpoint derived from mobile sensor data is fit-for-purpose for use in regulatory submissions for drug approvals.

EVIDENCE DOSSIER

The structure and content of a dossier to provide evidence supporting the use of a sensor-derived clinical endpoint is described. Sections include clinical endpoint definition and positioning, the concept of interest, the context of use, clinical validation and interpretation, study implementation, and analytical validity with sensor performance verification in support of the selected sensor.

CONCLUSIONS

In the absence of definitive regulatory guidance, this report provides a considered approach to compiling a comprehensive body of evidence to justify acceptance of mobile sensors for support of new drug applications.

Four-phase rhinomanometry: a multicentric retrospective analysis of 36,563 clinical measurements

Rhinomanometry can still be considered as the standard technique for the objective assessment of the ventilatory function of the nose. Reliable technical requirements are given by fast digital sensors and modern information technology. However, the xyimaging of the pressure-flow relation typically shows loops as a sign of hysteresis, with the need for resolution of the breath in four phases. The three pillars of 4-phase rhinomanometry (4PR) are the replacement of estimations by measurements, the introduction of parameters related to the subjective sensing of obstruction, and the graphical information regarding the disturbed function of the nasal valve. In a meta-analysis of 36,563 clinical measurements, we analyze the errors of the "classic" parameters (flow in 150 Pa) and reject the further use of these parameters as obsolete, because they correspond to an inaccurate estimation rather than proper measurement. In a pre-study of 1580 measurements, the logarithmic effective resistance (Reff) was found to have the highest correlation with values obtained from a visual analog scale. Next, we classify the inspiratory effective resistance in 20,069 measurements without treatment and 16,494 measurements after decongestion with xylometazoline 0.1 % spray in 20 % percentiles. The gradation of obstruction delivers not only "normal" values but also indications for the severity of the obstruction in adult Caucasian noses. Adoption of the distribution for the growing nose and analysis of the total nasal resistance is addressed, and typical findings of nasal valve phenomena are outlined.

Metabolic biomarkers for chronic kidney disease

Chronic kidney disease (CKD) is an increasingly recognized burden for patients and health care systems with high (and growing) global incidence and prevalence, significant mortality, and disproportionately high treatment costs. Yet, the available diagnostic tools are either impractical in clinical routine or have serious shortcomings impeding a well-informed disease management although optimized treatment strategies with proven benefits for the patients have become available. Advances in bioanalytical technologies have facilitated studies that identified genomic, proteomic, and metabolic biomarker candidates, and confirmed some of them in independent cohorts. This review summarizes the CKD-related markers discovered so far, and focuses on compounds and pathways, for which there is quantitative data, substantiating evidence from translational research, and a mechanistic understanding of the processes involved. Also, multiparametric marker panels have been suggested that showed promising diagnostic and prognostic performance in initial analyses although the data basis from prospective trials is very limited. Large-scale studies, however, are underway and will provide the information for validating a set of parameters and discarding others. Finally, the path from clinical research to a routine application is discussed, focusing on potential obstacles such as the use of mass spectrometry, and the feasibility of obtaining regulatory approval for targeted metabolomics assays.

Objective assessment of bradykinesia in Parkinson's disease using evolutionary algorithms: clinical validation

BACKGROUND

There is an urgent need for developing objective, effective and convenient measurements to help clinicians accurately identify bradykinesia. The purpose of this study is to evaluate the accuracy of an objective approach assessing bradykinesia in finger tapping (FT) that uses evolutionary algorithms (EAs) and explore whether it can be used to identify early stage Parkinson's disease (PD).

METHODS

One hundred and seven PD, 41 essential tremor (ET) patients and 49 normal controls (NC) were recruited. Participants performed a standard FT task with two electromagnetic tracking sensors attached to the thumb and index finger. Readings from the sensors were transmitted to a tablet computer and subsequently analyzed by using EAs. The output from the device (referred to as "PD-Monitor") scaled from - 1 to + 1 (where higher scores indicate greater severity of bradykinesia). Meanwhile, the bradykinesia was rated clinically using the Movement Disorder Society-Sponsored Revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS) FT item.

RESULTS

With an increasing MDS-UPDRS FT score, the PD-Monitor score from the same hand side increased correspondingly. PD-Monitor score correlated well with MDS-UPDRS FT score (right side: r = 0.819, P = 0.000; left side: r = 0.783, P = 0.000). Moreover, PD-Monitor scores in 97 PD patients with MDS-UPDRS FT bradykinesia and each PD subgroup (FT bradykinesia scored from 1 to 3) were all higher than that in NC. Receiver operating characteristic (ROC) curves revealed that PD-Monitor FT scores could detect different severity of bradykinesia with high accuracy (≥89.7%) in the right dominant hand. Furthermore, PD-Monitor scores could discriminate early stage PD from NC, with area under the ROC curve greater than or equal to 0.899. Additionally, ET without bradykinesia could be differentiated from PD by PD-Monitor scores. A positive correlation of PD-Monitor scores with modified Hoehn and Yahr stage was found in the left hand sides.

CONCLUSIONS

Our study demonstrated that a simple to use device employing classifiers derived from EAs could not only be used to accurately measure different severity of bradykinesia in PD, but also had the potential to differentiate early stage PD from normality.

Analytical and clinical performance of extended HPV genotyping with BD Onclarity HPV Assay in home-collected first-void urine: A diagnostic test accuracy study

BACKGROUND

Urine collection is a non-invasive self-sampling method offering the prospect of reaching women un(der)-screened for cervical cancer. The VALHUDES research framework was designed to address the lack of clinical accuracy data for high-risk (hr)HPV testing using urine samples.

OBJECTIVES

Here, we report on the analytical and clinical accuracy of hrHPV testing on first-void urine, collected at home, using an extended HPV genotyping assay.

STUDY DESIGN

Paired first-void urine (Colli-Pee with UCM, Novosanis; index test) and clinician-collected cervical samples (Cervex-Brush, Rovers in PreservCyt Solution, Hologic; comparator test) were collected from 492 women aged 19 to 72 years attending colposcopy (reference test, with histology if indicated) (VALHUDES; NCT03064087). Extended HPV genotyping was performed on paired samples with the BD Onclarity HPV Assay. Cut-offs defined for cervical samples were also applied for first-void urine.

RESULTS

HrHPV testing in first-void urine was similarly sensitive for both CIN2+ (ratio 1.00; 95% CI: 0.93-1.07) and CIN3 (ratio 0.98; 95% CI: 0.88-1.08), and marginally less specific for <CIN2 (ratio 0.92; 95% CI: 0.84-0.996) compared to cervical samples. HPV test agreement between sample pairs expressed as Cohen's Kappa (κ) was moderate to excellent for overall hrHPV and individual genotypes (or groups) (κ=0.56-0.85).

CONCLUSIONS

BD Onclarity HPV Assay on first-void urine has similar clinical sensitivity and somewhat lower specificity to detect cervical precancer to testing on clinician-collected cervical samples.

A novel minimal mathematical model of the hypothalamus-pituitary-thyroid axis validated for individualized clinical applications

The hypothalamus-pituitary-thyroid (HPT) axis represents a complex, non-linear thyroid hormone system in vertebrates governed by numerous variables. The common modeling approach until now aims at a comprehensive inclusion of all known physiological influences. In contrast, we develop a parsimonious mathematical model that integrates the hypothalamus-pituitary (HP) complex as an endocrinologic unit based on a parameterized negative exponential function between free thyroxine (FT4) as stimulus and thyrotropin (thyroid stimulating hormone, TSH) as response. Model validation with clinical data obtained from geographically different hospitals revealed a goodness-of-fit largely ranging between 90% < R² < 99%, each HP characteristic curve being uniquely defined for each individual akin to a fingerprint. Specifically, the HP model represents the afferent feedback limb of the HPT axis while the efferent limb is mathematically depicted by TSH input to the thyroid gland which responds by secreting T4 as its chief output. The complete HPT axis thus forms a closed loop system with negative feedback resulting in an equilibrium state or homeostasis under defined conditions illustrated by the intersection of the HP and thyroid response characteristics. In this treatise, we demonstrate how this mathematical approach facilitates homeostatic set points computation for personalized dosing of thyroid medications of patients to individualized euthyroid states.

Clinical genome sequencing in an unbiased pediatric cohort

PURPOSE

We report for the first time, the use of clinical genome sequencing (GS) in an unbiased pediatric cohort. We describe the clinical validation, patient metrics, ordering patterns, results, reimbursement, and physician retrieval of results for the first consecutive 80 cases.

METHODS

Clinical GS was performed for both inpatients and outpatients undergoing etiologic evaluations. Results were reported in the electronic medical record. Evidence of report retrieval by clinicians and whether interpretation was concordant with laboratory report was obtained through retrospective chart review.

RESULTS

Twenty definitive diagnoses were made in 19 patients (24%; n = 80). Except for two partial gene deletions, all diagnostic variants would have been detectable by our exome methods. Surprisingly, there was no documentation of communication of results to the family in the medical record for 17.5% of patients, and in 7.5%, physician and laboratory interpretations were discordant. Average insurance reimbursement was 30.2%, with yield for commercial payers significantly higher, at 54.1%.

CONCLUSIONS

The detection rate of GS is equivalent and potentially superior to exome sequencing (ES). Reimbursement rates were variable but overall satisfactory for commercial insurers, and poor for government entities. In addition, we identify opportunities for improvement in the communication of results to families, likely translatable to other tests and other institutions.

Comparison of radiographic joint space width and magnetic resonance imaging for prediction of knee replacement: A longitudinal case-control study from the Osteoarthritis Initiative

OBJECTIVE

To evaluate whether change in fixed-location measures of radiographic joint space width (JSW) and cartilage thickness by MRI predict knee replacement.

METHODS

Knees replaced between 36 and 60 months' follow-up in the Osteoarthritis Initiative were each matched with one control by age, sex and radiographic status. Radiographic JSW was determined from fixed flexion radiographs and subregional femorotibial cartilage thickness from 3 T MRI. Changes between the annual visit before replacement (T0) and 2 years before T0 (T-2) were compared using conditional logistic regression.

RESULTS

One hundred and nineteen knees from 102 participants (55.5 % women; age 64.2 ± 8.7 [mean ± SD] years) were studied. Fixed-location JSW change at 22.5 % from medial to lateral differed more between replaced and control knees (case-control [cc] OR = 1.57; 95 % CI: 1.23-2.01) than minimum medial JSW change (ccOR = 1.38; 95 % CI: 1.11-1.71). Medial femorotibial cartilage loss displayed discrimination similar to minimum JSW, and central tibial cartilage loss similar to fixed-location JSW. Location-independent thinning and thickening scores were elevated prior to knee replacement.

CONCLUSIONS

Discrimination of structural progression between knee pre-placement cases versus controls was stronger for fixed-location than minimum radiographic JSW. MRI displayed similar discrimination to radiography and suggested greater simultaneous cartilage thickening and loss prior to knee replacement.

KEY POINTS

• Fixed-location JSW predicts surgical knee replacement more strongly than minimum JSW. • MRI predicts knee replacement with similar accuracy to radiographic JSW. • MRI reveals greater cartilage thinning and thickening prior to knee replacement.

Application of PET Tracers in Molecular Imaging for Breast Cancer

PURPOSE OF REVIEW

Molecular imaging with positron emission tomography (PET) is a powerful tool to visualize breast cancer characteristics. Nonetheless, implementation of PET imaging into cancer care is challenging, and essential steps have been outlined in the international "imaging biomarker roadmap." In this review, we identify hurdles and provide recommendations for implementation of PET biomarkers in breast cancer care, focusing on the PET tracers 2-[18F]-fluoro-2-deoxyglucose ([18F]-FDG), sodium [18F]-fluoride ([18F]-NaF), 16α-[18F]-fluoroestradiol ([18F]-FES), and [89Zr]-trastuzumab.

RECENT FINDINGS

Technical validity of [18F]-FDG, [18F]-NaF, and [18F]-FES is established and supported by international guidelines. However, support for clinical validity and utility is still pending for these PET tracers in breast cancer, due to variable endpoints and procedures in clinical studies. Assessment of clinical validity and utility is essential towards implementation; however, these steps are still lacking for PET biomarkers in breast cancer. This could be solved by adding PET biomarkers to randomized trials, development of imaging data warehouses, and harmonization of endpoints and procedures.

The application of texture quantification in hepatocellular carcinoma using CT and MRI: a review of perspectives and challenges

Recently, radiomic texture quantification of tumors has received much attention from radiologists, scientists, and stakeholders because several results have shown the feasibility of using the technique to diagnose and manage oncological conditions. In patients with hepatocellular carcinoma, radiomics has been applied in all stages of tumor evaluation, including diagnosis and characterization of the genotypic behavior of the tumor, monitoring of treatment responses and prediction of various clinical endpoints. It is also useful in selecting suitable candidates for specific treatment strategies. However, the clinical validation of hepatocellular carcinoma radiomics is limited by challenges in imaging protocol and data acquisition parameters, challenges in segmentation techniques, dimensionality reduction, and modeling methods. Identification of the best segmentation and optimal modeling methods, as well as texture features most stable to imaging protocol variability would go a long way in harmonizing HCC radiomics for personalized patient care. This article reviews the process of HCC radiomics, its clinical applications, associated challenges, and current optimization strategies.

A new TaqMan method for the reliable diagnosis of Ehrlichia spp. in canine whole blood

Background

Ehrlichiosis is an important emerging infectious disease of the canid family and humans worldwide. To date, no extensive evaluation or validation of a molecular diagnostic test for ehrlichiosis has been published. Here, we present data for a newly designed TaqMan assay and compare its performance to a commercial technology (PCRun®). Both of these real-time methods of analysis were evaluated using a comprehensive number of prospective and retrospective samples collected from dogs exhibiting symptoms of ehrlichiosis.

Results

Whole blood samples collected from dogs, retrospectively in the United Kingdom and prospectively in Israel, were analysed for the presence of Ehrlichia canis and Ehrlichia minasensis DNA using the TaqMan PCR, developed specifically for this study. The results were compared to those of a real time commercial isothermal amplification method (PCRun® system developed by Biogal Galed Labs ACS, Galed, Israel). The sensitivity and specificity (CI: 95%) of the TaqMan PCR and PCRun® were both determined to be 100% and absolute, for all of the samples tested. Interestingly, both tests were demonstrated to be highly comparable, irrespective of differences in amplification chemistry or sequences targeted. Host differences, incidence of disease and geographical location of the isolates had little impact on the positivity recorded by each of the diagnostic methods.

Conclusions

It was evident that both amplification methods were equally suited for diagnosing canine ehrlichiosis and while the PCRun® clearly amplified all clinically relevant Ehrlichia species known to infect dogs and humans, the TaqMan method was more specific for E. canis and E. minasensis. This work demonstrates that despite good analytical sensitivities and specificities for Ehrlichia spp. neither method could fully account for the clinical diagnosis of thrombocytopenia.

Electronic supplementary material

The online version of this article (10.1186/s13071-018-2914-5) contains supplementary material, which is available to authorized users.

Clinical validation of a GPU-based Monte Carlo dose engine of a commercial treatment planning system for pencil beam scanning proton therapy

PURPOSE

To perform the validation of the GPU-based (Graphical Processing Unit based) proton Monte Carlo (MC) dose engine implemented in a commercial TPS (RayStation 10B) and to report final dose calculation times for clinical cases.

MATERIALS AND METHODS

440 patients treated at the Proton Therapy Center of Trento, Italy, between 2018 and 2019 were selected for this study. 636 approved plans with 3361 beams computed with the clinically implemented CPU-MC dose engine (version 4.2 and 4.5), were used for the validation of the new algorithm. For each beam, the dose was recalculated using the new GPU-MC dose engine with the initial CPU computation settings and compared to the original CPU-MC dose. Beam dose difference distributions were studied to ensure that the two dose distributions were equal within the expected fluctuations of the MC statistical uncertainty (s) of each computation. Plan dose distributions were compared with respect to the dosimetric indices D₉₈, D₅₀ and D₁ of all ROIs defined as targets. A complete assessment of the computation time as a function of s and dose grid voxel size was done.

RESULTS

The median over all mean beam dose differences between CPU- and GPU-MC was -0.01% and the median of the corresponding standard deviations was close to (√2s) both for simulations with an s of 0.5% and 1.0% per beam. This shows that the two dose distributions can be considered equal. All the DVH indices showed an average difference below 0.04%. About half of the plans were computed with 1.0% statistical uncertainty on a 2 mm dose calculation grid, for which the median computation time was 5.2 s. The median computational speed for all plans in the study was 8.4 million protons/second.

CONCLUSION

A validation of a clinical MC algorithm running on GPU was performed on a large pool of patients treated with pencil beam scanning proton therapy. We demonstrated that the differences with the previous CPU-based MC were only due to the intrinsic statistical fluctuations of the MC method, which translated to insignificant differences on plan dose level. The significant increase in dose calculation speed is expected to facilitate new clinical workflows.

A clinically validated whole genome pipeline for structural variant detection and analysis

Background

With the continuing decrease in cost of whole genome sequencing (WGS), we have already approached the point of inflection where WGS testing has become economically feasible, facilitating broader access to the benefits that are helping to define WGS as the new diagnostic standard. WGS provides unique opportunities for detection of structural variants; however, such analyses, despite being recognized by the research community, have not previously made their way into routine clinical practice.

Results

We have developed a clinically validated pipeline for highly specific and sensitive detection of structural variants basing on 30X PCR-free WGS. Using a combination of breakpoint analysis of split and discordant reads, and read depth analysis, the pipeline identifies structural variants down to single base pair resolution. False positives are minimized using calculations for loss of heterozygosity and bi-modal heterozygous variant allele frequencies to enhance heterozygous deletion and duplication detection respectively. Compound and potential compound combinations of structural variants and small sequence changes are automatically detected. To facilitate clinical interpretation, identified variants are annotated with phenotype information derived from HGMD Professional and population allele frequencies derived from public and Variantyx allele frequency databases. Single base pair resolution enables easy visual inspection of potentially causal variants using the IGV genome browser as well as easy biochemical validation via PCR. Analytical and clinical sensitivity and specificity of the pipeline has been validated using analysis of Genome in a Bottle reference genomes and known positive samples confirmed by orthogonal sequencing technologies.

Conclusion

Consistent read depth of PCR-free WGS enables reliable detection of structural variants of any size. Annotation both on gene and variant level allows clinicians to match reported patient phenotype with detected variants and confidently report causative finding in all clinical cases used for validation.

Electronic supplementary material

The online version of this article (10.1186/s12864-019-5866-z) contains supplementary material, which is available to authorized users.

Cardiothoracic ratio measurement using artificial intelligence: observer and method validation studies

Background

Artificial Intelligence (AI) is a promising tool for cardiothoracic ratio (CTR) measurement that has been technically validated but not clinically evaluated on a large dataset. We observed and validated AI and manual methods for CTR measurement using a large dataset and investigated the clinical utility of the AI method.

Methods

Five thousand normal chest x-rays and 2,517 images with cardiomegaly and CTR values, were analyzed using manual, AI-assisted, and AI-only methods. AI-only methods obtained CTR values from a VGG-16 U-Net model. An in-house software was used to aid the manual and AI-assisted measurements and to record operating time. Intra and inter-observer experiments were performed on manual and AI-assisted methods and the averages were used in a method variation study. AI outcomes were graded in the AI-assisted method as excellent (accepted by both users independently), good (required adjustment), and poor (failed outcome). Bland–Altman plot with coefficient of variation (CV), and coefficient of determination (R-squared) were used to evaluate agreement and correlation between measurements. Finally, the performance of a cardiomegaly classification test was evaluated using a CTR cutoff at the standard (0.5), optimum, and maximum sensitivity.

Results

Manual CTR measurements on cardiomegaly data were comparable to previous radiologist reports (CV of 2.13% vs 2.04%). The observer and method variations from the AI-only method were about three times higher than from the manual method (CV of 5.78% vs 2.13%). AI assistance resulted in 40% excellent, 56% good, and 4% poor grading. AI assistance significantly improved agreement on inter-observer measurement compared to manual methods (CV; bias: 1.72%; − 0.61% vs 2.13%; − 1.62%) and was faster to perform (2.2 ± 2.4 secs vs 10.6 ± 1.5 secs). The R-squared and classification-test were not reliable indicators to verify that the AI-only method could replace manual operation.

Conclusions

AI alone is not yet suitable to replace manual operations due to its high variation, but it is useful to assist the radiologist because it can reduce observer variation and operation time. Agreement of measurement should be used to compare AI and manual methods, rather than R-square or classification performance tests.

Wearable seizure detection devices in refractory epilepsy

Epilepsy affects 50 million patients and their caregivers worldwide. Devices that facilitate the detection of seizures can have a large influence on a patient's quality of life, therapeutic decisions and the conduct of clinical trials with anti-epileptic drugs. This article provides an up-to-date overview and comparison between wearable seizure detection devices (WSDDs), taking into account the newly proposed standards for testing and clinical validation of devices. 16 devices were included in our comparison. The F1-score, combining the device's accurate recall and precision, was calculated for each of these devices and used to evaluate their performance. The devices were separated by development phase and ranked by F1-score from highest to lowest. We describe 16 WSDDs: 6 of which were accelerometry (ACM)-based, 3 surface electromyography-based, 1 was a wearable application of EEG, 4 had multimodal sensors and 2 other types of sensors. We observed a significant inconsistency in the description of performance measures. The devices in the most advanced development phase with the highest F1-scores incorporated ACM- and sEMG-based sensors to detect tonic-clonic seizures. This review highlights the importance of implementing standards for an optimal comparison and, therefore, improving the research and development of WSDDs. WSDDs can improve the patient's care and quality of life, decrease seizure underreporting and they could potentially prevent sudden-unexpected-death in epilepsy. We discuss the central role of the neurologist in the use of WSDDs, and why a business to business to consumer model is better than the current business to consumer model of most WSDDs.