Individual Reference Intervals for Personalised Interpretation of Clinical and Metabolomics Measurements

Dynamic mirroring: unveiling the role of digital twins, artificial intelligence and synthetic data for personalized medicine in laboratory medicine

In recent years, the integration of technological advancements and digitalization into healthcare has brought about a remarkable transformation in care delivery and patient management. Among these advancements, the concept of digital twins (DTs) has recently gained attention as a tool with substantial transformative potential in different clinical contexts. DTs are virtual representations of a physical entity (e.g., a patient or an organ) or systems (e.g., hospital wards, including laboratories), continuously updated with real-time data to mirror its real-world counterpart. DTs can be utilized to monitor and customize health care by simulating an individual's health status based on information from wearables, medical devices, diagnostic tests, and electronic health records. In addition, DTs can be used to define personalized treatment plans. In this study, we focused on some possible applications of DTs in laboratory medicine when used with AI and synthetic data obtained by generative AI. The first point discussed how biological variation (BV) application could be tailored to individuals, considering population-derived BV data on laboratory parameters and circadian or ultradian variations. Another application could be enhancing the interpretation of tumor markers in advanced cancer therapy and treatments. Furthermore, DTs applications might derive personalized reference intervals, also considering BV data or they can be used to improve test results interpretation. DT's widespread adoption in healthcare is not imminent, but it is not far off. This technology will likely offer innovative and definitive solutions for dynamically evaluating treatments and more precise diagnoses for personalized medicine.

Advancing personalized medicine: Integrating statistical algorithms with omics and nano-omics for enhanced diagnostic accuracy and treatment efficacy

Medical laboratory services enable precise measurement of thousands of biomolecules and have become an inseparable part of high-quality healthcare services, exerting a profound influence on global health outcomes. The integration of omics technologies into laboratory medicine has transformed healthcare, enabling personalized treatments and interventions based on individuals' distinct genetic and metabolic profiles. Interpreting laboratory data relies on reliable reference values. Presently, population-derived references are used for individuals, risking misinterpretation due to population heterogeneity, and leading to medical errors. Thus, personalized references are crucial for precise interpretation of individual laboratory results, and the interpretation of omics data should be based on individualized reference values. We reviewed recent advancements in personalized laboratory medicine, focusing on personalized omics, and discussed strategies for implementing personalized statistical approaches in omics technologies to improve global health and concluded that personalized statistical algorithms for interpretation of omics data have great potential to enhance global health. Finally, we demonstrated that the convergence of nanotechnology and omics sciences is transforming personalized laboratory medicine by providing unparalleled diagnostic precision and innovative therapeutic strategies.

Use of hospital big data to optimize and personalize laboratory test interpretation with an application

BACKGROUND AND AIMS

In laboratory medicine, test results are generally interpreted with 95% reference intervals but correlations between laboratory tests are usually ignored. We aimed to use hospital big data to optimize and personalize laboratory data interpretation, focusing on platelet count.

MATERIAL AND METHODS

Laboratory tests were extracted from the hospital database and exploited by an algorithmic stepwise procedure. For any given laboratory test Y, an "optimized and personalized reference population" was defined by keeping only patients whose laboratory values for all Y-correlated tests fell within their own usual reference intervals, and by partitioning groups by individual-specific variables like sex and age category. The method was applied to platelet count.

RESULTS

Laboratory data were recorded for 28,082 individuals. At the end of the algorithmic process, seven correlated laboratory tests were chosen, resulting in a reference sample of 159 platelet counts. A new 95 % reference interval was constructed [152-334 × 109/L], notably reduced (27.2 %) compared to conventional reference values [150-400 × 109/L]. The reference interval was validated on a sample of 2,129 patients from another downtown laboratory, emphasizing the potential transference of the hospital-derived reference limits.

CONCLUSION

This method offers new perspectives in laboratory data interpretation, especially in patient screening and longitudinal follow-up.

Systematic identification of the role of gut microbiota in mental disorders: a TwinsUK cohort study

Mental disorders are complex disorders influenced by multiple genetic, environmental, and biological factors. Specific microbiota imbalances seem to affect mental health status. However, the mechanisms by which microbiota disturbances impact the presence of depression, stress, anxiety, and eating disorders remain poorly understood. Currently, there are no robust biomarkers identified. We proposed a novel pyramid-layer design to accurately identify microbial/metabolomic signatures underlying mental disorders in the TwinsUK registry. Monozygotic and dizygotic twins discordant for mental disorders were screened, in a pairwise manner, for differentially abundant bacterial genera and circulating metabolites. In addition, multivariate analyses were performed, accounting for individual-level confounders. Our pyramid-layer study design allowed us to overcome the limitations of cross-sectional study designs with significant confounder effects and resulted in an association of the abundance of genus Parabacteroides with the diagnosis of mental disorders. Future research should explore the potential role of Parabacteroides as a mediator of mental health status. Our results indicate the potential role of the microbiome as a modifier in mental disorders that might contribute to the development of novel methodologies to assess personal risk and intervention strategies.

Indirect estimation of pediatric reference interval via density graph deep embedded clustering

Establishing reference intervals (RIs) for pediatric patients is crucial in clinical decision-making, and there is a critical gap of pediatric RIs in China. However, the direct sampling technique for establishing RIs is resource-intensive and ethically challenging. Indirect estimation methods, such as unsupervised clustering algorithms, have emerged as potential alternatives for predicting reference intervals. This study introduces deep graph clustering methods into indirect estimation of pediatric reference intervals. Specifically, we propose a Density Graph Deep Embedded Clustering (DGDEC) algorithm, which incorporates a density feature extractor to enhance sample representation and provides additional perspectives for distinguishing different levels of health status among populations. Additionally, we construct an adjacency matrix by computing the similarity between samples after feature enhancement. The DGDEC algorithm leverages the adjacency matrix to capture the interrelationships between patients and divides patients into different groups, thereby estimating reference intervals for the potential healthy population. The experimental results demonstrate that when compared to other indirect estimation techniques, our method ensures the predicted pediatric reference intervals in different age and gender groups are closer to the true values while maintaining good generalization performance. Additionally, through ablation experiments, our study confirms that the similarity between patients and the multi-scale density features of samples can effectively describe the potential health status of patients.

Reference intervals: past, present, and future

Clinical laboratory test results alone are of little value in diagnosing, treating, and monitoring health conditions; as such, a clinically actionable cutoff or reference interval is required to provide context for result interpretation. Healthcare practitioners base their diagnoses, follow-up treatments, and subsequent testing on these reference points. However, they may not be aware of inherent limitations related to the definition and derivation of reference intervals. Laboratorians are responsible for providing the reference intervals they report with results. Yet, the establishment and verification of reference intervals using conventional direct methods are complicated by resource constraints or unique patient demographics. To facilitate standardized reference interval best practices, multiple global scientific societies are actively drafting guidelines and seeking funding to promote these initiatives. Numerous national and international multicenter collaborations demonstrate the ability to leverage combined resources to conduct large reference interval studies by direct methods. However, not all demographics are equally accessible. Novel indirect methods are attractive solutions that utilize computational methods to define reference distributions and reference intervals from mixed data sets of pathologic and non-pathologic patient test results. In an effort to make reference intervals more accurate and personalized, individual-based reference intervals are shown to be more useful than population-based reference intervals in detecting clinically significant analyte changes in a patient that might otherwise go unrecognized when using wider, population-based reference intervals. Additionally, continuous reference intervals can provide more accurate ranges as compared to age-based partitions for individuals that are near the ends of the age partition. The advantages and disadvantages of different reference interval approaches as well as the advancement of non-conventional reference interval studies are discussed in this review.

A new multimodal paradigm for biomarkers longitudinal monitoring: a clinical application to women steroid profiles in urine and blood

BACKGROUND

Most current state-of-the-art strategies to generate individual adaptive reference ranges are designed to monitor one clinical parameter at a time. An innovative methodology is proposed for the simultaneous longitudinal monitoring of multiple biomarkers. The estimation of individual thresholds is performed by applying a Bayesian modeling strategy to a multivariate score integrating several biomarkers (compound concentration and/or ratio). This multimodal monitoring was applied to data from a clinical study involving 14 female volunteers with normal menstrual cycles receiving testosterone via transdermal route, as to test its ability to detect testosterone administration. The study samples consisted of urine and blood collected during 4 weeks of a control phase and 4 weeks with a daily testosterone gel application.

RESULTS

Integrating multiple biomarkers improved the detection of testosterone gel administration with substantially higher sensitivity compared with the distinct follow-up of each biomarker, when applied to selected urine and serum steroid biomarkers, as well as the combination of both. Among the 175 known positive samples, 38% were identified by the multimodal approach using urine biomarkers, 79% using serum biomarkers and 83% by combining biomarkers from both biological matrices, whereas 10%, 67% and 64% were respectively detected using standard unimodal monitoring.

SIGNIFICANCE AND NOVELTY

The detection of abnormal patterns can be improved using multimodal approaches. The combination of urine and serum biomarkers reduced the overall number of false-negatives, thus evidencing promising complementarity between urine and blood sampling for doping control, as highlighted in the case of the use of transdermal testosterone preparations. The generation in a multimodal setting of adaptive and personalized reference ranges opens up new opportunities in clinical and anti-doping profiling. The integration of multiple parameters in a longitudinal monitoring is expected to provide a more complete evaluation of individual profiles generating actionable intelligence to further guide sample collection, analysis protocols and decision-making in clinics and anti-doping.

MetaboVariation: Exploring Individual Variation in Metabolite Levels

To date, most metabolomics biomarker research has focused on identifying disease biomarkers. However, there is a need for biomarkers of early metabolic dysfunction to identify individuals who would benefit from lifestyle interventions. Concomitantly, there is a need to develop strategies to analyse metabolomics data at an individual level. We propose "MetaboVariation", a method that models repeated measurements on individuals to explore fluctuations in metabolite levels at an individual level. MetaboVariation employs a Bayesian generalised linear model to flag individuals with intra-individual variations in their metabolite levels across multiple measurements. MetaboVariation models repeated metabolite levels as a function of explanatory variables while accounting for intra-individual variation. The posterior predictive distribution of metabolite levels at the individual level is available, and is used to flag individuals with observed metabolite levels outside the 95% highest posterior density prediction interval at a given time point. MetaboVariation was applied to a dataset containing metabolite levels for 20 metabolites, measured once every four months, in 164 individuals. A total of 28% of individuals with intra-individual variations in three or more metabolites were flagged. An R package for MetaboVariation was developed with an embedded R Shiny web application. To summarize, MetaboVariation has made considerable progress in developing strategies for analysing metabolomics data at the individual level, thus paving the way toward personalised healthcare.