Improving the Detection of Potential Cases of Familial Hypercholesterolemia: Could Machine Learning Be Part of the Solution?

BACKGROUND: Familial hypercholesterolemia (FH), while highly prevalent, is a significantly underdiagnosed monogenic disorder. Improved detection could reduce the large number of cardiovascular events attributable to poor case finding. We aimed to assess whether machine learning algorithms outperform clinical diagnostic criteria (signs, history, and biomarkers) and the recommended screening criteria in the United Kingdom in identifying individuals with FH-causing variants, presenting a scalable screening criteria for general populations. METHODS AND RESULTS: Analysis included UK Biobank participants with whole exome sequencing, classifying them as having FH when (likely) pathogenic variants were detected in their LDLR, APOB, or PCSK9 genes. Data were stratified into 3 data sets for (1) feature importance analysis; (2) deriving state-of-the-art statistical and machine learning models; (3) evaluating models' predictive performance against clinical diagnostic and screening criteria: Dutch Lipid Clinic Network, Simon Broome, Make Early Diagnosis to Prevent Early Death, and Familial Case Ascertainment Tool. One thousand and three of 454 710 participants were classified as having FH. A Stacking Ensemble model yielded the best predictive performance (sensitivity, 74.93%; precision, 0.61%; accuracy, 72.80%, area under the receiver operating characteristic curve, 79.12%) and outperformed clinical diagnostic criteria and the recommended screening criteria in identifying FH variant carriers within the validation data set (figures for Familial Case Ascertainment Tool, the best baseline model, were 69.55%, 0.44%, 65.43%, and 71.12%, respectively). Our model decreased the number needed to screen compared with the Familial Case Ascertainment Tool (164 versus 227). CONCLUSIONS: Our machine learning-derived model provides a higher pretest probability of identifying individuals with a molecular diagnosis of FH compared with current approaches. This provides a promising, cost-effective scalable tool for implementation into electronic health records to prioritize potential FH cases for genetic confirmation.

Application of wearables for remote monitoring of oncology patients: A scoping review.

Objective: This review aims to systematically map and categorize the current state of wearable applications among oncology patients and to identify determinants impeding clinical implementation. Methods: A Medline, Embase and clinicaltrials.gov search identified journal articles, conference abstracts, letters, reports, dissertations and registered studies on the use of wearables in patients with malignancies published up to 10 November 2021. Results: Of 2509 records identified, 112 met the eligibility criteria. Of these, 9.8% (11/112) were RCTs and 47.3% (53/112) of publications were observational. Wearables were investigated pre-treatment (2.7%; 3/112), during treatment (34.8%; 39/112), post-treatment (17.9%; 20/112), in survivors (27.7%; 31/112) and in non-specified or multiple treatment phases (17.0%; 19/112). Medical-grade wearables were applied in 22.3% (25/112) of publications. Primary objectives ranged from technical feasibility (8.0%; 9/112), user feasibility (42.9%; 48/112) and correlational analysis (40.2%; 45/112) to outcome change analysis (8.9%; 10/112). Outcome change was mostly investigated regarding physical activity improvement (80.0%; 8/10). Most publications (42.9%; 48/112) and registered studies (39.3%; 24/61) featured multiple cancer types, with breast cancer as the most prevalent specific type (22.3% in publications, 16.4% in registered studies). Conclusions: Most studies among oncology patients using wearables are focused on assessing the user feasibility of consumer-grade wearables, whereas rates of RCTs assessing clinical efficacy are low. Substantial improvements in clinically relevant endpoints by the use of wearables, such as morbidity and mortality are yet to be demonstrated.

Wearable based monitoring and self-supervised contrastive learning detect clinical complications during treatment of Hematologic malignancies.

Serious clinical complications (SCC; CTCAE grade ≥ 3) occur frequently in patients treated for hematological malignancies. Early diagnosis and treatment of SCC are essential to improve outcomes. Here we report a deep learning model-derived SCC-Score to detect and predict SCC from time-series data recorded continuously by a medical wearable. In this single-arm, single-center, observational cohort study, vital signs and physical activity were recorded with a wearable for 31,234 h in 79 patients (54 Inpatient Cohort (IC)/25 Outpatient Cohort (OC)). Hours with normal physical functioning without evidence of SCC (regular hours) were presented to a deep neural network that was trained by a self-supervised contrastive learning objective to extract features from the time series that are typical in regular periods. The model was used to calculate a SCC-Score that measures the dissimilarity to regular features. Detection and prediction performance of the SCC-Score was compared to clinical documentation of SCC (AUROC ± SD). In total 124 clinically documented SCC occurred in the IC, 16 in the OC. Detection of SCC was achieved in the IC with a sensitivity of 79.7% and specificity of 87.9%, with AUROC of 0.91 ± 0.01 (OC sensitivity 77.4%, specificity 81.8%, AUROC 0.87 ± 0.02). Prediction of infectious SCC was possible up to 2 days before clinical diagnosis (AUROC 0.90 at -24 h and 0.88 at -48 h). We provide proof of principle for the detection and prediction of SCC in patients treated for hematological malignancies using wearable data and a deep learning model. As a consequence, remote patient monitoring may enable pre-emptive complication management.

Familial Hypercholesterolemia: JACC Focus Seminar 4/4.

Detecting familial hypercholesterolemia (FH) early and "normalizing" low-density lipoprotein (LDL) cholesterol values are the 2 pillars for effective cardiovascular disease prevention in FH. Combining lipid-lowering therapies targeting synergistic/complementary metabolic pathways makes this feasible, even among severe phenotypes. For LDL receptor-dependent treatments, PCSK9 remains the main target for adjunctive therapy to statins and ezetimibe through a variety of approaches. These include protein inhibition (adnectins), inhibition of translation at mRNA level (antisense oligonucleotides or small interfering RNA), and creation of loss-of-function mutations through base-pair editing. For patients with little LDL receptor function, LDL receptor-independent treatment targeting ANGPTL3 through monoclonal therapies are now available, or in the future, antisense/small interfering RNA-based approaches offer alternative approaches. Finally, first-in-human studies are ongoing, testing adenovirus-mediated gene therapy transducing healthy LDLR DNA in patients with HoFH. Further development of the CRISPR cas technology, which has shown promising results in vivo on introducing PCSK9 loss-of-function mutations, will move a single-dose, curative treatment for FH closer.

A meta-analysis of medications directed against PCSK9 in familial hypercholesterolemia.

BACKGROUND AND AIMS: Several medications targeting PCSK9 reduce LDL-cholesterol (LDL-C) in heterozygous familial hypercholesterolemia (HeFH). We aimed to assess in patients diagnosed clinically as HeFH, whether LDL-C reduction varied by different therapeutic approaches to PCSK9-targeting or by the underlying genetic variant. METHODS: We conducted a random-effects meta-analysis of randomised clinical trials assessing PCSK9-targeting therapies, namely alirocumab, evolocumab and inclisiran, in patients with clinically diagnosed HeFH and restricted analyses to those patients in whom genotypic data were available. A search of MEDLINE and Embase identified eligible trials published between inception and June 29, 2020. We included trials of sufficient duration to allow for a stable treatment effect: ~12 weeks for monoclonal antibodies (mAbs) (alirocumab, evolocumab) and ~1 year for small interfering RNA (siRNA) (inclisiran). Single-moderator meta-regression comparing mean percentage LDL-C reduction between mAbs and siRNA as well as PCSK9-targeting therapies between different genotypes was used to assess heterogeneity. RESULTS: Eight trials of HeFH met our inclusion criteria, including 1887 genotyped patients. Among monogenic HeFH cases (N = 1347) the LDL-C reduction from baseline was 46.12% (95%CI 48.4-43.9) for siRNA and 50.4% (59.3-41.4) for mAbs compared to control, without evidence of significant heterogeneity between treatment (QM = 0.32, df = 1, p = 0.57). Irrespective of therapeutic approach to PCSK9-targeting, reductions in LDL-C were generally consistent across genetic variants (LDL-Receptor variants, LDL-Receptor variants of unknown significance, Apolipoprotein B variants, two variants and no variant) (QM = 8.3, df = 4, p = 0.08). CONCLUSIONS: Among patients with HeFH, the LDL-C-lowering effect of PCSK9-targeting medications did not show statistical heterogeneity across different drug-classes and across genetic variants.

Bempedoic acid, an inhibitor of ATP citrate lyase for the treatment of hypercholesterolemia: early indications and potential.

INTRODUCTION: The lowering of low-density lipoprotein cholesterol (LDL-C), regardless of the method used, results in a reduction of cardiovascular events. Bempedoic acid is a new and until now, the only approved adenosine triphosphate citrate lyase inhibitor that works through the cholesterol-synthesis pathway (similar to statins) that leads to a safe and effective reduction in LDL-C. AREAS COVERED: We review clinical phase 2 and 3 studies on bempedoic acid's lipid-lowering effect and approved indications. EXPERT OPINION: In the United States, bempedoic acid is currently approved for use in secondary prevention. In primary prevention, its approval is limited to individuals with heterozygous familial hypercholesterolemia (FH). However, its tolerability and safety profile may warrant its use in primary prevention besides FH. Even though its efficacy appears weaker than high-intensity statins, it may be a useful adjunct in individuals who achieve less than desirable LDL-C reductions with statins or who cannot tolerate statins, where bempedoic acid alone or in combination with ezetimibe may be useful alternatives.

Small interfering RNA to proprotein convertase subtilisin/kexin type 9: transforming LDL-cholesterol-lowering strategies.

PURPOSE OF REVIEW: Proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibition is a new strategy to reduce LDL cholesterol (LDL-C), that is currently pursued by mAbs. A promising novel approach to target PCSK9 is using small interfering RNAs to inhibit hepatic PCSK9 synthesis. The first small interfering RNA developed for this purpose is inclisiran. Here, we review its clinical trial data and potential impact on patient management. RECENT FINDINGS: Inclisiran achieves sustained, additional 50% LDL-C reduction in patients receiving background statin therapy. Resulting LDL-C changes can be maintained by an infrequent dosing regimen with twice per year injections, that appear safe and well tolerated. Thus far, inclisiran has been studied in patients with established cardiovascular disease, high-risk primary prevention and in patients with familial hypercholesterolemia. SUMMARY: High and very high-risk individuals may benefit from the additional LDL-C-lowering effect of inclisiran when added to current lipid-lowering therapies. Furthermore, the simple dosing regimen may improve the convenience for users and facilitate patient adherence to therapy. The safety and convenience of inclisiran may offer new opportunities for population health.

Familial hypercholesterolemia: is it time to separate monogenic from polygenic familial hypercholesterolemia?

PURPOSE OF REVIEW: This review explores the concepts of monogenic and the so-called polygenic familial hypercholesterolemia and how the identification of familial hypercholesterolemia as a monogenic condition and its separation from polygenic primary hypercholesterolemia may have implications for clinical practice. RECENT FINDINGS: Through genetic testing, a mutation in any of the three known autosomal dominant familial hypercholesterolemia-causing genes is found in 60-80% of cases with a clinical diagnosis of definite familial hypercholesterolemia. As individuals with a polygenic basis for their hypercholesterolemia do not follow the same inheritance pattern observed in monogenic familial hypercholesterolemia, the use of family-based cascade screening in individuals with a polygenic origin is not recommend, as only 30% of relatives have an elevated LDL-C compared to the 50% in monogenic families. The presence of a causative monogenic mutation associates the highest cardiovascular risk vs. not having a mutation or having a polygenic background, providing prognostic information independent of LDL-C. It may also help assess intensity of interventions. Treatment adherence also seems to be higher after monogenic confirmation of hypercholesterolemia. SUMMARY: Knowledge about the genetic status of an individual with clinical familial hypercholesterolemia (monogenic vs. polygenic) can provide a more informed understanding to evaluating risk, managing disease and opportunities for screening strategies.