"The impact and challenges of more #data: retraining of a Bayesian #genotype-#phenotype model on ~1M additional #patients to resolve #VUS" #TrevorWilliams #YuyaKobayashi et al. @GeneticsSociety #ASHG25 "We demonstrate the benefits & utility of retraining #ClinicalVariantModeling (a #machinelearning pipeline). RT resulted in large increases in available data that enhanced the global model." #access #reducedisparities #precisionmedicine #precisiontherapy #clinicaltrials #precisionprevention #precisionclassification @Invitae @Labcorp go.pheedloop.com/events/EVEA…

"VUS results do not lead to over utilization or increased cost of health care" in #patients w/ #genetictesting Study of >22K #patients w/ #breastcancer & #germlinetesting, those w/ VUS showed no difference from those w/ negative results in uptake of surgical, therapeutic, risk-reducing, surveillance procedures, also no higher screening or treatment costs in #patients w/ VUS vs negative. Because, "A variant of uncertain significance (VUS) should not be used in clinical decision making", per the @TheACMG c/w @WhitworthMD et al. jamanetwork.com/journals/jam… @JAMANetworkOpen #access #reducedisparities #universal #genetictesting #precisionmedicine #precisiontherapy #clinicaltrials #precisionprevention #AI #precisionclassification jnccn.org/view/journals/jncc… #SarahNielsen et al @JNCCN @Invitae @Labcorp @beitsch @AllisonKurian

595 high performance, scalable #MachineLearning/ #ML #ClinicalVariantModels/#CVMs using #genetictesting & clinical data from 3.5M #patients @Invitae, generated new evidence to improve classification for >5,000 VUS observed in >200K #patients, 94% would receive benign evidence. #ML #CVMs applied to just 17 genes (inc. #Lynchsyndrome, #DMD, #MEN1, #NF1, #RB1, #SMAD4, #STK11) reclassified >1K VUS, benefitting ~45K #patients reducing VUS rates by 9-49% per condition, w/ >99% of reclassifications VUS -> B/LB link.springer.com/article/10… #DrTobyManders et al. @Labcorp #HumanGenetics @SpringerNature #access #reducedisparities #reduceVUS #germlinetesting #AI #precisionclassification

"...our results do not support any tumor triage method...we recommend #universal #germlinetesting for all newly diagnosed #patients w/ #CRC..." Abu-Ghazaleh et al. Studies performing #universal #germlinetesting on all #patients w/ #CRC found 5.1% prevalence of #Lynchsyndrome "Prevalence of LS across multiple ethnic, geographic, and subject settings is remarkably similar." #SystematicReview & #MetaAnalysis supports #universal #germline #CRC screening for #Lynchsyndrome sciencedirect.com/science/ar… @GIMJournal @TheACMG c/w pubmed.ncbi.nlm.nih.gov/3385… @AGA_CGH @MayoClinic ascopubs.org/doi/full/10.120… @JCOPO_ASCO @ASCO @Invitae @Labcorp #access #reducedisparities #precisionmedicine #precisiontherapy #clinicaltrials #precisionprevention #AI #precisionclassification

"The impact of systematized generation, evaluation, and incorporation of #machinelearning algorithms for #clinical #variantclassification" in #germline #genetictesting, #LaureFresard et al. Just one of the reasons @Invitae/#LabcorpGenetics/@Labcorp has the highest published variant classification accuracy for #germlinetesting, exceeding the standards set by @TheACMG & @AMPath (jamanetwork.com/journals/jam… @JAMANetworkOpen ) #access #reducedisparities #precisionmedicine #precisiontherapy #clinicaltrials #precisionprevention #AI #precisionclassification medrxiv.org/content/10.1101/…

Study showed that #Sherloc (@Invitae) classifies a variant as likely pathogenic or likely benign with 99.9% accuracy, far exceeding the 90% confidence target set by @TheACMG/@AMPath. ...most powerful source of VUS reclassification was the application of #machinelearning to reanalyze existing data (database of >5 million #patients)...particularly among historically underrepresented race, ethnicity, and ancestry groups...#Machinelearning models co-developed by #clinicalgeneticists & #AI scientists. #ArtificialIntelligence #ML are poised to play a signficant role in advancing #genetictesting practices. the-scientist.com/genetic-va… #DrYuyaKobayashi @TheScientistLLC jamanetwork.com/journals/jam… @JAMANetworkOpen #access #reducedisparities #germlinetesting #precisionmedicine #precisiontherapy #clinicaltrials #precisionprevention #precisionclassification @Labcorp