BEGIN:VCALENDAR VERSION:2.0 PRODID:Linklings LLC BEGIN:VTIMEZONE TZID:America/Denver X-LIC-LOCATION:America/Denver BEGIN:DAYLIGHT TZOFFSETFROM:-0700 TZOFFSETTO:-0600 TZNAME:MDT DTSTART:19700308T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:-0600 TZOFFSETTO:-0700 TZNAME:MST DTSTART:19701101T020000 RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260422T000713Z LOCATION:506 DTSTART;TZID=America/Denver:20231112T144000 DTEND;TZID=America/Denver:20231112T145500 UID:submissions.supercomputing.org_SC23_sess430_ws_cafcw104@linklings.com SUMMARY:AI/ML-Derived Whole-Genome Predictor Prospectively and Clinically Predicts Survival and Response to Treatment in Brain Cancer DESCRIPTION:Sri Priya Ponnapalli (University of Utah); Penelope Miron and Kristy L. S. Miskimen (Case Western Reserve University School of Medicine) ; Kristin A. Waite (National Cancer Institute (NCI)); Nadiya Sosonkina (Hu dsonAlpha Clinical Services Lab LLC); Sara E. Coppens and Anthony C. Bryan (The Abigail Wexner Research Institute at Nationwide Children’s Hospital) ; Estevan P. Kiernan (Illumina, Inc.); Huanming Yang (Beijing Genomics Ins titute (BGI) -Shenzhen); Jay Bowen (The Abigail Wexner Research Institute at Nationwide Children’s Hospital); Ghunwa A. Nakouzi (HudsonAlpha Clinica l Services Lab LLC); Jill S. Barnholtz-Sloan (National Cancer Institute (N CI)); Andrew E. Sloan (Piedmont Healthcare, Case Western Reserve Universit y School of Medicine); Tiffany R. Hodges (Case Western Reserve University School of Medicine); and Orly Alter (University of Utah)\n\nCancer is comp lex, with contributing factors distributed across the entire genome affect ing every aspect of the disease. But typical artificial intelligence and m achine learning (AI/ML) would require 3B-patient training sets to generate predictive models from the whole 3B-nucleotide genome. As a result, tests remain limited to one to a few hundred genes. Prediction continues to rel y mostly on such factors as a tumor’s grade and the patient’s age. And the understanding and management of cancer continue to involve guesswork.\n\n A genome-wide pattern in glioblastoma brain cancer tumors was experimental ly validated in a retrospective clinical trial as the most accurate and pr ecise predictor of life expectancy and response to standard of care [1]. A pplicable to the general population, this predictor, the first to encompas s the whole genome, and predictors in lung, nerve, ovarian, and uterine ca ncers, were mathematically (re)discovered and computationally (re)validate d in open-source datasets from as few as 50–100 patients by using our AI/M L [2,3]. Data-agnostic, our algorithms, multi-tensor comparative spectral decompositions, extend the mathematics that underlies quantum mechanics to overcome typical AI/ML obstacles by not requiring large amounts of data, balanced data, or feature engineering. All other attempts to connect a gli oblastoma patient’s outcome with the tumor’s DNA copy numbers failed. For 70 years, the best indicator has been age. At 75–95% accuracy, our predict or is more accurate than and independent of age and all other indicators. Platform- and reference genome-agnostic, the predictor’s >99% precision is greater than the community consensus of <70% reproducibility based upon o ne to a few hundred genes. It describes mechanisms for transformation, and identifies drug targets and combinations of targets to sensitize tumors t o treatment.\n\nNow, in follow-up results from the trial we, first, show c orrect prospective prediction of the outcome of the five of the 79 patient s who were alive four years earlier, at the time of first results. Two pat ients, who were predicted to have shorter survival, lived less than five y ears from diagnosis, whereas of the three patients predicted to have longe r survival, one lived more than five, and the remaining two are alive >11. 5, years from diagnosis. Second, we demonstrate 100%-precise clinical pred iction for the 59 of the 79 patients with remaining tumor DNA by using who le-genome sequencing in a regulated laboratory. Third, we establish that t he risk that a tumor’s whole genome confers upon outcome, as is reflected by the predictor, is surpassed only by the patient’s access to radiotherap y.\n\nThis is a proof of principle that our AI/ML is uniquely suited for p ersonalized medicine. This also demonstrates that the inclusion of complet e genomes, and the normal diversity within, is, beyond fair AI/ML, a scien tific, engineering, and medical necessity, because a patient’s survival an d response to treatment are the outcome of their tumor’s whole genome. We conclude that our AI/ML-derived whole-genome predictors can take the guess work out of cancer. \n\n[1] Ponnapalli et al., APL Bioeng 4, 026106 (2020) ; https://doi.org/10.1063/1.5142559\n\n[2] Bradley et al., APL Bioeng 3, 0 36104 (2019); https://doi.org/10.1063/1.5099268\n\n[3] Alter et al., PNAS 100, 3351 (2003); https://doi.org/10.1073/pnas.0530258100\n\nTag: Applicat ions, State of the Practice\n\nRegistration Category: Workshop Reg Pass\n\ nSession Chairs: Lynn Borkon (Frederick National Laboratory for Cancer Res earch); Sally Ellingson (University of Kentucky); Sean Hanlon (National In stitutes of Health (NIH), National Cancer Institute (NCI)); Patricia Kovat ch (Icahn School of Medicine at Mount Sinai); and Eric Stahlberg (MD Ander son Cancer Center, University of Texas)\n\n END:VEVENT END:VCALENDAR