All round sensitivity of early cancer detection for novel blood-primarily based assay was 82% at 95% specificity particular cancer detection prices involve colorectal (96%), lung (85%), prostate (82%) and breast (73%)

Final results, which demonstrate the possible of Harbinger’s biology-informed, AI-driven method, at present getting validated in ten,000-topic CORE-HH clinical trial with the Sarah Cannon Study Institute

Findings published in the 2023 ASCO Annua Meeting Proceedings, a Journal of Clinical Oncology supplement

CAMBRIDGE, Mass., May possibly 25, 2023 /PRNewswire/ — Harbinger Well being, a biotechnology organization pioneering the detection of early cancer, announced these days information demonstrating the company’s biology-informed, AI-driven platform, HarbingerHx, has the possible for extremely sensitive multi-cancer diagnostic accuracy and especially these with early stage cancer. The abstract, “Novel blood-primarily based assay for detection of early stage multi-cancer,” was published in the 2023 American Society of Clinical Oncology (ASCO) Annual Meeting Proceedings, a Journal of Clinical Oncology supplement, as component of the 2023 ASCO Annual Meeting held practically and in Chicago, Illinois from June two-six, 2023.

Harbinger Well being (PRNewsfoto/Harbinger Well being)

“At Harbinger, we know that early detection is the 1st line of defense against cancer,” stated Stephen M. Hahn, M.D., CEO of Harbinger Well being. “These information gave us the conviction in our platform to type the organization and go just after the large unsolved difficulty of correct, informative early-stage cancer screening. We think we can advance the universally shared target of lowering cancer mortality with our blood-primarily based test that combines current genomic and epigenomic discoveries of early cancer with biology-informed artificial intelligence. This is the subsequent step in sophisticated cancer care.”

Abstract Specifics and Findings

Harbinger generated information from 1,046 subjects, 621 with newly diagnosed, therapy-naive cancer across 15 various kinds, and 425 folks with no history, diagnosis, or cancer symptoms. Making use of this sample cohort, the Organization created and applied a rigorous framework for predicting the likelihood of cancer utilizing a multi-layered logistic regression-primarily based machine understanding algorithm, producing final outputs of binary classification (cancer yes/no).

Story continues

All round sensitivity of cancer detection was 82% (95% self-assurance interval (CI): 72.7-91.%) at 95% specificity. Notably, the sensitivity was 74% (95% CI: 54.eight-92.7%) for stage 1 and 84% (95% CI: 65.three-one hundred%) for stage two. Additionally, the Organization was capable to properly predict cancer in 95% of patient samples with at least .037% tumor fraction. All round sensitivity for higher incident cancers had been breast (73%), prostate (82%), lung (85%), and colorectal (96%) at 95% specificity and the all round accuracy of tissue of origin prediction was 86% when the tumor fraction was higher than .1% in the best three most prevalent cancer kinds (breast, colorectal and lung). Importantly, technical variability was introduced when performing assay optimization tactics, but comparable functionality was observed when assessing subsets of stably processed samples, indicating robust functionality.

“As opposed to statistical approaches to recognize informative biomarkers, Harbinger’s method is informed by insights into particular biological events early in the course of tumorigenesis, permitting cancer detection in sufferers with quite low levels of circulating tumor DNA,” stated Professor Franziska Michor, Ph.D., abstract author, scientific advisor and co-founder of Harbinger Well being, and Professor of Computational Biology in the Division of Information Science at the Dana-Farber Cancer Institute, in the Division of Biostatistics at the Harvard T.H. Chan College of Public Well being, and in the Division of Stem Cell and Regenerative Biology at Harvard University. “We are pleased to share this information and appear forward to sharing additional updates as we progress our CORE-HH clinical trial.”

Harbinger’s platform is now getting validated in a ten,000-topic clinical trial, Cancer Origin Epigenetics – Harbinger Well being (CORE-HH, NCT05435066), in conjunction with the Sarah Cannon Study Institute. Presently on target to full enrollment by year finish, the CORE-HH study is created to enroll a diverse and representative population with and with no cancer at up to 40 U.S. web-sites.

About Harbinger Well being

Harbinger Well being is pioneering the detection of early cancer and enabling foundationally new approaches to cancer screening, diagnosis and management. The organization combines advances in artificial intelligence with proprietary insights into the biology of the beginnings of cancer to recognize cancer ahead of it is visible or symptomatic with the aim of creating a low-expense, multi-cancer blood test. Harbinger envisions a future exactly where, as an alternative of maintaining cancer from spreading, it could be kept from forming, producing a cancer diagnosis a routine well being difficulty to be addressed rather than a life-altering occasion to be feared with profound implications for folks, healthcare systems and societies. Harbinger was founded by Flagship Pioneering just after 3 years of foundational study in its Labs unit and launched in 2020. Find out additional about Harbinger by going to Harbinger-Well being.com or following us on Twitter (@harbingerhlth) and LinkedIn.

Media Speak to
press@harbinger-well being.com

Cision

View original content material to download multimedia:https://www.prnewswire.com/news-releases/harbinger-well being-announces-benefits-demonstrating-higher-sensitivity-of-early-stage-multi-cancer-detection-platform-at-asco-annual-meeting-2023-301835194.html

Supply Harbinger Well being