Notable Labs launches rolling blood cancer trial to test its AI system


Precision oncology firm Notable Labs is launching its first self-sponsored clinical trial, designed from the ground up to help validate its cancer patient matching platform over the long term. The observational study--which also represents the company's largest trial to date--aims to enroll up to 1,000 participants with a variety of blood cancers and will follow them for at least one year as they receive physician-led standard-of-care therapies at different sites across the U.S. and Canada. Separately, Notable's phenotypic and artificial intelligence-powered platform will be tested against multiple patient samples collected over time to provide a longitudinal view of its predictive value based on cancer mutations, drug responses and the outcomes of each participant. It will also search for patterns useful in the development of new treatments. The company combines AI approaches with automated lab processes to determine which drugs or combinations will be most effective for specific cancers.

Artificial intelligence and algorithms join the fight against cancer


BACKGROUND: Cancer is the uncontrolled growth of abnormal cells anywhere in a body. There are over 200 types of cancer. Anything that may cause a normal body cell to develop abnormally can potentially cause cancer. Inherited genetic defects, infections, environmental factors such as air pollution, and poor lifestyle choices such as smoking and heavy alcohol use can damage DNA and lead to cancer. Signs and symptoms depend on the specific type and grade of cancer and can include fatigue, weight loss, pain, skin changes, and changes in bowel or bladder function. There are many tests to screen and presumptively diagnose cancer, though the definite diagnosis is made by examination of a biopsy sample of suspected cancer tissue. Some cancers are diagnosed during routine screening examinations. Many cancers are discovered when a patient presents specific symptoms to their health care professional.

'Turbocharged artificial intelligence' could personalize combination therapy in pediatric leukemia


A team of UCLA bioengineers has demonstrated that its technology may go a long way toward overcoming the challenges of treatment for acute lymphoblastic leukemia, among the most common types of cancer in children, and has the potential to help doctors personalize drug doses. The five-year survival rate for individuals with pediatric acute lymphoblastic leukemia is about 85 percent, however those who experience a recurrence generally have a poor prognosis and a bone marrow transplant is their only option for a permanent cure. Conventional treatment for this leukemia includes a combination of drugs, which come with short- and long-term side effects. Two of these drugs, 6-mercaptopurine and methotrexate, can cause liver disease and other life-threatening infections. During the maintenance phase of treatment, which aims to keep individuals in remission, dosing for these two drugs is frequently adjusted through a system of trial and error, which is not always accurate.

Application of Deep Learning on Predicting Prognosis of Acute Myeloid Leukemia with Cytogenetics, Age, and Mutations Machine Learning

We explore how Deep Learning (DL) can be utilized to predict prognosis of acute myeloid leukemia (AML). Out of TCGA (The Cancer Genome Atlas) database, 94 AML cases are used in this study. Input data include age, 10 common cytogenetic and 23 most common mutation results; output is the prognosis (diagnosis to death, DTD). In our DL network, autoencoders are stacked to form a hierarchical DL model from which raw data are compressed and organized and high-level features are extracted. The network is written in R language and is designed to predict prognosis of AML for a given case (DTD of more than or less than 730 days). The DL network achieves an excellent accuracy of 83% in predicting prognosis. As a proof-of-concept study, our preliminary results demonstrate a practical application of DL in future practice of prognostic prediction using next-gen sequencing (NGS) data.

A death knell for relapsed leukemia?


Almost invariably, however, the disease returns and is often fatal. Relapse has been attributed to the expansion of preexisting leukemic clones that are resistant to therapy. In a preclinical study, Pan et al. investigated whether better efficacy might be achieved by using a class of drugs that work by inducing apoptotic cell death. They found that mice with drug-resistant AML showed dramatically extended survival after treatment with a combination of two drugs that promote apoptosis by distinct mechanisms. This combination therapy of venetoclax (a Bcl-2 inhibitor) and idasanutlin (a p53 activator) is now in clinical trials for relapsed AML.