Study findings could lead to new clinical laboratory diagnostics that give pathologists a more detailed understanding about certain types of cancer
New studies proving artificial intelligence (AI) can be used effectively in clinical laboratory diagnostics and personalized healthcare continue to emerge. Scientists in the UK recently trained an AI model using machine learning and deep learning to enable earlier, more accurate detection of 13 different types of cancer.
DNA stores genetic information in sequences of four nucleotide bases: A (adenine), T (thymine), G (guanine) and C (cytosine). These bases can be modified through DNA methylation. There are millions of DNA methylation markers in every single cell, and they change in the early stages of cancer development.
One common characteristic of many cancers is an epigenetic phenomenon called aberrant DNA methylation. Modifications in DNA can influence gene expression and are observable in cancer cells. A methylation profile can differentiate tumor types and subtypes and changes in the process often come before malignancy appears. This renders methylation very useful in catching cancers while in the early stages.
However, deciphering slight changes in methylation patterns can be extremely difficult. According to the scientists, “identifying the specific DNA methylation signatures indicative of different cancer types is akin to searching for a needle in a haystack.”
Nevertheless, the researchers believe identifying these changes could become a useful biomarker for early detection of cancers, which is why they built their AI models.
“Computational methods such as this model, through better training on more varied data and rigorous testing in the clinic, will eventually provide AI models that can help doctors with early detection and screening of cancers,” said Shamith Samarajiwa, PhD (above), Senior Lecturer and Group Leader, Computational Biology and Genomic Data Science, Imperial College London, in a news release. “This will provide better patient outcomes.” With additional research, clinical laboratories and pathologists may soon have new cancer diagnostics based on these AI models. (Photo copyright: University of Cambridge.)
The researchers then used a combination of machine learning and deep learning techniques to train an AI algorithm to examine DNA methylation patterns of the collected data. The algorithm identified and differentiated specific cancer types, including breast, liver, lung and prostate, from non-cancerous tissue with a 98.2% accuracy rate. The team evaluated their AI model by comparing the results to independent research.
In their Biology Methods and Protocols paper, the authors noted that their model does require further training and testing and stressed that “the important aspect of this study was the use of an explainable and interpretable core AI model.” They also claim their model could help medical professionals understand “the underlying mechanisms that contribute to the development of cancer.”
Using AI to Lower Cancer Rates Worldwide
According to the Centers for Disease Control and Prevention (CDC), cancer ranks as the second leading cause of death in the United States with 608,371 deaths reported in 2022. The leading cause of death in the US is heart disease with 702,880 deaths reported in the same year.
Globally cancer diagnoses and death rates are even more alarming. World Health Organization (WHO) data shows an estimated 20 million new cancer cases worldwide in 2022, with 9.7 million persons perishing from various cancers that year.
The UK researchers are hopeful their new AI model will help lower those numbers. They state in their paper that “most cancers are treatable and curable if detected early enough.”
More research and studies are needed to confirm the results of this study, but it appears to be a very promising line of exploration and development of using AI to detect, identify, and diagnose cancer earlier. This type of probing could provide pathologists with improved tools for determining the presence of cancer and lead to better patient outcomes.
As this therapeutic approach gains regulatory approval, clinical laboratory tests to determine condition of patient’s gut microbiota and monitor therapy will be needed
Some developments in the clinical laboratory industry are less about diagnostic tests and more about novel approaches to therapy. Such is the case with a new carbon bead technology developed by researchers from University College London (UCL) and the Royal Free Hospital intended to remove harmful bacteria toxins from the gut before they leak to the liver. The macroporous beads, which come in small pouches, are delivered orally and could be utilized in the future to treat a number of diseases.
Why is this relevant? Once a new treatment is accepted for clinical use, demand increases for a clinical laboratory test that confirms the therapy will likely work and to monitor its progress.
In collaboration with Yaqrit, a UK-based life sciences company that develops treatments for chronic liver disease, the UCL and Royal Free Hospital scientists engineered the carbon beads—known as CARBALIVE—to help restore gut health. They measured the technology’s impact on liver, kidney, and brain function in both rats and mice.
“The influence of the gut microbiome on health is only just beginning to be fully appreciated,” said Rajiv Jalan, PhD, Professor of Hepatology at UCL in a press release. “When the balance of the microbiome is upset, ‘bad’ bacteria can proliferate and out-compete the ‘good’ bacteria that keeps the gut healthy.
“One of the ways [the ‘bad’ bacteria] do this is by excreting endotoxin, toxic metabolites, and cytokines that transform the gut environment to make it more favorable to them and hostile to good bacteria,” he continued. “These substances, particularly endotoxin, can trigger gut inflammation and increase the leakiness of the gut wall, resulting in damage to other organs such as the liver, kidneys, and brain.”
“I have high hopes that the positive impact of these carbon beads in animal models will be seen in humans, which is exciting not just for the treatment of liver disease but potentially any health condition that is caused or exacerbated by a gut microbiome that doesn’t work as it should,” said Rajiv Jalan, PhD (above), Professor of Hepatology, University College London, in a press release. “This might include conditions such as irritable bowel syndrome (IBS), for example, which is on the rise in many countries.” Though not a clinical laboratory diagnostic test, new therapies like CARBALIVE could be a boon to physicians treating patients with IBS and other gastrointestinal conditions.
Developing the Carbon Beads
The team discovered CARBALIVE is effective in the prevention of liver scarring and injury in animals with cirrhosis when ingested daily for several weeks. They also found a reduced mortality rate in test animals with acute-on-chronic-liver-failure (ACLF).
After achieving success with CARBALIVE in animals, the researchers tested the technology on 28 cirrhosis patients. The carbon beads proved to be safe for humans and had inconsequential side effects.
“In cirrhosis, a condition characterized by scarring of the liver, it is known that inflammation caused by endotoxins can exacerbate liver damage,” Jalan explained. “Part of the standard treatment for cirrhosis is antibiotics aimed at controlling bad bacteria, but this comes with the risk of antibiotic resistance and is only used in late-stage disease.”
The beads, which are smaller than a grain of salt, contain an exclusive physical structure that absorbs large and small molecules in the gut. They are intended to be taken with water at bedtime as harmful bacteria is more likely to circulate through the body at night which could result in damage. The carbon beads do not kill bacteria, which decreases the risk of antibiotic resistance. They eventually pass through the body as waste.
“They work by absorbing the endotoxins and other metabolites produced by ‘bad’ bacteria in the gut, creating a better environment for the good bacteria to flourish and helping to restore microbiome health,” said Michal Kowalski, M.Sc.Eng, Director and VP of Operations at Yaqrit, in the UCL news release.
“This prevents these toxins from leaching into other areas of the body and causing damage, as they do in cirrhosis,” he added. “The results in animal models are very positive, with reduction in gut permeability, liver injury, as well as brain and kidney dysfunction.”
Additional Research
The researchers plan to perform further clinical trials in humans to determine if the carbon beads are effective at slowing the progression of liver disease. If the benefits that were observed in lab animals prove to be compelling in humans, the technology may become an invaluable tool for the treatment of liver disease and other diseases associated with poor microbiome health in the future.
According to the American Liver Foundation, 4.5 million adults in the US have been diagnosed with liver disease. However, it is estimated that 80 to 100 million adults have some form of fatty liver disease and are unaware of it. Liver disease was the 12th leading cause of death in the US in 2020 with 51,642 adults perishing from the disease that year.
According to BMC Public Health, globally there were 2.05 million new cases of liver cirrhosis diagnosed in 2019. In that year, 1.47 million people around the world died from the disease.
More research and clinical studies are needed before this novel technology can be used clinically. When and if that happens, the demand for clinical laboratory tests that measure microbiome deficiencies and monitor patient progress during therapy will likely be high.
As new diagnostic assays are cleared by regulators, clinical laboratories will play a key role in identifying appropriate patients for new less-invasive Alzheimer’s tests
With multiple companies racing to develop a blood-based test for Alzheimer’s disease (AD), clinical laboratories may soon have new less-invasive diagnostic assays for AD on their menus.
Why a race? Because a less-invasive clinical laboratory test that uses a venous blood draw (as opposed to a spinal tap)—and which has increased sensitivity/specificity—has a potentially large market given the substantial numbers of elderly predicted to develop Alzheimer’s over the next decade. It has the potential to be a high volume, high dollar diagnostic test.
In fact, Mordor Intelligence estimates that the market for Alzheimer’s disease therapeutics will grow from $7.7 billion in 2024 to $10.10 billion by 2029.
Alzheimers.gov, an official website of the US government, says, “Researchers have made significant progress in developing, testing, and validating biomarkers that detect signs of the disease process. For example, in addition to PET scans that detect abnormal beta-amyloid plaques and tau tangles [abnormal forms of tau protein] in the brain, NIH-supported scientists have developed the first commercial blood test for Alzheimer’s. This test and others in development can not only help support diagnosis but also be used to screen volunteers for research studies.”
Additionally, the US Food and Drug Administration (FDA) is clearing new Alzheimer’s drugs for clinical use. The pharma companies behind these drugs need clinical laboratory tests that accurately diagnosis the disease and confirm that it would be appropriate for the patient to receive the new therapeutic drugs, a key element of precision medicine.
“The big promise for blood tests is that they will eventually be accessible, hopefully, cost-effective, and noninvasive,” Rebecca Edelmayer, PhD (above), Vice President, Scientific Engagement, Alzheimer’s Association, told USA Today. “The field is really moving forward with use of these types of tests,” she added. Clinical laboratories may soon have these new assays on their test menus. (Photo copyright: Alzheimer’s Association.)
Companies in the Race to Develop Blood-based Alzheimer’s Tests
Researchers found that C2N’s blood test can detect brain amyloid status with “sensitivity, specificity, positive and negative predictive values that approximate those of amyloid positron emission tomography (PET) imaging,” according to a news release.
“The PrecivityAD2 blood test is intended for use in patients aged 55 and older with signs or symptoms of mild cognitive impairment or dementia who are undergoing evaluation of Alzheimer’s disease or dementia. Only a healthcare provider can order the PrecivityAD2 test,” the news release noted.
“The PrecivityAD2 blood test showed strong clinical validity with excellent agreement with brain amyloidosis by PET,” the researchers wrote.
The PrecivityAD2 test, which is mailed directly by C2N to doctors and researchers, is performed at the company’s CLIA-certified lab, according to USA Today, which added that the cost of $1,450 is generally not covered by insurance plans.
Expanding Test Access with IVD Companies
ALZpath, Inc. has a different approach to the Alzheimer’s disease test market. The Carlsbad, Calif.-based company, set up an agreement with in vitro diagnostics (IVD) company Roche Diagnostics for use of its phosphorylated tau (pTau)217 antibody “to develop and commercialize an Alzheimer’s disease diagnostic blood test that will be offered on the Roche Elecsys platform,” according to a news release.
Roche received FDA breakthrough device designation on the Elecsys pTau217 test earlier this year and will work with pharmaceutical company Eli Lilly to commercialize the test.
Estimates show 75% of dementia cases go undetected—a number which could grow to 140 million by 2050, according to data shared by Roche with Fierce Biotech.
“We plan to leverage our installed base of diagnostic systems, which is the largest in the world, to ensure we are able to create access to this test for those who need it the most,” Matt Sause, CEO, Roche Diagnostics, told Fierce Biotech.
Another IVD company, Beckman Coulter, recently signed an agreement to use ALZpath’s pTau217 antibody test in its DxI 9000 Immunoassay Analyzer. In a news release, Kathleen Orland, SVP and General Manager of the Clinical Chemistry Immunoassay Business Unit at Beckman Coulter, said that the test had “high performance in detecting amyloid pathology” and could “integrate into our advanced DxI 9000 platform to support broad-based testing.”
Clinical Laboratory Participation
The FDA is drafting new guidance titled, “Early Alzheimer’s Disease: Developing Drugs for Treatment” that is “intended to assist sponsors in the clinical development of drugs for the treatment of the stages of sporadic Alzheimer’s disease (AD) that occur before the onset of overt dementia.”
Pharma companies intent on launching new drugs for Alzheimer’s will need medical laboratory tests that accurately diagnosis the disease to confirm the medications would be appropriate for specific patients.
Given development of the aforementioned pTau217 antibody tests, and others featuring different diagnostic technologies, it’s likely clinical laboratories will soon be performing new assays for diagnosing Alzheimer’s disease.
Researchers have been exploring the role metabolites play in the development of disease for some time. Alzheimer’s is a progressive, degenerative brain disease typically linked to age, family history, and deposits of certain proteins in the brain that cause the brain to shrink and brain cells to eventually die. Alzheimer’s is the most common form of dementia, accounting for an estimated 60% to 80% of all dementia cases. It has no cure or proven method of prevention, according to the Alzheimer’s Association.
There are nearly seven million people living with Alzheimer’s in the US and 55 million people worldwide live with it or other forms of dementia. Patients are usually over the age of 65, but it can present in younger patients as well.
“Gut metabolites are the key to many physiological processes in our bodies, and for every key there is a lock for human health and disease,” said Feixiong Cheng, PhD (above), founding director of the Cleveland Clinic Genome Center, in a news release. “The problem is that we have tens of thousands of receptors and thousands of metabolites in our system, so manually figuring out which key goes into which lock has been slow and costly. That’s why we decided to use AI.” Findings from the study could lead to new clinical laboratory biomarkers for dementia screening tests. (Photo copyright: Cleveland Clinic Lerner Research Institute.)
Changes to Gut Bacteria
Metabolites are substances released by bacteria when the body breaks down food, drugs, chemicals, or its own tissue, such as fat or muscle. They fuel cellular processes within the body that may be either helpful or harmful to an individual’s health.
The Cleveland Clinic researchers believe that preventing detrimental interactions between metabolites and cells could aid in disease prevention. Previous studies have shown that Alzheimer’s patients do experience changes in their gut bacteria as the disease progresses.
To complete their study, the scientists used AI and machine learning (ML) to analyze more than 1.09 million potential metabolite-receptor pairs to determine the likelihood of developing Alzheimer’s.
They then examined genetic and proteomic data from Alzheimer’s disease studies and looked at different receptor protein structures and metabolite shapes to determine how different metabolites can affect brain cells. The researchers identified significant interactions between the gut and the brain.
They discovered that the metabolite agmatine was most likely to interact with a receptor known as CA3R in Alzheimer’s patients. Agmatine is believed to protect brain cells from inflammation and damage. They found that when Alzheimer’s-affected neurons are treated with agmatine, CA3R levels reduce. Levels of phosphorylated tau proteins, a biomarker for Alzheimer’s disease, lowered as well.
The researchers also studied a metabolite called phenethylamine. They found that it too could significantly alter the levels of phosphorylated tau proteins, a result they believe could be beneficial to Alzheimer’s patients.
New Therapies for Alzheimer’s, Other Diseases
One of the most compelling results observed in the study was the identification of specific G-protein-coupled receptors (GPCRs) that interact with metabolites present in the gut microbiome. By focusing on orphan GPCRs, the researchers determined that certain metabolites could activate those receptors, which could help generate new therapies for Alzheimer’s.
“We specifically focused on Alzheimer’s disease, but metabolite-receptor interactions play a role in almost every disease that involves gut microbes,” said Feixiong Cheng, PhD, founding director of the Cleveland Clinic Genome Center in the news release. “We hope that our methods can provide a framework to progress the entire field of metabolite-associated diseases and human health.”
The team plans to use AI technology to further develop and study the interactions between genetic and environmental factors on human health and disease progression. More research and studies are needed, but results of the Cleveland Clinic study suggest new biomarkers for targeted therapies and clinical laboratory tests for dementia diseases may soon follow.
Proof-of-concept study may eventually lead to new clinical laboratory urine tests for fast, non-invasive detection of cancer
Here is the latest example of researchers finding useful biomarkers in urine for diagnosing certain cancers. The discovery comes from the University of Michigan Health Rogel Cancer Center, where, in a proof-of-concept study, scientists developed a urine-based test that screens for circulating free DNA (cfDNA) fragments (aka, cell-free DNA) released by tumors in the head and neck. If they confirm these findings, it’s possible the technology could be adapted into a non-invasive clinical laboratory test for selected cancers.
One such cancer is human papillomavirus (HPV) which, though “widely recognized for causing cervical cancer” is “increasingly found to cause cancers in the mouth, throat, and other head and neck regions,” according to a U-M Medical School press release.
The U-M study findings could lead to an early, non-invasive test for the detection of cancer, as compared to traditional urine or blood-based liquid biopsy testing.
“In this study, we provide evidence to support the hypothesis that conventional assays do not detect ultrashort fragments found in urine since they are designed to support longer DNA fragments. Our team used an unconventional approach to develop a urine test for HPV-positive head and neck cancer ctDNA detection,” said Chandan Bhambhani, PhD (above), Research Lab Specialist Intermediate at University of Michigan and co-first author of the study, in a news release. Clinical laboratories may soon have a new urine-based test for detecting cancer. (Photo copyright: LinkedIn.)
According to the researchers, benefits of urine testing include:
Testing with urine is convenient for people who may be unable to access healthcare and phlebotomy services.
Urine has low biohazard risk and may be easily collected in large amounts, compared with blood.
Ongoing collection of urine could make way for TR-ctDNA “kinetics to be used as a high time-resolution biomarker” to monitor patients’ response to treatment.
However, urine, the researchers cautioned, must be analyzed in a different manner if it is to be comparable in efficiency to blood-based ctDNA testing.
“There have been mixed reports on the efficiency of TR-ctDNA detection compared with that of blood ctDNA. A potentially crucial factor for the analysis of TR-ctDNA is knowing the length of TR-ctDNA fragments present in urine, because this affects assay design for optimal sensitivity in TR-ctDNA detection,” the researchers explained.
New Assay Detects Ultrashort DNA Fragments
To complete their study, the U-M researchers developed an ultrashort HPV droplet digital PCR (polymerase chain reaction) assay that enabled detection of TR-ctDNA from HPV-associated oropharyngeal squamous cell carcinoma (HPV OPSCC), BioTechniques reported.
The assay was made to target the HPV16 E6 (Human papillomavirus 16) gene and to measure TR-ctDNA in patients with HPV OPSCC, the JCI Insight paper noted.
“The HPV16 E6 gene represents a highly recurrent ctDNA target in the population of patients with HPV OPSCC,” the researchers wrote in JCI Insight, adding:
Targeting ultrashort fragments was essential “for robust TR-ctDNA detection.”
Results in urine with patients with HPV OPSCC was consistent with results from plasma ctDNA.
The test, still in the discovery phase, was mailed to patients who were being treated for the disease and who reside within 100 miles of Ann Arbor, Mich. They returned urine samples for testing at the U-M lab and to get insights into possible post-treatment needs.
“Using longitudinal urine samples from a small case series, we showed proof of concept for early detection of cancer recurrence. Thus, our results indicate that by targeting ultrashort DNA fragments, TR-ctDNA becomes a viable approach for HPV OPSCC detection and potentially for cancer recurrence monitoring after treatment,” the authors wrote.
Further Studies, Possible Test Expansion
HPV infection—and especially HPV type 16—is a growing risk factor for oropharyngeal cancers, according to the National Cancer Institute.
The U-M Rogel Cancer Center scientists plan more studies to leverage the information urine may carry about an individual’s health. The researchers intend to expand the scope of their new test to other cancers including breast cancer and acute myeloid leukemia.
“The test that has been developed has detected cancer far earlier than would typically happen based on clinical imaging. As such, these promising results have given us the confidence to broaden the scope of this study, seeking to expanding distribution even further,” said J. Chad Brenner, PhD, Associate Professor of Otolaryngology-Head and Neck Surgery, U-M Medicine, and co-senior author of the study, in the news release.
The University of Michigan Health study exemplifies scientists’ commitment to new categories of biomarkers that can be used for medical laboratory tests and prescription drugs. And by focusing on urine, the researchers made it possible for patients to collect specimens themselves and send them to the medical laboratory for analysis and reporting.
Further research could eventually lead to clinical laboratory biomarkers and screening tests to identify infants whose gut bacteria may predispose them to neurodevelopment disorders later in life
Microbiologists and clinical laboratory scientists working with gut bacteria will be intrigued to learn that a study conducted by scientists from Linköping University in Sweden and the Department of Microbiology and Cell Science at the University of Florida (UFL) recently found that gut microbiota (aka, gut flora) in infancy can be correlated with developing a neurodevelopmental disorder (ND) later in life.
The researchers analyzed patient records from the 20-year All Babies in Southeast Sweden (ABIS) prospective cohort study into the etiology of obesity, diabetes, and other diseases. They found that “disturbances” in the microbiomes of children during the first years of life could be linked to later ND diagnoses, according to Neuroscience News.
“We’ve found associations with some factors that affect gut bacteria, such as antibiotic treatment during the child’s first year, which is linked to an increased risk of these diseases,” stated pediatrician Johnny Ludvigsson, MD, PhD, Senior Professor, Department of Biomedical and Clinical Sciences at Linköping University, who co-led the study, in a Linköping University news release.
“Analyzing over 16,000 children from the ABIS study, researchers identified significant biomarkers in cord blood and stool samples that correlate with future diagnoses of these disorders,” Neuroscience News reported.
This study adds evidence to the growing theory that every individual’s microbiome has much to do with his/her state of health and specific health conditions.
“We can see in the study that there are clear differences in the intestinal flora already during the first year of life between those who develop autism or ADHD and those who don’t,” said pediatrician and study co-author Johnny Ludvigsson, MD, PhD (above), Senior Professor, Department of Biomedical and Clinical Sciences at Linköping University, in a news release. Clinical laboratory scientists and microbiologists who work with gut microbiota will find these observations intriguing. (Photo copyright: Linköping University.)
Analysis of the ABIS Study
To conduct their study, the researchers analyzed the health records of 16,440 children born between 1997 and 1999 who participated in the ABIS study. The subjects were a close representation of the general Swedish population and were followed from birth into their twenties.
Research showed that 1,197 of the 16,440 children (7.28%) had been diagnosed with either autism, ADHD, communication disorders, or an intellectual disability.
The researchers also surveyed the ABIS study participants concerning their lifestyles and environmental factors during childhood. They analyzed substances found in the umbilical cord blood and stool bacteria collected at age one in some of the study participants. Cord blood remains in the placenta and umbilical cord after birth and is rich in stem cells.
“The remarkable aspect of the work is that these biomarkers are found at birth in cord blood or in the child’s stool at one year of age over a decade prior to the diagnosis,” said Eric Triplett, PhD, Professor and Chair of the Department of Microbiology and Cell Science at UFL and a co-leader of the study, in the Linköping University news release.
The investigation found that children who had numerous ear infections during the first year of life were more prone to receiving a diagnosis of a neurodevelopmental disorder later in life. The scientists surmised that it was not the infections that caused the issues. Rather, it was that repeated antibiotic treatments had disturbed the balance of healthy gut bacteria.
“We’re not trying to say that antibiotics are necessarily a bad thing,” stated Angelica Ahrens, PhD, Assistant Research Scientist in the Triplett Research Group at the University of Florida and first author of the study, in a UFL blog. “But perhaps overuse can be detrimental to the microbiome, and for some children, for whatever reason, their microbiome might not recover as readily.”
Deficits in Important Bacteria
The researchers discovered that the presence of Citrobacter bacteria increased the risk of a future ND diagnosis. According to ScienceDirect, “organisms of the genus Citrobacter are gram-negative bacilli that are occasional inhabitants of the gastrointestinal tract and are responsible for disease in neonates [newborns that are four weeks or younger] and debilitated or immunocompromised patients.”
They also discovered that the absence of Coprococcus bacteria increased the risk of getting an ND as well. One of the main producers of butyrate, Coprococcus is known to support gut barrier function, suppress harmful bacteria, and contain anti-inflammatory properties.
“Coprococcus and Akkermansia muciniphila have potential protective effects,” said Ahrens in the Linköping University news release. “These bacteria were correlated with important substances in the stool, such as vitamin B and precursors to neurotransmitters which play vital roles orchestrating signaling in the brain. Overall, we saw deficits in these bacteria in children who later received a developmental neurological diagnosis.”
Environmental/Behavioral Findings of the ABIS Study
Through the analysis of toxins present in study participants’ cord blood, the researchers confirmed that risk of developing an ND increases when babies are exposed to parents who smoke. The scientists also found that breastfeeding offers a protective effect against NDs.
More research is needed to determine whether gut flora in infants can have an effect on developing NDs later in life, and it is not yet known if similar findings will be detected in other populations. Nevertheless, the findings that many biomarkers for NDs can be observed in infancy may enable scientists to create clinical laboratory screening protocols, preventative measures, and innovative treatments for neurodevelopmental disorders.
Further research and studies linking certain microbiome factors to specific health conditions will create opportunities for microbiologists and clinical laboratories as well, to perform diagnostic testing that identifies if a patient—in this case a newborn or infant—has a microbiome that will lead to immediate or later neurological health conditions.
