The first AI-designed drugs have ended with disappointment.

Over the last year-plus, the first handful of molecules created by artificial intelligence have failed trials or been deprioritized. The AI companies behind these drugs brought them into the clinic full of fanfare about a new age of drug discovery — and have quietly shelved them after learning old lessons about how hard pharmaceutical R&D can be.

Earlier this month, UK-based Exscientia slipped into a pipeline update that a Phase I/II study of its cancer drug candidate EXS-21546 was winding down. That cut followed a decision last year by its partner Sumitomo Pharma to abandon another of its AI-designed drugs. In April, a test of BenevolentAI’s dermatitis drug fell short as well. And Recursion Pharmaceuticals — the third of AI’s early generation — hasn’t recorded a trial failure but has had a handful of clinical setbacks that don’t necessarily bode well.

There’s no shortage of AI naysayers, and the 0-for-3 start suggests that AI hype has set unrealistic expectations. Clinical wins are rarities in biotech, where an estimated 5% or 10% of drugs that head into human testing actually get approved.

“If you take the hype and PR at face value over the last 10 years, you would think it goes from 5% to 90%,” Patrick Malone, a principal at KdT Ventures, said of AI. “But if you know how these models work, it goes from 5% to maybe 6% or 7%.”

These three companies have been at this for roughly a decade, combining to rack up an accumulated deficit of over $1.5 billion.

Ivan Griffin

The reality check of the clinic, paired with a dour biotech market, has beaten up these first-generation biotechs that went public in 2021 or 2022. Their stock prices are all down at least 75%, underperforming the biotech market, even as new AI startups have continued to raise substantial sums of money. Generate:Biomedicines, Inceptive, Iambic and Genesis, for instance, have combined to raise $673 million over the past few months.

Executives at these first-generation companies say it’s too early for a verdict on whether or not AI boosts the odds of success, particularly given the vast likelihood that any drug candidate — AI-developed or not — will fail.

“The things that are easiest to show — speed and cost, particularly on the preclinical — have been done,” Ivan Griffin, BenevolentAI’s co-founder and chief operating officer, said in an interview with Endpoints News. Increasing the probability of success “will inevitably take the longest to prove out.”

Milestone moments to quiet cancellations

Back in the final days of January 2020, as Covid-19 was emerging as a pandemic threat, Exscientia CEO Andrew Hopkins hailed a “key milestone in drug discovery.”

His biotech announced the first AI-designed drug had entered the clinic. Its partner Sumitomo led the Phase I study in obsessive-compulsive disorder, with the Financial Times calling the trial’s start a “critical milestone for the role of machine learning in medicine.”

About two years later, in January 2022, Sumitomo disclosed they had abandoned the drug, which failed to meet the study’s criteria. In an interview, Hopkins said Exscientia’s job was to just design the molecule, with Sumitomo making the clinical decisions.

Exscientia had more control over the next drug, a cancer treatment called EXS-21546, which it brought into the clinic in December 2020. Earlier this month, the biotech said it was discontinuing an ongoing Phase I/II study with modeling suggesting “it will be challenging for ‘546 to reach a suitable therapeutic index.”

Andrew Hopkins

Hopkins said that the trial didn’t fail, as the biotech doesn’t have full results back.

“It wasn’t a clinical data decision,” he said. “It was a strategic decision” to prioritize two other cancer drugs that the company believes have better chances.

“We don’t want to be one of those companies that keeps pushing a program forward because it’s the only thing they have,” Hopkins said.

Clinical failure is less contestable for BenevolentAI, a fellow UK-based AI enthusiast whose lead drug was unable to beat a placebo in a Phase IIa atopic dermatitis study, the biotech announced earlier this year. That readout led to the drug’s end, a stock plunge, and sizable layoffs.

Recursion stands apart as the only one of the three to maintain a valuation above $1 billion today. (Exscientia is worth about $650 million, while BenevolentAI is valued at $117 million.)  The biotech has had several positive Phase I readouts centered on safety and tolerability, such as a C. diff drug recently clearing a healthy volunteer study of 42 people. The company used AI to identify existing compounds rather than design new drugs for its early pipeline.

Dylan Reid

“The market has never known what to make of these companies,” said Dylan Reid, a partner at Zetta Venture Partners. “They’ve been way too excited and way too down. At one point, they value the platform at X billions of dollars, and today, it’s probably a drag on valuation.”

While Recursion hasn’t had a clinical failure, its development plans have had hiccups. It quietly dropped a rare disease drug last year, citing “noise in the potency” and delays in getting a trial going. Earlier this month, the Salt Lake City-based biotech slimmed down an ongoing Phase II study for another of its drugs, dropping a placebo arm and cutting expected enrollment to 37 people. A spokesperson said the changes will help get to data and Phase III faster.

If a decade sounds too soon to judge these biotechs, consider Recursion’s leaders set that timeline themselves, publicly declaring the goal of discovering 100 clinical-stage drug candidates in its first 10 years. Roughly 10 years on, Recursion’s pipeline has four clinical-stage molecules.

Recursion expects two Phase II readouts in the second half of 2024. All three biotechs have ongoing partnerships with drugmaking giants like Merck, Bristol Myers Squibb, and Roche’s Genentech.

The AI clinical pipeline is full of other players as well, such as Verge Genomics’ ALS drug, BPGbio’s brain cancer treatment, Insilico Medicine’s idiopathic pulmonary fibrosis drug, Generate:Biomedicines’ Covid-19 antibody, and Auransa’s liver cancer therapy.

Strategies evolve as more players emerge

AI backers say successful programs like Moderna’s Covid-19 vaccine or Nimbus Therapeutics’ TYK2 inhibitor used AI to a degree. But those drugmakers don’t brand those medicines as AI-designed, while Exscientia, BenevolentAI, and Recursion market their approach as AI-driven or AI-enabled.

A decade in, leaders of the first-generation companies say they are still learning.

BenevolentAI, for instance, says its failed atopic dermatitis drug candidate didn’t use the company’s target identification approach, which is behind its ulcerative colitis drug that entered the clinic earlier this year.

Exscientia has incorporated more human tissue samples in its research process and hired experienced clinical hands like Michael Krams, Hopkins said.

“We’ve also now realized if we want to change the probability of success in the clinic, it’s not just better molecules,” Hopkins said. “We also need better translational models.”

Read More

Researchers at the University of Cincinnati have developed a lateral flow assay that can detect bacterial toxins from Porphyromonas gingivalis, the causative bacteria for gingivitis. The technology could make it easier and faster to identify early-stage gingivitis, which can lead to periodontitis and eventual tooth loss, as well as contributing to a variety of other diseases such as stroke and heart disease. The lateral flow assay requires a small saliva sample, and can provide results very quickly, but does require the saliva sample to be pre-treated with potato starch to deactivate salivary amylase, an enzyme that can interfere with the assay.

The humble lateral flow assay grew in prominence during the COVID-19 pandemic as a quick at-home method to check your COVID status, but this technology was already a staple of such applications as pregnancy testing. Now, researchers are increasingly aware of its utility as a rapid point-of-care diagnostic technology and are beginning to apply it to the detection of other diseases. In this instance, these researchers at the University of Cincinnati have developed a lateral flow assay to detect the bacteria responsible for gingivitis.

Gingivitis is caused by P. gingivalis, which typically starts as mild gum inflammation. However, this can spread to other parts of the periodontal tissue, causing damage to soft tissue and bone that stabilize our teeth. This damage can eventually lead to tooth loss. Moreover, researchers have also linked P. gingivalis to other conditions, including cardiovascular diseases, rheumatoid arthritis, and even neurodegenerative diseases such as Alzheimer’s disease.  

There are lab-based tests available to detect P. gingivalis, but compared with a lateral flow test, they are complex, slow, expensive, and lack portability. If a diagnostic technique is too expensive, time consuming and inconvenient, then patients or clinicians will only tend to seek it out or recommend it if symptoms have already developed. However, for routine testing and health screening, a convenient, rapid, and point-of-care test is much preferred. A lateral flow test for gingivitis, for example, could be administered by a dentist every time someone undergoes a routine dental checkup.   

The assay detects a bacterial endotoxin released into the saliva by P. gingivalis through a simple immunoassay, whereby antibodies capture and identify the toxin. An enzyme present in saliva called amylase can interfere with this, so the assay requires the saliva to be pretreated with potato starch to deactivate this enzyme. In the future, you may be able to use such lateral flow assays to conveniently detect a wide variety of pathogens and biomarkers, and you can thank SARS-CoV-2 for the privilege.  

Study in journal Sensors & Diagnostics: Salivary endotoxin detection using combined mono/polyclonal antibody-based sandwich-type lateral flow immunoassay device

Via: University of Cincinnati

Read More

When Daniel Judge, MD, director of the Cardiovascular Genetics program and fellowship director for cardiovascular disease at The Medical University of South Carolina (MUSC), moved over from John Hopkins a bit over five years ago, the use of genetics in clinical practice was absent in South Carolina.

“When you ask a clinician what genetics means, it’s often infants with phenylketonuria (PKU) or inherited disorders of metabolism,” Judge told Inside Precision Medicine. “But for adults with cardiovascular, pulmonary, or renal disease, not everyone uses genetics routinely in practice.”

So when the opportunity arose to work with population genomics company Helix on a large genomic screening program, Judge was incredibly enthusiastic.

“Here’s an opportunity for us and MUSC to stand apart from the large for-profit hospital centers that operate around the state, and for us to provide an academic and improved healthcare approach to things,” said Judge. “This was the opportunity for clinicians who have been doing good clinical work to bring genetics into their practice.”

In 2021, MUSC launched “In Our DNA, SC,” a first-of-its-kind genomics program to drive preventive, precision health care for South Carolinians. The statewide initiative has set out to enroll 100,000 patients in genetic testing over the next four years at no cost to the patients.

Walking the walk

Excluding COVID-19, the top cause of death in South Carolina is heart disease, followed by cancer. In the time that In Our DNA SC has been up and running, the cancer and heart disease rates haven’t yet begun to go down, but that’s the goal. Judge estimates that the program has detected about 175 positive results, which are split across breast cancer, colon cancer, and hyperlipidemia.

“Many of those patients with those results were quite surprised because they don’t have a family history of cancer or heart disease,” said Judge. “It’s nice to diagnose it because we’re seeing the ability to prescribe medications on the clinical side that are available more readily for people with familial hyperlipidemia or hypercholesterolemia.”

Judge is excited about the opportunity to provide an academic and improved healthcare approach. “It’s an opportunity for clinicians who have been doing great clinical work but haven’t integrated genetics into their primary care, or whatever that practice might be,” he said. “I think the use of genetics in clinical practice…is entering that mainstream. The patient benefits, and the family benefits.”

Judge’s experience in clinical genetics started about twenty years ago, and “I’ve always seen this pattern: once people see a successful use of something like this, they want to do more of it,” he said. “I think our clinicians will witness success with their patients and will want to see more.”

Making genomics part of real-time care

Behind the scenes of the In Our DNA SC initiative is a population genomics company called Helix that strives to accelerate the integration of genomic data into patient care and public health decision-making. Founder James Lu, MD, PhD, has dreamt of a world where genetic testing will provide a real-time response versus one that takes weeks. He set his sights on offering top-tier provider and patient experiences by making genomics a part of the healthcare fabric.

Lu thinks the best way to deliver genomics as a data stream or operating system within healthcare is to work with health systems. “Most of our partnerships are with large-scale health systems and are focused on large-scale, population-level programs,” said Lu. “Typically, a hundred thousand people plus, where they inevitably believe that this is going to become the standard of care over time.”

Lu believes that health systems will want to do this because it allows them to identify patients who are clearly at risk and are hiding in plain sight, such as carriers of well-established deleterious DNA variants. Helix went the route of whole exome sequencing (WES) rather than looking at an entire genome, as Lu believes that the coding regions contain almost all the relevant genomic data for clinical decision-making.

With its genomics data, Helix is currently narrowly focused on the diseases that are part of the CDC Tier 1 genomic application toolkit. This set addresses the nearly two million people in the United States who are at increased risk for adverse health outcomes because they have genetic mutations that predispose them to one of three conditions: hereditary breast and ovarian cancer syndrome (HBOC), Lynch syndrome (LS), or familial hypercholesterolemia (FH). The healthcare system currently has a poor understanding of these conditions, and many affected people and their families are unaware that they are at risk; however, early detection and intervention could significantly lower morbidity and mortality.

The type of work that Helix is doing enables health systems to create fresh population health and value-based care algorithms to manage the health of larger populations. Over time, health systems want to drive the cost of care down while improving the quality of patient experiences.

From sea to shining sea

In addition to In Our DNA, SC, Helix currently has five other major programs running across the country that represent over 100,000 people in total, representing potentially one of the largest-scale programs across America’s health systems.

In addition to South Carolina, Helix is located in Minnesota, where it is working with HealthPartners on a program called “myGenetics.” This large-scale community health research program, which launched in May 2022, is a first of its kind in Minnesota. Implementing this requires knowing what information to analyze, interpret, and communicate to patients, which Leslie Dockan, VP of primary care, clinic operations, and laboratories at HealthPartners, said aligns nicely with its and Helix’s core principle of providing clinicians with clear decision support.

The myGenetics program is free to the patient, given that it is a research project and the goal is to further biomedical understanding. “We wanted to create workflows that were easy for patients and weren’t disruptive to patient visits in the clinic, because our primary care clinicians have so many responsibilities and so many things that are happening,” said Dockan. This required HealthPartners to work closely with its electronic medical group and Helix to create a seamless workflow.

A day in the myGenetics life

After signing up and consenting electronically, the patient gets an automatic email to schedule their lab appointment at a pace that suits them. “They can take their time, ask questions, and review the information at a time when they’re comfortable, not feeling pressured to move into this,” said Dockan.

Once the appointment is scheduled, the patient gives a blood sample and receives clear information on what to expect, including how long sample processing takes and the information researchers will be looking for. From the results, HealthPartners shares information with the patient about the genes that it has screened for and what the results mean, in addition to facts about the patient’s ancestry and other genetic insights.

While this is happening, the clinical results for the CDC Tier 1 Genomic Applications Toolkit gets fed into the patient’s clinical record for any positive results. In the case of such a result, a nurse calls the patient and notifies them of the result, and offers a no obligation appointment with a genetic counselor to talk about their risk, what their results mean, and any additional testing that might be needed. Direct referrals get set up with a specialist. “If you need an oncology, cardiac, or a gastroenterology referral, we do that work for them, put the referral in, order any follow-up labs that may be needed, and set them on a clear clinical pathway,” said Dockan.

This information also goes to the primary care physician as part of the patient’s medical record, which impacts their future health maintenance, namely, how often screening occurs. “If they have a genetic variant, it doesn’t necessarily mean that they have the disease or will get the disease,” said Dockan. “So, we follow them closely and then have that as a part of their ongoing health maintenance and preventative care.”

Dockan said that genomics will be brought into everyday care, such as with pharmacogenomics. “Physicians will be able to see that there are drug-gene interactions,” said Dockan. “If your physician starts to order a drug that’s not going to be compatible with your genetic makeup and how you metabolize drugs, then we want to be able to alert your clinician at the time of order and have them be able to give you an alternative. Today, we have many people on drugs that just don’t work for them, and no one knows why.”

Outreach in every corner

As of June 2023, myGenetics has had 25,000 people consent, which is about the annual number its organizers are shooting for. “We’re starting to see positive results and have more people who are benefiting from this work in a positive way and learning things about themselves,” said Dockan. “We just identified our first early cancer—someone who was underage and not yet even at screening age came back positive for BRCA2. We ended up doing follow-up screening and an MRI, and we found cancer. She’s crediting the program with potentially saving her life.”

Dockan would like to see the next step of the program’s outreach be to everyone who’s due for their annual physical or a preventative exam. “We want to offer it with all of our mammography screenings,” said Dockan. “We have amazing screening rates for mammography, and this is just another layer that takes it even further.”

Dockan also wants to make sure that myGenetics is reaching underserved communities. She tells a story about a black woman in her fifties who has a long history of breast cancer in her family and found out that she was positive for one of the gene variants that put her at higher risk. Dockan thinks that this story can have a major influence on the communities of black women in Minnesota. Not only is there a benefit in getting the word out so that people get better immediate treatment, but the myGenetics team knows that patients of color are underrepresented in a lot of research databases and wants to help fuel new therapies and other ways of fighting disease in local underserved populations.

Judge laments that the program wasn’t in place several years earlier so that it could have worked in time for a famous South Carolina resident, Chadwick Bozeman. The actor developed metastatic colon cancer in his late thirties, well before colon cancer screening was done. Part of the plan for In Our DNA SC is to become one of the top-enrolled genomic screening programs for non-white participants. “We are in the southeast U.S., and while we are in the 15–16% range, we want to be like 30% of our participants who are non-white, predominantly black, representing our state,” said Judge. “When we look at what our goal is for inclusion in this program, we want the demographics in our publications to look like the state of South Carolina. We’re not there yet.”

Jonathan D. Grinstein’s wonder for the human mind and body led him to an undergraduate education in Neural Science and Philosophy and a doctorate in Biomedical science. He has 10 years of experience in experimental and computational research, during which he was a co-author on research articles in journals such as Nature and Cell. Since then, Jonathan hung up his lab coat and has explored positions in science writing and editing. Jonathan’s science writing work has been featured in Scientific American, Genetic Engineering and Biotechnology News (GEN), and NEO.LIFE.

The post Delivering a Real-time Genomics OS to Healthcare appeared first on Inside Precision Medicine.

Read More