English
Arabic
French
Spanish
Portuguese
Deutsch
Turkish
Dutch
Italiano
Russian
Romaian
Portuguese (Brazil)
Greek
This AI-designed drug for IBD was just given to human subjects for the first time
"We're excited to become a clinical-stage biotech company; it's exciting from an AI drug discovery standpoint," says Absci founder and CEO Sean McClain.AbsciArtificial intelligence has been working its way into the drug development process for years now, but with little to show so far in revamping the notoriously burdensome process.
While drugs are being developed using AI in a variety of ways, no drugs developed completely by AI, from start to finish, have so far made it over the finish line of regulatory approval. For that reason, every attempt by an AI drug to get approval is a landmark of sorts.
Tuesday, drug development startup Absci, based in Vancouver, Washington, announced such a landmark, the beginning of a Phase I clinical trial for a therapy it built from scratch using generative AI to treat irritable bowel disease.
The company announced it has "dosed" the first patients in Phase I trials, meaning, administering doses of its drug, ABS-101, to healthy volunteers. "This is a very big milestone for the company," said Sean McClain, founder and CEO of Absci, in a conversation with me via Google Meet Tuesday afternoon."We're excited to become a clinical-stage biotech company; it's exciting from an AI drug discovery standpoint," he said.Phase I is the first of three phases in a proposed drug's clinical trial process that must be completed in order for the drug to be considered for approval by regulators (the Food & Drug Administration in the US).
The purpose of Phase I is to prove that no adverse side effects result from the drug being put into humans for the first time.
Absci describes the process:The Phase 1 (ACTRN12625000212459p) randomized, double-blind, placebo-controlled, first-in-human study of single ascending doses of ABS-101 will evaluate safety, tolerability, pharmacokinetics (PK), and pharmacodynamics (PD) in healthy volunteers.
The study is expected to enroll approximately 40 healthy adult participants.
The primary endpoint is safety and tolerability, with PK, PD, and immunogenicity serving as secondary endpoints.
The Phase 1 interim data readout is expected in the second half of 2025. Absci has used AI to dramatically streamline the drug development and pre-clinical process, known as the "front end" of drug development, where the discovery of drugs happens, and the initial validation using in vitro and in vivo animal models, before being put into human subjects. ABS-101 was developed from scratch and brought to the clinic in just 24 months, and at a cost of $15 million."Because of AI, we got to the clinic in roughly half the time, from five years to just over 24 months," McClain told me, "and with an order-of-magnitude less cost, $15 million to get this asset into the clinic versus what typically costs $50 to $100 million.Absci's AI-driven software tools, combined with its own wet lab, are a virtual reinvention of laboratory procedures.The company uses generative AI "to predict antibodies from scratch that can bind to a target of interest," McClain said.
Traditionally, scientists in a wet lab would use an animal's immune system to generate an antibody.
With generative AI, the antibody can be created as a computer model.Absci's ABS-101 is the first drug the company has ever brought to the clinic after over a decade spent on fundamental computer work and wet lab work.
It is the company's lead drug candidate in its pipeline of drugs. The novel ABS-101 antibody, developed using generative AI, binds to the TL1A protein in immune cells whose over-expression has been linked to a variety of inflammatory autoimmune diseases. Absci AbsciNot only did AI cut time and costs, but it has brought other novel advantages, said Christian Stegmann, the company's head of drug development, on the same call."Others have brought antibodies to the clinic that have had shortcomings, which we've tried to address," he said.
A big issue has been that prior therapies "lead the patient to develop anti-drug antibodies, which can lead the patients to needing to switch treatments." The ABS-101, he said, is intended to have "reduced immunogenicity risk" by design, which will hopefully mean less drug resistance.In addition, the AI techniques allowed the company to go immediately to a "subcutaneous" method of administering the dose, rather than via a drip into the vein, as is standard in Phase I trials.
"That is unusual; it usually comes much later in clinical development settings," said Stegmann. Using a needle versus a drip is important because, ideally, the final drug will be self-administered by patients.
If the drug is already being tried out via needle rather than drip, it brings the therapy that much closer to its final form.
"That allows us to be quicker in the overall clinical development pipeline, and to gather data for the setting that is actually going to reach the market" in the drug's final form, assuming it is ultimately approved."This is an advantage of AI," McClain said, "this ability to model not just for affinity and potency, but also to optimize for the manufacturability and such -- to go to all the attributes you want in the first go-around; that really helps."The full Phase I clinical trial will extend well into next year, said McClain and Stegmann.
Gathering results is somewhat longer than for other Phase I trials because ABS-101 was designed to extend the time between doses. That is a benefit for patients as it makes less frequent dosing (less frequent needle pricks) possible, but it means the trial takes longer to carry out those dosages.
"We have a long half life we have to monitor for a while," explained Stegmann.Well before the completion of Phase I, later this year, McClain expects to have a meaningful "read-out" of initial data from the Phase I. "We are going to, in Phase I, understand important pieces [of the whole trial process], as well as confirming whether we see the extended half-life" of the dosage, said McClain.
"We will also get a look at the immunogenicity profile; there will be a lot of good information, as far as being able to show the efficacy" of ABS-101.Because of the incremental data Absci will get later this year, they will know enough to seek approval for Phase II and begin recruiting subjects before the completion of Phase I.
Phase II is where the intense work of measuring the drug's effectiveness takes place, said McClain.
"It's fair to say we will be moving faster into Phase II" than might otherwise be the case, he said. After ABS-101, McClain's next candidate approaching clinical trials is ABS-201, which has two indications of note, one for treating hair loss in the form of alopecia, and another for endometriosis.
ABS-201 is expected to enter a Phase I trial in the first half of next year, McClain said.By any measure, drug development needs an overhaul.
Creating new drugs, or even repurposing old ones, comes with an enormous cost.
A new drug takes, on average, 10 years to develop, from fundamental chemistry through clinical trials to regulatory approval.
It can cost almost $3 billion, and the failure rate of most new drug candidates is 96%. There has been a lot of activity so far, without a breakthrough AI drug. The US Food & Drug Administration's Center for Drug Evaluation and Research received over 500 drug applications through 2023 that used some sort of "AI component," according to CDER's materials on AI in drug development. But, as Nature Magazine's Melanie Senior reported in December, "No AI-enabled drug candidate has yet made it past regulators, despite several being in clinical trials."Aside from Absci, a small cohort of startups have made progress getting into trials even if they don't yet have a clinical result.
For example, BPGbio of Framingham, Mass., has a drug for pancreatic cancer, developed using AI approaches, that is working its way through Phase II clinical trials.
Beyond the results of ABS-101, and other trials, the goal of Absci is to ultimately "predict the biology." That means the company will seek to "actually start to predict where an antibody should bind to a target to give us the biological response that we want."Absci's stock is publicly traded on Nasdaq.
The shares have defied a tough stock market this year, rising 12% versus a 2% decline for the Nasdaq Composite Index.
After hours on Tuesday, as Absci issued its press release, the stock surged by as much as 25% in late trading. Featured
Source: https://www.zdnet.com/article/this-ai-designed-drug-for-ibd-was-just-given-to-human-subjects-for-the-first-time/" style="color: #0066cc;">https://www.zdnet.com/article/this-ai-designed-drug-for-ibd-was-just-given-to-human-subjects-for-the-first-time/
#this #aidesigned #drug #for #ibd #was #just #given #human #subjects #the #first #time
This AI-designed drug for IBD was just given to human subjects for the first time
"We're excited to become a clinical-stage biotech company; it's exciting from an AI drug discovery standpoint," says Absci founder and CEO Sean McClain.
AbsciArtificial intelligence has been working its way into the drug development process for years now, but with little to show so far in revamping the notoriously burdensome process.
While drugs are being developed using AI in a variety of ways, no drugs developed completely by AI, from start to finish, have so far made it over the finish line of regulatory approval. For that reason, every attempt by an AI drug to get approval is a landmark of sorts.
Tuesday, drug development startup Absci, based in Vancouver, Washington, announced such a landmark, the beginning of a Phase I clinical trial for a therapy it built from scratch using generative AI to treat irritable bowel disease.
The company announced it has "dosed" the first patients in Phase I trials, meaning, administering doses of its drug, ABS-101, to healthy volunteers. "This is a very big milestone for the company," said Sean McClain, founder and CEO of Absci, in a conversation with me via Google Meet Tuesday afternoon."We're excited to become a clinical-stage biotech company; it's exciting from an AI drug discovery standpoint," he said.Phase I is the first of three phases in a proposed drug's clinical trial process that must be completed in order for the drug to be considered for approval by regulators (the Food & Drug Administration in the US).
The purpose of Phase I is to prove that no adverse side effects result from the drug being put into humans for the first time.
Absci describes the process:The Phase 1 (ACTRN12625000212459p) randomized, double-blind, placebo-controlled, first-in-human study of single ascending doses of ABS-101 will evaluate safety, tolerability, pharmacokinetics (PK), and pharmacodynamics (PD) in healthy volunteers.
The study is expected to enroll approximately 40 healthy adult participants.
The primary endpoint is safety and tolerability, with PK, PD, and immunogenicity serving as secondary endpoints.
The Phase 1 interim data readout is expected in the second half of 2025. Absci has used AI to dramatically streamline the drug development and pre-clinical process, known as the "front end" of drug development, where the discovery of drugs happens, and the initial validation using in vitro and in vivo animal models, before being put into human subjects. ABS-101 was developed from scratch and brought to the clinic in just 24 months, and at a cost of $15 million."Because of AI, we got to the clinic in roughly half the time, from five years to just over 24 months," McClain told me, "and with an order-of-magnitude less cost, $15 million to get this asset into the clinic versus what typically costs $50 to $100 million.Absci's AI-driven software tools, combined with its own wet lab, are a virtual reinvention of laboratory procedures.The company uses generative AI "to predict antibodies from scratch that can bind to a target of interest," McClain said.
Traditionally, scientists in a wet lab would use an animal's immune system to generate an antibody.
With generative AI, the antibody can be created as a computer model.Absci's ABS-101 is the first drug the company has ever brought to the clinic after over a decade spent on fundamental computer work and wet lab work.
It is the company's lead drug candidate in its pipeline of drugs. The novel ABS-101 antibody, developed using generative AI, binds to the TL1A protein in immune cells whose over-expression has been linked to a variety of inflammatory autoimmune diseases. Absci AbsciNot only did AI cut time and costs, but it has brought other novel advantages, said Christian Stegmann, the company's head of drug development, on the same call."Others have brought antibodies to the clinic that have had shortcomings, which we've tried to address," he said.
A big issue has been that prior therapies "lead the patient to develop anti-drug antibodies, which can lead the patients to needing to switch treatments." The ABS-101, he said, is intended to have "reduced immunogenicity risk" by design, which will hopefully mean less drug resistance.In addition, the AI techniques allowed the company to go immediately to a "subcutaneous" method of administering the dose, rather than via a drip into the vein, as is standard in Phase I trials.
"That is unusual; it usually comes much later in clinical development settings," said Stegmann. Using a needle versus a drip is important because, ideally, the final drug will be self-administered by patients.
If the drug is already being tried out via needle rather than drip, it brings the therapy that much closer to its final form.
"That allows us to be quicker in the overall clinical development pipeline, and to gather data for the setting that is actually going to reach the market" in the drug's final form, assuming it is ultimately approved."This is an advantage of AI," McClain said, "this ability to model not just for affinity and potency, but also to optimize for the manufacturability and such -- to go to all the attributes you want in the first go-around; that really helps."The full Phase I clinical trial will extend well into next year, said McClain and Stegmann.
Gathering results is somewhat longer than for other Phase I trials because ABS-101 was designed to extend the time between doses. That is a benefit for patients as it makes less frequent dosing (less frequent needle pricks) possible, but it means the trial takes longer to carry out those dosages.
"We have a long half life we have to monitor for a while," explained Stegmann.Well before the completion of Phase I, later this year, McClain expects to have a meaningful "read-out" of initial data from the Phase I. "We are going to, in Phase I, understand important pieces [of the whole trial process], as well as confirming whether we see the extended half-life" of the dosage, said McClain.
"We will also get a look at the immunogenicity profile; there will be a lot of good information, as far as being able to show the efficacy" of ABS-101.Because of the incremental data Absci will get later this year, they will know enough to seek approval for Phase II and begin recruiting subjects before the completion of Phase I.
Phase II is where the intense work of measuring the drug's effectiveness takes place, said McClain.
"It's fair to say we will be moving faster into Phase II" than might otherwise be the case, he said. After ABS-101, McClain's next candidate approaching clinical trials is ABS-201, which has two indications of note, one for treating hair loss in the form of alopecia, and another for endometriosis.
ABS-201 is expected to enter a Phase I trial in the first half of next year, McClain said.By any measure, drug development needs an overhaul.
Creating new drugs, or even repurposing old ones, comes with an enormous cost.
A new drug takes, on average, 10 years to develop, from fundamental chemistry through clinical trials to regulatory approval.
It can cost almost $3 billion, and the failure rate of most new drug candidates is 96%. There has been a lot of activity so far, without a breakthrough AI drug. The US Food & Drug Administration's Center for Drug Evaluation and Research received over 500 drug applications through 2023 that used some sort of "AI component," according to CDER's materials on AI in drug development. But, as Nature Magazine's Melanie Senior reported in December, "No AI-enabled drug candidate has yet made it past regulators, despite several being in clinical trials."Aside from Absci, a small cohort of startups have made progress getting into trials even if they don't yet have a clinical result.
For example, BPGbio of Framingham, Mass., has a drug for pancreatic cancer, developed using AI approaches, that is working its way through Phase II clinical trials.
Beyond the results of ABS-101, and other trials, the goal of Absci is to ultimately "predict the biology." That means the company will seek to "actually start to predict where an antibody should bind to a target to give us the biological response that we want."Absci's stock is publicly traded on Nasdaq.
The shares have defied a tough stock market this year, rising 12% versus a 2% decline for the Nasdaq Composite Index.
After hours on Tuesday, as Absci issued its press release, the stock surged by as much as 25% in late trading. Featured
Source: https://www.zdnet.com/article/this-ai-designed-drug-for-ibd-was-just-given-to-human-subjects-for-the-first-time/
#this #aidesigned #drug #for #ibd #was #just #given #human #subjects #the #first #time
·53 Ansichten
