-
Categories
-
Pharmaceutical Intermediates
-
Active Pharmaceutical Ingredients
-
Food Additives
- Industrial Coatings
- Agrochemicals
- Dyes and Pigments
- Surfactant
- Flavors and Fragrances
- Chemical Reagents
- Catalyst and Auxiliary
- Natural Products
- Inorganic Chemistry
-
Organic Chemistry
-
Biochemical Engineering
- Analytical Chemistry
-
Cosmetic Ingredient
- Water Treatment Chemical
-
Pharmaceutical Intermediates
Promotion
ECHEMI Mall
Wholesale
Weekly Price
Exhibition
News
-
Trade Service
The industrial upgrading of the global pharmaceutical industry is starting around digitalization, intelligence, automation and other directions, AI-assisted drug research and development as one of the key technologies in recent years has been transformed into a number of major scientific research achievements, AlphaFold can predict almost all protein structures on the earth, deep learning tool ProteinMPNN can design new structural proteins in one second, AI technology platform accurately designs macrocyclic polypeptide molecules that can pass through cell membranes from scratch.
.
.
However, in the clinical stage, AI Pharmaceuticals has obtained fewer clinically validated drugs; So far, there are no approved drugs
on the market.
In this regard, Daphne, a senior expert in the field of AI pharmaceuticals
Speaking at this year's WuXi AppTec Health Industry Forum, Professor Koller said: "AI can make an important contribution to pharmaceuticals, enabling virtuous collaboration between machines and humans, helping to design experiments, generate data, and analyze outputs, but it may not be now, it will be decades away
.
”
In order to comprehensively analyze the development context and latest progress of AI pharmaceuticals over time, WuXi AppTec's content team systematically sorted out and analyzed 24 major AI pharmaceutical companies that apply AI to the field of drug development (quoted from Nature
Reviews Drug Discovery, see References [1]) through official channels disclosed preclinical and clinical stage pipelines, and strive to restore the real AI drug development portrait
.
List of 24 major AI pharmaceutical companies (Image source: Reference 1)
List of 24 major AI pharmaceutical companies (Image source: Reference 1)Analysis of the business and development stage of 24 AI companies
Analysis of the business and development stage of 24 AI companiesFirst of all, we divide the 24 major AI pharmaceutical companies according to the nature of AI business and AI drug pipeline stage, which is divided into end-to-end AI (AI technology application runs through the whole process of new drug discovery, from target discovery to lead compound determination to active compound screening to candidate drug identification), advanced AI platform/technology (with independent research and development pipeline), advanced AI tools (no independent research and development pipeline), and the AI drug pipeline stage is divided into clinical stage and preclinical stage
。 The most densely distributed are end-to-end AI companies with existing clinical pipelines (9); Among the AI platform/technology-based companies, 5 pipelines have entered the clinical stage, and 3 companies are in the preclinical stage, and the track development prospects are good; Among the AI tool companies, 5 companies did not disclose the pipeline of the clinical stage
.
▲WuXi AppTec content team mapping
▲WuXi AppTec content team mappingAnalysis of 67 AI pharmaceutical pipelines
Analysis of 67 AI pharmaceutical pipelinesAccording to incomplete statistics, there are currently 67 drug pipelines of these 24 major AI companies (Note: these companies independently develop their own pipelines and use AI to assist discovery or development, excluding the cooperative R&D pipelines listed on the official website) according to the research and development stage, 42 pre-clinical, 13 in phase 1, 11 in phase 2, and 1 in phase 2/3 (please see the table at the end of the article for details).
Although AI-assisted drugs are still some distance away from launch, and the results of clinical trials are also mixed, we will continue to pay attention to the development of these pipelines in the future to see if they can cross the finish line
of listing.
.
▲WuXi AppTec content team mapping
▲WuXi AppTec content team mappingIn the global AI pipeline, indications are concentrated on cancer (non-rare diseases) and rare diseases, accounting for 45% and 13%
respectively.
The remaining 42% of drug indications are for Alzheimer's disease, inflammatory bowel disease, hair loss and other diseases
for which there is currently no cure.
Developers look forward to using AI to discover new targets and design molecular structures to increase the potential for
discovering "first-in-class" and/or "best-in-class" drugs.
▲WuXi AppTec content team mapping
▲WuXi AppTec content team mappingAccording to the current publicly disclosed information, in the global AI pipeline, in addition to 22 drug targets may be new and unknown, 26 drugs target targets that have been approved therapies, accounting for 39%, including TYK2, PHD2, PI3Kα, S1P1 and other validated targets, most of which are in preclinical and phase
1 and 2.
While these targets are already approved for marketing, the therapeutic model of the pipeline may pursue more economical treatments, better drug efficacy, or better potential
to overcome drug-resistant mutations 。 19 drugs target targets under development, but there are no drugs for this target therapy on the market, accounting for 30%, including STING, MALT1, HPK1, GSK-3β and other popular targets, and most of the therapies are in the early clinical stage, among them, the advancement of some target research therapies is currently encountering great challenges, such as CD47, so it may be expected to promote the research of
related therapies through AI platforms.
▲WuXi AppTec content team mapping
▲WuXi AppTec content team mapping
▲WuXi AppTec content team mapping
▲WuXi AppTec content team mappingAmong the 24 major AI pharmaceutical companies, if you consider both preclinical and clinical pipelines, the largest number is Insilico
Medicine）（8）、Collaborations
Pharmaceuticals（7）、Aria
Pharmaceuticals (7), if only clinical pipelines are considered, the largest number is Recursion
Pharmaceuticals（6）、Relay
Therapeutics（3）、Exscientia（2）、Valo
Health（2）、Healx（2）、Berg（2）
。 Due to space constraints, this article will introduce some of the companies with large clinical pipelines and their representative pipelines
.
▲WuXi AppTec content team mapping
▲WuXi AppTec content team mappingRecursion Pharmaceuticals – 6 pipelines have entered the clinical stage
Recursion Pharmaceuticals – 6 pipelines have entered the clinical stageRecursion
Pharmaceuticals is a clinical-stage biotechnology company dedicated to industrializing
drug discovery by integrating technological innovations in biology, chemistry, automation, machine learning, and engineering.
In biology, Recursion uses tools such as CRISPR genome editing and synthetic biology for R&D, and in the course of laboratory research, Recursion automates repetitive and tedious laboratory work on a large scale with the help of advanced robotics; In phases outside of lab research, Recursion uses neural network architectures for iterative analysis and reasoning on large, complex internal data sets, using cloud solutions to increase the resilience
of high-performance computing.
REC-2282 (Phase 2/3 Clinical)
REC-2282 (Phase 2/3 Clinical)REC-2282 is a CNS permeable, orally bioavailable small molecule pan-HDAC inhibitor being developed to treat meningiomas
with NF2 gene mutations.
Recursion does this by leveraging its proprietary AI-driven drug discovery platform, Recursion
OS, discovered REC-2282 as a potential candidate
for the treatment of diseases caused by mutations in the NF2 gene.
This molecule appears to be well tolerated, including in patients who have been given for many years, and unlike other HDAC inhibitors, it may reduce cardiotoxicity
.
REC-4881 (Phase 2 Clinical)
REC-4881 (Phase 2 Clinical)REC-4881 is an oral, non-ATP competitive MEK1/MEK2 small molecule inhibitor being developed to reduce polyp burden and adenocarcinoma
in patients with familial adenomatous polyposis (FAP).
REC-4881 has been well tolerated in previous clinical studies, consistent with intended use, and has a characteristic gut-localized PK that may be beneficial for
FAP as well as potentially other APC-driven gastrointestinal tumors.
REC-4881 and REC-2282 have been granted
orphan drug and Fast Track designation by the U.
S.
FDA and the European Union.
Relay Therapeutics – Cholangiocarcinoma remission rate of 88%, Phase 2 clinical pipeline efficacy is excellent
Relay Therapeutics – Cholangiocarcinoma remission rate of 88%, Phase 2 clinical pipeline efficacy is excellent
Founded in 2016, Relay Therapeutics is a clinical-stage small molecule targeted drug research and development company
for precision tumors.
Relay
Therapeutics' Dynamo platform has protein dynamics at its core and is applied to three key stages
of exercise-based drug design.
The first stage is to understand how to drug the
protein target of interest.
For each target, the initial goal is to better understand the structure and conformational dynamics of all domains of the protein to identify potential novel allosteric binding sites
.
First, Relay
Therapeutics synthesizes full-length proteins
.
Next, a series of protein visualization methods, such as Cryo-EM and ambient temperature X-ray crystallography, are used to fully grasp the dynamic conformation of the target protein
.
The experimental dataset is then deployed in the computational platform to generate a virtual simulation
of the full-length protein.
The next step is to discover and generate lead compounds
.
The company integrates computational and experimental capabilities to design ligand-centric screening compounds
based on physiologically relevant activities.
These screened data provide input to the machine learning component of the Dynamo platform, which enables rapid identification of lead compounds and then optimization until candidate compounds
are selected.
RLY-4008 (Phase 2 Clinical)
RLY-4008 (Phase 2 Clinical)September 11, Relay
Therapeutics presented the latest clinical trial data for RLY-4008 in his ESMO presentation, which has excellent efficacy against FGFR2-positive cholangiocarcinoma with an objective response rate of up to 88%.
FGFR2 is a receptor tyrosine kinase that often mutates
in certain cancers.
However, small molecules that inhibit FGFR2 activity often also inhibit the activity of FGFR1 and other FGFR family members, and this off-target effect leads to many toxic side effects
.
Relay's proprietary technology platform developed the FGFR2-specific inhibitor RLY-4008
through detailed analysis of the homeostasis of protein conformation.
Preclinical studies have shown that RLY-4008 exhibits high selectivity for FGFR2 targets in cancer cell lines, with minimal
effect on other targets while shrinking tumors.
FGFR2-specific inhibitors are highly selective for FGFR2 targets
Exscientia – The first company to advance AI-designed drugs to the clinic
Exscientia – The first company to advance AI-designed drugs to the clinic
Founded in 2012, Exscientia is an AI-driven drug discovery company
.
The application scenarios of the Exscientia technology platform mainly focus on target discovery, drug design, biophysical screening, analysis and prediction of oncology clinical outcomes
.
For target discovery, Exscientia combines genetic data and global literature in machine learning models to predict and confirm disease-target associations, and identifies suitable targets
through druggability and handling assessments.
Regarding the drug design step, Exscientia uses sparse data to design molecules and active learning for targets, significantly improving the efficiency of drug candidate discovery; Balance molecular coding efficacy, safety, and bioavailability and dramatically improve your chances of
success in a clinical setting.
Fragment-based screening and X-ray techniques provide an ideal experimental basis for AI design, transduction groups and biophysical experiments enable the optimization of the design of this key target family, and high content data enables Exscientia to develop the widest range of machine learning models and AI design opportunities
.
In the field of precision medicine, Exscientia uses the power of living patient tissue and artificial intelligence to demonstrate improved
oncology clinical outcomes.
It is worth mentioning that Exscientia is the first company to advance AI-designed drugs to the clinic, but unfortunately, the world's first AI drug DSP-1181 to enter phase 1 clinical trials has been discontinued, and the company still has 2 clinical stage pipelines as of now
.
EXS-21546 (Phase 1 Clinical)
EXS-21546 (Phase 1 Clinical)EXS-21546 is an AI-designed A2A receptor antagonist
.
Some tumors produce high levels of adenosine, which binds to and activates A2A receptors on immune cells, thereby suppressing the immune system's antitumor activity
.
The EXS-21546 under study inhibits the ability of high concentrations of adenosine to activate A2A receptors, thereby promoting antitumor activity
of immune cells.
DSP-0038 (Phase 1 Clinical)
DSP-0038 (Phase 1 Clinical)DSP-0038 is designed as a single small molecule with the dual efficacy of a 5-HT2A receptor antagonist and a 5-HT1A receptor agonist, while selectively avoiding similar receptors and unnecessary targets such as dopamine D2 receptors
.
DSP-0038 is the third molecule
discovered by Exscientia using AI technology to enter clinical trials.
epilogue
epilogueAI pharmaceutical, the road resistance and long
AI pharmaceutical, the road resistance and longIn summary, AI pharmaceuticals are still an emerging young field, 82% of drugs are still in the initial stage of clinical phase 1 and earlier research and development, individual R&D failures are inevitable, and the underlying logic
of AI pharmaceuticals cannot be overturned.
In the clinical pipeline of AI drugs, RLY-4008 has a good anti-off-target effect, and REC-2282 and REC-4881 show novel mechanisms of action among similar inhibitors, which shows that AI discovery drugs can provide unique innovation in structure and pharmacological mechanism
.
Therefore, for AI pharmaceuticals, we should not be too deified, nor should we blindly sing about it, but should uphold a calm and objective attitude, see the impact and innovation brought by AI to the pharmaceutical industry, and accept the setbacks and lengths
that must be experienced in the development of new things.
.
We should not be overly deosified, nor should we blindly sing downfall, but should uphold a calm and objective attitude,
Schedule: 24 major AI pharmaceutical companies in the pipeline under development
Schedule: 24 major AI pharmaceutical companies in the pipeline under development
References:
References:
1.
Madura K.
P.
Jayatunga et al.
, (2022) AI in small-molecule drug discovery:
a coming wave? Doi: https://doi.
org/10.
1038/d41573-022-00025-1
Madura K.
P.
Jayatunga et al.
, (2022) AI in small-molecule drug discovery: a coming wave? Doi: https://doi.
org/10.
1038/d41573-022-00025-1
2.
Artificial Intelligence for Drug Discovery Landscape Overview Q1 2022,
Retrieved Octobor 14 2022, from
Artificial Intelligence for Drug Discovery Landscape Overview Q1 2022, Retrieved Octobor 14 2022, from
