How AI is Transforming the Pharmaceutical Industry

Use case 1:

AI in Pharmacovigilance (Adverse Event Identification, Validation and Reporting):

Pharma companies are pivoting towards beyond the pill solutions by harnessing the broad capabilities of digital and data analytics to research, develop, and craft experiences in new ways across entire value chain. Some pharma companies are using digitally native solutions that integrate with consumer/clinician activities and workflows to improve potential adverse events detection, accelerating potential remediation and improved outcomes. Identifying and reporting AEs is an extremely labor intensive process for pharma companies requiring significant manual effort and is prone to errors. Enormous number of potential AEs are reported annually however only a small fraction of them (5-20%: source FDA MedWatch System) end up being actually confirmed post investigation as an AE and linked to the use of pharma companies’ drug. Each potential AE needs to be assessed and adjudicated carefully to assess if it’s a true AE or not. Therefore, it is a perfect use case for using the power of AI to automate the process and make it more efficient/less prone to errors.

Solutions out there include chatbots/ apps on social media and other channels which consumers/ clinicians can interact with directly, and report their potential AEs that are linked to use of pharma companies’ drugs. These inputs are assessed by AI algorithms to ascertain if the symptoms being experienced were indeed linked to the manufacturer’s drug or potentially something else. This assessment needs intricate algorithms and some manual involvement initially by the AE teams to validate what the AI is adjudicating is correct. The system/data from several AE adjudication cases feeds self-learning whereby the manual verification overrides the AI recommendation and eventually the AI is optimized by being trained on data that becomes more available over time. If classified as an AE, it is sent to the company's Drug Safety database and eventually reported to the FDA. Interestingly these clinician/patient facing chatbots can also be used in a compliant fashion to gather valuable patient/caregiver/HCP data that can inform commercial teams with several key metrics

Outcome: Using AI in AE identification, validation and reporting helps differentiate true AEs from potential AEs, filter out false positive AEs in a highly efficient fashion by reducing millions of dollars of overhead costs linked to manually performing these processes. It also helps leverage end user inputs for improved clinical and commercial insights generation, and enhanced overall consumer experience.

Use case 2:

AI in drug discovery:

AI-based drug discovery in oncology has made significant advancements in recent years, transforming the way new cancer treatments are being developed. Drug screening, repurposing and target identifications are three common use cases. One of them which the author intends to share here was the topic of his PhD dissertation- the use of in silico/ AI based drug design to characterize and find ligands binding to tumor suppressor protein p53, to characterize and find ligands binding to tumor suppressor protein p53, a pioneering approach which is now being followed by multiple pharma companies to identify oncology therapeutics, one of which is currently in Phase 2 trials and granted FDA Fast Track Designation. p53 is a key protein in cell cycle regulation, the defect in which causes unchecked growth of more 50% of cancers in humans. In many tumors, p53 is inactivated directly by destabilizing mutations. The aim of the project was to rescue the function of p53 by binding of small-molecule compounds. A lot of efforts have been made in past to target p53 however none have ended up in the clinic. The strategy used here was to screen and design compounds which could bind to cavities on the mutated p53 protein surface, and thus shift the folding-unfolding equilibrium toward the native folded state. It is a very elegant and unique mechanism of action whereby the aim is to re-activate a mutated protein, and thus very different to traditional mechanisms of blocking protein targets which is more or less the standard in the industry.

In-silico/ AI based screening of small molecules that might have a stabilizing effect is a viable and attractive strategy to minimize the number of compounds to a few, which can then be experimentally tested. Using this philosophy, over 2.5 million compounds were computationally screened by employing several filters such as the Lipinski's rule of 5, pharmacophore models, small molecule docking and manual analysis of the resulting high-scoring small molecules. For the pharmacophore model, the structure of p53 core DNA binding domain with Y200C mutation (reported to affect 75,000 human cancers every year) was used as a starting point to identify small molecule compounds that bind to a the Y220C cavity. The small molecules were then tested experimentally using a battery of biophysical ‘approaches' but better use 'experimental assays/techniques' ranging from the less sensitive to the most sensitive ones. This included Calorimetry, Fluorimetry, Ultracentrifugation and NMR. The study identified a few compounds that were shown to bind p53 Y220C crevice. A lead compound Phi Kan 83 was found to bind to the Y220C cavity tightly enough with a potential to stabilize the protein. The same tumor agnostic approach targeting the identical crevice in the Y220C mutated p53 is being leveraged by pharma companies with potential impact on patients with various cancers, one of which is currently in Phase 2 trials and granted FDA Fast Track Designation. If this is approved it would be a paradigm shifting proof of concept for targeting tumor suppressor p53, by activation vs. inhibition, and a good validation of in silico-based drug discovery.

Phi Kan 83 was discovered 16 years ago. The computation capacity has significantly improved. The ligand docking simulations which took days now take hours in processing, cutting time by years in the discovery process. Insilico Medicine, a budding AI based drug discovery company combines AI with bioinformatics to identify novel drug candidates and biomarkers for aging and age-related diseases. In silico has two drug candidates identified through AI based/ computational workflows within Idiopathic Pulmonary Fibrosis (IPF). If marketed it will be another proof of concept which cut the cost of identification and commercialization by a fraction of the time to the traditional drug discovery process, and an industry changing event.  Google Deep Mind's Alphafold 3 (from Isomorphic labs) released a few weeks back has the power to predict the structure and interactions of all life's molecules with unprecedented accuracy. It has opened a plethora of opportunities to maximize discovery efforts in this golden age of tech and science convergence. AlphaFold 3 is already helping Isomorphic design new drugs for Eli Lilly and Novartis. There are other players like Benevolent, Exscientia and more who are following suit.

Another use case is a new platform being developed at University of California (San Diego), called POLYGON, is unique among AI tools for drug discovery in that it can identify molecules with multiple targets, while existing drug discovery protocols currently prioritize single target therapies. Multi-target drugs are of major interest to doctors and scientists because of their potential to deliver the same benefits as combination therapy, in which several different drugs are used together to treat cancer, but with fewer side effects (https://www.sciencedaily.com/releases/2024/05/240506131601.htm). The possibilities are virtually endless. No wonder this use case of AI in drug discovery has received the most attention/ funding in the last year

Use case 3:

AI in Commercial Operations:

Pharmaceutical marketing teams are increasingly leveraging Generative AI (Gen AI) to enhance their strategies and achieve more personalized and effective engagements with healthcare professionals (HCPs) and patients. Some use cases being used summarized below:

Custom Marketing Materials:

Gen AI tools can generate tailored content for different segments of healthcare professionals and patients. This includes creating personalized emails, brochures, and educational materials that address specific needs and preferences. Field force reps can upload an HCP profile and based on their therapeutic inclinations can carve out messages that would be better suited for them in order to ascertain positioning of different therapy options

Predictive Analytics:

AI models analyze vast amounts of data to predict which physicians or patients are most likely to respond to specific treatments or messages. This helps in targeting advertising efforts more precisely, improving the return on investment (ROI) for marketing campaigns:

o Early patient identification: Tools/ vendors out there using labs/EHR/Claims data are able to find patients as soon as they have a diagnosis and target their HCPs to educate them about potential therapeutic options they have that could help their patients. Each of these 3 datasets have slightly nuances offerings – while labs is great in cutting the data lag by 1-2 mths, EMR helps confirm a diagnosis but has limitations to be used for any targeting purposes. It is also highly unstructured/ difficult to decipher and relatively smaller. Claims has very rich data (both breadth and depth) however often comes with a time lag linked to the claims processing. Some vendors out there are using proprietary clinical algorithms to extract the medical diagnosis from the labs data and offer a much earlier patient identification which is a very novel use case. An application of this could be finding patients as soon as they turn metastatic and targeting them with therapies approved for metastatic disease

o Predict chances of metastasis based on longitudinal therapy history: Another novel approach is by using patient’s historical drug use, diagnosis, and social data spanning several years to predict when the patient will progress/ turn metastatic. The results have been surprisingly impressive with > 90% accuracy

o Alerting field force on targeting HCPs based on their prescribing patters: A more common approach within the industry has been to alert the field force based on HCP prescribing patterns of a specific drug or its competitor) e.g. going after HCPs who are recent decliners/ have reduced prescribing, or growers/increased in order to reinvigorate the decliners and encourage the growers
 
Market Research:

AI can analyze market trends and competitor activities, providing strategic insights that inform product positioning and marketing strategies.

o HCP interviews conducted as part of the market research projects can be uploaded to company specific Chat GPT like tools in an audio format. These can be easily converted into word documents and used to synthesis/summarize the key findings pertaining to the HCP preferences, unmet needs and how different therapies address those unmet needs

Digital HCP Analytics

Behavioral Analysis: AI tools analyze digital behavior patterns of HCPs and patients to understand their interests, preferences, and pain points. This insight allows for more effective segmentation and targeting

Sentiment Analysis: By analyzing social media, forums, and other online discussions, AI can gauge public sentiment towards certain drugs or treatments. This feedback is invaluable for adjusting marketing strategies and addressing concerns

Omni-Channel Campaigns: AI systems orchestrate and optimize campaigns across multiple channels (email, social media, webinars, etc.), ensuring a cohesive and effective marketing strategy that reaches customers wherever they are

This is just tip of the iceberg. There are ‘n’ number of uses cases that will be developed which will redefine the way pharmaceutical industry operates at every functional level. The end beneficiary would be all the players in the value chain- pharma companies who will benefit from more efficient and effective processes, HCPs who will get a personalized messaging for their needs at the right time, but most importantly patients who will get more superior treatments earlier especially with AI based drug discovery promising to cut the development and approval times by >70%. The golden dawn of life sciences made possible by merging of sciences and technology has only begun.

--Issue 04--