West Monroe is uniquely positioned to assist pharmaceutical manufacturers in implementing AI technologies so they can leverage genetic data when developing novel therapies. West Monroe has specific service offerings for digital innovation and business transformation in life sciences, as well as a dedicated technology practice with a focus on analytics and artificial intelligence. Our expertise in these areas allows us to serve as effective advisors as we collectively work to revolutionize the drug development process.
The global effort to find an effective treatment for COVID-19 has highlighted the need for advanced genetic sequencing and analysis technologies that contribute to speed and efficacy in the drug development process while potentially driving down costs. At least 17 of the 150 potential vaccines in development have started human testing. Of these, five are a novel kind called an RNA Vaccine. These are vaccines developed by injecting virus RNA in order to stimulate antigen production, a process that would not exist without recent advances in the fields of genomics and genetic testing.
Genetic testing, the data it unlocks, and the combination of emerging AI technologies will revolutionize the pharmaceutical industry by drastically reducing the cost and time needed to develop novel therapies. This stands to improve rates of approval and clinical trial success, accelerate the process of sorting through massive amounts of genetic data, and support better target selection and drug efficacy.
Genomics, the study of all genes and their interrelationships in order to identify their combined influence on growth and development, has been particularly impacted by innovations in genetic testing. Genetic testing examines an individual’s DNA for potential mutations or variations that may cause disease or illness, as well as identifying ancestral origins.
The field of genomics has undergone a major transformation since 2003 due to the completion of the Human Genome Project. We’ve seen the advent of home genetic testing kits make genomics accessible in a way unlike ever before while generating an unprecedented amount of genetic data that has helped bring about new developments in lab-based genetic testing. Next generation sequencing, for example, is a new form of genetic sequencing technology which works by running all sequencing reactions in parallel on a single surface, allowing billions of sequencing reactions to occur simultaneously. The Sanger Method, used during the Human Genome Project, took a decade to sequence the human genome; NGS methods now enable sequencing of the entire human genome in a single day, drastically increasing the amount of genetic data available to research organizations and pharmaceutical manufacturers. NGS is being used to drive the growth of precision medicine and pharmacogenomics.
Armed with such advanced capabilities, and recognizing the value of tapping into the data collected by genetic testing companies such as 23andMe, pharmaceutical companies and other healthcare players are becoming increasingly interested in partnerships that will afford them access to the data collected from customers for use in research and development. For example, 23andMe has signed a $300 million partnership agreement with pharmaceutical manufacturer GlaxoSmithKline (GSK) that gives GSK access to deidentified user genetic data for use in drug research and development. 23andMe also has a partnership with TrialSpark, a technology company focusing on accelerating clinical trials, to leverage 23andMe’s user data to identify patients who are an optimal match for TrialSpark clinical trials. Another major player in the genetic testing space, Ancestry DNA, recently concluded a study done with Google’s life-extension spinoff company, Calico.
The genetics testing market undoubtedly is growing. Valued at over $13 billion in 2019 and projected to reach $28.5 billion by 2026, with an anticipated compound annual growth rate (CAGR) of 12.2% from 2020 to 2026, this market will be at the forefront of drug development, particularly as data-sharing partnerships continue to be forged. But access to genetic information is just the beginning. Given the sheer quantity of data generated by sequencing even a single human genome, advanced analytics technologies are an absolute must if that data is to be useful to manufacturers and researchers.
Prior to moving to clinical trials, researchers must first work through the drug discovery phase to identify a target, the biochemical mechanism of a disease or molecular structure in the body that the novel medicine will be designed to treat. Approximately 50% of late-stage clinical trials fail due to ineffective drug targets, which are typically found using a time-consuming trial and error approach—methodically testing each identified potential target for promising results, eating up time and resources.
This is understandable to some degree, considering the sheer diversity and volume of genetic code variations. There are tens of thousands of genes in the human genome and researchers are just beginning to unlock each gene’s function and potential disease associations. Researchers from 37 different countries recently finished a decade-long study analyzing the whole genetic code of 2,658 different cancers. The study found that cancer contains, on average, between four and five mutations that drive the growth, and that more than one-fifth of those mutations occur years before the cancer develops. These findings present a real opportunity for developing early genetic diagnostic tests to catch driver mutations before the cancer ever forms and for developing new cancer treatments with these newly identified driver mutations as targets.
A group of researchers at GSK found that selecting drug targets supported by known gene-disease associations could double the drug success rate in the development process. These researchers also hypothesized that genetic mutations known to affect physical traits would be more responsive to drug intervention. The conclusion drawn here is that using genetic data to identify gene-disease associations will improve target selection, leading to a higher drug success rate and a faster, more cost-effective clinical development process.
This is where advanced analytics like AI come in. Including AI in drug development has the potential to more efficiently and accurately unlock previously unknown drug targets by helping analyze genetic data to identify potential variants and then determine if a variant is benign or a potential clinically-relevant target. Identifying variants leads to the discovery of new gene-disease associations and enhances the understanding of disease pathways, a proven method to increasing the drug success rate in clinical trials. Using genetic data and gene-disease associations to replace the trial-and-error target identification process will speed the development of successful new treatments, resulting in patients having access to potentially life-saving medications faster. In the future this will be the new standard for drug discovery and clinical development.
And we’re not just talking about quickly analyzing mountains of patient data—advanced analytics can also be deployed to scour information contained in scientific literature and databases to identify patterns and make connections between biomedical entities like disease targets and potential medicines, while reducing the opportunity for human bias and error. The Boston-based company Berg proved this when it used its AI platform to generate and analyze large amounts of biological and outcomes data from patients to understand the differences between healthy and cancerous cells. In doing so, they identified a previously unknown cancer mechanism and now have a potential treatment targeted to this mechanism in stage II clinical trials.
Genetic data will soon be a mainstream tool for precision medicine, an important part of which is pharmacogenomics, as it considers a genome’s impact on drug efficacy as well as an individual’s optimal dose and any potential adverse effects caused as a result of taking the medication.
Understanding the impact of genetics on drug response will be key in developing drugs of the future. Some companies have already begun leveraging AI and machine learning to identify more personalized treatment methods and are reaping the rewards. Biotech Sema4, which works with physicians to analyze genetic data using AI technology to identify patients’ optimal treatments, was recently valued at over $1 billion.
Pharmacogenomics is being employed by some manufacturers in the form of companion diagnostics, genomics-based diagnostic tools delivered with a drug in order to ensure the patient is receiving the proper treatment based on his or her genetic profile. Companion diagnostics are needed when treating conditions linked to specific genetic mutations; they are used to identify if a patient has the specific mutation and will therefore respond to treatment, and if a patient has any additional genetic variants that could cause an adverse reaction to the medication.
Companion diagnostics are important tools for improving patient quality of life and patient outcomes. As the use of real-world evidence and health economics and outcomes research by payers and PBMs grows, it will be in pharmaceutical manufacturers’ best interests to optimize drug efficacy rates and patient outcomes.
Companion diagnostics can and should be used to guide trial participant selection, ensuring that the individuals selected for the trial will respond to treatment with few, if any, adverse side effects. The ability to identify patient segments that are at the highest risk of adverse events or not responding to the treatment will decrease the probability of clinical trial failure, thereby increasing the likelihood of regulatory approval and future commercialization. This is an effective way to enhance clinical trial success and reduce cost during the drug development process, while directly improving patient experience during and after clinical trials.
In a similar vein, genetic data may be the key to improving drug approval success rates with the FDA, which is currently around 10%, costing pharmaceutical companies approximately $2.6 billion per approved medicine over an average of 10 years. This is incredibly costly and the scale is enormous—even a small increase in the approval rate would save millions of dollars for drug developers and manufacturers. With access to wider swaths of more precise genetic data and the application of advanced analytics capabilities to improve targeting and trial participant selection, the pharmaceutical industry stands to make tremendous gains in terms of cost savings, efficiency, and patient and provider experience.
The world of genomics and genetics testing, however, does raise some concerns that could hamper the widespread use of genetic data in drug development. Data privacy has become paramount to consumers in the digital age, and no data is more personal than one’s own genetic code. The prevalence of genetic testing, particularly direct-to-consumer tests sold by companies like 23andMe, has raised significant data privacy concerns. Pharmaceutical manufacturers looking to access valuable genetic data should be aware of the issues surrounding the industry, as it may impact their ability to access the data in the future.
The HIPAA security rule allows for the sale of patient data as long as it has been anonymized and scrubbed of all patient identifiers, but it is impossible to completely anonymize DNA because it is unique to each person. The Genetic Information Nondiscrimination Act (GINA) was passed to restrict the access of issuers of health insurance and employers to individuals' genetic information and prohibit genetic discrimination from these groups. However, the GINA does not restrict access of genetic information to other types of insurance providers.
Direct-to-consumer genetic testing companies claim to obtain “informed consent” from customers when they register their test kits but often bury privacy rules and information on how long they keep customer data deep in long terms and conditions documents. These concerns could deter consumers from purchasing tests or permitting their data to be included in research studies, which could limit the amount of data pharmaceutical manufacturers have access to, thereby hindering advancements in the drug development process.
Pharmaceutical companies must understand and address to the best of their abilities the privacy issues surrounding their use of valuable genetic data, treating privacy and security as a matter on par with drug efficacy. If these concerns are not addressed, it will likely become difficult to access the data as consumers become more aware of these problems and could do further damage to the reputation of an industry that has experienced more than its share of negative publicity.
Genetic testing and the use of AI will upend the drug development process and, by consequence, the pharmaceutical industry. The ability to better identify drug targets and move through the drug development process more quickly with a significantly higher rate of success will decrease the overall cost to bring a product to market. Pharmaceutical organizations could leverage the genetic data to make better informed decisions during drug development, potentially leading to innovative treatments and cures for many of our world’s most debilitating diseases, COVID-19 not least among them.
As for where we stand now in terms of our present pandemic, the highly touted Moderna vaccine currently entering Phase 3 clinical trials is one of the RNA vaccines we mentioned in the introduction. This vaccine would not be where it is in development without technologies like rapid genetic testing and analysis. While RNA vaccines have never before been tested in the real world, they show incredible promise in the lab; of Moderna’s 45 Phase 1 participants, all demonstrated immunity with limited side effects. This vaccine would be the first of its kind, revolutionizing the biotech and vaccine industries.