Drug development continues to be painfully slow and mind-bogglingly expensive.
In a 2010 article, Stephen Paul estimated that developing a new drug cost $1.8 billion and took thirteen years on average from discovery to regulatory approval. These excesses occurred because drug development is a fundamentally complex and risky process that is also subject to rigorous regulatory oversight and requirements.
Since then, despite well-intended efforts to streamline this process through digital solutions such as data warehouses and electronic submission formats, the productivity of pharma has continued to fall as costs rose. A study recently published in Drug Discovery Today found that large pharmaceutical companies now average $6.16 billion in spending per approved drug. Only nine of the sixteen largest pharma companies currently show positive R&D productivity; the rest had negative productivity, only offset by mergers and acquisitions.
At the same time, the pharmaceutical industry is undergoing unprecedented scrutiny regarding drug pricing. In particular, the Inflation Reduction Act (IRA; enacted in August 2022) shifts responsibility for drug costs away from beneficiaries and onto healthcare plans and manufacturers. Perhaps most significantly, this law allows the US federal government to negotiate prices for the first time (read about how sponsors can navigate the IRA in this white paper).
The double whammy of falling productivity and looming pricing scrutiny threatens the industry’s ability to deliver innovative new treatments to patients. In an excellent analysis of the crisis of pharma R&D productivity, Alex Zhavoronkov proposed that more academics need to research the root causes of this inefficiency and that investors should demand more accountability and transparency of the drug programs they’re funding. In this blog, I’ll suggest two potential technologies that drug developers should embrace to mitigate this threat.
Greater adoption of biosimulation
A computational modeling approach used in drug development, biosimulation involves conducting virtual clinical trials, which allows researchers to predict the safety and efficacy profile of drugs in various potential scenarios. Biosimulation can accelerate drug development in several ways:
1. Streamline clinical development: Biosimulation allows researchers to create predictive models of biological systems, enabling them to evaluate the effects of different drugs and dosages on the body before conducting expensive and time-consuming clinical trials. This reduces the number of clinical studies required, saving time and resources. For example, physiologically-based pharmacokinetic (PBPK) modeling and simulation has become an established biosimulation approach to assess drug-drug interaction (DDI) liabilities involving CYP enzymes. Successful application of PBPK models during regulatory review can be used instead of clinical trials and to inform drug product labels. In the recent NDA submission of asciminib (Scemblix®) by Novartis (Figure 1), PBPK simulations replaced over 10 clinical pharmacology studies and played an instrumental role in the approval of two additional doses by the US FDA with no additional clinical pharmacology studies at the date of approval (learn more about this case study by watching this webinar).
2. Improve the probability of success of clinical trials: Nine out of ten drugs in development don’t get to market, costing pharmaceutical companies billions every year. Even most drugs that get to Phase 3 clinical trials don’t get approved. These late-stage failures are especially costly as companies have already invested hundreds of millions of dollars in discovery, development, research, and testing. 试验失败的原因之一是试验设计欠佳。There are many variables to fine-tune to optimize the study design and maximize the probability of trial success. Biosimulation can be used to evaluate various clinical trial designs in silico to maximize the chances of trial success. Leveraging existing knowledge for a drug under study with simulation can help answer critical questions to increase the probability of meeting study endpoints.
3. Fail faster: One of the best ways to increase pharma R&D productivity is to halt the development of drug programs that are doomed to failure as early as possible. Certara scientists built predictive treatment-response models combined with an integrated metric of net patient value (Clinical Utility Index) to help Sanofi Aventis realize that it had little likelihood of competing with the standard of care. “…we stopped funding development of the compound,” said Frank Douglas, who was Aventis’ chief scientific officer and executive vice president of drug innovation and approval. “The ratio between the therapeutic benefit and the side effect demonstrated that this [compound] was not as beneficial as Evista.”Douglas … estimated that the [biosimulation] saved the company $50 million to $100 million, the cost of later-stage clinical trials. “We also avoided exposing a lot of women to a drug that ultimately would have failed,” he adds. “And we were able to switch to another project with a greater chance of success.”1
Overall, biosimulation has the potential to revolutionize drug development by accelerating the discovery and testing of new drugs, improving their efficacy and safety, and reducing the overall cost of bringing new drugs to market.
Embracing Artificial Intelligence technology
Artificial Intelligence (AI) is becoming increasingly useful in drug development due to its ability to process vast amounts of data quickly and accurately, identify patterns, and make predictions.
AI is everywhere you look these days, so what makes Certara’s approach distinct?
Where our competitors focus on general AI, applications, Certara’s solution combines unrivaled technology, data, and industry expertise to provide the leading life sciences-specialized AI platform.
Here are some unique ways that our AI platform, Certara.AI, can help increase the productivity of pharma R&D:
- 药物开发. 要了解一种分子进一步开发的可行性，需要对复杂的化学、生物、逻辑和计算数据进行分析。Certara.AI can comprehend complex data types so that discovery scientists spend less time running mundane computational analyses and research tasks.
- Clinical Research & Trial Execution. Eighty percent of the data clinical trial teams need resides in unstructured content. Pharmaceutical scientists spend a great deal of time manually searching this content. 在接受生物医学文献培训后，Certara.AI 可减少人工研究任务，加快收集团队所需的相关数据，以便成功地设计和实施试验。
- Medical Writing. Medical writers spend tremendous time developing drafts of the regulatory documents required for submissions to health authorities. With AI-generated content, medical writers streamline the early drafting process allowing more time for fine-tuning content and quality control, enabling more productive development of regulatory documents.
The continuously upward-spiraling cost of drug development hinders the pharmaceutical industry’s ability to continue delivering innovative treatments for patients’ unmet medical needs. True transformation of pharmaceutical R&D requires a fundamental shift in innovation models to achieve sustainable productivity gains. By embracing biosimulation and AI technology, the industry can improve its productivity and thus maintain its ability to continue delivering safer and more effective drugs to patients.
References:
- “I Zing the Body Electric.”Forbes.com [serial online]. 2002 年 10 月 7 日. Available at: https://www.forbes.com/asap/2002/1007/054.html
- Loisios-Konstantinidis I. Physiologically-based Pharmacokinetic Modeling & Simulation to support ASCIMINIB NDA submission & inform drug product label. Certara. 2023 年 8 月 7 日. Accessed 2023 年 10 月 24 日. https://www.certara.com/on-demand-webinar/physiologically-based-pharmacokinetic-modeling-simulation-to-support-asciminib-nda-submission-inform-drug-product-label/.
