The cost of innovation is high. Pharmaceutical drug development requires significant upfront investment and long-lead development programs. And yet, only a small fraction of these programs succeed in the form of new drug approvals.
Hurdles are also high. The new drug needs to be either more efficacious or safer than the standard of care (or better yet, fulfill both criteria). Providing evidence of the new drug’s benefit requires the selection of meaningful clinical endpoints in the appropriate patient population. A deep understanding of the natural disease progression and event rates is critical to qualify and quantify the new drug safety profile. In recent years, real-world evidence (RWE) has become an integral part of clinical development programs, developing innovative medicines faster and more cost-efficiently to improve patients' lives.
Regulatory agencies have set the framework and use cases for RWE. How can these increase the probability of success or efficiency in a pre-clinical and clinical drug development program?
- Target Identification and Validation: RWE can be used to identify potential new drug targets and to validate their importance in disease. For example, RWE can be used to identify genes or proteins that are differentially expressed in diseased versus healthy tissues or to identify biomarkers that are associated with disease progression or response to treatment.
- Pre-Clinical Trial Design: RWE can be used to inform the design of preclinical trials, such as by identifying the optimal dose and schedule of administration for a drug candidate or by identifying potential safety concerns. RWE can be used to identify the most relevant animal models for a disease or to identify potential drug-drug interactions.
- Translational Medicine: RWE can be used to bridge the gap between pre-clinical and clinical research by providing insights into the potential efficacy and safety of a drug candidate in humans. It can also be used to identify biomarkers that are predictive of response to treatment in humans or to identify potential safety concerns that may not be evident in animal models.
- Clinical Trial Design: RWE can be used to inform the design of clinical trials, such as by identifying the optimal patient population to enroll, selecting the most appropriate endpoints, and estimating the sample size required. RWE can be used to identify patients with a specific subtype of a disease who are most likely to benefit from a new drug or to identify the most clinically meaningful endpoints to measure in a trial.
- Clinical Trial Augmentation: RWE can be used to augment clinical trials by providing additional data on the efficacy and safety of a drug candidate by generating synthetic control groups for rare diseases or to identify patients who are at high risk for adverse events.
- Regulatory Decision-Making: Lastly, RWE can be used to support regulatory decision-making, such as by providing evidence of the efficacy and safety of a drug candidate in real-world settings like supporting a label expansion for a drug that is already approved for one indication or supporting the approval of a new drug for a rare disease.
To maximize these use cases, what are the criteria of a “good” RWE dataset?
- Fit-for-Purpose: The RWE dataset should be relevant to the research question being asked. For example, if the research question is about the efficacy of a new drug for cancer, the dataset should include data on patients with cancer who have been treated with the new drug.
- Completeness and Accuracy: A good RWE dataset should be as complete and accurate as possible. This means that the data should be missing as little information as possible and that the data should be recorded accurately. A robust dataset may also include unstructured data like patient-reported outcomes (PROs) and physical data in the form of biospecimen samples.
- Representative: The dataset should be representative of the population of interest. For example, if the research question is about the efficacy of a new drug in the general population, the dataset should include data on patients from all different backgrounds.
- Generalizable: The findings from an RWE dataset should be generalizable to other populations. This means that the findings should apply to other patients with the same condition, even if they are not part of the dataset.
- Transparency: All RWE datasets should be transparent and well-documented. This means that the source of the data, the methods used to collect the data, and the cleaning and pre-processing steps that were performed on the data should be documented.
In addition to these general criteria, here are some additional tips for selecting RWE datasets:
- A protocol-driven, retrospective and prospective disease state registry that’s differentiated from a constructed cohort;
- Deeply curated EMR data with complete structured and unstructured notes instead of pre-defined data dictionaries;
- Data with interventional physician and patient-reported outcomes, collection of biospecimen samples, and calculation/imputation of relevant disease activity scores;
- Ability to link the dataset to a large claims database to create more comprehensive and complete longitudinal patient journeys; and
- Ability to adjust the data for confounders through advanced statistical methodologies, particularly for smaller n-value datasets or complex causal inference studies.
Robust real-world data in combination with advanced statistical methods can provide fit-for-purpose RWE to augment new drug development programs, shorten time to new drug approval, and increase the probability of success and cost efficiency of these programs.
To learn more about how you can leverage real-world evidence solutions to help your product development, contact us at [email protected].