Clinical trials are central to modern medicine, meticulously designed to evaluate the safety and efficacy of new drugs and treatments. However, the carefully controlled environment of a clinical trial often encounters a significant real-world challenge: concomitant medications.
Concomitant medications, any drugs a participant is taking concurrently with the investigational product, can include medications taken for pre-existing conditions as well as pain relievers, antibiotics, or supplements.
While seemingly straightforward, the presence of concomitant medications introduces a complex web of interactions and variables that can significantly impact trial outcomes.
The generalisability gap
To mitigate the complexities of concomitant medications, clinical trials often employ strict exclusion criteria, barring patients on certain medications.
At a time when patient recruitment and retention remain critical challenges for the industry (80% of clinical trials fail to meet patient enrollment timelines), the rejection of potential participants can be a source of frustration.
And while barring patients with concomitant medications helps control variables, it introduces a new challenge: limiting the generalisability of trial findings to the broader patient population in real-world settings where people frequently take multiple drugs.
This creates a gap between trial results and clinical practice. What happens after a drug is approved and patients use concomitant medications in addition to their approved medication?
A complex landscape
Additional challenges posed by concomitant medications are diverse and far-reaching.
Drug-drug interactions (DDIs) are a primary concern. When multiple drugs are present in the body, they can interact in various ways, altering absorption, metabolism, distribution, and excretion.
Concomitant medications introduce numerous confounding variables that can mask or mimic the effects of the investigational drug. It becomes challenging to definitively attribute observed changes in a patient's condition (whether positive or negative) solely to the study drug when other active pharmaceutical ingredients (APIs) are also at play.
Patients in clinical trials often have co-morbidities requiring multiple medications. This inherent heterogeneity, while reflecting real-world patient populations, makes it harder to isolate the effect of the investigational drug.
Monitoring adherence to both the investigational drug and all concomitant medications can be incredibly difficult. Patients might forget to take a concomitant medication, or self-adjust dosages, introducing further variability into the study, making it challenging to interpret observed outcomes accurately.
The role of real-world data
Given these challenges, how can clinicians better understand and manage the impact of concomitant medications throughout studies – and once a drug moves beyond the controlled trial environment?
This is where real-world data (RWD), information from a variety of sources outside of traditional clinical trials, including electronic health records (EHRs), claims data, and wearable devices (data generated by patients themselves through wearables and mobile apps), emerges as an invaluable resource.
In particular, active longitudinal records – information collected over time from the same patients, outside of the context of a clinical trial – can help clinicians analyse concomitant medication data to identify issues that were not previously reported as adverse events (AEs) or explain AEs that were caused by drug interactions.
RWD can help researchers in addressing a range of issues
RWD can help researchers to identify previously unseen drug interactions. Clinical trials, due to limited sample sizes and controlled conditions, may not capture all possible DDIs. RWD, with its vast and diverse patient populations, can reveal rare or unexpected interactions that only become apparent when a drug is used more widely. By analysing patterns of AEs in patients taking specific combinations of drugs, researchers can identify potential DDIs that were missed in earlier development stages.
RWD can also help in understanding real-world effectiveness and safety. RWD provides insights into how a drug performs in patients with multiple comorbidities and medications. This directly addresses the generalisability gap created by strict trial exclusion criteria. Clinicians can better understand the true safety and efficacy profile of a drug when used alongside common concomitant medications in diverse patient groups.
By analysing large datasets, researchers can quantify the impact of specific concomitant medications on patient outcomes, such as treatment response, disease progression, and AE rates. This allows for a more nuanced understanding of how these co-prescriptions influence the overall effectiveness and safety of a new drug.
Insights gained from RWD analyses can also directly inform clinical guidelines and best prescribing practices. For example, if RWD consistently shows a negative interaction between a new drug and a commonly used concomitant medication, guidelines can be updated to recommend alternative therapies or closer monitoring for patients on that combination.
By leveraging machine learning (ML) and artificial intelligence (AI) on RWD, researchers can develop sophisticated predictive models to identify patients at higher risk of adverse drug reactions due to concomitant medications. This can enable proactive interventions and personalised treatment strategies.
RWD significantly enhances pharmacovigilance efforts by providing a broader and more continuous surveillance of drug safety in the post-marketing phase. This allows for the detection of AEs, including those related to concomitant medications, that may not have been observed during clinical trials.
The art of putting RWD in action
There’s great value in real-world data to gain insight to the impact of concomitant medications.
But the task of managing and assembling predominantly unstructured real-world data into a format that supports analysis and real-world evidence (RWE) generation is in itself a complex challenge.
For example, how do you derive clarity from data originating in scores of different EHR systems, contributed by tens of thousands of healthcare providers, across numerous speciality fields, from millions of patients?
Achieving this feat calls for a clinician-first approach, with teams of medical professionals, data scientists, technicians and specialists working together, assisted by AI, including Natural Language Processing (NLP) and ML. It takes experience and skill to extract, organise, and curate the data in a manner that ensures its underlying integrity.
Only then can curated data allow researchers to make the necessary inferences into patient populations, treatment patterns, and outcomes, including identifying and characterising concomitant medication use within specific patient groups for particular disease areas.
Better results in complex realities
Applying RWD to generate RWE about concomitant medications can have a range of significant impacts. These include informing clinical study design, enhancing the execution of clinical trials and generating robust evidence across the drug or device lifecycle, including post-market safety surveillance where concomitant medications are a key factor in assessing real-world therapy performance.
While concomitant medications present inherent challenges in the design and interpretation of clinical trials, their real-world presence is undeniable. RWD offers a powerful tool for gaining insights into how new drugs interact with other medications in diverse patient populations.
By embracing RWD, clinicians can gain a more comprehensive understanding of drug safety and efficacy, ultimately leading to more informed prescribing decisions and improved patient outcomes in the complex reality of polypharmacy.
