Clinical Researcher—June 2024 (Volume 38, Issue 3)

PEER REVIEWED

Deepika Khedekar, MPharm

 

 

 

This article delves into the critical role of clinical trials in combating major health crises. It elucidates the multifaceted challenges impeding clinical trials, including patient recruitment, funding, and data management issues, while also highlighting the transformative potential of emerging technologies such as artificial intelligence (AI), blockchain, and genomic sequencing. Specifically, it examines the impact of AI in enhancing trial data analysis and patient recruitment rates, the pivotal role of blockchain in securing data integrity, the dynamic use of wearable devices for continuous patient monitoring, and the cutting-edge application of genomic sequencing in personalized medicine. Furthermore, it addresses the inherent challenges these technologies pose, including data bias, security concerns, and the need for regulatory compliance. Ultimately, the article underscores a hopeful yet cautious outlook towards leveraging these innovations to advance clinical research, emphasizing the importance of ethical considerations and patient-centric approaches in navigating the future of clinical trials.

Global Health Crisis and Clinical Trials Overview

Annually, 67 million lives are lost globally.{1} Of these, 9.9 million deaths are attributed to cancer{2} and 18 million to cardiovascular diseases.{3} It’s staggering, isn’t it? Clinical trials, essential in battling these diseases, often falter due to an array of challenges, including but not limited to patient recruitment, funding disparities, staff shortage, scaling data analytics capabilities, and more.

Eighty-five percent of all clinical trials experience delays, and each day lost in these trials costs pharmaceutical companies between $600,000 to $8 million in U.S. dollars. It’s not just about the capital resources. Did you know that less than 4% of the U.S. population participates in these life-saving trials? Moreover, keeping participants engaged is another story, with only 7% of those who start a study seeing it through to the end.{4} However, hope is on the horizon with artificial intelligence (AI), blockchain, and genomic sequencing poised to transform clinical research.

Transforming Clinical Trials with AI-Driven Large Language Models

AI-driven large language models (LLMs) are revolutionizing the field of clinical trials, specifically addressing the challenges of slow patient recruitment and intricate data analysis. Rice University researchers have pioneered the use of these AI models, which are adept at processing and understanding vast amounts of text data, thereby producing synthetic data that mirror the complexity and variability of real-world clinical trial data.{5} This innovation is particularly useful in overcoming the hurdles posed by the need for extensive data in the face of strict privacy regulations like the Health Insurance Portability and Accountability Act in the U.S. and the General Data Protection Regulation (GDPR) in the European Union, which rightly limit the availability of sensitive patient information.

The Rice University team’s approach is twofold. Initially, it feeds publicly available trial data into a LLM to create artificial yet realistic trial data for deep learning analysis. These data encompass a wide range of inputs, from clinical notes to patient demographics, effectively training the algorithm to recognize and adapt to the nuances of clinical data. Subsequently, the synthetic data are processed by a deep learning algorithm on a secure server, ensuring patient privacy while enriching the model with a comprehensive dataset for nuanced analysis. This novel approach, recognized by the American Medical Informatics Association, mitigates issues related to data shortage and increases accessibility to trials. It also addresses the demand for high precision in patient-trial matching in a manner that opens the door to a future where the patient-matching process in these clinical trials can be largely automated, significantly accelerating the pace of medical discoveries and the development of new treatments.

While AI has proven instrumental in addressing challenges related to patient recruitment and data analysis, clinical trials also face challenges in ensuring data integrity and security.

Blockchain to Guard Clinical Data

Data form the lifeblood of clinical trials. Data integrity and security are paramount. Any compromise on these fronts can jeopardize the trial’s outcomes and, more importantly, patient safety. Here, blockchain technology, with its decentralized and tamper-proof ledgers, offers a powerful solution. Blockchain ensures data security and transparency, making it an invaluable tool in clinical trials.

Several trailblazing programs are already harnessing blockchain’s potential. The Mayo Clinic, in partnership with the Dutch blockchain startup Triall, embarked on a mission in 2022 to refine clinical trial design and enhance study data management.{6} In a similar manner, Medidata joined forces with the University of Oxford’s Cancer Research U.K. Clinical Trials Unit.{7} Their shared goal? To leverage blockchain technology to amplify the integrity and transparency of clinical trial data.

While blockchain addresses the intricacies of data security and integrity, another challenge looms large in clinical trials: the need for continuous, accurate, and real-time patient monitoring and engagement. With the aforementioned low retention rate for participants in trials, traditional methods for patient monitoring and engagement, often reliant on intermittent check-ins or self-reporting, can miss crucial datapoints, leading to gaps in understanding patient health.

Wearable Devices: Revolutionizing Patient Monitoring

Addressing the gaps left by traditional monitoring methods, wearable devices emerge as the next frontier in clinical trials. These devices, from smartwatches to fitness trackers, provide a solution to the challenges of intermittent check-ins and inconsistent self-reporting. They offer the advantage of continuous monitoring, capturing an array of health data ranging from heart rates to sleep patterns, ensuring researchers have access to comprehensive, real-time insights.

A testament to the transformative power of wearables comes from an initiative of the Scripps Translational Science Institute,{8} which seamlessly integrated devices like smartwatches and fitness trackers into a clinical trial, allowing for continuous monitoring of patient activity levels and sleep patterns. This innovative approach provided a holistic view of patient health, yielding richer data for analysis.

Further emphasizing the potential of wearables is the Apple Heart Study,{9} a collaboration between Apple and Stanford Medicine. This study utilized the Apple Watch to detect irregular heart rhythms in an impressive cohort of 400,000 participants. This research not only validated the efficacy of wearable devices in large-scale remote monitoring, it also heralded a new era of patient-centric, data-rich, and efficient clinical trials.

While wearable devices address the immediate and tangible aspects of patient health, there’s a deeper layer of understanding that remains largely uncharted: our genetic blueprint.

Genomic Sequencing: The Frontier of Personalized Medicine

Our DNA is the blueprint of our existence, holding secrets to our health and lifespan. To truly revolutionize clinical trials and medical treatments, it’s imperative to delve deep into this genetic code. Genomic sequencing emerges as a critical technology in this exploration, offering profound insights into an individual’s genetic makeup by allowing researchers to pinpoint specific genetic variations or mutations associated with diseases or treatment responses. A prime example is the National Cancer Institute-Molecular Analysis for Therapy Choice (NCI-MATCH) trial,{10} which harnessed genomic sequencing to align patients with advanced cancers to targeted therapies based on their tumor genetics.

Similarly, the 100,000 Genomes Project{11} in the United Kingdom aims to decode the genomes of 100,000 individuals. The objective? To unearth disease-causing genetic mutations and pave the way for tailored treatment strategies.

As we harness the power of our DNA to revolutionize treatments, we must also navigate the multifaceted challenges that arise with the integration of cutting-edge technologies in clinical research.

Challenges on the Horizon

While technological advancements have brought remarkable progress to the field of clinical trials, it is important to acknowledge that they are not without risks and limitations. The adoption of AI in clinical trials, for example, introduces concerns related to the reliability and interpretability of AI algorithms. There is a risk of bias in algorithmic decision-making if the training data used to develop AI models are not diverse or representative of the population under study. Vigilance is required to ensure that AI algorithms are transparent, explainable, and accountable, especially when making critical decisions regarding patient eligibility, treatment allocation, or adverse event prediction.

Similarly, while blockchain technology promises enhanced security and integrity in clinical trials, it also presents several challenges. Its decentralized nature, though ensuring data authenticity, can make updating data records extremely challenging at times. Further, despite its robust security, blockchain is not entirely immune to cyberattacks, which could have severe implications given the sensitive nature of clinical trial data. Additionally, as trial size increases, handling larger volumes of data could lead to scalability issues. Navigating these issues becomes even more challenging when considering compliance with data protection laws, such as GDPR.

As well, the use of wearable devices in clinical trials, though promising for patient monitoring, carries its own set of challenges. Data accuracy can be a significant concern, as the quality of data collected depends on the precision and reliability of the devices. These devices can also suffer from variability in measurements, making data interpretation challenging. Additionally, participants from rural areas may not have access to such devices, which means the trial team will need to provide these devices exclusively to these participants and the corresponding cost and training process will need to be factored into the trial design to ensure these trials are holistically inclusive.

Issues related to patient compliance and comfort with the devices must also be considered. For instance, a participant may forget to wear the device or use it incorrectly, impacting the data collected. Privacy and data security are other crucial aspects, as wearable devices generate large amounts of personal health data that need to be securely managed to prevent unauthorized access.

It is also imperative to address the potential for fraud, misconduct, and Type II errors as these emerging technologies are integrated into clinical trial design and operations. These considerations are essential for maintaining the integrity and reliability of trial outcomes. For example, the use of AI in data analysis and patient selection must be transparent and accountable, with checks in place to prevent bias and ensure fairness. Blockchain technology, while enhancing data security and traceability, requires careful implementation to prevent unauthorized access and ensure the authenticity of trial data.

Since these technologies are dependent heavily on devices and data, if either of those are either knowingly or unknowingly manipulated, the results may represent false and/or fraudulent outcomes hampering the integrity of the trial and raising risks for the health of the participants. Thus, the potential for fraud and misconduct requires stringent oversight and ethical guidelines to prevent manipulation of data or results.

Moreover, Type II errors are also an added concern for these trials. A Type II error occurs when a study fails to detect a true effect or benefit of a treatment or intervention, mistakenly concluding that there is no effect when, in fact, there is. This type of error poses significant risks in trials involving these technologies due to potential technological limitations or incorrect use. For example, if wearable devices are not accurately calibrated, they might fail to detect meaningful changes in a patient’s health status, leading to a false conclusion that a treatment is ineffective. Similarly, AI algorithms could miss significant patterns or outcomes if they are trained on incomplete or biased datasets, further compounding the risk of Type II errors.

Therefore, it is essential to establish robust protocols for the deployment of these technologies, such as rigorous testing of wearable devices for accuracy and reliability, and validating AI algorithms against diverse and comprehensive datasets to ensure they are capable of identifying true effects accurately. Additionally, clear protocols for data validation and result interpretation are essential to ensure that the data collected through these technologies are correctly analyzed and understood. By taking these steps, researchers can minimize the risks associated with integrating these technologies into clinical trials.

The Future of Clinical Trials: Innovations and Ethical Considerations

The integration of emerging technologies such as AI, blockchain, wearable devices, and genomic sequencing into clinical trials heralds a transformative era in clinical research. These advancements promise to enhance patient recruitment, streamline data analysis, secure data integrity, enable continuous patient monitoring, and pave the way for personalized medicine. However, the application of these technologies introduces complex challenges, including the potential for AI-induced biases, data rectification and scalability issues linked with blockchain, data accuracy and privacy concerns related to wearable devices, and the ethical considerations surrounding genomic sequencing.

Addressing these challenges requires a concerted effort to develop robust methodologies that ensure the equitable and unbiased application of AI, as well as scalable and secure data management frameworks that align with regulatory standards. Moreover, ensuring the accuracy and privacy of data collected through wearable technologies, alongside establishing ethical guidelines for the use of genomic information, is paramount. This necessitates a multidisciplinary approach that combines clinical expertise, data science, ethics, and regulatory knowledge, aiming to leverage technological advancements while safeguarding patient welfare and data integrity.

As the clinical trials landscape evolves, the successful integration of these technologies hinges on an unwavering commitment to ethical considerations and patient-centric approaches. The future of clinical research, while promising, demands vigilance, adaptability, and a comprehensive understanding of both the potential and the pitfalls of these emerging technologies. Embracing this future necessitates a holistic approach that not only navigates the balance between innovation and ethical responsibility but also ensures that advancements in clinical trials contribute to the broader goals of enhancing patient care, advancing medical science, and improving health outcomes globally.

References

1. How many people die and how many are born each year? Big Think.
2. Ferlay J, et al. 2021. Cancer statistics for the year 2020: An overview. International Journal of Cancer. doi:10.1002/ijc.33588. PMID:33818764.
3. Cardiovascular Diseases Fact Sheet 2021. World Health Organization.
4. Levy S. 2021. Patient-Centric Clinical Trials: The Acceleration Of Decentralized Patient Monitoring. Forbes.
5. Large language models could be the key to better patient-trial matching. 2023. Rice University, Department of Computer Science.
6. Mayo Clinic taps into blockchain technology for clinical trial design. 2022. Cointelegraph.
7. Medidata and Oxford University Innovation’s Clinical Outcomes Team Announce Addition of Gold Standard Questionnaires to eCOA Library. 2020. Oxford University Innovation.
8. Scripps Research invites public to join app-based DETECT study, leveraging wearable data to potentially flag the onset of viral illnesses. 2020. Scripps Research Initiatives.
9. Perez MV, Mahaffey KW, Hedlin H, et al. 2019. Large-Scale Assessment of a Smartwatch to Identify Atrial Fibrillation. N Engl J Med 381(20):1909–17. doi:10.1056/NEJMoa1901183.
10. Flaherty KT, et al. 2020. The Molecular Analysis for Therapy Choice (NCI-MATCH) Trial: Lessons for Genomic Trial Design. Journal of the National Cancer Institute 112(10):1021–9. doi:10.1093/jnci/djz245.
11. 100,000 Genomes Project, Genomics England.

 

Deepika Khedekar, MPharm, (deepika.khedekar@gmail.com) is a Centralized Clinical Lead with IQVIA Inc., Mumbai, India.