AI in clinical trials — the future is here

Clinical trials are the critical and necessary step to test and validate new treatments and medical devices, but their cost and complexity — particularly with enrollment — have grown exponentially.

With the explosion of technologies generating data about patients, clinical trial sponsors are contending with new challenges in managing and analyzing massive datasets. While eligibility criteria for clinical trial participants have remained relatively consistent, the volume of data generated from Phase III trials has tripled to 3.6 million data points from 10 years ago.

With a clear need for improved efficiency and modernization, AI in Life Sciences is revolutionizing these processes. Life sciences organizations are already experiencing the benefits of AI in clinical trials. For example, automating the most time-consuming aspects — such as matching and recruiting the right patients.

AI has the potential to enable more efficient clinical trials that ultimately allow patients to access new treatments faster. Life sciences organizations that need to maintain control over clinical trial data and patient interactions are turning to cloud-based platforms powered by AI to streamline the clinical trial journey.

AI's ability to improve patient matching accuracy and speed significantly reduces the time needed to recruit participants. Researchers use predictive modeling capabilities, a type of AI, to determine the optimal criteria when designing trials. They are also using the conversational power of AI agents to speed up trial enrollment and create personalized communication with trial participants.

With its ability to quickly analyze large volumes of structured and unstructured data, find patterns, and predict outcomes, the use of AI in clinical trials offers numerous benefits, including:

• Less time spent on site performance analysis: Because AI can swiftly analyze data and identify markers of a site's potential performance — such as enrollment rates, data entry timelines, data quality, and adherence to protocols — an AI-powered, data-first approach ensures the choice of the best site for each trial. AI models can be developed and refined based on historical and current site performance data, both structured and unstructured. This empowers study teams to quickly and easily assess risks associated with each site, understand why certain sites may exhibit risks, and pinpoint which sites are most likely to succeed.
• Data-driven site selection: AI enhances manual processes, eliminating guesswork and streamlining site selection and management throughout a trial. It enables you to focus on key information as needed, making it easier to identify and address studies or trials that aren't enrolling on schedule, are failing to meet diversity criteria, or are exhibiting data quality issues.
• Real-time visibility into trial progress and autonomous monitoring: Once sites are selected, they must be closely managed throughout the study to ensure the success of both the site and the study itself. For what was previously a time-consuming and labor-intensive process, predictive and generative AI tools now activate site data to help teams gain insight and identify risk markers. AI algorithms can provide continuous analysis of data from wearable devices and other sources to detect patient treatment adherence and other health indicators. Autonomous agents can also monitor health info in the background to identify and raise issues for the people leading the trials.
• Real time assistive collaboration: Study teams and site personnel can communicate with each other effectively and perform tasks collaboratively by using agentic AI with Agentforce to accelerate the study start-up and conduct processes.
• Targeted patient recruitment and enrollment: Enrolling the right participants for each trial can be challenging and may cause delays. AI patient matching offers study teams an intelligent, efficient, and accurate way to match prospective participants to eligible trials. For example, having AI pre-screen criteria helps teams automatically identify prospective trial participants and generate lists of patients who are ready to be contacted and recruited for clinical trials. This results in more precise matches and ensures patients are paired with the most appropriate trials.

AI is quickly gaining traction as a tool for extracting, organizing, and analyzing vast amounts of data from electronic health records (EHRs), medical claims, and other real-world data sources. Educating teams about the technology and preparing them to use it is important to ensure optimal results.

As more clinical research sponsors embrace AI technology, they seek AI tools and platforms that best meet their needs. To prepare for AI implementation in clinical trials, teams must standardize all available data from multiple sources — especially from clinical trial systems — to create a single unified data source. A data integration layer like Salesforce's MuleSoft unified integration and automation platform can provide tools to connect and map data to a unified clinical data model. Teams can use the power of MuleSoft to bring data into Data Cloud, which can automate data from external systems to create a common model.

Life Sciences Cloud is a secure end-to-end engagement platform that can help life sciences organizations provide clinical trial stakeholders with a single source of truth and insights about each trial.

For example, powered by AI, data, and automation, Life Sciences Cloud helps organizations streamline operations and personalize patient engagement and enable sites, sponsors, and CROs to collaborate seamlessly. It includes features that can help speed and simplify clinical trial management, from study participant recruitment to real-time support for sites and patients throughout the study lifecycle.

While today's life sciences organizations employ teams of data scientists and technical specialists who may have advanced AI knowledge, some researchers, clinicians, and other stakeholders may need more focused training to use AI tools effectively. To help study teams without extensive AI knowledge collaborate and interpret the results generated by AI systems — and prevent misinterpretations of results — it is essential to educate them about:

• How AI works and its potential impact on clinical research
• How to interpret and use AI-generated results
• How to identify potential biases
• How to integrate AI into trial design and study operations
• FDA guidelines, ethics committee guidelines, and other regulatory requirements

Study teams should also ensure that study participants understand how AI and other technologies will be used during the trial to create trust. They should demonstrate to participants the safeguards implemented to protect their data.

Despite being in the early stages of this technology, AI has already created a paradigm shift in how clinical research is conducted. For organizations considering adopting AI for clinical research, taking a systematic approach can help reduce the cost and complexity of research and ensure more efficient clinical trials.

Before launching AI in your clinical research, it's crucial to determine where it could play a key role. With its ability to collect and analyze large volumes of study data — including identifying the right patient populations, including identifying the right patient populations, selecting the right sites, and aligning protocols to ensure operational feasibility. AI transforms how clinical trials are conducted from start to finish.

Life sciences organizations are increasingly challenged to identify, engage, and retain the right candidates for enrollment in clinical trials. In fact, the National Library of Medicine reports that as many as 80% of clinical trials fail to meet participant enrollment targets . Most enrollment processes remain manual and are handled on a trial-by-trial basis, resulting in significant delays to market. This is why it's crucial to choose AI technology that can accelerate clinical trial processes. Here are some examples:

• Recruitment and enrollment. AI-powered customer relationship management (CRM) systems, such as Life Sciences Cloud, have enabled patient details, prescreening, and eligibility study criteria to automatically match eligible candidates to relevant clinical trials, followed by further screening as required by the study team. It can also streamline the enrollment process through customized e-consent and assessment forms. Tools like Agentforce can draw from structured and unstructured data, including demographics, diagnosis codes, and medication details, to identify the ideal clinical trial participants.
• Site selection. With their ability to analyze vast amounts of historical data to enable target questions that help assess a site’s ability to conduct a specific trial, AI tools can support the quick creation of site feasibility questionnaires development. Using analysis tools to analyze patient demographics, geographic locations, and past site performance, they can help teams identify the best potential sites.
• Streamlined workflows. Autonomous AI agents help clinical trial team members manage routine tasks and automate processes that provide personalized support throughout the trial lifecycle. These agents can function as a seamless, scalable workforce providing personalized support to clinical trial participants and sites at every touchpoint. This solution automates outreach for enrollment, follow-up appointments, and other key interactions.
• Data management. AI can help ensure data is collected and handled effectively during clinical trials. Data management tools help clinical research teams maintain clinical trial data accuracy and reliability throughout the trial process while ensuring compliance with regulatory standards to maintain integrity.

Choosing the right technology is critical to your success. The ideal AI platform for clinical trials should include tools that enable study teams to automate complex processes, from setting up and managing clinical sites to coordinating communication among contract research organizations (CROs) and sponsor personnel.

• It's crucial to look for a vendor that provides AI tools to automate site visit scheduling, offer visibility into trial progress and potential bottlenecks, and enable proactive management of the entire process. Life Sciences Cloud is a comprehensive platform featuring powerful AI tools that can enhance clinical trial use cases in important ways. For example:
• Data Cloud unifies data to enable candidate auto-matching and develop personalized portals that feature AI-curated content and chatbots for recruitment and onboarding.
• Agentforce AI capabilities simplify data analysis by matching eligible candidates to relevant clinical trials using patient details and study criteria. They not only enable personalized, timely interactions with trial participants, but they also provide insights and recommendations to encourage participant engagement.
• Built-in compliance — such as audit trails and electronic signature capabilities — help ensure regulatory compliance by automating compliance-related tasks.

A recent Forrester survey of decision-makers at health and life sciences organizations found that "selective long-term partnerships with vendors that have extensive capabilities and flexibility in applications is imperative" for reaching their expectations and goals in AI implementation.

AI can optimize patient recruitment, streamline data analysis, pinpoint potential safety concerns, and improve trial design with advanced pattern recognition. However, since the success of an AI implementation relies on the expertise of professionals to ensure optimal results, every investigator and study team member who will use the platform must have a clear understanding of AI and its capabilities.

When integrating AI in clinical trials, the onboarding process should ensure teams understand:

• AI limitations and ethical concerns. While AI can process data from diverse sources to identify potential trial candidates, it also has the potential for bias, lack of transparency, and issues related to privacy and ethics.
• The proper way to input data. Accurate data input ensures that your AI system can produce reliable results.
• How to interpret AI output. AI predictions are probabilities and should be reviewed alongside other clinical data.

AI is an important tool for ensuring the integrity of clinical trials and accelerating and streamlining clinical development. In addition to helping identify optimal trial sites, it can boost enrollment and enable proactive interventions to shorten trial timelines significantly. AI can cut the costs of crucial processes like developing trial documents and improving data quality and signal management.

By automating routine tasks, AI helps reduce operational costs and improve overall productivity. It can also bring the flexibility needed to adjust trial designs even while data is being collected. This can allow trial sponsors to detect treatment effects sooner and speed the delivery of effective treatments to the marketplace.

AI supports clinical trial progress monitoring by analyzing patient data from various sources (e.g., wearable devices, EHRs, and clinical assessments). For example, predictive modeling can help identify at-risk patients, reminders and educational content for personalized engagement can ensure compliance, and data from wearables can enable participant progress monitoring.

AI tools can monitor clinical trial progress in many ways, such as helping teams improve patient adherence to the study protocol by delivering personalized support. They can also analyze complex patient datasets to identify patterns and trends and enable digital diaries for patients to record real-time symptoms and side effects. Finally, they can analyze medical images to automatically detect abnormalities and assess disease progression.

As AI technology continues to integrate into clinical trials, life sciences organizations may benefit from new efficiencies and cost savings as well as face certain challenges when it comes to balancing the advantages and concerns with AI. Overcoming these obstacles will require adaptability and careful consideration when choosing the right AI tools for your organization.

For example, working with large data sets from diverse sources will require ongoing monitoring to assess the quality and relevance of that data. Implementing AI solutions will also mean new investments in expertise and technology, which can take time to deploy across your organization.

Researchers must also pay attention to the evolving regulatory landscape. While the FDA and other regulatory organizations encourage using AI to modernize the clinical trial process, they are also developing regulatory frameworks to ensure that AI usage is safe and effective. Organizations should pay close attention to compliance standards and be sure to choose trusted AI tools — like the Einstein Trust Layer — that prioritize data privacy and security.

AI is transforming how clinical trials are managed and helping life sciences organizations realize its potential for making clinical trial management more patient-centric and efficient. It holds great promise for key areas of clinical trials — from protocol development and patient recruitment to clinical operations analytics and enabling study team collaboration. AI adoption will drive transformational change and present a range of unique challenges. Working with a trusted, long-term partner with AI expertise can help life sciences organizations navigate these challenges, manage risks, ensure data security, protect their investments, and help them take full advantage of AI as it evolves.