Phesi CEO Warns of Impending Phase 3 Trial Failures Due to Site Oversaturation and Complex Design
核心洞察
Phesi (搜索) analysis reveals Phase 3 trials across multiple indications face avoidable failure due to overly complex design and lack of real-world data integration.
Atopic dermatitis (搜索) trials exemplify the crisis, with recruiting sites surging 205% in four years while one California investigator manages 14 concurrent AD studies.
AI-powered data analytics could simplify trial metrics and improve site selection, with EASI (搜索) and %BSA (搜索) showing near-perfect correlation in AD patient populations.
Clinical development companies risk wasting billions of dollars on Phase 3 trials that are destined to fail due to overly complex study designs and inadequate use of real-world data, according to new analysis from Phesi (搜索), a global provider of patient-centric data analytics. Dr. Gen Li, CEO and Founder of Phesi, warns that investigator site oversaturation and increased patient burden are creating systemic problems across multiple therapeutic areas.
The company's latest report, generated using its AI-powered Trial Accelerator™ platform, focuses on atopic dermatitis (搜索) (AD) as a case study of broader industry challenges. Analyzing industry-sponsored Phase 3 trials recruiting patients in June 2024, Phesi (搜索) found that AD ranks second highest in active trials after non-small cell lung cancer (搜索), with 18 ongoing studies despite affecting more than 200 million people globally.
Site Oversaturation Reaches Critical Levels
The analysis reveals alarming trends in investigator site utilization. The number of recruiting AD trial sites has increased 205% in just four years, rising from 1,944 sites in July 2021 to 3,981 sites in June 2025. This dramatic expansion exceeds global clinical development capacity for the indication.
Phesi (搜索) identified extreme examples of site oversaturation, including an investigator in Fountain Valley, California—a city of just 55,000 inhabitants—currently recruiting patients for 14 concurrent AD trials. Such concentration raises concerns about feasibility and data quality while highlighting the industry's failure to optimize site selection strategies.
"Phase 3 trials are high stakes for clinical development businesses. Late-phase programs are time consuming and involve significant investments, so carry considerable financial risk," commented Dr. Li. "Sponsors are under increasing pressure to deliver cost-effective therapies faster but risk failing against their commercial objectives if they continue to neglect real-world data and AI-driven clinical data science to support the design, execution and interpretation of trials."
Opportunities for Trial Simplification
Phesi (搜索)'s Digital Patient Profile (DPP) for adult AD reveals significant opportunities to streamline trial design without compromising scientific integrity. The analysis found near-perfect correlation between EASI (搜索) (Eczema Area and Severity Index) and %BSA (搜索) (percentage of affected body surface area) in the patient population.
While EASI (搜索) represents a complex scoring system involving multiple factors, %BSA (搜索) offers a simpler disease measure that could reduce patient burden while maintaining scientific rigor. This finding suggests that sponsors could simplify endpoint selection and measurement protocols to improve patient experience and trial efficiency.
"With so many active studies, there are exciting new treatments and modalities in development for AD. For sponsors, being first to market in a competitive space is critical – so designing smarter trials that mitigate recruitment issues and optimize endpoints will increase chances of success," Dr. Li explained.
Industry-Wide Implications
Jonathan Peachey, Phesi (搜索)'s Chief Operating Officer, emphasized that the challenges extend beyond atopic dermatitis (搜索) to all therapeutic areas, including rare diseases. "Failing to integrate clinical data sciences in late-phase studies is a significant missed opportunity to make trials smarter, and to increase the probability of technical and regulatory success," he stated.
The report identifies multiple areas where AI-driven approaches could improve trial outcomes, including reducing enrollment delays caused by investigator site saturation, minimizing sluggish site activation by contract research organizations, and addressing misalignment between feasibility data and actual site performance.
"The potential of AI in clinical development is huge – from precise scenario and prediction modelling, developing accurate digital twins, reduced complexity in trial design, and precision in selecting lead enrolling countries and investigator sites," Peachey noted. "The time is now for sponsors to adapt and evolve their approach to strategic and operational feasibility."
Data-Driven Solutions
Phesi (搜索)'s Trial Accelerator™ platform contains contextualized global data on more than 200 million patient records across over 4,000 indications. The company's Digital Patient Profiles incorporate key demographics including age, sex, comorbidities, outcome measures, and concomitant medications to help sponsors understand patient populations and design more effective protocols.
From these profiles, Phesi (搜索) derives Patient Access Scores, which enable sponsors to precisely select optimal investigator sites across 195 countries while eliminating non-active, non-enrolling locations. This approach could significantly improve recruitment efficiency and reduce the risk of late-stage trial failure.