Antimicrobial resistance remains a persistent public health challenge in the U.S., contributing to 35,000 deaths each year, costing billions in healthcare spending, and leading to patients suffering from drug-resistant skin infections, community-acquired bacterial pneumonia, and other illnesses with limited treatment options. In the effort to deliver life-saving antibiotics to those in need, Ran Li has provided critical technical expertise, transforming fragmented trial data into actionable insights, helping streamline regulatory pathways, and accelerating the delivery of important treatments. Her work, grounded in rigorous research and clinical experience, is helping to transform how the U.S. pharmaceutical industry addresses one of its greatest hurdles: translating raw data into reliable, FDA-ready evidence.
As a key technical contributor in clinical data science, Li has played an important role in the U.S. development of a novel aminomethylcycline antibiotic, engineered to target drug-resistant Gram-positive pathogens, a significant unmet need in clinical care. Her expertise spans Phase I-III clinical trials and New Drug Application submissions, focusing on two key conditions: acute bacterial skin and skin structure infections and CABP, which affect millions of Americans annually. For Li, the work is directly impactful: every dataset cleaned, every report refined, brings patients closer to accessible, effective treatments.
Multi-center clinical trials are notoriously complex, involving dozens of sites, conflicting data formats, mismatched units (mmol/L vs. mg/dL), and inconsistent decimal standards that can delay progress or introduce errors. Li addressed this industry pain point head-on by building a standardized data processing framework. Her team’s structured SDTM/ADaM datasets, integrated ISS/ISE reports, and regulatory-compliant Tables, Listings, and Figures (TLFs) not only meet FDA and CDISC standards but also form the backbone of the drug’s NDA eCTD submission, providing regulators with the transparent, traceable data needed to make timely approval decisions.
For late-stage CABP trials, Li tailored her approach to handle the complex data of critically ill patients, synthesizing clinical symptoms, lab results, and microbiology tests to support the drug’s safety and efficacy profile. She also extended her work to pediatric trials, developing child-specific data collection and processing protocols that helped secure the drug’s pediatric indication, addressing a longstanding gap in treating drug-resistant infections in children.
What sets Li apart is her ability to bridge academic innovation and real-world application. In two key studies, she has tackled pressing industry challenges:
In A Rule-Based Framework for Automated Table, Listing, and Figure (TLF) Generation in Clinical Study Reports, Li developed a rule-based system that automates TLF creation, a core requirement for FDA-mandated clinical reports. Tested on 5,248 patient records and 137 adverse event variables from a Phase III oncology trial, the framework generates 96 TLFs in 3.2 minutes, about 68% faster than manual scripting, with a 24.6% improvement in template matching accuracy and an error rate below 1.3%. For U.S. drugmakers, this translates to reduced labor costs and shorter development timelines.
Her second paper, Machine learning-enhanced clinical data anomaly detection: improving pharmaceutical data quality, targets hidden data flaws that could undermine trial integrity. By combining Autoencoder and Isolation Forest algorithms, Li’s AI model identifies numerical inconsistencies, logical conflicts, and missing fields in multi-center datasets (35 structured fields, 42,000 samples) with an average F1-score of 0.875 and AUC of 0.93, outperforming traditional methods. This allows for catching critical errors early, minimizing costly rework, and helping to ensure trial results are trustworthy.
These research breakthroughs are practical tools. Li translated her frameworks into custom SAS macros that integrate directly into clinical trial workflows, automating repetitive tasks and flagging anomalies in real time. The result: a 30%+ improvement in quality control efficiency, zero major data errors, and on-time delivery of all regulatory submissions. Colleagues refer to her as a “data steward”, a nod to her role in keeping trials on track and data reliable.
Li’s impact extends beyond a single drug. She collaborates closely with machine learning and cybersecurity experts to adapt cutting-edge technologies to clinical data’s unique demands, balancing innovation with FDA rules and industry best practices. Her scalable, replicable approach has become a reference point for U.S. clinical research, proving that efficiency and compliance can go hand in hand.
In an era where drug-resistant pathogens evolve faster than ever, Ran Li’s work reflects remarkable precision and clear practical purpose. She not only processes complex clinical data but also develops structured data workflows that strengthen U.S. capabilities in addressing antimicrobial resistance. For patients, pharmaceutical companies, and public health stakeholders, Li’s professional contributions demonstrate that targeted technical expertise, grounded in real-world clinical needs, can help generate meaningful progress in healthcare.
