Improving Early and Equitable Lung Cancer Detection With Sybil: Mary Pasquinelli, DNP, FNP-BC, CTTS, and Waqas Haque, MD, MPH

Motivation and the Promise of Early Detection

Dr. Pasquinelli has been in the field of oncology nursing for 40 years, and she sees patients throughout the care continuum from lung screening to diagnosis and treatment. Her work has been at the forefront of early lung cancer detection. Data from UI Health from the past 10 years has shown a decrease in stage IV lung cancer diagnoses from 60% to 40%, and an increase in stage I diagnoses from 20% to 40%. Her program’s data show significant improvements in early diagnosis, enabling more patients to be candidates for curative therapies.

Sybil: AI in Screening

Discussing AI risk prediction, Dr. Pasquinelli highlighted Sybil, the deep learning AI model developed by Massachusetts General Hospital/Massachusetts Institute of Technology and trained on 10 years of scans from more than 2,000 individuals in the National Lung Screening Trial (NLST). While the NLST consists of 92% White individuals, Dr. Pasquinelli pointed out that the population served at UI Health is 75% minority. “We know that African Americans have the highest incidence and mortality from lung cancer even though they have lower rates of cigarette smoking,” she says. “I thought Sybil was a game changer, but I wanted to be sure that AI models work for all populations.”

UI Health joined a Sybil Consortium along with several other institutions and performed a validation study, which found that Sybil is indeed effective in a diverse minority population. The area under the curve (AUC) was 0.94 at one year and 0.79 at six years. “For context, an AUC of 1 means a perfect ability to distinguish a high-risk case from a low-risk case,” Dr. Pasquinelli explains. “An AUC of 0.50 is a flip chance, above 0.8 is considered strong, and above 0.9, excellent,” situating Sybil’s one-year discrimination as excellent and six-year prediction as very good.

Integrating Liquid Biopsies and Alternative Modalities

Beyond low-dose computed tomography (CT), Dr. Pasquinelli sees promise in blood-based biomarkers and ctDNA. “We’re using ctDNA a lot after patients are diagnosed; we can actually see a rise even before changes are seen radiographically.” For global settings, AI on chest X-rays could expand access: “Technology has changed; maybe in the future, that will hold some promise,” particularly in low- and middle-income countries without broad CT capacity. She also underscores multivariable tools like PLCO_M2012, which has been shown to decrease disparities in lung cancer detection between men and women and between different races. Exposure-agnostic models like Sybil may capture complex risk-prediction determinants such as air pollution, radon, and secondhand smoke.

Programmatic Improvements

Operational design has proven pivotal in screening. Dr. Pasquinelli notes that her team has optimized their program to address barriers from referral to scheduling. They have a screening coordinator to help people get scheduled and have changed ordering protocol to enable program-led order placement. As a result, adherence rose from 40% to over 85%.

Global Priorities and Equity in Screening

As a newly elected member of the International Association for the Study of Lung Cancer (IASLC) Board of Directors, Dr. Pasquinelli highlighted three pillars: “collaborative science, global education, and promoting accessible multidisciplinary care.” Despite momentum—with more than 100 abstracts on screening at this year’s World Conference on Lung Cancer—the United States Preventive Services Task Force (USPSTF) criteria remain restrictive: 50% of individuals who get lung cancer would not have met screening guidelines. Globally, tailoring to local resources is essential; in Kenya, some lung cancers had been misattributed to TB, underscoring context-specific pathways.

A Moonshot for Universal Lung Checks

Dr. Pasquinelli’s moonshot is universal access to a “lung check.” She explains 20% of people who never smoked are getting lung cancer, which is the fifth leading cause of cancer death. Reducing stigma via population-health approaches and leveraging AI, low-dose CT, blood biomarkers, or AI-enhanced chest X-rays—depending on resources—could broaden equitable detection. “I think in the next year or two, we’re going to see an amazing increase in research.”

Key Takeaways

• Early-stage shift is feasible: UI Health doubled stage I detection and reduced stage IV from 60% to 40%.
• AI can be equitable: Sybil performed strongly in a diverse cohort (AUC 0.94 at 1 year; 0.79 at 6 years), with similar performance in African American patients.
• Multimodal future: Low-dose CT plus ctDNA—and potentially AI on chest X-rays—can expand access and sensitivity.
• Operations drive impact: Proactive scheduling and protocolized orders can raise adherence from 40% to >85%.
• Eligibility gaps persist: Many lung cancers occur outside USPSTF criteria; multivariable risk tools can reduce disparities.

About the Speakers

Mary Pasquinelli, DNP, FNP-BC, CTTS, is the Director of the Lung Screening Program and the Co-Director of Lung Screening and Prevention in the Oncology Service Line at the University of Illinois Chicago (UIC). She is also the Lead Advanced Practice Nurse in Pulmonary and Medical Oncology and an Adjunct Clinical Instructor in the College of Nursing at UIC. Dr. Pasquinelli has been in oncology nursing for 40 years. Currently, she leads a comprehensive lung screening and research initiative that provides continuity of care from initial screening through diagnosis, treatment, and survivorship. Dr. Pasquinelli serves on the IASLC Board of Directors and is passionate about reducing barriers to care, advancing health equity, and improving outcomes for individuals at risk for or diagnosed with lung cancer.

Waqas Haque, MD, MPH, is a second-year Hematology/Oncology Fellow at the University of Chicago. He recently completed his Internal Medicine Residency at New York University (NYU) in a Clinical Investigator Track. Dr. Haque has balanced his patient care work with a variety of research projects. During his fellowship training at the University of Chicago, he plans to further his work in innovative clinical trial design, value-based care delivery to cancer patients, and clinical investigation.