Boston University Chobanian & Avedisian School of Medicine has announced a major research funding award for a multi-institutional team focusing on early lung cancer detection. Joshua D. Campbell, PhD, associate professor of medicine at the medical school, secured a five-year, $3.4 million U01 grant from the National Cancer Institute’s Human Tumor Atlas Network (HTAN), which operates under the National Institutes of Health. The funded project carries the title “Integrative biomarkers to improve clinical management of lung ground glass nodules.” During the first phase of HTAN funding led by A. Spira, Campbell and fellow researchers successfully established the Lung Pre-Cancer Atlas, which mapped the cellular characteristics of lung cancer in its earliest stages. This new Phase II grant allows the team to take those foundational insights and translate them into actionable, clinically useful biomarkers.

Lung adenocarcinoma ranks as the most common form of lung cancer and continues to claim lives worldwide despite continuous medical progress in smoking cessation, early screening methods, targeted treatments, and immunological therapies. When low-dose CT scans reveal hazy spots known as ground glass nodules, medical teams face a difficult clinical dilemma. These spots can serve as early warning indicators for lung adenocarcinoma, yet figuring out which spots will resolve independently versus which ones will progress to invasive cancer remains extremely difficult. Misinterpreting these findings creates serious risks, ranging from unnecessary, invasive surgeries for benign abnormalities to missed opportunities for timely, life-saving interventions.

To eliminate this uncertainty and improve early detection, Campbell brought together a diverse group of experts spanning pulmonology, pathology, cancer biology, imaging science, and data science. The collaborative network includes researchers from Boston University, Roswell Park Comprehensive Cancer Center, the University of Colorado Anschutz, and the University of California, Los Angeles. Their work applies new robotic-assisted tools capable of safely reaching and sampling tiny, hard-to-access nodules located deep within human lung tissue. For those looking into admissions news or academic developments, applying robust Ivy League strategies to complex research frameworks highlights how cross-institutional team science drives modern medical breakthroughs.

According to Campbell, who also directs the Bioinformatics Program and serves on the faculty of Computing & Data Sciences at Boston University, the project combines advanced robotic biopsies with artificial intelligence to analyze scan images alongside high-resolution molecular tissue mapping. This approach creates an accurate method for predicting whether a nodule is cancerous or likely to grow, even when initial biopsy results are inconclusive. The work aims to change early-stage lung cancer management so that patients with aggressive tumors get immediate care while healthy individuals avoid the physical and psychological toll of unnecessary procedures.

The funding specifically targets part-solid ground glass nodules, which regularly appear on low-dose CT screening scans but pose major diagnostic hurdles. The research team will systematically evaluate emerging robotic-assisted bronchoscopy systems to determine how safely and accurately they can sample peripheral part-solid nodules. A biopsy achieves success when it retrieves enough tissue for a definitive answer identifying cancer, precancer, or benign tissue. The proportion of biopsies yielding such answers is known as diagnostic yield, serving as the main metric for evaluating biopsy tools. Because part-solid nodules are small, hazy, and less dense than solid nodules, they are difficult to reach and sample, often leading to inconclusive results that require repeat procedures or surgeries.

Interventional pulmonologists at Roswell Park Comprehensive Cancer Center in Buffalo, New York, led by co-principal investigator Nathaniel Ivanick, MD, along with specialists at the Chobanian & Avedisian School of Medicine and Boston Medical Center led by co-investigator and Assistant Professor Ehab Billatos, MD, will collect one of the largest cohorts of part-solid nodule biopsies to date. This collection enables the team to evaluate the diagnostic yield of different robotic-assisted bronchoscopy platforms. Every single biopsy will undergo review by co-principal investigator Daniel Merrick, MD, a pathologist at the University of Colorado Anschutz who specializes in early-stage lung cancer.

Furthermore, William Hsu, PhD, at UCLA will lead automated quantitative image analysis to extract advanced radiomic features from CT scans. This analysis aims to discover which specific characteristics of the nodule, surrounding lung tissue, and overall lung structure best predict successful biopsy diagnoses. Knowing these factors in advance will help physicians decide which nodules require biopsy, which instruments to select, and how to reach a definitive diagnosis through fewer procedures.

The collected cohort biopsies will then go to a biospecimen team led by co-principal investigator Sarah Mazzilli, PhD, assistant professor of medicine at Boston University and Director of the BU Spatial Biology Core. These biopsy samples will undergo spatial transcriptomics profiling, a technology that maps gene activity across different histological regions within a tissue sample. Researchers will combine these molecular profiles with CT imaging data and artificial intelligence risk scores to construct predictive models capable of diagnosing cancer from inconclusive biopsies and identifying nodules likely to progress.

Mazzilli notes that because every biopsy sample is tiny, every single piece of tissue must be utilized effectively. The team established a specialized pipeline to optimally preserve, process, profile, and map gene activity at near single-cell resolution. This allows researchers to pinpoint early cancer warning signs even when they remain indistinguishable under standard microscopy. Additionally, every generated profile feeds into a public atlas that researchers everywhere can access to deploy novel methods for understanding how lung cancer begins and how to target it effectively.

The Human Tumor Atlas Network functions as a National Cancer Institute initiative designed to build three-dimensional atlases detailing the dynamic cellular, morphological, and molecular features of human cancers as they evolve from initial precancerous lesions into advanced diseases. Readers can read the full policy/details and check the official announcement directly on the official portal.

News is prepared from public university announcements for general guidance. Policies change frequently; confirm current details with each university admissions office.