Biomarkers in lung cancer diagnosis and bronchoscopy: Current landscape and future directions

Cancer Biomark 2025 AI 5 Explanations View Original
Original Paper (PDF)

Unable to display PDF. Download it here or view on PMC.

Plain-English Explanations
Pages 1-2
Overview: The Evolving Role of Biomarkers in Lung Cancer Care

The Lung Cancer Challenge With approximately 2 million deaths per year globally and only 23% five-year overall survival, lung cancer remains the deadliest cancer. Only 16% of cases are diagnosed at early (localized) stage - where survival climbs to 56-80% - making improved early detection a critical unmet need that biomarkers could help address.

What This Review Covers This comprehensive review examines biomarkers across the entire lung cancer care timeline: from risk assessment and screening eligibility to diagnosis of lung nodules, prognosis prediction, treatment response monitoring, and surveillance for recurrence after treatment.

Biomarker Development Framework The CDC's ACCE model provides a structured framework for validating biomarkers through 5 stages: discovery, analytical validation, clinical validation, clinical utility assessment, and implementation. Most lung cancer biomarkers are in early stages, with only a few reaching clinical use.

Focus on Bronchoscopic Biomarkers A particular emphasis is placed on biomarkers applicable in the bronchoscopic setting - where many lung cancer diagnoses are made via biopsy - especially for guiding management when initial biopsy results are inconclusive.

TL;DR: This review surveys the complete landscape of lung cancer biomarkers across risk assessment, diagnosis, prognosis, and surveillance, with a focus on clinically validated blood-based and bronchoscopic tests.
Pages 2-3
Biospecimen Sources: From Blood to Breath

Blood-Based Biomarkers Blood is the most accessible biospecimen and enables repeated sampling. Measurable molecules include autoantibodies (produced in response to tumor antigens), serum proteins and antigens (like carcinoembryonic antigen), circulating tumor DNA (ctDNA), circulating microRNAs, and complement pathway products. Each has distinct performance characteristics for different clinical applications.

Exhaled Breath Analysis Volatile organic compounds (VOCs) in exhaled breath reflect tumor metabolism and can be measured non-invasively. Over 3,000 VOCs have been linked to potential lung cancer associations, though individual VOC specificity is low - signatures of multiple VOCs together show more promise.

Bronchoscopic Specimens During bronchoscopy, samples including endobronchial brush specimens, bronchoalveolar lavage (BAL) fluid, and transbronchial biopsies can be analyzed beyond standard pathology. Bronchial epithelial gene expression in particular captures a 'field of injury' effect where airway cells throughout the respiratory tract show cancer-related gene expression changes.

Other Sources Sputum DNA methylation, urine metabolites, and airway epithelium gene expression represent additional biospecimen types under investigation, each with unique logistical advantages and biological windows into lung cancer biology.

TL;DR: Lung cancer biomarkers can be derived from blood (proteins, ctDNA, miRNA), exhaled breath (VOCs), bronchoscopic specimens (epithelial gene expression), and other sources - each providing complementary biological information.
Pages 4-5
Clinically Available Biomarkers for Pulmonary Nodule Evaluation

The Intermediate-Risk Nodule Problem Approximately 1.6 million incidental pulmonary nodules are identified annually in the U.S., with 5.2% representing lung cancer. Guidelines stratify nodules as low, intermediate, or high risk. Intermediate-risk nodules are the challenge - they cannot be confidently dismissed or immediately biopsied, yet guidelines on their management are least definitive.

Nodify XL2 (Biodesix) - Rule-Out Test This blood-based proteomics test measures the ratio of LG3BP and C163A proteins. The PANOPTIC trial showed that combined with the Mayo nodule risk calculator, Nodify XL2 achieved 97% sensitivity and 98% negative predictive value - making it a powerful 'rule-out' test to safely avoid biopsies in benign nodules.

EarlyCDT-Lung/Nodify CDT - Rule-In Test This autoantibody panel tests for 7 antibodies produced against tumor antigens. It has 98% specificity and 78% positive predictive value - a 'rule-in' test that, when positive, strongly suggests malignancy and warrants aggressive investigation.

Percepta Genomic Sequencing Classifier When bronchoscopy yields inconclusive biopsy results, the Percepta GSC can be applied to an endobronchial brush from normal-appearing mainstem bronchial mucosa. Using whole-transcriptome RNA sequencing, it reclassifies malignancy risk, reducing unnecessary second procedures in approximately 34% of non-diagnostic bronchoscopies.

TL;DR: Three FDA-cleared biomarkers are available for lung nodule evaluation: Nodify XL2 (rule-out), Nodify CDT (rule-in), and Percepta GSC (post-bronchoscopy risk reclassification) - each occupying a distinct clinical niche.
Pages 4-5
Biomarkers Across the Lung Cancer Care Timeline

Risk Assessment Most lung cancer risk calculators use demographic and smoking factors. Biomarkers could augment these calculators to better identify never-smokers (15% of NSCLC patients) or capture molecular risk beyond tobacco exposure - potentially guiding who is offered LDCT screening beyond current USPSTF criteria.

Treatment Response Monitoring Circulating tumor DNA (ctDNA) is well-established for monitoring response to targeted therapy in advanced NSCLC. Its ability to detect emerging resistance mutations in real time allows treatment adjustments before radiographic progression becomes apparent.

Recurrence Surveillance Post-treatment CT surveillance is recommended but has significant false positive and false negative rates for recurrence. Serial ctDNA monitoring could detect molecular relapse before CT findings appear, enabling earlier salvage therapy for potentially curable recurrences.

Prognostic Biomarkers Genetic markers like EGFR, KRAS, and p53 mutational status provide prognostic information - EGFR and ERRC mutations favor better prognosis while KRAS and p53 mutations indicate worse outcomes. This molecular stratification increasingly supplements traditional TNM staging.

TL;DR: Biomarkers serve different roles at each point in the lung cancer journey - from screening eligibility and nodule triage to treatment selection, response monitoring, and recurrence detection.
Pages 5-7
Challenges and Future Directions in Biomarker Development

The Diversity Gap Clinical validation has been performed predominantly in white populations, yet lung cancer disparities are severe - African American men have the highest lung cancer mortality, and Native Hawaiian individuals have high incidence. Biomarkers validated only in white populations may perform differently in underrepresented groups.

ctDNA Limitations in Early Stage Circulating tumor DNA shows excellent performance in advanced/metastatic disease but poor sensitivity (~15%) in stage I cancer. Small early-stage tumors shed minimal ctDNA into the bloodstream, and less invasive subtypes (lepidic pattern adenocarcinomas) may be particularly low ctDNA shedders.

Clinical Implementation Barriers Even validated biomarkers face implementation challenges: clinician unfamiliarity with test interpretation, patient anxiety about ambiguous results, cost-effectiveness uncertainty, and lack of clarity about how to integrate results into shared decision-making conversations.

Future Directions The authors anticipate expanding clinical use of existing validated tests (especially the Percepta GSC), development of BAL-based ctDNA testing that achieves higher sensitivity than blood-based approaches, and AI-assisted integration of multiple biomarker types into unified risk scores that guide clinical decisions.

TL;DR: Key challenges include limited diversity in validation cohorts, poor ctDNA sensitivity in early-stage disease, and implementation barriers; future directions focus on BAL-based liquid biopsy and AI-integrated multi-biomarker scores.
Citation: Open Access, 2025. Available at: PMC12288387.