Abstract Rationale Indeterminate pulmonary nodules (IPNs) provide a diagnostic dilemma for clinicians attempting to distinguish benign from malignant nodules. Current risk prediction tools are imperfect, which can result in delays in diagnosis and unnecessary invasive workups. There is growing interest in the development of non-invasive biomarkers to aid in the diagnosis of IPNs. We undertook this systematic review to rigorously identify non-invasive molecular biomarkers designed to provide risk stratification for pulmonary nodules. Methods A systematic review of molecular biomarkers for malignancy risk stratification of IPNs was conducted in accordance with PRISMA guidance. PubMed, CINAHL Complete, and Scopus were searched from date of inception through December 27, 2024. English language restrictions were applied. References were screened using Covidence. Studies eligible for inclusion evaluated non-invasive molecular biomarkers for the evaluation of pulmonary nodules, defined as lesions 30mm, in adults. Only studies that assessed clinical validation or clinical utility were included. Risk of bias assessment was performed using the QUADAS-2 tool. Results Database and citation searching identified 1,577 eligible studies between 2007 and 2024. After abstract and full-text screening, 106 studies were included in the review. Less than half of studies validated the biomarker in an external cohort (n = 48), while of the remaining studies the majority (n = 47) used a split-validation design. The size of the validation cohort ranged from 29 to 1,679 with a mean of 177 and median of 112.5. Protein biomarkers were the most commonly evaluated (n = 58), followed by genetic biomarkers (n = 39) (including RNA and micro-RNA (miRNA) (n = 18), cell-free DNA (cfDNA) (n = 14), and circulating genetically abnormal cells (CACs) n = 8), and auto-autoantibody biomarkers (n = 15). Fourteen studies reported an AUC 0.9 and 42 studies reported either a sensitivity or specificity 90%. Utilizing the QUADAS-2 tool, only 9 studies were found to be at low risk of bias across all 4 domains, while many studies had an unclear risk of bias due to lack of reporting or clarity. Conclusions In conclusion, this review highlights the growing and diverse field of molecular biomarker development for the evaluation of pulmonary nodules. Although there have been many potential biomarker candidates identified, the vast majority remain in the early stages of development. Improvements must be made to the completeness of reporting in the field to address issues with risk of bias and applicability of study results. This abstract is funded by: None
Long et al. (Fri,) studied this question.