Abstract
Distinguishing between multiple primary lung cancers and intrapulmonary metastases is crucial for staging, therapeutic planning, and prognosis. Traditional histological assessment provides a foundation for diagnosis, which can be limited when tumors showed identical or similar histological types. This systematic review and independent validation study aimed to evaluate the performance of next-generation sequencing (NGS)-based molecular algorithms alongside histological methods for the classification of multiple lung adenocarcinomas (MLAs). We conducted a literature search to identify relevant studies and selected algorithms for validation using a cohort of patients with MLAs. Our analysis included 27 patients with MLAs and compared histological assessment using Martini and Melamed criteria and comprehensive histologic assessment combined with a low-grade lepidic component (CHA & lepidic) with NGS data. We found a high consistency between CHA & lepidic and NGS-based diagnoses, although some discrepancies remained, particularly in cases with no somatic mutations or distant metastases. NGS-based molecular algorithms offer a high degree of accuracy in determining the origin of MLAs, supporting or challenging histological diagnoses. However, histological methods remain valuable, especially when NGS data are inconclusive. This study underscores the complementary nature of histology and molecular diagnostics in the precise classification of MLAs.