ATCC的患者来源二维和三维癌症模型使转化肿瘤学成为科学界的现实

Background: The Human Cancer Models Initiative (HCMI) is a global effort led by the National Cancer Institute (NCI) to advance translational oncology through patient-derived cancer models. Traditional cell lines often fail to capture the complexity of human tumors, limiting their relevance in drug discovery. HCMI addresses this by generating biologically relevant 2-D and 3-D models from patient samples, emphasizing genomic fidelity and clinical relevance. ATCC contributes to the initiative by developing, manufacturing, and distributing these models worldwide. Over 300 models have been released across 28 tissue types, including colorectal, pancreas, brain, and esophagus as well as rare cancers such as Wilms tumor and Ewing's sarcoma. These models span diverse diagnoses, age groups, and racial backgrounds, supporting research into tumor heterogeneity and health disparities. Comparative analyses show strong alignment with The Cancer Genome Atlas (TCGA), retaining over 80% of oncogenic drivers and preserving key transcriptional and epigenetic features. HCMI models offer a robust platform for precision oncology, enabling drug screening, biomarker discovery, and personalized medicine. Results and discussion: To date, the portfolio includes 329 models spanning common and rare cancers, comprising 91 colon, 54 pancreas, 50 brain, and 37 esophagus. Diversity includes clinical stage, age, and racial representation, enhancing translational relevance. Genomic profiling confirms strong fidelity to patient tumors. TCGA comparisons show >80% retention of oncogenic drivers and high concordance for mutations like BRAF (SKCM), KRAS (PAAD), APC (COAD/READ), and TP53 (ESCA). Concordance exceeds 90% for several cancer types. Additionally, 95% of tumor-model pairs show significant similarity in DNA methylation, and >80% align transcriptionally, preserving epigenetic and transcriptomic features. Models include primary (61%), metastatic (31%), and recurrent (7%) tumors, with samples from varied racial and age groups. This heterogeneity supports research into tumor evolution, drug response variability, and disparities in cancer care. Conclusion: HCMI offers a next-generation resource for oncology research. By providing clinically relevant, patient-derived models with rich molecular annotation, it enables drug discovery, biomarker validation, and personalized therapy development, bridging the gap between preclinical studies and clinical outcomes.