The development and characterization of patient derived xenografts from non-small cell lung cancer brain metastases

Abstract

Introduction: NSCLC brain metastasis cell lines and in vivo brain metastasis models are not widely accessible. The purpose of this study was to establish and characterize a direct-from patient-derived xenograft (PDX) model of non-small cell lung cancer (NSCLC) brain metastases. Methods: Surgically obtained tissue was implanted subcutaneously and as orthotopic intracranial implants into immunodeficient mice. Histology and DNA loci were compared between original tumor and subsequent PDX passages. Tumors underwent RNA and DNA sequencing and relevant therapeutic targets were identified. Tumor growth rates were assessed following treatment with radiation, MEK inhibitor selumetinib, or MET inhibitor savolitinib. Cell lines were established. Results: Nine NSCLC brain metastases PDXs were established. Morphologically, strong retention of cytoarchitectural features was observed between original patient tumor and subcutaneous and intracranial tumors. Preservation of thyroid transcription factor 1 expression was seen in all xenografts. Short tandem repeat analysis demonstrated strong concordance between patient tumors and subsequent PDX passages. RNA sequencing analysis revealed high correlation between matched patient and PDX samples. Significant growth inhibition was shown with radiation, with selumetinib in tumors harboring KRAS G12C mutation and with savolitinib in tumors with MET exon 14 skipping mutation. Significant tumor growth delay was observed with the combination of radiation and savolitinib compared to radiation or savolitinib alone in the PDX harboring MET exon 14 skipping mutation. Early passage cell strains showed high consistency between patient and PDX tumors. Conclusion: We have established a robust human xenograft model system for investigating NSCLC brain metastases. These PDXs and cell lines show strong phenotypic and molecular correlation with the original patient tumors and provide a valuable resource for testing preclinical therapeutics.