Abstract 4244
Background
The Tumor Inflammation Signature (TIS) is an investigational use RNA expression assay on the NanoString nCounter Dx Analysis System, which provides a measure of tumor inflammation across multiple solid tumor types. TIS measures immune genes in tumors from multiple origins, and it is possible that inclusion of tissue-specific interferents, such as non-tumor lymphoid aggregates (NTLA) could influence TIS performance. Here we describe the validation of the reproducibility of the TIS assay starting from FFPE tissues and robustness of the TIS across 8 potential tissue interferents.
Methods
TIS includes both review of an H&E tumor slide by a Pathologist and sample processing of unstained slides by an assay user. Analytical validation of the reproducibility of the TIS assay from 3 different Pathologists and 3 different assay users was performed using at least 10 patient specimens for 11 different tumor types (>110 independent tumors tested) from excisional and core biopsies. The robustness of the TIS assay to potential tissue interferents (genomic DNA, adjacent non-tumor tissue, NTLA, mucin, hemorrhagic, necrotic, and fibrotic tissue) was also assessed.
Results
The assay was validated to be reproducible with ≥ 95% concordance in assay results between independent pathologists. The total standard deviation of the TIS score was less than 2% of the score range from tissue including different pathology review and users. The interference studies demonstrated that the presence of mucin, necrotic, hemorrhagic and fibrotic tissue did not influence TIS results. However, if not properly removed, contamination with large concentrations of genomic DNA, non-tumor tissue, and NTLA can reduce biomarker concordance by increasing the TIS score.
Conclusions
The analytical performance of the NanoString TIS assay has been validated to be reproducible between users and pathologists when potential interferents are removed as instructed in the assay procedures. TIS is well suited for decentralized clinical testing and use as a potential biomarker to enrich for patients based on their inflamed phenotype across multiple tumors.
Clinical trial identification
Legal entity responsible for the study
NanoString Technologies, Inc.
Funding
NanoString Technologies, Inc.
Editorial Acknowledgement
Disclosure
B. Wallden, S. Church, S. Zimmerman, S. Popa, A. Sullivan, C. Ngouenet, E. Harris, N. Dowidar, A. Bergdahl, S. Ferree: Employee, Stockholder: NanoString Technologies, Inc. I. Pekker, P. Danaher: Employee, Stockholder at the time of the study: NanoString Technologies, Inc. C. Schaper: Paid consultant: NanoString Technologies, Inc.