Validation of the Nanostring of the Banff Human Organ Transplant Gene Panel in Archival Data from Human Kidney Transplants and its Modelling Complexity

Background: RNA gene expression of renal transplantation biopsies is commonly used to identify rejection. Mostly done with microarrays, seminal findings describe and define the patterns of genes associated with types of rejection and non-rejection kidney allograft diagnoses. To make gene expression more accessible for pathology laboratories, the Molecular Diagnostics Working Group of the Banff Foundation for Allograft Pathology and NanoString Technologies partnered to create the Banff Human Organ Transplant Panel (BHOT), a gene panel set of 770 genes as a substitute for microarrays (~ 50,000 genes). The advantage of this platform is that gene expressions are quantifiable on formalin fixed and paraffin embedded archival tissue samples. This new technology, thus, makes gene expressions cheaper and accessible to more laboratories and investigators. The purpose of this report is to validate the BHOT panel as a surrogate for microarrays and test the accuracy of the modelled BHOT data. Results This limited NanoString gene set readily identifies renal rejection and non-rejection diagnostic patterns using in silico statistical analyses of seminal archival databases derived from renal transplant RNA expression arrays. Multiple modelling algorithms show a highly variable pattern within the error matrices per sample. The discrepancies within the error matrices are most likely related to the gene expression heterogeneity of samples within a given pathological diagnosis. This was confirmed by clustering the data into 8 groups, which modelled with fewer misclassifications. Conclusion This report validates gene expression of human renal allografts using the Banff Human Organ Transplant Panel as a surrogate for microarrays and confirms the its modelling complexity.