Development of the molecular diagnostic (MDx) DLBCL Lymphoma Subtyping Test (LST) on the nCounter Analysis System.

Abstract

8536 Background: Diffuse large B-cell lymphoma (DLBCL) is an aggressive non-Hodgkin’s lymphoma with two distinct molecular cell-of-origin (COO) subtypes known as germinal center B-cell (GCB) or activated B-cell (ABC). DLBCL subtypes have been reported to be prognostic and potentially predictive of treatment benefit, underscoring the need for a precise and accurate MDx test. NanoString’s LST is based on the Lymph2Cx gene expression profile (GEP). These studies describe the development of the LST and analytical robustness and clinical accuracy of the Linear Predictor Score (LPS) and DLBCL subtypes (ABC and GCB). Methods: 51 banked formalin fixed, paraffin embedded (FFPE) DLBCL specimens were used to recalibrate the algorithm using a clinical grade assay. The subtype accuracy of the final locked algorithm was verified by testing 68 independent specimens with gold standard (GS) GEP results. Analytical precision was measured across 2 operators and 3 reagent lots by testing 10 FFPE DLBCL RNA samples. Reproducibility was measured across 2 operators and pathologists by testing replicate tissue sections from 64 FFPE DLBCL blocks following independent pathology review of H&E slides. Following guided macrodissection of pathologist identified tumor tissue, isolated RNA was tested on the NanoString nCounter system. The assay was evaluated across the assay RNA input range (62.5-1000 ng) and with the inclusion of adjacent non-tumor tissue. Results: The estimated misclassification rate compared to the GS result was 5.9% (95% CI: 4.6%-7.1%). The total standard deviation in LPS was < 2% of the score range, including all sources of assay variation, with no significant differences between operators or reagent lots. Average LST subtype concordance with independent pathology review was > 95% with no GCB to ABC misclassifications (or vice versa). The assay was robust across the specified range and against the inclusion of tissue interferents. Conclusions: The NanoString LST is a highly precise and accurate method for determining the COO from FFPE DLBCL tissue. The assay is well suited to clinical applications and is being used to select ABC patients for a Phase III study investigating lenalidomide (REVLIMID) in DLBCL.