Abstract 161P
Background
The MammaPrint (MP) 70-gene risk of recurrence and BluePrint (BP) 80-gene molecular subtyping tests are used in personalized treatment planning of early-stage breast cancer (EBC). MP and BP are offered both centrally, on the microarray (MA) platform and as decentralized solution based on a targeted Next Generation Sequencing (NGS) kit. Recently, we translated MP and BP to a Whole-Transcriptome (WT) -NGS based platform. Here we present the comparison between MA and WT-NGS based test results and report the performance on the equivalence.
Methods
A set of 151 Formalin Fixed Paraffin Embedded (FFPE) EBC samples was processed both on WT-NGS and on standard MA diagnostics platform. Libraries were prepared with the TruSeq Total RNA stranded 75 bp single-read protocol on a NextSeq machine. A set of 709 FFPE samples from patients with stage II-III HER2 positive EBC, partly treated were available for WT-NGS and MA processing to confirm the validation. Comparisons of the MP (High-, Low-risk) and BP (Basal-, Luminal-, HER2-type) results from WT-NGS and MA were based on the overall concordance and index correlation.
Results
WT-NGS MP results are highly comparable with the MA predicate [96% concordance, positive percent agreement (PPA)= 99% (95%CI: 93.5 – 99.8), negative percent agreement (NPA) = 93% (95%CI: 83.9 – 96.8)]. Similarly, high concordance of 96.7 % is observed between WT-NGS and MA BP results. MP and BP indices are highly correlating between MA and WT-NGS results (Pearson correlation coefficient () of 0.97 for MP, Basal, Luminal, and 0.88 for HER2). Concordance is also confirmed in the HER2 set (MP =97.9%, PPA = 99.1 % (95 % CI: 98.0 – 99.6), NPA = 81.3% (95 % CI: 68.1 – 89.8) and BP = 92.0%), with high index correlation = 0.92 for MP and Basal, = 0.97 for Luminal and = 0.81 for HER2).
Conclusions
WT-NGS test results are highly comparable with the predicate MA as also independently confirmed on a large set of clinical samples. The successful translation of the MA to an WT-NGS-based test highlights the robustness of the biology behind the MP/BP test, which prove to be reliable tool for personalized medical care, irrespective of the technology platform used.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Agendia NV.
Funding
Agendia NV.
Disclosure
R. Bhaskaran, Y. Bijl, D. Israeli, S. Jong-Raadsen, E. van montfort, A. Gallagher, T. Piel, A. Witteveen, M. Kleijn, A. Glas, L. Mittempergher: Financial Interests, Personal, Full or part-time Employment: Agendia NV. S. Bao, S. Mee, T. Cavness, J. Falk: Financial Interests, Personal, Full or part-time Employment: Agendia Inc. All other authors have declared no conflicts of interest.