Abstract LB-327: Analytical performance of the nCounter analysis system for gene expression cancer signatures
Bibliographic record
Abstract
Abstract Background: The nCounter Analysis System is a platform for performing highly multiplexed, digital quantification of hundreds of different nucleic acid species in a single reaction. The system is being developed for use as a platform for in vitro diagnostic applications. The current study aimed to evaluate the analytical performance of the system by implementing a 50-gene signature used for determining the intrinsic subtype and prognostic risk of a breast cancer tumor from formalin-fixed, paraffin-embedded (FFPE) tissue samples. The gene expression profiles can be used to divide breast cancer into four intrinsic subtypes: Basal-like, Luminal A, Luminal B, and HER2 enriched. A normal-like subtype identifies tumor specimens contaminated with a high percentage of normal tissue. A prognostic Risk-Of-Relapse (ROR) score is also calculated. Methods: The nCounter assay is performed by direct, multiplexed hybridization of molecular barcodes to target mRNAs. For this study, a CodeSet containing probes for 50 classification genes and 8 normalizing genes was developed. The intrinsic subtyping algorithm was trained by supervised hierarchical clustering of data from 538 samples. Prototypical centroids for tumor subtypes were chosen as statistically significant clusters, while the normal-like centroid was trained from reduction mammoplasty samples. All samples were correlated to the five prototypical centroids and assigned the subtype with the largest positive correlation. The analytical precision of the assay was measured by testing the same sample across replicates. The precision of the subtyping test was determined by assaying 40 different samples across 9 different combinations of reagent lots. The performance of the assay was also evaluated with varying input levels of RNA. Results: The precision of the nCounter assay was found to be driven by Poisson noise in the digital measurements at low expression levels. When samples were subtyped across multiple reagent lots, the correlations to each of the five subtypes were narrowly distributed. One sample varied in the categorical subtype call between Luminal B and HER2-enriched across all lots, however, this sample was similarly correlated between both Luminal B and HER2-enriched across all lot combinations. Not surprisingly, the variability for the ROR was minimal, even for the case where the categorical subtype call changed. Finally, the subtype call and ROR score were stable across a 10x range of RNA input. Conclusions: The NanoString nCounter system has the sensitivity and precision to be implemented as a distributed platform for multiplexed gene expression profiling of tumors. The sensitivity and direct digital detection without the need for enzymatic amplification make the technology compatible with FFPE tissue samples. The precision should make the technology robust under the many varied conditions seen in testing labs. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr LB-327. doi:10.1158/1538-7445.AM2011-LB-327
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".