Intra- and Interlaboratory Reproducibility of the Sensitivity to Endocrine Therapy Assay for Stage II/III Breast Cancer.

Abstract

BACKGROUND:

The sensitivity to endocrine therapy assay (SET2,3) predicts treatment outcomes in Stage II-III breast cancer. SET2,3 measures transcription related to estrogen and progesterone receptors (SETER/PR index) and the molecular subtype (RNA4: ESR1, PGR, ERBB2, AURKA) from formalin-fixed paraffin-embedded (FFPE) tissue sections.

METHODS:

We designed a nested study across 3 pathology laboratories, each testing 60 breast cancers twice in controlled batches. Laboratories macrodissected and directly homogenized the unstained FFPE tumor sections, then performed the QuantiGene Plex bead-based hybridization assay. SET2,3 was calculated centrally using predefined statistical R-scripts and applying pre-defined cutpoints. Concordance correlation coefficient (CCC) was calculated from continuous measurements and Kappa statistic from categorical results. A mixed-effects model estimated contributions to bias (fixed effects) and variance (random effects) from the replicated design.

RESULTS:

Intralaboratory (CCC 0.96-0.99) and interlaboratory (CCC 0.98-0.99) SET2,3 results were concordant, with rates of agreement for high/low categorization within (Kappa 0.83-0.93) and between laboratories (Kappa 0.87-0.88). The relative contributions to overall variance of SET2,3 measurements were 96.90% from biological differences between cancers, 0.67% from interlaboratory variability, and 2.44% from residual causes including intralaboratory replicates. Similar results were obtained with SETER/PR, the baseline prognostic index calculated using pathological or clinical tumor and nodal staging information, and the 4 individual genes (ESR1, PGR, ERBB2, and AURKA).

CONCLUSION:

Intra- and interpathology laboratory measurements of SET2,3 and its components were highly reproducible when tested from FFPE tumor sections.

• Authors: Veerle Bossuyt, Rosanna Lau, Brandon Young, John Greg Howe, Fengmin Zhao, Brian Leyland-Jones, Lili Du, Tiffany Foli, Christos Hatzis, W Fraser Symmans
• Journal: Clinical chemistry (Clin Chem) Vol. 67 Issue 9 Pg. 1240-1248 (09 01 2021) ISSN: 1530-8561 [Electronic] England
• PMID: 34374711 (Publication Type: Journal Article, Research Support, Non-U.S. Gov't)
• Copyright: © American Association for Clinical Chemistry 2021.

Chemical References

• Biomarkers, Tumor
• Receptors, Progesterone
• Aurora Kinase A

Topics

• Aurora Kinase A
• Biomarkers, Tumor (genetics)
• Breast Neoplasms (drug therapy, genetics, pathology)
• Female
• Humans
• Prognosis
• Receptors, Progesterone (genetics)
• Reproducibility of Results