# Reproducibility of Immunohistochemical Testing of Estrogen Receptors, Progesterone Receptors, Human Epidermal Growth Factor Receptor-2 (HER2) and Ki-67 in Vietnam

## TL;DR

This study evaluates how consistently Vietnamese hospitals test breast cancer samples for four key markers, finding that results are mostly reliable but with some variation, especially for Ki-67 and low HER2 expression.

## Contribution

The paper presents the first external quality assessment ring study in Vietnam for ER, PR, HER2, and Ki-67 immunohistochemical testing.

## Key findings

- ER testing showed the least variation (kappa 0.822, CV 4.8%), while Ki-67 had the most (kappa 0.647, CV 17%).
- HER2-low expression reproducibility was poor (kappa 0.323), compared to overall HER2 testing (kappa 0.794).
- 84% and 81% agreement for Ki-67 evaluations using clinically relevant cut points of ≥30% and ≥20%, respectively.

## Abstract

Immunohistochemical (IHC) testing of estrogen receptors (ER), progesterone receptors (PR), HER2 and Ki-67 on breast cancer samples is carried out in the majority of clinical departments to predict response to therapies and to determine prognosis. Issues surrounding the reproducibility of testing are well documented and guidelines recommend laboratories participate in external quality assessment (EQA) in order to ensure reliability of results.

To assess the reproducibility of IHC testing for these markers in hospitals from the south, north, and centre of Vietnam, estimated to be approximately half of all clinical hospitals in the country performing these tests.

As cases are referred for testing between hospitals, an EQA ring study was designed that included the testing of samples from all participating laboratories. Participants were provided with unstained slides of invasive breast carcinomas with different expression levels for ER, PR, HER2 and Ki-67.

There was a significant level of reproducibility for all four biomarkers, with ER testing giving the least variation in results (kappa 0.822, coefficient of variation [CV] 4.8%) and Ki-67 the greatest variation (kappa 0.647 CV 17%). However, 328/392 (84%) and 317/392 (81%) of the Ki-67 evaluations were in agreement when employing the clinically relevant cut points of ≥30% and ≥20%, respectively. The reproducibility of testing for HER2-low expression was relatively poor (kappa 0.323, 95% CI 0.223–0.424), compared to overall agreement for HER2 testing (kappa 0.794, 95% CI 0.753–0.836).

This is the first EQA ring study held within Vietnam for ER, PR, HER2 and Ki-67 and sets the base line as to the current level of reproducibility in the country. Continued participation in the program will help ensure the reliability of testing for clinical use.