Posted on Sep 28, 2023
To date, we have had little evidence of whether immunohistochemistry (IHC) can reliably diagnose HER2-low breast cancers. Now a new study has assessed the concordance between expert pathologists using IHC to assess HER2 levels – and the results will do little to reassure patients.
As an assay, immunohistochemistry (IHC) for HER2 protein was designed to identify high levels of HER2 expression, dividing breast cancer patients into either ‘HER2-positive’ or ‘HER2-negative’.
But last year, the groundbreaking DESTINY-Breast04 trial identified that the antibody-drug conjugate Enhertu was beneficial for patients with lower levels of HER2-expression – the so-called ‘HER2-low’ population. Ever since the announcement of the results at ASCO in 2022, a new challenge for HER2 testing has been set.
IHC was never designed to discriminate between high, low, and no expression of HER2. We have little evidence to date that IHC is a reproducible and reliable method for pathologists to identify HER2-low cancers. Yet it is still the main method used to type breast cancer tumours.
So will these patients get a reliable diagnosis of HER2-low, no matter which hospital they’re treated at? It’s a vital question, given that ‘HER2-low’ group represents up to half of all breast cancers, and that Enhertu can delay progression and extend life for patients in this group.
Concordance of HER2-low scoring in breast cancer among expert pathologists.
A new study led by Dr Abeer Shabaan of University of Birmingham, UK, set out to answer this question, and assess the concordance of HER2-low scoring among pathologists.
Dr Shabaan assembled a group of 16 expert pathologists from the UK National Coordinating Committee for Breast Pathology. 50 invasive breast cancer core biopsies, enriched for cases with lower HER2 expression, were selected from the archives of Queen Elizabeth Hospital in Birmingham, a reference centre for HER2 IHC testing. All slides were stained using Ventana Autostainer, and digitally scanned using the Leica Aperio AT2 slide scanner. They were asked to assess the degree of HER2 staining with IHC, which ranges from 0 (no staining), 1+, 2+, to 3+ (highest degree of staining).
In short, each of the expert pathologists looked at the same breast cancer samples, stained and imaged in the same way, using the same ‘gold-standard’ test that they’ve used countless times in their careers.
So what were the results?
Overall agreement is low
Absolute agreement (where all 16 pathologists agreed on the degree of HER2-staining) occurred in only 3 of the 50 biopsies (6%), all of which were 3+ staining. No absolute agreement was achieved for tumours scored as 0, 1+ or 2+ (which includes the ‘HER2-low’ group).
High agreement (where more than 75%, or 12 out of the 16, pathologists agreed) occurred in a further 23 of the 50 cases (46%). So, in total, for only 26 of the 50 cases – little over half of the breast cancers – was there high or absolute agreement on the HER2-staining among 16 expert pathologists.
Concordance increased when categories were combined. When distinguishing between high expression (3+), low expression (1+ or 2+), and no expression (0), there was high or absolute agreement in 42 cases (84%).
Overall, agreement among the pathologists was lower for ‘HER2-low’ cases than for all breast cancers.
Of great importance were situations where incorrect scoring could lead to incorrect treatment, which occurred in 8 of the 50 tumours (16%). These discrepancies in the eight tumours (from six pathologists) include three tumours scored negative while the consensus was HER2-positive, and five tumours scored positive while the consensus was HER2-negative.
The authors conclude that “this current study highlights that a proportion of breast cancers can be challenging to designate as immunohistochemistry score 1+ versus 0 even for expert breast pathologists.”
In terms of remedies, the authors recommend double scoring for borderline cases, and discussions with colleagues to reach consensus. They also say pre-analytical factors, which are well-known to influence the results of IHC, will be crucial.
Disappointing news for breast cancer patients
From the patient’s perspective, the results of the study are not reassuring. This study suggests a woman whose breast cancer is HER2-low has a reasonable chance, depending on where she is treated, that she will not be diagnosed accurately with IHC. The danger is she might miss out on treatments which could delay progression and extend her life.
The pathologists taking part in this study do not lack experience or skills. These expert pathologists are among the most experienced in the UK and Ireland – what’s more, they train other pathologists and advise the NHS and Royal College of Pathology.
The results also confirm the findings of other studies, such as one led by Dr Ximena Baez-Navarro from Erasmus Medical Center in The Netherlands. In that study, 106 HER2-negative cases were scored by 16 pathologists according to ASCO/CAP guidelines. Absolute concordance between the pathologists was 4.7% overall, which was only improved when staining categories were combined to compare 0 vs ultralow/1+/2+.
For us at Cerca Biotech, the results of these studies provide another example of the shortcomings of immunohistochemistry, and how it is letting down pathologists and patients alike. IHC might have been sufficient in the days of a binary diagnosis of ‘HER2-positive’ or ‘HER2-negative’. But with more advanced treatments requiring more subtle assessment of HER2 expression, it’s clear to us that IHC as a tool is no longer good enough to help pathologists do their job.
**MammaTyper® is a diagnostic tool based upon quantitative PCR, a proven, standardised method where agreement between laboratories is at least as good (if not better) than that for IHC. What’s more, it is far simpler to carry out, reliably reproducible, much faster, and a fraction of the cost.
Could MammaTyper® help you diagnose your patients? Contact us to find out for yourself: https://www.cercabiotech.com/contact-us**
Reference: Zaakouk M, Quinn C, Provenzano E, Boyd C, Callagy G, Elsheikh S, Flint J, Millican-Slater R, Gunavardhan A, Mir Y, Makhija P, Di Palma S, Pritchard S, Tanchel B, Rakha E, Atallah NM, Lee AHS, Pinder S, Shaaban AM. (2023) “Concordance of HER2-low scoring in breast carcinoma among expert pathologists in the United Kingdom and the republic of Ireland –on behalf of the UK national coordinating committee for breast pathology.” The Breast, 70, p82-91; doi: 10.1016/j.breast.2023.06.005 Link: https://www.thebreastonline.com/article/S0960-9776(23)00509-X/fulltext