Discordance in pathological testing for breast cancer: A complex situation requiring more international collaboration

Giuseppe Viale

Head of Department of Pathology and Laboratory Medicine, European Institute of Oncology, University of Milan, Italy;
Chairman of the Central Pathology Office of the International Breast Cancer Study Group (IBCSG), and Lead Pathologist
of the Breast International Group (BIG)

Despite several decades of experience with immunohistochemical (IHC) and molecular assays to identify prognostic and predictive biomarkers in breast cancer, worryingly high rates of discordance persist among laboratories and pathologists in histopathological reports. Giuseppe Viale, a specialist in breast cancer pathology, notes that, “This is a very real issue. It is happening now and it is a major clinical problem that needs to be addressed. The development of international testing guidelines has been a significant milestone, but further international collaboration is needed to address the many challenges that still lie ahead. It is a complex situation.”

Viale has reviewed tens of thousands of breast cancer samples over the last 15 years while running the Central Pathology Laboratory associated with adjuvant breast cancer trials such as ALTTO (McCullough AE, et al. Breast Cancer Res Treat. 2014) and APHINITY (von Minckwitz G, et al. N Engl J Med. 2017). Clinical trials with strict patient eligibility criteria provide most of the evidence for discordance in pathology assessment, as in order to ensure patients do not receive inappropriate treatment, and to reduce the risk of diluting results with ineligible patients, trials need to confirm local pathology results by central review. He described how central pathology review in the ALTTO and APHINITY trials identified relatively high rates of discordance, with false-negative rates for oestrogen receptor (ER) and progesterone receptor (PgR) of ~20% and ~15%, respectively, and false-positive rates for human epidermal growth factor receptor 2 (HER2) of ~9–14%. Similar data were reported in the neoadjuvant trial BERENICE (Swain SM, et al. Ann Oncol. 2018) (~11% false-positive rate for HER2). “Importantly, these data are from global trials, suggesting that the significant variability we observe is an issue worldwide and not confined to specific regions or single laboratories.”

So what does Viale think are the main causes of discordance in pathological review? “Theoretically, I believe there are two main reasons for this, the first involving technicalities, such as the timing and type of fixation, and the specificity and sensitivity of the assay; and the second involving ‘human’ factors, such as the experience and expertise of the reader, and a lack of interest and involvement in standard testing procedures at a senior level. Historically, at least, the importance of these biological parameters to the management of early breast cancer may not have been fully appreciated by some pathologists and, as a consequence, the necessary care has not always been taken with the assessments.” In terms of the contribution of technical factors, it is widely acknowledged that different assays can produce variable results, and this is clearly demonstrated when discordant findings are observed between experienced central laboratories, as reported in the ALTTO trial. Viale notes that, “The issue is further complicated by the fact that some laboratories mix antibodies and detection systems from different kits,” and he feels strongly that, “Standardisation of assays and testing protocols across laboratories must be improved.”

The clinical implications of discordant results in early breast cancer can be very serious. In a recent retrospective analysis of early breast tumours, high rates of discordance between local and central HER2 and ER/PgR assessments resulted in modifications to systemic prescriptions in a significant number of patients: 12% due to changes in HER2 status and 17% due to changes in ER/PgR status (Orlando L, et al. Breast. 2016). "The clinical consequences of a discordant pathology result can be hugely harmful to a patient with early breast cancer,” explains Viale. “Based on trial data, there is a ~20% risk that a patient with an ER-negative result will have an ER-positive tumour. The only treatment option for ER-negative breast cancer is chemotherapy, so, in reality, a patient with early breast cancer may receive chemotherapy, and be denied endocrine therapy, for a hormone-responsive tumour, and this can be detrimental to patient outcome.”

There is clearly much room for improvement in pathological testing, and international collaborative efforts will be needed to help change the situation. The last few years have seen some progress in this regard; Viale was one of a panel of pathologists and clinicians invited by the American Society of Clinical Oncology (ASCO) and the College of American Pathologists (CAP) to contribute to the development of recommendations for optimal laboratory testing, initially for ER/PgR (in 2010) and subsequently for HER2 (in 2013). However, while these recommendations are currently considered the gold standard for early breast cancer pathological assessment, they have not led to acceptable rates of concordance. Viale believes that, “If these guidelines were followed by everyone, we could potentially standardise testing and achieve at least 95% concordance between local and central pathological review.” One of the major obstacles to this goal is cost. “Cost is an ongoing issue and often dictates the type of assay available in a specific institution, regardless of the pathologist’s preference that is informed by the guidelines.”

Looking to the future, the use of alternative assays based on mRNA rather than IHC assessment may hold the key to reducing variability and increasing reproducibility in ER/PgR and HER2 testing. “However, until we have more experience with mRNA assays the jury is still out on their relative benefits. The current guidelines state that there is no evidence that they are any more accurate or reproducible than IHC methods,” he mentions.

Whatever the future holds, Viale feels that further international collaboration among the pathology community is vital in order to raise awareness and find opportunities to improve the accuracy and reproducibility of pathological testing. “Ultimately, all patients should be given the opportunity to receive timely and effective treatment for their disease and we need to work together to get this right.”