*For medical professionals only
In the era of the new generation of ADC drugs, can the current standard test meet the treatment needs of HER2 low-expression breast cancer?
HER2 low expression accounts for about half of breast cancer, but in the past, due to the lack of corresponding targeted therapy, it was often considered HER2 negative to be classified as HR+/HER2- or TNBC. With the advent of a new generation of ADC drugs, HER2 low expression has become a direction of clinical attention. As HER2 low expression embarks on the "stage" of new classification of breast cancer, the accurate detection of HER2 low expression has also become a key consideration for scholars in the field of breast cancer. Recently, JAMA Oncol published a research report "Examination of Low ERBB2 Protein Expression in Breast Cancer Tissue", which explored whether the current HER2 standard detection method can accurately identify HER2 low-expression breast cancer. The key content is specially sorted out as follows for the benefit of the reader.
The Age of ADCs, HER2
Precision detection of low-expression breast cancer is a major focus
ADC refers to a new type of anti-tumor drug that conjugates cytotox drugs to monoclonal antibodies through ligon couplings, and targets cytotoxic drugs to tumor cells by specifically binding to antigens on the surface of tumor cells. The new generation of ADC drugs with protease cleavable liquefiable ligates and highly active carriers with membrane permeability can exert a strong bystander effect based on this, and have been shown to have significant anti-tumor activity in HER2-low-expression breast cancer. In contrast, trastuzumab requires HER2 amplification (high expression) to predict therapeutic benefit. Breast cancer HER2 expression status can be divided into three expression ranges, but immunohistochemistry (IHC), invented more than 30 years ago, was optimized for the detection of HER2 high expression as a companion diagnosis. Patients with breast cancer with IHC 3+ or IHC 2+ and HER2 fluorescence in situ hybridization (FISH) amplification can benefit significantly from trastuzumab therapy and are interpreted as HER2 positive. Breast cancer with IHC 0 or IHC 1+ is interpreted as HER2 negative because it is not sensitive to trastuzumab therapy. At present, although the new generation of ADC drugs is effective for HER2 low-expression breast cancer (IHC 1+), the detection method designed for HER2-positive breast cancer is used in clinical trials, and theoretically, this method may not be a good detection of HER2 low-expression breast cancer, so there may be some problems in clinical application.
Multiple large randomized clinical trials are currently evaluating the efficacy, safety and tolerability of next-generation ADC drugs in patients with HER2-low-expression breast cancer. Based on the effectiveness of these novel drugs in HER2-low-expression breast cancer, if the current companion detection methods cannot well identify the HER2 low-expression population or the effective population of the new drug, it may be that some patients who can benefit from it cannot receive effective treatment (false negative), and may also make some patients who cannot benefit from these drugs receive unnecessary treatment (false positive). Therefore, it is important to explore more accurate detection methods.
Existing HER2 testing standard scheme,
Whether it is possible to accurately distinguish between 0 and 1+
The study reported in the literature aims to determine whether current companion diagnostic testing techniques can identify HER2 low-expression breast cancer. The study hypothesized that there were differences between the beneficiary populations assessed in the current trial and those who actually benefited from the new generation of ADC drugs, resulting in fewer patients receiving treatment than those who actually benefited from treatment (i.e., false negatives), or that some patients continued to receive medication despite limited benefit (i.e., false positives).
■ Research methods
The study evaluated the results of the 2019 and 2020 American Pathological Society (CAP) study of HER2 expression levels in breast cancer. Participating laboratories (n=1391-1452) received 10 HER2 thick needle aspiration biopsy specimens from 2 tissue microarrays (TMAs) twice a year. HER2 testing is then performed using standard test methods in the laboratory. These TMAs are then scored and returned to cap as part of their quality assessment program. The total survey dataset covered data from the HER2 immunohistochemical score (20 rough needle biopsies twice a year, for a total of 80 rough needle biopsies) in 1391-1452 laboratories each year, as well as supplementary questions about the methods used. The investigators pooled the relative frequency and distribution of scores for all individual HER2 pathological biopsies. Scoring consistency between labs is defined as 70% or higher, which means that 70% or higher of evaluators score the same chip on the same score on a 4-subscale.
The second independent analytical dataset comes from breast biopsy specimens collected by Yale's Department of Pathology starting in 2018. The films were read by 18 pathologists certified by the Commission and mostly with more than 5 years of experience and participated in a consistency study between evaluators. This part of the data comes from hematoxylin-ebony staining and HER2 IHC digital scan images of 170 independent cases. Pathologists score cases as 0, 1+, 2+, or 3+. 17 or 18 raters scored more than 90% of a single sample on a 4-point scale, defined as acceptable values. The Yates-corrected χ2 test was used to compare all cases in the Yale cohort with a "0" or "3+" score with at least 1 case with a rating of "0" or "3+". All tests were bilateral with a significance level of 0.05.
■ In the CAP survey, 1/4 of the 0 and 1+ pathological chip interpretation consistency is less than 70%
The consistency rate was 90% or higher among 65% (52/80; 26 in 2019 and 26 in 2020) of the 80 HER2 pathology specimens evaluated in the CAP survey (Figure A). This high degree of consistency is limited to 0 and 3+ scores. ERBB2 0 had the lowest consistency with 1+ cases. Of the 80 pathological specimens, 56 were considered negative (HER2 score 0 or 1+). In 25% of the pathological specimens, the consistency was less than 70% (n=15; Figure B). The CAP study evaluated both herr2 IHC samples before and during the analysis process. However, if you want to evaluate the variables before analysis, you need to have a calibrated reference gold standard. Since CAP uses consistency rather than a separate gold standard laboratory, consistency among pathology reading specialists is first assessed to determine a reference standard when evaluating pre-analysis parameters for sample analysis.
Her2 IHC score distribution in 80 pathological samples in the A.CAP survey
The percentage is the proportion of 0, 1+, 2+ or 3+ interpreted by the reader for each specimen, and the black part is a 15/80 specimen with a consistent < 70%.
Figure B.THE DISTRIBUTION OF HER2 IHC scores among the 15 pathological specimens with the highest inconsistencies in 0 and 1+ in the B.CAP survey
■ For the same pathological specimen, the interpretation consistency of HER2 0 is only 26%
To assess consistency among pathology readers, it is first necessary to have multiple pathologists score the same biopsy specimen of the same case, rather than TMAs, to assess the consistency of pathologist readings in a way that simulates the real world. To assess this parameter, HER2 IHC was evaluated by 18 pathologists from 15 institutions on whole-tissue slices of 170 independent biopsy cases collected from Yale University. Of the 170 cases, 92 were interpreted as 0 by at least 1 pathologist. Using a defined value of 90% or higher as an acceptable consistency, in 24 (26%) of these 92 cases, the interpretation was endorsed by at least 17 of the 18 assessors (i.e., a 26% agreement for HER20). In contrast, at least one pathologist interpreted 45 patients as 3+, also using a defined value of 90% or higher, and 26 (58%) of these 45 cases were endorsed by at least 17 of the 18 evaluators (Figure C), i.e. HER2 3+ with a 58% agreement. Comparisons of 0/1+ consistent cases with 2+/3+ consistent cases showed statistically significant differences (χ2=12.07, P<0.001).
Figure C. Hertiotic slices of 170 cases in the Yale cohort were distributed by HER2 IHC scores read by 18 pathologists
How to distinguish between 0 and 1+ to accurately detect
HER2 is lowly expressive and needs further exploration
In the study, which used the current standard HER2 test, CAP survey data from more than 1400 laboratories around the world was used, and the results showed that the current standard HER2 test had poor consistency in evaluating 0 and 1+ cases. In an independent Yale cohort analysis, the study found that the inconsistency between 0 and 1+ was significantly greater than the inconsistency between 2+ and 3+. Although the agreement on 2+ and 3+ was only 58% for reading specialists, its clinical impact was small, as 2+ cases required FISH testing prior to treatment to guide treatment decision-making, while cases 0 and 1+ did not have such a interpretation verification process. At the same time, it should be noted that the consistency of pathologists in interpreting 0 and 1+ cases is not enough to define the reference standard, so future research needs to explore the problems before analysis.
conclusion:
This study shows that the current HER2 detection method can not accurately distinguish between 0 and 1+ cases, that is, it cannot effectively distinguish between HER2-negative and HER2 low-expression breast cancer, which has a very broad application prospect given that the new generation of ADC drugs has shown unprecedented therapeutic benefits in HER2 low-expression breast cancer. Future treatment decisions that remain based on the results of the current standard HER2 test can result in a significant proportion of patients being misled or unable to receive treatment. To solve this problem, a novel assay that can detect cases of low EXPRESSION of HER2 is required, and the new assays need to be standardized using standard tablets.
bibliography:
[1] Fernandez AI,Liu M,Bellizzi A,et al. Examination of Low ERBB2 Protein Expression in Breast Cancer Tissue.JAMA Oncol.2022 Feb 3:e217239.
*This article is supported by AstraZeneca and is intended for medical professionals only
Approval number: CN-94817 is valid until: 2023-4-24
*The copyright of this article belongs to the original author, if you need to reprint, please contact the original author for authorization