Assessment of HER2 using the 2018 ASCO/CAP guideline update for invasive breast cancer: a critical look at cases classified as HER2 2+ by immunohistochemistry

Concordance of RT-qPCR with immunohistochemistry and its beneficial role in breast cancer subtyping

This study was to compare the concordance of transcription-quantitative polymerase chain reaction (RT-qPCR) with immunohistochemistry (IHC) in determining estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and tumor proliferation index (Ki67) status in breast cancer, and to assess the prognosis based on different subtypes. Totally 323 breast cancer patients were selected, including 216 in the training set and 107 in the validation set. Logistic regression models were constructed using 5-fold cross-validation with the mRNA expression of each biomarker as the predictor and the corresponding IHC expression level as the binary response variable. Receiver operating characteristic curve was used to determine the cutoff value. When the thresholds of ER, PR, HER2, and Ki67 were 0.764, 0.709, 0.161, and 0.554, there existed high concordance rates between IHC and RT-qPCR in ER (94.4%), PR (88.0%) and HER2 (89.4%) and a medium concordance rate in Ki67 (67.8%), which were further confirmed in the validation set (ER: 81.3%, PR: 78.3%, HER2: 80.4%, and Ki67: 69.1%). Based on the subtyping stratified by RT-qPCR, the 5-year recurrence-free interval rates of patients with luminal, HER2-enriched, and triple-negative subtypes were 88% (95% CI: 0.84-0.93), 82% (95% CI: 0.73-0.92) and 58% (95% CI: 0.42-0.80), respectively, which were similar to those assessed by IHC (88%, 78% and 47%). RT-qPCR may be a complementary method to IHC, which can not only provide additional useful information in clinic, but also show more advantages over IHC in determining certain subtypes of breast cancer.

HER2 Equivocal (Score = 2+) Breast Carcinoma Cases Identified by Immunohistochemistry at a South African Hospital. What is the Impact of Fluorescent In Situ Hybridization Testing?

The American Society of Clinical Oncology and the College of American Pathologists (ASCO/CAP) guidelines are used for human epidermal growth factor receptor 2 (HER2) reporting in breast carcinoma. Cases that demonstrate weak to moderate complete membrane immunohistochemical staining in >10% of the tumor are scored as 2+ (equivocal). This study aimed to determine what proportion of HER2 immunohistochemistry (IHC) score = 2+ breast carcinomas were confirmed to be positive by HER2 fluorescent in situ hybridization (FISH). There were 241 HER2 IHC score = 2+ breast carcinomas included. Most (74.3%) carcinomas were estrogen and progesterone receptor-positive. Invasive breast carcinoma of no special type (89.2%) was the commonest histologic subtype. Most tumors were grade 2 (64.3%). As per the FISH report, at the time of diagnosis, 27 cases (11.2%) were HER2 FISH positive. All HER2 FISH equivocal cases and one FISH positive case assessed using the 2013 ASCO/CAP HER2 criteria were reclassified to HER2 FISH negative when the 2018 criteria were applied. There was a high level of agreement (κ = 0.979) between HER2 FISH results obtained using the 2013 and the 2018 criteria. This study provides insight into the frequency of HER2 FISH positivity (11.2%) among HER2 IHC score = 2+ breast carcinomas and the impact of modifications to the ASCO/CAP HER2 guidelines. Elimination of the HER2 FISH equivocal category by the 2018 guidelines has reduced the need for repeat testing and simplified clinical management. Reclassification of previous HER2 FISH positive to negative has resulted in some patients being ineligible for costly anti-HER therapy.

Clinicopathologic Correlation of CD44 + /CD24 - Expression in Breast Cancer: a Report from Tertiary Care Medical University in India

CD44 + /CD24 - phenotype has been associated with stem cell-like characteristics with enhanced invasive properties, radiation resistance, and with distinct genetic profiles suggesting a correlation to adverse prognosis in western literature. The aim of this study was to study CD44 + /CD24 - phenotype as an adverse prognostic marker in Indian breast cancer patients. N = 61 breast cancer patients included in a tertiary care facility in India were evaluated for receptor studies (estrogen receptor ER, progesterone receptor PR, Herceptin antibody Her2 neu receptor, CD44 & CD24 stem cell markers). CD44 + /CD24 - phenotype was statistically related to adverse factors like estrogen and progesterone receptors non-expression, her 2 neu expression, and triple-negative breast cancer. Of the 39 patients with ER-ve status, 33 (84.6%) were found to have CD44 + /CD24 - phenotype and 82.5% of all the CD 44 + /CD24 - patients were ER negative (p = 0.001). Thirty-four (75.5%) of the PR-ve patients showed the CD44 + /CD24 - phenotype, and of all the CD 44 + /CD24 - patients, 85% of were PR negative (p = 0.006). Thirty-six (75%) of Her-2-Neu + ve were CD44 + /CD24 - . Approximately 90% of the Her 2 Neu patients expressed CD44 + /CD24 - and 76.9% of all the triple-negative patients were found to be CD44 + /CD24 - expression (p = 0.001). CD44 + /CD24 - had a significant association with adverse prognostic factors like stage of disease, hormonal receptor status, and molecular subtypes in Indian breast cancer patients like the Western data.

Using Deep Learning to Predict Final HER2 Status in Invasive Breast Cancers That are Equivocal (2+) by Immunohistochemistry

Invasive breast carcinomas are routinely tested for HER2 using immunohistochemistry (IHC), with reflex in situ hybridization (ISH) for those scored as equivocal (2+). ISH testing is expensive, time-consuming, and not universally available. In this study, we trained a deep learning algorithm to directly predict HER2 gene amplification status from HER2 2+ IHC slides. Data included 115 consecutive cases of invasive breast carcinoma scored as 2+ by IHC that had follow-up HER2 ISH testing. An external validation data set was created from 36 HER2 IHC slides prepared at an outside institution. All internal IHC slides were digitized and divided into training (80%), and test (20%) sets with 5-fold cross-validation. Small patches (256×256 pixels) were randomly extracted and used to train convolutional neural networks with EfficientNet B0 architecture using a transfer learning approach. Predictions for slides in the test set were made on individual patches, and these predictions were aggregated to generate an overall prediction for each slide. This resulted in a receiver operating characteristic area under the curve of 0.83 with an overall accuracy of 79% (sensitivity=0.70, specificity=0.82). Analysis of external validation slides resulted in a receiver operating characteristic area under the curve of 0.79 with an overall accuracy of 81% (sensitivity=0.50, specificity=0.82). Although the sensitivity and specificity are not high enough to negate the need for reflexive ISH testing entirely, this approach may be useful for triaging cases more likely to be HER2 positive and initiating treatment planning in centers where HER2 ISH testing is not readily available.

Patterns and Prevalence of BRCA1 and BRCA2 Germline Mutations Among Patients with Triple-Negative Breast Cancer: Regional Perspectives

Background

Among all subtypes, patients with triple-negative (TN) breast cancer is known for their poor outcome and their higher risk of harboring BRCA1 or BRCA2 pathogenic mutations. Identification of such mutations has clinical impact on breast and ovarian cancer prevention and treatment decisions. We here report on patterns and prevalence of BRCA1 and BRCA2 mutations among Arab patients diagnosed with TN subtype.

Patients and Methods

Patients with TN-breast cancer (n=197) were enrolled regardless of their age or family history. Following a detailed genetic counseling, BRCA1/2 testing was performed at reference labs. BRCA1 and BRCA2 variants were classified as negative, pathogenic/likely pathogenic (positive) and variants of uncertain significance (VUS).

Results

Median age of enrolled patients was 42 (range, 19–74) years and 27 (13.7%) were non-Jordanian Arabs. Among the study group, 50 (25.4%) were tested positive for BRCA1 (n=36, 18.3%) or BRCA2 (n=14, 7.1%), while 14 (7.1%) others had VUS. Compared to older ones, mutation rates were higher among patients <40 years (32.9%, P= 0.034), those with close relatives with breast, ovarian, pancreatic or prostate cancer (37.8%, P=0.002) and those with two or more breast cancers (41.4%, P=0.032). Among eligible patients, 23 (63.9%) patients underwent prophylactic mastectomy, while 19 (52.8%) patients had risk-reducing salpingo-oophorectomy. None of the patients with VUS underwent any prophylactic surgery.

Conclusion

Arab patients with TN-breast cancer have relatively high BRCA1 or BRCA2 mutation rates. Young age at diagnosis and personal and family history of breast cancer further increase this risk.