The Role of Artificial Intelligence in Accurate Interpretation of HER2 Immunohistochemical Scores 0 and 1+ in Breast Cancer

Wu, Si; Yue, Meng; Zhang, Jun; Li, Xiaoxian; Zhang, Huina; Wang, Xinran; Han, Xiao; Cai, Lun; Shang, Jiuyan; Jia, Zhanli; Wang, Xiaoxiao; Li, Jinze; Liu, Yueping

doi:10.1016/j.modpat.2022.100054

Cited by 37 publications

(33 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A multi-institutional study showed improved accuracy of HER2 0 and 1 + tumor interpretation when pathologists were assisted by AI compared to not being assisted (0.93 vs. 0.8). Importantly, the accuracy was significantly improved even with the presence of heterogeneity (0.68–0.89) ( Wu et al, 2023a ). Given the complications of HER2 evaluation in the presence of heterogeneity and the higher prevalence of HER2 heterogeneity in HER2 1 + and 2 + than in HER2 3 + tumors ( Seol et al, 2012 ), this methodology could assist pathologists in a more precise HER2-low evaluation ( Wu et al, 2023a ).…”

Section: Novel Complementary Methodologies and Future Directionsmentioning

confidence: 99%

“…Importantly, the accuracy was significantly improved even with the presence of heterogeneity (0.68–0.89) ( Wu et al, 2023a ). Given the complications of HER2 evaluation in the presence of heterogeneity and the higher prevalence of HER2 heterogeneity in HER2 1 + and 2 + than in HER2 3 + tumors ( Seol et al, 2012 ), this methodology could assist pathologists in a more precise HER2-low evaluation ( Wu et al, 2023a ). Gustavson et al used deep learning-based image analysis to generate a novel HER2 Quantitative Continuous Score (QCS).…”

Section: Novel Complementary Methodologies and Future Directionsmentioning

confidence: 99%

“…However, there is a lack of consensus to define HER2 heterogeneity for IHC ( Ahn et al, 2020 ; Sajjadi et al, 2022 ). It should be noted that intratumoral HER2 IHC heterogeneity (i.e., uneven distribution of HER2 expression or different intensities of HER2 staining in tumor cells) is more frequent in the HER2-low (2 + or 1 + IHC score) samples ( Seol et al, 2012 ; Wu et al, 2023a ). This condition may contribute to the poor consistency of HER2-low interpretation ( Wu et al, 2023a ).…”

Section: Current Challenges: How Far From the Perfect Test?mentioning

confidence: 99%

“…It should be noted that intratumoral HER2 IHC heterogeneity (i.e., uneven distribution of HER2 expression or different intensities of HER2 staining in tumor cells) is more frequent in the HER2-low (2 + or 1 + IHC score) samples ( Seol et al, 2012 ; Wu et al, 2023a ). This condition may contribute to the poor consistency of HER2-low interpretation ( Wu et al, 2023a ). Moreover, whether the presence of intratumor heterogeneity may affect the efficacy of ADCs is yet to be further investigated ( Wu et al, 2023b ).…”

Section: Current Challenges: How Far From the Perfect Test?mentioning

confidence: 99%

See 3 more Smart Citations

Pathological identification of HER2-low breast cancer: Tips, tricks, and troubleshooting for the optimal test

Sajjadi

Guerini-Rocco

Camilli

et al. 2023

Front. Mol. Biosci.

View full text Add to dashboard Cite

The introduction of novel anti-HER2 antibody-drug conjugates (ADC) for the treatment of HER2-low breast cancers has transformed the traditional dichotomy of HER2 status to an expanded spectrum. However, the identification of HER2-low (i.e., immunohistochemistry (IHC) score 1 + or IHC score 2+, without gene amplification) tumors is challenged by methodological and analytical variables that might influence the sensitivity and reproducibility of HER2 testing. To open all possible therapeutic opportunities for HER2-low breast cancer patients the implementation of more accurate and reproducible testing strategies is mandatory. Here, we provide an overview of the existing barriers that may trouble HER2-low identification in breast cancer and discuss practical solutions that could enhance HER-low assessment.

show abstract

Section: Novel Complementary Methodologies and Future Directionsmentioning

confidence: 99%

Section: Novel Complementary Methodologies and Future Directionsmentioning

confidence: 99%

Section: Current Challenges: How Far From the Perfect Test?mentioning

confidence: 99%

Section: Current Challenges: How Far From the Perfect Test?mentioning

confidence: 99%

See 2 more Smart Citations

Pathological identification of HER2-low breast cancer: Tips, tricks, and troubleshooting for the optimal test

Sajjadi

Guerini-Rocco

Camilli

et al. 2023

Front. Mol. Biosci.

View full text Add to dashboard Cite

show abstract

“…Assessing the HER2 expression involved immunohistochemical (IHC) staining and an in situ hybridization test if the former was deemed equivocal [10]. However, recent study has suggested that clinical HER2 interpretation could suffered from low concordance among pathologists, while with the assistance of several AI algorithms, it became more likely for the pathologists to reach an agreement [11–12]. The urgent need of AI adaption in the HER2 expression analysis was thereof apparent.…”

Section: Introductionmentioning

confidence: 99%

Cross-institutional HER2 assessment via a computer-aided system using federated learning and stain composition augmentation

Yang,

Chen,

Chang

et al. 2024

Preprint

View full text Add to dashboard Cite

The rapid advancement of precision medicine and personalized healthcare has heightened the demand for accurate diagnostic tests. These tests are crucial for administering novel treatments like targeted therapy. To ensure the widespread availability of accurate diagnostics with consistent standards, the integration of computer-aided systems has become essential. Specifically, computer-aided systems that assess biomarker expression have thrusted through the widespread application of deep learning for medical imaging. However, the generalizability of deep learning models has usually diminished significantly when being confronted with data collected from different sources, especially for histological imaging in digital pathology. It has therefore been challenging to effectively develop and employ a computer-aided system across multiple medical institutions. In this study, a biomarker computer-aided framework was proposed to overcome such challenges. This framework incorporated a new approach to augment the composition of histological staining, which enhanced the performance of federated learning models. A HER2 assessment system was developed following the proposed framework, and it was evaluated on a clinical dataset from National Taiwan University Hospital and a public dataset coordinated by the University of Warwick. This assessment system showed an accuracy exceeding 90% for both institutions, whose generalizability outperformed a baseline system developed solely through the clinical dataset by 30%. Compared to previous works where data across different institutions were mixed during model training, the HER2 assessment system achieved a similar performance while it was developed with guaranteed patient privacy via federated learning.

show abstract

Immunohistochemical assessment of HER2 low breast cancer: interobserver reproducibility and correlation with digital image analysis

Sun,

Kang,

Chen

et al. 2024

Breast Cancer Res Treat

View full text Add to dashboard Cite

The Role of Artificial Intelligence in Accurate Interpretation of HER2 Immunohistochemical Scores 0 and 1+ in Breast Cancer

Cited by 37 publications

References 27 publications

Pathological identification of HER2-low breast cancer: Tips, tricks, and troubleshooting for the optimal test

Pathological identification of HER2-low breast cancer: Tips, tricks, and troubleshooting for the optimal test

Cross-institutional HER2 assessment via a computer-aided system using federated learning and stain composition augmentation

Immunohistochemical assessment of HER2 low breast cancer: interobserver reproducibility and correlation with digital image analysis

Contact Info

Product

Resources

About