AI Discussion of CAP TODAY, HER2 Low, August 2023

https://www.captodayonline.com/low-level-limbo-in-her2-breast-cancer/2/

Summarize in detail for an expert; 600 words.  Include a focus on disagreements about "low Her2" or "Her2 negative," one of the points of the journalist.

The two-part article discusses the evolving landscape of HER2 testing in breast cancer, focusing on the emerging challenge of categorizing cases with low levels of HER2 expression. The article acknowledges the complex nature of this issue and highlights differing opinions among experts regarding the use of terms like "HER2 low" or "HER2 negative."

Part 1 introduces the concept of "HER2 low" in breast cancer and its implications. The introduction of trastuzumab-deruxtecan (T-DXd) as a therapeutic option for HER2-low metastatic breast cancer has led to the need for more precise classification methods. The article highlights the challenges faced by pathologists in accurately identifying cases with low HER2 expression levels, especially those that were previously considered HER2 negative.

The American Society of Clinical Oncology (ASCO) and the College of American Pathologists (CAP) released an updated guideline to address these challenges. This update provides best practices to help pathologists navigate HER2 testing challenges brought about by antibody-drug conjugates. The article emphasizes the need for accurate classification due to the availability of targeted therapies that could benefit patients with low HER2 expression.

Part 2 delves deeper into the debates and perspectives surrounding the classification of HER2 expression levels. Some experts express concerns about the terminology "HER2 low" and suggest that it might lead to confusion, as it's not yet a well-defined clinical category. Despite the complexities, the term has gained popularity and is being used in clinical discussions.

Some pathologists advocate for separating HER2 0 and 1+ cases and assigning distinct labels to better reflect the differences in expression. There's a discussion about the challenges of implementing such changes due to the interplay between HER2 expression levels and various treatment responses. Experts acknowledge the limitations of the current scoring system and the evolving nature of the terminology.

Dr. Kimberly Allison, for instance, suggests that using a report comment or footnote with proposed wording could help convey the significance of different HER2 expression levels to oncologists without introducing new and potentially misleading terminology.

Dr. Stuart Schnitt suggests that rather than introducing new terminology, the emphasis should be on distinguishing between 0 and 1+ cases and ensuring consistency among pathologists in their interpretations. There's a discussion about the ongoing Destiny-B06 trial and its potential to shed light on the response of HER2 0 cases to treatments.

The article also highlights the role of technology, such as artificial intelligence, in improving the accuracy and consistency of HER2 testing. Dr. David Rimm proposes a shift from qualitative reading to quantitative measurement of HER2 expression levels. He discusses the development of assays that can measure protein expression using sophisticated image analysis tools.

In conclusion, the article highlights the ongoing debates within the medical community regarding the terminology and classification of HER2 expression levels in breast cancer. Experts have varying opinions on whether to adopt terms like "HER2 low" or stick to the traditional "HER2 negative." The challenges of accurately categorizing cases with low HER2 expression, along with evolving treatment options, emphasize the need for continued research and collaboration to optimize patient care.

User

Are there any references to automated image analysis, digital pathology, or artificial intelligence?  If so, please reflect these and discuss.

ChatGPT

Yes, the article does mention the role of technology, including automated image analysis, digital pathology, and artificial intelligence, in addressing the challenges of HER2 testing and classification. Here's how these aspects are discussed:

Artificial Intelligence (AI) and Quantitative Measurement: Dr. David Rimm discusses the shift from qualitative reading to quantitative measurement of HER2 expression levels. He mentions that new methods have been developed that allow for the measurement of protein expression on slides using fully quantitative methods. This approach involves advanced image analysis tools that can measure the amount of protein on a slide using a standard curve, similar to a biochemistry assay. Dr. Rimm emphasizes the importance of adopting the idea of measuring rather than reading and suggests that AI and quantitative methods could play a significant role in improving the accuracy and consistency of HER2 testing.

Automated Image Analysis and Digital Pathology: While the article doesn't extensively focus on automated image analysis and digital pathology, it does highlight the potential of these technologies. Dr. Rimm's approach of measuring protein expression using image analysis tools is an example of leveraging digital pathology for quantitative assessment. The article discusses how these advanced tools could help measure HER2 expression and other biomarkers with greater precision, providing a more objective evaluation compared to manual interpretation.

Improving Interobserver Variability: The article briefly mentions the use of education and online tutorials, such as those provided by the USCAP, to improve interobserver variability in HER2 testing. While not explicitly stated, these educational resources could include digital pathology techniques and automated image analysis examples, contributing to better standardization among pathologists.

Potential Role in the Future: Dr. Rimm also suggests that artificial intelligence and technology could be instrumental in the future of HER2 testing. As these tools continue to evolve, they might offer solutions to some of the challenges posed by HER2 expression variability and interpretation.

In summary, while the article doesn't extensively explore automated image analysis, digital pathology, or artificial intelligence, it does acknowledge their potential importance in addressing the challenges of HER2 testing and classification. Dr. Rimm's work in developing a high-sensitivity HER2 test using quantitative methods and image analysis tools exemplifies the potential role of technology in improving the accuracy and objectivity of HER2 expression measurement. These technologies have the potential to contribute to more standardized and consistent HER2 testing, ultimately benefiting patient care.