Discussion with the Scientists: Interview with VP of Product & Translational Research, Ben Glass, and the Vision for PathExplore™ Fibrosis

Building PathExplore™ Fibrosis: AI-Powered Fibrotic Microenvironment Insights for Oncology We sat down with VP of Product & Translational Research, Ben Glass , to discuss how AI-powered pathology is revolutionizing fibrosis quantification in oncology and to learn more about the development of PathAI’s latest product: PathExplore™ Fibrosis.

As precision medicine advances, understanding novel biomarkers like fibrosis has potential to optimize cancer therapies and predict patient outcomes. Fibrosis biomarkers, such as fiber and collagen parameters, play a significant role in the tumor microenvironment, in some cases correlating with disease progression and treatment response. However, traditional fibrosis quantification methods often face limitations due to the need for specialized microscopy and challenges in scalability.

In our discussion, Ben sheds light on how PathExplore Fibrosis offers a cutting-edge solution by bringing precision, scalability, and reproducibility to fibrosis assessment.

Learn more about PathExplore™ Fibrosis and demo the product here

Q: Ben, could you tell us a bit about why fibrosis quantification would be important for oncology research?

Ben Glass: Absolutely. Fibrosis, which is essentially an excessive buildup of collagen and other extracellular matrix components, has a significant role in tumor progression and therapy response. Accurately measuring fibrosis is vital because it gives us a clearer understanding of disease prognosis and can help us tailor effective therapies. Traditional methods, however, tend to be subjective, aren't scalable, and often miss the spatial complexities in the tumor microenvironment. This is because traditional methods for imaging fibrosis require access to additional tissues, stains, and specific microscopy equipment built specifically for this use case. That’s where we saw a real opportunity for innovation with PathExplore Fibrosis.

Q: PathExplore Fibrosis originated from internal research. Could you share how this product evolved from concept to reality?

Ben Glass: PathExplore Fibrosis started with our work on quantitative multi-modal anisotropic imaging (QMAI). We developed the imaging method to quantify collagen fibers directly from H&E-stained tissue samples, and we used these as the basis for training our machine learning model. The idea was to create a tool that could accurately assess fibrosis in H&E whole-slide images. After extensive validation and refinement, we transformed this technology into an off-the-shelf product that researchers and clinicians can use right out of the box. It’s been an exciting journey from concept to a fully realized product.

Q: What are some of the specific applications of PathExplore Fibrosis in oncology?

Ben Glass: PathExplore Fibrosis has several impactful applications in oncology. One major area is predicting treatment response—fibrosis can influence how well a patient responds to certain therapies, and our tool can help identify patients who may benefit more or less from specific treatments. It may also be useful as a prognostic biomarker; the extent and distribution of fibrosis can provide valuable insights for patient stratification and treatment planning. In drug development, PathExplore Fibrosis allows researchers to evaluate the impact of new therapies on fibrosis, potentially accelerating the development of more effective treatments. To name a few: 

• Predicting Treatment Response: Fibrosis can influence the efficacy of various cancer therapies. By characterizing the quantity, nature and spatial orientation of fibrosis in the tumor microenvironment, PathExplore Fibrosis can discover novel biomarkers toidentify patients who are more or less likely to respond to specific therapies.  
• Prognostic Biomarker: The extent and spatial distribution of fibrosis can provide valuable prognostic information, aiding in patient stratification and treatment planning.
• Drug Development: PathExplore Fibrosis can be used to evaluate the impact of novel therapies on fibrosis, accelerating drug development.

Q: What sets PathExplore™ Fibrosis apart from traditional methods of fibrosis quantification?

Ben Glass: PathExplore Fibrosis is unique in that it leverages advanced AI to provide objective, reproducible, and spatially resolved measurements of fibrosis. Unlike traditional approaches that may only offer a global assessment, our tool provides a highly detailed spatial map of fibrosis across the whole tumor. This allows researchers to see exactly where fibrosis is distributed and how it might be influencing the tumor’s behavior. And because it’s AI-driven, it automates the quantification process, which saves time and allows for high-throughput analysis of large datasets—something that’s really hard to achieve with manual methods. Here are the key benefits: 

• Enhanced Quantification: Identifying every individual collagen fiber in an H&E WSI would have been infeasible for human expert annotators. Instead, PathExplore Fibrosis utilizes deep learning models trained on a massive dataset of images annotated using our proprietary  QMAI microscope, ensuring unparalleled accuracy in identifying and quantifying collagen and fibrotic regions.
• Spatial Resolution: Unlike traditional methods that provide only global assessments, PathExplore Fibrosis offers detailed spatial information of individual fibers, enabling researchers to understand the distribution and patterns of fibrosis within the tumor microenvironment.
• Efficiency and Scalability: PathExplore Fibrosis automates the quantification process, significantly reducing analysis time and enabling high-throughput analysis of large cohorts.
• Standardized and Reproducible: By eliminating additional sources of variability, PathExplore Fibrosis ensures consistent and reproducible results across different studies and laboratories.

Q: You mentioned that PathExplore Fibrosis offers unprecedented resolution. How does it achieve this level of accuracy?

Ben Glass: The accuracy of PathExplore Fibrosis stems from its unique ground truth data. We trained the model on an extensive dataset of images that were annotated using our own QMAI imaging technology that we built in house. This allowed us to capture high-quality data from H&E-stained slides, focusing on collagen and fibrosis regions with incredible precision. The result is a product that can reliably identify and quantify fibrosis with a level of detail that hasn’t been possible until now: in contrast to a human annotator, our AI model doesn’t get tired and will highlight every single collagen fiber in a piece of tissue. There can be tens of thousands of them!

Q: Can you explain the concept of spatial resolution and why it's important in fibrosis research?

Ben Glass: Spatial resolution in this context refers to the ability to map where fibrosis occurs within the tumor environment. Knowing just the amount of fibrosis isn’t always enough; understanding its distribution helps us see patterns. Is fibrosis concentrated in specific areas or spread throughout the tumor? This insight can have major implications for understanding how fibrosis impacts tumor progression and for developing effective therapies that take these spatial dynamics into account.

Q: What impact does PathExplore Fibrosis have on efficiency and scalability in research settings?

Ben Glass: One of the main advantages is that PathExplore Fibrosis is designed for scalability. Manual fibrosis assessment is not only time-consuming but also requires specialized expertise, limiting how many samples can be processed in a given time. With PathExplore Fibrosis, we’ve automated much of this process, allowing labs to analyze large cohorts faster and without sacrificing accuracy. This makes it feasible to incorporate fibrosis quantification into high-throughput drug development pipelines.

Q: How does PathExplore Fibrosis ensure standardization and reproducibility in fibrosis quantification?

Ben Glass: By using AI to unlock these features directly from H&E images, PathExplore Fibrosis delivers standardized and reproducible results across different studies and labs. Consistency in fibrosis quantification is key for comparing results across studies, and that’s something PathExplore Fibrosis handles seamlessly, allowing researchers to rely on the data without worrying about variability between operators or labs.

Q: How does PathAI see the role of AI evolving in pathology, particularly with products like PathExplore Fibrosis?

Ben Glass: AI has tremendous potential to reshape pathology, and with PathExplore Fibrosis, we’re seeing it bring quantitative precision to areas that were once unable to be accessed with standard pathology workflows. PathAI’s mission is to transform pathology to improve patient care, and products like PathExplore Fibrosis are a part of that vision. By providing tools that allow for a more nuanced understanding of disease and treatment outcomes, we’re equipping clinicians and researchers with the insights they need to make more informed decisions, ultimately advancing the fight against cancer.

Learn more about PathExplore™ Fibrosis and demo the product here

PathExplore™ and PathExplore™ Fibrosis are for research use only. Not for use in diagnostic procedures.