The digitization of pathology slides enables image analysis algorithms to extract diagnostically relevant information from the pixels of the image. The creation of this data can, in turn, help pathologists diagnose cancer with greater precision. The technology developed from Proscia’s active collaborations with research departments at major academic medical centers worldwide addresses the need for diagnostic tests currently unmet by genomic or molecular assays alone and pushes the boundaries of our understanding of cancer.
The era of precision medicine was ushered in via high-throughput molecular assays, identifying patients with certain phenotypes and genotypes that correspond with significant improvement in overall and disease-free survival after certain neoadjuvant treatments. These molecular biomarkers effectively stratify patient populations for treatments such as PD-L1 checkpoint inhibition, which has shown to significantly improve overall and disease-free survival for advanced melanoma patients with CD8+ T-cells located in the invasive tumor margin1.
At Proscia, our vision is to use the image analysis capabilities of our digital pathology platform as a complement to the capabilities of molecular assays — leveraging data encoded in images to drive precise surgical procedures and therapeutic plans. In order to do this, we are working with major medical centers around the world to combine our deep learning technology with high order, high-value data to create diagnostic tests that address unmet needs for patients.
These cancer-specific image-based diagnostic tests represent a potential shift in how we think about digital image analysis. Most commercial image analysis products — including Proscia’s own digital pathology platform — offer chromogenic immunohistochemistry (IHC) quantification that has an inherently limited dynamic range due to compression of the visible spectrum and enzymatic reactions of antibodies with proteins. While yielding IHC quantification that is concordant with what a pathologists reads is effective for automating monotonous tasks, and allowing pathologists to focus on the difficult cases is a value proposition for throughput, the future of image analysis will rely on supplying pathologists with supplementary tests for precision diagnosis. This is why our partnerships and collaborations with research organizations are so important. Digital pathology gives us the chance to not only make pathology “faster and easier” for pathologists but also to make it vastly more useful as a prognostic and diagnostic tool.
We are still early in the adoption of digital pathology, but image-based diagnostic assays could be the tipping point to drive proliferation of digital pathology into the clinical workflow. We envision a time when “digital pathology” becomes the standard and are actively looking for institutional partners and collaborators that share our vision to join our current studies and start new investigations into the effect of image-based diagnostic assays on the precision diagnostic workflow.
We invite pathologists, researchers, and clinicians to reach out and tell us about their personal experiences with difficult diagnoses, and to share their knowledge of treatment plans that could be informed by quantified histology using Proscia’s deep learning enabled technology. Our vision can only come to fruition if we tackle this opportunity with trusted partners.
 Tumeh PC, Harview CL, Yearley JH, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature. 2014;515(7528):568-571. doi:10.1038/nature13954.