HistoIndex Explores the Clinical Utility of Stain-free AI Digital Pathology Platform in 388 Patients with Triple-Negative Breast Cancer (TNBC)

Assessing the morphological and architectural changes in collagen fibers with the platform could potentially aid TNBC clinical trials in categorizing…

Assessing the morphological and architectural changes in collagen fibers with the platform could potentially aid TNBC clinical trials in categorizing patients and monitoring therapeutic responses.

SINGAPORE, Jan. 24, 2021 /PRNewswire/ — HistoIndex‘s stain-free AI digital pathology platform, incorporating Second Harmonic Generation (SHG), holds promise as a quantitative tool in the assessment of morphological and architectural changes in collagen fibers within the tumor-stromal microenvironment in patients with TNBC. This will allow clinicians to better interpret the role of collagen remodeling in tumor progression and its prognostic value. On a larger scale, the tool may greatly aid existing and future TNBC studies that are currently exploring new therapies for targeted treatments.

TNBC is an incredibly challenging and aggressive form of breast cancer compared to other subtypes and holds a relatively poor prognosis primarily due to a lack of targeted treatments. In cancer, collagen fibers play a significant role in the tumor microenvironment, with remodeling of the extracellular matrix (ECM) that is often more collagen-rich with increased ‘stiffness’ [1]. As a component of the ECM, collagen may also influence cancer cell behavior [2]. Particularly in TNBC, collagen remodeling is seen in the stromal compartment [3].

Assessing Collagen Features at a Finer Level of Detail

In a collaborative study involving scientists from the Institute of Molecular and Cell Biology (IMCB) in Singapore and TNBC pathologists from the Singapore General Hospital (SGH), unstained biopsies from 388 TNBC patients were scanned using HistoIndex’s AI-based SHG platform and analyzed to extract different collagen features from the SHG images at a finer level of detail. Findings published in the leading peer-reviewed oncology journal, Breast Cancer Research [3], showed a strong correlation between several imaging features and clinicopathological characteristics. Aggregation of collagen fibers, collagen fiber density and the length of dispersed thin collagen fibers were key collagen-associated parameters revealed to be of prognostic value based on the patient cohort and clinical outcomes. Furthermore, analyzing the aggregated thick collagen (ATC) fibers and dispersed thin collagen (DTC) fibers (as shown in Figure 1) provided a novel understanding of collagen remodeling during cancer progression.

Figure 1: Picture A shows a patient’s TNBC biopsy scanned by HistoIndex’s stain-free AI digital pathology platform. The collagen fibers as well as changes in collagen structures are highlighted in green when detected by SHG and shows a comparison of the tissue area occupied by the ATC and DTC fibers based on intensity, texture, and morphology. Picture B shows an overlay of the collagen structure acquired by SHG onto an image stained with Haemotoxylin and Eosin (H&E), which is commonly used in conventional pathology. With SHG imaging, the ATC and DTC fibers were revealed to be of prognostic value based on the patient cohort and clinical outcomes. And when analyzed separately, the key collagen-associated parameters provided a novel understanding of collagen remodeling during cancer progression [3]. Image Credits: Institute of Molecular and Cell Biology.

Says Professor Tan Puay Hoon, Chairman, Division of Pathology, and Senior Consultant, Department of Anatomical Pathology, SGH, and lead pathologist of the study, «Critical biomarkers in TNBC are needed to stratify patients and predict clinical outcomes. Technological advances in pathology such as SHG assessment may improve the characterization of detailed and minute changes in important collagen features within the tumor stromal microenvironment – such as the collagen structure, density and length. These are important parameters that could possibly enhance pathological assessment and allow for a clearer understanding of the relationship between collagen features and tumor progression.»

Evaluating Therapeutic Efficacy with Key Collagen Parameters

The advantages of these novel collagen parameters make the platform a valuable asset in existing and future TNBC studies that are currently monitoring therapeutic responses in their exploration of targeted treatments. For instance, an ongoing collaboration between HistoIndex and a team at the Memorial Sloan Kettering Cancer Center (MSK), led by Professor Linda Vahdat, Chief of Medical Oncology and Clinical Director of Cancer Services at the MSK Physicians at Norwalk Hospital, is currently investigating influencing the tumor microenvironment with anti-copper therapy (copper depletion) for patients with breast cancer who are at a high risk of a relapse.

Having spent many years examining copper depletion in TNBC studies, Prof. Vahdat has previously explained the role of copper in triggering metastasis, and how the collagen scaffolding that houses the tumor breaks down once copper is pulled out of the system [4]. Says Prof. Vahdat, «Collagen within the tumor microenvironment represents an under-explored predictor of treatment outcome. Preliminary data from our group suggests that we can normalize the collagen microenvironment with a copper depletion strategy rendering an inhospitable environment for metastases. With this collaboration with HistoIndex, we hope to be able to predict those primary tumors that are amenable to this treatment strategy.»

Photo – https://www.luxembourgnewstoday.com/wp-content/uploads/2021/01/histoindex-explores-the-clinical-utility-of-stain-free-ai-digital-pathology-platform-in-388-patients-with-triple-negative-breast-cancer-tnbc-1.jpg