Aeschlimann, Bettina Leonore (2024). Artificial Intelligence and Immunohistochemical Analysis of the BRAF V595E Mutation in Canine Lower Urinary Tract and Prostate Tumors. (Thesis). Universität Bern, Bern
|
Text
24aeschlimann_bl.pdf - Thesis Available under License Creative Commons: Attribution (CC-BY 4.0). Download (4MB) | Preview |
Abstract
Canine urothelial carcinoma (UC) and prostate carcinoma (PC) frequently exhibit the BRAFV595E mutation, akin to the BRAFV600E mutation common in various human cancers. Since the initial discovery of the BRAF mutation in canine cancers in 2015, PCR has been the standard method for its detection in both liquid and tissue biopsies. During the scope of this doctoral thesis, we evaluated two different techniques for detection of BRAFV595E in tumors of the canine lower urinary tract and prostate. Recent advances in digital pathology and the power of artificial intelligence (AI) have opened up new possibilities for the detection of genetic alterations through AI histology, and offer a wide range of new opportunities in the field of diagnostic and predictive tumour marker detection. The aim of the first study was to test the efficacy of AI histology to predict the presence of BRAFV595E in canine bladder carcinomas and to assess its intratumoral heterogeneity. We used a commercially available AI histology software to predict BRAFV595E in whole slide images (WSI) of bladder UC stained with haematoxylin and eosin (HE), based on a training (n = 81) and a validation set (n = 96). Among 96 WSI, 57 showed identical PCR and AI-based BRAF predictions, resulting in a sensitivity of 58% and a specificity of 63%. The sensitivity increased substantially to 89% when excluding small or poor-quality tissue sections. Test reliability depended on tumour differentiation (p < 0.01), presence of inflammation (p < 0.01), slide quality (p < 0.02) and sample size (p < 0.02). This is the first study to utilize AI histology to predict BRAF mutational status in canine UC. Considering the similarity between the canine BRAFV595E and human BRAFV600E mutations, we hypothesized that immunohistochemistry (IHC) using a BRAFV600E-specific antibody could effectively identify the canine mutant BRAFV595E protein. For this second study, we tested 122 canine UC (bladder n=108, urethra n=14) and 21 PC using both IHC and digital droplet PCR (ddPCR) and included benign tissue as control as well. The results from ddPCR and IHC were concordant in 99% (135/136) of the tumors. Using IHC, BRAFV595E was detected in 72/122 (59%) UC and 14/21 (65%) PC. Staining of all benign bladder and prostate tissues was negative. If present, mutant BRAF staining was homogenous, with rare intratumor heterogeneity in three (4%) cases of UC. BRAFV595E was more prevalent in tumors with urothelial morphology, and less common in glandular PC or UC with divergent differentiation. This study establishes that BRAFV600-specific IHC is a reliable and accurate method for detecting the mutant BRAFV595E protein in canine UC and PC. Moreover, the use of IHC, especially with tissue microarrays, provides a cost-efficient test for large-scale screening of canine cancers for the presence of BRAF mutations. This advancement paves the way for further research to define the prognostic and predictive role of this tumor marker in dogs and use IHC to stratify dogs for the treatment with BRAF inhibitors. Keywords: Urothelial carcinoma, prostate carcinoma, BRAF, artificial intelligence, immunohistochemistry, PCR
| Item Type: | Thesis |
|---|---|
| Dissertation Type: | Cumulative |
| Date of Defense: | 22 February 2024 |
| Subjects: | 600 Technology > 610 Medicine & health 600 Technology > 630 Agriculture |
| Institute / Center: | 05 Veterinary Medicine > Department of Infectious Diseases and Pathobiology (DIP) > Institute of Animal Pathology |
| Depositing User: | Sarah Stalder |
| Date Deposited: | 21 Nov 2024 10:49 |
| Last Modified: | 02 Dec 2024 11:11 |
| URI: | https://boristheses.unibe.ch/id/eprint/5642 |
Actions (login required)
 |
View Item |