Development of a semi-automated method for tumor budding assessment in colorectal cancer and comparison with manual methods

Abstract

Tumor budding is an established prognostic feature in multiple cancers but routine assessment has not yet been incorporated into clinical pathology practice. Recent efforts to standardize and automate assessment have shifted away from haematoxylin and eosin (H&E)-stained images towards cytokeratin (CK) immunohistochemistry. In this study, we compare established manual H&E and cytokeratin budding assessment methods with a new, semi-automated approach built within the QuPath open-source software. We applied our method to tissue cores from the advancing tumor edge in a cohort of stage II/III colon cancers (n=186). The total number of buds detected by each method, over the 186 TMA cores, were as follows; manual H&E (n=503), manual CK (n=2290) and semi-automated (n=5138). More than four times the number of buds were detected using CK compared to H&E. A total of 1734 individual buds were identified both using manual assessment and semi-automated detection on CK images, representing 75.7% of the total buds identified manually (n=2290) and 33.7% of the total buds detected using our proposed semi-automated method (n=5138). Higher bud scores by the semi-automated method were due to any discrete area of CK immunopositivity within an accepted area range being identified as a bud, regardless of shape or crispness of definition, and to inclusion of tumor cell clusters within glandular lumina ('luminal pseudobuds'). Although absolute numbers differed, semi-automated and manual bud counts were strongly correlated across cores (ρ=0.81, p<0.0001). Despite the random, rather than 'hotspot', nature of tumor core sampling, all methods of budding assessment demonstrated poorer survival associated with higher budding scores. In conclusion, we present a new QuPath-based approach to tumor budding assessment, which compares favorably to current established methods and offers a freely-available, rapid and transparent tool that is also applicable to whole slide images.

The relationship between primary colorectal cancer histology and the histopathological growth patterns of corresponding liver metastases

“…As such, the results of this study should serve more as a stepping stone for a deep-learning digital-pathology approach in a larger cohort. Several deep-learning models already exist for the automated detection and classification of individual markers, for instance peritumoural budding [43], TIL density [44], and the fibrotic stroma type [45]. Additionally, studies have shown deep-learning on histopathology capable of predicting relevant outcomes in a hypothesisfree manner, i.e., not training the model to predict specific markers but instead let the model identify relevant features itself for accurate prediction of the outcome of interest.…”

Semi-supervised learning to automate tumor bud detection in cytokeratin-stained whole-slide images of colorectal cancer

“…Therefore, IHC is frequently used to help pathologists identify tumor buds (TB) [2]. Several studies involving experienced gastrointestinal pathologists have reported on the prognostic value of TB based on H&E staining, as well as on IHC [3][4][5]. At the same time, moderate to substantial inter-observer variability has been reported for TB scoring both in H&E and IHC, including our recent work [6].…”

“…Within these ROIs, tumor bud detection is achieved using classical image analysis operations, sometimes in combination with some form of machine learning. Fischer et al [4] use QuPath software to classify tumor buds in digitized CRC tissue microarrays (TMA) based on color and measured cluster size (area instead of nuclei number) and compare the results with manually obtained tumor buds yields in H&E and CK per count and in a bud-by-bud fashion. They analyze the impact of all methods on patient survival.…”