Comparing computer-generated and pathologist-generated tumour segmentations for immunohistochemical scoring of breast tissue microarrays

Abstract

Background: Tissue microarrays (TMAs) have become a valuable resource for biomarker expression in translational research. Immunohistochemical (IHC) assessment of TMAs is the principal method for analysing large numbers of patient samples, but manual IHC assessment of TMAs remains a challenging and laborious task. With advances in image analysis, computer-generated analyses of TMAs have the potential to lessen the burden of expert pathologist review.

Methods: In current commercial software computerised oestrogen receptor (ER) scoring relies on tumour localisation in the form of hand-drawn annotations. In this study, tumour localisation for ER scoring was evaluated comparing computer-generated segmentation masks with those of two specialist breast pathologists. Automatically and manually obtained segmentation masks were used to obtain IHC scores for thirty-two ER-stained invasive breast cancer TMA samples using FDA-approved IHC scoring software.

Results: Although pixel-level comparisons showed lower agreement between automated and manual segmentation masks (κ=0.81) than between pathologists' masks (κ=0.91), this had little impact on computed IHC scores (Allred; =0.91, Quickscore; =0.92).

Conclusions: The proposed automated system provides consistent measurements thus ensuring standardisation, and shows promise for increasing IHC analysis of nuclear staining in TMAs from large clinical trials.

Figure 1

Examples of Type 1 (red), Type 2 (green) and Type 3 (blue) disagreements. Annotations drawn by pathologist A (purple) and pathologist B (orange) are shown on the right overlaid on the original image.

Figure 2

A TMA spot (left) and colour-coded images showing types of disagreement between the two pathologists' manual annotations (top row), pathologist A and the algorithm trained by that pathologist (middle row), and pathologist B and the algorithm trained by that pathologist (bottom row).

Figure 3

Pie charts showing distribution of agreements, and Type 1, Type 2 and Type 3 disagreements between manual and automated segmentations.

Figure 4

Bland–Altman plot of percentage of positive cells identified in the Aperio software. TMA spots are shown by black dots (pathologist A) and blue diamonds (pathologist B).

Figure 5

Histogram plots of Allred scores and Quickscores extracted from manual and automated segmentations.