Astronomical algorithms for automated analysis of tissue protein expression in breast cancer

Abstract

Background:High-throughput evaluation of tissue biomarkers in oncology has been greatly accelerated by the widespread use of tissue microarrays (TMAs) and immunohistochemistry. Although TMAs have the potential to facilitate protein expression profiling on a scale to rival experiments of tumour transcriptomes, the bottleneck and imprecision of manually scoring TMAs has impeded progress.Methods:We report image analysis algorithms adapted from astronomy for the precise automated analysis of IHC in all subcellular compartments. The power of this technique is demonstrated using over 2000 breast tumours and comparing quantitative automated scores against manual assessment by pathologists.Results:All continuous automated scores showed good correlation with their corresponding ordinal manual scores. For oestrogen receptor (ER), the correlation was 0.82, P<0.0001, for BCL2 0.72, P<0.0001 and for HER2 0.62, P<0.0001. Automated scores showed excellent concordance with manual scores for the unsupervised assignment of cases to ‘positive' or ‘negative' categories with agreement rates of up to 96%.Conclusion:The adaptation of astronomical algorithms coupled with their application to large annotated study cohorts, constitutes a powerful tool for the realisation of the enormous potential of digital pathology.

Performance of automated scoring of ER, PR, HER2, CK5/6 and EGFR in breast cancer tissue microarrays in the Breast Cancer Association Consortium

“…Use of study-specific cut-off points for negative versus positive scores did not substantially improve the measures of agreement (data not shown). Measures of relative performance of automated versus visual scoring were similar when we used the Ali et al [15] method to select a cut-off point for positive and negative automated score (data not shown).The kappa statistics for HER2 Herceptest score showed substantial agreement for both inter-rater and automated to visual comparisons (kappa 5 0.62-0.71; Table 4). Although the agreement for the HER2 21 dichotomous classification was substantial for both inter-rater and rater-automated comparisons, the agreement for HER2 31 was only moderate for one of the raters.…”

“…For ER, PR and HER2, they were lower than previously reported by our group using the Ariol system [14], or an automated scoring algorithm adapted from astronomy [15], possibly reflecting the greater variability in tissue preparation related to multiple specimen sources. Bolton et al [14] used TMAs stained by ER, PR and HER2 from PBCS, and Ali et al [15] used ER and HER2 stained TMAs from SEARCH. While these TMAs were also included in our study, the automated image analysis of the TMAs was done independently using different methods.…”

Factors predicting pasireotide responsiveness in somatotroph pituitary adenomas resistant to first-generation somatostatin analogues: an immunohistochemical study

“…Expression of AIP was scored semi-quantitatively (H-score), taking into account both the percentage of positive cells (0-100%) and the staining intensity (0-3), total range 0-300 (34).…”

SERS-active Au/SiO_2 clouds in powder for rapid ex vivo breast adenocarcinoma diagnosis

“…This is why the use of gold decorated SiO 2 nanoparticles has been widely applied to SERS studies for biomedical purposes. In fact, this Au/SiO 2 powder is especially useful in the case of breast cancer tissue, where the expression of proteins plays an important role [42,43]. Moreover carboxylic groups of fatty acids may be covalently incorporated into silanol groups present on the surface of the Au/SiO 2 structures by Si-O-Si bonds [16,31,32,38,40,44].…”