Laboratory validation studies in Ki-67 digital image analysis of breast carcinoma: a pathway to routine quality assurance

Artificial intelligence scale-invariant feature transform algorithm-based system to improve the calculation accuracy of Ki-67 index in invasive breast cancer: a multicenter retrospective study

BACKGROUND

Ki-67 is a key indicator of the proliferation activity of tumors. However, no standardized criterion has been established for Ki-67 index calculation. Scale-invariant feature transform (SIFT) algorithm can identify the robust invariant features to rotation, translation, scaling and linear intensity changes for matching and registration in computer vision. Thus, this study aimed to develop a SIFT-based computer-aided system for Ki-67 calculation in breast cancer.

METHODS

Hematoxylin and eosin (HE)-stained and Ki-67-stained slides were scanned and whole slide images (WSIs) were obtained. The regions of breast cancer (BC) tissues and non-BC tissues were labeled by experienced pathologists. All the labeled WSIs were randomly divided into the training set, verification set, and test set according to a fixed ratio of 7:2:1. The algorithm for identification of cancerous regions was developed by a ResNet network. The registration process between paired consecutive HE-stained WSIs and Ki-67-stained WSIs was based on a pyramid model using the feature matching method of SIFT. After registration, we counted the nuclear-stained Ki-67-positive cells in each identified invasive cancerous region using color deconvolution. To assess the accuracy, the AI-assisted result for each slice was compared with the manual diagnosis result of pathologists. If the difference of the two positive rate values is not greater than 10%, it was a consistent result; otherwise, it was an inconsistent result.

RESULTS

The accuracy of the AI-based algorithm in identifying breast cancer tissues in HE-stained slides was 93%, with an area under the curve (AUC) of 0.98. After registration, we succeeded in identifying Ki-67-positive cells among cancerous cells across the entire WSIs and calculated the Ki-67 index, with an accuracy rate of 91.5%, compared to the gold standard pathological reports. Using this system, it took about 1 hour to complete the evaluation of all the tested 771 pairs of HE- and Ki-67-stained slides. Each Ki-67 result took less than 2 seconds.

CONCLUSIONS

Using a pyramid model and the SIFT feature matching method, we developed an AI-based automatic cancer identification and Ki-67 index calculation system, which could improve the accuracy of Ki-67 index calculation and make the data repeatable among different hospitals and centers.

Ki-67 Proliferation Index Assessment in Gastroenteropancreatic Neuroendocrine Tumors by Digital Image Analysis With Stringent Case and Hotspot Level Concordance Requirements

OBJECTIVES

The Ki-67 proliferation index is integral to gastroenteropancreatic neuroendocrine tumor (GEP-NET) assessment. Automated Ki-67 measurement would aid clinical workflows, but adoption has lagged owing to concerns of nonequivalency. We sought to address this concern by comparing 2 digital image analysis (DIA) platforms to manual counting with same-case/different-hotspot and same-hotspot/different-methodology concordance assessment.

METHODS

We assembled a cohort of GEP-NETs (n = 20) from 16 patients. Two sets of Ki-67 hotspots were manually counted by three observers and by two DIA platforms, QuantCenter and HALO. Concordance between methods and observers was assessed using intraclass correlation coefficient (ICC) measures. For each comparison pair, the number of cases within ±0.2xKi-67 of its comparator was assessed.

RESULTS

DIA Ki-67 showed excellent correlation with manual counting, and ICC was excellent in both within-hotspot and case-level assessments. In expert-vs-DIA, DIA-vs-DIA, or expert-vs-expert comparisons, the best-performing was DIA Ki-67 by QuantCenter, which showed 65% cases within ±0.2xKi-67 of manual counting.

CONCLUSIONS

Ki-67 measurement by DIA is highly correlated with expert-assessed values. However, close concordance by strict criteria (>80% within ±0.2xKi-67) is not seen with DIA-vs-expert or expert-vs-expert comparisons. The results show analytic noninferiority and support widespread adoption of carefully optimized and validated DIA Ki-67.

Digital Image Analysis of Ki-67 Stained Tissue Microarrays and Recurrence in Tamoxifen-Treated Breast Cancer Patients

Purpose

The proliferation marker Ki-67 has been used as a prognostic marker to separate low- and high-risk breast cancer subtypes and guide treatment decisions for adjuvant chemotherapy. The association of Ki-67 with response to tamoxifen therapy is unclear. High-throughput automated scoring of Ki-67 might enable standardization of quantification and definition of clinical cut-off values. We hypothesized that digital image analysis (DIA) of Ki-67 can be used to evaluate proliferation in breast cancer tumors, and that Ki-67 may be associated with tamoxifen resistance in early-stage breast cancer.

Patients and Methods

Here, we apply DIA technology from Visiopharm using a custom designed algorithm for quantifying the expression of Ki-67, in a case–control study nested in the Danish Breast Cancer Group clinical database, consisting of stages I, II, or III breast cancer patients of 35–69 years of age, diagnosed during 1985–2001, in the Jutland peninsula, Denmark. We assessed DIA-Ki-67 score on tissue microarrays (TMAs) from breast cancer patients in a case–control study including 541 ER-positive and 300 ER-negative recurrent cases and their non-recurrent controls, matched on ER-status, cancer stage, menopausal status, year of diagnosis, and county of residence. We used logistic regression to estimate odds ratios and associated 95% confidence intervals to determine the association of Ki-67 expression with recurrence risk, adjusting for matching factors, chemotherapy, type of surgery, receipt of radiation therapy, age category, and comorbidity.

Results

Ki-67 was not associated with increased risk of recurrence in tamoxifen-treated patients (ORadj =0.72, 95% CI 0.54, 0.96) or ER-negative patients (ORadj =0.85, 95% CI 0.54, 1.34).

Conclusion

Our findings suggest that Ki-67 digital image analysis in TMAs is not associated with increased risk of recurrence among tamoxifen-treated ER-positive breast cancer or ER-negative breast cancer patients. Overall, our findings do not support an increased risk of recurrence associated with Ki-67 expression.

EndoPredict® in early hormone receptor-positive, HER2-negative breast cancer

Purpose

Evaluating consecutive early breast cancer patients, we analyzed both the impact of EndoPredict^® on clinical decisions as well as clinico-pathological factors influencing the decision to perform this gene expression test.

Methods

Hormone receptor (HR)-positive and human epidermal growth factor receptor 2 (HER2)-negative early breast cancer patients treated between 2011 and 2016 were included in this study to investigate the role of EndoPredict^® (EPclin) in the treatment of early breast cancer. A main study aim was to analyze the changes in therapy recommendations with and without EPclin. In addition, the impact of clinico-pathological parameters for the decision to perform EPclin was examined by Pearson's chi-squared test (χ²-test) and Fisher's exact test as well as univariate and multivariate logistic regressions.

Results

In a cohort of 869 consecutive early HR-positive, HER-negative breast cancer patients, EPclin was utilized in 156 (18.0%) patients. EPclin led to changes in therapy recommendations in 33.3% (n = 52), with both therapy escalation in 19.2% (n = 30) and de-escalation in 14.1% (n = 22). The clinico-pathological factors influencing the use of EPclin were age (P < 0.001, odds ratio [OR] 0.498), tumor size (P = 0.011, OR 0.071), nodal status (P = 0.021, OR 1.674), histological grade (P = 0.043, OR 0.432), and Ki-67 (P < 0.001, OR 3.599).

Conclusions

EPclin led to a change in therapy recommendations in one third of the patients. Clinico-pathological parameters such as younger age, smaller tumor size, positive nodal status, intermediate histological grade and intermediate Ki-67 had a significant influence on the use of EndoPredict^®.

Electronic supplementary material

The online version of this article (10.1007/s10549-020-05688-1) contains supplementary material, which is available to authorized users.