A Multi-Institutional Study to Evaluate Automated Whole Slide Scoring of Immunohistochemistry for Assessment of Programmed Death-Ligand 1 (PD-L1) Expression in Non-Small Cell Lung Cancer.

Assessment of programmed death-ligand 1 (PD-L1) expression is a critical part of patient management for immunotherapy. However, studies have shown that pathologist-based analysis lacks reproducibility, especially for immune cell expression. The purpose of this study was to validate reproducibility of the automated machine-based Optra image analysis for PD-L1 immunohistochemistry for both tumor cells (TCs) and immune cells. We compared conventional pathologists' scores for both tumor and immune cell positivity separately using 22c3 antibody on the Dako Link 48 platform for PD-L1 expression in non-small cell lung carcinoma. We assessed interpretation first by pathologists and second by PD-L1 image analysis scores. Lin's concordance correlation coefficients (LCCs) for each pathologist were measured to assess variability between pathologists and between pathologists and Optra automated quantitative scores in scoring both tumor and immune cells. Lin's LCCs to evaluate the correlation between pathologists for TC was 0.75 [95% confidence interval (CI), 0.64-0.81] and 0.40 (95% CI, 0.40-0.62) for immune cell scoring. Pathologists were highly concordant for tumor scoring, but not for immune cell scoring, which is similar to previously reported studies where agreement is higher in TCs than immune cells. The LCCs between conventional pathologists' read and the machine score were 0.80 (95% CI, 0.74-0.85) for TCs and 0.70 (95% CI, 0.60-0.76) for immune cell population. This is considered excellent agreement for TCs and good concordance for immune cells. The automated scoring methods showed concordance with the pathologists' average scores that were comparable to interpathologist scores. This suggests promise for Optra automated assessment of PD-L1 in non-small cell lung cancer.

An international multicenter study to evaluate reproducibility of automated scoring for assessment of Ki67 in breast cancer.

The nuclear proliferation biomarker Ki67 has potential prognostic, predictive, and monitoring roles in breast cancer. Unacceptable between-laboratory variability has limited its clinical value. The International Ki67 in Breast Cancer Working Group investigated whether Ki67 immunohistochemistry can be analytically validated and standardized across laboratories using automated machine-based scoring. Sets of pre-stained core-cut biopsy sections of 30 breast tumors were circulated to 14 laboratories for scanning and automated assessment of the average and maximum percentage of tumor cells positive for Ki67. Seven unique scanners and 10 software platforms were involved in this study. Pre-specified analyses included evaluation of reproducibility between all laboratories (primary) as well as among those using scanners from a single vendor (secondary). The primary reproducibility metric was intraclass correlation coefficient between laboratories, with success considered to be intraclass correlation coefficient >0.80. Intraclass correlation coefficient for automated average scores across 16 operators was 0.83 (95% credible interval: 0.73-0.91) and intraclass correlation coefficient for maximum scores across 10 operators was 0.63 (95% credible interval: 0.44-0.80). For the laboratories using scanners from a single vendor (8 score sets), intraclass correlation coefficient for average automated scores was 0.89 (95% credible interval: 0.81-0.96), which was similar to the intraclass correlation coefficient of 0.87 (95% credible interval: 0.81-0.93) achieved using these same slides in a prior visual-reading reproducibility study. Automated machine assessment of average Ki67 has the potential to achieve between-laboratory reproducibility similar to that for a rigorously standardized pathologist-based visual assessment of Ki67. The observed intraclass correlation coefficient was worse for maximum compared to average scoring methods, suggesting that maximum score methods may be suboptimal for consistent measurement of proliferation. Automated average scoring methods show promise for assessment of Ki67 scoring, but requires further standardization and subsequent clinical validation.

Use and validation of epithelial recognition and fields of view algorithms on virtual slides to guide TMA construction.

While tissue microarrays (TMAs) are a form of high-throughput screening, they presently still require manual construction and interpretation. Because of predicted increasing demand for TMAs, we investigated whether their construction could be automated. We created both epithelial recognition algorithms (ERAs) and field of view (FOV) algorithms that could analyze virtual slides and select the areas of highest cancer cell density in the tissue block for coring (algorithmic TMA) and compared these to the cores manually selected (manual TMA) from the same tissue blocks. We also constructed TMAs with TMAker, a robot guided by these algorithms (robotic TMA). We compared each of these TMAs to each other. Our imaging algorithms produced a grid of hundreds of FOVs, identified cancer cells in a stroma background and calculated the epithelial percentage (cancer cell density) in each FOV. Those with the highest percentages guided core selection and TMA construction. Algorithmic TMA and robotic TMA were overall approximately 50% greater in cancer cell density compared with Manual TMA. These observations held for breast, colon, and lung cancer TMAs. Our digital image algorithms were effective in automating TMA construction.

Semi-automated imaging system to quantitate Her-2/neu membrane receptor immunoreactivity in human breast cancer.

BACKGROUND: Immunocytochemical methods for quantitating Her-2/neu immunoreactivity rest on subjective semi-quantitative interpretations with resulting interobserver, intraobserver, and fatigue variability. METHODS: To standardize and quantitate measurements of Her-2/neu immunoreactivity, we created epithelial-recognition and specific membrane-recognition algorithms, which could image breast cancer cells against a background of stroma, compartmentalize the cancer cell into nucleus, cytoplasm and membrane, and quantitate the degree of Her-2/neu membrane immunoreactivity based on both gray scale intensity and RGB colorimetric determinations. Image acquisition utilized either scanner or microscope with attached camera with a resolution of 20 pixels/10 microm. Areas of 150 whole slides were screened and the regions of interest manually selected for image processing. Three hundred TMA cores were directly processed. Images were acquired by jpg conversion of svs virtual slides or direct jpg photomicrograph capture. Images were then assessed for quality and processed. RESULTS: The digital algorithms successfully created a semi-automated imaging system whose algorithm-based ordinal values for Her-2/neu both strongly correlated with the subjective measurements (intraclass correlation: 0.84; 95% confidence interval: 0.79-0.89) yet exhibited no run variability. In addition, the algorithms generated immunocytochemical measurements of Her-2/neu on an expanded continuous scale, which more reliably distinguished true Her-2/neu positivity from true Her-2/neu negativity (determined by FISH) than subjective or algorithmic ordinal scale measurements. Furthermore, the continuous scale measurements could better resolve different levels of Her-2/neu overexpression than either subjective or algorithmic ordinal interpretation. Other semi-automated analysis systems have been used to measure Her-2/neu and other cellular immunoreactivities, but these either have required proprietary hardware or have been based on luminosity differences alone. In contrast, our algorithms are independent of proprietary hardware and are based on not just luminosity but also many other imaging properties including epithelial recognition and membrane morphology. CONCLUSION: These features provide a more accurate, versatile, and robust imaging analysis platform.