Self-report. Breast cancer treatment decisions are based not only on tumor size and lymph node condition, but also on tumor tissue analysis of histological grade and four biomarkers. Expression of estrogen receptor (ER) is primarily predictive of response to endocrine therapy, whereas expression of progesterone receptor (PR) is primarily predictive of value. When the gene is amplified for the tumor growth gene HER2 (human epidermal growth factor 2), increased expression of the HER2 receptor is observed, which is a predictor of anti-HER2 therapy. Additionally, diffusion marker Ki67 is analyzed.
There are standardized international guidelines for analysis and evaluation, and national quality assurances are organized nationally through collaboration between the breast disease group within KVAST (the Quality and Standards Committee within the Swedish Pathology Society), the Swedish Breast Cancer Group, NordiQC and Equalis. However, it cannot be said that the external quality assurance programs reflect the quality of clinical practice. Quality assurance evaluations are usually performed by a more experienced breast specialist. In the clinical routine, pathologists work with varying levels of competence, and evaluations are made under time pressure. For this reason, we wanted to study the analysis of histopathology and biomarkers in clinical routine and how different laboratory results influence treatment decisions.
We examined the results of the ER, PR, HER2, and Ki67 analyzes and the histopathology assessment in Swedish pathology clinics (29). All primary breast cancer patients diagnosed in Sweden between 2013 and 2018 were identified in the National Register of Quality of Breast Cancer, and 43,261 cases with complete data related to ER, PR, HER2, Ki67, histological score and treatment were selected for the study.
The proportion of ER positive cases in the different laboratories ranged between 84.2 and 97.6%, and 6 of the 29 laboratories had median values outside the 95% confidence interval (95KI). The proportion of people with PR disease ranged from 64.8 to 86.6 percent (7 laboratories outside of 95KI). The proportion of people with HER2 ranged from 9.4 to 16.3 percent (3 laboratories outside of 95KI), Fig. 1. The mean value of Ki67 in the different laboratories varied between 15 and 30 percent, and 19 of the 29 laboratories had significant variation between different years studied. The proportion of cases with grade II tumors ranged from 42.9 to 57.1 percent (13 labs outside of 95KI). There was a clear correlation between the proportion of cases assessed as ER and HER2 positive and the proportion of patients who received endocrine therapy and HER2 antagonist, respectively.
In ER, PR and HER2 analyzes, limited variation was observed between laboratories and between years of analysis. However, in the few laboratories that highlight (outliers), a clinically significant effect on the proportion of patients receiving treatment was observed, which is why further improvement in prognosis is needed. The high contrast in histological score assessment and Ki67 is of concern and underscores the importance of continuing education but also the introduction of image analysis techniques as decision support for routine pathology.
La Cartdingen 17-18/2021