Determination of threshold values for determining the size of the fraction of steroid hormone receptor-positive tumor cells in paraffin-embedded breast carcinomas.

BACKGROUND: For 4 years we used a multiparameter DNA flow cytometric (MP-FCM) technique to assess steroid hormone receptor expression in the diagnostic workup of routinely processed formalin-fixed, paraffin-embedded breast carcinomas as an alternative to immunohistochemistry (IHC) for the quantification of hormone receptor-positive cells. In all cases a positive fraction of hormone receptor-expressing epithelial cells was detected. This observation raised the question of what the cutoff value might be to distinguish receptor-negative from receptor-positive tumors. METHODS: In our search for a possible threshold value of positivity for estrogen receptor (ER) and progesterone receptor (PR) in MP-FCM, we developed four steps. First, we compared IHC results in our own laboratory with the results obtained by MP-FCM on a small series of breast tumors (n = 42). Second, after collecting our first 843 tumors, we made a comparison with the literature of the distribution of receptor positivity according to age classes. Third, using the most likely threshold that resulted from this comparison, we compared a subset of 340 node-negative tumors for their combined ER/PR profiles with the data from a similar group of node-negative tumor cases from the National Cancer Institute's Surveillance, Epidemiology and End-Result (SEER) study. Fourth, with the results of these comparisons, we prospectively collected IHC data and MP-FCM results of the same tumor samples for a period of 1 year. In this way, we collected data for an additional 180 tumors. RESULTS: The first step in this process resulted in an previous publication where 20% of steroid hormone receptor-positive cells seemed to be an acceptable cutoff point for positivity. However, the second step provided the best correlation at approximately 35% of ER reactive cells in the cytokeratin-positive cell population. With this cutoff, the distribution of combined ER/PR profiles in our patient population of node-negative breast cancers also showed a distribution similar to the data from the SEER study. The fourth step, using the 35% threshold value, resulted in a good correlation (r = 0.85, P < 0.0001) for ER and PR between IHC and MP-FCM in the 180 tumors investigated. CONCLUSION: By comparing in-house data with those from large external data collections in the literature, a threshold percentage can be defined that distinguishes steroid hormone receptor-negative from hormone receptor-positive tumors. As a result, information about DNA content and cell cycle distribution can be obtained. This observational study provides additional support to our opinion that MP-FCM is an alternative for IHC determination of ER and PR positivity. It is more objective and quantification can be done more appropriately. The additional value of this approach is that we generate continuous variables of ER/PR content instead of categorical classes, which can be used at different threshold levels for evaluation of clinical relevance.

Multiparameter flow cytometry as a tool for the detection of micrometastatic tumour cells in the sentinel lymph node procedure of patients with breast cancer.

AIM: To investigate whether multiparameter flow cytometry (MP-FCM) can be used for the detection of micrometastasis in sentinel lymph nodes (SLNs) in breast cancer. METHODS: Formalin fixed, paraffin wax embedded sentinel lymph nodes (n = 238) from 98 patients were analysed. For each lymph node, sections for haematoxylin and eosin (H&E) staining and immunohistochemistry (IHC) for cytokeratin (MNF116) were cut at three levels with a distance of 500 microm. The intervening material was used for MP-FCM. Cells were immunostained with MNF116, followed by an incubation with fluorescein isothiocyanate (FITC) labelled goat antimouse immunoglobulin. DNA was stained using propidium iodide. From each lymph node 100,000 cells were analysed on the flow cytometer. RESULTS: Thirty eight of the 98 patients with breast carcinoma showed evidence of metastatic disease in the SLN by one ore more of the three methods. In 37 of 38 cases where metastatic cells were seen in the routine H&E and/or IHC, more than 1% cytokeratin positive cells were detected by MP-FCM. In 24 patients, metastatic foci were more than 2 mm (macrometastasis) and in 14 these foci were smaller than 2 mm (micrometastasis). In three of these 14 cases, MP-FCM revealed positive SLNs, although this was not seen at first glance in the H&E or IHC sections. After revision of the slides, one of these three remained negative. However, MP-FCM analysis of the cytokeratin positive cells showed an aneuploid DNA peak, which was almost identical to that of the primary breast tumour. Duplicate measurements, done in 41 cases, showed a 99% reproducibility. In five of 14 patients with micrometastasis, one or two metastatic foci were found in the non-SLN. However, in 15 of 24 macrometastases multiple non-SLNs were found to have metastatic tumour. All micrometastases except for the remaining negative one mentioned above showed only diploid tumour cells, despite the fact that their primary tumours contained both diploid and aneuploid tumour cells. In primary tumours with more than 60% aneuploid cells, predominantly aneuploid macrometastasis were found, whereas diploid primary tumours only showed diploid micrometastases or macrometastases in their SLN. Aneuploid SLN macrometastases were associated with non-SLN metastases in five of seven patients, whereas diploid cases showed additional non-SLN metastases in only seven of 16 patients. CONCLUSION: In all cases, MP-FCM was sufficient to detect micrometastatic tumour cells in a large volume of lymph node tissue from SLNs. In some cases it was superior to H&E and IHC staining. Approximately 30% of SLN micrometastases are accompanied by additional non-SLN metastases. The size of the aneuploid fraction (> 60%) in the primary tumour may influence the risk of having both SLN and non-SLN metastases.