summary 1
Yung Sheng Lin
Dr. med
Detection of Tumor Cells in Leukapheresis Products from Patients with Breast Cancer
Using Immunocytochemical Staining Method
Geboren am 06, 03, 1965 in Taipei, Taiwan
Reifeprüfung am 01, 09, 1983 in Taipei, Taiwan
Studiengang der Fachrichtung Medizin vom WS 1983 bis WS 1988, WS 1990 bis SS 1995
Physikum am 01, 09, 1988 an der Universität Taichung, Department of Medical Technology,
Klinisches Studium in Tongji medizinischer Universität (1990-1994)
Praktisches Jahr in 1994-1995
Staatsexamen am 01, 07, 1995 an der Universität Tongji
Promotionsfach: Innere Medizin
Doktorvater: Prof. Dr. med. R. Haas
We have used a combination of four monoclonal antibodies (BM7, BM8 against MUC1, 5D3
aganist CK8, 18, 19 and HEA125 against human epithelial antigen) in a sensitive
immunocytochemical assay to identify breast tumor cells in PBPC. In model experiments in
which small numbers of MCF7 human breast tumor cells were added to PBMN cells, we
estimate that there is a 99,7% probability of detecting tumor cells at a prevalence of one tumor
cell per million cells by exassessing 4 slides with 1x106 MNC. The probability to detect tumor
cells at the same concentration assessing only one slide is 77 %. For evaluation of specificity,
we applied the APAAP staining using cocktail as first antibody to PB from healthy donors and
the LP products from patients with haematological malignancy. In some cases, cross-
staining of non-epithelial cells could be detected. The cross-reacting cells could be
distinguished from tumor cells by the combination of staining pattern and cytological details
of the cells in question.
In clinical specimens, the proportion of patients with tumor cell positive LP products in this
study was 14,3% (6/42) in the adjuvant treatment group, 18,2% (2/11), in the neoadjuvant
treatment group and 20,1% (9/43) in the group of patients with metastatic disease.
We analyzed the relationship between the number of PBPC collections and contamination of
tumor cells. By statistical analysis, a great number of leukapheresis products performed and
tested was significantly correlated to the probability to find a tumor positive LP (P<0,05). The
summary 2
number of leukaphereses depends on the yield of CD34+ cells per leukapheresis. To support
one cycle high-dose therapy, 2.5 x 106 CD34+ cells / kg body weight are needed to gurantee a
rapid haematological reconstitution . Our protocols include 2-3 cycle of high-dose therapy, at
least 5-7.5 x 106 cells/kg body weight should be harvested. On the other hand, the
heterogeneity in the patient population and pretreatment of chemotherapeutic agents might
result in difference in the mobilization capacity. Therefore, different number of LP are needed
for each patient.
Using cytospin preparations, the tumor cells in LP products can be quantified. Tumor cell
counts ranged from 0,25 to 5 cell(s) per 106 cells per LP products. Only in patients with
metastatic disease tumor cell concentrations greater than 1,25 cells/106 cells could be
detected. In the patients with stage II / III disease, tumor cell counts ranged from 0,25 to 1,25
cells per 1x106 cells. As a consequence, the median tumor cell concentration was higher in
specimens from patients with metastatic disease (median=0,96) than in specimens from
patients in the adjuvant and neoadjuvant treament groups (median = 0,5 and 0,75).
We found no significant difference between epithelial cell positive group and epithelial cell
negative group with respect to tumor size, LN involvement, tumor grade, histological type and
receptor.
In this study, the effectiveness of the Isolex 300 SA used for selection of CD34+ haemopoietic
propenitor and stem cells from blood-derived autografts of patients with high-risk breast
cancer. The Isolex 300 was found to be efficient in reducing the amount of contaminating
tumor cells. Prior to selection, five LP products contained malignant cells in a concentration
between one and two positive cells per 1x106 MNCs. All selected products obtained were free
of tumor cells. Since the selection procedure leads to a median 100-fold reduction in the
number of mononuclear cells and as the median purity of the selected fraction was >90%, a
purging efficienty of approximately 3 logs can be achieved.
We conclude that properly performed and controlled immunocytochemical staining of PB
cytospins is a sensitive and simple way to detect and quantitate breast cancer cells in PBPC
for autotransplantation. Isolex 300 SA for selection of CD34+ haemopoietic propenitor and
stem cells from blood-derived autografts resulted in reduction of tumor cells in PBPC. This
technique is a potent and efficient tool for the selection of CD34+ haemopoietic propenitor
and stem cells of patients with high-risk breast cancer.
