BONE MARROW TRANSPLANT FOR DYSKERATOSIS CONGENITA

Detection of plasma cells (PC) by flow cytometry is a problem because B lymphocytes gradually lose specific B-cell markers and surface-bound immunoglobulins during the differentiation pathway. In 1994 a monoclonal antibody (moAb), named B-B4 (CD138) and able to specifically mark PC among haemopoietic cells, was described (Wijdenes et al, 1994). CD138 recognizes syndecan-1, a heparan sulphate proteoglycan expressed on myeloma cells but not on other haemopoietic cells, has been suggested to play a role in the pathobiology of myeloma (Dhodapkar et al, 1998). We describe our experience in using CD138 by flow cytometry to detect PC in the preand post-CD34-positive selection leukaphereses products of 27 patients affected by multiple myeloma. All patients underwent peripheral blood stem cell harvesting after high-dose cyclophosphamide (7 or 4 g/m) þG-CSF at a median time of 2 months (range 1–6) after the end of first-line chemotherapy. A median of two apheresis per patient (range one to four) collected a median of 11 × 10/kg CD34 cells (range 7–25). For each patient a median of 8·8 × 10/kg CD34 cells (range 5–16) were positively selected with an avidin-biotin immunoaffinity device (Ceprate, CellPro) and a final median of 5 × 10/kg CD34 cells (range 2–10) was obtained. The median yield of the selection procedure was 58% (range 21–93%) and the median percentage of CD34 positivity in the final product was 80 (45–92). To evaluate PC contamination, the method described by Van Zaanen et al (1995) was used, with some modifications. Samples of 1 × 10 cells were first treated with FACS lysing solution, washed twice and incubated with CD138 (Immunotech, Marseille, France) alternatively coupled with CD38 and intracellular kappa and lambda light chains in 100 ml of a PBS saponine solution (0·02%) for 15 min in separate samples. Cells were washed twice and analysed with a FACScalibur cytofluorometer (Becton Dickinson, Calif.). At least 100 CD138/38 events were acquired and the intracellular light chain Ig pattern of CD138 cells was subsequently analysed; only PC having the same isotype of PC at the diagnosis in each patient were considered. Preselection samples contained a median percentage of 0·25 (range 0·003–1·6) PC, i.e. a median of 0·83 × 10/kg of PC (range 0·009–7·8), whereas in the post-selection products a median percentage of 0·1 (0·004–0·93), which corresponded to a median of 0·008 × 10/kg PC (range 0·0008–0·049) was detected. 16/27 patients had already been autotransplanted with a median of 2·8 × 10/kg CD34 positively selected cells (range 1·4–5); the median number of PC reinfused was 0·0032 × 10/kg (range 0·0008–0·03). After transplantation the median time to 1·0 × 10/l neutrophils was 12 d and was 13 d to 20 × 10/l platelets. A bone marrow biopsy performed 3 months after transplantation did not detect monoclonal PC in 12/16 (80%) patients. Gertz et al (1997) recently reported a borderline relationship between the absolute number of PC in the unmanipulated stem cell harvest and the relapse-free survival in 33 patients with myeloma. CD34-positive selection procedures are widely used to purge leukaphereses of patients affected by myeloma, since contaminating PCs can be a potential source of relapse. The method we used again demonstrated PC contamination in leukaphereses and the effectiveness of positive selection in decreasing the tumour load. Flow-cytometry enables analysis of large numbers of cells; the wide availability and the good reliability of this technique in quantifying small cell populations can contribute to acquiring more data about PC contaminating the stem cells harvests and to define the possible relationship between the quantity of PC reinfused and the clinical outcome in patients with multiple myeloma.