In vitro and in vivo complement activation by dmso as a possibi.e cause of adverse reactions observed during reinfusion of frozen autologous peripheral stem cells

Dimethylsulfoxide (DMSO) is widely used as a cryoprotectant for human cells in the clinical setting of peripheral stem cell (PSC) support after high dose chemotherapy. Although PSC support is considered to be a safe procedure in general, it is nevertheless frequently associated with serious side effects like nausea, vomiting, flushing, acute abdominal pain, dyspnea, headache and hemoglobinuria, which might in part be due to complement activation. Since we have shown that DMSO is capable of activating the complement cascade in cellfree plasma, this study was designed to evaluate the in vitro and in vivo effects of DMSO on the generation of the activated complement fragment C3a in the settings of human autologous bone marrow or peripheral stem cell transplantation. Samples of frozen, DMSO-cryoprotected, autologous bone marrow and stem cells were drawn into EDTA prior to reinfusion of the autografts. Patients' blood samples were taken in EDTA before and 0,5h, 1,5h, 3,5h, 24h and 48h after the beginning of the reinfusion. Plasma was separated by centriftigation and stored at -80C until analysis by radioimmunoassay. Patients were clinically monitored for 48 hours and laboratory data were obtained by routine procedures. Before reinfusion of the autografts the patients' C3a plasma concentrations were above 1000 ng/ml (5545 ng/ml ±2258 ng/ml) in one group and under 1000 rig/ml (324 ng/ml ±221 ng/ml) in the other. In DMSO-protected autografts C3a-levels were much higher (5819 ng/ml ±4548 ng/ml in the first and 4302 ng/ml ±5352 ng/ml in the second group). Reinfusion of these autografts resulted in an up to 15,5-fold increase of the patients' C3a plasma concentrations (549 ng/ml to 8513 ng/ml) after 3,5 hours. This reflects an in vivo complement activation. Patients' maximum C3a plasma levels seem to correlate with the amounts of DMSO infused (r=0,527). The incidence of adverse reactions apparently correlates with the patients' C3a levels as most patients showed moderate side effects but those with large volumes of DMSO (172 ml, 78 ml) and high C3a plasma levels (14360 ng/ml, 25626 ng/ml) had serious reactions as fever, tachycardia, tachypnea and somnolence. This was also confirmed by the measurement of elevated levels of bilirubin, lipase, LDH, IL-6 and TNFa in patients who had received large volumes of DMSO. Although further experiments are needed to clarify the mechanisms of complement activation by DMSO, these results give rise to some concern regarding the increasing use of DMSO-protected frozen cells in various malignancies-.