1. ¹Blood and Marrow Transplant Service, Westmead Hospital, Westmead, New South Wales, Australia
2. ²Sydney Cellular Therapies Laboratory, Westmead Hospital, Westmead, New South Wales, Australia
3. ³Western Clinical School, University of Sydney, Westmead, New South Wales, Australia

Correspondence: K Bradstock, E-mail: bradstok@icpmr.wsahs.nsw.gov.au

We wish to comment on the feasibility of peripheral blood stem cell (PBSC) harvest in patients with cold agglutinins, as little has been published on this. High-dose chemotherapy with autologous haemopoietic stem cell rescue is the accepted best treatment of chemotherapy-sensitive relapsed large cell non-Hodgkin's lymphoma (NHL).¹ Collection of autologous stem cells by leukapheresis is dependent on adequate venous access and blood flow through extra-corporeal tubing, and requires anticoagulation to prevent clotting within the circuit. Uncommonly, patients with NHL and other B-cell lymphoproliferative disorders may have circulating cold agglutinins, predisposing their blood to clumping when below 37°C, which would interfere with stem cell collection by apheresis.²

We performed successful PBSC harvest and reinfusion in a patient with a cold agglutinin related to relapsed diffuse large B-cell lymphoma (DLBCL). There was no published information on PBSC harvest in patients with cold agglutinins, and, since anecdotal reports from clinicians at other institutions were not encouraging, the procedure was approached cautiously.

A 62-year-old otherwise well male was initially treated for DLBCL with six cycles of CHOP chemotherapy and achieved a 14-year complete remission. He suffered a splenic relapse, confirmed on fine-needle biopsy, at which time an asymptomatic cold agglutinin was noted. Laboratory testing revealed a cold autoantibody with a titre of 512 for cord cells at room temperature and one in four at 37°C. The serum contained a monoclonal paraprotein of IgG lambda type measuring 4 g/l, but no IgM paraprotein. He had no clinical or biochemical evidence of haemolysis and was otherwise asymptomatic.

The patient received chemotherapy with intravenous ifosfamide, carboplatinum, etoposide and mesna, with radiological evidence of partial response.³ He was not treated with Rituximab, despite CD20 positivity, as this drug was not licensed for this indication at the time. Cold agglutination persisted on room temperature blood films after chemotherapy. Following the second cycle, he was given 10 g/kg G-CSF subcutaneously daily until PBSC harvest on day 15 postchemotherapy. There are no reports of PBSC harvest in patients with cold agglutinins, but successful total plasma exchange has been performed using continuous flow technology, warmed replacement fluids and a high room temperature.⁴ In light of this, our patient underwent plasmapheresis via a femoral venous catheter in an attempt to decrease the cold agglutinin titre and confirm that the procedure could be undertaken without blood clotting in the circuit. A neonatal cot overhead heater was positioned above the apheresis machine (Cobe Spectra), the room temperature increased to 30°C and the patient covered with a warming blanket. The return line passed through a blood warmer at 37°C. A 2000 ml plasma exchange with 4% albumin was performed without complication. The patient tolerated the procedure well.

The following day, a PBSC harvest was performed. In addition to the above measures, the collection bag was enclosed into a sterile bag placed in a 37°C water bath. A total of 12 l of blood was processed over 200 min with a collection volume of 194 ml. The collected cells were transported to the laboratory in the water bath. A total CD34⁺ cell harvest of 6.3 10⁶/kg was obtained prior to freezing. All laboratory equipment and reagents were prewarmed to 37°C. Prior to any manipulation, the collected cells were washed in warm saline. The specimen was then volume reduced by centrifugation and a plasma extractor. The cells were frozen in a rate controlled freezer at a final cell concentration of 19.6 10⁷/ml in freezing mix made up at room temperature containing a final DMSO concentration of 10%, and stored in liquid nitrogen. To check CD34 viability and clumping post-thaw, a test thaw was performed on a 10 ml pilot bag 48 h after freezing. No clumping of cells was seen after thawing at 37°C. Using a published flow cytometry assay, the viable CD34 count post-thaw was 3.8 10⁶/kg, representing a CD34 recovery of 68%.⁵

The patient underwent conditioning chemotherapy for autologous transplantation with BEAM (carmustine, etoposide, cytarabine and melphalan). A total of 3.8 10⁶ CD34/kg was reinfused 50 days after freezing. There were no adverse effects during the infusion, no evidence of immediate or delayed haemolysis, and other than a febrile episode at day +4, there were no post-transplant complications. Neutrophil engraftment (>1.5 10⁹/l) was achieved on day +14 and platelet engraftment (>50 10⁹/l) on day +28. The splenic lesion remained PET positive following transplant, and although antibody titres were not reassessed, autoagglutination persisted on blood films. The spleen was surgically removed 5 months post-transplant. The patient remains asymptomatic and in clinical remission 15 months post-transplant, with negative PET scans.

Cold agglutinins are observed occasionally in patients with lymphoproliferative disorders.² Unsuccessful attempts at stem cell harvest may result in the failure to offer potentially life-saving high-dose therapy. Our experience indicates that the presence of a significant cold agglutinin should not preclude autologous stem cell rescue. Successful harvest, storage and reinfusion of stem cells is possible in patients if close attention is paid to temperature control at all stages of management.