Richter's syndrome after allogeneic stem cell transplantation for chronic lymphocytic leukaemia successfully treated by withdrawal of immunosuppression, and donor lymphocyte infusion.

Development of high-grade non-Hodgkin's lymphoma is a possible complication of chronic lymphocytic leukaemia/small lymphocytic lymphoma, known as Richter's syndrome (RS). Treatment for RS includes systemic chemotherapy and, recently, allogeneic stem cell transplantation (SCT). We describe a patient with B-chronic lymphocytic leukaemia who developed RS 4 months after allogeneic SCT from an HLA-identical sibling. The RS presented with systemic symptoms, lymphadenopathy, pancytopenia and serum lactate dehydrogenase elevation. The patient was treated with immunosuppressive drug withdrawal and a donor lymphocyte infusion (DLI) of 1 x 10(7) CD3/kg, leading to the disappearance of all symptoms and the attainment of complete donor chimerism. After 18 months of the therapeutic DLI, the patient continues in complete remission.

Central and extrapontine myelinolysis following allogeneic peripheral haematopoietic progenitor cell transplantation. Favourable outcome in a patient with chronic myeloid leukaemia.

A 48-year-old-man in the first chronic phase of chronic myeloid leukaemia developed a central nervous system complication on day +57 after HLA-identical peripheral blood progenitor cell (PBPC) transplantation. The clinical picture evolved to a reversible pseudobulbar palsy requiring mechanical ventilation. MRI examination disclosed lesions typical of central and extrapontine myelinolysis (CEPM), which disappeared on a repeat examination 20 days later. The patient had received cyclosporine A (CsA) as GVHD prophylaxis and severe hyponatremia was detected 7 days after the first neurological sign. CEPM has been described in alcohol-induced liver disease, following rapidly corrected hyponatremia and associated with CsA in orthotopic liver transplantation. This is the first reported case of CEPM in PBPC transplantation, and CsA seems to have played a role in the development of this very serious complication.

Combined isolation of CD34+ progenitor cells and reduction of B cells from peripheral blood by use of immunomagnetic methods.

BACKGROUND: Malignant cells may contribute to relapse after autologous hematopoietic cell transplantation The effectiveness of a double immunomagnetic purging strategy combining CD34-positive with B-negative cell selection to purge peripheral blood progenitor cells (PBPCs) from patients with chronic lymphoproliferative disorders has been analyzed. STUDY DESIGN AND METHODS: Twenty-two CD34+ cell selections from patients with follicular lymphoma (n = 14), chronic lymphocytic leukemia (n = 6), mantle cell lymphoma (n = 1), and splenic marginal zone lymphoma (n = 1) were performed by use of a magnetic cell selector followed by a negative cell selection step with anti-CD19 monoclonal antibody bound to immunomagnetic beads. RESULTS: The PBPC components contained median CD34+ cells of 1.24 percent (range, 0.38-3.92%) and CD19+ cells of 1.83 percent (range, 0.06-69.7%). After positive selection (n = 22), 49 percent (range, 16-72%) of CD34+ cells were recovered with a purity of 93 percent (range, 24-99%). The double-positive and -negative selections (n = 20) yielded 57.5 percent of CD34+ cells (range, 33.4-79.4%) with a purity of 95 percent (range, 63-99%). Logarithms of B-cell reduction in the CD34+-cell-enriched B-cell-depleted component had a median value of 3.63 (range, 2.74-4.84 log) and CD19+ and CD5+ cells for chronic lymphocytic leukemia patients with more than 4.56 log (>3.6-5.6 log). Of 13 PBPC components that had a tumor-specific clonal signal, 10 became PCR negative after the double-selection procedure. CONCLUSION: Combined positive and negative magnetic cell selection achieves a high grade of tumor cell reduction with up to 77 percent of the grafts being negative for tumor-specific clonal signal by PCR analysis. This technique preserves an adequate recovery of progenitor cells able to engraft.

Frequency of HLA-DPB1 disparities detected by reference strand-mediated conformation analysis in HLA-A, -B, and -DRB1 matched siblings.

The role of HLA-DPB1 as transplantation antigen is controversial. The frequency and relevance of HLA-DPB1 mismatch in hematopoietic stem cell transplantation are unknown. To ascertain the rate of HLA-DBP1 mismatch in siblings that had been matched for HLA-A, -B, and -DRB1, reference strand mediated conformation analysis (RSCA) a high resolution HLA typing method was used. Locus-specific primers were used to amplify the HLA-DPB1 locus. The PCR product was then hybridized with two fluorescein-labeled references and the duplexes were analyzed after electrophoresis in a short polyacrylamide gel. Among the 113 pairs of individuals tested, six HLA-DPB1 mismatches were identified, which corresponds to a frequency of 5.31 % (95% confidence interval 3.20%-7.42 %).

Individually adjusted prophylactic donor lymphocyte infusions after CD34-selected allogeneic peripheral blood stem cell transplantation.

T cell depletion of the graft increases graft failure and relapse rate in allogeneic PBSC transplantation. Delayed lymphocyte add-back after T cell-depleted transplants might prevent these complications. We present 22 consecutive allogeneic PBSC transplants from related histocompatible donors with positive selection of CD34+ cells. Recipients received prophylactic donor lymphocyte infusions (DLI) depending on their risk of relapse and of developing GVHD. Patients were considered at high risk of relapse with AML > first CR, ALL > second CR, and CML in accelerated or blastic phase. Patients were considered at high risk of developing GVHD if older than 35 years, or with a donor sensitized through previous pregnancy or blood transfusion. Patients at high risk of relapse and low risk of GVHD were scheduled to receive three DLI. Patients at low risk of relapse and high risk of GVHD did not receive DLI. The remaining patients were scheduled to receive two DLI. The DLI were administered on days +28 (2 x 10(5)/kg), +60 (2 x 10(5)/kg) and +90 (2 x 10(6)/kg) after transplant. G-CSF mobilized peripheral stem cells from healthy donors were positively selected by an immunomagnetic method. The mean CD34+ cells and CD3+ cells infused were 4.4 x 10(6)(range 1.9-10.6) and 0.085 x 10(5) (range 0.01-0.67). Cyclosporin A was given to prevent GVHD. All the patients engrafted. Twenty-two prophylactic DLI were performed in 12 patients: seven developed acute GVHD (one case grade III-IV) and none presented pancytopenia. At a mean follow-up of 585 days (range 89-1103), 14 patients were alive in CR, one patient was alive in relapse, four patients had died of relapse and three had died of transplant-related complication. Individually adjusted prophylactic DLI at the doses we used with an escalating schedule allowed an acceptable GVHD rate and a good engraftment of donor hematopoiesis.

Disparity for the minor histocompatibility antigen HA-1 is associated with an increased risk of acute graft-versus-host disease (GvHD) but it does not affect chronic GvHD incidence, disease-free survival or overall survival after allogeneic human leucocyte antigen-identical sibling donor transplantation.

Disparity for the minor histocompatibility antigen HA-1 between patient and donor has been associated with an increased risk of acute graft-versus-host disease (GvHD) after allogeneic human leucocyte antigen (HLA)-identical sibling donor stem cell transplantation (SCT). However, no data concerning the impact of such disparity on chronic GvHD, relapse or overall survival are available. A retrospective multicentre study was performed on 215 HLA-A2-positive patients who received an HLA-identical sibling SCT, in order to determine the differences in acute and chronic GvHD incidence on the basis of the presence or absence of the HA-1 antigen mismatch. Disease-free survival and overall survival were also analysed. We detected 34 patient-donor pairs mismatched for HA-1 antigen (15.8%). Grades II-IV acute GvHD occurred in 51.6% of the HA-1-mismatched pairs compared with 37.1% of the non-mismatched. The multivariate logistic regression model showed statistical significance (P: 0.035, OR: 2.96, 95% CI: 1.07-8.14). No differences were observed between the two groups for grades III-IV acute GvHD, chronic GvHD, disease-free survival or overall survival. These results confirmed the association between HA-1 mismatch and risk of mild acute GvHD, but HA-1 mismatch was not associated with an increased incidence of chronic GvHD and did not affect relapse or overall survival.

Do corticosteroids add any benefit to standard GVHD prophylaxis in allogeneic BMT?

In a retrospective study, we compared 15 patients who received cyclosporine (CsA), methotrexate (MTX) and prednisone (PDN) and 15 patients who received CsA-MTX for GVHD prophylaxis after allogeneic BMT (HLA-identical sibling (n = 22), related one HLA mismatch (n = 1), unrelated matched donors (n = 6), unrelated one HLA mismatch (n = 1)). The primary objectives of this study were to compare the incidence of GVHD and post-transplantation complications. Secondary objectives were to compare relapse rate, transplant-related mortality and overall survival. The incidence of acute GVHD grade III-IV was similar between the two groups (P = 0.66), as was the incidence of chronic GVHD (P = 0.67). Incidence of arterial hypertension was significantly higher in patients who received prophylactic PDN, (P = 0.03) and more insulin treatment was required in this group (P = 0.003). We observed no differences in the incidence of infections or upper digestive tract bleeding. Musculoskeletal complications appeared earlier in the group which received PDN. With a median follow-up of 4.4 years, patients in the CsA-MTX group had better overall survival, 46.7% vs 13.3% (P = 0.026). Relapse was a more frequent cause of death in the CsA-MTX group, whereas procedure-related mortality was more frequent in the CsA-MTX-PDN group (P = 0.013). These results suggest that prophylactic prednisone when combined with cyclosporine and methotrexate adds no benefit in acute or chronic GVHD prevention and may increase the morbidity of allogeneic transplantation. Corticosteroids may be reserved for GVHD treatment.

Post-transplant lymphomas: a 20-year epidemiologic, clinical and pathologic study in a single center.

BACKGROUND AND OBJECTIVES: To study the incidence, clinical presentation, pathologic features and outcome of post-transplant lymphomas (PTL) during the past 20 years. DESIGN AND METHODS: We undertook a descriptive study of all biopsy-proven cases of PTL diagnosed in our hospital from 1979 through 1999. The average annual incidence rate of PTL was analyzed at 5-year intervals from 1979 to 1999. Risk ratios were estimated by comparing the incidence of PTL among transplanted patients with that of lymphoma observed in the general population of the region. Survival analysis was performed at the univariate level using the Kaplan Meier technique and at the multivariate level by Cox hazard models. RESULTS: Seventeen of 1,860 transplanted patients developed a PTL (0.9%). The risk of PTL was calculated to be almost 8-fold higher than the risk of lymphoma in the general population. The risk was highest among those who had received a heart transplant (RR=35.6). The mean time between transplant and the diagnosis of PTL was 31 +/- 29 months. Of all PTL, 88% were of B-cell origin and 53% of the cases tested were Epstein-Barr virus (EBV)-positive. The median survival was 24 months. The majority of patients with allograft involvement died within the 2 months following diagnosis (hazard ratio 5.3; 95% CI 1.4-20.7). INTERPRETATION AND CONCLUSIONS: Organ transplantation is a major risk factor for the development of lymphoma, a disease with a particularly bad prognosis when it develops at the site of the allograft. Early diagnosis and more specific treatment may improve PTL survival.

Combined positive and negative cell selection from allogeneic peripheral blood progenitor cells (PBPC) by use of immunomagnetic methods.

Twenty-four mobilized peripheral blood products from healthy donors for allogeneic transplantation were positively selected for CD34(+) cells and depleted of CD4(+) and CD8(+) cells (+/- selection) by combining clinical grade immunomagnetic methods. A sequential, "two-step" strategy combining positive selection of CD34(+) cells by use of the Isolex 300i (versions 1 and 2) device and T cell depletion (TCD) using the MaxSep device and a simultaneous, "one-step" method of CD34(+)cell selection and TCD using the Isolex 300i (software versions 1 and 2) have been investigated. Using these magnetic bead separation systems, two groups of sequential +/- selection (Isolex 300i version 1/MaxSep and Isolex 300i version 2/MaxSep) and two groups of simultaneous +/- selection (Isolex 300i versions 1 and 2) were analysed. In the sequential +/- selection, logarithms of TCD (CD3(+) cell depletion) obtained by the positive selection step had median values of 3.7 with the version 1 (n = 5) and 4.5 with version 2 software of the Isolex 300i (n = 5) (P = 0.07). Version 2 also gave a higher CD34(+) cell purity and yield than did version 1 (92% vs77%, P < 0.05 and 55% vs 34%, P = 0.3, respectively). Additional TCD obtained in the second step with the MaxSep device for the two groups had a median value of 0.9 log and 7% CD34(+)cell losses. In the simultaneous +/- selection, the Isolex 300i version 2 (n = 10) gave a median TCD of 5.1 log and version 1 (n = 4) of 4 log (P < 0.005). Higher CD34(+)cell purity and yield were also obtained with version 2 than with version 1 (97% and 76%, P < 0.005 and 57% and 39%, P = 0.07, respectively). These data indicate that simultaneous, "one-step" +/- selection in the Isolex 300i version 2 achieves a high TCD with a high CD34(+) cell purity and an acceptable CD34(+) cell yield.

Autologous stem cell transplantation for clinically aggressive non-Hodgkin's lymphoma: the role of preparative regimens.

We investigated the impact of the most commonly used preparative regimens on the outcome of 395 patients with diffuse large cell lymphoma (DLCL), consecutively reported to the registry of the Spanish GEL/TAMO. Among them, 139 (35%) were autografted in 1st CR, 86 (22%) in 2nd/3rd CR, 124 (31%) had chemosensitive disease and 46 (12%) had chemoresistant disease. Conditioning consisted of chemotherapy-only in 348 patients (BEAM, 164; BEAC, 145; and CBV, 39) and radiochemotherapy with CY and TBI in 47. Median times to granulocyte, platelet recovery and to discharge were significantly shorter in the chemotherapy-only group. Early transplant-related mortality was significantly higher when using CY-TBI. After a median follow-up of 28 months, overall survival (OS) at 8 years of patients conditioned with BEAM or BEAC (58% (95% CI 50-66%)) was more favorable than with CBV (40% (95% CI 24-56%)), and significantly better than with CY-TBI (31% (95% CI 18-44%)). Multivariate analysis revealed that patients conditioned with chemotherapy-only regimens had improved OS, disease-free (DFS) and relapse-free survival (RFS) when compared to those conditioned with CY-TBI. Status at transplant was also a powerful prognostic indicator. We conclude that preparative regimens consisting of chemotherapy-only seem more efficacious than CY-TBI as conditioning for DLCL, because of faster engraftment and greater anti-lymphoma effect, as indicated by improved OS, DFS and RFS.

[Incidence and survival of leukemias according to the different histologic subsets, in Tarragona, Spain, between 1980-1994]. / Incidencia y supervivencia de las leucemias según el tipo histológico en Tarragona, en el período 1980-1994.

BACKGROUND: To study the incidence and survival of leukemias according to the different histologic types in Tarragona, Spain. MATERIAL AND METHOD: Analysis of the information obtained from the Cancer Registry of Tarragona (Spain) between 1980-1994. The leukemias were classified in: acute lymphoblastic leukemia (ALL), acute myelogenous leukemia (AML), chronic lymphoid leukemia (CLL) and chronic myeloid leukemia (CML). The estimated incidence rates have been adjusted to the worldwide population, the percentage of the incidence annual change through Poisson regression models and the relative survival using the registry of death rate of Catalonia. RESULTS: The adjusted rate for leukemias between the period 1990-1994 was 8.0 per 100,000 inhabitants in men and of 5.2 in women, being the CLL the most frequent subtype. Regarding the trend of incidence an increase of the CLL of 2.2% annual (CI 95%, 0.1-6.6) in men and of 7.7% (CI 95%, 1.4-14.4) in women was observed. In the remaining subtypes, there was no trend, but the non-classified leukemias decreased a -10.8% annual (CI 95%, -15.0 to -6.4) in men and a -9.9% annual (CI 95%, 15.4 to -4.0) in women. 5-year relative survival (RS5y) for the total leukemias was 37.7% in men and 45.3% in women. It stands out the CLL with a RS5y of 64.8% in men and of 75.7% in women and childhood ALL with a RS5y of 83.0% in boys and of 84.9% in girls. CONCLUSIONS: In Tarragona, Spain, an increase of the CLL incidence has been observed suggesting an improvement in the diagnosis, parallel to a decrease of the non-classified leukemias. The survival in this cohort of patients was similar to the that reported in other european registries.

Autologous stem cell transplantation (ASCT) with immunologically purged progenitor cells in patients with advanced stage follicular lymphoma after early partial or complete remission: toxicity, follow-up of minimal residual disease and survival.

The role of autologous stem cell transplant (ASCT) in indolent lymphomas is a controversial issue. From 1994 to 1999, we performed ASCT with immunologically purged progenitor cells in 15 patients with advanced stage follicular lymphoma (FL) after early partial or complete remission. Results of the purging strategy and follow-up of minimal residual disease after transplant were analyzed with PCR amplification of bcl-2/IgH rearrangement for the t(14;18) translocation. A comparison of transplanted patients with a group of controls was carried out to evaluate differences in progression-free survival and overall survival. Eighty percent of patients received one chemotherapy regimen before ASCT and were in first remission. All the patients received cyclophosphamide plus hyperfractionated total body irradiation as the conditioning regimen. Nine patients were transplanted with bone marrow (BM) and six with peripheral blood progenitor cells (PBPC). Engraftment was delayed in one patient transplanted with BM. Two patients died during the transplant procedure. Ten of 12 evaluable patients were PCR positive in the BM for bcl-2 rearrangement at diagnosis. Six of them (60%) were still positive after chemotherapy, and one patient was transplanted with a positive hematopoietic product after purging. With a median follow-up of 27 months, six of eight evaluable patients still remain PCR negative in the BM. With a median follow-up of 4.7 years from diagnosis, progression-free survival was 83% (95% CI: 63-100). The risk of disease progression of non-transplanted patients was 19.2 times higher than that of transplanted patients (P = 0.01), but no differences were found in overall survival. Regarding patients in first remission, the risk of relapse was 12.6 times higher in non-transplanted than in transplanted patients (P = 0.04). This procedure seems to offer a good chance to achieve a PCR-negative state and prolonged freedom from recurrence. According to these results, prospective randomized trials are warranted.

HLA-B44 subtyping in the Catalan population using reference strand mediated conformation analysis. Implications for the selection of unrelated bone marrow donors.

Cytotoxic T lymphocytes (CTLs) reactive against the disparity between HLA-B*4402 and HLA-B*4403 have been reported after unrelated donor bone marrow transplantation. These CTLs have been associated with acute graft-versus-host disease and graft rejection. This study describes the HLA-B44-subtyping in the Catalan population using reference-strand mediated conformation analysis. It has been performed on 297 unrelated HLA-B44+ cord blood units from the Barcelona Cord Blood Bank (Barcelona, Spain). We have found a predominance of HLA-B*4403 (66.04%) over HLA-B*4402 (33.02%), whereas the predominant HLA-B44 allele in Northern Europe and the United States is HLA-B*4402. This inverted proportion between HLA-B44 subtypes in Mediterranean populations compared with other Caucasian populations suggests that HLA-B44 subtyping should be performed when an HLA-B44+ unrelated donor marrow is identified.

Genomic typing of minor histocompatibility antigen HA-1 by reference strand mediated conformation analysis (RSCA).

Disparities in minor histocompatibility antigen (mHAg) HA-1 are involved in the development of acute graft-versus-host disease (GvHD) in adult recipient after HLA-identical sibling donor hematopoietic stem cell transplantation. The mHAg HA-1 is an HLA-A*0201-restricted nonapeptide, which derives from the cleavage of a protein encoded at chromosome 19. The sequence analysis of HA-1 cDNA identified two alleles, termed HA-1H and HA-1R, which differ in only two nucleotides at 3' end of exon A, at positions 500 and 504. DNA-based methods for HA-1 typing were developed in 1998, using polymerase chain reaction with sequence-specific primers (PCR-SSP) and restriction fragment length polymorphism (PCR-RFLP). Here, we report the usefulness of reference strand mediated conformation analysis (RSCA), which was developed for mutation detection and typing of polymorphic loci, to discriminate between the two HA-1 alleles. We performed genomic typing of HA-1 locus in 203 HLA-A*0201-positive samples using RSCA and we confirmed these results by PCR-SSP. The results demonstrate the high reproducibility of this method and their strong correlation with the results obtained by PCR-SSP (99%). Only two samples showed disparity between the RSCA typing and the PCR-SSP. Direct sequencing of these samples confirmed that the correct allele assignment was that obtained by the RSCA typing. Furthermore, HA-1- RSCA-based typing provides additional information about the intronic structure of both alleles. With this approach, we describe the almost constant presence (99.2%) of a 5-bp deletion at intronic position 214-218 associated to the HA-1H allele, previously unidentified. We conclude that HA-1 genomic typing by RSCA is easy to perform and that could be used as a routine typing method.

Evaluation of engraftment of ex vivo expanded cord blood cells in humans.

Cord blood transplants (CBT) result in high rates of engraftment in patients transplanted because of inherited diseases even across marked HLA disparities, mostly in children, with less severe manifestations of GVHD than BM and PBSC transplants. Evaluation of engraftment potential of CBT based on early progenitor content is difficult due to their inaccurate quantification. Instead, post-thaw nucleated cell counts (Pt-NCC) are commonly used for this purpose. We have analyzed engraftment as a function of pre-freeze nucleated cell counts (Pf-NCC) in patients receiving CBT because of inherited diseases. We have observed median times to engraftment of 26 days or less, shortest times ranging 8 to 13 days, late engraftment or graft failures tending to be associated with age >15 years and infusions of <3.7 x 10(7)/Pf-NCC/kg. These data may be appropriate references to evaluate engraftment of CBT performed with previously ex vivo expanded cells. CBT performed with units of which one aliquot has been previously culture-expanded have resulted in times to engraftment similar to the ones observed in the above-mentioned analysis. In these trials it is not possible to trace the actual origin of the early engrafting cells because the pre-cultured cells lack differentiating markers. To better evaluate the engraftment dynamics of culture-expanded CB cells in humans, we have used a model of simultaneously transplanting cells from two different donors to the same patient. Preliminary results of patients that have simultaneously received one uncultured CB unit and culture-expanded purified CB CD34+ cells obtained from a second one show no significant contribution of cultured cells to early engraftment, and no prohibitive unfavorable immunological problems have been observed.

Follow-up of chimerism status after allogeneic HLA-mismatched stem cell transplantation by detection of non-shared HLA alleles.

BACKGROUND AND OBJECTIVES: Chimerism studies after allogeneic transplantation are usually performed using cytogenetic analysis, PCR-VNTR or PCR-STR. Here, we report an alternative method for following the chimerism status after an HLA-mismatched stem cell transplantation (SCT), detecting the presence of non-shared HLA alleles by reference-strand mediated conformation analysis (RSCA). DESIGN AND METHODS: We tested this new approach on allogeneic related haploidentical SCT, unrelated cord blood transplantation, and HLA-mismatched unrelated donor SCT. The quantification of the chimerism was performed by laser detection of fluorescent-labeled primers on an automated DNA sequencer. RESULTS: In all cases this technique was able to detect mixed chimeras. The technique detected above 5% of residual cells when the analysis was based on HLA-class I and above 3% for HLA-class II. This sensitivity is similar to that of the PCR-VNTR analysis. INTERPRETATION AND CONCLUSIONS: This method avoids the need to search for an informative locus (which is essential for PCR-VNTR or -STR). Moreover, we did not find the phenomenon of preferential amplification that is observed with most VNTR, thus avoiding the need for construction of standard curves to quantify mixed chimeras. We conclude that the detection of the non-shared HLA alleles by RSCA is a useful approach for chimerism follow-up after HLA-mismatched SCT.

Isolation of CD34+ progenitor cells from peripheral blood by use of an automated immunomagnetic selection system: factors affecting the results.

BACKGROUND: The isolation of CD34+ cells from mobilized peripheral blood is being increasingly used in the setting of allogeneic or autologous hematopoietic cell transplantation. Investigation of variables that may influence the effectiveness of CD34+ cell selection is of interest. STUDY DESIGN AND METHODS: Fifty-one CD34+ cell selections from peripheral blood progenitor cells (PBPCs) (39 allogeneic and 12 autologous) were performed using a magnetic cell separator (Isolex 300i, Baxter), including version 2.0 software. The results obtained were analyzed for different processing variables. The feasibility of transplanting these isolated CD34+ cells was also analyzed. RESULTS: The isolated CD34+ cell fraction had a median purity of 88.9 percent (range, 47.8-98.3). The median recovery of CD34+ cells was 45.1 percent (13.8-76.2), and the median colony-forming unit- granulocyte-macrophage (CFU-GM) content was 17. 2 percent (0.8-58.6). Logarithms of T- and B-cell depletion had median values of 3.7 and 2.8, respectively. The version 2.0 software of the Isolex 300i gave a higher CD34+ cell recovery in the enriched cell fraction (median 57.8%) than did version 1.11 (39.4%) or 1.12 (44.4%) (p = 0.01). The use of recombinant human deoxyribonuclease I during cell processing yielded more CD34+ cells (53% vs. 41%, p = 0. 01) and higher purity (92.8% vs. 87%, p = 0.03). There was a correlation between the percentage of CD34+ cells labeled with the monoclonal antibody 8G12 clone and the percentage of CD34+ cells labeled with the monoclonal antibody used during the processing technique (9C5 clone) in the initial, enriched, and depleted CD34+ cell fractions (R(2) = 0.95; 0.92; 0.78, p< 0.005, respectively). Median times for recovering >0.5 x 10(9) per L of granulocytes and >20 x 10(9) per L of platelets were 13 and 16 days in the allograft patients and 13 and 14 days in the autograft patients. CONCLUSION: CD34+ cells can be highly and effectively isolated from allogeneic and autologous grafts by use of this automated technique, with a high grade of T- and B-cell depletion. These purified CD34+ cell components can engraft normally.