Engraftment and molecular monitoring of CD34+ peripheral-blood stem-cell transplants for follicular lymphoma: a pilot study.

PURPOSE: A pilot study to validate the use of CD34+ selection (Ceprate SC) of blood stem-cell collection in patients with advanced follicular lymphoma receiving myeloablative chemoradiotherapy. PATIENTS AND METHODS: Seventeen patients were entered onto the protocol. Thirteen of 17 patients have undergone transplantation; the median age is 42.5 years (range, 33 to 51), seven of 13 are stage IVB, two stage IVA, three stage IIIB, and one stage IIB. All except two patients were treated after first or subsequent relapses after receiving cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) chemotherapy to achieve a good partial (six of 13) or complete (seven of 13) response before stem-cell mobilization with cyclophosphamide 3 g/m2 and filgrastim 300 microg once daily. RESULTS: Eleven of 13 patients had a detectable t(14;18) by nested polymerase chain reaction (PCR). Median CD34+ count before selection was 2.9 x 10(6)/kg (range, 1.17 to 11.3) and after CD34+ selection was 1.54 x 10(6)/kg (range, 0.88 to 7.6) with a median CD34+ yield of 62.4% (range, 17% to 95%) and purity of 60% (range, 39.3% to 73%). Of the 11 patients known to have t(14;18), 10 had PCR-detectable contamination of stem-cell harvests that became PCR negative in six of the 10 after CD34+ selection. Engraftment was rapid with a median day to absolute neutrophil count (ANC) greater than 0.5 x 10(9)/L of 13 days (range, 11 to 21) and platelet count greater than 20 x 10(9)/L of 14 days (range, 10 to 44). With a median follow-up duration of 15 months, three patients have remained persistently PCR-positive, two of whom received PCR-positive stem cells. Two have relapsed. Of the seven patients who received PCR-negative stem cells, five have had no PCR-detectable disease in posttransplant bone marrow samples. CONCLUSION: Longer follow-up duration is required to determine the significance of these findings, but we have confirmed the feasibility of CD34+ selected cells to deplete peripheral-blood stem cells of tumor cells from patients undergoing high-dose therapy for follicular lymphoma.

Mobilisation of peripheral blood stem cells with IVE and G-CSF improves CD34+ cell yields and engraftment in patients with non-Hodgkin's lymphomas and Hodgkin's disease.

The transplantation of mobilised peripheral blood stem cells is associated with more rapid engraftment than marrow transplantation. We have previously reported that G-IVE (G-CSF, ifosphamide, VP-16, epirubicin) improves the yield of CD34+ cells mobilised in patients with lymphoproliferative disorders compared with cyclophosphamide 3 g/m2 and G-CSF (G/CYCLO). In this study we have extended these observations to a larger series of patients including different lymphoma subtypes. Ninety-seven patients undergoing stem cell mobilisation were studied. Forty-two patients with lymphoproliferative disorders received G-IVE for mobilisation and 55 patients G/CYCLO. The median number of mobilised CD34+cells per leucapheresis was significantly higher for those patients receiving G-IVE: 5.82 x 106/kg (0.19-36) compared with 1.2 x 106/kg (0.04-17), P < 0.001 which resulted in a significantly reduced number of leucapheresis procedures performed in the G-IVE group. When patients were analysed dependent on underlying disease G-IVE mobilised significantly more CD34+cells per leucapheresis for all lymphoma types reaching 8.41 x 10(6)/kg (0.2-32) compared to 1.32 x 10(6)/kg (0. 06-17) for patients with high-grade NHL mobilised with G-IVE and C-GCSF respectively (P = 0.012). For patients with low-grade NHL 3. 12 x 10(6)/kg (0.10-24.39) compared to 1.08 x 10(6)/kg (0.04-9.74) were collected (P = 0.04) and for patients with Hodgkin's disease 3.02 x 10(6)/kg (1.48-36) and 1.04 x 10(6)/kg (0.1-12.3) (P = 0.001). Mobilisation with G-IVE resulted in the achievement of clinically significant CD34+ cell thresholds in a significantly higher proportion of patients compared to cyclophosphamide and G-CSF reaching >2.5 x 10(6)/kg CD34+ cells in 88% vs 62% (P = 0.004), >5 x 10(6)/kg in 67% vs18% (P = 0.001) and >10 x 10(6)/kg in 31% vs 14.5% (P = 0.05). Furthermore, an analysis of engraftment demonstrated that there was a significant reduction in the time to achieve platelet counts of >20 and >50 x 10(9)/l in patients receiving each incremental dose of CD34+ cells. We conclude that G-IVE mobilizes significantly more CD34+cells than G/CYCLO in patients with lymphoproliferative disorders. This effect is consistent in patients with high-grade NHL, low-grade NHL and HD and results in fewer failed stem collections and increased CD34+ cells available for transplantation which results in significantly accelerated platelet engraftment post transplant.

Stem cell mobilisation in lymphoproliferative diseases.

A number of different regimens have evolved for the mobilisation of peripheral blood stem cells for autologous transplantation in patients with lymphoma or myeloma. A successful regimen could be defined as one which consistently resulted in the collection of an optimal number of CD34+ cells with a minimum number of apheresis procedures with minimal toxicity. Initial protocols, which used chemotherapy alone as a mobilising agent, have now been replaced by regimens involving the use of haematopoietic growth factors either alone or in combination with variable doses of cyclophosphamide. Although there is good evidence that high-dose cyclophosphamide (6-7 g/m2) is an effective mobilising agent it is associated with significant toxicity and many groups have now utilised lower doses of cyclophosphamide with reduced toxicity which have still proven to be effective in the majority of patients. More recently a number of 'second generation' combined salvage chemotherapy and mobilisation regimens have been reported for use in the lymphomas which have the advantage of avoiding a specific stem cell mobilisation step and at the same time appear more consistently effective at mobilising stem cells than cyclophosphamide and G-CSF. These regimens are associated with fewer 'poor-mobilisers' and indeed some patients who have failed previous mobilisation with cyclophosphamide and G-CSF have been successfully re-mobilised. It is clear that in both lymphoma and myeloma patients the success of PBSC mobilisation is affected by the amount and type of previous chemotherapy and radiotherapy and probably other pre-treatment factors as exemplified by variability seen in normal donors mobilised with G-CSF alone. In myeloma most groups have utilised cyclophosphamide in variable doses in combination with G-CSF or GM-CSF. However, recent randomised studies have confirmed that G-CSF alone is an effective and nontoxic alternative although it appears that the efficacy of G-CSF as a single agent is related to the dosage used with daily doses of 16 microg/kg/day or greater being most effective. Thus, disease-specific mobilisation strategies appear to be emerging and these will undoubtedly be modified further as more is understood concerning the biology of blood stem cell mobilisation.

The effect of the serotherapy regimen used and the marrow cell dose received on rejection, graft-versus-host disease and outcome following unrelated donor bone marrow transplantation for leukaemia.

Unrelated donor (UD) transplantation is the only potentially curative therapy for many leukaemia patients but is associated with a high mortality and morbidity. We sought to identify factors that could be optimised to improve outcome following UD transplantation in adults. Data was retrospectively analysed on 55 patients sequentially receiving UD transplants for CML or acute leukaemia (AL), all of whom received serotherapy for the prevention of GVHD and rejection. All patients received standard conditioning regimens. The first 28 patients transplanted also received combined pre- and post-transplant serotherapy with Campath 1G (days -5 to +5) and standard dose CsA plus MTX as GVHD prophylaxis (protocol 1). The subsequent 27 patients received a 5-day course of pre-transplant serotherapy alone either with ATG (CML patients) or Campath 1G (AL patients) on days -5 to -1 inclusive, with high-dose CSA plus MTX (protocol 2). The incidence of acute GVHD was low with no patient receiving either protocol developing > grade 2 disease. The use of protocol 2 and the administration of a bone marrow cell dose above the median (2.17 x 10(8)/kg) were the most important factors predicting engraftment (P = 0.03 and P = 0.001, respectively) but this only remained significant for cell dose in multivariate analysis (P = 0.03). Overall survival for the group was 45% at 3 years and was influenced by both age (P = 0.02) and disease status at transplantation (P = 0.001). Receiving a cell dose above the median was also associated with a trend towards better survival (P = 0.08), due primarily to a reduction in the TRM to 8.2% compared with 54.5% in those receiving a lower cell dose (P = 0.002). We conclude that pre-transplant serotherapy alone is highly effective at preventing acute GVHD following UD BMT and that additional post-transplant serotherapy does not confer any benefit. Furthermore, a high marrow cell dose infused has a major effect in reducing transplant related mortality following UD BMT.

Low incidence of acute graft-versus-host disease and recurrent leukaemia in patients undergoing allogeneic haemopoietic stem cell transplantation from sibling donors with methotrexate and dose-monitored cyclosporin A prophylaxis.

One of the major aims of allogeneic haemopoietic stem cell transplantation has been the effective suppression of graft-versus-host disease (GVHD) without loss of a graft-versus-leukaemia effect. For GVHD suppression, one of the most frequently used regimens has been the combination of cyclosporin (CsA) and a short course of methotrexate (MTX) although the optimal usage of these agents remains unclear. Here, we report the results of 55 patients with standard risk leukaemia who have undergone allogeneic transplantation using either bone marrow (n = 48) or G-CSF mobilised peripheral blood stem cells (n = 7) using CsA and MTX for GVHD prophylaxis where the dosage of CsA was regularly adjusted to maintain a trough whole blood level of 95-205 ng/ml for the first 50 days post-transplant. To achieve this level of CsA in the immediate post-transplant period, over 40% of patients required dose adjustments of CsA as a result of sub-therapeutic levels on day +1 post-transplant. The achievement of CsA levels within the therapeutic range was expedited following the introduction of a sliding scale for dose adjustment. With this regimen we have observed a low incidence of acute GVHD with only 11% of patients developing > or =grade II disease. With a median follow-up of 66 months (range 8-132) the probability of relapse is only 6.6%. The disease-free survival probability for all patients was 72% at 5 years. These results demonstrate that effective GVHD prevention with CsA and MTX can be achieved without a high risk of recurrent leukaemia provided that rapid attainment of therapeutic CsA levels is achieved and maintenance within a low therapeutic range may help to maximise this effect.

Accelerated telomere shortening following allogeneic transplantation is independent of the cell source and occurs within the first year post transplant.

Telomere shortening has been documented in the blood cells of recipients of allogeneic bone marrow transplants compared with their donors. Allogeneic peripheral blood progenitor cells (PBPCs) have been increasingly used as an alternative to bone marrow. Their advantages include earlier engraftment and immune reconstitution following transplantation. We have measured telomere length of neutrophils and T cells in fully engrafted recipients of allogeneic bone marrow (n = 19) and allogeneic PBPC (n = 17) and also measured sequential telomere length in four patients after transplantation. Overall, significant telomere shortening occurred in recipients in neutrophils (0.3 kb, P < 0.001) and T cells (0.2 kb, P = 0.045). The data demonstrate that first, the degree of shortening was the same for BM and PBPC transplants and was not related to the time taken to engraft neutrophils and platelets and second, telomere shortening occurs in the first year post transplant without further shortening during the period of observation. These data suggest that the superiority of engraftment seen in PBPC transplants is independent of telomere shortening and other mechanisms such as homing or seeding may be more important.

Stem cell mobilization in normal donors for allogeneic transplantation: analysis of safety and factors affecting efficacy.

The use of peripheral blood stem cells instead of bone marrow as the source of haemopoietic cells for allogeneic transplantation is being increasingly explored. We have analysed data from 17 normal donors who underwent stem cell mobilization for allogeneic transplantation with an identical protocol using G-CSF at a dose of 10 micrograms/kg/d, with the first leukapheresis (LP) on the day following the fourth dose of G-CSF. Both G-CSF administration and leukapheresis were well tolerated. Donors under-went a median of two leukaphereses (range one to three) and a median of 6.80 x 10(6) CD34+ cells/kg recipient weight (range 2.4-15.6 x 10(6)) were collected. The median number of CD34+ cells per kg donor weight was 6.05 x 10(6), when corrected for a 12 litre leukapheresis, this gave a median total of 3.89 x 10(6) CD34+ cells/kg donor weight. When analysed with respect to factors which might influence the efficacy of mobilization, male donors were associated with a superior yield. The median number of CD34+ cells/kg/LP harvested was 4.96 x 10(6) in males and 2.79 x 10(6) in females (P < 0.05). The results suggested that, given a recipient of 75 kg, in a male donor a single 12 litre leukapheresis should yield sufficient CD34+ cells (4 x 10(6)/ kg), whereas a female donor would be likely to need two leukaphereses. Age was not found to affect donor yield. In summary, these data confirm that leukapheresis is a safe procedure in normal donors and suggest that males may be more efficient mobilizers of stem cells than females.

Stem cell mobilization in resistant or relapsed lymphoma: superior yield of progenitor cells following a salvage regimen comprising ifosphamide, etoposide and epirubicin compared to intermediate-dose cyclophosphamide.

We analysed the factors influencing the efficacy of peripheral blood stem cell (PBSC) collection in patients with lymphoma. Sixty-six patients underwent initial PBSC collection following mobilization with chemotherapy plus recombinant granulocyte colony-stimulating factor (300 microg/d). Patients were mobilized with one of two chemotherapy regimens, either cyclophophamide (3 g/m2 or 4 g/m2) (n = 50) or ifosphamide, etoposide and epirubicin (IVE; n = 16). The target of collecting > 2.0 x 10(6) CD34+ cells/kg was achieved in 43/66 (65%) patients with a median of two apheresis procedures. The IVE plus G-CSF mobilization regimen gave a significantly higher median yield of CD34+ cells (8.62 x 10(6)/kg) compared with cyclophosphamide plus G-CSF (3.59 x 10(6)/kg) (P = 0.045). The median yield of CD34+ cells per leukapheresis was almost twice as high in patients receiving IVE (1.94 x 10(6)/kg) compared to cyclophosphamide (1.03 x 10(6)/kg) (P = 0.035). In a univariate analysis of the factors affecting mobilization, the subtype of lymphoma (high-grade NHL) and the mobilization regimen were the only factors associated with high CD34+ cell yield. However, in a multivariate analysis of factors affecting mobilization including age, lymphoma subtype, previous chemotherapy and radiotherapy, only the use of the IVE protocol was predictive of a high yield of CD34+ cells. In 13 patients undergoing a second mobilization procedure the use of IVE was associated with a significantly higher yield of CD34+ cells compared to cyclophosphamide; three patients who failed cyclophosphamide plus G-CSF mobilization were able to proceed to transplantation following successful mobilization with IVE + G-CSF. These results demonstrate that IVE is a highly effective mobilization regimen which is superior to cyclophophamide and has the benefit of being effective salvage therapy for lymphoma patients.

Allogeneic haemopoietic stem cell transplantation for multiple myeloma or plasma cell leukaemia using fractionated total body radiation and high-dose melphalan conditioning.

We have evaluated the outcome of allogeneic haemopoietic stem cell transplantation for multiple myeloma using a conditioning regimen comprising fractionated total body irradiation and high-dose melphalan (110 mg/m2). The study comprised 25 patients (median age 49 years) who had been transplanted by either bone marrow (n = 13) or G-CSF mobilized peripheral blood stem cells (n = 12). Overall transplant-related mortality was 30% but was lower for patients < 50 years of age at transplant (21%). The main cause of treatment-related mortality was viral infection. Of the 19 patients evaluable post-transplant, 17 have so far achieved complete remissions. Currently, with a median follow-up of 3.4 years, 18 out of 25 patients are alive, of whom 15 are in continuing complete remission (CR) and 2 in second remission after suffering localized relapses, which were treated with radiotherapy and donor leucocyte infusions. Patients transplanted after 1 line of previous therapy, < 50 years of age and receiving peripheral blood stem cells (PBSC) rather than bone marrow (BM) had a superior outcome, although there was no statistically significant factor. We conclude that allogeneic transplantation should be considered as a potentially curative option for younger patients with myeloma and that the regimen using fractionated total body irradiation and melphalan has a high CR rate and a relatively low risk of treatment-related mortality, particularly in younger patients.

Generation of anti-idiotype immune responses following vaccination with idiotype-protein pulsed dendritic cells in myeloma.

Myeloma cells produce immunoglobulin which is unique to the malignant clone and presents antigenic determinants, or idiotypes, which may function as a tumour-specific antigen. The availability of significant quantities of idiotype protein in the serum makes immunotherapeutic strategies utilizing this protein to generate an anti-idiotype immune response an attractive prospect. We treated two patients with advanced refractory myeloma with a series of four vaccinations using autologous idiotype-protein pulsed dendritic cells combined with adjuvant GM-CSF. The vaccinations were well tolerated with a mild fever post-vaccination in one patient. An idiotype-specific T-cell proliferative response developed in both patients. This T-cell response was associated with the production of gamma-interferon, indicating a TH-1-like response. Furthermore, one patient developed anti-idiotype IgM antibodies. However, no idiotype-specific cytotoxic T-cell response could be demonstrated. Further investigation is warranted to define the optimal conditions for dendritic cell culture and priming to maximize the anti-tumour immune response.

A comparison of two different systems for CD34+ selection of autologous or allogeneic PBSC collections.

This study compared CD34 selection procedures using the CellPro CEPRATE and the Baxter Isolex 300 systems. Thirty-two procedures were performed, 19 CEPRATE and 13 Isolex. Median starting CD34 percentages were (CEPRATE/Isolex) 0.80% (range 0.24%-7.73%) and 0.85% (range 0.27%-10.17%), respectively (p = 0.788). After selection, there was a highly significant difference in purity of the product (CEPRATE/Isolex), 54% and 82% respectively (p < 0.0001). There was no significant difference in median recovery (CEPRATE/Isolex), 43% and 50%, respectively (p = 0.383). The starting CD34 percentage influenced the purity of the final product, and at high and low starting percentages, the Isolex produced superior purity. Improved efficacy of T cell depletion was observed with the Isolex, a median log depletion of 3.4 compared with 2.9 for the CEPRATE system (p = 0.012). In conclusion, the Isolex 300i produced a significantly higher purity CD34+ fraction, even at starting CD34+ levels of <0.5%, with no significant difference in recovery when compared with the CEPRATE system. The associated log T cell depletion is significantly improved with the Isolex system, with possible implications for use in CD34-selected allogeneic transplants.

Comparative serial quantitative measurements of chimaerism following unmanipulated allogeneic transplantation of peripheral blood stem cells and bone marrow.

Serial samples were collected from 38 patients following allogeneic transplantation using either unmanipulated peripheral blood stem cells (PBSC) (n = 18) or bone marrow (BM) (n = 20) to assess the incidence of mixed chimaerism using PCR amplification of five VNTR regions. After amplification, products were analysed using the Applied Biosystems ABI PRISM 377 Automated DNA Sequencer and GeneScan software GenoTyper program to determine a quantitative measure of chimaerism. The sensitivity of detection using this method was 0.1%. In the immediate post-transplant period (up to day 30) a significantly lower incidence of mixed chimaerism (MC) occurred in recipients of PBSC compared to BM (P < 0.0005). Between 1 and 6 months there was a significantly lower incidence of low-level MC in patients receiving PBSCT compared to BMT (4/14 v 8/11 respectively, P = 0.04) in patients who had not rejected their grafts or relapsed. Similarly, beyond 6 months 0/9 PBSCT patients compared to 4/9 BMT patients showed MC (P = 0.02). Beyond day 30 13/33 (39%) patients showed intermittent low-level MC, but this was not predictive for subsequent relapse. A rapidly increasing proportion of recipient haemopoiesis was predictive of graft rejection or relapse. Stable continuous MC without relapse was seen in one patient transplanted with PBSC for severe aplastic anaemia. These results suggest that the incidence of intermittent low-level MC is relatively high in the first 6 months following unmanipulated haemopoietic stem cell transplantation but reduces with time and is significantly lower in recipients of PBSC.