Search

Allogeneic stem cell transplantation (SCT) is an established therapy for patients with relapsed lymphoma, but the role of positron emission tomography (PET) scanning preallogeneic and postallogeneic SCT is uncertain. We investigated whether pretransplantation PET status predicted outcome after allogeneic SCT and whether PET surveillance after transplantation provided additional information compared with computed tomography (CT) scanning. Eighty consecutive patients with lymphoma who received a reduced-intensity allogeneic SCT were entered onto a prospective trial. PET and CT scans were performed before transplantation and up to 36 months after transplantation. Forty-two patients were PET-positive before transplantation. Pretransplantation PET status had no significant impact on either relapse rate or overall survival. Thirty-four relapses were observed, of which 17 were PET-positive with a normal CT scan at relapse. Donor lymphocyte infusion (DLI) was administered in 26 episodes of relapse and was guided by PET alone in 14 patients. These findings suggest that, in contrast to autologous SCT, pretransplantation PET status is not predictive of relapse and survival after allogeneic SCT for lymphoma. Posttransplantation surveillance by PET detected relapse before CT in half of episodes, often allowing earlier administration of DLI in patients with recurrent lymphoma, and permitted withholding of potentially harmful DLI in those with PET-negative masses on CT scans.

This study aimed to determine the outcome of patients with relapsed or refractory lymphoma who have an inadequate response to first-line salvage therapy (1 degrees ST) and who subsequently receive a second-line salvage regimen (2 degrees ST) with the intention of ultimately proceeding to high-dose therapy. The outcome of 57 patients [Hodgkin's Lymphoma 17, histologically-aggressive non-Hodgkin's Lymphoma (NHL) 26, histologically-indolent NHL 14] who received more than one modality of conventional-dose salvage therapy was analysed. Sixteen patients had a partial response (PR) to 1 degrees ST, but subsequently received 2 degrees ST because the PR was judged to be inadequate (iPR) because of persisting disease bulk or marrow infiltration. Of these 16 patients, 10 (63%) continued to respond to 2 degrees ST. Of the 15 patients who had stable disease following 1 degrees ST, 5 (33%) responded to 2 degrees ST. Only one of the 24 (4%) with progressive disease (PD) following 1 degrees ST, responded to 2 degrees ST. 25 of the 57 patients ultimately underwent stem cell transplantation. The 2-year progression-free survival (PFS) and the 3-year overall survival (OS) for all patients was 24% and 31%, respectively. Long-term survival was highly dependent on response to 1 degrees ST (P = 0.0001); in patients with PD following 1 degrees ST, the PFS and OS at 3 years was only 4%. This analysis indicates that patients with malignant lymphomas, who have PD on 1 degrees ST, are not rescued by subsequent salvage regimens. They should either be treated palliatively or novel approaches should be explored.

The role of allogeneic bone marrow transplantation in lymphoma remains uncertain. We have analyzed 1185 allogeneic transplants for lymphoma reported to the EBMT registry between 1982 and 1998 and compared the results with those of 14687 autologous procedures performed over the same period. Patients receiving allogeneic transplants were subdivided according to histology: low-grade non-Hodgkin's lymphoma (NHL) 231 patients; intermediate-grade NHL 147 patients; high-grade NHL 255 patients; lymphoblastic NHL 314 patients; Burkitt's lymphoma 71 patients; and Hodgkin's disease 167 patients. These patients received allogeneic transplants as their first transplant procedure. Actuarial overall survival (OS) at 4 years from transplantation was: low-grade NHL 51.1%; intermediate-grade NHL 38.3%; high-grade NHL 41.2%; lymphoblastic lymphoma 42.0% years; Burkitt's lymphoma 37.1%; and Hodgkin's disease 24.7% years. These outcomes are relatively poor because of the high procedure-related mortality associated with these procedures, particularly in patients with Hodgkin's disease (51.7% actuarial procedure-related mortality at 4 years). Multivariate analysis showed that for all lymphomas apart from Hodgkin's disease, status at transplantation significantly affected outcome. A matched analysis was performed: for all categories of lymphoma, OS was better for autologous than for allogeneic transplantation. Relapse rate was better in the allogeneic group for low-, intermediate- and high-grade, and lymphoblastic NHL. It was equivalent for Burkitt's lymphoma and worse in the allogeneic group for Hodgkin's disease. Allogeneic transplantation appears to be superior to autologous procedures in terms of producing a lower relapse rate. The toxicity of allogeneic procedures must however be reduced before this translates into an improvement in OS.

Peripheral blood progenitor cells (PBPCs) have become the stem cell source of choice in autologous transplantation. In a prospective randomised trial, we previously demonstrated that autologous transplantation using filgrastim-mobilised PBPCs resulted in faster haematopoietic recovery with shorter hospitalisation and reduced platelet transfusions compared to bone marrow transplant (BMT). This study is a follow-up analysis evaluating the long-term clinical outcome. Seventy-two patients with advanced Hodgkin's disease or high-grade lymphoma were randomised to receive either filgrastim-mobilised PBPCs (n = 37) or bone marrow (n = 35) after BEAM chemotherapy. Fourteen patients withdrew from the study before commencing high-dose chemotherapy. Fourteen of the 58 patients who received treatment with chemotherapy and transplant have died, 6 (19%) in the ABMT arm and 8 (30%) in the PBPC transplant (PBPCT) arm. Twenty-five patients (81%) in the ABMT arm and 17 (63%) in the PBPCT arm, who received treatment, were in complete remission at the date of last follow-up. Progression-free survival and overall survival (OS) were similar for both arms (OS 81% at 46 months for ABMT versus 63% for PBPC; p = 0.38). Further prospective studies with larger number of patients need to be done to assess which source of stem cells may translate into a long-term clinical benefit for the patient.

Cyclophosphamide 1.5 g m(-2) followed by granulocyte colony-stimulating factor (G-CSF) is an effective peripheral blood stem cell (PBSC) mobilizing regimen, but has limited anti-lymphoma activity. We therefore assessed the mobilizing potential of ESHAP (etoposide, ara-C, methylprednisolone and cisplatin), a potent second-line lymphoma regimen followed by G-CSF. The results were compared in 78 patients with relapsed or resistant lymphomas with the use of cyclophosphamide 1.5 g m(-2) followed by G-CSF in a matched pairs analysis, matching the ESHAP recipients (for predetermined prognostic factors) from a cohort of 178 lymphoma patients mobilized with cyclophosphamide and G-CSF. The total numbers of mononuclear cells collected at apheresis was similar with both regimens but ESHAP plus G-CSF resulted in a significantly higher percentage of CD34+ cells, absolute number of CD34+ cells and GM-CFC (all with P-values < 0.001). The number of patients requiring only one apheresis harvest to achieve a CD34+ cell yield of > 2.0 x 10(6) kg(-1) was greatly increased in the ESHAP recipients (56/78 vs 17/78, P < 0.001). The total number of progenitor cells collected was not significantly different with the two mobilization regimens because of this higher number of apheresis in the cyclophosphamide group. The proportion of patients who failed to achieve a minimum CD34+ cell target of 1 x 10(6) kg(-1) with the pooled harvests was less in the ESHAP arm (four patients vs nine patients) despite an increased number of aphereses in the cyclophosphamide recipients. ESHAP plus G-CSF is well tolerated and is an excellent mobilization regimen in patients with pre treated lymphoma.

We have performed a pilot study of MDR-1 gene transfer in patients receiving CD34-selected peripheral blood stem cell (PBSC) transplant for lymphoma. To ensure minimum engraftment thresholds and facilitate CD34 purification, mobilisation of > 2 x 10(6) CD34 cells/kg was a condition for recruitment. Of 11 patients counselled for study entry, only five achieved this target in a single apheresis. In three consenting patients, purified CD34 cells were exposed to A12M1 MDR-1 retroviral supernatant for 6 h, cryopreserved then thawed and readministered following ablative chemotherapy. No delay in engraftment was observed, although one patient received additional back-up cells. Gene transfer was demonstrated by polymerase chain reaction (PCR) for vector-derived MDR-1 cDNA sequence in all cases. Analysis of peripheral blood and bone marrow cells after transplant has, however, shown no evidence of in vivo gene transfer with a follow-up of 12, 15 and 18 months. The effect of MDR-1 substrate drugs has not yet been tested as all patients remain in clinical and radiological remission of their lymphoma. These results confirm the difficulty of achieving in vivo gene transfer in human haemopoietic cells and indicate major logistical constraints in PBSC mobilisation in patients with relapsed and resistant disease in whom initial studies are appropriate.

We have undertaken a prospective randomized study in 90 patients with relapsed or resistant lymphomas to assess the value of G-CSF (lenograstim) in the acceleration of myeloid recovery after peripheral blood stem cell transplantation (PBSCT). A common regimen of cyclophosphamide 1.5 g/m2 on day 1 and lenograstim 263 microg s.c. on days 2-10 with two aphereses on days 10 and 11 was used for stem cell mobilization. 77% of patients achieved an adequate PBSC collection in two harvests (> 2 x 10(8) MNC/kg or > 2 x 10(6) CD34+ cells/kg). 65 patients went on to receive high-dose BEAM chemotherapy and engraftment data was available for 62. 34 patients had been randomized to receive lenograstim 263 microg/d s.c. and 28 to no growth factor. The median time to ANC > 0.5 x 10(9)/l was 9 d in the lenograstim arm versus 12.5 d in the no-lenograstim arm (P=0.0001). This was associated with a median duration of time in hospital post PBSCT of 13 d in the lenograstim arm versus 15.5 d in the no-lenograstim arm (P=0.0002). Median days to platelet independence, platelet transfusions, incidence of infection and red cell transfusion were the same in both arms. These data indicate that lenograstim significantly accelerated myeloid recovery after PBSCT and shortened the duration of hospital stay.

Purpose: Treatment of selected patients with poor-prognosis lymphomas with high-dose chemotherapy and marrow or peripheral stem-cell rescue improves prognosis. A second course of myeloablative chemotherapy has been given to some patients, but few data are available on the indications, morbidity, and overall survival associated with this approach. This study was undertaken to evaluate morbidity and identify subgroups of patients who may benefit from a second transplant., Patients and Methods: Thirty-four patients with lymphoma given two cycles of myeloablative chemotherapy and entered onto the European Blood and Bone Marrow Transplant (EBMT) registry between 1982 and 1995 were included in this study: Hodgkin's disease (HD), n = 12; intermediate/high-grade non-Hodgkin's lymphoma (HG-NHL), n = 17; and low-grade non-Hodgkin's lymphoma (LG-NHL), n = 5. The reason for second transplant, status at transplant, conditioning regimen, morbidity, and both progression-free survival (PFS) and overall survival (OS) were assessed., Results: The second procedure was performed for the following reasons: (1) elective double procedure in four patients, (2) relapse after first transplant in 20, (3) partial remission (PR) after first transplant in eight, and (4) refractory disease after first transplant in two. The OS rate at 2 years for patients who underwent two transplants (estimated from the date of second transplant) was 49%, with a median follow-up time of 44 months. The OS rate at 2 years by histologic subtype was as follows; HD, 50%; HG-NHL, 60%; and LG-NHL, 0%. Seven of 15 patients with HD or HG-NHL who relapsed after they had achieved a posttransplant complete remission (CR) remain in CR 13 to 36 months after the second transplant, compared with two of 10 patients in CR (at 6 and 19 months after second transplant) who achieved a PR or had refractory disease after the first transplant. There were eight deaths (24%) before 3 months, of which three (9%) were transplant-related and the remainder due to persistent disease. Three late toxic deaths occurred: two of cardiovascular disease and one of secondary leukemia., Conclusion: Selected patients with HD and HG-NHL whose disease recurs after one transplant may benefit from a second transplant. Patients with refractory disease and LG-NHL did not benefit from a second transplant.

Purpose: To define parameters that predict for rapid engraftment after peripheral-blood stem-cell (PBSC) transplantation, progenitor thresholds, the proportion of patients who achieve these thresholds with a standardized mobilization regimen, and the factors that predict for mobilization efficiency., Patients and Methods: One hundred and one patients with pretreated lymphoma were mobilized with cyclophosphamide 1.5 g/m2 and granulocyte colony-stimulating factor (G-CSF), with the first apheresis performed when the recovery WBC count was > or = 5.0 x 10(9)/L. The relationship between the number of progenitor cells collected and patient age, sex, diagnosis, prior radiotherapy, and time since last chemotherapy was determined by multivariate analysis. The relationship between these factors, progenitor numbers returned, post-PBSC G-CSF, and hematologic recovery was performed in 81 patients following chemotherapy with carmustine (BCNU), etoposide, cytarabine, and melphalan (BEAM protocol)., Results: No BEAM recipients had delayed neutrophil recovery beyond 28 days. Delayed platelet recovery occurred in 7.4% and minimum and optimum thresholds of 1 x 10(6) and 3.5 x 10(6) CD34+ cells/kg and 1 x 10(5) and 3.5 x 10(5) granulocyte-macrophage colony-forming cells (GM-CFC)/kg were established. Hematologic recovery was adversely affected by prior treatment with mini-BEAM, and neutrophil recovery was accelerated by post-PBSC G-CSF. The minimum GM-CFC threshold was achieved with a single apheresis in 83% of patients and in 90% with two aphereses. The optimal threshold was achieved with two leukaphereses in 69% of patients. Prior radiotherapy adversely affected mobilization., Conclusion: Hematopoietic recovery following PBSC is dependent on progenitor-cell number infused and affect of previous chemotherapy on progenitor quality. Progenitor-cell mobilization is adversely affected by prior radiotherapy. The minimum threshold of GM-CFC required is achieved in most patients with a single apheresis, but an optimal collection usually requires at least two harvests.

Hodgkin's disease and non-Hodgkin's lymphomas can be treated and, in a large number of cases, cured by first-line chemotherapy or radiotherapy. Unlike many other malignancies, relapse is not uniformly fatal but the treatment is usually markedly myelotoxic with the high doses of chemotherapy used in relapse. Haematopoietic reconstitution with either autologous marrow or peripheral stem cells postchemotherapy has made high-dose chemotherapy relatively safe with mortality rates as low as 2% in some centres. The clinical indications for high-dose therapy in lymphoma management for patients with relapsed and bad prognosis disease are reviewed. The advantages of autologous bone marrow and peripheral stem cell transplants are compared and current peripheral stem cell mobilization and harvesting practice is discussed.

Gonadal damage, often associated with irreversible failure, is an invariable effect of high-dose myeloablative chemotherapy used for allogeneic or autologous bone marrow transplantation. Although not life threatening, the psychological consequences are significant. Therefore semen cryopreservation is advisable prior to initial therapy, but oocyte storage is not yet possible and embryo cryopreservation seldom feasible. Treatment protocols should be carefully selected to maximise cure rates whilst limiting injury to the reproductive system. Involvement of a reproductive endocrinologist is thus ideal from presentation, and certainly necessary pretransplant and for long-term follow-up in order to provide the endocrine assessment, counselling and fertility management which are essential for adequate transplant care. Improvement in bone marrow transplant technology and more rigorous patient selection have resulted in a higher proportion of long-term survivors. This has increased the awareness of long-term effects, such as gonadal failure, which may become apparent several months or years after transplantation. Premature ovarian and testicular failure may be overlooked by physicians managing the life-threatening disease, but for patients, the physical, emotional, psychological and social consequences of functional castration consequent on high-dose therapy can be devastating. The pathophysiology, diagnosis and management of gonadal failure are reviewed here.

We have previously demonstrated a dose response relationship in Hodgkin's disease for the combination of BCNU, VP16, Ara C and Melphalan, with the superior efficacy of the BEAM regimen requiring haemopoietic support, compared with miniBEAM. To further exploit this, we have attempted to escalate the VP16 dose in BEAM. The standard etoposide dose is 200 mg/m2 IV for four days. Thirty seven patients with refractory lymphoma received 400 mg/m2/day of etoposide, and 13 patients 600 mg/m2/day, in addition to BCNU, cytarabine, and melphalan. Toxicity and outcome parameters were compared in the preceding 40 patients, who received 200 mg/m2/day etoposide. The toxic mortality with 400 mg/m2/day of etoposide (3%) was identical to that for the standard BEAM regimen (5%). Two procedure related deaths occurred in the highest VP16 dose group (15%). The morbidity of the lower etoposide dose regimens was comparable, but 600 mg/m2/day induced significantly greater gastrointestinal toxicity. Twelve of the 13 patients receiving this dose suffered grade II-IV mucositis, with stomatitis, dysphagia and prolonged diarrhoea; 5 haemodynamically significant gastrointestinal haemorrhage, and 1 fatal toxic colitis. Granulocyte colony stimulating factor did not influence the nonhaematological toxicity. The three month response rates were similar (91%) in all dose cohorts. The maximum tolerable etoposide dose within the BEAM regimen is thus 400 mg/m2 for four days.

Thirteen post-pubertal male patients aged 17-25 years were assessed for pituitary-gonadal function 0-3 months prior to and 2-3 months post-bone marrow transplantation for haematological malignancy. All patients had multiagent cytotoxic treatment prior to transplantation and 30% were found to have germ cell dysfunction with abnormal semen parameters before high-dose therapy indicating damage to the germinal epithelium. They also had evidence of reduced Leydig cell reserve even before transplantation. During transplantation all patients sustained sustained gonadal injury, the effect on their germ cells being more pronounced than on the Leydig cells. Fifty per cent had reduction in testicular volume and all had azoospermia 2-3 months post-transplantation. Our results indicate that short-term chemoradiotherapy causes profound damage to the germ cell compartment of the testis, with less severe damage to the Leydig cells, but no overt injury to the anterior pituitary gland. The changes appeared to be identical in patients conditioned with total body irradiation-based protocols and those who received only high-dose chemotherapy prior to bone marrow transplantation.

A case-controlled analysis was performed to assess the effect of stem-cell source on autograft in a group of patients with malignant lymphoma reported to the European Bone Marrow Transplant Group (EBMT). The study was performed matching 83 patients autografted with peripheral blood stem cells (PBSC) with 83 autologous bone marrow transplantation (ABMT) patients. The case-matching was carried out following selection of the main prognostic factors for progression-free survival by multivariate analysis. The progression-free survival was similar in both types of transplants (38.5% PBSCT vs. 36.4% ABMT). The overall relapse and progression rate for the PBSCT was 51.2% compared with 50.1% for the ABMT patients. The differences were not significant statistically. The transplant-related mortality was 6% for both groups. The neutrophil and leucocyte recovery occurred faster in the peripheral blood stem-cell transplantation (PBSCT); the platelet recovery was not significant. A higher number of interstitial pneumonitis and fungal infection episodes were observed in the ABMT group. In conclusion, in these closely matched groups, there is no difference in PFS between patients undergoing PBSCT and those undergoing ABMT. However, the patients autografted with PBSC have a more rapid engraftment and a lower toxicity.

Granulocyte colony-stimulating factor (G-CSF) has been shown to be effective in accelerating neutrophil recovery after BMT. In this single centre study, we have examined the effect of delaying the start of treatment with G-CSF (5 micrograms/kg) until 8 days after BM infusion in a cohort of 17 patients with malignant lymphomas undergoing autologous BMT. In comparison with historical controls, neutrophil recovery to > 0.5 x 10(9)/l was shortened from 22 to 14 days (p < 0.01). Patients receiving G-CSF required iv antibiotics for a median 13 days (control 17, p < 0.05) and were discharged a median 28 days post-ABMT (control 32.5, p = NS). Patients received G-CSF for a median of 10 days only. Although these results require confirmation in a randomised trial they suggest that G-CSF administration could be delayed until 8 days after BMT without compromising efficacy and with an accompanying reduction in treatment costs.

Patients receiving multiple transfusions are at risk of acquiring hepatitis C (HCV) infection from a donor population which is unscreened for hepatitis C antibodies (anti-HCV). Prior to the introduction of blood donor screening for anti-HCV in the U.K., a group of patients undergoing therapy for haematological malignancies, with repeatedly abnormal liver function tests, were investigated for acute HCV infection. Thirty-two patients had repeatedly raised serum transaminases, and eight of these (25%) had evidence of an acute HCV infection. The diagnosis was made by the detection of HCV-RNA in the patients' serum using a complementary DNA/polymerase chain reaction (cDNA/PCR) procedure. All eight patients had received myeloablative chemotherapy and three had undergone bone marrow transplantation. HCV infection contributed significantly to the morbidity of this group of patients in the short term whilst they were undergoing treatment for their underlying haematological condition. The long-term effects have yet to be evaluated. In an attempt to decrease hepatic damage due to HCV, three patients were placed on interferon therapy. None showed a sustained reduction in serum transaminases or HCV viraemia. It is hoped that the introduction of anti-HCV screening of blood donors, will reduce the frequency of transfusion-acquired HCV infections. Early observations suggest that this is the case, as we have seen no new cases of HCV infection in our unit since the introduction of donor screening in September 1991.

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is active in enhancing the production of mature myeloid cells in vitro and several phase I/II clinical trials have suggested that its administration may accelerate neutrophil recovery after autologous bone marrow transplantation (ABMT). We have conducted a multicentre randomized double-blind placebo controlled trial in patients with poor prognosis malignant lymphoma receiving an identical high-dose combination chemotherapy regimen with ABMT. 61 patients were entered and 29 in each arm of the trial were evaluated. Treatment with GM-CSF did not affect the period of severe neutropenia (absolute neutrophil count (ANC) of < 0.1 x 10(9)/l) but accelerated recovery to an ANC of 0.5 x 10(9)/l (median 14 d v 20 d in controls, P = 0.001). There was no significant difference in platelet recovery between the groups (GM-CSF group platelet dependent for 25 d v control 19 d, P = NS). The number of positive blood cultures was similar in both groups (GM-CSF 14 v placebo 13) and there were no differences in days of fever > 37.5 degrees C (median 8 v 6) or days on parenteral antibiotics (11 v 10). Patients receiving GM-CSF had a median period of hospitalization following BMT of 24 d (control 25). No significant major toxicity attributable to GM-CSF administration was detected. We have confirmed in a randomized trial that GM-CSF accelerates neutrophil but not platelet recovery following ABMT. We were unable to demonstrate any accompanying changes in clinical outcome and believe that further trials are necessary to assess the clinical value of GM-CSF in BMT.

The pattern of changes in neutrophil myeloperoxidase (MPO) before, during and after bacteraemia was studied in 34 patients recovering from autologous bone marrow transplantation for relapsed Hodgkin's disease and non Hodgkin's lymphomas. Thirteen patients received haemopoietic growth factors (7 received M-CSF, 3 received G-CSF and 3 GM-CSF). The mean peroxidase index (MPXI) produced as part of a routine FBC performed by a flow cytochemistry blood autoanalyser (Technicon H*1) was used as a parameter to assess the MPO and subsequently the azurophil degranulation. The manufacturer's normal values for MPXI range from -10 to +10. Median MPXI on the day of documented bacteraemia was just below normal in the control and M-CSF groups (-10.8 and -8.9 respectively), but it was much below normal in the G-CSF (-16.5, P < 0.05) and even lower in the GM-CSF group (-39.6, P < 0.02); this correlated well with the decreased bacteraemia incidence in the last two groups. Although contact of neutrophils with bacterial chemoattractants resulted in primary degranulation in all groups, the pattern of changes in MPO content was different, suggesting that neutrophils primed in vivo with various haemopoietins respond to the challenge of microbial agents via different pathways.

Eighty-eight patients with haematological malignancies requiring Hickman catheters for intensive chemotherapy were randomized to receive either one single bolus intravenous injection of teicoplanin or no teicoplanin immediately before insertion of a double lumen Hickman catheter. There were lower incidences of catheter-related Gram-positive sepsis in patients receiving prophylactic teicoplanin. There were less exit site infections, tunnel infections and catheter-related Gram-positive septicaemia. The benefit of prophylactic teicoplanin was observed especially among patients who were already neutropenic at the time of catheterization. All Gram-positive organisms isolated from the infected skin sites or from blood cultures taken from the Hickman catheters were sensitive to teicoplanin. No adverse reaction was reported in any of the patients in the prophylactic group. Prophylactic teicoplanin therefore may be used routinely for patients requiring insertion of Hickman catheters for intensive chemotherapy to reduce the early incidence of catheter-related sepsis, especially during the associated period of neutropenia following chemotherapy.

The prognosis of patients with advanced refractory lymphoma remains poor. We have carried out a Phase II study of continuous high dose intravenous recombinant interleukin 2 alone without LAK cells in this group of patients. Eight patients have so far been treated, 4 with non-Hodgkins lymphoma and 4 with Hodgkins disease. Of the 7 evaluable patients, the maximum response observed was stable disease in 2 patients (1 with NHL and 1 with HD). The other patients' diseases progressed in the face of immune activation following the rIL-2 infusion. Adverse reactions were common but no life-threatening occurred. These results are disappointing. Whether or not lymphokine-activated killer (LAK) cells are needed to improve response rate deserve further investigations.

A case of T-cell lymphoma with an unusual phenotype is described. The majority of malignant tumour cells did not form E-rosettes, and lacked surface immunoglobulin but reacted with an antiserum to T cells and thymocytes. Localised acid phosphatase and non-specific alpha naphthol acid esterase activities and a prominent, convoluted, nuclear pattern in some neoplastic cells also supported the T cell derivation. The results demonstrate the importance of using a panel of markers for identification of the cellular derivation of some lymphomas.

50 cases of advanced, intermediate (18) and high grade (32) non-Hodgkin's lymphoma (NHL) including 16 with Burkitt lymphoma have been treated with very high dose chemotherapy and autologous bone marrow transplantation (ABMT). These cases represent a retrospective analysis of the combined experience of a recently established collaborative group. 31 patients were treated with a protocol used in Lyon, 12 with that used in Marseille and seven with that used in London. Although the details of drug administration differed, each protocol was based on high dose alkylating agent (cyclophosphamide or melphalan), BCNU and cytosine arabinoside. 16 patients had drug resistant progressive NHL. Of these 11 responded to high dose treatment (nine CR, two PR). The duration of CR in this group was short (median 104d) and only one patient was in CR at 1 year. 19 patients had relapsed on previous therapy but were still responding to conventional rescue therapy. Following high dose therapy 47% of these patients are in continuous CR with a median time of observation of 300 d (73-962 d). Seven patients were partial responders to conventional induction therapy. Of these, six had a CR with high dose treatment and are still in CR (range 39-1230 d, median 200 d). Eight patients received high dose therapy as intensification after a long delay to CR with conventional treatment. Of these, four are alive and in remission 124-763 d after treatment. The high dose protocols produced significant morbidity with 25 patients (50%) having major or minor treatment-related complications, and there were seven treatment related deaths (14%). However, these results indicate that durable responses can be obtained with high dose chemotherapy in patients who have been heavily treated and indicate a role for this type of treatment at an earlier stage in advanced non-Hodgkin's lymphoma.

Seventeen patients with advanced lymphoma were treated with high-dose chemotherapy with autologous bone marrow rescue. In 11 patients with non-Hodgkin's lymphoma (NHL) there were 2 complete remissions (CRs) and 2 partial remissions (PRs), and in 6 patients with Hodgkin's disease there were 5 CRs. Three patients remain well in unmaintained remission (days 874, 446 and 351), and a further 2 are alive and still receiving treatment (days 650 and 558). This type of therapy appears useful and should now be considered earlier in the course of the disease.

This short review of autologous bone marrow transplantation (ABMT) has a distinct clinical emphasis and concentrates particularly on adult acute leukaemia and lymphoma in which the greatest amount of current clinical experience lies. In the early part of the review we discuss how escalations of dose of chemoradiotherapy might allow ablation of both marrow-derived (leukaemia) and non-marrow-derived disease (lymphoma and solid tumour) provided that haemopoiesis is reintroduced into the host in the form of autologous marrow stem cells, and how cryopreservation techniques have allowed this to proceed. Whilst discussing ABMT in acute leukaemia we describe initial results possibly suggestive of an improvement on current consolidation/maintenance chemotherapy regimens but emphasis that we are dealing only with heterogeneous registry data and not randomised controlled trials. We also suggest that there is no useful data as yet as to the value of purging autologous acute leukaemia marrow. The lymphoma data is described which may suggest a useful role of high dose therapy with ABMT in relapsed disease-timing of ABMT may need to differ profoundly in HD from NHL. Current ABMT data in lymphoma suggests that local relapse at sites of previous disease remain the major problem and emphasises the difficulties of finding satisfactory ablative regimens and timing the selection of patients at particular points in the natural history of their disease. Finally, we emphasise that although solid tumours may numerically represent the largest group of potential candidates for ABMT, the picture in this area remains essentially one of failure to be able to ablate the underlying disease despite increments in chemoradiotherapy and ABMT.

Twenty-three cytopenic episodes in 18 patients undergoing ablative chemotherapy for the treatment of leukaemia or lymphoma were monitored from commencement of treatment until recovery, by automated differential counts using the Technicon H*1 Autoanalyser, with particular reference to abnormal white cell flags and large unstained cell (LUC) percentage. The blast flag was indicated in this recovery phase in 100% of patients and in 85% this preceded bone marrow recovery (defined as neutrophil count greater than 0.5 X 10(9)/l) by a mean of 10 days. On average the blast flag was indicated for 8 days in total. Bone marrow function continued to improve in all patients with no evidence of relapse. An increase in the LUC percentage on the differential count reached a maximum at 18 days, 6 days prior to marrow recovery. The ability to detect impending marrow recovery by means of the positive blast flag, may be of great value when patients have been cytopenic for many days.