Dose-escalated CHOEP for the treatment of young patients with aggressive non-Hodgkin's lymphoma: II. Results of the randomized high-CHOEP trial of the German High-Grade Non-Hodgkin's Lymphoma Study Group (DSHNHL).

BACKGROUND: The addition of etoposide to combination chemotherapy with cyclophosphamide, doxorubicin, vincristine and prednisone [etoposide to combination chemotherapy with cyclophosphamide, vincristine and prednisone (CHOEP)] improved outcome of young patients with good-prognosis aggressive lymphoma. To improve results further, the maximal dose-escalated version of CHOEP-21 tolerable without stem-cell support (high CHOEP: cyclophosphamide 1400 mg/m2, doxorubicin 65 mg/m2, vincristine 2 mg, etoposide 175 mg/m2 x3, prednisone 100 mg x5) was compared with CHOEP-21. PATIENTS AND METHODS: Intention-to-treat analysis of 389 young (18-60 years) patients with good-prognosis (age-adjusted International Prognostic Index = 0, 1) aggressive lymphoma randomized to CHOEP-21 (n = 194) or high CHOEP (n = 195). RESULTS: There was no difference in 3-year event-free (64% versus 67%; P = 0.734) or overall survival (83% versus 87%; P = 0.849). Neither low-risk nor low-intermediate risk patients benefited from high CHOEP. High CHOEP was more toxic than CHOEP-21 (grades 3 and 4 leukocytopenia 100% versus 87.2%, P < 0.001; thrombocytopenia 80.8% versus 9.6%, P < 0.001; infections 35% versus 11%, P < 0.001; therapy-associated deaths 3.1% versus 0%, P = 0.03). CONCLUSION: Dose-escalated CHOEP-21 does not provide clinical benefit for young patients with good-prognosis aggressive lymphomas. Since differences between chemotherapy regimens are compressed by the addition of rituximab, the results of this trial have bearing on strategies aiming to improve outcome of good-prognosis aggressive lymphomas in the rituximab era.

Severe lymphocytopenia and interstitial pneumonia in patients treated with paclitaxel and simultaneous radiotherapy for non-small-cell lung cancer.

PURPOSE: In a phase II trial with paclitaxel and simultaneous radiotherapy in non-small-cell lung cancer (NSCLC) patients, an unexpected high incidence of interstitial pneumonias was observed. The type of immunodeficiency associated with this treatment approach is characterized. PATIENTS AND METHODS: Fifteen patients with inoperable stage IIIA/B NSCLC were treated with paclitaxel as a 3-hour infusion on day 1 in weeks 1 to 3 and 6 to 8 at dose levels between 50 mg/m2 and 86 mg/m2 and with simultaneous radiotherapy in daily doses of 2 Gy, 5 days per week, in weeks 1 to 3 and 6 to 8 up to a total dose of 56 Gy. Hematologic parameters and lymphocyte subsets were monitored. RESULTS: Fourteen patients are assessable for response. The overall response rate was 78%, with four major responses, six partial remissions, and four minor responses. The major toxic effect observed was a moderate to severe protracted lymphocytopenia (380 +/- 310/microL) in all patients. Seven patients developed moderate to severe interstitial pneumonia; one had an additional herpes zoster infection, while an eighth patient had a cytomegalovirus infection. During treatment, all lymphocyte subsets were reduced, as follows (n = 9, mean +/- SD): CD4+ T cells (100 +/- 90/microL), CD8+ T cells (130 +/- 160/microL), natural killer (NK) cells (70 +/- 80/microL), and B cells (20 +/- 10/microL). Thus, the most pronounced toxicity was seen in CD4+ T and B cells. There was no recovery of lymphocyte subsets during a 3-month follow-up period. CONCLUSION: Paclitaxel with simultaneous radiation induces lymphocytopenia and promotes opportunistic infections. Long-term antibiotic and antimycotic prophylaxis is recommended. Whether the lymphocytopenia is an additive effect of paclitaxel and radiation or whether it can be induced by low-dose weekly paclitaxel alone remains to be determined.

Modulation of functions of granulocytes by recombinant human GM-CSF and possible complications of GM-CSF therapy.

Besides its effect on bone marrow progenitors, GM-CSF is able to modulate functions of mature cells such as neutrophils. It inhibits random migration and chemotaxis through action on both cells and chemotactic factors, and stimulates oxidative metabolism as well as elastase release. Furthermore, it strongly enhances the response of the cells to the usual stimulants such as f-Met-Leu-Phe and phorbol esters. The role of neutral proteinases and activated oxygen species in different diseases such as ARDS, emphysema, coagulation defects, arthritis, and inflammation, is recognized. The remarkable in vitro release of neutral proteinases and activated oxygen species from granulocytes after GM-CSF stimulation may be of importance in vivo. This should be considered in clinical application of GM-CSF, particularly with high-dose therapy.

Altered expression of the retinoblastoma gene product in human high grade non-Hodgkin's lymphomas.

The retinoblastoma gene (RB) is a growth suppressor gene on the human chromosome 13q14. It encodes a 105 kDa phosphoprotein (p105), with DNA-binding capacity. P105 is thought to be involved in cell cycle control. Inactivation of RB is responsible for the development of retinoblastomas and occurs frequently in osteosarcomas and small cell lung cancer. In this study we looked at the RB-structure and expression in cell lines and primary lymphoma samples from patients with high grade non-Hodgkin's lymphoma (NHL). Forty five primary high grade NHL, the B-lymphoblastoid cell line IM-9 and the NHL cell line WSU-NHL were studied for RB structure by Southern blotting and for RB-expression by Northern blotting, Western blotting and immunocytochemistry. In all experiments freshly cryopreserved material was used. Southern and Northern experiments were performed with the 0.9 kb and 3.8 kb RB-cDNA probe. For the detection of p105 two different anti-p105-monoclonal antibodies were used in immunocytochemistry and Western blotting experiments. No RB mRNA and no p105 could be found in IM-9 cells. Twenty six high grade NHL samples (58%) showed no p105 expression. In the subgroup of centroblastic lymphomas 16 out of 21 and in Burkitt's lymphomas five out of eight showed no p105-expression. P105 expression is absent in 58% of high grade NHL, particularly in centroblastic and Burkitt's lymphomas, suggesting that inactivation of RB may play a crucial role in the pathogenesis of high grade NHL.

Limited sampling models for reliable estimation of etoposide area under the curve.

Limited sampling models are able to estimate the area under the concentration-time curve (AUC) from plasma concentrations measured at only a few time points. The purpose of this study was to establish a model estimating etoposide AUC independently of specific chemotherapy protocols, underlying malignancies, concomitant diseases and age. Pharmacokinetic parameters were measured in 30 patients treated with polychemotherapy including etoposide (80-150 mg/m2). Etoposide analysis was performed by thin layer chromatography and consecutive quantitative sample detection by 252Cf-plasma desorption mass spectrometry. Data from the first 15 patients formed the training set. Based on the training data, five different models were generated, with the multiple regression coefficient r ranging from 0.91 to 0.96. The following model was selected as "most accurate": AUC = 343 (min)C4h(micrograms/ml) + 650(min)C8h(micrograms/ml) + 1252 (min micrograms/mol), where C4h is the plasma concentration of etoposide at 4 h after the end of infusion and C8h at 8 h. This model was validated on the test set, comprising the data of the remaining 15 patients. The mean predictive error (MPE) was -0.2% and the root mean square predictive error (RMSE) was 4.7%. When used for a large number of patients, this practicable and simple model is an instrument for use in prospective studies, to measure a correlation between drug dosage and efficacy or toxicity of the drug.

The use of dose-intensified chemotherapy in the treatment of metastatic nonseminomatous testicular germ cell tumors. German Testicular Cancer Study Group.

With the use of a cisplatin-based chemotherapy, metastatic testicular cancer has become a model for a highly curable malignant disease. Current data show that 70% to 80% of patients with this disease will achieve long-term survival following cisplatin/etoposide/bleomycin therapy. The role of high-dose chemotherapy with autologous stem cell support is being investigated in metastatic germ cell cancer in attempts to improve outcome for patients whose disease relapses after standard-dose chemotherapy and for those who present initially with advanced metastatic disease. Prognostic categories for patients receiving high-dose salvage chemotherapy have recently been developed: cisplatin-refractory disease, beta-human chorionic gonadotropin values greater than 1,000 U/L, and primary mediastinal germ cell tumors are factors characterizing patients who will derive less benefit from high-dose chemotherapy than those with chemosensitive disease at relapse. While standard-dose salvage chemotherapy achieves only a 20% long-term survival rate, high-dose salvage chemotherapy may yield a cure rate of approximately 40%. A randomized study comparing high-dose therapy with conventional-dose therapy (IT94 coordinated by the European Group for Blood and Marrow Transplantation) in patients with relapsed disease is ongoing to substantiate this observation. The use of dose-intensive therapy as first-line treatment is currently being studied by several institutions. High-dose therapy may be better tolerated when used first line compared with its use in the salvage situation, and may also achieve a rapid initial cell kill before cytostatic drug resistance develops. The German Testicular Cancer Study Group has developed a sequential high-dose combination regimen of cisplatin/etoposide/ifosfamide given with granulocyte colony-stimulating factor and peripheral blood stem cell support for four cycles every 3 weeks. This ongoing study, started in 1990, had accrued 218 patients with advanced testicular germ cell tumors as of June 1997. Of 141 evaluable patients receiving dose levels 1 through 5, 82 (58%) have achieved complete remission with no evidence of disease and 32 (23%) have achieved partial remission with marker normalization. The early death rate was 8%. Overall and event-free survival rates at 2 years are 78% and 73%, respectively, with a projected 5-year overall survival rate of 74%. Despite favorable preliminary results, this approach cannot be considered standard treatment. Currently, high-dose chemotherapy with peripheral blood stem cell transplantation should be administered to patients with testicular cancer only within controlled clinical trials to allow long-term cure rates and treatment-related late side effects to be evaluated.

Multiple myeloma of the serosa coat.

We report on a 47-year-old male patient with IgA-kappa plasmacytoma, who 12 months subsequent to diagnosis developed marked ascites. On light and electron microscopy morphologically identical plasma cells with bizarre intracytoplasmatic material were found in the bone marrow, pleural exudate, and ascites fluid. This kind of extramedullary spread is extremely rare and usually resists therapy. The mean survival rate for the 9 patients with malignant plasmacellular ascites whose cases have been documented so far was 2 months. After progression under VCMP regimens our patient has been successfully treated over a period of more than 4 years with i.v. VP16 monotherapy with additional intermittent local administration of methotrexate and systemic poly-chemotherapy according to the VAD scheme.

Epidermal growth factor receptor expression, proliferation, and colony stimulating activity production in the urinary bladder carcinoma cell line 5637.

Addition of epidermal growth factor (EGF) to cultures of the urinary bladder carcinoma cell line 5637 regulated proliferation and production of colony stimulating activity (CSA). The optimal concentration range of EGF for stimulation of cell proliferation was 5-20 ng/ml EGF and for production of CSA 2-20 ng/ml EGF. High EGF concentrations (100-200 ng/ml) showed inhibitory effects on proliferation and to a greater extent on CSA production. Also, EGF binding sites of high affinity (kd:3.25 nM) were demonstrated on the cell surface. In the optimal concentration range for stimulation (5-20 ng/ml EGF) EGF binding sites were occupied half-maximally. The loss in EGF binding after long incubation at 37 degrees C was prevented by the lysosomal inhibitory agent, chloroquine. Nonspecific binding of EGF was very low, the amount of maximally bound EGF was 1430 fmol/mg protein (130,000 bound EGF molecules/cell). A strong band of approximately 170,000 daltons could be detected by means of an anti-erbB serum which recognizes the EGF receptor protein. The protein became phosphorylated upon addition of gamma-32P ATP. The data suggest that EGF initiates its action by binding to specific high affinity receptors and plays a role in growth regulation and differentiation of the urinary bladder carcinoma cell line 5637.

Drug monitoring of etoposide (VP16-213). Correlation of pharmacokinetic parameters to clinical and biochemical data from patients receiving etoposide.

Pharmacokinetic parameters established in 15 patients receiving parenterally administered etoposide (80-120 mg . m-2) are reported. The etoposide assay by means of mass spectrometry after sample separation by thin-layer chromatography or high-pressure liquid chromatography used in this study has been described elsewhere. Peak plasma levels (9.5-63.3 micrograms . ml-1), the area under the curve (AUC) (2707-10192 micrograms . ml-1 . min-1), the mean transit time MTT (2.7-10.6 h), etoposide half-lives t1/2 alpha (0.10-0.52 h) and t1/2 beta (2.18-8.17 h), the volume of distribution at steady state (Vdss) (2.5-15.1 . l/m-2) and the systemic clearance (Cls) (10.1-35.1 ml min-1 . m-2) with the resulting mean values and standard deviations were determined. Our findings are compared with those of other authors, especially with regard to the method of detection used. This comparison indicates similar individual deviations and shorter half-lives with increasing specificity of the employed assay. Four patients studied on 3 consecutive days and, in one instance, during two different cycles of chemotherapy showed no sign of accumulation or of accelerated excretion of etoposide. There was little intrapatient variability. The pharmacokinetic parameters were correlated to clinical and laboratory findings. Statistical analysis indicated that the AUC was increased by prior cisplatin therapy and in patients with elevated levels of alkaline phosphatase. The Cls was decreased by prior cisplatin therapy, in obese patients, and by elevated alkaline phosphatase. The t1/2 beta of etoposide was increased in older patients. Linear regression analysis yielded a greater Vdss in patients with lower serum albumin levels, but this correlation has not yet been found to be statistically significant.

Drug monitoring of etoposide (VP16-213). I. A combined method of liquid chromatography and mass spectrometry.

Drug monitoring is performed by means of sample extraction, sample purification by high-performance liquid chromatography (HPLC), and sample detection by time-of-flight mass spectrometry. This mass spectrometry utilizing 252Cf fission fragment-induced ionization and desorption of nonvolatile compounds is suitable as a universal, nondestructive detector in HPLC. Liquid chromatography and mass spectrometry are combined, so that mass analysis can be operated online and offline to the fractional sampling of the effluent and the samples can still be recovered. As an alternative to HPLC separation, samples can be purified by thin-layer chromatography (TLC), resulting an offline TLC + MS combination. Preliminary pharmacokinetic data for etoposide (VP16-213) together with calibration data are presented, and are discussed with reference to the sensitivity and detection limit of the new experimental method.

Pharmacokinetics of high-dose etoposide after short-term infusion.

The pharmacokinetics of high-dose etoposide (total dose, 2100 mg/m2 divided into three doses given as 30-min infusions on 3 consecutive days) were studied in ten patients receiving high-dose combination chemotherapy followed by autologous bone marrow transplantation. In addition to etoposide, all subjects received 2 x 60 mg/kg cyclophosphamide and either 6 x 1,000 mg/m2 cytosine arabinoside (ara-C), 300 mg/m2 carmustine (BCNU), or 1,200 mg/m2 carboplatin. Plasma etoposide concentrations were determined by 252Cf plasma desorption mass spectrometry. In all, 27 measurements of kinetics in 10 patients were analyzed. According to graphic analysis, the plasma concentration versus time data for all postinfusion plasma etoposide values were fitted to a biexponential equation. The mean values for the calculated pharmacokinetic parameters were: t1/2 beta, 256 +/- 38 min; mean residence time (MRT), 346 +/- 47 min; AUC, 4,972 +/- 629 micrograms min ml-1 (normalized to a dose of 100 mg/m2); volume of distribution at steady state (Vdss), 6.6 +/- 1.2 l/m2; and clearance (CL), 20.4 +/- 2.4 ml min-1 m-2. A comparison of these values with standard-dose etoposide pharmacokinetics revealed that the distribution and elimination processes were not influenced by the dose over the range tested (70-700 mg/m2). Also, the coadministration of carboplatin did not lead to significant pharmacokinetic alterations. Although plasma etoposide concentrations at the time of bone marrow reinfusion (generally at 30 h after the last etoposide infusion) ranged between 0.57 and 2.39 micrograms/ml, all patients exhibited undelayed hematopoietic reconstitution.

Pharmacokinetics of etoposide: correlation of pharmacokinetic parameters with clinical conditions.

The pharmacokinetic parameters of etoposide were established in 35 patients receiving the drug parenterally within the framework of different polychemotherapy protocols. A total of 62 data for 24-h kinetics were analysed. After sample extraction and high-performance liquid chromatography (HPLC) or thin-layer chromatographic (TLC) separation, etoposide was measured by means of [252Cf]-plasma desorption mass spectrometry (PDMS). This highly specific detection system proved to be very practicable and reproducible. The present study comprised two parts that were absolutely comparable in terms of clinical and pharmacokinetic parameters. In part II of the study, sensitivity was improved by modifying the analytical technique. After the exclusion of patients who had previously been given cisplatin or who exhibited renal impairment and of one patient who showed extremely high levels of alkaline phosphatase, gamma-GT and SGPT, the mean values calculated for the pharmacokinetic parameters evaluated were: beta-elimination half-life (t 1/2 beta), 4.9 +/- 1.2 h; mean residence time (MRT), 6.7 +/- 1.4 h; area under the concentration-time curve (AUC), 5.43 +/- 1.74 mg min ml-1; volume of distribution at steady state (Vdss), 6.8 +/- 2.7 l/m2; and clearance (Cl), 18.8 +/- 5.3 ml min-1 m-2. The pharmacokinetic parameters were correlated with 12 different demographic or biochemical conditions. Impaired renal function, previous application of cisplatin and the age of patients were found to influence etoposide disposition to a statistically significant extent. We suggest that the dose of etoposide should be reduced in elderly patients and/or in individuals with impaired renal function, especially in those exhibiting general risk factors such as reduced liver function with regard to the polychemotherapy.

Relationship of calcitonin mRNA expression to the differentiation state of HL 60 cells.

Raised plasma levels of immunoreactive human calcitonin (ihCT) can be found in patients with myeloid leukemia and seem to indicate a poor prognosis. High levels were found in acute undifferentiated and acute myeloblastic leukemia. To test whether CT expression could be a marker of myeloid differentiation, we used the promyelocytic leukemia cell line HL 60 which also expresses ihCT as a model system for myeloid differentiation. Exponentially growing HL 60 cells as well as differentiation induced HL 60 cells expressed a single 1.0 Kb CT transcript. The induction of HL 60 cell differentiation along the granulocytic lineage by DMSO or HMBA had no effect on the level of CT transcripts. Induction of monocytic/macrophagic differentiation by TPA resulted in a transient, about 10-fold elevated expression of CT steady state mRNA after 24 h. In contrast to TPA, induction of HL 60 cell differentiation along the monocytic pathway by Vit D3 had no detectable effect on the level of the CT in RNA expression at corresponding time points. These findings suggest that the transient induction of CT steady state mRNA expression by TPA is rather a direct effect of the phorbol ester than commitment along the monocytic line of differentiation.

Biochemical characterization of a novel autocrine transferrin-like growth factor in acute myeloblastic leukemia.

Serum-free conditioned media (CM) of acute myelogenous leukemia cell lines induce growth stimulation of other leukemic cell lines. A transferrin (Tf)-like iron binding kD 50 protein was found to be the main growth factor of these CM's. Purification and further characterization of this protein was performed by chromatographic methods, SDS-PAGE, peptide mapping and amino acid analysis. In addition m-RNA of leukemic cell lines was analysed by Northern blot analysis with oligonucleotide probes specific for human serum Tf and melanotransferrin (p97) respectively. No hybridization signals for these probes have been detected. Biochemical results show significant differences between human serum Tf and kD 50 growth factor with respect to retention time on reversed phase FPLC, pattern of digestion peptides in SDS-PAGE and amino acid composition, which suggests, that this protein is an additional tumor associated member of the transferrin superfamily.

Concomitant primary low grade non-Hodgkin's lymphoma of the spleen and breast carcinoma.

A case is described of a 71 year old woman with scirrhous breast cancer and multiple hypoechogenic lesions in the spleen. The patient was treated successfully by mastectomy and splenectomy. Morphologic examination of the spleen showed a primary, centroblastic-centrocytic, low grade non-Hodgkin's lymphoma. Eight months after splenectomy the patient still remains in complete remission.

Cutaneous monoblastic leukemia as a first sign of relapse six years after autologous bone marrow transplantation for acute leukemia.

A patient with acute monoblastic leukemia (AML, M5A) was treated successfully in December 1987. In 1993 after 6 years in complete remission, she presented with an intracutaneous nodular mass on her right upper arm which was resected in toto and shown to be undifferentiated monoblastic leukemia. Two further chloroma lesions were excised in July 1994 and March 1995 respectively. Bone marrow cytology and histology always showed a continuing complete remission with no evidence of leukemia relapse. In July 1995 she presented with a disseminated skin infiltrate and a relapse with 80% monoblasts in the bone marrow. After one course of chemotherapy (Idarubicin/Ara-C), a second complete remission was achieved and her leukemic skin infiltrate disappeared completely. This case illustrates that chloromas of the skin can occur as late as 6 years after treatment for AML and also emphasizes that the occurrence of a chloroma does not necessarily mean immediate leukemia relapse. It also stresses that a second complete remission can be achieved with standard AML-induction therapy despite widespread leukemic skin infiltrates in such patients.

The role of decreased retinoblastoma protein expression in acute myelomonocytic and monoblastic leukemias.

The results of different investigators show that lack of p105 expression is relatively common in human myeloid leukemias, especially in monocytic leukemias. This suggests that loss of p105 expression could contribute to the altered growth control of these cells. So far no clear data exist which show that low p105 levels in AML blasts predict a poor therapy outcome. Therefore it is not very likely that p105 expression will become a strong prognostic factor for the different treatment strategies in AML.

Prognosis of high dose chemotherapy/autologous bone marrow transplantation candidates not receiving this treatment after failure of primary therapy of Hodgkin's disease.

In a multicenter study on the therapy of Hodgkin's disease, in 88 out of 297 patients with primary advanced stages IIIB/IV, a failure to the treatment with the alternating chemotherapy COPP/ABVD +/- radiation was recorded. The cause of failure was as follows: tumor progression under current therapy (PD) 23/88, partial response at the end of therapy (PR) 28/88, early nodal relapses 13/88, late nodal relapses 16/88, extranodal relapses 7/88, undetermined localization 1/88.36 months after manifestation of the failure to treatment, 45% of all patients were still alive. In cases of primary PD the prognosis was the worst of all. Only 1/23 of these patients received a long-term continuous complete remission (cCR) with the salvage therapy. 11 patients with only a nodal relapse received a cCR with irradiation alone. These cases could be regarded as low risk relapses. For the high risk relapse group (n = 57) an indication for high dose chemotherapy with subsequent autologous bone marrow transplantation (HDC/ABMT) would have been imperative, following the present-day definition. The probability of survival of these patients who, however, only received a conventional salvage therapy was up to 38% (95% confidence interval 22-54%). Comparing these data with the literature our results seem not to be substantially worse than those for patients who underwent HDC/ABMT. Only in a randomized comparison can the decision be made on whether HDC/ABMT would be superior to high dose conventional chemotherapy supported by hematopoietic growth factors. It is suggested that such a therapy study be performed as soon as possible.

[Recurrence of Hodgkin's disease after advanced primary stages. German Hodgkin's Study Group]. / Morbus-Hodgkin-Rezidive nach primär fortgeschrittenen Stadien. Deutsche Hodgkin-Studiengruppe.

In a multicentre study on the treatment of Hodgkin's disease, 88 out of 297 patients with primary advanced stages IIIB/IV failed to respond to alternating COPP/ABVD chemotherapy +/- radiotherapy. They may be broken down as follows: tumour progression under current therapy (PD) 23/28, partial remission at the end of treatment (PR) 28/88, early nodal recurrence 13/88, late nodal recurrence 15/88, extranodal recurrence 7/88, unclear localisation 1/88. Thirty-six months after noting failure of treatment, 45% of all patients were still alive. The prognosis was poorest in the case of primary PD. Only 1/23 of these patients experience lasting complete remission thanks to salvage treatment (cCR). Eleven patients with an exclusively nodal recurrence experienced a cCR on treatment with radiation alone, and may be considered a low-risk recurrence group. For a high-risk recurrence group (n = 57), indication for high-dose chemotherapy with subsequent autologous bone marrow transplantation (HDC/ABMT) should have been recognized on the basis of the present definition. The survival probability of these patients, who only received conventional salvage treatment, was 38% after 30 months (95% confidence limit, 22 to 54%). These data would not appear to be appreciably poorer than those reported in the literature for comparable patients receiving HDC/ABMT. Only a randomized comparison would be capable of showing whether HDC/ABMT is superior to high-dose conventional chemotherapy with haematopoietic growth factors. It is proposed that such a therapeutic trial should be initiated as soon as possible.

Toxicity-reduced, myeloablative allograft followed by lenalidomide maintenance as salvage therapy for refractory/relapsed myeloma patients.

Relapse after dose-reduced allograft in advanced myeloma patients remains high. To reduce the risk of relapse, we investigated a myeloablative toxicity-reduced allograft (aSCT) consisting of i.v. BU and CY followed by lenalidomide maintenance therapy in 33 patients with multiple myeloma (MM) who relapsed following an autograft after a median of 12 months. The cumulative incidence of non-relapse mortality at 1 year was 6% (95% confidence interval (CI): 0-14). After a median interval of 168 days following aSCT, 24 patients started with a median dose of 5 mg (r, 5-15) lenalidomide without dexamethasone. During follow-up, 13 patients discontinued lenalidomide owing to progressive disease (n=6), GvHD (n=3), thrombocytopenia (n=2), or fatigue (n=2). Major toxicities of lenalidomide were GvHD II-III (28%), viral reactivation (16%), thrombocytopenia (III-IV°,16%), neutropenia (III/IV°, 8%), peripheral neuropathy (I/II°, 16%), or other infectious complication (8%). Cumulative incidence of relapse at 3 years was 42% (95% CI: 18-66). The 3-year estimated probability of PFS and OS was 52% (95% CI: 28-76) and 79% (95% CI: 63-95), respectively. Toxicity-reduced myeloablative allograft followed by lenalidomide maintenance is feasible and effective in relapsed patients with MM, but the induction of GvHD should be considered.