& 2005 Nature Publishing Group All rights reserved /05 $

Transcription

1 (2005) 35, & 2005 Nature Publishing Group All rights reserved /05 $30 Conditioning regimens Reduced-intensity allogeneic stem cell transplantation in relapsed and refractory Hodgkin s disease: low transplant-related mortality and impact of intensity of conditioning regimen P Anderlini 1, R Saliba 1, S Acholonu 1, G-J Okoroji 1, M Donato 1, S Giralt 1, B Andersson 1, NT Ueno 1, I Khouri 1, M De Lima 1, C Hosing 1, A Cohen 1, C Ippoliti 2, J Romaguera 3, MA Rodriguez 3, B Pro 3, L Fayad 3, A Goy 3, A Younes 3 and RE Champlin 1 1 Department of Blood and Marrow Transplantation, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; 2 Department of Pharmacy, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; and 3 Department of Lymphoma, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Summary: A total of 40 patients with relapsed/refractory Hodgkin s disease (HD) underwent reduced-intensity conditioning (RIC) allogeneic stem cell transplantation (allo-sct) from an HLA-identical sibling (n ¼ 20) or a matched unrelated donor (n ¼ 20). The median age was 31 years (range 18 58). Disease status at allo-sct was refractory relapse (n ¼ 14) or sensitive relapse (n ¼ 26). The conditioning regimens were fludarabine-cyclophosphamide7antithymocyte globulin (n ¼ 14), a less intensive regimen, and fludarabinemelphalan (FM) (n ¼ 26), a more intensive one. The two groups had similar prognostic factors. The median time to neutrophil recovery (ie absolute neutrophil count X500/ll) was 12 days (range 10 24). The median time to platelet recovery (ie platelet count X20 000/ll)was 17days(range 7 132). Day 100 and cumulative (18-month) transplantrelated mortalities (TRMs) were 5 and 22%. Twenty-four patients (60%) are alive (14 in complete remission or complete remission, unconfirmed/uncertain) with a median follow-up of 13 months (4 78). In all, 16 patients expired (TRM n ¼ 8, disease progression n ¼ 8). FM patients had better overall survival (73 vs 39% at 18 months; P ¼ 3), and a trend towards better progression-free survival (37 vs 21% at 18 months; P ¼ ). RIC allo-sct is feasible in relapsed/refractory HD patients with a low TRM. The intensity of the preparative regimen affects survival. (2005) 35, doi:1038/sj.bmt Published online 4 April 2005 Keywords: Hodgkin s disease; Hodgkin s lymphoma; allogeneic stem cell transplantation; bone marrow transplantation; peripheral blood stem cell transplantation Correspondence: Dr P Anderlini, The UTMD Anderson Cancer Center, Department of Blood and Marrow Transplantation, 1515 Holcombe Boulevard Unit 423, Houston, TX 77030, USA; panderli@mdanderson.org Received 17 December 2004; accepted 27 January 2005 Published online 4 April 2005 Although many patients with Hodgkin s disease (HD) are successfully treated with chemotherapy and/or radiation therapy, the management of patients with relapsed or refractory disease remains a challenge. 1,2 There is considerable experience with high-dose chemotherapy and autologous stem cell transplantation in these cases, with a sizable fraction of patients successfully salvaged and presumably cured. 3 5 Data on allogeneic stem cell transplantation (allo-sct) in the management of relapsed or refractory HD are, on the other hand, more sketchy and far less encouraging. 6 8 Published reports of allo-sct included a majority of patients who were extensively pretreated and/ or had advanced, chemoresistant disease. Despite high rates of transplant-related mortality (TRM) (50 60%), a subset of patients (15 18%) did achieve durable remissions. The presence of a graft-versus-hd effect was postulated in some of these reports, 7,8 raising the possibility that the outcome could be significantly more favorable in patients transplanted with less intensive preparative regimens and/ or earlier in the course of their disease. Recent data seem to support the existence of a graft-versus-hd, despite the persistence of a high TRM. 9 Allo-SCT following reduced-intensity conditioning (RIC) regimens is yielding very promising preliminary results in several hematological malignancies. 10,11 Experience with RIC in HD is gradually accumulating. 12,13 We elected to explore the feasibility of such an approach in patients with relapsed and refractory HD. Preliminary results of this study were reported previously 14 and are expanded upon here. Patients and methods Patients All consecutive patients with relapsed or refractory HD who underwent allo-sct during a 6-year period (8/ /2003) at the MD Anderson Hospital (MDAH) following conditioning with one of two fludarabine-based preparative regimens (fludarabine-cyclophosphamide7antithymocyte

2 944 globulin and fludarabine-melphalan (FM)) are included in this report. Patients were deemed eligible for allotransplantation if they had chemosensitive or stable disease (SD) after salvage chemotherapy, no active or uncontrolled infection as well as adequate cardiac, pulmonary, renal and hepatic function. Patients not meeting these criteria (eg progressive disease, poor organ function) were, however, also considered on a case-by-case, compassionate basis (ie off protocol). Patients needed to have either an HLAidentical related donor or a fully serologically matched unrelated donor willing, and needed to be capable of donating filgrastim-mobilized peripheral blood progenitor cells (PBPCs: HLA-identical related donors) or bone marrow (matched unrelated donors). The treatment plan for these patients was approved by the MDAH Surveillance Committee. All patients and donors were required to sign written informed consent. Unrelated donors were recruited, consented and harvested through the National Marrow Donor Program. All individual patient data presented in this report have been de-identified. count X20 000/ml. Patients were evaluable for engraftment if they survived at least 30 days following their transplant and had a chimerism assay performed a minimum of 30 days after transplant. Chimerism was determined beginning at day post-transplant on bone marrow or peripheral blood samples by restriction fragment length polymorphism (RFLP) or, more recently, with PCR-based microsatellite polymorphism analysis using established criteria. 15 Acute and chronic graft-versus-host disease was graded according to established criteria. 16,17 Patients were considered evaluable for acute GVHD if they had achieved engraftment and for chronic GVHD if they had survived beyond day 100. TRM included all causes of death other than disease progression (PD) or relapse occurring at any time after transplant. Relapse-related mortality included all deaths in patients with active disease after transplant. However, in patients reinduced into remission (eg with salvage chemotherapy and/or donor leukocyte infusions (DLIs)) after PD or relapse, deaths were considered as transplant-related. Conditioning regimens The conditioning regimen in the first group of patients (n ¼ 14) consisted of fludarabine (25 mg/m 2 intravenously daily for 5 days), cyclophosphamide (1 g/m 2 intravenously daily for 3 days; FC: n ¼ 7) and, for matched unrelated donor transplants (as well as two sibling transplants in the initial phase of the study), antithymocyte globulin (ATG; 20 mg/kg intravenously daily for 3 days) (FC þ ATG: n ¼ 7). ATG was added to maximize engraftment. In view of the high early progression rate experienced in this initial cohort of patients, the conditioning regimen was changed in the successive 24 patients to a more intensive regimen, namely fludarabine (25 mg/m 2 intravenously daily for 5 days) and melphalan (70 mg/m 2 intravenously daily for 2 days) (FM), in order to provide more effective pretransplant cytoreduction. In two additional FM patients, the melphalan dose was 90 mg/m 2 intravenously daily for 2 days, either because treated on a compassionate basis (n ¼ 1) or because enrolled in a separate protocol (n ¼ 1). Post-transplant immunosuppression All patients received tacrolimus intravenously beginning 2 days before transplantation, dosed to maintain therapeutic serum levels (4 12 ng/ml) and switched to oral administration as soon as oral intake was feasible. In the absence of persistent or progressive disease, tacrolimus was ordinarily continued for a minimum of 6 months and subsequently tapered. Methotrexate (5 mg/m 2 intravenously) was administered on days 1, 3, 6 and (for the unrelated donor transplants) day 11 post transplant. Criteria for study evaluation Engraftment was defined as the first of three consecutive days of an absolute neutrophil count (ANC) X500/ml. Platelet recovery was considered to have occurred on the first of seven consecutive days of an unsupported platelet Response definitions A complete remission (CR) was defined as the disappearance of all clinical and radiological evidence of active disease in all known sites for a minimum of four consecutive weeks. Complete remission, unconfirmed/ uncertain (CRU) was defined as the presence of residual radiographic abnormalities of unclear clinical significance, unchanged or decreased in size during an observation period of least 4 weeks and non-gallium-avid or negative on PET scan (if initially gallium-avid or positive on a PET scan). CRUpatients were considered as CRs after one full year without PD. Partial remission (PR) was defined as an at least 50% decrease in the sum of the products of diameters of any measurable lesion persisting for at least 4 weeks. SD was defined as any response not meeting PR criteria or lack of evidence of progressive disease. Progressive disease was defined as at least a 50% increase in measurable disease or the appearance of new disease sites. Early death was defined as death prior to day 100 after allo-sct. A sensitive relapse was defined as the achievement of at least a partial response to salvage chemotherapy, whereas failure to achieve at least a partial response qualified as refractory relapse. Both PD and relapse were considered as progressive disease (PD). Patients with evidence of PD and no active GVHD ordinarily had their immunosuppression tapered and/or withdrawn and were eligible, at the discretion of the investigator, to receive DLIs, with or without preceding salvage chemotherapy and /or radiation therapy. 18 Statistical end points In keeping with the pilot study design, the main end points were engraftment (ie neutrophil and platelet recovery as well as chimerism), and early (ie before day 100) TRM. Secondary end points were acute and chronic GVHD, cumulative TRM, PD or relapse, overall survival (OS) and

3 progression-free survival (PFS). Actuarial rates of OS, PFS and acute GVHD were estimated by the method of Kaplan Meier. 19 The cumulative incidence method was used to estimate the rates of TRM, PD and chronic GVHD. Death attributed to disease was considered a competing risk for TRM, and death in remission or without chronic GVHD was considered a competing risk for the later two events, respectively. Cox proportional hazards model 20 was used to evaluate outcomes according to type of conditioning regimen, stem cell source and response status pre-transplant. The effect of acute and/or chronic GVHD on PD was evaluated considering GVHD as a timedependent variable. Only univariate analysis was possible, given the small sample size. OS and PFS were measured in months from the day of transplantation until disease relapse or progression. To minimize a potential selection bias in data reporting and make this report more clinically meaningful, all patients were included in the analysis, regardless of whether they were enrolled in the protocol. Results Patient characteristics Characteristics of the 40 patients are displayed in Table 1. The median age at transplant was 31 years (range 18 58). The median number of chemotherapy regimens received prior to allo-sct was five (2 9). In all, 30 (75%) patients had received prior radiotherapy. A total of 30 (75%) had undergone a prior autologous stem cell transplant (auto- SCT), and the median time to progression after auto-sct was 9 months (3 52). Disease status at transplant was refractory relapse (n ¼ 14) and sensitive relapse (n ¼ 26). Eight patients (20%) were transplanted on a compassionate basis. All of 945 Table 1 Patient characteristics Patient Age at SCT Time from diagnosis to allogeneic SCT (months) Number of prior chemotherapy regimens Prior radiation therapy Prior autologous SCT Time to progression post prior autologous SCT (months) Disease status at SCT 1 a Yes Yes 14 Relapse sensitive Yes Yes 3 Relapse resistant No No N/A Relapse sensitive No No N/A Relapse sensitive No No N/A Relapse sensitive Yes Yes 52 Relapse sensitive 7 a No No N/A Relapse resistant 8 a Yes Yes 3 Relapse resistant Yes Yes 4 Relapse resistant Yes Yes 6 Relapse resistant Yes Yes 3 Relapse sensitive Yes Yes 5 Relapse sensitive Yes Yes 9 Relapse sensitive Yes Yes 12 Relapse sensitive 15 a Yes No N/A Relapse resistant Yes Yes 7 Relapse resistant No No N/A Relapse sensitive No Yes 11 Relapse sensitive No No N/A Relapse sensitive Yes Yes 38 Relapse sensitive Yes Yes 7 Relapse sensitive No Yes 3 Relapse sensitive Yes Yes 8 Relapse resistant Yes Yes 7 Relapse sensitive Yes Yes 4 Relapse sensitive 26 a Yes Yes 16 Relapse sensitive 27 a Yes Yes 5 Relapse resistant 28 a Yes Yes 12 Relapse resistant 29 a Yes No N/A Relapse resistant Yes Yes 20 Relapse sensitive Yes Yes 11 Relapse sensitive Yes No N/A Relapse sensitive No Yes 9 Relapse sensitive Yes Yes 12 Relapse sensitive Yes No N/A Relapse resistant Yes Yes 9 Relapse sensitive Yes Yes 4 Relapse resistant No Yes 22 Relapse sensitive Yes Yes 10 Relapse resistant Yes Yes 6 Relapse sensitive SCT ¼ stem cell transplant; N/A ¼ not applicable. a Patients transplanted on a compassionate (ie off protocol) basis.

4 946 them were transplanted with PD, except one who had inadequate pulmonary function. Stem cell source PBPCs were employed in 21 patients (matched sibling transplants n ¼ 20; matched unrelated donor transplants n ¼ 1) and bone marrow in 19 patients (all unrelated donor transplants; Table 2). One matched related donor chose to donate PBPCs over bone marrow. The median CD34 þ cell dose infused was /kg (2.1 29). Engraftment and GVHD Hematological recovery data are presented in Table 2. The median time to neutrophil recovery (ie ANC X500/ml) was 12 days (range 10 24). The median time to platelet recovery was 17 days (range 7 132). Chimerism data at day indicate 100% donor-derived engraftment in 9/13 (69%) evaluable FC patients and 26/26 (100%) FM patients. One FC patient (patient 10) did not have chimerism studies performed at day , although peripheral blood studies done at day 20 revealed mixed chimerism (80% donor). Of the four FC patients with mixed chimerism, two had received sibling and two unrelated donor allografts. Two of these patients converted from mixed to full donor chimerism after DLIs. All of the long-term responders have ongoing 100% donor-derived engraftment, which has lasted a minimum of 6 months. The actuarial incidences of grade II IV and grade III IV acute GVHD were 38% (95% CI 25 55) and 10% (95% CI 4 25), respectively. The cumulative incidence of chronic Table 2 Engraftment and chimerism data Patient Age at SCT Conditioning regimen Donor type Stem cell source CD34+ cell dose ( 10 6 /kg) ANC 500 (days) PLT 20K (days) Chimerism day (% donor) 1 a 42 FC SIB PBPC FC SIB PBPC FC SIB PBPC FC SIB PBPC FC SIB PBPC FC SIB PBPC a 58 FC SIB PBPC a 29 FC+ATG SIB PBPC FC+ATG SIB PBPC FC+ATG MUD BM Unknown FC+ATG MUD BM FC+ATG MUD BM FC+ATG MUD BM FC+ATG MUD BM a 40 FM SIB PBPC FM SIB PBPC FM SIB PBPC FM SIB PBPC FM SIB PBPC FM SIB PBPC FM SIB PBPC FM SIB PBPC FM SIB PBPC FM SIB PBPC FM SIB PBPC a 30 FM MUD BM a 48 FM MUD BM a 32 FM MUD BM a 31 FM MUD BM FM MUD BM FM MUD BM FM MUD BM FM MUD BM N/A FM MUD BM FM MUD BM FM MUD BM FM MUD BM FM MUD BM FM MUD BM No recovery FM MUD PBPC PBPC ¼ peripheral blood progenitor cells; BM ¼ bone marrow; SIB ¼ HLA-identical sibling donor; MUD ¼ matched unrelated donor; FC ¼ fludarabinecyclophosphamide; ATG ¼ antithymocyte globulin; FM ¼ fludarabine-melphalan; N/A ¼ not applicable; ANC 500 ¼ first of three consecutive days of an absolute neutrophil count X500/ml; PLT 20K ¼ the first of seven consecutive days of an unsupported platelet count X20 000/ml. In eight FC patients the platelet count never dropped below /ml after the conditioning regimen, hence the figure 0. a Patients transplanted on a compassionate (ie off protocol) basis.

5 GVHD at 18 months was 69% (SE 13%). There were no significant differences between the FC and FM groups in the incidence of grade II IV acute GVHD (P-value ¼ ), grade III IV acute GVHD (P-value ¼ 5) or chronic GVHD (P-value ¼ ). Patient outcome: OS, PFS and PD Median follow-up in the whole group was 12 months (range 2 78). OS, PFS (actuarial estimates) and PD (cumulative incidence) at 18 months were 61% (95% CI 42 76), 32% (95% CI 17 47) and 55% (s.e. 9%), respectively (Figures 1 and 2). At the time of the latest follow-up (March 2004), 24 patients (60%) are alive, with a median follow-up of 13 months (4 78). Of these, 14 patients are in CR or CRU. Three of them were reinduced into remission after initial PD (patients 16, 20 and 23). All three are alive and their median CR/CRUduration is 9 months (1 43). The median follow-up is 23 months (4 41) for the 11 patients alive and always in remission after allo-sct. Cumulative proportion surviving Figure 1 OS PFS Kaplan Meier estimates for OS and PFS for the whole group. Patient outcome: TRM and causes of death Patient outcome data are presented in Table 3. In all, 16 patients expired (40%). Of them, two died before day 100 (viral and bacterial pneumonia, respectively), accounting for a day 100 TRM of 5%. Also, 14 patients expired after day 100 (PD n ¼ 8, GVHD n ¼ 3, pulmonary nocardiosis n ¼ 1, intracranial hemorrhage n ¼ 1, diffuse alveolar hemorrhage n ¼ 1). Cumulative incidence of TRM at 18 months was 22% (s.e. 7%) (Figure 2). Patient outcome according to conditioning regimen, stem cell source and response status pretransplant FC and FM patients were compared according to several known prognostic variables, as well as response status (ie chemosensitivity) pretransplant and stem cell source, and no statistically significant difference was observed between the two groups (Table 4). When OS and PFS were stratified according to the preparative regimen received, FM patients enjoyed an OS advantage (73%; 95% CI vs 39%; CI at 18 months, P ¼ 3), as well as a trend towards better PFS (37%; 95% CI vs 21%; 95% CI 5 45 at 18 months, P ¼ ; Figures 3 and 4). There was a nonsignificant trend towards a lower TRM in the FM group (18%; s.e. 9 vs 30%; s.e. 12% at 18 months, P ¼ ; Figure 5). Progression rates were similar in the FM and FC patients (53%; s.e. 11 vs 57%; s.e. 14%, respectively, at 18 months, P ¼ ). However, PD occurred later in FM patients (range 2 34 months) than in FC patients (range 13 months) (Figure 6). In addition, with comparable follow-up time after progression, the FM group experienced a lower death rate after progression, although this did not reach statistical significance (3/12 progressors in the FM group vs 7/8 progressors in FC group; hazard ratio and P ¼ ). Cell source, chemosensitivity and the presence of B symptoms at transplant were evaluated as prognostic factors for OS, PD and PFS by univariate analysis. The effect of donor type was not evaluated separately as it was highly correlated with the stem cell source. None of these factors had a significant impact on outcome (data not shown). 947 Cumulative incidence PD TRM Figure 2 Cumulative incidence of PD and TRM for the whole group. Impact of acute and/or chronic GVHD on disease relapse/progression The impact of GVHD (the earliest of acute or chronic) on PD was evaluated considering GVHD as time-dependent variable. There was a trend for lower relapse rate after the occurrence of GVHD; however, this was not statistically significant (hazard ratio ; P ¼ ). Similar results were obtained when acute GVHD grade 1 was not considered. All patients who had grade 2 4 acute GVHD survived up to at least day 100 post transplant. Donor leukocyte infusions (DLIs) Eight patients with PD received one (or more) DLIs, with (n ¼ 5) or without (n ¼ 3) preceding salvage chemotherapy and/or radiation therapy (Table 3). The response rate (PR plus CR/CRU) was 3/8 (38%). Four patients (50%) survive, although only one (Patient 20) has an ongoing

6 948 Table 3 Patient outcome Patient Donor type Regimen Disease relapse/progression post SCT Salvage therapy Last follow-up after allo-sct (months) Alive Disease status at last follow-up 1 a SIB FC Yes DLI 9 No CRU 2 SIB FC Yes Tacrolimus withdrawal 12 No PD 3 SIB FC Yes Chemo 4 No PD 4 SIB FC No N/A 36 Yes CR 5 SIB FC Yes Chemo/DLI/XRT 23 No PD 6 SIB FC No N/A 24 Yes CR 7 a SIB FC Yes Chemo 4 Yes PD 8 a SIB FC+ATG Yes DLI 9 No PD 9 SIB FC+ATG Yes DLI 53 No PD 10 MUD FC+ATG No N/A 2 No ED 11 MUD FC+ATG No N/A 8 No CRU 12 MUD FC+ATG No N/A 6 No CRU 13 MUD FC+ATG Yes Chemo 5 No PD 14 MUD FC+ATG No N/A 41 Yes CR 15 a SIB FM Yes XRT 13 Yes PD 16 SIB FM Yes Surgery 78 Yes CR 17 SIB FM Yes Chemo/DLI/XRT 12 No CRU 18 SIB FM No N/A 24 Yes CR 19 SIB FM No N/A 24 Yes CR 20 SIB FM Yes Chemo/DLI 13 Yes CRU 21 SIB FM Yes Chemo/DLI 12 Yes PD 22 SIB FM No N/A 11 Yes CRU 23 SIB FM Yes Tacrolimus withdrawal 8 Yes CRU 24 SIB FM No N/A 6 Yes CRU 25 SIB FM Yes Tacrolimus withdrawal 5 Yes PR 26 a MUD FM No N/A 15 No CR 27 a MUD FM No N/A 21 Yes CR 28 a MUD FM Yes XRT 5 No PD 29 a MUD FM No N/A 11 Yes PD 30 MUD FM No N/A 32 Yes CR 31 MUD FM Yes Chemo/DLI 25 Yes PD 32 MUD FM Yes Chemo 21 No PD 33 MUD FM No N/A 2 No ED 34 MUD FM Yes XRT 22 Yes PR 35 MUD FM Yes Chemo 20 Yes SD 36 MUD FM No N/A 13 Yes CR 37 MUD FM Yes Tacrolimus withdrawal 5 No SD 38 MUD FM Yes Tacrolimus withdrawal 12 Yes PD 39 MUD FM No N/A 4 Yes CRU 40 MUD FM Yes Chemo/DLI 5 Yes SD SCT ¼ stem cell transplant; FC ¼ fludarabine-cyclophosphamide; ATG ¼ antithymocyte globulin; FM ¼ fludarabine-melphalan; SIB: HLA-identical sibling donor; MUD: matched unrelated donor. Chemo ¼ chemotherapy; DLI ¼ donor leukocyte infusion; XRT ¼ radiation therapy; CR ¼ complete remission; CRU ¼ complete remission, unconfirmed/uncertain; SD ¼ stable disease; PD ¼ progressive disease; ED ¼ early death; N/A ¼ not applicable. See text for disease status (ie response) definitions (CR, CRU, SD, PD). Patient 14 received a DLI for mixed chimerism post BMT. a Patients transplanted on a compassionate (ie off protocol) basis. Table 4 patients Comparison of prognostic factors between FC and FM Patient characteristics FC FM P-value N ¼ 14 N ¼ 26 Sex (M/F) 12/2 18/8 Age (median, range) 31 (26 59) 31 (18 57) No. of prior chemotherapy 5.5 (3 9) 5 (2 8) regimens (median, range) Prior autologous SCT (N, %) 10 (71%) 20 (77%) Matched sibling donor (N, %) 9 (64%) 11 (42%) Chemosensitivity at SCT (N, %) 9 (64%) 17 (65%) Stage III IV at SCT (N, %) 7 (50%) 11 (42%) SCT ¼ stem cell transplant; FC ¼ fludarabine-cyclophosphamide; FM ¼ fludarabine-melphalan. Staging as per Urba and Longo. 1 Cumulative proportion surviving P value 3 FM FC Figure 3 Kaplan Meier estimates for OS in FC and FM groups.

7 Cumulative proportion surviving progression free Figure 4 Cumulative incidence TRM P value FM FC Kaplan Meier estimates for PFS in FC and FM groups. P value Figure 5 Cumulative incidence PO response (CRU) 9 months after his DLI. As all but one patient developed GVHD after the DLI, any correlation between GVHD and response could not be evaluated statistically. There was no obvious correlation between FC FM Cumulative incidence of TRM in FC and FM groups Figure 6 Cumulative incidence of PD in FC and FM groups. FC FM P value CD3 þ cell dose infused and disease response (data not shown). Our DLI experience in these patients has been reported in greater detail elsewhere. 18 Discussion The main purpose of this study was to explore in a singlecenter contest the feasibility of reduced-intensity, fludarabine-based conditioning for allografting from matched related and unrelated donors in advanced HD. The results proved that such approach is indeed feasible, with reliable engraftment rates (particularly in the FM patients) and a very low early (ie day 100) TRM (5% in this patient group as a whole). Cumulative TRM was also low at 22%, although this result will require a longer follow-up for proper evaluation. The feasibility of this approach extended to matched unrelated donor transplants. The role of allo-stem cell transplantation in the management of relapsed/refractory HD remains controversial. Literature and registry data on mainly related donor transplants (Table 5) indicate virtually prohibitive early as well cumulative TRM and relapse rates (6 8). Despite that, a subset of patients (15 18%) in these reports achieved long-term PFS. A more recent report, 9 while suggesting a somewhat more favorable long-term patient outcome (event-free survival 26% at 10 years), still listed a day 100 and overall TRM of 32 and 43%, respectively. Allo-SCT following a less intensive or even nonmyeloablative conditioning regimens is yielding very promising early results in several hematological malignancies, presumably due to its ability to harness a graft-versus-leukemia or lymphoma effect within the contest of a reduced transplantrelated toxicity. 10,11 Heavily pretreated patients with relapsed and refractory HD may be ideal candidates for this approach, as their ability to tolerate conventional myeloablative preparative regimens is limited. In addition, thanks to its beneficial impact on early TRM, such approach may facilitate the detection of a graft-versus- HD effect. Peggs et al presented an update of their experience on 41 patients (26 allografted from sibling donors, 15 from matched unrelated donors) conditioned with a FMalemtuzumab regimen. In all, 19 patients (46%) required DLIs, primarily for persistent or progressive disease. The cumulative TRM was 20% (11% DLI-related), with a projected 4-year PFS of 34%, not dissimilar from the one presented in this study. 12 Robinson et al 13 recently reported data on 90 HD patients who received an allograft (mainly from HLA-identical siblings) following a variety of preparative regimens (but primarily fludarabine-based conditioning), with an early (day 100) TRM of 11% with a 1-year PFS and OS of 45 and 60%, respectively. Burroughs et al have recently reported data with singledose total body irradiation (TBI)7fludarabine conditioning in 27 patients (18 allografted from sibling donors, nine from matched unrelated donors). Day 100 and 1-year nonrelapse mortality were 7 and 35%, respectively, but the 1-year PFS was only 18%. 21 A double transplantation approach (auto/allo) was also found to provide promising results, albeit in a smaller cohort of patients

8 950 Table 5 Published experience with allogeneic stem cell transplantation in Hodgkin s disease (HD) Reference Year of transplants No. of cases Early TRM Cumulative TRM Relapse DFS/PFS Graft-vs-HD effect? IBMTR % 61% (3-year) 65% (3-year) 15% (3-year) No EBMT % 48% (4-year) 61% (4-year) 15% (4-year) Possibly FHCRC N/A 49% (5-year) 65% (5-year) 18% (5-year) Possibly Johns Hopkins % (chemosensitive) 32% (chemosensitive) 53% (10-year) 26% (10-year) Possibly 39% (chemoresistant) 53% (chemoresistant) IBMTR¼ International Bone Marrow Transplant Registry; EBMT ¼ European Group for Blood and Marrow Transplantation: FHCRC ¼ Fred Hutchinson Cancer Research Center; TRM ¼ transplant-related mortality; DFS/PFS ¼ disease-free survival/progression-free survival; N/A ¼ not available. The FHCRC series includes nine patients transplanted from mismatched related and matched unrelated donors. The feasibility of reduced-intensity, fludarabine-based conditioning in matched unrelated donor allografts is particularly noteworthy, as inclusion of unrelated donor allografts would allow a far larger number of patients to potentially benefit from this strategy. Our results also suggested that, within the broad definition of reducedintensity or nonmyeloablative, the intensity of the preparative regimen does affect patient outcome. Despite the limited sample size of this study, patients receiving FM had significantly better OS than patients receiving FC, a clearly less intensive regimen. PFS was superior as well, although the difference did not reach statistical significance, possibly because of the small sample size. We interpret these data as suggesting that, while the intensity may indeed be reduced, the preparative regimen still needs to have significant cytoreductive activity to impact patient outcome. This interpretation is also supported by a similar patient outcome reported by other investigators, 12,13 with preparative regimens comparable to the one we employed. It is also consistent with the very poor 1-year PFS reported with low-dose TBI7fludarabine (ie a regimen with virtually no activity against HD). 21 Alternative explanations for these results do exist and include a possible deleterious effect of the ATG included in the FC regimen, unrecognized prognostic differences between the two groups, and improved supportive care modalities in the FM patients. However, all of these remain speculative. While the cumulative relapse rate per se was seemingly unaffected by the intensity of the preparative regimen, FM patients relapsed later and survived longer after PD. With regard to the issue of a graft-versus-hd effect, in view of the very high TRM (as well as relapse rates) reported in the literature (Table 5), it is hardly surprising that definitive proof of the very existence of a graft-versus- HD effect has largely eluded investigators, even if some have postulated it. Milpied et al 7 showed such an effect with Xgrade 2 acute GVHD. Anderson et al 8 reported a lower relapse rate in HD patients who received an allo-sct when compared to recipients of an autologous SCT, and Akpek et al 9 showed a trend towards lower relapse rate in HD patients transplanted in chemosensitive relapse and patients who developed GVHD. These findings, as well as several case reports of disease responses following DLIs, 12,23 support the existence of a graft-versus-hd effect. Our data do not allow a definitive conclusion on this issue. We evaluated the possible impact of GVHD (acute and/or chronic) on the development of disease relapse or progression. There was a trend for lower relapse rate after the occurrence of GVHD, although this did not reach statistical significance, possibly because of the sample size of the study. DLI data in these patients, although limited, are also suggestive of a graft-versus-hd effect, as discussed in greater detail elsewhere, 18 although definitive proof is still lacking. The trend towards a higher TRM in the FC group (a less intensive regimen), while not reaching statistical significance, defies an immediate explanation. As discussed previously, this finding could be ascribed to the use of ATG in these patients and/or unrecognized prognostic differences increasing the risk of TRM in the FC group. While a formal comparison is not possible, it may be noteworthy to point out that HD patients conditioned with low-dose TBI7fludarabine (another minimally intensive regimen) also experienced a surprisingly high 1-year nonrelapse (ie transplant-related) mortality of 35%. 21 In conclusion, allo-stem cell transplantation with reduced-intensity, fludarabine-based conditioning is feasible in high-risk, heavily pretreated patients with relapsed and refractory HD, with very encouraging early TRM and engraftment rates. This approach seems well tolerated and deserves to be explored further in a larger number of patients, and ideally in patients with less advanced disease. While the short follow-up, small sample size, sequential nature of our study and patient heterogeneity all suggest some caution in data interpretation, we found that the intensity of the preparative regimen affects patient outcome. PD remains a major cause of treatment failure, and this will need to be addressed in the planning of future studies. Longer follow-up data will allow a more meaningful evaluation of the ultimate impact of this approach. Acknowledgements We are indebted to Muriel Giese for her secretarial assistance. References 1 Urba WJ, Longo DL. Hodgkin s disease. N Engl J Med 1992; 326: Aisenberg AC. Problems in Hodgkin s disease management. Blood 1999; 93: Linch DC, Goldstone AH. High-dose therapy for Hodgkin s disease. Br J Haematol 1999; 107: Forman SJ. Role of High-Dose Therapy and Stem Cell Transplantation in the Management of Hodgkin s Disease.

9 American Society of Clinical Oncology (ASCO) Educational Book: Alexandria, VA, 1997, pp Johnston LJ, Horning SJ. Autologous hematopoietic cell transplantation in Hodgkin s disease. Biol Blood Marrow Transplant 2000; 6: Gajewski JL, Phillips GL, Sobocinski KA et al. Bone marrow transplants from HLA-identical siblings in advanced Hodgkin s disease. J Clin Oncol 1996; 14: Milpied N, Fielding AK, Pearce R, et al, for the European Group for Blood and Marrow Transplantation. Allogeneic bone marrow transplant is not better than autologous transplant for patients with relapsed Hodgkin s disease. J Clin Oncol 1996; 14: Anderson JE, Litzow MR, Appelbaum FR et al. Allogeneic, syngeneic and autologous marrow transplantation for Hodgkin s disease: the 21-year Seattle experience. J Clin Oncol 1993; 11: Akpek G, Ambinder RF, Piantadosi S et al. Long-term results of blood and marrow transplantation for Hodgkin s lymphoma. J Clin Oncol 2001; 19: Martino R, Sierra J. Allogeneic hematopoietic stem cell transplantation after immunosuppressive but non-myeloablative conditioning: miniallografts are no small matter. Haematologica 1998; 83: Champlin R, Khouri I, Kornblau S et al. Allogeneic hematopoietic transplantation as adoptive immunotherapy: induction of graft-vs-malignancy as primary therapy. Hematol Oncol Clin North Am 1999; 13: Peggs KP, Thomson K, Chopra R et al. Long term results of reduced intensity transplantation in multiply relapsed and refractory Hodgkin s lymphoma: evidence of a therapeutically relevant graft-versus-lymphoma effect (abstract). Blood 2003; 102: 198a. 13 Robinson SP, Sureda A, Mackinnon S et al. Reduced intensity allogeneic stem cell transplantation for Hodgkin s disease: an analysis from the European Group for Blood and Marrow Transplantation (abstract). Blood 2002; 100: 144a. 14 Anderlini P, Giralt S, Andersson B et al. Allogeneic stem cell transplantation with fludarabine-based, less intensive conditioning regimens as adoptive immunotherapy in advanced Hodgkin s disease. Bone Marrow Transplant 2000; 26: Yam P, Petz L, Knowlton R et al. Use of DNA restriction fragment length polymorphism to document marrow engraftment and mixed hematopoietic chimerism following bone marrow transplantation. Transplantation 1987; 43: Glucksberg H, Storb R, Fefer A et al. Clinical manifestations of graft versus host disease in human recipients of marrow from HLA matched sibling donors. Transplantation 1974; 18: Schulman HM, Sullivan KM, Weiden PL et al. Chronic graftversus-host disease syndrome in man. A long term clinical pathological study of 20 Seattle patients. Am J Med 1980; 69: Anderlini P, Acholonu S, Julia-Okoroji GJ et al. Donor leukocyte infusions in relapsed Hodgkin s lymphoma following allogeneic stem cell transplantation: CD3+ cell dose, GVHD and disease response. Bone Marrow Transplant 2004; 34: Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: Cox DR. Regression models and life tables (with discussion). J Roy Stat Soc 1972; B34: Burroughs LM, Maris MB, Sandmaier BM et al. HLAmatched related or unrelated donor non-myeloablative conditioning and hematopoietic cell transplant for patients with advanced Hodgkin s disease (abstract). Biol Blood Marrow Transplant 2004; 10 (Suppl. 1): Carella AM, Cavaliere M, Lerma E et al. Autografting followed by non-myeloablative immunosuppressive chemotherapy and allogeneic peripheral-blood hematopoietic stem-cell transplantation as treatment of resistant Hodgkin s disease and non-hodgkin s lymphoma. J Clin Oncol 2000; 18: Porter DL, Luger SM, Duffy KM et al. Allogeneic cell therapy for patients who relapse after autologous stem cell transplantation. Biol Blood Marrow Transplant 2001; 7: