Long-term follow-up of high-dose treatment with autologous ...

Leukemia (2007) 21, 2324–2331 & 2007 Nature Publishing Group All rights reserved 0887-6924/07 $30.00 www.nature.com/leu

ORIGINAL ARTICLE Long-term follow-up of high-dose treatment with autologous haematopoietic progenitor cell support in 693 patients with follicular lymphoma: an EBMT registry study S Montoto1, C Canals2, AZS Rohatiner1, G Taghipour3, A Sureda2, N Schmitz4, C Gisselbrecht5, L Fouillard6, N Milpied7, C Haioun8, S Slavin9, E Conde10, C Fruchart11, A Ferrant12, V Leblond13, H Tilly14, TA Lister1 and AH Goldstone15, for the EBMT Lymphoma Working Party 1

Cancer Research UK Medical Oncology Unit, Institute of Cancer, Centre for Medical Oncology, Barts and the London, Queen Mary’s School of Medicine and Dentistry, London, UK; 2Clinical Hematology Division, Hospital Santa Creu i Sant Pau, Barcelona, Spain; 3EBMT Central Registry Office, London, UK; 4Department of Hematology, AK St Georg, Hamburg, Germany; 5Department of Hemato-oncology, Hoˆpital St Louis, Paris, France; 6Department of Hematology, Hoˆpital Saint Antoine, Paris, France; 7 Department of Hematology, Hotel Dieu, Nantes, France; 8Department of Hematology, Hoˆpital Henri Mondor, Creteil, France; 9 Department of Bone Marrow Transplantation, Hadassah University Hospital, Jerusalem, Israel; 10Department of Hematology, Hospital de Valdecilla, Universidad de Cantabria, Santander, Spain; 11Department of Hematology, Centre François Baclesse, Caen, France; 12Department of Hematology, UCL, Cliniques Universitaires Saint-Luc, Brussels, Belgium; 13Department of Hematology, Groupe Hospitalier Pitie´ Salpeˆtrie`re, Paris, France; 14Department of Hematology, Centre Henri Becquerel, Rouen, France and 15Department of Hematology, University College London Hospital, London, UK

To evaluate the outcome of a large series of patients who received high-dose treatment (HDT) for follicular lymphoma (FL), 693 patients undergoing HDT (total-body irradiation (TBI)containing regimen: 58%; autologous bone marrow (BM)/ peripheral blood progenitor cells (PBPCs): 378/285 patients) were included in the study. A total of 375 patients (54%) developed recurrent lymphoma, 10-year progression-free survival (PFS) being 31%. On multivariate analysis, younger age (P ¼ 0.003) and HDT in first complete remission (CR1) (Po0.001) correlated with longer PFS. With a median follow-up of 10.3 years, 330 patients died. Ten-year overall survival (OS) from HDT was 52%. Shorter OS was associated on multivariate analysis with older age (Po0.001), chemoresistant disease (Po0.001), BM þ PBPC as source of stem cells (P ¼ 0.007) and TBI-containing regimens (P ¼ 0.004). Thirty-nine patients developed secondary myelodysplastic syndrome/acute myeloid leukaemia (MDS/AML), in 34 cases having received TBI as the conditioning regimen. The 5-year non-relapse mortality (NRM) was 9%. On multivariate analysis, older age (Po0.001), refractory disease (Po0.001) and TBI (P ¼ 0.04) were associated with a higher NRM. This long follow-up study shows a plateau in the PFS curve, suggesting that a selected group of patients might be cured with HDT. On the downside, TBI-containing regimens are associated with a negative impact on survival. Leukemia (2007) 21, 2324–2331; doi:10.1038/sj.leu.2404850; published online 19 July 2007 Keywords: high-dose therapy; autologous stem cell; follicular lymphoma

Introduction

comparison with other subtypes, this disease remains incurable, with a continuous relapse pattern, the rate and duration of response decreasing in subsequent recurrences.2 Numerous attempts to improve the outcome of patients with FL have been made, including consolidation of the response with high-dose treatment (HDT) and autologous stem cell (ASC) support.3–9 The use of HDT with ASC support has increased considerably over the last 25 years.10 In recent years, three randomized trials in newly diagnosed11,12 or relapsed patients13 have shown significantly prolonged progression-free survival (PFS) in patients receiving HDT in comparison with conventional chemotherapy. A survival advantage has also been demonstrated in one of the studies.13 However, the potential long-term toxicity of the procedure, especially the development of secondary myelodysplasia,14–16 remains a matter of concern. In addition, there is the problem of recurrence. Allogeneic transplant, the only potentially curative option for patients with FL, is associated with a prohibitive high toxicity in a relatively old and heavily pretreated population. The advent of reducedintensity conditioning regimens (RIC) has increased the number of patients eligible for this procedure, who might also be candidates for HDT with ASC rescue. As a result and despite the widespread use of HDT, the indications for this treatment in FL and its optimal timing are still highly controversial. The aim of this retrospective study was therefore to report the data from a large series of patients with FL who received HDT with ASC rescue before 1995, to characterize the outcome of this population with long follow-up.

Follicular lymphoma (FL) is the second commonest type of nonHodgkin’s lymphoma in Western countries.1 Although the median survival for patients with FL is relatively long in

Patients and methods

Correspondence: Dr S Montoto, Cancer Research UK Medical Oncology Unit, Institute of Cancer, Centre for Medical Oncology, Barts and the London, Queen Mary’s School of Medicine and Dentistry, St Bartholomew’s Hospital, 45 Little Britain, London EC1A 7BE, UK. E-mail: [email protected] Dedicated to the memory of Felix Reyes. Received 16 April 2007; revised 4 June 2007; accepted 7 June 2007; published online 19 July 2007

From 1979 to 1995, 878 patients with FL who underwent HDT with ASC rescue were reported to the EBMT registry. Patients with histological transformation at the time of HDT, those undergoing a second transplant and those with a follow-up of less than 5 years were excluded. Thus, 693 patients were included in the analysis. The clinical characteristics of the patients at the time of diagnosis and HDT are summarized in Table 1. Central review of the pathology was not carried out.

Patient population

Long follow-up of HDT in FL S Montoto et al

2325 Table 1

Main characteristics of the patients at the time of diagnosis and at HDT, and characteristics of the procedures

At diagnosis Male Age, median (range) B symptoms Bulky disease (X10 cm) BM involvement Stage III–IV At the time of HDT Age, median (range) Age X55 years Treatment lines 1 2 X3

Whole series (n ¼ 693)

HDT in CR1 (n ¼ 131)

Response at HDT other than CR1 (n ¼ 562)

No./no. assessable

No./no. assessable

No./no. assessable

372 42 91/434 73/409 261/485 507/589

(54%) (17–63) (21%) (18%) (54%) (86%)

45 (17–68) 94/693 (14%) 133/442 (30%) 275/442 (62%) 34/442 (8%)

Disease status CR1 CR2 CR3 PR1 VGPR1 Chemosensitive recurrence Chemoresistant recurrence Untested recurrence Primary refractory

131 140 43 124 46 151 20 15 23

Stem cell source BM PB BM+PB

73 44 24/105 14/108 62/117 111/124

(56%) (20–59) (23%) (13%) (53%) (90%)

45 (22–59) 17/131 (13%) 73/118 (62%) 39/118 (33%) 6/118 (5%)

299 42 67/329 59/301 199/368 396/465

(53%) (17–63) (20%) (20%) (54%) (85%)

46 (17–68) 77/562 (14%) 60/324 (18%) 236/324 (73%) 28/324 (9%)

131 F F F F F F F F

F 140 (25%) 43 (8%) 124 (22%) 46 (8%) 151 (27%) 20 (4%) 15 (2%) 23 (4%)

378 (55%) 285 (41%) 30 (4%)

67 (51%) 56 (43%) 8 (6%)

311 (55%) 229 (41%) 22 (4%)

Ex vivo manipulation

283/633 (45%)

53/124 (43%)

230/509 (45%)

Conditioning regimen TBI-containing Chemotherapy-based BEAM BEAC CBV Other

401/690 (58%) 289/690 (42%) 163 39 11 76

71/130 (55%) 59/130 (45%) 41 5 4 9

330/560 (59%) 230/560 (41%) 122 34 7 67

(19%) (20%) (6%) (18%) (7%) (22%) (3%) (2%) (3%)

Abbreviations: BEAC, carmustine, etoposide, cytarabine, cyclophosphamide; BEAM, carmustine, etoposide, cytarabine, melphalan; BM, bone marrow; CBV, cyclophosphamide, carmustine, vincristine; CR, complete response; HDT, high-dose treatment; PB, peripheral blood; PR, partial response; TBI, total-body irradiation; VGPR, very good partial response.

The median time from diagnosis to HDT was 2.2 years (range, 1 month to 22 years), with 24% of the patients receiving HDT during the first year after diagnosis. One hundred and ninetyeight patients (29%) had never achieved a complete remission (CR) before HDT. One-fifth of the patients received HDT after achievement of first CR (CR1), 38% of them having needed more than one chemotherapy line to achieve CR1. The clinical characteristics of these patients at diagnosis and at HDT are listed in Table 1. Five hundred and sixty-three of the patients (81%) were treated after 1990. The follow-up for patients reported to the EBMT registry is updated annually. The followup was updated in the last 3 years for 238 of 363 (66%) living patients.

Treatment The source of progenitor cells and the conditioning regimens are detailed in Table 1. Ex vivo manipulation of the bone marrow (BM) or peripheral blood progenitor cells (PBPCs) was carried

out in 283 of 633 patients with known data (45%). The commonest method of purging (75%) was negative selection with a combination of monoclonal antibodies and complement. The proportion of patients who received a total-body irradiation (TBI)-containing regimen was 51% in patients treated before 1990 and 60% in patients treated thereafter. Supportive care was given according to local protocols at each centre.

Definitions

According to the EBMT guidelines,17 complete response (CR) was defined as the disappearance of tumour masses and diseaserelated symptoms, as well as normalization of initially abnormal tests and/or biopsies. Partial response (PR) was considered when measurable lesions decreased by at least 50%. A very good partial response (VGPR) was defined as a reduction in tumour burden of at least 90%. For the purpose of this analysis, patients achieving a PR or a VGPR were grouped together, since these definitions predate the 1999 consensus statement.18 Recurrence Leukemia


2326 was defined as the occurrence of new sites of disease after a CR lasting for 3 months or longer, whereas it was considered progression when the CR had lasted less than 3 months or CR had not been achieved. Recurrence or progression was considered to be ‘chemosensitive’ if at least PR was achieved after salvage treatment, otherwise ‘chemoresistant’. The recurrence/progression was deemed ‘untested’ if the patient had no further treatment after recurrence.17 Overall survival (OS) was defined as the time from HDT to death from any cause, and surviving patients were censored at last follow-up. PFS was defined as time from HDT to recurrence, progressive disease or death from any cause. Non-relapse mortality (NRM) was defined as death from any cause without progression.

Statistical analysis The probabilities of OS and PFS were estimated from the time of HDT using Kaplan–Meier curves, and compared by the log-rank test. Estimates of NRM and recurrence or progression were calculated using cumulative incidence (CI) rates taking into account the competing risk structure, and compared by univariate Cox regression models. The risk of secondary malignancies after HDT was investigated by CI considering death from any cause as a competing event. The univariate analysis was carried out for the main characteristics at diagnosis and at the time of HDT. For the multivariate analysis, potential prognostic factors for NRM, recurrence, PFS and OS were first entered into the model; covariates found not to be significant at the 0.10 level were removed from the Cox proportional hazards model in a stepwise backward way. In each model, the assumption of proportional hazards was tested for each variable using time-dependent covariates.19,20 As the effect of TBI on outcome changed over time (nonproportional hazard), models were constructed to break the post-HDT time course into two periods, using the most appropriate break point. CIs were calculated using the NCSS 97 software. All other computations were carried out in the SPSS 13.0 statistical package.

Results

Risk of recurrence and progression-free survival The median follow-up for living patients was 10.3 years (range: 5–19.5). Three hundred and seventy-five patients (54%) developed recurrent disease at a median of 1.5 years after HDT (range: 1 month–13.5 years). On univariate analysis, patients who received a TBI-containing regimen had a lower risk of recurrence than the remainder (43% at 5 years versus 53%). The risk of recurrence also related to disease status and interval from diagnosis to HDT. On multivariate analysis (Table 2), disease status (Po0.001) and conditioning regimen (P ¼ 0.05) retained predictive value for the risk of recurrence. PFS for the overall series at 5, 10 and 15 years was 44, 31, and 27%, respectively (Figure 1a). The variables that predicted for a shorter PFS on univariate analysis were age at HDTX45 years, disease status at HDT other than CR1, time from diagnosis to HDTX1 year and more than two lines of treatment before HDT. Age (P ¼ 0.003) and disease status (Po0.001) remained significant on multivariate analysis (Table 2). PFS at 5, 10 and 15 years for patients treated in CR1 was 59, 48 and 39%, respectively, whereas the corresponding figures for patients treated in responses other than CR1 were 41, 28 and 26%, Leukemia

respectively (Figure 1b). The number of previous therapy lines had no impact on PFS for patients treated in CR1.

Overall survival and causes of death A total of 330 patients died, the median OS being 12 years. Two hundred and thirty-seven died following disease progression: 233 due to disease and 4 due to a second malignancy. The causes of death for the 93 remaining patients were as follows: treatment-related toxicity: 61 (66%), second malignancies: 25 (27%) and other causes: 7 (7%). OS at 5, 10 and 15 years was 64, 52, and 47%, respectively (Figure 2a). Older age (X45 years), disease status at HDT other than CR1, interval from diagnosis to HDTX1 year, more than two lines of treatment before HDT and a TBI-containing regimen were all associated with shorter survival on univariate analysis. The variables that retained significance on multivariate analysis (Table 2) were age (Po0.001), disease status (Po0.001), source of the stem cells (P ¼ 0.007) and conditioning regimen (P ¼ 0.004). After 5 years of follow-up, the OS was shorter for patients receiving a TBIcontaining regimen, no difference being apparent during the first 5 years of follow-up (Figure 3). OS at 5, 10 and 15 years for patients treated in CR1 was 75, 72 and 60%, respectively; when those patients were excluded from the analysis, the corresponding figures were 62, 48 and 44%, respectively (Figure 2b). Amongst patients treated in CR1, the number of previous therapy lines did not influence OS.

Non-relapse mortality Ninety-three patients died without evidence of disease progression, 53 (57%) less than 1 year after HDT. The causes of death for the latter were: infection: 36 (68%) (bacterial: 12, nonspecified: 9, interstitial pneumonitis: 8, viral: 5, Pneumocystis carinii: 1, fungal: 1); organ failure: 4 (7%); other causes: 5 (9%) (cardiac toxicity: 2 and veno-occlusive disease, adenocarcinoma of the jejunum, and graft failure, 1 each) (9%) and nonspecified treatment-related: 8 (15%). Forty patients died without disease progression more than 1 year after HDT. Their causes of death were second malignancy: 24 (60%), HDT-related causes: 9 (22%) (non-specified treatment related: 4, organ failure: 2 and cardiac toxicity, viral infection and interstitial pneumonitis, 1 each). Seven patients died from other causes (18%) (not specified: 4, myocardial infarction: 2 and bronchiolitis: 1). NRM at 1 and 5 years was 6 and 9%, respectively, for both the overall series and for patients who were not in CR1 at the time of HDT. There was no significant difference in NRM in relation to time period. The variables that influenced NRM on univariate analysis were age, disease status and conditioning regimen. With regard to TBI, the NRM at 5 and 10 years was 4 and 6% for patients not receiving TBI and 11 and 16% for the remainder (P ¼ 0.002). On multivariate analysis (Table 2), age at HDTX45 years (Po0.001), chemoresistant disease at HDT (Po0.001) and a TBI-containing regimen were predictive for a higher NRM. TBI-containing regimens did not have an impact on early NRM, but were associated with a higher NRM after 5 years of follow-up (P ¼ 0.04).

Second malignancies Sixty-four patients (9%) developed a second malignancy at a median of 7 years after HDT (range: 1–17). The CI of second malignancies at 5, 10 and 15 years was 2, 5 and 21%, respectively, for patients treated in CR1, whereas it was 4, 9 and 15% for the remainder (P ¼ NS). The most frequent was


2327 Table 2 Multivariate analysis for risk of recurrence, PFS, OS, NRM, risk of second malignancies, risk of secondary MDS/AML and risk of dying of MDS/AML Relative risk (95% CI)

P-value

Adverse prognostic factors for recurrence or progression Chemoresistant disease at HDT versus CR1 Chemosensitive disease at HDT versus CR1 Non-TBI regimens

3.5 (2.3–5.3) 1.5 (1.2–2.1) 1.2 (1.0–1.5)

o0.001 0.003 0.05

Adverse prognostic factors for PFS Chemoresistant disease at HDT versus CR1 Chemosensitive disease at HDT versus CR1 Age at HDT X45 years

3.8 (2.6–5.4) 1.5 (1.2–2.0) 1.3 (1.1–1.6)

o0.001 0.001 0.003

3.7 1.8 1.5 2.0

(2.4–5.9) (1.3–2.5) (1.2–1.9) (1.2–3.2)

o0.001 0.001 o0.001 0.007

F 2.1 (1.3–3.6)

F 0.004

Adverse prognostic factors for OS Chemoresistant disease at HDT versus CR1 Chemosensitive disease at HDT versus CR1 Age at HDT X45 years BM+PBPC as source of stem cells TBI-containing regimens First 5 years after HDT: no impact on OS X5 years after HDT Adverse prognostic factors for NRM Refractory disease at transplant Age at HDT X45 years TBI-containing regimens First 5 years after HDT: no impact on NRM X5 years after HDT

o0.001 o0.001

4.6 (2.1 - 10.5) 2.9 (1.8–4.6) F 3.1 (1.1–9.0)

F 0.04

Adverse prognostic factors for second neoplasms XTwo previous lines of treatment Age at HDT X45 years TBI-containing regimens

2.4 (1.0–5.3) 2.2 (1.3–3.7) 3.6 (1.7–7.4)

0.04 0.003 0.001

Adverse prognostic factors for secondary MDS/AML TBI-containing regimens Age at HDT X45 years

4.1 (1.5–10.9) 2.4 (1.2–4.7)

0.005 0.01

Adverse prognostic factors for dying of second MDS/AML TBI-containing regimens Age at HDT X45 years HDT p1990

4.7 (1.6–13.3) 2.7 (1.3–5.8) 2.2 (1.0–4.9)

0.004 0.009 0.04

1.0

1.0

1.0

1.0

0.8

0.8

0.8

0.8

0.6

0.6

0.6

0.6

0.4

0.4

0.4

0.4

0.2

0.2

0.2

0.2

0.0

First CR n=131

Other responses n=562

0.0 0

3 6 9 12 Years after HDT

15

Overall Survival

Progression-free Survival

Abbreviations: BM, bone marrow; CR, complete response; HDT, high-dose treatment; MDS/AML, myelodysplastic syndrome/acute myeloid leukaemia; NRM; non-relapse mortality OS, overall survival; PBPC, peripheral blood progenitor cells; PFS, progression-free survival; TBI, total-body irradiation.

Other responses n=562

0.0

0.0 0

3

6

9

12

15

Years after HDT

First CR n=131

0

3

6

9

Years after HDT

12

15

0

3

6

9

12

15

Years after HDT

Figure 1 (a) Progression-free survival in 693 patients who received HDT. (b) Progression-free survival according to disease status at HDT.

Figure 2 (a) Overall survival in 693 patients who received HDT. (b) Overall survival according to disease status at HDT.

secondary myelodysplastic syndrome/acute myeloid leukaemia (MDS/AML). The specific diagnoses are listed in Table 3. The median time from HDT to diagnosis of a second malignancy was 5 years for MDS/AML and 9 for the remaining second neoplasms. The diagnoses of second malignancy were made before progression of lymphoma in 53% of the patients for both MDS/AML and other malignancies. Forty-six patients with a

second malignancy (72%) died (29 due to the second neoplasm and 17 due to lymphoma progression). Thirty-three of 40 patients with secondary MDS/AML died (82%), 20 due to MDS/ AML (17 without having developed disease progression). Second malignancies were diagnosed in 54 patients (13.5%) who had received a TBI-containing regimen, contrasting with 10 patients (3.5%) who received a chemotherapy-only regimen Leukemia


2328 1.0

p = 0.004

Table 3 Second malignancies diagnosed in 693 patients receiving HDT according to use of TBI in the conditioning regimen

Overall Survival

0.8 No TBI n=289 MDS/AMLa Lung/breast/colon/prostate Cutaneous non-melanoma Other Not specified

0.6 0.4 TBI n = 401

Alive TBI yes/no

Dead TBI yes/no

Total TBI yes/no

6/1 2/2 2/1 3/0 0/1

29/4 2/0 1/0 5/1 4/0

35/5 4/2 3/1 8/1 4/1

Abbreviations: HDT, high-dose treatment; MDS/AML, myelodysplastic syndrome/acute myeloid leukaemia; TBI, total-body irradiation. a Including one myeloproliferative disorder.

0.2 0.0 0

3

6 9 Years after HDT

12

15

Figure 3 Overall survival for 693 patients treated with HDT according to conditioning regimen.

(Po0.001). Likewise, the proportion of patients developing secondary MDS/AML was significantly higher in those who received TBI as part of the HDT: 34 of 401 (8.5%) versus 5 of 289 (1.7%) (Po0.001). A multivariate analysis for the risk of developing a second neoplasm was carried out (Table 2), including gender, age at HDT, number of previous lines of chemotherapy, interval from diagnosis to HDT, source of stem cells, year of HDT and conditioning regimen. The variables that predicted for a higher risk of second malignancies were (1) the use of a TBI-containing regimen (Po0.001), (2) age at HDTX45 years (P ¼ 0.003) and (3) more than two previous therapy lines (P ¼ 0.04). When the same analysis was repeated for patients who developed MDS/AML, the use of TBI (P ¼ 0.005) and age at HDTX45 years (P ¼ 0.01) were associated with a higher risk. Finally, a multivariate analysis for the risk of dying of secondary MDS/AML was carried out (Table 2) including age at HDT, gender, interval from diagnosis to HDT, year of HDT, previous lines of chemotherapy, source of stem cells and conditioning regimen. The use of a TBI-containing regimen was associated with a higher risk of dying due to MDS/AML (P ¼ 0.004), as well as age at HDTX45 years (P ¼ 0.009) and year of HDT o1990 (P ¼ 0.04).

Discussion This analysis describes the outcome of the largest series of patients with FL treated with HDT with a very long follow-up, indispensable to draw any meaningful conclusion in a disease with a long median survival such as FL. On the downside, it has the caveats inherent to a retrospective registry study, namely lack of histological review, heterogeneous management of patients in terms of treatment and follow-up and missing data. Nonetheless, the patients included can be regarded as representative of a standard group of patients with FL as their characteristics at diagnosis do not differ substantially from the main characteristics of patients with de novo FL,1 apart from being younger, in keeping with other series of HDT in FL. Unfortunately, the retrospective nature of this registry study, precluded obtaining accurate information on risk factors at diagnosis, including cytological grade. However, the characteristics of the patients at the time of HDT may be somewhat different from other series given that 1/5 of the patients in the Leukemia

current study received HDT in CR1. This is predominantly a consequence of the large proportion of patients reported from French centres, where at that time there was an open study and subsequently two randomized studies analyzing the effect of HDT in patients with FL in CR1.11,21,22 Nonetheless, although not formally compared, there do not seem to be significant differences between the clinical characteristics of patients treated in CR1 and the remainder (Table 1). In the current study, after a median follow-up of 10 years, the OS and PFS at 15 years were 47 and 27%, respectively. The results here reported, even after excluding patients treated in CR1 (OS and PFS at 15 years, 44 and 26%, respectively), are better than those achieved with conventional chemotherapy, as reported in other series comparing the outcome of patients treated with HDT with those of historical controls.8,9 A randomized study confirmed these results: the ‘CUP’ trial assigned patients with recurrent FL to receive chemotherapy, HDT with a ‘purged’ stem cell collection or ‘unpurged’ HDT.13 In this study, as in the present one, ex vivo manipulation did not influence the outcome. In the current study, the OS and the PFS stabilized after 10 years of follow-up (OS at 10 and 15 years: 52 and 47%; PFS at 10 and 15 years: 31 and 27%, respectively) with a suggestion of a plateau that has also been reported in another series with a long follow-up.23 It seems, thus, that there is a subgroup of patients with FL who remain alive and free of disease after HDT. The group that apparently obtains the greatest benefit from HDT in the current series was younger patients and those treated in CR1. Unfortunately, the nature of the EBMT registry prevented us from obtaining reliable data on variables such as haemoglobin level or number of nodal areas involved at diagnosis to be able to retrospectively calculate the follicular lymphoma international prognostic index (FLIPI) score and to assess its prognostic value. The good prognosis of patients treated in CR1 is not surprising. A reduced number of previous lines of chemotherapy4,6 and a short interval from diagnosis5 have been associated with a better outcome in other series. Whether it follows that patients with FL should receive HDT at first response is not a question to be addressed in a retrospective analysis. However, three randomized trials have explored this possibility. The Groupe Ouest Est d’Etude des Leucemies Aigues et Maladies du Sang (GOELAMS) study and the German Low Grade Lymphoma Study Group randomized patients to receive ‘standard’ chemotherapy or HDT. In both studies, patients in the HDT arm had a significant longer event-free survival (EFS) than patients in the chemotherapy arm, although no difference in OS was detected.11,12 In contrast, a third randomized trial led by the Groupe d’Etude des Lymphomes de l’Adulte did not show significant differences in OS or (EFS) between the two arms after a long follow-up.22 Notwithstanding the good results of HDT,


2329 the German and the ‘GOELAMS’ studies were associated with a higher risk of second malignancies in the HDT group11,24 Fwhich might have accounted for the lack of difference in OSFunacceptable in a population with a median survival at diagnosis of 10 years. Of note, the patients included in the latter two studies had received relatively little treatment before HDT and the conditioning regimen included TBI in both. A critical result of the current analysis is the relationship between the diagnosis of second malignancies, especially secondary MDS/AML, and the use of a TBI-containing regimen. The diagnosis of second malignancies after HDT has been a matter of concern for many years.14–16 The factors associated with this event have been related to characteristics of the patients,14,15,25 previous treatment25–28 and characteristics of the HDT procedure.16,28 TBI-based regimens have been associated with a higher risk of MDS/AML in some studies,14,27 but not in others.26,28 The higher risk of second malignancies associated with TBI in the current series cannot be attributed to a shorter follow-up of patients who received a chemotherapy-only treatment, as the proportion of patients who received TBI before 1990 was similar to that of patients treated subsequently. Moreover, in a multivariate analysis for the risk of developing MDS/AML, including the year of HDT and number of previous lines of therapy amongst other variables, treatment with a TBIcontaining regimen was associated with a fourfold increased risk of developing MDS/AML. Of note, there were not significant differences in the risk of developing MDS/AML between patients treated in CR1 and the remainder. Thus, although the use of TBI was associated with a lower incidence of recurrence in this and other studies,29 its benefit might be outweighed by the higher risk of MDS/AML. The effect of TBI on survival was a consequence of late mortality, as there were no differences in OS during the first 5 years of follow-up. In this regard, TBIcontaining regimens were not only associated with a higher risk of developing MDS/AML, but also with a higher risk of dying of MDS/AML resulting in the increase in late mortality. In conclusion, HDT results in prolonged PFS in selected groups of patients. However, the good results obtained with HDT are counterbalanced by late toxicity. In this sense, TBI-containing regimens should be avoided given their strong association with the diagnosis of secondary MDS/AML. Nevertheless, these results should be viewed against the perspective of major advances in the treatment of FL since these patients were treated, namely the use of monoclonal antibodies and the advent of RIC allograft.

Acknowledgements We are indebted to all transplantation centres, physicians, nurses and data managers for their continued help in EBMT studies. A list of the transplant centres and investigators of the European Blood and Marrow Transplantation Group appears in the ‘Appendix’. SM was kindly supported by a grant from Olivia Walduck’s family.

References 1 Armitage JO, Weisenburger DD. New approach to classifying nonHodgkin’s lymphomas: clinical features of the major histologic subtypes. Non-Hodgkin’s Lymphoma Classification Project. J Clin Oncol 1998; 16: 2780–2795. 2 Johnson PW, Rohatiner AZ, Whelan JS, Price CG, Love S, Lim J et al. Patterns of survival in patients with recurrent follicular lymphoma: a 20-year study from a single center. J Clin Oncol 1995; 13: 140–147.

3 Freedman AS, Ritz J, Neuberg D, Anderson KC, Rabinowe SN, Mauch P et al. Autologous bone marrow transplantation in 69 patients with a history of low-grade B-cell non-Hodgkin’s lymphoma. Blood 1991; 77: 2524–2529. 4 Rohatiner AZ, Johnson PW, Price CG, Arnott SJ, Amess JA, Norton AJ et al. Myeloablative therapy with autologous bone marrow transplantation as consolidation therapy for recurrent follicular lymphoma. J Clin Oncol 1994; 12: 1177–1184. 5 Bastion Y, Brice P, Haioun C, Sonet A, Salles G, Marolleau JP et al. Intensive therapy with peripheral blood progenitor cell transplantation in 60 patients with poor-prognosis follicular lymphoma. Blood 1995; 86: 3257–3262. 6 Bierman PJ, Vose JM, Anderson JR, Bishop MR, Kessinger A, Armitage JO. High-dose therapy with autologous hematopoietic rescue for follicular low-grade non-Hodgkin’s lymphoma. J Clin Oncol 1997; 15: 445–450. 7 Freedman AS, Neuberg D, Mauch P, Soiffer RJ, Anderson KC, Fisher DC et al. Long-term follow-up of autologous bone marrow transplantation in patients with relapsed follicular lymphoma. Blood 1999; 94: 3325–3333. 8 Apostolidis J, Gupta RK, Grenzelias D, Johnson PW, Pappa VI, Summers KE et al. High-dose therapy with autologous bone marrow support as consolidation of remission in follicular lymphoma: long-term clinical and molecular follow-up. J Clin Oncol 2000; 18: 527–536. 9 Brice P, Simon D, Bouabdallah R, Belanger C, Haioun C, Thieblemont C et al. High-dose therapy with autologous stemcell transplantation (ASCT) after first progression prolonged survival of follicular lymphoma patients included in the prospective GELF 86 protocol. Ann Oncol 2000; 11: 1585–1590. 10 Gratwohl A, Baldomero H, Passweg J, Frassoni F, Niederwieser D, Schmitz N et al. Hematopoietic stem cell transplantation for hematological malignancies in Europe. Leukemia 2003; 17: 941–959. 11 Deconinck E, Foussard C, Milpied N, Bertrand P, Michenet P, Cornillet-LeFebvre P et al. High-dose therapy followed by autologous purged stem-cell transplantation and doxorubicinbased chemotherapy in patients with advanced follicular lymphoma: a randomized multicenter study by GOELAMS. Blood 2005; 105: 3817–3823. 12 Lenz G, Dreyling M, Schiegnitz E, Forstpointner R, Wandt H, Freund M et al. Myeloablative radiochemotherapy followed by autologous stem cell transplantation in first remission prolongs progression-free survival in follicular lymphoma: results of a prospective, randomized trial of the German Low-Grade Lymphoma Study Group. Blood 2004; 104: 2667–2674. 13 Schouten HC, Qian W, Kvaloy S, Porcellini A, Hagberg H, Johnson HE et al. High-dose therapy improves progression-free survival and survival in relapsed follicular non-Hodgkin’s lymphoma: results from the randomized European CUP trial. J Clin Oncol 2003; 21: 3918–3927. 14 Darrington DL, Vose JM, Anderson JR, Bierman PJ, Bishop MR, Chan WC et al. Incidence and characterization of secondary myelodysplastic syndrome and acute myelogenous leukemia following high-dose chemoradiotherapy and autologous stem-cell transplantation for lymphoid malignancies. J Clin Oncol 1994; 12: 2527–2534. 15 Milligan DW, Ruiz De Elvira MC, Kolb HJ, Goldstone AH, Meloni G, Rohatiner AZ et al. Secondary leukaemia and myelodysplasia after autografting for lymphoma: results from the EBMT. EBMT Lymphoma and Late Effects Working Parties. European Group for Blood and Marrow Transplantation. Br J Haematol 1999; 106: 1020–1026. 16 Friedberg JW, Neuberg D, Stone RM, Alyea E, Jallow H, LaCasce A et al. Outcome in patients with myelodysplastic syndrome after autologous bone marrow transplantation for non-Hodgkin’s lymphoma. J Clin Oncol 1999; 17: 3128–3135. 17 www.ebmt.org/4Registry/registry3.html MED-AB Manual 2.pdf, June 2007. 18 Cheson BD, Horning SJ, Coiffier B, Shipp MA, Fisher RI, Connors JM et al. Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas. NCI Sponsored International Working Group. J Clin Oncol 1999; 17: 1244–1253. 19 Klein JP, Rizzo JD, Zhang MJ, Keiding N. Statistical methods for the analysis and presentation of the results of bone marrow transplants. Leukemia


2330 20

21

22

23

24

Part I: unadjusted analysis. Bone Marrow Transplant 2001; 28: 909–915. Klein JP, Rizzo JD, Zhang MJ, Keiding N. Statistical methods for the analysis and presentation of the results of bone marrow transplants. Part 2: regression modeling. Bone Marrow Transplant 2001; 28: 1001–1011. Colombat P, Cornillet P, Deconinck E, Tourani JM, Gardembas M, Delain M et al. Value of autologous stem cell transplantation with purged bone marrow as first-line therapy for follicular lymphoma with high tumor burden: a GOELAMS phase II study. Bone Marrow Transplant 2000; 26: 971–977. Sebban C, Mounier N, Brousse N, Belanger C, Brice P, Haioun C et al. Standard chemotherapy with interferon compared with CHOP followed by high-dose therapy with autologous stem cell transplantation in untreated patients with advanced follicular lymphoma: the GELF-94 randomized study from the Groupe d’Etude des Lymphomes de l’Adulte (GELA). Blood 2006; 108: 2540–2544. Rohatiner AZ, Davies AJ, Apostolidis J, Neuberg D, Matthews J, Nadler L et al. Myeloablative therapy with autologous bone marrow transplantation for follicular lymphoma at the time of second or subsequent remission. Long-term follow-up. J Clin Oncol 2007; 25: 2554–2559. Lenz G, Dreyling M, Schiegnitz E, Haferlach T, Hasford J, Unterhalt M et al. Moderate increase of secondary hematologic malignancies after myeloablative radiochemotherapy and

Appendix Agnes Buzyn, Hoˆpital Necker, Paris, France [FR][160] 3; Roelof Willemze, Leiden University Hospital, Leiden, The Netherlands [NL][203] 7; Jane Apperley, Imperial College, Hammersmith Hospital, London, United Kingdom [GB][205] 5; Niels Jacobsen, Rigshospitalet, Copenhagen, Denmark [DK][206] 10; J Maertens, University Hospital Gasthuisberg, Leuven, Belgium [BE][209] 4; Norbert C Gorin, Hopital Saint Antoine, Paris, France [FR][213] 38; Stephen Mackinnon, Royal Free Hospital and School of Medicine, London, United Kingdom [GB][216] 1; Kurt Leibundgut, University Hospital Bern, Bern, Switzerland [CH][221] 1; Bernard Rio, Hotel Dieu, Paris, France [FR][222] 8; Antony H Goldstone, University College London Hospital, London, United Kingdom [GB][224] 5; John M Davies, Western General Hospital, Edinburgh, United Kingdom [GB][228] 1; Didier Blaise, Unite´ de transplantation et de the´rapie cellulaire, Marseille, France [FR][230] 22; Roberto Foa, Univ.‘La Sapienza‘, Rome, Italy [IT][232] 6; Eric Deconinck, Hopital Jean Minjoz, Besancon, France [FR][233] 10; Augustin Ferrant, Cliniques Universitaires St Luc, Brussels, Belgium [BE][234] 13; A Schattenber, University Medical Center St Radboud, Nijmegen, The Netherlands [NL][237] 8; Leo F Verdonck, University Medical Centre, Utrecht, The Netherlands [NL][239] 3; Michele Baccarani, Bologna University, S Orsola-Malpighi, Bologna, Italy [IT][240] 5; Pierre Biron, Centre Leon Berard, Lyon, France [FR][241] 8; Arturo Iriondo, Hospital U Marque´s de Valdecilla, Santander, Spain [ES][242] 15; H Van den Berg, Academisch Ziekenhuis bij de Universiteit, Amsterdam, The Netherlands [NL][247] 1; Oumedaly Reman, Centre Hospitalier Universitaire, Caen, France [FR][251] 1; Catherine Cordonnier, Hoˆpital Henri Mondor, Creteil, France [FR][252] 27; JL Harousseau, Hotel Dieu, Nantes, France [FR][253] 37; Shimon Slavin, Hadassah University Hospital, Jerusalem, Israel [IL][258] 19; Anna Sureda, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain [ES][ 260] 5; JP Vernant, Hopital PitieSalpetriere, Paris, France [FR][262] 12; F Guilhot, Hopital La Miletrie, Poitiers, France [FR][264] 2; Gunnar Oberg, University Hospital, Uppsala, Sweden [SE][266] 3; Noel Milpied, CHU Leukemia

25

26

27

28

29

autologous stem-cell transplantation in patients with indolent lymphoma: results of a prospective randomized trial of the German Low Grade Lymphoma Study Group. J Clin Oncol 2004; 22: 4926–4933. Micallef IN, Lillington DM, Apostolidis J, Amess JA, Neat M, Matthews J et al. Therapy-related myelodysplasia and secondary acute myelogenous leukemia after high-dose therapy with autologous hematopoietic progenitor-cell support for lymphoid malignancies. J Clin Oncol 2000; 18: 947–955. Metayer C, Curtis RE, Vose J, Sobocinski KA, Horowitz MM, Bhatia S et al. Myelodysplastic syndrome and acute myeloid leukemia after autotransplantation for lymphoma: a multicenter case-control study. Blood 2003; 101: 2015–2023. Hosing C, Munsell M, Yazji S, Andersson B, Couriel D, de Lima M et al. Risk of therapy-related myelodysplastic syndrome/acute leukemia following high-dose therapy and autologous bone marrow transplantation for non-Hodgkin’s lymphoma. Ann Oncol 2002; 13: 450–459. Krishnan A, Bhatia S, Slovak ML, Arber DA, Niland JC, Nademanee A et al. Predictors of therapy-related leukemia and myelodysplasia following autologous transplantation for lymphoma: an assessment of risk factors. Blood 2000; 95: 1588–1593. van Besien K, Loberiza Jr FR, Bajorunaite R, Armitage JO, Bashey A, Burns LJ et al. Comparison of autologous and allogeneic hematopoietic stem cell transplantation for follicular lymphoma. Blood 2003; 102: 3521–3529.

Bordeaux, Pessac, France [FR][267] 37; P Colombat, Hopital Bretonneau, Tours, France [FR][272] 10; Jacques-Olivier Bay, Fe´de´ration de Greffe de Moelle et de Therapie Cellulaire, Clermont-Ferrand, France [FR][273] 3; GH Jackson, Royal Victoria Infirmary, Newcastle- upon-Tyne, United Kingdom [GB][276] 1; JP Jouet, Hopital Claude Huriez, Lille, France [FR][277] 7; Carlos Solano, Hospital Clıńico Universitario, Valencia, Spain [ES][282] 1; DW Milligan, Birmingham Heartlands Hospital, Birmingham, United Kingdom [GB][284] 3; Ignazio Majolino, Ospedale S Camillo-Forlanini, Rome, Italy [IT][287] 1; Keith Wilson, University of Wales, Cardiff, United Kingdom [GB][303] 2; Alberto Bosi, Ospedale di Careggi, Firenze, Italy [IT][304] 9; Giuseppe Leone, Universita Cattolica S Cuore, Rome, Italy [IT][307] 3; Urs Hess, Kantonsspital St Gallen, St Gallen, Switzerland [CH][324] 1; Rainer Haas, Heinrich Heine Universita¨t, Du¨sseldorf, Germany [DE][390] 1; Rosanna Scime`, Ospedale V Cervello, Palermo, Italy [IT][392] 4; Filomena Floristan, Hospital de Cruces, Barakaldo (Bilbao), Spain [ES][393] 1; Tapio Nousiainen, Kuopio University Hospital, Kuopio, Finland [FI][396] 1; Tapani Ruutu, Helsinki University Central Hospital, Helsinki, Finland [FI][515] 2; Martin Hildebrandt, Charite´ Universita¨tsmedizin Berlin, Berlin, Germany [DE][518] 1; Nicole Gratecos, Hoˆpital de l‘ARCHET I, Nice, France [FR][523] 1; M Janvier, Centre Rene Huguenin, Saint Cloud, France [FR][551] 3; Harry Schouten, University Hospital Maastricht, Maastricht, The Netherlands [NL][ 565] 8; Bjarne Jensen, Herlev Hospital Anker, Herlev, Denmark [DK][568] 3; GJ Ossenkoppel, VU University Medical Centere, Amsterdam, The Netherlands [NL][588] 1; Axel A Fauser, Klinik fu¨r Knochenmarktransplantation, Idar -Oberstein, Germany [DE][592] 1; W Feremans, ULBFHopital Erasme, Brussels, Belgium [BE][596] 2; Fabio Benedetti, Policlinico GB Rossi, Verona, Italy [IT][623] 3; R Schots, University Hospital VUB, Brussels, Belgium [BE][630] 1; Elli Koivunen, Tampere University Hospital, Tampere, Finland [FI][635] 1; Norbert Ifrah, CHRU, Angers, France [FR][650] 6; Tiziano Barbui, Ospedale Bergamo, Bergamo, Italy [IT][658] 2; Thierry de Revel, Hoˆpital Percy, Clamart, France [FR][665] 3; Jean HenriBourhis, Institut Gustave Roussy, Villejuif, France [FR][666] 3; Mauricette


2331 Michallet, Hopital E Herriot, Lyon, France [FR][671] 9; NH Russell, Nottingham City Hospital, Nottingham, United Kingdom [GB][717] 2; Vladimir Koza, Charles University Hospital, Pilsen, Czech Republic [CZ][718] 3; J Besalduch, Hospital Universitari Son Dureta, Palma de Mallorca, Spain [ES][722] 1; Dolores Caballero, Hospital Clıńico, Salamanca, Spain [ES][727] 2; M Komarnicki, K Marcinkowski University of Medical Science, Poznan, Poland [PL][730] 1; Anders Wahlin, Umea University Hospital, Umea, Sweden [SE][731] 2; Jose´ M Moraleda Jimenez ,Hospital Morales Meseguer, Murcia, Spain [ES][735] 6; Elisabeth Koller, Hanuschkrankenhaus, Vienna, Austria [AT][743] 2; Marek Trneny, Charles University Hospital, Prague, Czech Republic [CZ][745] 3; Jean-Pierre Lotz, Hoˆpital Tenon, Paris, France[FR][747] 1; GJ Mufti, GKT School of

Medicine, London, United Kingdom [GB][763] 1; John Gribben, St Bartholomew‘s and The Royal London Hospital, London, United Kingdom [GB][768] 114; Joerg Schubert, University of Saarland, Homburg, Germany [DE][785] 1; Massimo F Martelli, University of Perugia, Perugia, Italy [IT][794] 21; Francesco Rodeghiero, S Bortolo Hospital, Vicenza, Italy [IT][797] 2; C Gisselbrecht, Hoˆpital St Louis, Paris, France [FR][805] 49; Jean Francois Rossi, CHU Lapeyronie, Montpellier, France [FR][926] 7; H Tilly, Centre Henri Becqueral, Rouen, France [FR][941] 11; Ghandi Damaj, University of Amiens: CHU Amiens, Amiens, France [FR][955] 1; Alain Delmer, Hoˆpital Robert Debre, Reims, France [FR][959] 6; Laurent Degos, Institut Universitaire d‘He´matologie, Paris, France [FR][960] 2; Dominique Bordessoule, CHRU Limoges, Limoges, France [FR][977] 2.

Leukemia