¹Dept of Hematology, Hospital Clinic, Barcelona, Spain

²Dept of Hematology, AK St Georg, Hamburg, Germany

³Dept of Hematology, University Medical Center St. Radboud, Nijmegen, The Netherlands

⁴Dept of Hematology, Ospedale San Martino, Genova, Italy

⁵Dept of Onco-Hematology, Ospedale Civile, Ravenna, Italy

⁶Dept of Hematology, Oncology, Transfusion Med, University Hospital Benjamin Franklin, Berlin, Germany

⁷Dept of Hematology, BMT Unit, Hôpital St. Louis, Paris, France

⁸Abt Pädiatrie II, University of Ulm, Ulm, Germany

⁹Dept of Hematology, Hôpital Henri Mondor, Créteil, France

¹⁰Reumatologische Universitätsklinik, Felix Platter Spital, Basel, Swizerland

¹¹Dept of Pediatrics, University Hospital, Tübingen, Germany

¹²Dept of Hematology, Huddinge University Hospital, Huddinge, Sweden

¹³Dept of Hematology, Kantonsspital, Basel, Switzerland

¹⁴Dept of Haematology, Imperial College School of Medicine, Hammersmith Hospital, London, UK

¹⁵Div Internal Med II, University of Leipzig, Germany

Correspondence to: A Bacigalupo, Department of Haematology, Ospedale San Martino, Viale Benedetto XV, 16132 Genova, Italy

The Accreditation Sub-Committee of the EBMT regularly publishes special reports on current practice of haemopoietic stem cell transplantation for haematological diseases, solid tumours and immune disorders. Major changes have occurred since the last report in 1998. Haemopoietic stem cell transplantation today includes allogeneic and autologous stem cells derived from bone marrow, peripheral blood and cord blood. With reduced intensity conditioning regimens in allogeneic transplantation, the age limit has increased, permitting the inclusion of older patients. New indications have emerged, such as autoimmune disorders and AL amyloidosis for autologous, and solid tumours for allogeneic transplants. Other indications, such as autologous transplantation for breast cancer have been challenged. An updated report with revised tables and operating definitions is presented here.

Bone Marrow Transplantation (2002) 29, 639-646. DOI: 10.1038/sj/bmt/1703535

haemopoietic transplantation; indications; recommendations; practice; Europe

In 1996, the Accreditation Sub-Committee of the EBMT published a special report in which allogeneic (allo) and autologous (auto) haemopoietic stem cell transplant (HSCT) procedures were classified in accordance with prevailing practice in Europe.¹ The published table categorised haemopoietic transplant procedures by patient diagnosis, stage of disease, patient age and source of stem cells. Procedures involving predominantly adults and those involving children were classified in the same table. In 1998, the same group published an updated article in which new indications for transplant procedures were recognised and certain indications were modified.² In addition, these issues were addressed separately in adults and children. Since then, some new indications for transplant procedures have been recognised and certain accepted indications have been modified. Allogeneic stem cell transplantation with reduced intensity conditioning may be used for patients not eligible for a standard allogeneic transplant because of age and/or clinical condition. The increasing use of two alternative sources of progenitor cells for allogeneic transplants, ie peripheral blood and cord blood, is also discussed. The updated classification is presented below (Table 1). As in the two previous articles, it was not our intention to review the literature in detail nor to provide reasoned arguments in favour of or against the decision to offer a given patient a transplant in a particular clinical situation. Rather, we have attempted to summarise the opinions and practice of clinicians working in transplant centres in Europe in 2001.^1,2,3

Definitions

Haemopoietic stem cell transplant (HSCT)

HSCT refers to any procedure where haematopoietic stem cells of any donor type and any source are given to a recipient with the intention of repopulating/replacing the haematopoietic system in total or in part. Chimerism studies are used to monitor the replacement. The goal of the procedure should be defined beforehand and documented informed consent from the patient (and donor) should be obtained prior to the procedure.

Donor lymphocyte infusions (DLI)

DLI refers to any situation where lymphocytes from a previous donor of haemopoietic stem cells are given to the same recipient with the intention of treating or preventing relapse, or to shift the balance between donor and recipient haemopoiesis towards donor type. The goal of the procedure should be defined beforehand and documented informed consent from the patient and donor should be obtained prior to the procedure.

Patient age

The age of an individual patient remains one of the most important determinants of outcome following both allogeneic and autologous HSCT procedures. As a broad generalisation it seems reasonable to recommend limits of 65 years for autograft procedures, of 60 years for allograft procedures using HLA-identical sibling donors and of 50 years for unrelated donor transplants. In some cases it might be reasonable to exceed these arbitrary limits. For instance, the use of reduced intensity conditioning regimens in allogeneic transplantation appears to be a reasonable option for patients up to 70 years of age. As in previous reports, for the purpose of this document, patients up to the age of 16 years are classified as children.

Stem cell source

Bone marrow and G-CSF-mobilised peripheral blood progenitor cells are standard sources of haemopoietic stem cells. For autologous HSCT, peripheral blood has become the preferred choice in view of its more rapid haemopoietic reconstitution. For allogeneic HSCT, both sources are used. Both methods have their specific advantages and disadvantages. There are marked differences in toxicity concerning the donors. For recipients, the final issue concerning long-term outcome remains open. Peripheral blood is associated with more rapid engraftment in the recipient. A major concern with allogeneic peripheral blood transplantation (allo-PBT) is the high incidence of chronic GVHD. Although it has been shown in some studies that for patients with advanced disease transplant-related mortality and disease-free survival are better with allo-PBT than with allo-BMT, these results have not been confirmed in individuals transplanted with early disease. For selected situations, such as high stem cell dose allografts from non-identical donors, peripheral blood remains the sole choice. Preliminary results fail to show an advantage for peripheral blood compared to bone marrow in severe aplastic anaemia. In general, cord blood transplantation is recommended when patients require allogeneic transplantation and do not have an HLA identical or a one-antigen mismatched blood or marrow donor, or the patient's condition requires prompt transplantation. Cord blood must contain at least 2 ´ 10⁷ nucleated cells/kg before thawing. HLA identity is preferable but not mandatory. Cord blood units should be selected firstly according to nucleated cell dose and secondly according to the number of HLA mismatches. Major or minor ABO incompatibility is not a contraindication but ABO compatibility is preferable. Whenever possible, autologous haemopoietic stem cells should be collected and frozen. The indications for the use of cord blood as a source for stem cells in children are identical to the indications listed in Table 2.

Allogeneic transplants with reduced intensity conditioning regimen

Reduced intensity conditioning regimens can be used in the allogeneic setting with the intention of shifting the balance between risk of transplant-related mortality and risk of relapse. Long-term results are still unknown. Allo-HSCT with reduced intensity conditioning may be used for patients up to the age of 70 years, within approved clinical protocols.

Categorisation of transplant procedures

The categories into which different transplant procedures were allocated were defined in the 1996 publication and raised some discussion. These categories were recapitulated with some modifications in the 1998 publication, and have now been modified according to some new developments:

Standard (S)

Transplants categorised as 'standard' are carried out in many centres in Europe, often without entering the patient into an institutional or national study. The results of such transplants are, in general, reasonably well defined and compare favourably (or are superior to) results of non-transplant treatment approaches. The term 'standard' has been incorporated in place of 'routine'. Obviously, defining a transplant as 'standard' does not mean that it is necessarily the optimal therapy for a given patient in all clinical circumstances. 'Standard' transplants may be performed in any specialist centre with experience with HSCT procedures provided they have an appropriate infrastructure as defined by the EBMT guidelines.⁴

Clinical protocol (CP)

The value of transplants for patients included in this category needs further evaluation. Patients are therefore offered the opportunity of undergoing allo- or auto-HSCT in the context of a study that has been designed specifically to cover a series of patients who satisfy defined diagnostic criteria. The protocol may have been written for a single institution or may reflect national or international multicentre collaboration. A research ethics committee or an institutional review board should have approved the protocol. It is implied that the results of the study are intended for presentation to and/or publication for the medical community at large. The clinical protocol may be a randomised comparison of two or more approaches to treatment. Clinical protocols may be performed in a specialist centre with experience with HSCT procedures, with appropriate infrastructure as defined by EBMT guidelines and, optimally, shall meet JACIE standards.^4,5 The EBMT has prepared a document for Rules and Regulations for EBMT Clinical Trials, which could be used as a guideline (http://www.ebmt.org).

Developmental (D)

Transplants have been classified as developmental if there is little experience with this particular type of transplant. In general, such transplants will involve single cases or small pilot series undertaken by transplant units with acknowledged expertise in the management of that particular disease. The category also covers fundamentally new approaches to the management of a disease that, in a different status or phase, may already be classified under standard or clinical protocol, as above. Protocols for 'developmental' transplants will have been approved by local research ethics committees and must follow the FDA's 'Good Clinical Practice' guideline (www.mcclurenet.com/FedRegisterPDFs/E6.pdf). Again, it is implied that the results of 'developmental' transplants are intended for presentation or publication. Centres performing transplants under the category of 'developmental' should meet JACIE standards.⁵ The document for Rules and Regulations for EBMT Clinical Trials could also be used as a guideline.

Not generally recommended (NR)

This category covers procedures contemplated for a disease in a phase or status in which patients are not conventionally treated by HSCT. Clearly, there will be some overlap between 'NR' and 'D'. It also includes early disease stages when results of conventional treatment do not justify the additional risk of transplant-related mortality, or when the disease is so advanced that the chance of success is so small that the risk of the harvest procedure for the normal donor is difficult to justify. 'NR' may not apply to specific situations where a syngeneic donor exists.

Status of transplants in specific diseases in adults

The updated classification of HSCT procedures in adults is shown in Table 1. A version of the text published in 1996 and 1998^1,2 with appropriate modifications appears below:

Acute myeloid leukaemia (AML)

Patients with AML in first remission may be treated by allo- or auto-HSCT on an individual basis or within the context of a clinical study. Some groups discourage allograft transplants for patients in first remission with cytogenetically 'favourable' subtypes: t(8;21) and inv(16). Allograft procedures for patients in first remission with t(15;17) are not recommended. However, in those patients diagnosed with M3 with molecular PML/RARA persistence after consolidation or salvage treatment, an allogeneic transplant from an HLA identical sibling (S), or from an unrelated donor (CP) is recommended. Patients in first remission considered as 'high risk group' due to specific cytogenetic abnormalities (eg monosomy 7) are candidates for allo-HSCT from either an HLA-identical sibling or an unrelated donor. Patients who fail to achieve complete remission after one course of induction chemotherapy may be treated by allo-HSCT with an HLA-identical sibling, or if time permits, from a matched unrelated donor. Patients with refractory AML in early relapse, or in second or later remission may also be treated by allo-HSCT. Patients in established relapse are not generally recommended for allo-HSCT. Transplants involving unrelated donors for AML in remission should proceed largely in the context of a clinical protocol. Patients with AML in first remission, early relapse or later remission may be treated by auto-HSCT with or without purging of the graft. The results of transplant procedures for AML must be compared with results of contemporary chemotherapy regimens.

Acute lymphoblastic leukaemia (ALL)

Patients with standard risk ALL who relapse after chemotherapy are candidates for allo-HSCT from an HLA-identical sibling in second remission. Allo-HSCT from an unrelated donor may be a reasonable alternative, in the context of a clinical protocol. Patients with ALL with poor prognostic features, eg t(9;22) or t(4;11), are currently considered candidates for allo-HSCT from either an HLA-identical sibling or an unrelated donor. As with AML, ALL patients who fail to achieve complete remission after one course of induction chemotherapy may receive an allo-HSCT from an HLA-identical sibling, or if time permits, from an unrelated donor in first CR.

Chronic myeloid leukaemia (CML)

Patients with CML under the age of 60 years may receive an allo-HSCT from a sibling donor, or when under the age of 50 an unrelated donor may be used as standard practice. The transplant should ideally be performed in chronic phase within 1 year from diagnosis. Patients in advanced phases may be offered transplants on an individual basis. The timing of transplantation for patients with compatible donors who achieve a major or complete cytogenetic response to interferon-alfa or STI571 must be determined. Autografting for patients in chronic phase up to an age of 65 years may be performed for selected patients within approved clinical protocols.

Myeloproliferative disorders other than CML

A small number of patients with primary polycythaemia, or primary myelofibrosis have received auto- or allo-HSCT. It would seem reasonable, however, that any patient in this category with 'high-risk' disease who has an HLA-identical sibling could be considered for allografting within a clinical protocol. Allo-HSCT using alternative donors or auto-HSCT for patients in this category should be regarded as 'developmental'.

Myelodysplastic syndromes (MDS)

Allo-HSCT is considered the treatment of choice for patients with MDS or secondary AML, and offers a good chance of long-term disease-free survival if the transplant is performed at an early stage of the disease or if the patient is transplanted in complete remission after chemotherapy. Auto-HSCT is currently under consideration for MDS patients who achieve complete remission after treatment with chemotherapy.

Chronic lymphocytic leukaemia (CLL)

Patients under the age of 60 years with HLA-identical sibling donors may receive allogeneic transplantation as standard therapy, since allo-HSCT is the only therapy with proven curative potential in CLL to date. Such patients will usually be adults with poor prognostic features. Auto-HSCT should currently be performed in the context of a clinical protocol. It is an option for patients under 65 years who have achieved complete remission or very good partial remission after conventional therapy. Allo-HSCT using a matched unrelated donor should be explored within developmental protocols, particularly in younger patients with poor prognostic features who are refractory to conventional therapy, and in whom the prognosis is anticipated to be very poor.

Non-Hodgkin's lymphoma (NHL)

Auto-HSCT is currently offered to patients with relapsed indolent or aggressive NHL, or to patients with aggressive NHL who have two or more poor prognostic features according to the Shipp Index, to consolidate first complete response to conventional chemotherapy. Lymphoblastic lymphoma may be consolidated in remission by auto-HSCT. Allo-HSCT for lymphoblastic lymphoma might be considered for young adults in first remission. Since patients with mantle cell lymphoma do particularly poorly with chemotherapy alone, they should be offered treatment by auto-HSCT or allo-HSCT as soon as possible.

Hodgkin's disease

Auto-HSCT is indicated to treat patients with early or late relapse after first-line therapy including chemotherapy. It may also be offered to patients who do not enter complete remission or who are beyond first relapse. In general, allo-HSCT does not produce better results than auto-HSCT, but it may be considered in the context of clinical protocols for patients with an HLA-identical sibling donor who have relapsed post auto-HSCT, had a short first remission (<1 year), or achieved no remission with first-line treatment.

Myeloma

Auto-HSCT is an indication for patients under 70 years of age who respond to first-line treatment. Allo-HSCT should be considered for patients under 60 years of age who have responded to first-line treatment. Allo-HSCT could be considered after response to second- or third-line treatment, or after partial response to auto-HSCT. Allo-HSCT should not be performed when progressive disease is present or in heavily pretreated, non-responsive patients. Allo-HSCT using an unrelated donor should only be performed as part of selected experimental protocols.

Acquired severe aplastic anaemia (SAA)

Allo-HSCT from an HLA-identical sibling is the treatment of choice for patients with SAA under the age of 45 years. The conditioning regimen should not include radiation because of the high risk of secondary tumours. In older patients, or in the absence of a matched sibling, an initial course of a combination of antithymocyte globulin and cyclosporin should be given. The median time for response after this treatment is 2-3 months. One should therefore, wait at least 4 months for assessment of response before a transplant is undertaken especially if from an alternative donor. Unrelated donor transplants are still associated with significant morbidity and should be undertaken within experimental protocols.

Constitutional SAA, including Fanconi's anaemia

Allo-HSCT is the only curative treatment for patients with constitutional SAA. For patients lacking an HLA-identical sibling donors, transplantation should be considered from a matched unrelated donor in the context of a clinical trial. The conditioning regimen should preferably not include radiation.

Paroxysmal nocturnal haemoglobinuria (PNH)

Small numbers of patients with PNH have been treated with allo-HSCT, which seems to be the only curative approach. It would seem reasonable to treat patients with high-risk disease who have an HLA-identical sibling with allografting, within developmental protocols.

Solid tumours

Despite much research in this field, no single solid tumour can be regarded as a standard indication for auto-HSCT, except for germ cell tumours after relapse. Most autografts should be undertaken as a part of approved clinical studies and most such protocols should now be randomised. Adjuvant autografting for patients with breast cancer who have had four or more positive nodes should be conducted as part of a prospective study; autografting for patients with metastatic disease can be undertaken on a 'D' basis. Prospective European studies of auto-HSCT for patients with metastatic and inflammatory breast cancer, germ cell tumours in relapse or incomplete remission, small cell lung cancer and ovarian cancer with minimal residual disease, are in progress.

Autoimmune disorders

Auto-HSCT following appropriate conditioning is being considered on clinical protocols for selected patients with severe multiple sclerosis, rheumatoid arthritis, and systemic sclerosis. Auto-HSCT for other autoimmune disorders is being considered on a developmental basis. Dependency of high steroid doses above the 'Cushing threshold' and causing skeletal damage could be an indication. In auto-HSCT the use of T cell depletion (and possibly B cell depletion) should be considered.

AL amyloidosis

A number of patients with AL amyloidosis have been treated by auto-HSCT, both in North America and in Europe. Some patients have obtained clear-cut clinical benefit, but the associated transplant-related mortality is cause for concern. Such studies continue in the context of a clinical protocol. Allo-HSCT should be discussed for individual patients on a developmental basis.

Status of transplants in specific diseases in children

Acute myeloid leukaemia (AML)

The results of transplant procedures for AML must always be compared with results of contemporary chemotherapy regimens, which, in general, are better in children than in adults. This is why some groups, such as the I-BFMSG, only recommend HSCT in first remission for children with 'unfavourable' FAB types. The question of whether allo-HSCT leads to better results than does auto-HSCT is also unanswered. In second remission, results of chemotherapy are poor, which is why all types of HSCT are recommended.

Acute lymphoblastic leukaemia (ALL)

Selected patients with poor prognostic factors are currently considered for allo-HSCT in first remission if they have a sibling donor, or in some centres, with an unrelated matched donor. Auto-HSCT, however, cannot be recommended. The situation with allo-HSCT in CR2 is somewhat complicated. Children who relapse very early (during the first year of treatment) or who relapse with T-ALL have no chance of cure with conventional chemotherapy. Thus, allo-HSCT from a sibling or an unrelated donor is recommended. For all later relapses occurring up to 4 years after the first diagnosis, allo-HSCT using a sibling donor can be recommended. At a later stage of the disease, auto-HSCT in CR2 is only recommended within a clinical protocol.

Chronic myeloid leukaemia (CML)

The recommendations for children with CML are identical to those for adults. Due to the lower complication rate in children, the use of unrelated donors may be considered more readily. Allo-HSCT from a sibling donor is the treatment of choice for children with chronic myelomonocytic leukaemia.

Malignant lymphoma

The prognosis of Hodgkin's disease in children is extremely good and relapses are rare. In such cases auto-HSCT can be recommended. Stem cell transplantation in NHL should usually be undertaken as in ALL.

Myelodysplastic syndrome (MDS)

Allo-HSCT from a sibling donor is the treatment of choice for children with standard MDS, as well as secondary AML. Since the chance of cure is extremely poor with conventional chemotherapy, allo-HSCT from an unrelated donor can also be recommended. The role of auto-HSCT in children with MDS remains controversial. A limited number of children with advanced stage MDS has been treated by autografting. If no sibling donor is available, then auto-HSCT may be considered, especially in children who are expected to enter complete remission with relative ease, such as those with a short history, normal or 'favourable' cytogenetics.

Immunodeficiency syndromes/inborn errors

For severe immunodeficiency syndromes such as severe combined immunodeficiency (SCID), allo-HSCT from a matched related or unrelated donor, or a haploidentical HSCT from a parent, is recommended.

Aplastic anaemia, pure red cell aplasia (Blackfan-Diamond) and Fanconi anaemia

For children with acquired severe aplastic anaemia and an HLA-identical family donor, allo-HSCT is the treatment of choice. For patients who lack a compatible family member donor, a course of intensive immunosuppressive therapy (IS) (antilymphocyte globulin and cyclosporin A), possibly with the addition of a haemopoietic growth factor (G-CSF), is indicated. For children who fail their first course of IS, a search should be initiated for an unrelated donor (UD) while they receive a second course of IS. If a 'matched' unrelated donor is identified, the patients should be treated in the context of a clinical protocol. Children with Blackfan-Diamond anaemia who have a matched sibling should be transplanted if they do not respond to steroids and also if they do not eventually become independent of these drugs, since long-term administration is associated with the risk of severe side-effects in children. Children with Fanconi anaemia (FA) should be transplanted if they have a normal HLA-identical sibling donor. For FA patients who lack a genetically HLA-identical donor, a transplant should be considered with a 'matched' unrelated donor or with cord blood stem cells in the context of a clinical protocol.

Haemoglobinopathies

For young patients with homozygous thalassaemia who have an unaffected HLA-identical sibling, allo-HSCT is probably the treatment of choice. Allo-HSCT using an HLA-identical sibling donor should also be considered for young patients with homozygous SS disease who have one of the major complications listed in the 1996 Special Report.¹ Such complications include recurrent pain episodes, pulmonary manifestations, osteoporosis, transient neurological deficits or problems concerning future medical care such as limited availability of medical care or blood products. Studies using unrelated donors are underway as research protocols.

Solid tumours

Despite much research in this field, no single solid tumour can be regarded as a standard indication for blood or marrow stem cell autografting in children. The only exception might be neuroblastoma (stage 4 beyond the age of 1 year, or high-risk factors in lower stages), but autografting should be performed in accordance with approved protocols. An improvement in survival and even disease-free survival can be expected. Autografts for solid tumours should always be undertaken as part of approved clinical studies. Children with solid tumours may benefit from auto-HSCT following high-dose chemotherapy in the following situations:

• Germ cell tumours: after a relapse or with progressive disease
• Ewing's sarcoma: stage IV and after a localised relapse
• Soft tissue sarcoma: stage IV or after a non-resectable relapse
• Neuroblastoma: stage 4 in a child older than 1 year or for lower stages with chromosome 1p deletion and/or n-myc amplification
• Wilms tumour: high-risk histology or relapse
• Osteogenic sarcoma: the value of HSCT is not yet clear
• Brain tumours: children with medulloblastoma and high-grade gliomas responsive to chemotherapy

In general, allo-HSCT cannot be recommended in children with solid tumours. Allo-HSCT may be undertaken, however, on a developmental basis in specialist centres.

Allo-HSCT with reduced conditioning

For patients otherwise not treatable (eg severe infections, heavy burden of chemotherapy, second transplants) allo-HSCT with reduced or minimal conditioning may be carried out within CP. Developmental protocols in particular indications (ie solid tumours) may be undertaken as pilot protocols in specialised centres.

Haploidentical transplantation

The use of haploidentical donors in children may be indicated when no other donor can be found and an otherwise curative approach is not available. For this purpose development or pilot studies can be approved in specialised centres.

Acknowledgements

The authors are grateful for the advice and helpful comments received from a number of individuals across Europe specialising in the use of haemopoietic stem cell transplantation to treat patients with haematological disorders. Special thanks to T Ruutu, JJ Ortega, G Sanz, F Prosper, E Montserrat, P Corradini and P Dreger.

1 Schmitz N, Gratwohl A, Goldman JM. Allogeneic and autologous transplantation for haematological diseases, solid tumours and immune disorders: current practice in Europe in 1996 and proposals for an operational classification. Bone Marrow Transplant 1996; 17: 471-477. MEDLINE

2 Goldman JM, Schmitz N, Niethammer D, Gratwohl A. Allogeneic and autologous transplantation for haematological diseases, solid tumours and immune disorders: current practice in Europe in 1998. Bone Marrow Transplant 1998; 21: 1-7. MEDLINE

3 Gratwohl A, Hermans J, Baldomero H et al. Indications for haemopoietic precursor cell transplants in Europe. Br J Haematol 1996; 92: 35-43. MEDLINE

4 Link H, Schmitz N, Gratwohl A, Goldman JM. Standards for specialist units undertaking blood and marrow stem cell transplants - recommendations from the EBMT. Bone Marrow Transplant 1995; 16: 733-763. MEDLINE

5 Joint Accreditation Committee EBMT-EuroISHAGE (JACIE) Accreditation Manual.

Table 1 Proposed classification of transplant procedures for adults - 2001

Table 2 Proposed classification of transplant procedures for children - 2001