Allogeneic bone marrow transplantation for systemic AL amyloidosis


Dr Gillmore Immunological Medicine Unit, Division of Medicine, Imperial College School of Medicine, Hammersmith Hospital, Du Cane Road, London W12 0NN.


Low-intensity chemotherapy is ineffective in most patients with AL amyloidosis, probably because clinical benefit requires regression of the amyloid deposits, and this occurs only very gradually after the underlying plasma cell dyscrasia has been suppressed. We report the first successful allogeneic bone marrow transplant (allo-BMT) for AL amyloidosis, which after 3 years was associated with complete clinical recovery. This supports the idea that there may be a brief window of opportunity in patients with AL amyloidosis during which dose-intensive chemotherapy is feasible and most likely to produce clinical benefit.

Amyloidosis is characterized by extracellular deposition of protein in a damaging abnormal fibrillar form. The fibrils in systemic AL amyloidosis are derived from monoclonal immunoglobulin light chains, which in most patients are produced by relatively low-grade clonal plasma cell dyscrasias. Myeloma-type chemotherapy remains the mainstay of treatment for AL amyloidosis, although the majority of patients do not benefit from traditional low-intensity regimens ( Kyle et al, 1997 ).

Prospective scintigraphic studies with 123I-labelled serum amyloid P component (SAP), a specific quantitative method for imaging amyloid deposits in vivo ( Hawkins et al, 1990 ), have confirmed that amyloid deposits of AA, AL and other types regress in most patients in whom the supply of the respective amyloid fibril precursor protein can be reduced ( Hawkins, 1994). This provides a new optimistic perspective on the clinical management of amyloidosis, which encourages increasingly vigorous efforts to treat the various underlying disorders.

We report here the first successful case of allo-BMT for systemic AL amyloidosis and show that complete clinical remission of the patient was associated with regression of the amyloid deposits.


A 53-year-old man presented in April 1994 with a 6-month history of periorbital bruising and progressive macroglossia. Initial investigations including full blood count, biochemical profile and a clotting screen were normal, but the serum albumin was 28 g/l (normal range 35–45) and there was proteinuria of 1.4 g/24 h. There was an IgG lambda serum paraprotein of 18.6 g/l with immuneparesis (IgA 0.35 and IgM 0.22). Bone marrow aspirate revealed 24% plasma cells with abnormal morphology. Skeletal X-ray survey identified two asymptomatic radiolucent areas in the left femoral head. The clinical suspicion of AL amyloidosis was confirmed on rectal biopsy. The diagnosis was therefore of systemic AL amyloidosis associated with indolent multiple myeloma. He was treated initially with six cycles of C-VAMP chemotherapy following which the serum paraprotein fell to 1.5 g/l. SAP scintigraphy at this stage identified amyloid deposits in the kidneys and spleen and showed that his whole body amyloid load was moderate.

In view of his good remission and the availability of an HLA-matched sibling, he proceeded to allo-BMT in December 1994. Conditioning comprised total body irradiation (12 Gy in six fractions) and melphalan (110 mg/m2). Cyclosporin and a short course of methotrexate (days 1, 3, 6, 11) were given as prophylaxis for graft-versus-host disease. The post-transplant course was uneventful and the patient was discharged home on day 25. A bone marrow aspirate at 3 months post-transplant showed normal haemopoiesis and < 1% plasma cells.

At follow-up 6 months post-transplantation the patient's clinical condition had stabilized. There was no detectable monoclonal component in the serum or urine, and the serum albumin and 24 h urinary protein had returned to normal levels. SAP scintigraphy was unchanged. However, at his latest review in May 1997 the patient was completely asymptomatic and had returned to full-time employment. Examination was normal as were all serological and urinary investigations. SAP scintigraphy showed only a modest amount of amyloid in the spleen, the deposits elsewhere having regressed (Fig 1).

Figure 1.

23I-SAP scintigraphy. The scan at presentation (left) shows modest amyloid deposits in the spleen and kidneys which had diminished at follow-up (right) 2 years after allogeneic bone marrow transplantation. Tracer elsewhere represents the normal distribution of SAP throughout the blood pool.


The prognosis of systemic AL amyloidosis is only 12–15 months ( Kyle & Gertz, 1995). In the absence of any therapy specific for amyloid, treatment is aimed at suppressing the underlying plasma cell dyscrasia and thereby production of the amyloid fibril precursor protein in a bid to reduce disease progression. Systematic serial studies with radiolabelled SAP scintigraphy have recently indicated that gradual mobilization of amyloid is the usual mechanism which underlies favourable clinical responses to this approach ( Hawkins, 1994; Tan et al, 1995 ). However, only 20% of patients with AL amyloidosis respond to traditional low-intensity myeloma chemotherapy ( Gertz et al, 1991 ) and therefore many therapeutic challenges remain: the plasma cell dyscrasias in most patients with AL amyloidosis are low grade and may be poorly chemosensitive; many patients have multi-system amyloidosis at diagnosis; it is impossible to quantify treatment response in patients whose monoclonal protein cannot be measured, and clinical benefit is typically delayed for months or even years since it depends on the variable capacity of an individual to mobilize amyloid after the underlying plasma cell dyscrasia has been suppressed.

Dose-intensive chemotherapy regimens have not yet been tested in controlled trials in AL amyloidosis, but promising results have been reported in open studies of VAD ( Persey et al, 1996 ), and high-dose melphalan followed by autologous peripheral blood stem cell rescue (PBSCT) ( Comenzo et al, 1996 ). A single patient with systemic AL amyloidosis underwent syngeneic bone marrow transplantation with an excellent clinical outcome supported by evidence of substantial amyloid regression on SAP scintigraphy ( Van Buren et al, 1995 ). Allo-BMT has been reported only once in systemic AL amyloidosis, in a patient who died 14 d later of sepsis ( Mehta et al, 1992 ).

Although advantages of allo-BMT include an uncontaminated harvest, potential graft-versus-tumour effect and the possibility of immunizing donors with tumour antigen, it has not yet been shown to improve survival in myeloma more than autologous transplantation. However, the reduced frequency of minimal residual disease ( Corradini et al, 1997 ), along with a higher and more prolonged rate of complete remission amongst myeloma survivors who have undergone allo-BMT ( Bjorkstrand et al, 1996 ), suggest that advantages may yet emerge. On the other hand, in the EBMT study transplant-related mortality in myeloma was 41% for allo-BMT and 13% for autologous PBSCT ( Bjorkstrand et al, 1996 ).

We have therefore confirmed that allo-BMT is feasible in AL amyloidosis, and in our case was followed by the gratifying resolution of clinical disease in association with regression of the amyloid deposits. This supports the concept that there may be a brief window of therapeutic opportunity in AL amyloidosis when rapid suppression of the monoclonal gammopathy is most likely to produce clinical benefit. However, given the risk of the procedure, evidence is required that allo-BMT has substantial long-term advantages over autologous PBSCT in the treatment of clonal plasma cell dyscrasias before it used more widely in AL amyloidosis.