BONE-MARROW TRANSPLANTATION IN SEVERE APLASTIC ANÆMIA

Saturday 30 October 1976

BONE-MARROW TRANSPLANTATION IN SEVERE APLASTIC ANÆMIA U.C.L.A. BONE MARROW TRANSPLANT TEAM* U.C.L.A. School of Medicine, Center for the Health Sciences, Los Angeles, California 90024, U.S.A.

Twenty-one patients with severe aplastic anæmia were evaluated at a single hospital between 1972 and 1975. Patients without histocompatible donors were treated conventionally with androgens, corticosteroids, and HLA-matched platelet and granulocyte transfusions. Bone-marrow transplantation was performed in patients with HLA-identical siblings. The two groups had comparable clinical and hæmatological prognostic indicators and received similar supportive therapy. All nine patients who were not transplanted died. Median survival in these patients was 82 days (range 21-545). Seven of the twelve (58%) transplant recipients were alive at 120 days—>930 days (P<0·04). The one-year actuarial survival of the transplant group was 53%. This study shows that bone-marrow transplantation is a rational alternative to conventional therapy in selected patients with aplastic anæmia. Summary

Introduction APLASTic anaemia is characterised by bone-marrow hypoplasia and resultant pancytopenia. In most cases the xtiology remains obscure. The outlook in patients with severe disease is grave, with a median survival of less than 6 months.l2 The mainstays of therapy are elimination of potential myelotoxins, maintenance of hoemoglobin and platelet levels by transfusion, and treatment of infection. Protected environments and granulocyte transfusions have been used but their efficacy is uncertain,3 4 Attempts at stimulating haematopoiesis have, in general, been unrewarding. Androgens have either not been adequately evaluated or have proved ineffective in

controlled trials. 12

Transplantation of normal bone-marrow

may be of defective haematopoiesis. Preliminary survival data from several centres are encouraging.5-9 The efficacy of marrow transplantation should be evaluated in controlled clinical trials. Unfor-

benefit in patients with absent

*Members of the

or

team were: R. P. GALE, M. J. CLINE, J. L. FAHEY, S. FEIG, G. OPELZ, L. YOUNG, M. TERRITO, D. GOLDE, R. SPARKES, N. NAEIM, G. JUILLARD, C. HASKELL, F. FAWZI, G. SARNA, P. FALK.

tunately, the limited number of patients in any one transplantation centre has necessitated data pooling. Inter-centre variability in patient selection and support facilities may significantly affect patient survival. To circumvent this problem, we analysed the clinical course of twenty-one patients with severe aplastic anxmia referred

to U.C.L.A. from 1972 to 1975. Transplantation was performed when an HLA-identical sibling was available and clinical and haematological data indicated a poor prognosis. All patients whether or not they received transplants were given comparable blood-product and antibiotic therapy. Survival amongst trans-

plant recipients was significantly improved. Methods Patient Selection All patients with aplastic anaemia referred to U.C.L.A. medical centre from September, 1972, to December, 1975, were evaluated. Criteria for inclusion in the study included: a

granulocyte-count < 500/fLI, reticulocyte-count <1%, platelet20 000/jjd, two or more bone-marrow biopsy specimens indicating moderate to severe aplasia, and/or bone-marrow aspirate with ≥70% non-myeloid cells. The corrected reticulocyte-count and proportion of non-myeloid cells were calculated as described.’ Additional data were obtained about possible aetiological factors, interval from onset of symptoms to first clinic visit, presenting symptom(s), and previous therapy. HLA typing was performed on all patients, parents, and siblings by means of standard techniques.1O HLA-identical siblings were tested in mixed lymphocyte culture (M.L.c.)." The decision regarding future clinical management was based solely on results of histocompatibility testing. count <

Clinical Management Most patients received a trial course (1-3 months) of oxymetholone (3-5 mg/kg) unless contraindicated. Prednisone (10 mg/m2) was given at the discretion of the senior physician. Patients with granulocyte-counts <500/1 were maintained in reverse isolation and received oral non-absorbable antibiotics (vancomycin, colistin, and nystatin). Febrile patients were randomly allocated to receive carbenicillin and an aminoglycoside antibiotic. Granulocyte transfusions were given when patients remained febrile after 72-96 hours of antibiotic therapy. Platelet transfusions were given for counts <%20 000/1 or for significant haemorrhage. Random HLA-unmatched platelets were used initially in most patients. When these were ineffective, matched platelets from either family members or unrelated individuals were used. When transfusions were clinically indicated frozen red blood-cells were used preferentially.

Transplantation Patients with HLA and

M.L.c.

identical siblings

were

given

922 bone-marrow transplants when conventional therapy was felt to be unsuccessful. Donor buffy-coat-rich plasma was given on day -5, cyclophosphamide, 50 mg/kg intravenously, on days - 4, -3, -2, and -1; and donor bone-marrow on day 0.12 Two patients were prepared by 1000 rad total-body irradiation from a single 60CO source (5 to 8 rad/min). Methotrexate was administered at 15 mg/m2 on day 1 and 10 mg/m2 on days 3, 6, 11, and weekly thereafter for graft-versus-host disease (GVHD) until day 100.9 Marrow donors or family members provided granulocyte and platelet support. Post-transplant bloodproducts were irradiated with 1500 rad. Engraftment was followed by measuring the increase in white blood-cell and platelet counts, peripheral-blood and bone-marrow chromosome analysis, and red blood cell and leucocyte antigens and isoenzymes. Transplant patients received cotrimoxazole twice weekly between days +50 and +150 as prophylaxis for Pneumocystis carinii.

Graft-versus-host Disease The diagnosis of GVHD was based upon the presence of dermatitis, hepatitis, and/or diarrhoea in engrafted patients. Skin biopsy specimens were routinely examined by light and electron microscopy. Liver and rectal biopsy specimens were obtained when clinically indicated. Criteria for diagnosis and staging of GVHD are given elsewhere.9 Skin biopsy specimens in three patients with skin rash without hepatitis or diarrhoea were not diagnostic of GVHD. Patients with stage 11 GVHD were treated with prednisone (100 mg/m2) daily. The dose was reduced by 20 mg/m2 on each of 3 days to a dose of 20 mg/m2 and decreased slowly over several weeks. Patients considered to be at risk of GVHD were those with documented grafts >30 days post-transplant. Data

Analysis Data were analysed by Student’s two-tailed t test. Survival data were analysed by means of a life table and CLINICAL AND LABORATORY DATA

Actuarial survival in patients who received bone-marrow transplants and in those who were managed conventionally.

survival-rate (BMD 11S of the U.C.L.A. Health Sciences Computing Facility)." Results

,

Twenty-one patients with severe aplastic anxmia evaluated. Nine patients without histocompatible donors were treated conventionally while twelve patients received bone-marrow transplants from HLA and M.L.C, identical siblings. The groups were similar with respect to age, sex, presenting symptoms, time from onset of symptoms to first clinic visit, setiology, and prognostic clinical and haematological parameters. Previous therapy was comparable with a majority in both groups having received androgens and corticosteroids (see accompanying table). The nine patients managed conventionally all died, four from infection and five from haemorrhage. Median survival-time from diagnosis was 82 days (range 21-54) days). Three deaths from infection were associated with gram-negative bacteraemia and one with blastomycosis, The patient who survived the longest received twiceweekly HLA-matched platelet transfusions for 18 months with only slight hsemorrhagic complications but succumbed to gram-negative bacteraemia. Five patients received HLA-matched platelet transfusions from family

were

,

unrelated donors. The actuarial survival of conventionally managed patients is shown in the accom-

members

or

panying figure.

*Medlan (range). F.c.v.=first clinic visit. fSee methods. w.B.c.white blood-cells.

p.M.N.polymorphonuclear leucocytes.

The twelve patients in the experimental group received transplants a median of 37 days after diagnosis (range 22-214 days). Ten were prepared with cyclophosphamide and two by irradiation. Patients received a mean dose of 3.9 x 10* nucleated marrow cells per kg body-weight (±0 8, S.E.M.). The procedure was well tolerated, and the mean length of time in hospital was 60

923 are alive 120+ to 950+ days survival at one year was 53%. Actuarial after diagnosis. to full returned have Five patients activity. One paobstructive bronchitis are restricted uent’s activities by from the and one has been recently discharged hospital.

avs. Seven patients (58%)

One

patient with aplasia-associated HBsAg-negative

168 of liver failure. Hepatitis rèsùlved pre-transplantation but recurred on day +150. rhe bone-marrow was normal at the time of death. Four patients rejected their grafts. One patient retransplanted from a second HLA-identical sibling after repeat cyclophosphamide conditioning is alive 277

hepatitis

died

day

on

days post-transplant. Three patients retransplanted from their original donor after repeat immunosuppression with total-body irradiation died too early for critical evaluation. in four of eight patients at risk. One receive methotrexate developed patient GVHD. Stage-II GVHD developed in a second pastage-ui tient. Both had skin, hepatic, and gastrointestinal involvement and were treated with corticosteroids. Late rashes (> 100 days post-transplant) developed in two patients. Dermal biopsy specimens showed evidence of chronic GVHD. Hepatic and gastrointestinal-tract abnormalities were not observed in these two patients. Eight transplants were between ABO-identical pairs and five were mismatched. Of the latter, two were 0-A, one 0-B, and one B-0. Pre-transplant plasma in case (to be reported in used the latter was exchange GVHD

developed

who did

not

detail).

compatibility had no effect graft rejection, GVHD, or survival. ABO

upon

engraftment,

Discussion This study confirms the brief course of severe aplastic anxmia. The median survival of conventionally managed patients was 82 days despite treatment with androgens, corticosteroids, and histocompatible platelet and granulocyte support. Bone-marrow transplantation, in contrast, was associated with a 53% one-year survival. Survival differences were unrelated to patient selection factors or differences in clinical management. Our transplantation results were comparable to several uncontrolled series,6in which grafts failed in approximately 25% and GVHD occurred in 65% The American

College of Surgeons/National Institutes of Health Transplant Registry reported 45% long-term survival in thirty-eight patients.’ Early transplantation, age O21 years and minimal exposure to blood-transfusions

associated with

an

improved

outcome.

were

In the present

series, long-term survival was 58% and graft rejection occurred in 39%. GVHD occurred in 25% of patients at nsk. The

importance of a controlled trial from a single centre is clear. Supportive therapy of these patients is complex, and institutional variability may account for significant differences in patient survival. We attempted to give comparable support to all patients. Platelet support for non-transplant patients was provided by HLA:natched family members or HLA-matched unrelated honors. Transplant patients received platelet and granuIxvte support from marrow donors and family members st-transplant. While HLA-matched blood-products om unrelated donors may not be as effective as those ::’Jm related donors, they represent the best alternative.

Infectious complications were managed similarly in all patients. Laminar-!Mr-now environments were not used. Their value over conventional protective isolation and non-absorbable antibiotics is controversial4 14 and practical considerations limit their use. Moreover, their use without attempting correction of the pathogenesis of the aplasia presupposesi spontaneous restoration of normal haematopoieisis. Proof of this hypothesis is lac-

king. The long-term fate of patients receiving bone-marrow transplants is unknown. Despite successful initial engraftment, four patients subsequently rejected their grafts. While loss of engrafted marrow may be the result of non-immunological (microenvironmental) factors, the time course of engraftment and successful retransplantation suggests an immunological mechanism in most patients. The establishment of normal donor haematopoiesis also accords with defective or absent stem cells in most patients. ABO compatibility was not a significant factor in transplant outcome. This is in contrast to results in skin 15 and kidney grafting16 but similar to those in corneal grafting." In one patient preformed anti-ABO antibodies had to be removed before marrow infusion. This was accomplished by plasma exchange and transfusion of donor red blood-cells Bone-marrow transplantation is useful in treating selected patients with severe aplastic anaemia. Survival of transplant recipients was better than that of patients managed by conventional therapy. Hopefully advances in our understanding of the xtiology of aplastic anxmia, in histocompatibility testing, and in a more effective use of immunosuppressive agents will increase the survival of aplastic patients undergoing transplantation and broaden its applicability. This study was supported by grants CA 12800 and CA 15688 and RR 00865 from the National Institutes of Health, Bethesda, Maryland. R.P.G. ts a scholar of the Leukemia Society of America.

Requests for reprints should be addressed to R. P. Gale, Departof Medicine, U.C.L.A. School of Medicine, Center for Health Sciences, Los Angeles, California 90024, U.S.A.

ment

REFERENCES

1. 2. 3. 4. 5.

Lynch,

R.

E., Williams, D. M., Reading, J. C, Cartwright, G. E. Blood,

1975, 45, 517. Lewis, S. M. Br. med. J. 1965, i, 1027. Preisler, H. D., Bjornsson, S. Semin. Oncol. 1975, 2, 369.

Boggs, D. R. New Engl. J. Med. 1974, 290, 1055. Storb, R., Thomas, E. D., Buckner, C. D., Clift, R. A., Johnson, F. L., Fefer, A., Glucksberg, H., Giblett, E. R., Lerner, K. G., Neiman, P. Blood,

1974, 43, 157. J., Gale, R. P., Stiehm, E. R., Opelz, G., Young, L. S., Feig, S. A., Fahey, J. L. Ann. intern. Med. 1975, 83, 691. 7 Advisory Committee of the Bone Marrow Transplant Registry (U S A.), J. Am Med. Ass. 1976, 236, 1113 8. Camitta, B. M., Thomas, E. D., Nathan, D. G., Santos, G., Gordon-Smith, E. C., Gale, R. P. Blood, (in the press). 9. Thomas, E. D., Storb, R., Clift, R. A., Fefer, A., Johnson, F. L., Neiman, P. E., Glucksberg, H., Buckner, C. D. New Engl. J. Med. 1975, 292, 832,

6. Cline, M.

895. 10.

Mittal, K. K., Mickey, M. R., Singal, D. P., Terasaki, 1968, 6, 913. Sengar, D. P. S., Terasaki, P. I ibid 1971, 11, 260.

P. I.

Transplantation,

11. 12 Santos, G W Semin. Hemat 1974, 11, 341 13. Dixon, W. J. (editor) BMD Biomedical Computer Programs, X-series suppl. Berkley, 1969. 14. Levine, A. S., Siegel, S. E., Schreiber, A. D , Hauser, J., Preisler, H., Goldstein, I. M., Seidler, F , Simon, R., Perry, S , Bennett, J. E., Henderson, E S. New Engl J. Med. 1973, 288, 477. 15. Rapaport, F. T., Dausset, J., Legrand, L , Barge, A., Lawrence, H S., Converse, J. M. J. clin. Invest 1968, 47, 2206 16. Starzl, T. E., Marchioro, T L , Holmes, J. H., Herman, G., Brittain, R. S., Stonington, O. H., Talmage, D. W , Waddell, W. R Surgery, 1964, 55, 195. 17. Batchelor, J. R., Casey, T A., Werb, A , Gibbs, D. C., Prasad, S. S , Lloyd, D. F., James, A. Lancet, 1976, i, 551.