Fred Hutchinson Cancer Research Center and the University of Washington School of Medicine, Seattle, Washington, USA

Marrow grafting from genotypically HLA-identical siblings for the treatment of chronic myelocytic leukemia (CML) began at the Fred Hutchinson Cancer Research Center in the early 1970s (1,2). A review of results in the first 629 patients transplanted shows event-free survivals at 10 years which are at slightly above 60% for patients transplanted in chronic phase, 38% for those in second chronic phase, 30% for those in accelerated phase, and 12% for those in blast crisis. Major problems encountered were acute graft-versus-host disease (GVHD), cytomegalovirus-associated interstitial pneumonia, and leukemic relapse. Relapse was seen in approximately 25% of patients transplanted in chronic phase, 45% of those in accelerated phase, and 75% of those transplanted in blast crisis. During subsequent years, studies were carried out addressing each of the three major problems. The introduction of a combination of methotrexate and cyclosporine in lieu of either methotrexate or cyclosporine alone led to a significant reduction in the incidence of acute GVHD and transplant-related mortality (3,4). The incidence of leukemic relapse remained at 20%. Event-free long-term survival for patients transplanted in chronic phase of the disease rose to 70-75% at 7 years. In regards to the prevention of cytomegalovirusassociated pneumonia, a randomized placebo controlled study showed an acyclovir derivative, ganciclovir, to be effective in preventing serious disease (5). Ganciclovir prophylaxis should have an impact on survival of future patients. Other studies asked whether the risk of leukemic relapse could be altered by alterations in the conditioning programs. In the first study, 116 patients were randomized to receive either cyclophosphamide and 12 ay of fractionated TBI or cyclophosphamide and 15.75 ay of fractionated TBI (6). None of the patients given 15.75 Gy TBI relapsed compared to 25% of those given 12 ay of TBI. With the increase in TBI dose, however , increases in transplanted-related toxicity and mortality were seen, offsetting the gain made by the reduction in leukemic relapse. Accordingly, long-term event-free survival for both groups of patients was the same, approximately 70% . The next study randomized 115 patients to receive either cyclophosphamide and 12 Gy of TBI or cyclophosphamide and 14 mg of busulfan per kg (7). All patients were given methotrexate/cyclosporine for avHD prevention. Preliminary results show 80% survival at 2 years for both groups of patients and a relapse rate of so far 10%. Thus, busulfan may not improve the problem of leukemic relapse. In regards to patients with accelerated phase, small improvements have been made. Patients were either given a combination ofbusulfan/cyclophosphamide along with 12 ay of TBI or cyclophosphamide and 15.75 ay of TBI (R. Clift et al., unpublished). The intensification of the conditioning program decreased the relapse rate in both groups of patients to less than 20% .Survival of patients receiving the busulfan-containing regimen was 55% at 3 years compared to a 48% survival among the patients given cyclophosphamide and 15.75 ay of TBI. Only 35% of patients with CML have an HLA-identical sibling donor. In less than 10% of patients, another suitably matched family member can be identified, either a phenotypically HLA-matched donor or a I-HLA-Iocus mismatched haploidentical donor (8,9). Patients in chronic phase of CML have been conditioned with cyclophosphamide and 12 Gy TBI. The risk of graft failure increased from less than 2% to 7% , and that of avHD from 30% to 70%. Despite the increase in transplant-related problems, survival of patients with haploidentical donors was comparable to that of patients with genotypically HLA-identical siblings, because of a significant decrease in the risk of leukemic relapse. For 30% of patients with CML, either fully HLAmatched or 1-HLA-locus mismatched unrelated donors can be identified, although this figure may increase as the number of volunteer donors in the international registries increases (10,11). Patients with CML in chronic phase have been conditioned with cyclophosphamide and 12 ay TBI while those in more advanced phase of their disease have been administered cyclophosphamide and either 13.2 ay (> 18 years of age) or 14.4 Gy «18 years of age) ofhyperfractionated TBI. Graft failure was seen in 5% , and GVHD in 75 % of cases, but relapse was rare, presumably the result of a graft-versus-Ieukemia effect. Largely because of increased transplant-related mortality, event-free survival at 3 years for recipients of marrow from unrelated donors, either fully phenotypically matched or 1-HLA-Iocus mismatched has been on the order of 50% compared to the 75% seen with genotypically HLAidentical sibling donors. Review of these results and those obtained in other centers worldwide shows the frequent inability of the conditioning programs to eradicate the last leukemic cell to be a persistent major problem. At our Center , preclinical animal experimental work has focused on different ways of delivering TBI and on the use of radiolabeled monoclonal antibodies directed against hematopoietic cells. Most of the clinically used TBI programs deliver radiation at low dose rates, between 5 and 8 cay per minute. We explored whether higher dose rates might offer advantages. Specifically, we compared marrow toxicity , gastrointestinal toxicity , immunosuppressive properties, and late organ toxicities in dogs given single-dose versus fractionated-dose TBI administered either at low dose rates of 5 to 10 cay per minute or at high dose rates of 70 to 80 cay per minute. At 5 to 10 cay per minute, single-dose and fractionated TBI had comparable marrow toxicity (12). Also, the early gastrointestinal toxicity of fractionated TBI was similar to that seen with single-dose TBI (13). However, late organ toxicity was significantly reduced, and long-term survival improved with fractionation. While the sparing of nonhematopoietic tissues by fractionation of radiation is desirable, we were concerned whether fractionated TBI was as immunosuppressive as single-dose TBI. A study in recipients of marrow from DLA-identical littermates showed fractionated TBI delivered at a dose rate of 7 cay per minute to be significantly less immunosuppressive than single-dose TBI as measured by the criterion of graft rejection (14,15). We hypothesized that the greatest benefit of fractionation compared to single-dose TBI may be obtained at higher dose rates. Results (unpublished) showed that single-dose TBI delivered at 80 cay per minute was considerably more toxic to the gastrointestinal tract than TBI given at 5 to 10 cay per minute. The LDo dose of single-dose TBI fell from 14 ay at 5 cay per minute to 7 ay at 80 cay per minute. By comparison, the LDo dose with fractionated TBI fell from 14 ay at 5 cay per minute to only 10 ay at 80 cay per minute. This suggests a relative sparing of gut epithelial cells with fractionation when TBI is given at high dose rates. What was true for gastrointestinal toxicity was also found to be true for the other two parameters studied. Dogs given 300 cay of TBI and no subsequent marrow infusion served to study the question of marrow toxicity . Seven of 21 animals given single-dose TBI delivered at 10 cay per minute survived compared to six of ten given fractionated-dose TBI, a difference which was not statistically significant. None of five dogs given single-dose TBI delivered at 75 cay per minute survived, compared to seven of ten given fractionated TBI, a result which was statistically significantly different. Thus, single-dose TBI at 75 cay per minute is significantly more toxic to the marrow than at 10 cay per minute while fractionated TBI at 75 cay and 10 cay per minute has comparable toxicity .Results suggest that fractionation at 75 cay per minute spares the myeloid marrow compartment. Results on immunosuppression were similar to those on marrow toxicity .To study the immunosuppressive effect of TBI, dogs were given 450 cay of TBI and marrow grafts from DLA-identicallittermates. None of 15 dogs given either single-dose or fractionated-dose TBI delivered at 7 cay per minute showed sustained allogeneic engraftment. Six of seven dogs given single-dose TBI at 70 cay per minute engrafted, a result which was significantly better than that seen at 7 cay per minute (none of 10 dogs engrafted). Only two of ten dogs given fractionated TBI at 70 cay per minute showed sustained allogeneic engraftment, a result which was significantly worse than that seen with single-dose TBI at 70 cay per minute and barely better than that seen with fractionated TBI delivered at 7 cay per minute (none of 5 dogs engrafted). We concluded from these studies that fractionated TBI delivered at dose rates of 70 to 80 cay per minute spared in an equal manner the gastrointestinal epithelial cells, the myeloid compartment of the marrow, and the lymphoid system when compared to single-dose TBI given at 70 to 80 cay per minute. At low dose rates of 5 to 10 cay per minute, gastrointestinal and myeloid toxicities of single-dose and fractionated TBI were indistinguishable, and significant sparing effects with fractionation were seen only with regard to the immune system and late nonhematopoietic organ toxicities. Findings imply that fractionated TBI delivered at high dose rates of 70 to 80 cay per minute is not likely to improve the therapeutic ratio of TBI. One way to deliver radiation to hematopoietic tissues while avoiding dose-limiting toxicity to other organs might be to target radiotherapy specifically using monoclonal antibody coupled to radionuclides. In pursuit of this objective, we have carried out studies in dogs using 131I-labeled anti-la and anti-CD44 antibodies. We chose 1311 as the radionuclide because it is readily available, inexpensive, attaches easily to the antibody and does not harm it, and because it has gamma and beta components which enable use of the same radionuclide for imaging and therapy. Antibodies to la and CD44 were selected because the antigens they recognize are expressed in high numbers on most lymphohematopoietic cells in dogs (16-18). We found that, when trace labeled, the antibodies localized to spleen, marrow, and lymph node more than to any other organ, and that when labeled at high activity , the antibodies ablated marrow function, an effect that could be reversed by infusing autologous marrow. Another antibody, DM5, directed against a wide spectrum of myeloid precursor cells and mature granulocytes, appeared even more effective than the other two antibodies, particularly when used with cold antibody pretreatment before the infusion of the radiolabeled antibody (19). Based on studies in preclinical models, a clinical study has been initiated in patients with advanced leukemia, either in relapse or in second or third remission of acute leukemia (20). A 131I-labeled antibody to CD45 is used. CD45 is expressed on all hematopoietic cells. Doses to the marrow have ranged from 403 to 892 cay, doses to the spleen from 769-2216 cay, and those to the liver have been approximately 350 cay. Treatment with radiolabeled antibody was followed 7-10 days later by the standard conditioning program for an allogeneic transplant, which included 60 mg of cyclophosphamide per kg on each of two successi ve days and 6 x 200 cay of TBI. Preliminary results suggest that 131I-labeled anti-CD45 antibody can target marrow effectively, and at the doses used so far, can be combined with a conventional conditioning program without untoward toxicity .Optimal dose and dose schedule are as yet unknown. It is likely that combinations of antibody radionuclide conjugates and conventional conditioning therapy will effectively reduce the risk of leukemic relapse after marrow transplantation with tolerable toxicity.

ACKNOWLEDGMENTS

This work was supported by grants HL36444, CA18221, CA18029, CA31787, CA18105,
and CA15704 from the National Institutes of Health, DHHS.
The author thanks Bonnie Larson and Harriet Hefton for typing the manuscript.

REFERENCES

1.Clift RA, Buckner CD, Thomas ED, et al. The treatment of chronic granulocytic leukaemia in chronic phase by allogeneic marrow transplantation. Lancet 1982;ii:621-623.

2. Thomas ED, Clift RA, Fefer A, et al. Marrow transplantation for the treatment of chronic myelogenous leukemia. Ann Intern Med 1986;104:155-163.

3. Storb R, Deeg HJ, Whitehead J, et al. Methotrexate and cyclosporine compared with cyclosporine alone for prophylaxis of acute graft versus host disease after marrow transplantation for leukemia. N Engl J Med 1986;314:729-735.

4. Storb R, Pepe M, Deeg HJ, et al. Long-term follow-up of a controlled trial comparing a combination of methotrexate plus cyclosporine with cyclosporine alone for prophylaxis of graft-versus-host disease in patients administered HLA- identical marrow grafts for leukemia. (Letter to the Editor). Blood 1992;80:560-561.

5. Goodrich JM, Mori M, Gleaves CA, et al. Early treatment with ganciclovir to prevent cytomegalovirus disease after allogeneic bone marrow transplantation. N Engl J Med 1991;325:1601-1607.

6. Clift RA, Buckner CD, Appelbaum FA et al. Allogeneic marrow transplantation in patients with chronic myeloid leukemia in the chronic phase. A randomized trial of two irradiation regimens. Blood 1991 ;77: 1660-1665.

7. Buckner CD, Clift RA, Appelbaum FR, Thomas ED. A randomized study comparing two transplant regimens for CML in chronic phase (CP) .Blood 1992;80 (Suppl 1):72a.

8. Anasetti C, Amos D, Beatty PG, et al. Effect of HLA compatibility on engraftment of bone marrow transplants in patients with leukemia or lymphoma. N Engl J Med 1989;320: 197-204.

9. Anasetti C, Beatty PG, Storb R, et al. Effect of HLA incompatibility on graft-versus-host disease, relapse,and survival after marrow transplantation for patients with leukemia or lymphoma. Hum Immunol 1990;29:79-91.

10. Beatty PG, Hansen JA, Longton GM, et al. Marrow transplantation from HLA-matched unrelated donors for treatment of hematologic malignancies. Transplantation 1991;51:443-447.

11. Beatty PG, Anasetti C, Hansen JA, et al. Marrow transplantation from unrelated donors for treatment of hematologic malignancies: effect of mismatching for one HLA locus. Blood 1993;81 :249-253.

12. Storb R, Raff RF, Graham T et al. Marrow toxicity of fractionated versus single dose total body irradiation is identical in a canine model. Int J Radiat Oncol BioI Phys (in press)

13. Deeg HJ, Storb R, Longton Get al. Single dose or fractionated total body irradiation and autologous marrow transplantation in dogs: Effects of exposure rate, fraction size and fractionation interval on acute and delayed toxicity .Int J Radiat Oncol BioI Phys 1988;15:647-653.

14. Storb R, Raff RF, Appelbaum FR, et al. What radiation dose for DLA-identical canine marrow grafts? Blood 1988;72:1300-1304.

15. Storb R, RaffRF, Appelbaum FR, et al. Comparison of fractionated to single-dose total body irradiation in conditioning canine littermates for DLA-identical marrow grafts. Blood 1989;74:1139-1143.

16. Appelbaum FR, Badger C, Deeg HJ, Nelp WB, Storb R. Use of iodine-131-labeled anti-immune response-associated monoclonal antibody as preparative regimen prior to bone marrow transplantation: Initial dosimetry .NCI Monogr 1987;3:67-71.

17. Appelbaum FR, Brown P, Sandmaier Bet al. Antibody-radionuclide conjugates as part of a myeloblative preparative regimen for marrow transplantation. Blood 1989;73:2202-2208.

18. Appelbaum FR, Badger CC, Bemstein ID et al. Is there a better way to deliver total body irradiation? Bone Marrow Transplantation 1992; 10 (Suppl 1):77-81.

19. Bianco JA, Sandmaier B, Brown Pet al. Specific marrow localization of an 1311-labeled anti-myeloid antibody in normal dogs: Effects of a "cold" antibody pretreatment dose on marrow localization. Exp Hematol 1989; 17:929-934.

20. Matthews DC, Appelbaum FR, Eary JF et al. Use of radioiodinated anti-CD45 antibody to augment marrow irradiation prior to marrow transplantation for acute leukemia. Blood 1992;80 (Suppl 1):335a.