Allotransplantation In

Although allotransplantation adds immunotherapeutic effects to the cytotoxic effects and might thus be curative, its use in patients with CLL has been difficult and contentious, partly because many patients with CLL are older and/or have indolent disease, which does not justify aggressive treatment. Even in experienced centers, the treatment-related mortality (TRM) of allogeneic SCT in patients with CLL has been reported to be as high as 36% (6,7,10,11). A recent update of the EBMT database comprising 209 allografted patients with CLL showed a TRM of 40% at 36 mo after transplantation, which is much greater than after standard indications such as acute leukemia or chronic myeloid leukemia (5,13). The causes of these discouraging results are not completely clear, but patient age, selection of poor risk patients with advanced disease and extensive pretreatment, and the CLL-associated incompetence of the immune system may all contribute to the high TRM observed. The recent development of conditioning regimens with reduced intensity may help to improve the tolerability of allogenic SCT in patients with CLL (25,36). The information available to date is too limited to justify the investigation of allografting for CLL in a large phase III multicenter study.

The EBMT has recently updated data on the outcome of 209 allogeneic transplants from their registry (13) (Table 1). There were 163 men (78%) and 46 women (22%), with a median age of 42 yr (range, 22-64 yr). The median interval between diagnosis and transplantation was 45 mo (range, 5-198 mo) and at diagnosis, 76% of patients studied were in stages B or C; 22% of patients had received one conventional line of therapy, 28% two lines, and 50% three lines. These lines included fludarabine for 44 of 209 patients (21%), chlorambucil in 30 of 209 patients (14%), and 32 of 209 (15%) received ChOP. At transplantation, 172 patients were evaluated for response to therapy: 19 of 172 patients (11%) were in CR, 78 of 172 (45%) in PR, and 75 of 172 (44%) in PD. Ninety patients (43%) received BM, 115 PBSCs (55%), and 4 BM and PBPCs (2%). One hundred and sixty-six patients (83%) received an allotransplant from HLA-identical sibling donors, 6 from syngeneic donors (3%), 16 (8%) from matched and mismatched related donors, and 12 (6%) from unrelated donors. For pretransplantation conditioning, 125 of 194 patients (64%) received a TBI-containing regimen and for graft-vs-host disease (GVHD) prophylaxis, 38 of 171 patients (22%) received a T-depleted graft.

After alloT, 183 of 197 patients (93%) were engrafted [within 16 d (range, 0-100 d), achieving 0.5 g/L neutrophils and 25 d (range, 0-214 d), achieving platelets > 50 g/L]. Sixty-five patients (34%) out of 190 who were evaluable developed acute GVHD of grade 2 or higher, and 47 patients (49%) out of 95 who were evaluable developed chronic GVHD (28 limited and 19 extensive). After transplantation, 141 patients were evaluated for disease response: 101 patients (72%) achieved a CR, 19 patients (13%) a PR, and 21 patients (15%) were in stable disease or progressed. Univariate analysis showed that the projected 3-yr survival was 55% (SE = 5%), which was significantly worse than after autoT [79% (SE = 3%); (p < 0.01)] (Fig. 3). At 5 mo after alloT, 75% of patients were still alive. The projected 3-yr TRM was 40% (SE = 5%), which was significantly worse than after autoT [11% (SE = 3%); (p < 0.01)] and the risk of relapse at 3 yr was 27% (SE = 7%) (Fig. 4).

Using log-rank comparisons, we found no significant association between survival and gender or age of recipient. Even though it concerned only a small subset of patients, we found a significant association between survival and stage at diagnosis, with worse survival for stage C patients. Regarding conventional therapy before transplantation, we could not prove any survival difference after alloT and autoT between patients treated with ChOP and others. After alloT, we found a significant association between survival and fludarabine (p = 0.03), survival and chlorambucil (p = 0.02) given before transplantation. Finally, we showed a significant association between survival and stem cell source in favor of PBPCs (p = 0.05). The results of multivariate analysis using the same parameters as after autoT demonstrated a significant association between survival and (1) fludarabine (HR = 0.48) (Fig. 5) and (2) year of transplantation (HR = 0.87).

In an analysis from the IBMT registry, Esteve et al. (14) demonstrated similar results for 46 allogeneic transplantations with 70% CR after transplantation, a TRM of 28%, an overall survival of 56 ± 7%, and a risk of relapse of 23 ± 13%. In a multivariate analysis, the authors showed a significant association between survival and disease status, stage before transplantation, and chemosensitivity.

In a retrospective analysis from EBMT registry, Dreger et al. (36) showed that 72 patients were transplanted after a reduced conditioning regimen for CLL in Europe. The median age was 54 yr (range, 37-65 yr) and the interval between diagnosis and transplantation was 49 mo (range, 8-

MONTHS

Fig. 3. Survival in CLL transplantations. Data from EBMT.

MONTHS

Fig. 3. Survival in CLL transplantations. Data from EBMT.

MONTHS

Fig. 4. Risk of relapse in CLL transplantation. Data from EBMT database.

MONTHS

Fig. 4. Risk of relapse in CLL transplantation. Data from EBMT database.

146 mo). The status at transplant was 3 patients in stage A, 21 patients in stage B, and 46 patients in stage C; all patients were treated before transplantation with a median of three therapeutic lines, 79% received fludarabine, 68% had a short lynmphocyte doubling time, 73% had a high initial lymphocyte count, and 13% were previously autotransplanted. Most of the patients (65/72) received PBSCs as grafts from HLA-identical donors (59/72) and fludarabine associated with an alkylating agent (70%). With this strategy, it takes time to achieve a complete donor chimerism and a disease response (Figs. 6 and 7). The cumulative incidence of acute GVHD higher than grade II was 31% (15% for grades III and IV), and that of chronic GVHD was 51%. The probability of overall survival was 74% (range, 31-87%), the risk of relapse at 2 yr was 27% (range, 14-40%), with a median follow-up of12 mo (range, 1-37 mo). There was a significant association between TRM and GVHD of grade II or higher (p = 0.005) and TBI used during the conditioning (p = 0.05); there was also a significant association between relapse and disease status (p = 0.001),

p<0.01
Fig. 5. Allogeneic transplants for CLL survival: effect of fludarabine given (adjusted for calandar year). Data from EBMT database.
Fig. 6. Allotransplantations after reduced conditioning regimens for CLL (P. Dreger): time to >95% chimerism. Data from EBMT database.

nongenoidentical donor (p = 0.003), short lymphocyte doubling time (p = 0.07), and stage C (p = 0.07).

Was this article helpful?

0 0

Post a comment