Clinical Comparison of Non-Myeloablative Conditioning with Anti-Thymocyte Globulin and Fludarabine for Patients with Hematologic Malignancies

Background

OBJECTIVE:

In this study, we retrospectively compared hematopoietic reconstitution, GVHD, complications, primary disease relapse, survival, and quality of life (QOL) in patients treated with non-myeloablative conditioning regimens based on FLU or ATG. We observed that NST based on FLU conditioning resulted in earlier hematopoietic reconstitution, lower aGVHD and extensive cGVHD incidence, better QOL, higher relapse risk, and lower NRM in comparison with ATG-based conditioning NST followed by donor lymphocyte infusion.

Material and Methods

CLINICAL DATA OF DONORS AND RECIPIENTS:

We retrospectively studied patients undergoing allo-HSCT for hematologic malignancies in the Department of Hematology of Guangzhou First People’s Hospital, Guangzhou Medical University and the Department of Hematology of Nanfang Hospital, Southern Medical University between January 1, 1999 and May 31, 2011. No patient with liver, kidney, or heart dysfunction or older than 50 received myeloablative conditioning. In order to study a homogenous cohort, we also excluded patients who received allo-HSCT twice, and those who received chemotherapy for a relapse subsequent to allo-HSCT. Apart from chronic myeloid leukemia (chronic phase) and myelofibrosis, all patients underwent transplantation during their first complete hematological remission period. Patients with CML (chronic phase) and primary myelofibrosis were not treated before transplantation.

Donor hematopoietic stem cells were derived from low-resolution serological HLA-matched (6/6 or 5/6 loci) siblings. All donors and recipients were tested for antigen pp65 and antibody for cytomegalovirus (CMV). Before transplantation, there were 2 and 3 positive cases of CMV-IgG in ATG and Flu groups, respectively. Ganciclovir was used for prophylaxis of cytomegalovirus. Ganciclovir combined with human blood immunoglobulin was used to treat cytomegalovirus infection.

The study protocol received approval from the Ethics Committees at each center (Guangzhou First People’s Hospital, Guangzhou Medical University and Nanfang Hospital, Southern Medical University, 2012-SYL-077). Written informed consents were acquired from all participants or their families according to institutional guidelines.

CONDITIONING REGIMENS:

Fifty-six patients received non-myeloablative conditioning regimens based on immunosuppression with either ATG or FLU, combined with reduced-dose busulfan and cyclophosphamide. The preparative regimen consisted of rabbit anti-human thymocyte globulin (Pasteur Merieux) (3 mg/kg/d×3 d), or FLU (30–35 mg/m²/d×5 d), busulfan (injection; 1.6 mg/kg/d×4 d) or busulfan (capsule; 2 mg/kg/d×4 d), and cyclophosphamide (60 mg/kg/d×2 d).

Both regimens were used in both centers according to clinicians’ choices during treatment. In Guangzhou First People’s Hospital, ATG regimen was used for 18 cases while FLU regimen was used for 20 cases; in Nanfang Hospital 6 were treated with ATG regimen while 12 cases were treated with FLU regimen. No significant time period difference exists for the choice of regimen.

MOBILIZATION AND COLLECTION OF STEM CELLS:

For bone marrow only graft, bone marrow was collected without G-CSF-mobilization and was infused. For peripheral blood graft, donors received subcutaneous injections of 5 μg/kg G-CSF once a day for 5 consecutive days. Peripheral blood stem cells were collected and infused on the fifth day. For graft with both bone marrow and peripheral blood, donors received consecutive subcutaneous injections of 5 μg/kg G-CSF once a day for 5 days. Bone marrow was collected and infused on the fourth day. Peripheral blood stem cells were then collected and infused on the fifth day.

GVHD PROPHYLAXIS:

For prophylaxis of GVHD, CsA and methotrexate (MTX) were administered. MTX was administrated at 15 mg/m² (+1d), and 10 mg/m² (+3d, +6d, +11d). For patients receiving ATG, CsA was administered by intravenous infusion over 16 h at 3 mg/kg/d. When patients were able to eat, they were given 3–5 mg/kg/d CsA orally, divided into 2 doses per day, to maintain a trough plasma concentration of 0.1663–0.3326 μM (1 μM=1201.9 μg/L). In contrast, for patients receiving FLU, when the donor stem cells were engrafted at 3 weeks post-transplantation, CsA was reduced to a dose of 2.5 mg/kg/d, and then adjusted to maintain a trough plasma concentration of 0.0832–0.1633 μM. When aGVHD did not occur, the CsA dose was reduced after 2 weeks to 1 mg/kg/d, and then adjusted to maintain a trough concentration below 0.0832 μM. If GVHD did not occur, CsA dose was gradually tapered and stopped. When GVHD occurred, the CsA dose was increased or low-dose methylprednisolone (≤160 mg/d) or mycophenolate mofetil was added for 1–2 weeks.

DONOR LYMPHOCYTE INFUSION:

If mixed chimerism was achieved without GVHD in the ATG group, patients received the first donor lymphocyte infusion at 4 weeks post-transplant. According to the escalated-dose regimen, donor lymphocyte infusion was carried out at intervals of 20–30 days, 2 to 9 times. Before donor lymphocyte apheresis, donors received mobilization with G-CSF at 5 μg/kg/d for 3 consecutive days.

METHODS OF DETECTING CHIMERISM:

Whole peripheral mononuclear cells were used for chimera analysis. Donor cells were DNA fingerprinted, and sex chromosome and RBC typing was performed twice weekly post-transplantation. Recipient-to-donor cell ratio was quantitatively assessed by fluorescent short tandem repeat [15] weekly in the second and third month after transplantation, and then once every 2 weeks until patients achieved FDC.

EVALUATION OF HEMATOPOIETIC RECOVERY:

We recorded the time at which patients achieved an ANC ≥0.5×10⁹/L, and a platelet count ≥20×10⁹/L and ≥50×10⁹/L in peripheral blood. When the level of hemoglobin exceeded 70 g/L and the platelet count reached ≥20×10⁹/L, infusions of red blood cells (RBC) and platelets were ceased. The infused volumes of RBC and platelets were calculated after transplantation conditioning. The profile of peripheral blood post-transplant was measured every day until ANC≥0.5×10⁹/L and BPC≥50×10⁹/L for hematologic recovery.

EVALUATION OF GRAFT-VERSUS-HOST DISEASE:

GVHD was evaluated according to classic and NIH criteria [15]. We first evaluated aGVHD according to the 1994 consensus conference on aGVHD grading [16]. cGVHD was further subcategorized as limited (involving only localized skin and/or hepatic dysfunction) or extensive, according to the Seattle criteria [17]. The latter category included generalized or localized skin and/or hepatic dysfunction; liver histology indicating chronic active hepatitis; bridging necrosis or cirrhosis; eye dryness: Schirmer test ≤5 mm; lip mucosa biopsies indicating mild dysfunction of salivary glands and oral mucosa; and other organ dysfunctions [17].

aGVHD may present after 3 months and manifestations of aGVHD and cGVHD can be present simultaneously [6,18]. We further analyzed GVHD 100 days post-transplant according to NIH criteria [19]. Briefly, aGVHD is defined as the absence of diagnostic or distinctive features of cGVHD. Therefore, aGVHD includes classic aGVHD occurring within 100 days after transplantation and persistent, recurrent, or late aGVHD (features of aGVHD occurring beyond 100 days). Categories of cGVHD include classic cGVHD (without features or characteristics of aGVHD) and an overlap syndrome in which diagnostic or distinctive features of cGVHD and aGVHD appear simultaneously.

Mouth ulcers caused by conditioning drugs and agranulocytosis at early stage after transplantation were rarely seen after 100 days post-transplant, and therefore was excluded due to our method of comparing GVHD complications in the ATG and FLU groups.

FOLLOW-UP AND QUALITY OF LIFE:

We evaluated the relapse rate of primary disease, infection, survival, and QOL (according to the Karnofsky criteria) at 6, 12, and 24 months after stem cell transplantation.

Overall survival (OS) was defined as survival from first day after transplant to the final visit (follow-up term).

STATISTICAL ANALYSIS:

All statistical analyses were performed with SPSS15.0 software (IBM SPSS Inc., Chicago, US). Data found to be normally distributed are expressed as mean ±SD. For comparison between 2 groups, the independent samples t/t’-test was used. Non-normally distributed data were expressed as median, lower, and upper quartiles (P50, P25, P75). The non-parametric Mann-Whitney U test was used for comparison between groups. Categorical data are shown as frequency (f) and constituent ratio or percentage, analyzed by the Pearson chi-square test. Four-fold table data were switched to Fisher’s exact test method. Survival times are shown as mean and 95% confidence interval (CI). Kaplan-Meier survival curves were calculated for long-term survival, and compared using the log rank test. Data were censored for patients alive at their last follow-up visit. A P value ≤0.05 was considered significant.

Results

BASELINE CLINICAL CHARACTERISTICS OF STUDIED SUBJECTS:

The demographic and clinical characteristics of all patients are presented in Table 1. Apart from those diagnosed with chronic myeloid leukemia and myelofibrosis, all patients underwent transplantation during their first complete hematological remission period. There was no significant difference in patient age, sex, diagnosis, HLA mismatch, sex match, or ABO match between patients receiving FLU and ATG.

POST TRANSPLANTATION PERIPHERAL BLOOD PROFILE AND INFUSION OF BLOOD COMPONENTS:

The post-transplantation peripheral blood profiles of all patients and volume of infused RBCs and blood platelets are listed in Table 1. The lowest white blood cell count was significantly lower in the FLU group than in the ATG group (P=0.0242), the peripheral blood platelet count exceeded 50×109/L earlier in the FLU group than in the ATG group (P=0.0111), and the volumes of infused RBCs and platelets were lower in the FLU group than in the ATG group (P=0.0303 and P=0.0010). However, the time at which the absolute neutrophil count (ANC) exceeded 0.5×109/L post-transplantation did not differ significantly between the 2 groups (P=0.5687).

TRANSFORMATION OF BLOOD TYPE AND CHIMERISM:

Median ABO blood type transformation occurred significantly earlier in the FLU group (45 days post-transplantation) than in the ATG group (90 days post-transplantation) (P=0.019) (Table 2).

All patients achieved successful engraftment of donor stem cells, and we observed no conditioning-associated mortality within 100 days post-transplantation. All patients treated with FLU achieved a chimeric ratio of 95–100% donor-to-recipient cells within 30 days, and all achieved full donor chimerism (FDC) within 45 days post-transplant. However, 30 days post-transplant, patients treated with ATG achieved a chimerism ratio of only 50–80%, and the median time to achieve FDC was 120 (135±105) days (Table 2).

GRAFT-VERSUS-HOST DISEASE:

As shown in Table 3, 4 (16.7%) patients in the ATG group and 16 (50.0%) patients in the FLU group were diagnosed with aGVHD within 100 days after transplantation (P<0.05, Table 3); however, after 100 days there was no significant difference in the fraction of patients in each group diagnosed with acute, classic cGVHD, or overlapping GVHD, according to the new NIH criteria (P>0.05) [19]. A significantly higher fraction of patients treated with ATG (37.5%) than FLU (9.4%) experienced extensive cGVHD (P=0.019), but fewer patients treated with ATG (12.5%) experienced oral complications of GVHD than patients treated with FLU (43.8%) (p=0.018).

OTHER COMPLICATIONS:

We observed the hepatic toxicity in the course of conditioning. As shown in Table 3, although the incidence of liver dysfunction was no statistical difference in the 2 groups (41.7% vs. 68.8%, p=0.058), the incidence of elevated total bilirubin level in the ATG group was significantly higher than that of the FLU group (50.0% vs. 18.8%, p=0.021<0.05). The incidence of sever infection, including pulmonary aspergillosis, interstitial pneumonia, cytomegalovirus infection or septicemia, in the ATG group was higher than that of the FLU group (50% vs. 25%, p=0.024). One (3.1%) and 3 (8.3%) patients had complicated EB virus-associated post-transplant lymphoproliferative disease (PTLD) in the FLU and ATG group, respectively, and there was not statistically significant difference (p=0.565).

FOLLOW-UP OF SURVIVAL:

As shown in Figure 1, the overall survival (OS) expectation of 67.1 months (95% CI, 53.8–65.3) in the ATG group did not differ significantly from the 77.0 months (95%CI, 65.8–80.5) (P>0.05) in the FLU group. The 60.5 months (95% CI, 55.2–71.3) disease-free survival (DFS) expectation in the ATG group did not differ significantly from the 72.1 months (95% CI, 65.4–78.3) (P> 0.05) in the FLU group.

Non-relapse-related mortality (NRM) was significantly less frequent in the FLU group than in the ATG group (P<0.05). The primary relapse rate was significantly higher in the FLU group than in the ATG group (P<0.05) (Table 4). NRM was mainly attributed to severe infections, including pulmonary aspergillosis, interstitial pneumonia, and cytomegalovirus infection.

In 10 recorded relapse cases, 1 case occurred in the ATG group at the stage of mixed chimerism. Complete remission and full donor chimerism was achieved after chemotherapy and donor lymphocyte infusion. All the other cases were in the FLU group. At the time of relapse, 3 cases had complete recipient chimera. Hematologic remission was achieved after chemotherapy. Another 3 cases had mixed chimerism. After chemotherapy and donor lymphocyte infusion, 1 case achieved complete hematologic remission and full donor chimerism. The other 2 cases failed to achieve remission.

We scored QOL according to the Karnofsky criteria at 6, 12, and 24 months after HSCT (Table 4). Although QOL scores did not differ significantly between the 2 groups at 6 months post-transplant, scores were higher in the FLU group at 12 and 24 months after transplantation (P<0.05, P<0.01), and QOL scores in the ATG group declined between 6 and 12 and 6 and 24 months (P<0.05, P<0.01) (Table 4).

Discussion

We retrospectively studied 56 patients who underwent allo-HSCT for hematologic malignancies and we compared hematopoietic reconstitution following immunosuppressant conditioning with either FLU or ATG, combined with reduced-dose busulfan and cyclophosphamide. We observed no regimen-related mortality. All patients achieved successful engraftment. At an early stage post-transplant, mixed and FDC were achieved in both the ATG and the FLU groups. The FLU regimen achieved earlier hematopoietic reconstitution, lower incidence of aGVHD within the first 100 days post-transplant, and extensive cGVHD, better QOL, higher relapse risk, and lower NRM in comparison with the ATG regimen. Although the FLU regimen achieved faster hematopoietic reconstitution, it was associated with a relatively high incidence of severe aGVHD within 100 days and a higher rate of relapse than the ATG regimen. In contrast, the ATG regimen was associated with higher NRM rate and therefore did not translate into better overall or disease-free survival.

During pretreatment, we discovered pretreatment-related toxicity. Although the incidence of liver dysfunction in the 2 groups was considerable, the incidence of elevated total bilirubin level in the ATG group was significantly higher than that of the FLU group.

RBC and platelet counts both decreased significantly in patients treated with ATG, but patients treated with FLU achieved recovery of platelet count earlier than those in the ATG group, and patients treated with FLU required infusion of less RBC and platelets. This observation may reflect FDC occurring at an earlier stage in patients treated with FLU, and may also be related to the previously described impact of ATG on the platelet and RBC requirements during RIC before allo-HSCT [20]. Although complete chimerism occurred far later in the ATG group, hematopoietic stem cells of both donor and recipient collaboratively performed hematopoiesis under conditions of immunological tolerance. As a result, there was no significant difference in the time taken for the peripheral blood profile to recover.

Mattsson et al. reported that severe aGVHD occurred less frequently in patients who gradually converted to FDC[21]. In our sample, aGVHD and severe aGVHD within the first 100 days post-transplant also occurred less frequently in the ATG group than in the FLU group, perhaps due to the early mixed chimerism achieved by the ATG group in contrast to the full chimerism observed in the FLU group.

Recently, other research groups have reported that although the incidence of grade II to IV aGVHD in NST was lower than in conventional HSCT, there was no significant difference in the incidence of GVHD after 100 days [6]. Mixed chimerism after NST was associated with primary disease relapse. When conventional donor lymphocyte infusion was used, primary disease relapse decreased. However, donor lymphocyte infusion after NST resulted in increased cGVHD pseudomorphs [22].

We analyzed the GVHD features 100 days after transplant and found that the incidence of aGVHD, classic cGVHD, and overlap GVHD did not significantly differ between patients treated with FLU and ATG; however, the incidence of oral GVHD was significantly higher in patients treated with FLU than ATG. The oral GVHD was resolved by increasing cyclosporine (CsA) dose and administration of glucocorticoid and mycophenolate mofetil.

One study indicated that the incidence of GVHD, especially cGVHD, correlated with the graft-versus-leukemia effect after allo-HSTC [23]. We observed a lower relapse rate in the ATG group, indicating that a strategy of NST followed by donor lymphocyte infusion achieved a stronger graft-versus-leukemia or tumor effect following administration of ATG than FLU.

The influence of cGVHD on NRM remains controversial. Vigorito et al. conducted a large-scale clinical study of allo-HSTC and reported that survival, NRM, and systemic immunotherapy were not correlated with the clinical features of cGVHD [24]. However, prolonged immunosuppressant use for late aGVHD was correlated with NRM. Persistent cGVHD or aGVHD beyond 100 days post-transplant were both independent predictors for poor survival [24]. Beyond 100 days after transplant, we observed no significant difference in the incidence of aGVHD or cGVHD between patients treated with ATG and FLU. There was also no difference detected in overall survival (OS) expectations between the 2 groups, as would be expected according to the similar persistent GVHD [25]. Refractory GVHD complicated by donor lymphocyte infusion resulting from prolonged immune suppression increased infection and mortality in the ATG group. In this study, NRM was more common in the ATG group than in the FLU group because of higher sever infection such as pulmonary aspergillosis, interstitial pneumonia, cytomegalovirus infection, or septicemia in the ATG group, but more patients died of primary diseases in the FLU group than in the ATG group. It is well known that EBV reactivation is frequently observed after ATG administration. As in the FLU group, the ATG group was also susceptible to EB virus-associated post-transplant lymphoproliferative disease (PTLD), but a statistically significant difference was not found because of the small sample size of this study. Three-year DFS and OS did not differ significantly between these groups, possibly as a result of the 2 risk factors offsetting one another.

Few previous reports have described long-term QOL of HSCT patients. Some studies suggested that the QOL of patients receiving RIC stem cell transplantation was superior to that of those who underwent myeloablative allogeneic and autologous HSCT [26–28], but systemic comparison of QOL in different NST strategies was not reported. In the present study we compared QOL between patients treated with ATG and FLU after stem cell transplantation according to Karnofsky scoring criteria. After 6 months, no statistically significant difference between the 2 groups was observed. Despite a lower incidence of primary disease relapse in the ATG group at 12 and 24 months, the QOL scores of patients in the FLU group were superior to those of the ATG group. In the ATG group, the higher incidence of severe infection, GVHD, and NRM may have been offset by the higher incidence of mortality resulting from primary disease relapse in the FLU group. This observation agrees with a previous study that reporting that refractory late aGVHD and cGCHD may have compromised attitude and behavior, increasing NRM and decreasing QOL scores at 12 and 24 months after transplantation [29]. We also observed lower scores at 12 and 24 months than at 6 months post-transplant in the ATG group.

Conclusions

The conclusions of this small experimental study are limited by the retrospective nature of the study and the limited sample size. Due to analysis of only a few patients at 2 sites, the study was insufficiently powered to make broad generalizations about the applicability of any observed differences between the groups to a larger sample. In summary, the ideal conditioning regimen before allo-HSCT has not yet been established. We attempted to develop novel RIC or non-myeloablative conditioning (NMAC) strategies capable of reducing the incidence and severity of GVHD without compromising the beneficial impact of graft-versus-leukemia activity or immune reconstitution after transplantation, and to improve long-term DFS and QOL in patients with hematologic malignancies. The FLU regimen achieved faster hematopoietic reconstitution, but was associated with a relatively high incidence of severe aGVHD and a higher rate of relapse than the ATG regimen. A recent report showed that, acute rejection rates, renal function, CMV infection, and the adverse effects of low-dose rabbit-ATG induction may be superior to high-dose rabbit-ATG induction therapy in renal transplantation [30]. Use of low-dose ATG effectively improved the survival rate for the partially-matched unrelated stem cell transplantation because the use of ATG lowered the cumulative incidence of sever acute GVHD chronic GVHD [31]. The sample size in these study was probably too small and lacked homogeneity. To explore a better balance between efficacy and complications of immune-suppressants in conditioning regimens, we conclude that further, larger clinical trials of improved regimens will be required to establish a successful conditioning regimen.