| Part□:Clinical study on the treatment of relapsing leukemia after allogeneic hematopoietic stem cell transplantation with granulocyte colony-stimulating-factor through induction of graft-versus-leukemia effectObjective:To explore the efficacy and safety of granulocyte colony stimulating factor (G-CSF) regimen (G-CSF alone and in combination with low-dose cytosine arabinoside, Ld Ara-C) for the treatment of leukemia relapsing after allogeneic hematopoietic stem cell transplantation (Allo-HSCT).Methods:We retrospectively analyzed 2 patients who relapsed following Allo-HSCT and were treated with G-CSF alone before 2008, and starting in January 2008, we initiated a prospective clinical trial of G-CSF in combination with Ld Ara-C for relapsed leukemia post-transplantation, and all 16 consecutive patients who fulfiled the inclusion criteria at our BMT unit were enrolled in this study. There were eighteen patients with leukemia relapsing following Allo-HSCT (10 with acute lymphoblastic leukemia,4 with acute myeloid leukemia,1 with mixed-lineage acute leukemia, and 3 with chronic myeloid leukemia). All the 15 patients with acute leukemia were classified as having a hematological relapse, and 2 and 1 of 3 chronic myeloid leukemia (CML) patients were classified as having a blast crisis and cytogenetic relapse, respectively. Two patients were treated with G-CSF (150μg/d) alone, and the other 16 patients received G-CSF (150-300μg/d) in combination with Ld Ara-C at a dose of 30-50 mg/d. Immunosuppressant agents were abruptly withdrawn in 14 patients at the time of relapse. G-CSF had to be taped in case of WBC>30×109/L and discontinued in case of acute GVHD of≥gradeⅡor extensive chronic GVHD. Ld Ara-C was stoped in case of reduction of blast cells in bone marrow to< 5%. Patients were removed from this study in case of increasing numbers of blast cells in the peripheral blood or bone marrow, or in case of no reduction of blast cells in bone marrow on the day-28 evaluation after the therapy. When patients obtained complete remission, the dose of G-CSF was adjusted for WBC count.Results:□Fourteen (77.8%) of eighteen patients responded to G-CSF regimen at a median of 18.5 days (range from 12 to 95 days).11 (61.1%) of 18 patients achieved complete remission (CR) including complete cytogenetic remission (CCyR) in 2 CML patients with blast crisis at a median of 16 days (range from 12 to 54 days). Long term leukemia-free survival was 11.1%.□The patient with AML who received G-CSF alone achieved CR and the other patient with ALL did not respond to it.13 out of 16 patients (81.3%) responded to the combination of G-CSF and LD Ara-C after a median of 18.5 days (12-95 days) on treatment, among whom,10 entered CR (62.5%) at a median time of 16.5 days (12-54 days) after starting G-CSF therapy.□The values of ANC (absolute neutrophil count) were evaluable in 16 patients. All 12 patients with ANC ratio (peak value of ANC post-treatment/ base-line value of ANC pre-treatment)≥2 responded to the treatment (100%), as compared with 1 (25%) in 4 patients with ANC ratio < 2 (P<0.01).④12 patients developed G-CSF-induced graft-versus-host disease (GI-GVHD) with acute GVHD in 8 and chronic GVHD in 4. Median time of the onset of GI-GVHD was 30.5 days (range from 5 to 56 days) of initiation of G-CSF administration. The response rate in GI-GVHD group was 100%(12/12), as compared with 33.3%(2/6) in non GI-GVHD group (P=0.005), and CR rates were 83.3%(10/12) in GI-GVHD group and 16.7% (1/6) in non GI-GVHD group (P=0.013). All the 5 patients who received G-CSF therapy within day+100 post-transplant developed gradeⅡ～ⅣGI-aGVHD (100%) with involvement of gastrointestinal tract in 5 patients (100%), and 3 of 7 patients who received G-CSF therapy beyond day+100 post-transplant developed GI-aGVHD (42.9%) with grade I GI-aGVHD in 2 and involvement of gastrointestinal tract in 1 (14.3%). Severe GI-aGVHD contributed death in 4 out of 5 patients in whom G-CSF was given less than 100 days after Allo-HSCT (80%) in contrast to 0 out of 7 patients in whom G-CSF was given more than 100 days after Allo-HSCT (0%).⑤ The median follow-up was 161 days (range from 19 to 565 days).2 patients with CML in blast crisis had durable cytogenetic remission for 565 days and 426 days, respectively.2 patients were withdrawn from this trial and switched to donor lymphocyte infusion (DLI) or second transplant at a median time of 161 days and 179 days after G-CSF therapy, respectively.9 patients died from leukemia and 4 from severe GVHD. Granulomatous pneumonitis was responsible for death in the remainining one case. All 5 patients with extramedullary relapse (EMR) were died of recurrent leukemia (100%), as compared with 4 (30.8%) in 13 patients without EMR.Conclusion:This study suggests that G-CSF exerts its anti-leukemic activity through its ability to generate a graft-versus-leukemia effect, and combination of G-CSF and Ld Ara-C was highly effective in relapsing leukemia after allo-HSCT because of their significantly synergistic anti-leukemic activity. The occurrence of GI-GVHD and higher ANC ratio (≥2) were predictors for response. The most common and significant toxicity attributable to G-CSF regimen is GVHD. Initiation of G-CSF within day+100 post-transplant was predictive of developing severe acute GVHD. In terms of CR rate, combination of G-CSF and Ld Ara-C has efficacy in the treatment of relapsed leukemia following the transplantation is comparable with donor lymphocyte infusion (DLI), and therefore, it could be considered as an alternative to DLI. Although combination of G-CSF and Ld Ara-C attained a higher CR rate, the price to pay in terms of GVHD-related death was too high, and therefore, It warrants to be further improved with regard to the therapeutic strategy. PartⅡ:Experimental study on the impact of the administration of granulocyte colony-stimulating-factor after allogeneic bone marrow transplantation on graft-versus-host disease and graft-versus-leukemia effect in a murine modelObjective To establish murine models of nonlethal graft-versus-host disease (GVHD) and of post-transplant WEHI-3 myelomonocytic leukemia, and to explore the impact of the administration of granulocyte colony-stimulating-factor (G-CSF) after Allo-BMT on GVHD and graft-versus-leukemia (GVL) effect in the models and its possible mechanisms. The result from this murine study will provide the rationale for the clinical application of G-CSF for relapsing leukemia after allo-transplantation.Methods1. Male C57BL/6 (H-2b,8～10 weeks) and BALB/c (H-2d,8～10 weeks) mice were allogeneic and syngeneic donor mice, respectively, and female BALB/c (H-2d, 8～10 weeks) mice were recipient mice. Recipient mice were conditioned by a single dose of total body irradiation (TBI,8Gy,0.74 Gy/min). Transplantation was performed within 6h after TBI, by a single injection of 0.2mL of PBS containing 1×107 nucleated bone marrow cells (BMC) plus various numbers of splenocytes (SpC,0,2×106,5×106, and 1×107) through the tail vein of recipient mice. Mice of TBI control group received 0.2mL PBS. Flow cytometic (FCM) was used to assay the percentage of H2-Kb positive cells in bone marrow of two recipient mice at day+30 after allogeneic graft.2. The murine model of post-transplant WEHI-3 myelomonocytic leukemia were established by co-infusion of 1×106 WEHI-3 cells with 1×107 BMC+2.5×106 SpC from donor mice via tail vein.3. The study of the impact of post-transplant G-CSF on GVHD, recipient mice were divided into 6 groups:Syn-BMT control group (Cs, n=10), post-Syn-BMT G-CSF administration group (Gs, n=10), Allo-BMT 1 control group (single transplantation of 1×107 BMC, CM, n=10), post-Allo-BMT1 G-CSF administration group (GM, n=10), Allo-BMT2 control group (transplantation of 1×107 BMC+5×106 SpC, CA, n=22), post-Allo-BMT2 G-CSF administration group (GA, n=22). Mice of control groups and G-CSF administration groups were subcutaneous injection of 0.1mL normal saline (NS) and 0.1mL NS containing 2μg G-CSF per day from day+1, respectively. The levels of serum IL-2, IL-4, IFN-γand TNF-αof CA and GA group mice were detected by ELISA at day+10 (n=7). FCM was used to assay the immunophenotypes of splenocytes at day+10 (n=5).4. The study of the impact of G-CSF on GVL effect, BALB/c mice were also divided into 6 groups:non-transplant leukemia control group (C1, n=9) and its G-CSF treatment group (G1, n=9), post-Syn-BMT leukemia control group(C2, n=9)and its G-CSF treatment group (G2, n=9), post-Allo-BMT leukemia control group (C3, n=10) and its G-CSF treatment group (G3, n=10). Mice of control groups and G-CSF treatment groups were subcutaneous injection of 0.1mL normal saline and 0.1mL normal saline containing 2μg G-CSF per day from day+1, respectively. The therapeutic efficacy of G-CSF was evaluated by survival time of mice of different groups.Results1. Mice of TBI group were all died of hematopoietic failure with a mean survival of 12.0±1.9d. All mice in Syn-BMT group were 100% surviving without GVHD at day+60. The GVHD mortality of 9.1%,72.7%,100% and 100%, and a mean survival of 57.3±6.6d,46.9±9.5d,34.5±6.0d, and 16.6±4.1d were observed in the group of irradiated recipient mice who received C57BL/6 mice 1×107BMC, 1×107BMC+2×106 SpC, 1×107BMC+5×106 SpC, and 1×107BMC+1×107 SpC, respectively (P<0.05).99.8% and 99.4% of the percentage of H2-Kb positive cells were detected in bone marrow of two allogeneic recipient mice at day+30 respectively. The mean survival of post-Syn-BMT and post-Allo-BMT leukemic mice were 17.1±1.2d and 25.0±1.8d respectively (P=0.000).2. There was no impact of G-CSF on the survival of mice who underwent Syn-BMT and transplantation of single allogeneic marrow cells. The mean survival of 19.8±6.1d and 34.8±4.5d were observed in GA and CA group respectively (P=0.000). Moreover, post-transplant G-CSF can amplify the totol nucleated cells count, NK cells subset, and DC1/DC2 ratio in the spleen with over 99% of donor chimerism rate at day+10. There were no differences in the levels of serum IL-2, IL-4, IFN-γand TNF-αbetween CA and GA group mice at day+10.3. G-CSF treatment can not improve the survival of mice who were inoculated with WEHI-3 myelomonocytic leukemia and underwent Syn-BMT, but can prolong the the survival of leukemic mice received Allo-BMT (31.4±3.9 vs.25.0±1.8, P=0.000).Conclusion The administration of G-CSF after allo-BMT can induce GVHD and augment GVL effect in a murine model. The expansion of NK cells stimulated by G-CSF may be involved in the mechanism of generating alloreactivity against leukemic cells. The result from this study may provide the theoretical basis for a novel therapeutic strategy of G-CSF:1. G-CSF is exploited to treat relapsing leukemia following allo-BMT through inducton of GVL effect; 2. G-CSF can be used to prevent leukemia relapse following allo-BMT by enhanced GVL effect; 3. There may be potential risk of evoking or aggravating acuteGVHD if G-CSF is administered in the early stage of allo-BMT.
|
PDF Full Text Request