|Home | About | Journals | Submit | Contact Us | Français|
Consolidation with allogeneic hematopoietic stem cell transplant (allo-HSCT) provides a survival benefit to patients with acute lymphoblastic leukemia (ALL). We have previously reported comparable survival and relapse rates after T-cell depleted (TCD) allo-HSCT compared to unmodified transplants for acute myelogenous leukemia, myelodysplastic syndrome, and non-Hodgkin lymphoma with significantly decreased graft-versus-host disease (GVHD). We performed a 56 patient retrospective study to evaluate TCD allo-HSCT for the treatment of ALL following myeloablative TBI-based therapy. The 2-year and 5-year OS for patients with ALL after TCD-HSCT was 0.39 (95% CI: 0.26-0.52) and 0.32 (95% CI: 0.19-0.44), respectively. The 2-year and 5-year DFS was 0.38 (95% CI: 0.25-0.50) and 0.32 (95% CI: 0.20-0.44). There was a trend toward improved survival of patients who entered TCD allo-HSCT in first complete remission, compared with other remission states. The cumulative incidence of grade II-IV acute GVHD at one year was 0.20 (95% CI: 0.10-0.31), and no patients developed grade IV acute GVHD. The cumulative incidence of chronic GVHD in 41 evaluable patients at 2 and 5 years was 0.15 (95% CI: 0.04, 0.26), and that of extensive chronic GVHD at 2 and 5 years was 0.05 (95% CI: 0, 11.6). We demonstrate OS and DFS rates that compare favorably to unmodified allo-HSCT with lower rates of GVHD.
Approximately one-third of the nearly 4000 acute lymphoblastic leukemia (ALL) cases in the United States occur in adults (1). Although complete remission (CR) rates after induction and consolidation are nearly equivalent in children (> 90%) and adults (> 78%) (1), the overall cure rate is < 40% in adults compared with 80% in children (1). Patients with relapsed or refractory disease have a dismal prognosis with a median survival of six months (2-4). A number of studies have demonstrated the benefit of allogeneic hematopoietic stem cell transplantation (allo-HSCT) in patients with ALL in CR2 or high-risk CR1 (5-10). In other studies, however, a similar benefit was not observed (11, 12). A recent meta-analysis of seven published studies in ALL reported a significant advantage for sibling allogeneic transplant in high-risk patients compared with those lacking a sibling donor (13). More recent studies have also demonstrated the benefit of allo-HSCT in patients with standard risk ALL (14, 15).
We have established the efficacy of T-cell depleted (TCD) allo-HSCT as post-remission therapy in patients with AML, MDS or NHL, including an almost complete elimination of acute and chronic graft-versus-host disease (GVHD) without compromising the antileukemic efficacy of the allograft (16-19). We now report the results of a single-center retrospective review of 56 adult patients with ALL who underwent TCD allo-HSCT, and we demonstrate an acceptable overall survival (OS) and disease-free survival (DFS) with a low risk of GVHD.
Fifty-six consecutive adult patients with ALL underwent an allogeneic TCD-HSCT at Memorial Sloan-Kettering Cancer Center (MSKCC) from May 1997 through December 2008. Written informed consent for treatment was obtained from all patients and donors. Approval for this retrospective review was obtained from the Institutional Review and Privacy Board. Eligibility criteria for transplant included a diagnosis of relapsed or refractory ALL or high risk ALL in first remission; age less than 65 years; availability of an HLA-matched or single-allele mismatched donor; absence of active infection; and lack of coexisting cardiac, pulmonary, hepatic, or renal dysfunction that would preclude administration of the cytoreductive regimen. HLA matching was established by DNA sequence-specific oligonucleotide typing for HLA-A, -B, C, DR-B1 and DQ-B1 loci.
All patients received myeloablative cytoreduction comprising hyperfractionated total body irradiation (HFTBI), followed by thiotepa and high-dose cyclophosphamide (18 patients) or thiotepa and fludarabine (38 patients) (16, 17, 19). Patients received the fludarabine regimen on a protocol designed to study if the additional immune suppression due to fludarabine could eliminate the need for anti-thymocyte globulin (ATG) in TCD allo-HSCT using related donors (17). Otherwise, patients received a cyclophosphamide based regimen. HFTBI (total dose of 1375 cGy or 1500 cGy, with lung blocking plus chest wall and testicular boosts) was administered as previously described (16). After HFTBI, thiotepa 5mg/kg IV was administered daily for two days. In the 18 patients who received the first regimen, administration of thiotepa was followed by high-dose cyclophosphamide 60mg/kg IV daily for two days (16). In the other 38 patients, fludarabine 25 mg/m2 IV was administered daily for five days beginning on the first day of thiotepa (17).
T cells were removed from bone marrow (BM) grafts by sequential soybean lectin agglutination and sheep red blood cell (sRBC)-rosette depletion (16, 20). T cell depletion of granulocyte colony stimulating factor (G-CSF)-mobilized peripheral blood stem cells (PBSC) was accomplished by positive selection of CD34+ stem cells using the ISOLEX 300i Magnetic Cell Separator and subsequent sRBC-rosette depletion (17). T-cell depleted marrow or PBSC was infused within 24-48 hours after completion of the chemotherapy. Seventeen patients received BM and 38 received PBSC transplants, all TCD. One patient received a combined TCD BM and PBSC allograft.
Recipients of an HLA-matched related donor graft who were treated with HFTBI, thiotepa, and fludarabine (n=11), and two patients under the age of 20 did not receive any rejection prophylaxis. Equine (30 mg/kg/dose) or rabbit (5 mg/kg/dose) ATG prior to (n=30) or post (n = 13) stem cell infusion provided graft rejection prophylaxis in all other patients (16, 17, 21). Patients who received 2 daily doses of ATG prior to stem cell infusion were given high-dose methylprednisolone (1 mg/kg/day) with each dose. The patients who received 4 doses of ATG post stem cell infusion received methylprednisolone with each dose of ATG and steroids were rapidly tapered off. All patients received supportive care and prophylaxis against opportunistic infections according to standard guidelines. No pharmacologic GVHD prophylaxis was given.
Myeloid engraftment was defined as an absolute neutrophil count (ANC) > 500μl on 3 consecutive days post-transplant. Platelet engraftment was defined as an untransfused platelet count > 20,000/μl for at least 3 consecutive days. Primary graft failure was defined as the absence of neutrophil recovery (≥ 500/μl) by day 28 and BM biopsy with ≤ 5% cellularity. Secondary graft failure was defined as loss of ANC to < 500/μl after primary engraftment with BM biopsy showing ≤ 5% cellularity (18). GVHD was diagnosed clinically, confirmed pathologically by biopsy whenever possible, and classified according to standard criteria (22). Patients who engrafted were evaluable for acute GVHD, and patients surviving at least 100 days were evaluable for chronic GVHD (23). Cause of death was determined using a standard algorithm (24).
Analyses were performed as of 12/31/2011. The OS and DFS probabilities were calculated using the Kaplan-Meier method, and the differences between levels within a covariate were tested using the log rank statistic (25). The cumulative incidence of relapse was calculated using competing risk methods. The following pre-transplant variables were assessed for their effects on OS and DFS: disease status, cytogenetic risk stratification (high, standard, low) (26), presence of extramedullary disease, stem cell source, HLA match, and use of a cyclophosphamide versus fludarabine-containing preparative regimen. Univariate analyses were performed using the log-rank test. Because only one association between pre-transplant variables and survival endpoints reached statistical significance in the univariate analysis, no multivariate analysis was performed.
Pre-transplant characteristics of the 56 patients are listed in Table 1. The median age was 36 years with 11 patients older than 50. Twenty-nine patients (52%) were transplanted in CR > 1 with 11 patients (20%) in CR3 or greater. Forty-three percent of the patients had poor risk cytogenetics. A significant proportion of patients received alternative donor grafts with 28 patients (50%) receiving grafts from unrelated donors (13 mismatched) and an additional 6 patients receiving grafts from mismatched related donors. Patients were considered to be mismatched if they did not match at 10/10 alleles.
All 54 patients evaluable for engraftment achieved initial engraftment. However, 2 patients who received a mismatched graft from an unrelated donor experienced secondary graft failure at day + 35 and +67, respectively. Both patients eventually died with multi-organ failure. The median time to neutrophil engraftment was 13 days for the entire group, and 15 days and 12 days for recipients of BM grafts and PBSC grafts, respectively. The median time to platelet engraftment for the cohort was 14 days. It was 25 days and 13 days for recipients of BM grafts and PBSC grafts, respectively. Four patients (2 BM, 2 PBSC) did not achieve platelet engraftment. Data on engraftment times were not available for 2 patients (BM) and 2 patients (PBSCs) were not evaluable for engraftment due to early death.
The cumulative incidence of grade II-IV aGVHD at one year in 54 evaluable patients was 0.20 (95% CI: 0.10-0.31, figure 1a). No patient developed grade IV acute GVHD. The organs involved included skin only (7 patients), skin and GI tract (2 patients), GI tract only (1 patient), and skin and liver (1 patient). Two of the patients with aGVHD received BM or a PBSC graft from a matched related donor (MRD), and the other 9 from alternative donors (4 matched unrelated donors and 5 mismatched related or unrelated donors). The cumulative incidence of chronic GVHD in 41 evaluable patients at 2 and 5 years was 0.15 (95% CI: 0.04, 0.26; figure 1b), and that of extensive chronic GVHD at 2 and 5 years was 0.05 (95% CI: 0, 11.6).
Four patients received unselected DLI following their initial transplant, and 1 patient received EBV CTLs. DLI was administered for infection (n = 2), minimal residual disease (n = 1), and mixed BM chimerism (n = 1). The latter patient subsequently developed chronic GVHD and died.
Thirty patients were documented to have CMV positive serology. Of the 28 patients evaluable for viral reactivation (2 premature deaths) 9 (32%) reactivated CMV. One of these patients died from CMV pulmonary disease. Of 40 patients who received surveillance with EBV PCRs, 7 patients (18%) were noted to have PCR evidence of EBV reactivation. Four of these patients required therapy with rituximab. One further patient, who was transplanted prior to routine EBV PCR surveillance testing, was noted to have an EBV lymphoproliferative disorder on a lung biopsy and received DLI and rituximab.
The causes of death included infection (10 patients), GVHD (8 patients), organ failure (5 patients), and non-engraftment (2 patients). Both patients with non-engraftment had evidence of reemergence of host cells in bone marrow evaluations. In the patients who died of infection, 5 patients had a documented bacterial infection (VRE = 4, C. difficile colitis = 1), and 3 patients had a documented viral infection (1 patient each with CMV, EBV and adenovirus). Three of the 4 patients who died with VRE infection were noted to have VRE colonization prior to transplantation. The remaining patient was transplanted prior to instituting routine VRE surveillance practices. We have previously described the virulence of VRE in unmodified and TCD allo-HSCT (27). One patient each was diagnosed with a fungal infection (Aspergillus flavus) and toxoplasmosis. Of the 5 patients who died of organ failure, 4 died from pulmonary toxicity and 1 died from veno-occlusive disease. Each of the 4 patients who died from pulmonary toxicity suffered progressive pulmonary failure that culminated in ARDS. Seven of the 8 patients classified as dying from GVHD died from infections in the setting of immune suppressive treatment (1 bacterial, 3 fungal, 2 viral, 1 toxoplasmosis). The eighth patient had biopsy-proven GVHD and died of concurrent demyelination. He was, however, classified as dying from GVHD based on the algorithm of Copelan et al (24).
The non-relapse mortality (NRM) at 2 years for patients who received allo-HSCT in ≥ CR3 (0.63) was significantly higher than patients in CR1 or CR2 (0.33, p =0.030) with 8/11 patients transplanted in ≥ CR3 dying from transplant-related causes. This resulted in an NRM rate at 2 years of 0.39 for the entire group. Because the rate of death of patients from GVHD-related causes was different than our previous experiences with TCD allo-HSCT (16-19), we evaluated the GVHD-related mortality and NRM not related to GVHD by disease status. The 2 year NRM without GVHD was related to disease status (CR1= 0.19, CR2= 0.28, CR3+= 0.55, p=0.089), while the GVHD-related mortality was not (p=0.569).
The cumulative incidence of relapse for the entire group was 0.23. The cumulative incidence of relapse was not affected by disease status (p=0.522) or cytogenetic risk group (p= 0.408). With a median follow-up of 6.1 years, the 2-year and 5-year DFS for the entire cohort of patients was 0.38 (95% CI: 0.25-0.50) and 0.32 (95% CI: 0.20-0.44), respectively (Figure 2a). The 2-year and 5-year OS was 0.39 (95% CI: 0.26-0.52) and 0.32 (95% CI: 0.19-0.44), respectively (figure 2b). Disease-free and overall survival at 2 and 5 years by disease status is described in table 2. Patients in CR1 had a higher DFS (p = 0.062, figure 2c) and OS (p = 0.056, figure 2d) than patients in CR2+. When pre-transplant prognostic factors were analyzed, an improved median OS was observed in patients with a MRD compared with other donors (p = 0.041). A trend for improved DFS was also noted (p = 0.051). No difference was seen with either DFS or OS based on cytogenetic risk group, the presence of extramedullary disease, or the age at transplant. No differences were observed in DFS or OS when we compared outcomes for the two different regimens (TBI-thiotepa-cyclophosphamide vs. TBI-thiotepa-fludarabine) or for graft source, as previously observed in patients with NHL treated with the same approach (19).
Our results represent the first published experience describing ex-vivo TCD allo-HSCT for the treatment of ALL in adults without post-transplant pharmacologic GVHD prophylaxis. Recognizing the limitations of a retrospective study that encompasses many years, we are nevertheless able to demonstrate an acceptable OS, DFS, and relapse-rate with low rates of GVHD in the context of the historical experience with T-cell replete allo-HSCT (5-10). Notably, our results for patients in CR1 are promising with a plateau in the OS and DFS curves for these patients at approximately 2 years.
Fifty-six adult patients underwent a TCD allo-HSCT for the treatment of ALL. This patient population was particularly high risk, with only 7 patients < 20 years old and 24 patients with poor risk cytogenetic classifications (43%). Slightly more than half of the patients were transplanted in > CR1 (29 patients, 52%). Sixteen (29%) patients had extramedullary disease at presentation. For the entire cohort, the 2 year OS and DFS were comparable at 0.39 and 0.38 to published experiences with T cell replete transplants in this high risk patient population. For patients in CR1, OS and DFS were 0.48 at both 2 years and 5 years. In the absence of post-transplant pharmacologic GVHD prophylaxis, the cumulative incidence of aGVHD at 1 year was only 0.20 (without any incidences of grade IV aGVHD), consistent with our previous data in TCD allo-HSCT, and despite the fact that 34% of the patients received grafts from mismatched donors. Only 15% of patients were diagnosed with cGVHD at 2 years.
For the whole group, the 2 year NRM of 0.39 was higher than our previously published experiences with TCD allo-HSCT in other diseases (16-18), with infection being the most prevalent non-relapse cause of death. Five of the 10 deaths attributed to infection resulted from early bacterial deaths (range d+5 to d+66), of which 4 cases were due to VRE bacteremia. It is thus difficult to attribute these deaths to TCD of the graft. Furthermore, we have previously demonstrated that the risk of VRE blood stream infections is not increased by TCD (27). Furthermore, the increase in NRM with disease status at allo-HSCT suggests that the heavy pre-treatment of the patients, especially those with advanced disease, was a likely contributor to NRM. In addition, while OS and DFS were associated with remission status, the cumulative incidence of relapse was not, further suggesting that the degree of pre-treatment may contribute to post-HSCT outcomes for these patients. It is also striking that there were 8 deaths resulting from GVHD in the context of a low prevalence of GVHD, an expected distribution of involved organs, and no incidence of grade IV GVHD. This high mortality attributable to GVHD also contrasts with our previous experiences with TCD (16-18). It is possible that the significant treatment prior to transplant of our cohort (and ALL patients in general), including extensive exposure to glucocorticoids, resulted in patients having increased susceptibility to complications resulting from GVHD and its immune suppressive treatments. It is notable that the GVHD-related mortality does not appear to be related to disease status. This finding may be related to the small number of patients who suffered GVHD following TCD allo-HSCT or it may imply that the common use of steroids or some other pre-transplant characteristic rather than the amount of pre-treatment increase the risks of complications associated with GVHD and its therapy for ALL patients.
There are interventions that could potentially decrease the NRM for ALL patients undergoing TCD allo-HSCT. First, acknowledging the small size of our CR3+ cohort, one could offer a reduced intensity approach for patients with ALL > CR2, given the high rate of NRM in these patients with the intense regimens required for TCD. In addition, preemptive use of antibiotics active against VRE for ALL patients who are colonized could be considered. Alternative non-transplant options may also need to be considered for patients > CR2.
We compared our experience with TCD allo-HSCT with other published experiences of T-cell replete allo-HSCT for ALL. Goldstone et al recently published the largest randomized experience in adult ALL in the MRC UKALL XII/ECOG E2993 trial (14). A donor versus no donor analysis of Philadelphia chromosome negative (Ph-) patients in CR1 revealed that patients with a donor had a 5 year OS of 53%. Thomas et al also studied the optimal post induction therapy for patients in CR1 in the LALA-94 trial (8). Patients who were high risk were assigned to receive allo-HSCT if an HLA-identical sibling was identified. For high risk Ph- patients who received allo-HSCT (without central nervous system [CNS] disease), the 5 year DFS rate was 44% with a 5 year OS of 51% and for Ph+ patients the 3 year OS and DFS were 36% and 34%, respectively. Our 5 year OS of 48% in this study compares quite favorably to these findings, especially given that 16 (59%) of our patients in CR1 were Ph+.
Outcome data for patients who receive an allo-HSCT in > CR1 are more limited. However, it appears survival outcomes after TCD allo-HSCT are also comparable to T-cell replete strategies in this patient population. Fielding et al published an analysis of the patients who relapsed after their initial therapy in the MRC UKALL XII/ECOG E2993 trial (2). For patients who received an allo-HSCT in the relapsed setting, the 5 year OS was 23% for recipients of matched related donor transplants and 16% for recipients of matched unrelated donor transplants.
We have previously published that select patient populations can have favorable survival after TCD transplant with a decreased risk of GVHD in the absence of post allo-HSCT pharmacologic prophylaxis (16-19). A recent, cooperative group-led multicenter trial of ex vivo TCD for patients with AML in CR1 or CR2 supports this center's experience (28). The risk of aGVHD grades II-IV in the multicenter trial was comparable at 22.7%, with an incidence of extensive cGVHD of only 6.8% at 24 months. Thus, TCD offers appropriate patients comparable survival with the decreased morbidity associated with GVHD.
Our findings should be confirmed prospectively as this study has some limitations common to retrospective studies. We cannot rule out a selection bias. The study encompasses many years where many clinical changes have occurred. Also, the small sample size of our study limits our conclusions.
In summary, this first published experience of adult patients who received a TCD allo-HSCT for ALL demonstrates a survival and relapse rate that compares favorably with T-cell replete transplants, but importantly, with a decreased rate of GVHD. The survival outcomes for patients in CR1 are promising with a plateau noted at approximately 2 years. Further studies are needed to identify subpopulations of patients with ALL who would benefit most from a TCD rather than a T-cell replete strategy. Patients with an advanced disease status may have improved outcomes with a reduced intensity strategy. In addition, studies should assess if the addition of post transplant maintenance therapy may improve outcomes after TCD allo-HSCT in certain patient populations. Finally, minimal residual disease (MRD) monitoring should be incorporated into future studies to assess if MRD impacts on clinical outcomes following TCD allo-HSCT.
We gratefully acknowledge the expert care provided to these patients by the fellows, housestaff, and nurses of Memorial Sloan-Kettering Cancer Center.
Financial disclosure: Supported in part by NIH P01 CA23766. Additional support was also received from When Everyone Survives, Cycle for Survival, the New York Community Trust, and the Experimental Therapeutics Center of Memorial Sloan-Kettering Cancer Center funded by Mr. William H. and Mrs. Alice Goodwin and the Commonwealth Foundation for Cancer Research (MAP).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.