Effects of post-transplant maintenance therapy with decitabine prophylaxis on the relapse for acute lymphoblastic leukemia

In adults with acute lymphoblastic leukemia (ALL), post-transplant relapse is a major risk factor for mortality after allogeneic hematopoietic stem cell transplantation (allo-HSCT). Our study investigated the efficacy and safety of decitabine (dec) with ALL patients post-transplantation. We performed a retrospective cohort study to assess the efficacy of decitabine (dec) with post-transplant ALL at the First Affiliated Hospital of Zhengzhou University from February 2016 to September 2021. A total of 141 consecutive ALL patients were analyzed and divided into decitabine (dec, n = 65) and control (ctrl, n = 76) groups based on whether they were treated with decitabine after allo-HSCT. The 3-year cumulative incidence of relapse (CIR) rate in the dec group was lower than that in the ctrl group (19.6 vs. 36.1%, p = 0.031), with a hazard ratio of 0.491 (95% confidence interval [CI], 0.257–0.936). Additionally, subgroup analyses revealed that the 3-year CIR rate of T-ALL and Ph-negative B-ALL patients in the dec and ctrl groups was 11.7 vs. 35.9% and 19.5 vs. 42.2% (p = 0.035, p = 0.068) respectively. In summary, ALL patients, especially those with T-ALL and Ph-negative B-ALL, may benefit from decitabine as maintenance therapy following allo-HSCT.


INTRODUCTION
Adult acute lymphoblastic leukemia (ALL) is an aggressive malignancy originating from B or T lymphocytes, and it accounts for~15-20% of all acute leukemias. In spite of the fact that patients with ALL have greatly benefited from allogeneic hematopoietic stem cell transplantation (allo-HSCT), relapse remains the most common cause of death in patients after allo-HSCT. Among patients who relapsed after transplantation, the 1-year and 2-year overall survival (OS) rates were only 17 and 10%, respectively [1]. Donor lymphocyte infusion (DLI) is currently the most common treatment for post-transplantation relapse. However, the outcome after DLI remains poor regardless of the disease type and donor source [2]. Similarly, patients with refractory or relapsed B acute lymphoblastic leukemia showed a 70-90% complete remission (CR) rate in clinical trials utilizing CD19 or CD22 chimeric antigen receptor (CAR) T-cell treatment; however, a substantial number of these patients experienced relapses within 1 year [3]. Therefore, the prevention of leukemia relapse after allo-HSCT remains a high priority in patients with ALL.
Previous studies have demonstrated that acute myeloid leukemia (AML) relapse after transplantation was associated with the dysregulation of pathways that may influence immune function, including the downregulation of major histocompatibility complex(MHC) class II genes by epigenetic silencing, which allows leukemic cells to escape the graft-versus-leukemia effect, and MHC class II gene expression is downregulated as a result of the hypermethylation of the class II transactivator (CIITA) [4][5][6].
In an in vivo human T-cell leukemia study, CIITA hypermethylation resulted in a lack of MHC class II expression on leukemic T cells, and the treatment of leukemic T-cell lines with demethylating agents led to the re-expression of CIITA and HLA-DR, which was expressed in only 5-17% of all T-ALL [7]. Similarly, the pattern of DNA methylation in B-ALL has been extensively studied in relation to survival, and the methylation levels in malignant cells were higher than those in normal hematopoietic B-cell precursors and in remission samples from the same patients in several studies [8][9][10].
The purpose of this single-center retrospective cohort study was to evaluate whether decitabine reduces relapse rates and improves long-term survival in patients with ALL receiving allo-HSCT.

MATERIALS AND METHODS Patients
One hundred forty-one consecutive patients with ALL who underwent allo-HSCT at the First Affiliated Hospital of Zhengzhou University between January 2016 and September 2021 were included. The end date of the follow-up was September 30, 2022. This time frame was chosen to provide a minimum follow-up of 12 months for all living patients. The diagnostic criteria for ALL were based on the WHO 2016 guidelines [11].
This study was approved by the ethics committee of the First Affiliated Hospital of Zhengzhou University, and patients gave their written informed consent prior to treatment in accordance with the Helsinki declaration.

Conditioning regimen
All patients received either a modified Busulfan (Bu)/Cyclophosphamide (Cy) or a total body irradiation (TBI)/Cy conditioning regimen. The modified Bu/Cy was provided as follows: for the HLA-haploidentical donors (HID) population, cytarabine (Ara-c) was administered intravenously at 4 g/m 2 / per day on days −10 and −9, Bu (3.2 mg/kg/day on days −8 to −6), Cy (1.8 g/m 2 /day on days −5 to −4), and semustine (250 mg/m 2 /day on day −3). For the HLA-unrelated donors (URD) and HLA-identical sibling donor (ISD) population, Ara-c was administered at 2 g/m 2 /per day on days −10 and −9. Correspondingly, TBI/Cy was provided as follows: a total of 9 Gy of TBI on days −8 and −7 and a total dose 3.6 g/m 2 of Cy on days −4 and −3.

Decitabine maintenance therapy
Decitabine maintenance therapy was defined as the administration of decitabine more than 30 days after transplantation for prevention of relapse. Generally, if patients had hematopoietic reconstitution and no active GVHD, decitabine was started between 30 and 180 days after transplantation. The dose of decitabine was 10 mg/per day for consecutive 3 days approximately once a month for up to 8 cycles. If the patient's MRD test turned positive during the final phase of maintenance therapy, the number of cycles was increased based on the requests of physicians and patients, and each cycle's interval duration was also adjusted on the basis of hematological toxicity. Once patients who received maintenance therapy relapsed or developed serious graft-versus-host disease (grade III-IV aGVHD) after transplantation, decitabine was discontinued immediately.

Parameter definitions
Assessment of risk classification and stratification with ALL based on GRAALL-2005 Regimen and NCCN guidelines [12,13]. Complete remission (CR) was defined as less than 5% bone marrow (BM) blasts, the absence of circulating blasts, and no active extramedullary sites of disease. The data of minimal residual disease (MRD) was assessed in all patients undergoing a subsequent bone marrow examination for treatment response analysis by either multiparametric flow cytometry (FCM) with a sensitivity of 10 −4 (1 malignant cell/10,000 normal cells) or real-time quantitative polymerase chain reaction (QT-PCR). MRD negativity was defined as absence of leukemia cells in BM as determined by FCM, and/or the absence of leukemia-associated fusion gene in BM determined by QT-PCR. Samples for MRD were obtained at the time of achieving CR (~day 21-30 of the induction therapy) and subsequently at 1-month intervals during the course of consolidation and maintenance therapy prior to allo-HSCT, 30 days following allo-HSCT, and at 2-month intervals during the first year after transplantation, and then every 3-6 months as feasible. In accordance with common morphological criteria, leukemia relapse was classified as the reappearance of blasts in the blood or bone marrow or in an extramedullary site after a complete remission. Acute graft versus host disease (aGVHD) and chronic graft versus host disease (cGVHD) were evaluated with reference to standard consensus criteria [14,15]. Grading of hematological toxicity during maintenance therapy was based on the National Cancer Institute Common Toxicity Criteria, version 4.0.

Statistical analysis
Categorical variables were compared using χ 2 or Fisher's exact test and continuous variables were compared using Mann-Whitney tests. The competing risk model (Fine and Gray model) was used to calculate the CIR, aGVHD, cGVHD and Transplantation-related mortality (TRM). Disease-free survival (DFS) and overall survival (OS) rates were determined by Thirty-one T-ALL, including 2 ETP-ALL.
Kaplan-Meier method and were compared by using the log-rank test and Cox proportional hazard modeling. p values < 0.05 were considered statistically significant. Statistical analyses were performed using SPSS (version 26.0) and R software package (version 4.2.0).

Patients and characteristics
One hundred forty-one consecutive patients with ALL undergoing allo-HSCT were enrolled in this study. The patient characteristics are provided in Table 1. A total of 103 (73.0%) patients were males and 38 (27.0%) were females. The median age for the whole population was 23 (range, 14-55) years. In total, 32 (22.7%) cases were classified as standard risk and 109 (78.6%) as high risk. 88 (62.4%) patients were B-ALL while 53 (37.6%) patients were T-ALL. In total, 32 (22.7%) patients were Philadelphia chromosomepositive (Ph-positive), and 109 (77.3%) patients were Philadelphia chromosome-negative (Ph-negative or Ph−). All patients received the induction chemotherapy scheme including VD(CL)P regimen (vincristine, daunorubicin, cyclophosphamide, L-asparaginase or asparaginase, prednisone; n = 121) or Hyper-CVAD protocol (cyclophosphamide, vincristine, doxorubicin and dexamethasone alternating with methotrexate and cytarabine; n = 20). Thirteen (9.2%) patients who failed to achieve CR received a reinduction regimen (VDCP-based regimen or augmented Hyper-CVAD protocol). After achieving CR, 136 (96.5%) patients received consolidation therapy consisting of high-dose methotrexate, high-dose cytarabine, L-asparaginase or asparaginase. Among them, 93 patients received 1-4 courses of consolidation therapy and then the remaining continued to receive chemotherapy until they received an allotransplant. Maintenance therapy included 6-mercaptopurine and methotrexate. Thirty-two patients with BCR-ABL fusions were treated in the same manner with L-asparaginase or asparaginase omitted and a TKI (imatinib, 400 mg/day, n = 17; dasatinib, 100 mg/day, n = 12; flumatinib, 600 mg/day, n = 3) added from diagnosis or combination with chemotherapy. All patients received at least one intrathecal injection with intrathecal methotrexate, cytarabine and dexamethasone for central nervous system (CNS) prophylaxis. All patients were treated with myeloablative protocol, including 111 (78.7%) patients receiving mBu/Cy conditioning regimen and 30 (21.3%) patients receiving TBI/Cy conditioning regimen. Expect for 3 patients in control group failed to achieve hematopoietic reconstitution, the remaining were engrafted successfully. There were no significant differences in baseline characteristics between two cohorts except for the conditioning regimen and transplant donor.

CIR, TRM, OS, and DFS
The estimated 3-year cumulative incidence rate of relapse in the dec group was 19.6% (95% CI, 10.0-29.3%), compared with 36.1% (95% CI, 25.3-46.9%; p = 0.031) in the ctrl group (Fig. 2). Relapse occurred in 13 (20.0%) of 65 patients in dec group with a median time of 7.6 months (range, 4.8-53.6 months), including 10 patients who relapsed during maintenance therapy and 3 patients who relapsed after the end of maintenance therapy. There were eight cases of hematological relapse, three cases of extramedullary relapse and two cases of hematological and extramedullary relapse simultaneously. Contrastingly, 27 (35.5%) patients in the ctrl group relapsed at a median duration of 6.2 months (range, 1.9-28.3 months), including 21 cases of hematological relapse, 5 cases of extramedullary relapse and 1 case of both hematological and extramedullary relapse.

Subgroup analysis
In addition, we conducted subgroup analyses based on different disease subtypes. Among the patients with T-ALL, two patients (n = 2/22) relapsed in the dec group, and the relapse time was 5 months and 29 months, respectively. In contrast, 11 patients (n = 11/31) in the control group relapsed, with a median time to relapse of 7.6 months (range, 1.9-28.3 months). The 3-year cumulative incidence of relapse in the dec group was lower than in the ctrl group (11.7% [95% CI: 0-25.2%] vs. 35.9% [95% CI, 19.1-52.8%]; p = 0.035). The 3-year DFS rates in the dec and ctrl groups were 79.2% (95% CI, 62.4-100%) and 41.5% (95% CI, 27.2-63.2%; p < 0.01), the 3-year OS rates in the dec and ctrl groups were 86.1% (95% CI, 72.7-100%) and 47.8% (95% CI, 33.0-69.3%; p < 0.01). Similarly, we took into account the effect of oral TKI factors in Ph-positive patients after transplantation in the analysis of B-ALL, therefore, only Ph-negative B-ALL patients were analyzed. There were 7 patients (n = 7/32) in the dec group who relapsed with a median time of 16.7 months (range, 4.8-53.6 months), compared to 10 patients (n = 10/24) in the  Leukopenia was the most common hematological adverse event during maintenance therapy. Forty-one patients (grade I-II, n = 36, grade III-IV, n = 5) with leukopenia but almost all patients' white blood cell counts could be restored with the prompt administration of G-CSF. No patients suffered severe complications during the period of myelosuppression.

Univariate and multivariate analyses
The competing risk model (Fine and Gray model) was used to determine hazard ratios with 95% CI of relapse, and the Cox regression model was used to screen variables affecting OS and DFS. The results are shown in Table 2. We found that patients receiving maintenance therapy reduced relapse with a hazard ratio of 0.491 (95% CI, 0.257-0.936), improved OS and DFS after adjustment for possible confounders (p = 0.019, p < 0.01, p < 0.01), as well as MRD status before transplantation (p < 0.01, p = 0.027, p = 0.030). Moreover, a significant correlation was found between III-IV aGVHD and inferior OS (p = 0.011), whereas patients with limited cGVHD showed a higher OS and DFS (p = 0.017, p < 0.01).

DISCUSSION
In the setting of B-ALL, there is an abnormal hypermethylation pattern within the promoter regions of certain tumor suppressor genes, which inhibits the apoptotic capacity of leukemic cells [16][17][18][19]. Also, in adult T-ALL, preleukemic clones may eventually evolve toward complete transformation in a particular lineage with changes in DNA methylation pattern targeting a common lymphoid progenitor [20,21]. As one of the first trials to investigate the safety and clinical activity of decitabine in ALL, Benton et al. conducted a two-part phase 1 study to administer decitabine with or without Hyper-CVAD in relapsed/refractory ALL. They noted some patients who had disease progression to prior Hyper-CVAD alone achieved a complete response with the addition of dec and a substantial number of patients obtained meaningful responses [22]. As a result, we have attempted to apply the advantages of dec to patients with ALL.
To date, this was the largest cohort study to our knowledge of decitabine for prevention of post-transplant relapse with acute lymphoblastic leukemia. In this study, we demonstrated that decitabine as a maintenance therapy option after allo-HSCT could effectively reduce relapse and improve OS and DFS for adult ALL patients when compared with a contemporaneous control cohort.
T-cell lineage has previously been considered a high-risk feature in ALL and T-ALL is frequently linked to a poor response, the clinical prognosis of primary resistant leukemia is incredibly bad with overall survival rates of~20%. Abnormal DNA methylation is more common in adult ALL and TP53 hypermethylation, but not TP53 mutation, is also frequently observed in ALL [23]. Zhang et al. found that decitabine could inhibit cell proliferation, induce cell cycle arrest in G2 phase and induce apoptosis in molt4 cells which were stimulated in vitro, moreover, decitabine induced mitochondrial damage and lipid droplets production on subcellular level [24]. Although currently there is some evidence that T-ALL can be demethylated by hypomethylating agents (HMAs) in preclinical research, studies supporting the use of HMAs as monotherapy in clinical practices are limited. The majority of these clinical cases are often combined with additional treatment in relapsed/ refractory T-ALL or ETP-ALL [25][26][27][28][29]. In our study, dec treatment showed a significant effect in terms of preventing T-ALL recurrence, as the 3-year CIR rate in the dec group was 11.7%, while in the ctrl group it was 35.9%. Only two patients in the dec group experienced relapse, including one patient with ETP-ALL. Although the ETP-ALL patient experienced extramedullary recurrence after the end of maintenance treatment, the DFS was more than 28 months, which was encouraging.
In this retrospective study with B-ALL, the 3 years of CIR rate in dec group was 23.7% (95% CI, 11.0-36.4%), compared with 36.0% (95% CI, 22.0-51.0%; p = 0.19) in ctrl group. Four patients relapses occurred in TKI plus decitabine group (n = 4/11) and six patients relapses occurred in control group (n = 6/21). The results showed that decitabine combined with TKI maintenance therapy did not demonstrate an advantage over TKI maintenance therapy alone among Ph+ population. Therefore, timely quantification of BCR-ABL and adjustment of TKI based on results may be more meaningful than decitabine [30]. After excluding the contribution   [31]. Although we were unable to determine the specific expression of Hck between the two cohort, our results also laterally confirm the clinical activity of decitabine in Phnegative ALL. Several clinical studies have previously reported the prognostic relevance of MRD for predicting relapse risk and survival in adult ALL [32][33][34][35][36]. The monitoring of MRD has not only been shown to directly correlate with clinical outcome but also provided meaningful information for treatment decisions. In our univariate and multivariate analysis, we demonstrated that the risk of relapse was nearly three times higher in patients with pre-transplant MRD+ than in those with pre-transplant MRD− (HR 3. 20   patients before transplant was three times as high as that of MRD− patients, which is similar to our findings [37]. Our observations, which are in accordance with the bulk of the MRD literature, have exhibited that achieving CR, MRD-prior to transplant predicted a better clinical outcome among both dec and ctrl groups. Gao et al. revealed that patients with HR-AML after allo-HSCT did not experience an increase in the incidence of cGVHD when receiving G-CSF combined with the dec maintenance therapy (rhG-GSF 100 ug/m 2 on days 0-5, dec at 5 mg/m 2 on days 1-5). Following the second cycle of therapy, the lymphocyte subset analysis showed a substantial increase in the proportion of Treg cells, suppressing GVHD without decreasing the GVL effect [38,39]. The same clinical results were obtained in our study, with no obvious differences between two cohorts in either the incidence of aGVHD or cGVHD. Also, we found minimal-dose decitabine maintenance therapy was well tolerated, which showed less hematological toxicity because the drug dose and the number of administration days were reduced in comparison with above-mentioned studies.
The retrospective nature of our study introduces typical limitations. It is possible for additional imbalances to exist even after adjusting for baseline differences including transplant donor and conditioning regimen in the multivariate analysis. Additionally, cases in subgroup were still not sufficient to reflect stable differences. Furthermore, the median age of the overall patients was 23 years, meaning that the population was more comprised of adolescent and young adult patients than adult patients, therefore, larger multicenter studies are needed to further validate our findings.
Despite these limitations, our data exhibit that decitabine maintenance therapy after allo-HSCT may serve as a new strategy to prevent leukemia relapse. A larger-scale, prospective clinical trial with longer follow-up is required to determinate the efficacy and safety of dec maintenance therapy.

DATA AVAILABILITY
The data that support the findings of this study are available from the corresponding author upon request.