Incidence and Risk Factors for Central Nervous System Relapse after Allogeneic Hematopoietic Stem Cell Transplantation in Adult Patients with Acute Lymphoblastic Leukemia

Background Acute lymphoblastic leukemia (ALL) is frequently accompanied with the involvement of the central nervous system (CNS). The clinical prognosis for ALL patients with CNS disease is generally unfavorable, and prophylactic measures such as intrathecal chemotherapy are highly recommended. In adult ALL patients, hematopoietic stem cell transplantation (HSCT) presents a potential curative approach. However, there are currently only a few reports available on the risk factors and epidemiology of CNS relapse after HSCT.

Method We conducted a retrospective single-center cohort study involving 1043 patients diagnosed with either acute lymphoblastic leukemia (ALL) or mixed phenotype acute leukemia (MPAL). These patients were treated at our center using the same treatment protocol between 2009 and 2022. During the remission induction phase, we utilized a modified hyper-CVAD regimen, along with six rounds of prophylactic intrathecal (IT) chemotherapy. High-risk patients who had suitable donors underwent allogeneic hematopoietic stem cell transplantation (HSCT) as part of their post-remission therapy. Diagnosis of central nervous system (CNS) leukemia was established when neurologic symptoms were observed in correlation with CNS lesions on imaging studies, or when leukemic blasts were detected in the cerebrospinal fluid (CSF). In our study, we examined the incidence, risk factors, and survival outcomes of patients who experienced CNS relapse after transplantation.

Results Among the entire cohort of 1043 patients, CNS involvement was observed in 3 (0.3%) patients at initial diagnosis and 12 (1.2%) during chemotherapy. We proceeded with allogeneic hematopoietic stem cell transplantation (allo-HSCT) in 755 consecutive patients. Among these 755 patients, 3 had CNS leukemia prior to transplantation. Among the group of 755 patients, a total of 224 (29.6%) experienced relapse. Of these relapses, 142 (18.8%) were bone marrow relapses alone, 54 (7.1%) were extramedullary relapses alone (with CNS involvement in 21 [38.8%]), and 28 (3.7%) had both bone marrow and extramedullary relapses (with CNS involvement in 13 [46.4%]). Overall, extramedullary relapse was observed in 82 (10.8%) patients, with CNS involvement in 34 (4.5%) cases, following allo-HSCT. The 5-year cumulative incidence of CNS relapse was 4.6%, which was significantly higher in patients with Philadelphia chromosome-positive (Ph-positive) ALL (9.4% vs. 1.0%, p <0.001), as well as those with a high leukocyte count at diagnosis (8.3% vs. 2.6%, p = 0.001). Among the Ph-positive ALL cases, a high leukocyte count (n=151) was associated with a higher prevalence of pre-HSCT minimal residual disease (MRD) > 0.1% (n=88), which exhibited a significantly increased incidence of CNS relapse (16.6% vs. 6.8%, p = 0.009). Factors such as age, number of intrathecal chemotherapy treatments, conditioning intensity, and dose of anti-thymocyte globulin did not have a significant impact. Among the 34 patients with CNS relapse, 21 (61.7%) had isolated CNS relapse, and 7 of them showed evidence of positive MRD. With the exception of one patient who refused further treatment, the remaining 33 patients received initial CNS local therapy, including intrathecal chemotherapy combined with radiotherapy in 27 (82.0%) cases, intrathecal chemotherapy alone in 6 cases, and radiotherapy alone in 4 cases. Out of these patients, 22 were treated with dasatinib or ponatinib, while 6 received salvage chemotherapy. The median overall survival for patients with CNS relapse was 11.2 months, which was 19.8 months for those with isolated CNS relapse and 10.9 months for those with CNS relapse accompanied by bone marrow relapse (p = 0.077). The subsequent 2-year cumulative incidence of CNS relapse was 30%.

Conclusion Based on our findings, the current prophylaxis strategy employed for the prevention of post-HSCT CNS relapse in adult ALL appears to be effective. However, our data indicates that MRD-positive Ph-positive ALL requires distinct management due to its significantly elevated risk of CNS relapse. New strategies including prophylactic radiotherapy for this high-risk subgroup must be considered to prevent post-HSCT CNS disease.