Searching the central nervous system for relapsed acute promyelocytic leukemia

The introduction of all-trans retinoic acid (ATRA) transformed acute promyelocytic leukemia (APL) from the most fatal to the most curable subtype of acute myeloid leukemia (AML) in adults. The treatment of APL has evolved relatively rapidly, from combination therapy with ATRA and cytotoxic chemotherapy, to non-chemotherapy regimens that use ATRA and arsenic trioxide (ATO) alone [1–3]. The successes seen in the treatment of APL mirror the victories with combination chemotherapy for Hodgkin lymphoma and pediatric acute lymphoblastic leukemia and tyrosine kinase inhibition for chronic myeloid leukemia (CML).

Despite the remarkable progress in extending survival in APL, occasional patients – most of them with high-risk disease at diagnosis – relapse. The most common site of extramedullary relapse is the central nervous system (CNS). In this issue of Leukemia and Lymphoma, Gupta and colleagues report five patients with CNS relapse among a cohort of 115 patients seen at the Mayo Clinic between 1995 and 2012 [4]. Three of these five patients had isolated CNS disease at the time of relapse, with no evidence of systemic disease. Two out of four patients (the fifth did not have adequate information) had low-risk disease, with a white blood cell count less than 10 000/μL at the time of diagnosis.

Previous reports from large clinical trials in APL confirm that the CNS is the most common site of extramedullary relapse. Specchia and colleagues reported data from the AIDA-0493 trial, performed by the Italian GIMEMA (Gruppo Italiano Malattie Ematologiche Ddell’Adulto) cooperative group, that the CNS was the most frequent site of extramedullary relapse; 88% of the patients had hematologic relapse, 8% of patients relapsed in the CNS and 4% of patients relapsed at other extramedullary sites [5]. In a subsequent report from de Botton and colleagues, 806 patients from the French APL91 and APL93 trials and the Spanish PETHEMA (Programa para el Estudio de la Terapéutica en Hemopatía Maligna) group's LPA 96 trial were followed for relapse [6]. Of the 169 patients who relapsed after first remission, 10 relapsed in extramedullary sites and nine of those patients had a CNS relapse. Montesinos and colleagues reported on the combined rate of CNS relapse in the LPA 96 and LPA 99 trials performed by the PETHEMA group. In addition to the five CNS relapses previously reported from the LPA 96 trial, an additional six patients treated in the LPA 99 trial had a CNS relapse [7].

Why is the CNS the most common site of extramedullary relapse? It may be related to increased CD56 expression (neural crest adhesion factor) on the leukemic promyelocyte, and the (justified) reluctance of clinicians to perform diagnostic lumbar punctures on patients with APL, all of whom are at high risk of bleeding due to the coagulopathy associated with APL [8]. Intriguingly, there are other unusual sanctuary sites such as the auditory canal and mastoid process that are described in the literature [9]. Before the introduction of ATRA, when induction and consolidation regimens relied on anthracyclines and cytarabine, overall survival was markedly worse, but CNS relapse was uncommon. This trend was recognized by Specchia, when she compared the rate of CNS relapse in the LAP 0398 trial, a GIMEMA protocol for newly diagnosed APL performed pre-ATRA, and the AIDA 0493 trial, which combined ATRA and idarubicin. Only one patient in LAP 0398 (1%) had a CNS relapse, compared to 8% of the patients on the AIDA 0493 protocol. It may be that the increased use of regimens with no or minimal chemotherapy, particularly cytarabine, and the low penetration of ATRA and ATO into the CNS, allows the CNS to harbor leukemic promyelocytes [10].

Because of the shift to non-chemotherapy based treatment strategies for APL, CNS relapse may become a more prevalent clinical problem in the future. Results presented by Lo Coco and colleagues from APL 0406 – a prospective randomized phase III clinical trial for patients with low-risk disease – demonstrated that ATRA and arsenic without any chemotherapy led to a non-inferior event-free survival (EFS) compared with standard treatment (idarubicin and ATRA-based induction and consolidation) [3]. After 31 months, the primary endpoint, EFS, was achieved in 97% of patients in the experimental arm and 86.7% in the standard arm. Overall survival was 98.7% in the experimental arm and 91.1% in the standard arm. This clinical trial may well change the standard of care in patients with newly diagnosed, low-risk APL.

This most recent report on CNS relapse from Gupta et al. and the likely shift to ATRA and ATO induction therapy for patients with low-risk disease raise important questions. In the modern era of APL treatment, will we see a rise in the incidence of CNS relapse? If so, perhaps we should consider administration of prophylactic intrathecal chemotherapy, to both low-risk and high-risk patients, once the coagulopathy associated with APL has resolved. Future clinical trials of APL, particularly in patients with high-risk disease, should include a careful evaluation for CNS relapse.

Potential conflict of interest:

Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.