Haematologica 2002; 87:(04)ELT19
[Medline] [prev] [index] [next]Late relapse of childhood acute lymphoblastic leukemia and plymerose chain reaction-monitoring of minimal residual disease: how much time can elapse between "molecular" and clinical relapse?
Maurizio Aricò,§ Giuseppe Germano,* Laura del Giudice,* Ottavio Ziino,§ Franco Locatelli,** Giuseppe Basso*
§Pediatric Onco-Hematology, Ospedale dei Bambini G. Di Cristina, Palermo; *Department of Pediatrics, University of Padua; **Department of Pediatrics, IRCCS Policlinico S.Matteo, Pavia, Italy.
Correspondence: Maurizio Aricò, OncoEmatologia Pediatrica, Ospedale dei Bambini "G. Di Cristina", 90100 Palermo, Italy. Tel. +39-091-6666131. Fax +39-091-6666202. Email: arico@ospedalecivicopa.org
Leukemia relapse, the most frequent cause of treatment failure in childhood acute lymphoblastic leukemia (ALL), usually occurs in the bone marrow within three years after achieving complete remission (CR). Together with flow-cytometry and analysis of breakpoint fusion regions of chromosome aberrations, detection of clone-specific immunoglobulin and T-cell receptor gene rearrangements by PCR offers the opportunity to identify leukemic cells undetectable at light microscopy,1 allowing detection of leukemic cells during CR even at very low concentration (<=10-4). Repeated measurement of PCR-defined minimal residual disease (MRD) is currently considered one of the most promising strategies for treatment tailoring and understanding ALL biology.2-4 The growth pattern of relapsing ALL clones remains unclear. In particular no information is available on the results of PCR-defined MRD in children who suffer a late relapse.Case report. MS presented at 3.2 years with WBC 26,980/mm3, hemoglobin 4.7 gr/dL, platelets 48000/mm3. B-precursor ALL (CD10+, CD19+, CD20+, CD22+, CD24+, CyIgm+, HLADR+; DNA index 1.0) was diagnosed; molecular screening ruled out t(9;22) and t(4;11) translocations. According to the AIEOP-ALL-95 study,5 due to insufficient clearance of circulating blasts (1,230/mm3) on day 8 after a steroid prophase (prednisone poor response; PPR), the girl was assigned to the high risk group. Day 15 bone marrow confirmed the slow clearance (15% blasts). She achieved CR by the end of induction therapy (day 43) and in June 1997, after a total duration of 24 months, treatment was electively withdrawn. In November 2000, 5.3 years after CR was achieved, she developed bone pain, with WBC 33,250/mm3, Hb 9.7 gr/dL, platelets 162,000/mm3. Isolated marrow relapse by a B-precursor clone mirroring the initial one was diagnosed. Following a four-drug induction therapy, the child promptly achieved second complete remission. With the informed consent of the parents, patients in the AIEOP-ALL-95 study underwent an additional bone marrow aspirate one year off-treatment aimed at possible MRD monitoring; since the results of any PCR-based study performed on this sample were not allowed for immediate clinical use, the samples were checked for morphologic remission and then stored for future evaluation. Thus, DNA from the whole series of bone marrow samples, collected during and after treatment completion and at the time of the leukemia relapse, was studied for MRD evaluation by PCR analysis of Ig/TCR gene rearrangement.6 A homoduplex band corresponding to the VH1-JHc was observed both at diagnosis and relapse and the same for VgI-Jg1.3/2.3 rearrangement. Both primers had 1x10-4 sensitivity. Results of RQ-PCR analysis are summarized in Figure 1 and Table 1.
Our patient developed a late leukemia relapse, over five years after initial CR had been achieved. Treatment with a full course of high-risk, BFM-type chemotherapy, due to slow initial response to chemotherapy documented by PPR and incomplete clearance at day-15 marrow, may have delayed, but not prevented, ALL relapse. A tempting question is whether a late ALL is a true relapse or otherwise a secondary tumor. Whenever an adequate molecular target is available, true late relapse has usually been confirmed.7,8 In a recent study, a total of 1,134 of 2,746 children had survived 10 years or more in first CR. Of those, 12 subsequently relapsed. Their relapse blast cells were shown to express an identical clonal IgH or TCR gene rearrangement on PCR analysis of DNA from diagnosis and relapse. These data raise the question on how blasts can survive quiescent for so long but lack of information on MRD study during remission prevented any further information.9 In our patient, despite initial complete response defined by two markers at the sensitivity level of 10-4, initial re-growth of the leukemic clone could be identified, at least by one of the two markers, one year after treatment withdrawal. Nevertheless the child remained asymptomatic until leukemia relapse occurred 2.5 year later, suggesting that her "molecular relapse" of B-lineage childhood ALL remained pre-symptomatic for a prolonged time. Whether such slow leukemic clone re-growth was due to a slow doubling time cannot be defined since cell cycle studies were not performed in this case. Molecular monitoring has been recently applied in genetic subgroups of childhood ALL. In the t(12;21) TEL/AML subgroup, evidence of "molecular relapse" has been associated with clinical relapse shortly thereafter, suggesting that molecular monitoring is prognostically relevant.10 Monitoring of bcr/abl fusion product in adult CML showed controversial results with some patients bearing repeated molecular evidence of residual leukemia despite long-lasting clinical and morphologic remission, while use of quantitative RT-PCR assay appears more promising. The counterpart of this observation has not been thoroughly investigated in childhood Ph+ ALL so far. This case suggests that in childhood, B-lineage high-risk ALL, relapse may occur after a prolonged period of clinical and "molecular remission". At present we cannot define how late molecular relapse may occur and, even more importantly, what time interval may elapse between "molecular relapse" and clinical relapse. Only prospective evaluation, by use of quantitative PCR technology, of patients treated front-line in large cooperative trials, as in the in current AIEOP-BFM-ALL-2000, may provide detailed information on unselected groups of patients and clarify most of these issues. Patients in whom leukemia relapse could be reliably identified in a preclinical stage could be addressed to proper treatment intensification, which in some cases can be resolutory.
Acknowledgments: Fondazione Città della Speranza, La Jolla, Ca, USA.
References