Abstract
Early measurement of blast clearance is a relevant prognostic indicator in childhood acute lymphoblastic leukemia (ALL). To this purpose we measured, by four-colour flowcytometry (FC), the percentage of blast cells in bone marrow samples from Italian patients enrolled in the multicentre AIEOP-BFM ALL 2000 trial. Samples were collected on day 15 (after 14 days of steroids, and one dose of IT-MTX, vincristine, daunorubicine, asparaginase) and shipped overnight to the reference laboratory. The data were compared to PCR-MRD performed, by study design, on day +33 and +78 BM samples. We report the results of patients enrolled between December 2000 and October 2004. The 561 patients studied were not different from the remaining ones (with no available material) including their cumulative incidence of relapse (SE): 17.3% (1.9) vs. 18.1% (1.5) in 850 patients not studied. According to the results of FC-MRD, 5 groups were defined: negative (blast count <0.01%, n=143), <0.1% (n=94), <1% (n=149), 1–10% (n=119), >10% (n=56). Their cumulative 5-year risk of relapse was: 4.1% (1.9), 9.3% (4.0), 14.3% (3.2), 26.5% (5.5), 53.7% (7.4), respectively. By PCR-MRD, the same patients were stratified as follows: 177 were standard risk and had 5-year risk of relapse of 4.1% (1.7), 233 at intermediate risk had a relapse risk of 24.2% (3.4), 37 at high risk had a relapse risk of 58.1% (9); the remaining 124 patients (21.6%) were not stratified by PCR-MRD due to lack of 2 sensitive (≥10−4) markers. Of 177 patients classified as standard risk by PCR (double negative), 110 fell within the 2 subgroups with lower FC-MRD (<0.1%), 46 had <1%, 19 had <10%, only 2 >10% of blasts. Of the 233 patients stratified as PCR-MRD intermediate risk (d78 <10−3), FC-MRD related groups had the following probabilities of EFS: 93.5% (3.6; n=47), 83.3%(8.0; n=30), 80.5%(5.1; n=70), 66.5%(10.8; n=57), 39.2%(11.8; n=29). We conclude that very early measurement of FC-MRD on day 15 bone marrow is feasible in our multicentre cooperative setting. On the basis of our data we suggest the following risk groups: standard, when <0.1% blasts on day 15 BM; intermediate for 0.1 to <10%; high, for >10% blasts. These groups had a risk of relapse of 6.2% (1.9), 19.5% (3), and 53.7% (7.4), respectively. Since it is fast, reproducible, relatively cheap and applicable to virtually all patients, our group decided to apply it prospectively on all ALL patients to integrate PCR-based stratification. Our findings showed that:
early (d15) MRD detection by FCM identifies different patients than PCR on d33 and d78; FCM may be very useful to identify earlier the highly sensitive ALL with low relapse risk (even though long-term follow-up is still missing), whereas later timepoints may be accessible for PCR and the identification of HR patients.
Flow cytometric detection of leukemic blasts in cerebrospinal fluid predicts risk of relapse in childhood acute lymphoblastic leukemia: a Nordic Society of Pediatric Hematology and Oncology study
