The PML-RAR alpha transcript in long-term follow-up of acute promyelocytic leukemia patients

Abstract

BACKGROUND AND OBJECTIVES: Detection of PML-RAR alpha transcripts by RT-PCR is now established as a rapid and sensitive method for diagnosis of acute promyelocytic leukemia (APL). Although the majority of patients in long-term clinical remission are negative by consecutive reverse transcription polymerase chain reaction (RT-PCR) assays, negative tests are still observed in patients who ultimately relapse. Conversion from negative to positive PCR has been observed after consolidation and found to be a much stronger predictor of relapse. This study reports on 47 APL patients to determine the correlation between minimal residual disease (MRD) status and clinical outcome in our cohort of patients. DESIGN AND METHODS: The presence of PML-RAR alpha t transcripts was investigated in 47 APL patients (37 adults and 10 children) using a semi-nested reverse transcriptase-polymerase chain reaction to evaluate the prognostic value of RT-PCR tests. RESULTS: All patients achieved complete clinical remission (CCR) following induction treatment with all-trans retinoic acid (ATRA) and chemotherapy (CHT) or ATRA alone. Patients were followed up between 2 and 117.6 months (median: 37 months). Relapses occurred in 11 patients (9 adults and 2 children) between 11.4 and 19 months after diagnosis (median: 15.1 months) while 36 patients (28 adults and 8 children) remained in CCR. Seventy-five percent of patients carried the PML-RAR alpha long isoform (bcr 1/2) which also predominated among the relapsed cases (9 of 11) but did not associate with any adverse outcome (p= 0.37). For the purpose of this analysis, minimal residual disease tests were clustered into four time-intervals: 0-2 months, 3-5 months, 6-9 months and 10-24 months. INTERPRETATION AND CONCLUSIONS: Children showed persisting disease for longer than adults during the first 2 months of treatment. At 2 months, 10 (50%) of 20 patients who remained in CCR and 4 (80%) of 5 patients who subsequently relapsed were positive. Patients who remained in CCR had repeatedly negative results beyond 5.5 months from diagnosis. A positive MRD test preceded relapse in 3 of 4 tested patients. The ability of a negative test to predict CCR (predictive negative value, PNV) was greater after 6 months (>83%), while the ability of a positive test to predict relapse (predictive positive value, PPV) was most valuable only beyond 10 months (100%). This study (i) highlights the prognostic value of RT-PCR monitoring after treatment of APL patients but only from the end of treatment, (ii) shows an association between conversion to a positive test and relapse and (iii) suggests that PCR assessments should be carried out at 3-month intervals to provide a more accurate prediction of hematologic relapses but only after the end of treatment.