BACKGROUNDReal-time quantitative RT-PCR (RQ-PCR) assay has become a vital tool to monitor residual disease of leukemia. However, the complexity and standardization of RQ-PCR should never be overlooked and the results should be interpreted cautiously in clinical conditions. We aimed to assess the methodology of RQ-PCR and its clinical applications in monitoring molecular kinetics of 36 newly diagnosed cases of acute promyelocytic leukemia patients with t (15; 17) from October 2004 to December 2005.

METHODSAll the TaqMan probe-based RQ-PCR reactions and analysis were performed on an ABI-PRISM 7,500 platform. The quantitation of PML-RARalpha transcripts was represented by the normalized quotient, that is, PML-RARalpha transcript copies divided by ABL transcript copies. According to induction therapy, the patients were classed into two groups: group 1 (n = 23), three-drug combination including arsenics, all-trans retinoic acid and mitoxantrone; and group 2 (n = 13), two-drug combination from all-trans retinoic acid, arsenics and mitoxantrone.

RESULTSThe sensitivity of RQ-PCR was 1 per 10(5) cells and 5 copies of the PML-RARalpha transcript could be reproducibly detected. No false positive results occurred in 40 non-acute promyelocytic leukemia samples. Optimal amplification efficiency could be attained, which was determined by the slope of the standard curves (slope: -3.2 - -3.7). The inter-assay and intra-assay variation coefficients of the method were 1.01% and 0.56% respectively. Although the time to attain hematological complete remission was similar in both groups, the time to achieve molecular remission of group 1 was significantly shorter than that of group 2 (61 days vs 75 days, P = 0.034). The rate of molecular remission within 70 days was higher in group 1 than in group 2 (75.00% vs 38.46%, P = 0.036). Compared with pretreatment, median reduction of the PML-RARalpha transcript before first consolidation therapy differed significantly between group 1 and group 2 (log scale, 3.15 vs 2.31, P = 0.024). Interestingly, we found that PML-RARalpha transcript levels temporarily increased in bone marrow (7 patients) and peripheral blood (22 patients) samples of patients during induction therapy in both groups.

CONCLUSIONSThe RQ-PCR assay is reliable for the detection of PML-RARalpha transcripts. Arsenics, all-trans retinoic acid and mitoxantrone triad induction treatment of acute promyelocytic leukemia is superior to two-drug combination induction therapy in terms of the molecular response.