In acute myeloid leukemia (AML), the level of MRD as determined by multiparametric flow cytometry (MPFC) has been shown to impact on remission duration and survival. BM is the most common source to perform MRD assessment. We have previously showed that BM MRD negativity after consolidation was associated with a significantly longer relapse free survival (RFS) and overall survival (OS). However, in children with T acute lymphoid leukemia, it has been reported that PB might be used as an alternative source to BM for MRD studies. Based on this, we investigated whether PB might substitute for BM to monitor MRD in adult AML patients, showing the same prognostic value. Forty adult patients with AML were enrolled into the EORTC/GIMEMA protocols AML10/AML12 (age <61 yrs) or AML13/AML15 (age>61 yrs), all consisting in intensive induction and consolidation cycles, and, for patients aged <61 years, autologous or allogeneic stem cell transplantation. Median age was 48 years (range 21–73), all FAB subtypes were represented with the exception of M3 cases. We used MPFC to compare MRD measurements in 40 and 38 pairs of BM and PB after induction and consolidation, respectively. Findings in BM e PB were highly concordant after induction and consolidation therapy. In fact, median value of BM and PB residual leukemic cells (BMRLC and PBRLC, respectively) after induction, was 5.75x10−3 (range 1x10−4–1.64x10−1) and 4.7x10−3 (range, 3x10−5–9.3x10−2), respectively (r=0.84, P<0.001). After consolidation, the median value of BMRLC and PBRLC was 6.8x10−3 (range 2x10−5–6.3x10−2) and 7.7x10−3(range 3.5x10−5–1.34x10−1), respectively (r=0.82, P<0.001). The cut-off value of PBRLC which correlated with the clinical outcome was 1x10−4; in fact, 27 of 37 (73%) patients with PBRLC >1x10−4 after induction had a relapse whereas, the 3 patients with <1x10−4 PBRLC did not (P=0.028). After consolidation, using the same threshold, 32 patients were considered MRD positive (MRDCons+) and 81% (26) of them experienced a relapse; the remaining 6 patients, who were MRD negative (MRDCons−) are still in complete remission (P=0.00033). Duration of RFS was significantly longer in the MRDCons− group (median not reached, range 1.7–71 months) as compared to the MRDCons+ one (median 8.8 months, range 1–22) (P=0.005); in multivariate analysis, PB MRD status at the end of consolidation was an independent factor impacting on RFS (P=0.009). In conclusion: 1) PB may be used to monitor MRD in patients with AML, allowing closer monitoring of leukemia while sparing patients the discomfort of BM aspiration; 2) the level of MRD in the PB after consolidation therapy, may provide useful prognostic informations. Our results warrant further studies in a larger group of patients recruited to different treatment protocols and monitored at different time-points.