We analyzed minimal residual disease (MRD) by consensus polymerase chain reaction (PCR), quantitative PCR (qPCR), and flow cytometry in 40 patients with chronic lymphocytic leukemia (CLL) who underwent stem cell transplantation; 97.4%, 89%, and 100% of the patients could be studied by consensus PCR, qPCR, and flow cytometry, respectively. Overall, 164 of 248 samples were negative for MRD by consensus PCR. Among those, CLL cells were detected by qPCR and by flow cytometry in 77 (47%) and 39 (23%) of the 164 samples, respectively. All 84 samples positive on PCR had detectable CLL cells by qPCR and flow cytometry. A good correlation was seen between MRD levels by flow cytometry and by qPCR (n = 254; r = 0.826; P < .001). Fifteen of 25 patients receiving autografts suffered a relapse, with increasing levels of MRD being observed before relapse in all of them. MRD detection within the first 6 months after autologous transplantation identified patients with a high relapse risk. In contrast, in allografted patients (n = 15) MRD did not correlate with outcome. In conclusion, quantitative methods to assess MRD (flow cytometry and qPCR) are more accurate than consensus PCR to predict clinical evolution. These results might be useful to investigate treatments aimed at preventing relapse in patients with CLL who have received an autograft.