First results of a randomized trial (APL91 trial) and other randomized or non-randomized studies have shown that ATRA followed by chemotherapy significantly increased event-free survival (EFS) and survival, and decreased the incidence of relapse by comparison to chemotherapy alone in newly diagnosed APL. We present here long-term follow-up of the APL91 trial. In this trial, 101 patients had been randomized between ATRA followed by three courses of daunorubicin-AraC chemotherapy (ATRA group) and the same chemotherapy alone (chemotherapy group). Results were reanalyzed 73 months after closing of patient entry. Updated results of APL 91 trial found a Kaplan-Meier estimate of EFS and relapse rate at 4 years of 63% and 31% in the ATRA group, as compared to 17% and 78% in the chemotherapy group (P= 10(-4) and relative risk 2.95, P= 10(-4) and relative risk 3.68, respectively). Kaplan-Meier survival at 4 years was 76% in the ATRA group and 49% in the chemotherapy group (P= 0.026, relative risk 2.7). In the chemotherapy group, seven of the 27 relapses occurred after 18 months, but no relapse was seen after 43 months. In the ATRA group, four of the 17 relapses occurred after 18 months, including two late relapses (at 58 and 74 months). In the chemotherapy group, 23 of the 25 patients who relapsed achieved a second CR with ATRA, and the Kaplan-Meier estimate of second relapse was 40% at 30 months. In the ATRA group, the 10 patients who relapsed and were retreated with ATRA achieved a second CR. In conclusion, long-term results of APL91 trial confirm the superiority of the combination of ATRA and chemotherapy over chemotherapy alone in newly diagnosed APL, and that ATRA should be incorporated in the front-line treatment of APL.