Potential curability of newly diagnosed acute promyelocytic leukemia without use of chemotherapy: the example of liposomal all-trans retinoic acid

Several years ago we reported that liposomal all-trans retinoic acid (L-ATRA) used alone might cure some patients with untreated acute promyelocytic leukemia (APL).^1,2  The median follow-up was 1.5 years from complete remission (CR) date. Because the risk of relapse does not decrease appreciably until considerably later,³  we herein update the study. The L-ATRA dose was 90 mg/m² every other day until CR, after which this dose was given 3 times weekly for 9 months. Using a sensitivity level of 10 (-4), polymerase chain reaction (PCR) testing for the promyelocytic leukemia/retinoic acid receptor{alpha} (PML-RAR) (fusion protein) was done every 3 months for 2 years from CR date. If positive, the test was repeated 2 to 4 weeks later. If again positive (“molecular relapse”), patients received 12 mg/m² idarubicin 3 times daily every 4 to 5 weeks for 3 courses. Thirty-four patients, median age 49 years, median white blood cell count (WBC) 2000/μL, were treated: 8 were high risk (WBC count > 10 000/μL) using Sanz et al's system.⁴  The CR rate was 26 of 34, but only 3 of 8 in high-risk patients. Of the 34 patients, 10 (29%, 95% CI 15%-47%) remain alive in first CR at a median of 4.5 years (range, 3.0-5.5 years) from CR date despite never receiving other anti-APL therapy. In each case, the most recent PCR test is negative, having been obtained a median of 3.2 years (range, 1.4-5.5 years) years from CR date. The remaining 16 patients who entered CR received idarubicin; in one of these, the reports of the relevant PCRs were later changed to negative and in another only a single PCR test was positive. Thus, 14 patients “required” idarubicin: 8 had molecular relapse (at a median of 9 months from CR date), 2 failed to achieve molecular CR despite 6 months of L-ATRA monotherapy, and 4 had simultaneous clinical and molecular relapse (at a median of 17 months from CR date). The 10 patients who never received treatment other than L-ATRA each had a presenting WBC count less than 10 000/μL, as did 13 of the 14 who required idarubicin. The 2 groups also had similar distributions of initial WBC count, platelet count, and type of PML-RAR (isoform and age). The PCR status at CR was of no discriminatory value since all patients but one were PCR positive at CR, with 24 of 26 becoming negative by 3 months. Of the 10 patients whose first indication of failure was molecular, 9 received idarubicin in (hematologic) CR. Of those, 6 remain in hematologic CR, while 3 had hematologic relapse, which occurred within 1 year of molecular failure.

The immediate significance of our results is limited. L-ATRA is unavailable commercially. The 3 of 8 CR rate in high-risk patients seems extraordinarily low. Furthermore, while possibly sparing two fifths (ie, 10 of 26) of low-risk patients the need for chemotherapy, L-ATRA was not free of toxicity^1,2  and required 3 intravenous infusions weekly. Nonetheless, the observation that patients can be potentially cured of APL without use of chemotherapy should encourage further attempts in the same direction as, for example, in our current trial using ATRA and arsenic trioxide.⁵