Characteristics and Outcome of Older Patients with Acute Promyelocytic Leukemia Front-Line Treated with or without Arsenic Trioxide — an International Collaborative Study

Background: Characteristics and outcome of older patients (pts) with acute promyelocytic leukemia (APL) are unclear due to lack of clinical data.

Aims: To describe a large series of older APL pts and compare outcome according to treatment strategy.

Methods: We retrospectively studied 475 APL pts (median age, 73.8 yrs; range, 70-90.3 yrs), treated between 1990 and 2018 within four study groups/institutions of the US and Europe (Acute Leukemia French Association, n=228; Programa Espanol de Tratamientos en Hematologia, n=211; Study Alliance Leukemia, n=28; Johns Hopkins School of Medicine, Baltimore, n=8). APL was confirmed either by cytogenetics, fluorescence in situ hybridization and/or polymerase chain reaction. For analysis, pts were grouped according to treatment: i) chemotherapy/all-trans retinoic acid (CTX/ATRA, n=260; consisting of daunorubicin/idarubicin and ATRA for induction and different CTXs+ATRA for consolidation), ii) ATO/ATRA±CTX, n=177 (according to Lo-Coco F, et al. NEJM, 2013, n=23 or CTX/ATO/ATRA, n=154), iii) less intensive therapy, n=26 (reduced CTX, n=2 or ATRA only, n=24) and iv) no treatment/unknown, n=12.

Results: Median white blood cell (WBC) and platelet counts at diagnosis were 1.5/nl (range, 0.1-242/nl) and 37/nl (range, 2-261/nl), respectively. Two-hundred twenty-nine pts (48%) were female. Cytogenetic analysis was available in 408 pts and 85 (21%) had additional abnormalities. BCR3 was positive in 138 (44%) of 316 available pts. Only 15 (22%) of 69 tested pts were FLT3-ITD positive. One hundred (22%) of 464 pts had a WBC count >10/nl. Median WBC was significantly lower (P<0.001) in the ATO/ATRA±CTX group (1.2/nl) as compared to the CTX/ATRA (2.05/nl) or less intensive (2.8/nl) group. Median age was comparable in CTX/ATRA (73.5 yrs) and ATO/ATRA±CTX (73.6 yrs) but significantly higher (P<0.001) in the less intensive group (79.6 yrs). Median platelet counts were comparable (P=0.08) in the three groups (CTX/ATRA, 34.5/nl; ATO/ATRA±CTX, 41/nl; less intensive, 39/nl). Response data were available in 459 (97%) pts.

Complete remission (CR) after induction therapy was achieved in 75% (194/259) of the CTX/ATRA group, in 93% (162/174) of the ATO/ATRA±CTX group, and in 50% (13/26) of the less intensive group. Two pts of the CTX/ATRA group and one patient after ATRA only were refractory. Early death rates were 24% (n=63) after CTX/ATRA, 7% (n=12) after ATO/ATRA±CTX and 46% (n=12) in the less intensive group. A logistic regression model revealed age above 75 yrs (odds ratio [OR], 0.53; P=0.02) higher WBC (OR, 0.35; P<0.001), less intensive treatment (OR, 0.46; P=0.08) and therapy with ATO/ATRA±CTX (OR, 4.2; P<0.001) as significant factors for achievement of a CR. Median follow-up was 4.7 yrs (95%-CI, 4.2-5.4 yrs) and 5-yrs overall survival (OS) 62% (95%-CI, 57-67%). Treatment with ATO/ATRA±CTX was associated with significantly higher 5-yrs relapse-free (RFS; 96% [95%-CI, 92-99%] vs. 88% [95%-CI, 83-93%]; P=0.007) and OS rates (78% [95%-CI, 72-85] vs. 58% [95%-CI, 52-65%]; P<0.001) as compared to CTX/ATRA. Of note, higher WBC count (>10/nl) was associated with an inferior RFS in CTX/ATRA (P<0.001), but not in ATO/ATRA±CTX (P=0.47). When restricting the analysis according to treatment period after the year 2000, ATO/ATRA±CTX was still associated with a significant better RFS (P=0.05) and OS (P<0.001) as compared to CTX/ATRA. Overall, only five pts relapsed after ATO/ATRA±CTX as compared to 33 after CTX/ATRA. As expected, higher WBC (>10/nl; hazard ratio [HR], 2.36; P<0.001), age >75 yrs (HR, 2.07; P<0.001) and therapy with ATO/ATRA±CTX (HR, 0.48; P<0.001) were significant factors for OS.

Conclusions: The ATO-based regimen for first line treatment of elderly APL pts was associated with excellent and sustained response rates. Our data demonstrate the important potential of ATO/ATRA in the primary management of older APL pts.