Leukemia (2020) 34:2333–2341 https://doi.org/10.1038/s41375-020-0758-4 ARTICLE Acute myeloid leukemia Outcome of older (≥70 years) APL patients frontline treated with or without arsenic trioxide—an International Collaborative Study Sabine Kayser1,2,3 Ramy Rahmé4 David Martínez-Cuadrón5,6 Gabriel Ghiaur7 Xavier Thomas8 Marta Sobas9 ● ● ● ● ● ● Agnes Guerci-Bresler10 Ana Garrido11 Arnaud Pigneux12 Cristina Gil13 Emmanuel Raffoux14 Mar Tormo15 ● ● ● ● ● ● Norbert Vey16 Javier de la Serna17 Olga Salamero18 Eva Lengfelder19 Mark J. Levis 7 Pierre Fenaux4 ● ● ● ● ● ● Miguel A. Sanz5,6 Uwe Platzbecker1 Richard F. Schlenk3,20 Lionel Adès4 Pau Montesinos 5,6 ● ● ● ● Received: 30 November 2019 / Accepted: 10 February 2020 / Published online: 19 February 2020 © The Author(s) 2020. This article is published with open access Abstract Data on outcome in older (≥70 years) patients with acute promyelocytic leukemia after treatment with arsenic trioxide (ATO) compared with standard chemotherapy (CTX) is scarce. We evaluated 433 patients (median age, 73.4 years) treated either with ATO+ all-trans retinoic acid (ATO/ATRA; n = 26), CTX/ATRA + ATO during consolidation (CTX/ATRA/ATO; 1234567890();,: 1234567890();,: n = 148), or with CTX/ATRA (n = 259). Median follow-up for overall survival (OS) was 4.8 years. Complete remissions (CR) were achieved in 92% with ATO/ATRA and 82% with CTX/ATRA; induction death rates were 8% and 18%, respectively. For analysis of postremission outcomes we combined the ATO/ATRA and CTX/ATRA/ATO groups (ATO/ ATRA ± CTX). Cumulative incidence of relapse (CIR) was significantly lower after ATO/ATRA ± CTX compared with CTX/ATRA (P < 0.001). The same held true when restricting the analysis according to the treatment period after the year 2000. OS of patients in CR1 was not different between ATO/ATRA ± CTX compared with CTX/ATRA (P = 0.20). High (>10 × 109/l) white blood cell (WBC) counts at diagnosis were associated with higher CIR (P < 0.001) compared with lower WBC in the CTX/ATRA group, but not in the ATO/ATRA ± CTX group (P = 0.48). ATO, when added to ATRA or CTX/ ATRA is feasible and effective in elderly patients for remission induction and consolidation, particularly in patients with high WBC at diagnosis. Introduction therapy has shown to partly equalize the risk of relapse between APL-risk groups [5, 7]. Upfront arsenic trioxide (ATO) and all-trans retinoic acid To date, reports on outcome of older APL patients (≥70 (ATRA) combination has been proven to be highly effective years) with ATO/ATRA are scarce [8, 9]. Although the in acute promyelocytic leukemia (APL) and has become approval of ATO in the USA and Europe was based on the standard therapy in younger adult, nonhigh-risk patients randomized APL0406 study in patients 18–70 years of age [1, 2]. Within the pre-ATO era, high white blood cell [1, 2] no upper age limit for the treatment with ATO/ATRA (WBC) counts were identified as a risk factor for relapse [3– was included in both labels. Thus, clinical data on patients 6]. After initial induction treatment with ATRA and idar- above the age of 70 years treated with different regimens ubicin (AIDA), subsequent risk-adapted consolidation will help clinicians to identify best treatment options for their older APL patients. In addition, a high mortality before start and during the first day of treatment is still a major issue in APL, particu- These authors contributed equally: Lionel Adès, Pau Montesinos larly in older patients according to the Swedish adult acute Supplementary information The online version of this article (https:// leukemia registry [10]. Early death rates after ATRA± doi.org/10.1038/s41375-020-0758-4) contains supplementary anthracycline-based induction therapy were as high as 60% material, which is available to authorized users. in patients above the age of 80 years and still 18.8% in * Sabine Kayser patients aged 50–59 years [10]. Furthermore, despite dose s.kayser@dkfz-heidelberg.de reduction of chemotherapy (CTX) in older patients [11–13] Extended author information available on the last page of the article non-relapse mortality during postremission therapy remained 2334 S. Kayser et al. high, with 10–18.6% dying mainly due to infections, (n = 34) and APL2000 [22] (n = 41) trials, as well as whereas it was much lower (6.9%) after therapy with ATO standard AIDA-based CTX [23] (n = 9). Twenty-six (6%) [8]. In contrast to CTX/ATRA, only single cases with a patients were treated with ATO/ATRA [1] (including one secondary malignancy are reported after ATO treatment [8]. dose of idarubicin, n = 1 or daunorubicin for ATRA-syn- The objectives of our study were to evaluate character- drome, n = 1; and 5 days of decitabine, n = 1). In addition, istics and outcome of older (≥70 years) APL patients treated 148 (34%) patients were treated with CTX/ATRA + ATO with different treatment strategies in a large, international according to the French APL2006 trial [24] (n = 144), or cooperative cohort of patients. according to Gore et al. [25] (n = 4). Response and treat- ment failure were assessed according to Cheson et al. [15]. Patients and methods Statistical analyses Patients and treatment Overall survival (OS) was defined as recommended [15]. Induction death was defined as death occurring before Data on 475 patients aged ≥70 years, reported either to the complete remissions (CR) evaluation, occurring at the latest multicenter, multinational registries German Intergroup within 43 days after treatment initiation in our cohort. Napoleon registry (n = 28), Spanish PETHEMA (Programa Comparisons of patient characteristics were performed Español de Tratamientos en Hematología, n = 211), or with the Kruskal–Wallis rank sum test for continuous French APL Group (n = 228) or to the local data base at the variables and Fisher’s exact test for categorical variables, Johns Hopkins University of Maryland, USA (n = 8), were respectively. The median follow-up time was computed collected in a large international collaboration. All patients using the reverse Kaplan–Meier estimate [26]. The were treated between 1990 and 2018. Patients who were Kaplan–Meier method was used to estimate the distribution treated less intensively with single-agent ATRA or in of OS [27]. Confidence interval (CI) estimation for survival combination with low-dose cytarabine or comparable CTX curves was based on the cumulative hazard function using (n = 38) or patients with lacking data on response and Greenwood’s formula for variance estimation. Logrank tests survival (n = 4) were excluded from analysis. The final were employed to compare survival curves between groups. study cohort consisted of n = 433 patients. Diagnosis of For cumulative incidence of relapse (CIR) and death (CID), APL was based on genetic analysis as well as on equality of cumulative incidences between the treatment French–American–British Cooperative Group criteria [14], groups was evaluated using the Gray’s test [28]. A cause- and, after 2003, on revised International Working Group specific Cox proportional hazards regression model was criteria [15]. Chromosome banding was performed using used to identify prognostic variables for relapse [29]. The standard techniques, and karyotypes were described following variables were included in the Cox model: age at according to the International System for Human Cytoge- diagnosis, gender, APL treatment (ATO/ATRA ± CTX vs. netic Nomenclature [16]. The diagnosis was confirmed by CTX/ATRA), and WBC count (dichotomized, ≤10 × 109/l either reverse-transcriptase polymerase chain reaction (RT- vs. >10 × 109/l). All statistical analyses were performed qPCR) or fluorescence in situ hybridization detection by with the statistical software environment R, version 3.3.1, standard methods. FLT3 mutation screening for internal using the R packages, and survival, version 2.39-5 [30]. tandem duplications (ITD) and point mutations within the tyrosine kinase domain (TKD) was carried out as previously described [17, 18]. Data collection and analysis were Results approved by the local Institutional Review Boards. In total, 433 older APL patients, diagnosed between 1990 Treatment and 2015 from four study groups/institutions in the USA and Europe, were included. Median age was 73.4 years Two-hundred and fifty-nine patients (60%) were treated (range, 70–89 years). with ATRA and an anthracycline (daunorubicin or idar- Risk categorization based on WBC count at diagnosis ubicin) as induction and different CTX in combination with was available in 429 (99%) of the 433 patients and was ATRA as consolidation therapy according to treatment low-/intermediate-risk (WBC ≤ 10 × 109/l) in 339 (79%) and protocols active in various institutions and cooperative high-risk (WBC > 10 × 109/l) in 90 (21%) patients. groups (Supplementary Table 1). These protocols included Information on cytogenetics was available in 381 (88%) the PETHEMA LPA96 [19] (n = 23), LPA99 [20] (n = 45), patients. In 26 (7%) of the 381 patients t(15;17) could only LPA2005 [21] (n = 76), and LPA2012 (ClinicalTrials.gov be detected by FISH and/or RT-qPCR, whereas cytoge- Identifier: NCT02020161; n = 31), the French APL93 [7] netics showed a normal karyotype. In 238 (67%) of the Outcome of older (≥70 years) APL patients frontline treated with or without arsenic trioxide—an. . . 2335 remaining 355 patients, the balanced t(15;17) translocation Table 1 Comparison of presenting clinical and laboratory findings according to the applied therapy of older patients with acute was the sole abnormality, whereas in 117 (33%) patients, promyelocytic leukemia. the translocation was accompanied by additional cytoge- netic abnormalities, most frequently trisomy 8 (n = 7; 6%) Characteristics CTX/ATRA CTX/ATRA/ ATO/ATRA P value n = 259 ATO n = 26 or t(15;17) within a complex karyotype (≥2 cytogenetic n = 148 abnormalities in addition to t(15;17); n = 12; 10%). Regarding FLT3-ITD and TKD mutations, information Age, years was available in only 56 (13%) of the 433 patients. Of Median 73.4 73.2 76 0.07 those, eight (14%) patients had FLT3-ITD and two (4%) Range 70–88.7 70–89 71–87 had a FLT3-TKD mutation. Patients with FLT3-ITD Gender, no. (%) mutations had significantly higher WBC at diagnosis as Male 123 (47) 80 (54) 18 (69) 0.07 compared with FLT3 wild type patients (P < 0.001). Female 136 (53) 68 (46) 8 (31) Information on the PML-RARA transcript isoform WBC, ×109/l (breakpoint cluster region (BCR)) was available in 292 Median 2.1 1.3 0.8 <0.001 patients; 125 (43%) had the short isoform (BCR3). Baseline Range 0.1–764 0.2–136 0.2–3.6 characteristics according to the frontline strategy are shown Missing 4 0 0 in Table 1. Hemoglobin, g/dl Median age was in trend higher in the ATO/ATRA group Median 9.2 9.7 9.2 0.20 as compared with the CTX/ATRA and CTX/ATRA/ATO Range 4.0–15.2 4.6–14.3 6.0–14.0 groups. Median WBC, percentage of peripheral blood and Missing 14 2 17 bone marrow blasts as well as risk categorization were Platelet count, ×10 /l 9 significantly lower in the ATO/ATRA ± CTX group as Median 34 41 43 0.08 compared with the CTX/ATRA group. Range 3–213 4–261 11–195 Missing 8 3 2 Response to induction therapy Percentage of PB blastsa Median 61 6 1 <0.001 Overall, 356 patients achieved a CR (82%), two patients Range 0–100 0–99 0–67 had refractory disease (0.5%) and 75 patients died during induction therapy (17%). The rate of death during induction Missing 53 15 7 therapy was lower in patients receiving ATO/ATRA (n = 2/ Percentage of BM blastsa 26; 8%) as compared with those receiving CTX/ATRA Median 80 74 50 0.005 (n = 73/407; 18%) as induction therapy (P = 0.28). The rate Range 13–100 12–98 3–90 of deaths during induction therapy decreased over time from Missing 42 0 11 28% (17 of 60 patients) in 1990–1999, to 15% (29 of 188 High-riskb 72 (28) 18 (12) 0 <0.001 patients) in 2000–2009 and 16% (29 of 185 patients) in Missing 4 0 0 2010–2018 (P = 0.07). Main causes of deaths were bleed- Percentages may not add to 100 because of rounding. ing/hemorrhage (n = 16), infections (n = 17), differentia- ATO arsenic trioxide, ATRA all-trans retinoic acid, BM bone marrow, tion syndrome (n = 9), multi-organ failure (n = 3), as well CTX chemotherapy, PB peripheral blood, WBC white blood cell as other causes (not further specified; n = 30). counts. a CRs were achieved after induction therapy in 82% with Blast cells included malignant promyelocytes. CTX/ATRA (n = 332/407) and in 92% with ATO/ATRA b High risk: WBC > 10 × 109/l. (n = 24/26; Table 2; P = 0.29). In a logistic regression model on response to induction therapy WBC count above after induction therapy and two patients relapsed, whereas 10 × 109/l (P < 0.001) and age above 75 years (P = 0.04) no induction death or relapse occurred after ATO/ATRA were unfavorable factors, whereas treatment with ATO/ treatment. ATRA was a beneficial, but not a significant factor (P = 0.34; Table 3). APL was refractory after CTX/ATRA Survival analysis treatment in two patients, who died after 51 and 477 days from diagnosis, respectively. None of the patients in the Median follow-up for survival was 4.8 years (95% CI, ATO/ATRA group was refractory. 4.3–5.6 years). Overall, the estimated 5-year OS rate was Of twelve patients with a complex karyotype, eight were 66% (95% CI, 61–71%). As we were interested in the effect treated with AIDA-based regimen, whereas four patients of ATO on CIR and CID, as well as survival of CR patients received ATO/ATRA. Two patients treated with AIDA died we combined ATO/ATRA and CTX/ATRA + ATO in one 2336 S. Kayser et al. group (ATO/ATRA ± CTX). Treatment with ATO/ATRA ± at diagnosis (>10 × 109/l) were associated with an increased CTX was associated with significantly reduced CIR after 5 CIR in CTX/ATRA (P < 0.001; Fig. 3a), but not in ATO/ years (4% vs. 18%; P < 0.001; Fig. 1a) with no significant ATRA ± CTX (P = 0.48; Fig. 3b). differences in CID (12% and 13%; P = 0.73; Fig. 1b) and Overall, five patients relapsed after ATO/ATRA ± CTX OS rates of CR patients (84% (95% CI, 78–91) vs. 77% (median time from CR to relapse, 18.2 months) as com- (95% CI, 71–84%); P = 0.20; Fig. 2) as compared with pared with 33 after CTX/ATRA (median time from CR to CTX/ATRA. When restricting the analysis according to the relapse, 5 months). Of those, risk categorization was high in treatment period after the year 2000, ATO/ATRA ± CTX one patient of the ATO/ATRA ± CTX group as compared was still associated with significantly lower CIR (P = 0.02) with 22 patients of the CTX/ATRA group. as compared with CTX/ATRA. Of note, higher WBC counts Only two patients experienced besides a hematological a CNS relapse. Both relapses occurred after AIDA-based Table 2 Response to induction therapy according to treatment therapy. Radiotherapy was applied in one patient, whereas strategy. the other patient received no treatment. Both patients died % (N) CTX/ATRA ATO/ATRA shortly after relapse (13 days and 98 days). N = 407 N = 26 Of the CTX/ATRA group eight patients received ATO CR 82 (332) 92 (24) and additionally five patients ATO/ATRA as salvage ther- RD 0.5 (2) – apy. Overall, 12 of the 13 relapsed patients of the CTX/ ID 18 (73) 8 (2) ATRA group achieved a second remission, whereas one patient died during salvage therapy. Multivariate analysis Missing data, n = 3 (CTX/ATRA). Percentages may not add to 100 showed that higher WBC count (>10 × 109/l; hazard ratio because of rounding. (HR), 7.90; P < 0.001) had an adverse impact on relapse, ATO arsenic trioxide, ATRA all-trans retinoic acid, CR complete remission, CTX chemotherapy, ID induction death, N numbers, RD whereas therapy with ATO/ATRA ± CTX (HR, 0.23; P = resistant disease. 0.003) was beneficial (Table 4). Fifty-five patients died in remission (CTX/ATRA, n = 37; ATO/ATRA ± CTX, n = 18), mainly due to infections Table 3 Logistic regression model on response to induction therapy. (n = 12; CTX/ATRA, n = 7 and ATO/ATRA + CTX, n = Regression model on response 5), bleeding/hemorrhage (n = 9; CTX/ATRA, n = 2 and to induction therapy ATO/ATRA + CTX, n = 7), cardiac arrest + aneurysm OR P value rupture in the brain (ATO/ATRA + CTX, n = 1), ence- phalitis/cerebellitis (ATO/ATRA + CTX, n = 2), prostate Age above 75 years 0.55 0.030 cancer (ATO/ATRA + CTX, n = 1), and unknown causes WBC (>10 × 109/l) 0.26 <0.001 (n = 30; CTX/ATRA, n = 28 and ATO/ATRA + CTX, ATO/ATRA 2.21 0.30 n = 2). Male gender 0.72 0.22 Among patients achieving CR after intensive therapy, the ATO arsenic trioxide, ATRA all-trans retinoic acid, CTX chemotherapy, median follow-up time was 3.4 years for ATO/ATRA ± OR odds ratio, WBC white blood cell count. CTX and 7.2 years for CTX/ATRA. Data on the A B 1.0 1.0 P<0.001 P=0.73 Cumulative incidence of relapse Cumulative incidence of death 0.8 0.8 CTX/ATRA n=192 CTX/ATRA n=192 ATO/ATRA±CTX n=162 ATO/ATRA±CTX n=162 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 Time (years) Time (years) Fig. 1 Cumulative incidence of relapse (CIR, a) and cumulative incidence of death (CID, b) according to treatment strategy. ATO arsenic trioxide, ATRA all-trans retinoic acid, CTX chemotherapy, n number. Outcome of older (≥70 years) APL patients frontline treated with or without arsenic trioxide—an. . . 2337 development of secondary neoplasms were available in 163 ATRA as compared with CTX/ATRA, primarily due to a patients. Of those, 11 (7%) patients developed a secondary lower rate of induction deaths, which is in line with pub- neoplasm (gastro-intestinal cancer, n = 4; lung cancer, n = lished data in younger APL patients [1, 9]. The rate of 3; neuroendocrine carcinoma, n = 1; and secondary acute induction deaths decreased from 28% between 1990 and myeloid leukemia, n = 3) after a median of 61.5 months 1999 to 15% and 16% of the following decades, which may (range, 5.5–85 months) from the time point of CR. All of be attributable to improved supportive care and awareness them had been treated with CTX/ATRA. In a competing of APL as a medical emergency. This finding suggests that risk analysis among patients in CR after intensive therapy, the clear guidance in international recommendations about the rate of secondary malignancies was significantly lower early supportive care has had an effect in recent years [31]. after ATO/ATRA ± CTX as compared with CTX/ATRA Thus, an effective strategy to overcome induction mortality (P = 0.02). is due to standardized guidelines along with consultative support and sharing of expertise [32]. Main causes of induction death after CTX/ATRA included bleeding/ Discussion hemorrhage and infections, which is in line with previously published data [10, 33, 34]. However, 30 patients experi- Our retrospective analysis of a large cohort of 433 older enced induction death due to other causes, which might APL patients, spanning a time period of almost 28 years, simply be related to the fact, that older patients might have shows a higher response rate after treatment with ATO/ more frequent and severe comorbidities and thus lower capacity to tolerate CTX [10]. The main limitation of CTX/ ATRA treatment is the high rate of induction death, whereas 100 ATO/ATRA±CTX n=162 treatment with ATO/ATRA seems to be safer and being associated with a low-induction death rate. Thus, treating 75 Overall survival (%) patients with ATO/ATRA during induction seems to be CTX/ATRA n=192 rational rather than in consolidation, when the greatest risk 50 has passed [35]. However, in our cohort the number of patients treated with ATO/ATRA during induction was very 25 low and induction mortality was high after CTX/ATRA, P=0.20 particularly in the earlier treatment period before 1999. 0 Therefore, a comparison of OS between the CTX/ATRA 0 1 2 3 4 5 6 7 8 9 10 and ATO/ATRA groups is hampered. Thus, we focused in Time (years) time-to-event analyses on patients in first CR. As a potential selection bias, none of the patients treated Fig. 2 Overall survival. Overall survival of patients achieving a complete remission after induction therapy according to treatment with ATO/ATRA in induction therapy had high risk (WBC strategy. count >10 × 109/l) at diagnosis as compared with 28% in the A B P=0.48 1.0 1.0 P<0.001 Cumulative incidence of relapse Cumulative incidence of relapse WBC ≤ 10 109/l, n=146 WBC ≤ 10 109/l, n=146 0.8 0.8 WBC > 10 x 109/l, n=16 WBC > 10 x 109/l, n=43 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 Time (years) Time (years) Fig. 3 Cumulative incidence of relapse according to risk category incidence of relapse according to risk category (white blood cell and treatment. a Cumulative incidence of relapse according to risk count ≤10 × 109/l vs. >10 × 109/l) in patients treated with arsenic category (white blood cell count ≤10 × 109/l vs. >10 × 109/l) in patients trioxide-based regimens. treated with all-trans retinoic acid and chemotherapy. b Cumulative 2338 S. Kayser et al. Table 4 Cause-specific Cox model on relapse risk. neoplasm or solid tumor after ATO-based regimens as Cause-specific Cox model on compared with 11 (7%) patients after CTX/ATRA, includ- relapse risk ing four patients, who developed a secondary malignancy HR P value within 2.5 years after achievement of CR. Although the median follow-up time was shorter after ATO-based regi- Age above 75 years 1.07 0.86 mens as compared with CTX/ATRA, the reported latency WBC (>10 × 109/l) 7.90 <0.001 period between APL diagnosis and the development of a ATO/ATRA ± CTX 0.23 0.003 secondary malignancy of 6.6 months strongly argues for a Male gender 0.99 0.99 true lower incidence after ATO/ATRA ± CTX [36]. The ATO arsenic trioxide, ATRA all-trans retinoic acid, CTX chemotherapy, low-relapse rate after ATO-based therapy in our cohort is in HR hazard ratio, OR odds ratio, WBC white blood cell count. line with others [8, 35]. Moreover, 12 relapsed patients of the CTX/ATRA group were successfully salvaged with ATO-based therapy, confirming the tolerability of this CTX/ATRA group, which is a known risk factor for treatment option in elderly patients. In contrast to the pub- bleeding diathesis [3]. In addition, as a further potentially lication of Powell et al. [35] presenting a subgroup analysis selection bias, percentage of peripheral blood and bone in older patients, treatment with ATO/ATRA was not marrow blasts were also unequally distributed between the superior to the standard CTX/ATRA-approach in the low-/ CTX/ATRA and ATO-based groups. intermediate-risk in our cohort. However, a clear and sig- Regarding postremission outcome, results of the ran- nificant benefit was seen in the high-risk group. Thus, our domized North American Leukemia Intergroup Study data suggest that ATO-based regimens are effective in older C9710 on 481 APL patients (age range, 15–79 years patients independent of risk classification. including 77 patients above the age of 60 years) evalu- Regarding biological characteristics, additional cyto- ating ATO in first-line therapy during consolidation genetic abnormalities were present in 25% of the patients, demonstrated that ATO further reduced the risk of relapse most frequently trisomy 8, which is in line with published and improved survival as compared with consolidation data [37, 38]. In addition, t(15;17) could also be found with daunorubicin/cytarabine [35]. In contrast, a subgroup within a complex karyotype, suggesting that older age analysis of the randomized phase-III AML17 trial of the might lead to higher chromosomal instability and thus UK National Cancer Research Institute Acute Myeloid higher rate of abnormalities, as has been described [39]. Leukaemia Working Group on 49 older (age range, Although data were limited to a low number of patients, ≥60–77 years) patients showed no significant difference of FLT3-ITD mutations seem to be less frequent in older the 4-year OS rate after treatment with ATO/ATRA (74%; (14%) as compared with all APL patients (31%) [35]. As n = 25) as compared with the AIDA-based regimen (80%; expected, FLT3-ITD mutated patients had significantly n = 24) [9]. However, this was in contrast to the total higher WBC count at diagnosis as compared with FLT3 study cohort with significantly better event-free and wild type patients [37, 40–43]. To date, there are still relapse-free survival after ATO/ATRA as compared with conflicting data regarding the impact of additional chro- the AIDA-based regimen. In our cohort postremission mosomal or genetic abnormalities on outcome in APL outcomes were comparable after CTX/ATRA and ATO- patients [37, 40–49]. In our large cohort, the presence of based treatment in patients with low-/intermediate-risk, additional cytogenetic abnormalities had no impact on OS which is in contrast to previously published data in (P = 0.99), whereas FLT3-ITD was associated with an younger adults [1, 2]. Contrary to the data of the AML17 adverse impact on OS (P < 0.001). The latter issue, trial [9], the ATO-based treatment resulted in significant however, should be interpreted with precaution due to the lower CIR in patients with high-risk. low availability of FLT3 mutational status in our cohort. There was no difference between CTX/ATRA and ATO- Regarding the distribution of risk category according to based regimens with regard to deaths in remission with WBC count at diagnosis, published data are again con- relatively high rates of 12% and 13%, which is probably tradictory [49, 50]. Sanz et al. [49] reported that older due to the advanced age of the cohort. Besides bleeding patients seem to be more likely to present with non-high- infections were the most frequent cause of death, particu- risk APL as compared with their younger counterparts larly after CTX/ATRA treatment, arguing for a CTX-free (37% vs. 18%), which in part may account for the low approach. Other causes were most probably related to the relapse rate observed in their publication. In contrast, natural death rate. Lengfelder et al. reported on 31% (n = 28/91) of older In addition, there was a significant difference in the patients with high-risk APL [50]. In our cohort, the pro- occurrence of secondary malignancies, since none of the portion of high-risk patients in the CTX/ATRA group was patients developed a secondary therapy-related myeloid comparable with the latter publication [50], whereas it was Outcome of older (≥70 years) APL patients frontline treated with or without arsenic trioxide—an. . . 2339 lower in the ATO/ATRA/ ± CTX group due to study final results of the randomized Italian-German APL0406 trial. J inclusion criteria [24] as well as reservation to use ATO- Clin Oncol. 2016;35:605–12. 3. Testa U, Lo-Coco F. Prognostic factors in acute promyelocytic based therapy in high-risk patients. As expected, higher leukemia: strategies to define high-risk patients. Ann Hematol. WBC count above 10 × 109/l was an unfavorable factor for 2016;95:673–80. response to induction therapy as well as CIR and CID. Of 4. Sanz MA, Lo Coco F, Martín G, et al. Definition of relapse note, no difference in outcome was observed in high-risk risk and role of nonanthracycline drugs for consolidation in patients with acute promyelocytic leukemia: a joint study of the patients after treatment with ATO/ATRA ± CTX, whereas PETHEMA and GIMEMA cooperative groups. Blood. 2000;96: this was not the case in patients after treatment with CTX/ 1247–53. ATRA, suggesting that the ATO-based regimen is also 5. Sanz MA, Martin G, Gonzalez M, et al. Risk-adapted treatment of highly active within the high-risk group. acute promyelocytic leukemia with alltrans-retinoic acid and anthracycline monochemotherapy: a multicenter study by the In conclusion, ATO, when added to ATRA or CTX/ PETHEMA group. Blood. 2004;103:1237–43. ATRA, is a feasible and efficacious treatment and leads to 6. Fenaux P, Chastang C, Chevret S, et al. A randomized comparison good outcomes in the primary management of older APL of all transretinoic acid (ATRA) followed by chemotherapy and patients. Thus, it seems to be prudent to expand this ATRA plus chemotherapy and the role of maintenance therapy in newly diagnosed acute promyelocytic leukemia. The European approach to older APL patients. APL Group. Blood. 1999;94:1192–200. 7. Lo-Coco F, Avvisati G, Vignetti M, et al. Front-line treatment of Author contributions SK, PM, and RFS were responsible for the acute promyelocytic leukemia with AIDA induction followed by concept of this paper, contributed to the literature search data collec- risk-adapted consolidation for adults younger than 61 years: tion, analyzed and interpreted data, and wrote the paper. UP and LA results of the AIDA-2000 trial of the GIMEMA Group. Blood. contributed patients, interpreted data and critically revised the paper. 2010;116:3171–9. RR, DMC, GG, XT, MS, AGB, AG, AP, CG, ER, MT, NV, JS, OS, 8. Zhang Y, Zhang Z, Li J, et al. Long-term efficacy and safety of EL, MJL, PF, and MAS contributed patients and critically revised the arsenic trioxide for first-line treatment of elderly patients with paper. All authors reviewed and approved the final paper. newly diagnosed acute promyelocytic leukemia. Cancer. 2013;119:115–25. Funding Open Access funding enabled and organized by Projekt 9. Burnett AK, Russell NH, Hills RK, et al. Arsenic trioxide and all- DEAL. trans retinoic acid treatment for acute promyelocytic leukaemia in all risk groups (AML17): results of a randomised, controlled, phase 3 trial. Lancet Oncol. 2015;16:1295–305. Compliance with ethical standards 10. Lehmann S, Ravn A, Carlsson L, et al. Continuing high early death rate in acute promyelocytic leukemia: a population-based Conflict of interest SK was supported by the Olympia-Morata fel- report from the Swedish Adult Acute Leukemia Registry. Leu- lowship program from the Medical Faculty of the Heidelberg Uni- kemia. 2011;25:1128–34. versity. MJL is supported by a grant from the NCI (NCI Leukemia 11. Ades L, Chevret S, De Botton S, et al. Outcome of acute pro- SPORE P50 CA100632). UP has received research support from myelocytic leukemia treated with all-trans retinoic acid and che- TEVA. All other authors declare no competing conflict of interest. motherapy in elderly patients: the European group experience. Leukemia. 2005;19:230–3. Publisher’s note Springer Nature remains neutral with regard to 12. Mandelli F, Latagliata R, Fazi P, Rodeghiero F, Leoni F, et al. jurisdictional claims in published maps and institutional affiliations. Treatment of elderly patients (> or =60 years) with newly diag- nosed acute promyelocytic leukemia. Results of the Italian mul- Open Access This article is licensed under a Creative Commons ticenter group GIMEMA with ATRA and idarubicin (AIDA) Attribution 4.0 International License, which permits use, sharing, protocols. Leukemia. 2003;17:1085–90. adaptation, distribution and reproduction in any medium or format, as 13. Disperati P, Minden MD, Gupta V, et al. Acute promyelocytic leu- long as you give appropriate credit to the original author(s) and the kemia in patients aged 70 years and over—a single center experience source, provide a link to the Creative Commons license, and indicate if of unselected patients. Leuk Lymphoma. 2007;48:1654–8. changes were made. The images or other third party material in this 14. Bennett JM, Catovsky D, Daniel MT, et al. Proposed revised article are included in the article’s Creative Commons license, unless criteria for the classification of acute myeloid leukemia. A report indicated otherwise in a credit line to the material. If material is not of the French–American–British Cooperative Group. Ann Intern included in the article’s Creative Commons license and your intended Med. 1985;103:620–5. use is not permitted by statutory regulation or exceeds the permitted 15. Cheson BD, Bennett JM, Kopecky KJ, et al. Revised recom- use, you will need to obtain permission directly from the copyright mendations of the International Working Group for diagnosis, holder. To view a copy of this license, visit http://creativecommons. standardization of response criteria, treatment outcomes, and org/licenses/by/4.0/. reporting standards for therapeutic trials in acute myeloid leuke- mia. J Clin Oncol. 2003;21:4642–9. 16. Mitelman F. ISCN: an international system for human cytogenetic References nomenclature. Basel: S. Karger; 1995. 17. Yokota S, Kiyoi H, Nakao M, et al. Internal tandem duplication of 1. Lo-Coco F, Avvisati G, Vignetti M, et al. Retinoic acid and the FLT3 gene is preferentially seen in acute myeloid leukemia arsenic trioxide for acute promyelocytic leukemia. N Engl J Med. and myelodysplastic syndrome among various hematological 2013;369:111–21. malignancies. A study on a large series of patients and cell lines. 2. Platzbecker U, Avvisati G, Cicconi L, et al. Improved outcomes Leukemia. 1997;11:1605–9. with retinoic acid and arsenic trioxide compared with retinoic acid 18. Thiede C, Steudel C, Mohr B, et al. Analysis of FLT3-activating and chemotherapy in non-high-risk acute promyelocytic leukemia: mutations in 979 patients with acute myelogenous leukemia: 2340 S. Kayser et al. association with FAB subtypes and identification of subgroups leukemia: North American Leukemia Intergroup Study C9710. with poor prognosis. Blood. 2002;99:4326–35. Blood. 2010;116:3751–7. 19. Sanz MA, Martín G, Rayón C, et al. A modified AIDA protocol 36. Pagano L, Pulsoni A. Second malignancy after treatment of acute with anthracycline-based consolidation results in high antileu- promyelocytic leukemia: experience of GIMEMA trials. Blood. kemic efficacy and reduced toxicity in newly diagnosed PML/ 2002;100:1514–5. RARalphapositive acute promyelocytic leukemia. PETHEMA 37. Schlenk RF, Germing U, Hartmann F, et al. High-dose cytarabine group. Blood. 1999;94:3015–21. and mitoxantrone in consolidation therapy for acute promyelo- 20. Sanz MA, Montesinos P, Vellenga E, et al. Risk-adapted treatment cytic leukemia. Leukemia. 2005;19:978–83. of acute promyelocytic leukemia with all-trans retinoic acid and 38. Lou Y, Suo S, Tong H, et al. Characteristics and prognosis ana- anthracycline monochemotherapy: long-term outcome of the LPA lysis of additional chromosome abnormalities in newly diagnosed 99 multicenter study by the PETHEMA Group. Blood. acute promyelocytic leukemia treated with arsenic trioxide as the 2008;112:3130–4. front-line therapy. Leuk Res. 2013;37:1451–6. 21. Sanz MA, Montesinos P, Rayon C, et al. Risk-adapted treatment 39. Grimwade D, Walker H, Harrison G, et al. The predictive value of of acute promyelocytic leukemia based on all-trans retinoic acid hierarchical cytogenetic classification in older adults with acute and anthracycline with addition of cytarabine in consolidation myeloid leukemia (AML): analysis of 1065 patients entered into therapy for high risk patients: further improvements in treatment the United Kingdom Medical Research Council AML11 trial. outcome. Blood. 2010;115:5137–46. Blood. 2001;98:1312–20. 22. Adès L, Chevret S, Raffoux E, et al. Is cytarabine useful in the 40. Kiyoi H, Naoe T, Yokota S, et al. Internal tandem duplication of treatment of acute promyelocytic leukemia? Results of a rando- FLT3 associated with leukocytosis in acute promyelocytic leu- mized trial from the European Acute Promyelocytic Leukemia kemia. Leukemia Study Group of the Ministry of Health and Group. J Clin Oncol. 2006;24:5703–10. Welfare (Kohseisho). Leukemia. 1997;11:1447–52. 23. Mandelli F, Diverio D, Avvisati G, et al. Molecular remission in 41. Noguera NI, Breccia M, Divona M, et al. Alterations of the FLT3 PML/RAR alpha-positive acute promyelocytic leukemia by gene in acute promyelocytic leukemia: association with diag- combined all-trans retinoic acid and idarubicin (AIDA) therapy. nostic characteristics and analysis of clinical outcome in patients Gruppo Italiano-Malattie Ematologiche Maligne dell’Adulto and treated with the Italian AIDA protocol. Leukemia. 2002;16: Associazione Italiana di Ematologia ed Oncologia Pediatrica 2185–9. Cooperative Groups. Blood. 1997;90:1014–21. 42. Shih LY, Kuo MC, Liang DC, et al. Internal tandem duplication 24. Rahmé R, Ades L, Thomas X, et al. Reducing mortality in newly and Asp835 mutations of the FMS-like tyrosine kinase 3 (FLT3) diagnosed standard-risk acute promyelocytic leukemia in elderly gene in acute promyelocytic leukemia. Cancer. 2003;98: patients treated with arsenic trioxide requires major reduction of 1206–16. chemotherapy: a report by the French Belgian Swiss APL group 43. Lucena-Araujo AR, Kim HT, Jacomo RH, et al. Internal tandem (APL 2006 trial). Haematologica. 2018;103:e519–21. duplication of the FLT3 gene confers poor overall survival in 25. Gore SD, Gojo I, Sekeres MA, et al. Single cycle of arsenic patients with acute promyelocytic leukemia treated with all-trans trioxide-based consolidation chemotherapy spares anthracycline retinoic acid and anthracycline-based chemotherapy: an Interna- exposure in the primary management of acute promyelocytic tional Consortium on Acute Promyelocytic Leukemia study. Ann leukemia. J Clin Oncol. 2010;28:1047–53. Hematol. 2014;93:2001–10. 26. Schemper M, Smith TL. A note on quantifying follow-up in 44. De Botton S, Chevret S, Sanz M, et al. Additional chromosomal studies of failure time. Control Clin Trials. 1996;17:343–6. abnormalities in patients with acute promyelocytic leukaemia 27. Kaplan E, Meier P. Nonparametric estimation from incomplete (APL) do not confer poor prognosis: results of APL 93 trial. Br J observations. J Am Stat Assoc. 1958;53:457–81. Haematol. 2000;111:801–6. 28. Gray RJ. A class of k-sample tests for comparing the cumulative 45. Cervera J, Montesinos P, Hernández-Rivas JM, et al. Additional incidence of a competing risk. Ann Stat. 1988;168:1141–54. chromosome abnormalities in patients with acute promyelocytic 29. Cox DR. Regression models and life tables (with discussion). J R leukemia treated with all-trans retinoic acid and chemotherapy. Stat Soc. 1972;34:187–220. Haematologica. 2010;95:424–31. 30. R Development Core Team. R: a language and environment for 46. Pantic M, Novak A, Marisavljevic D, et al. Additional chromo- statistical computing. Vienna: R Foundation for Statistical Com- some aberrations in acute promyelocytic leukemia: characteristics puting; 2014. and prognostic influence. Med Oncol. 2000;17:307–13. 31. Sanz MA, Grimwade D, Tallman MS, et al. Management of acute 47. Poiré X, Moser BK, Gallagher RE, et al. Arsenic trioxide in front- promyelocytic leukemia: recommendations from an expert panel on line therapy of acute promyelocytic leukemia (C9710): prognostic behalf of the European LeukemiaNet. Blood. 2009;113:1875–91. significance of FLT3 mutations and complex karyotype. Leuk 32. Jillella AP, Kota VK. The global problem of early deaths in acute Lymphoma. 2014;55:1523–32. promyelocytic leukemia: a strategy to decrease induction mortality 48. Cicconi L, Divona M, Ciardi C, et al. PML-RARα kinetics and in the most curable leukemia. Blood Rev. 2018;32:89–95. impact of FLT3-ITD mutations in newly diagnosed acute pro- 33. Martínez-Cuadrón D, Montesinos P, Vellenga E, et al. Long-term myelocytic leukaemia treated with ATRA and ATO or ATRA and outcome of older patients with newly diagnosed de novo acute chemotherapy. Leukemia. 2016;30:1987–92. promyelocytic leukemia treated with ATRA plus anthracycline- 49. Sanz MA, Vellenga E, Rayón C, et al. All-trans retinoic acid and based therapy. Leukemia. 2018;32:21–9. anthracycline monochemotherapy for the treatment of elderly 34. de la Serna J, Montesinos P, Vellenga E, et al. Causes and patients with acute promyelocytic leukemia. Blood. 2004;104: prognostic factors of remission induction failure in patients with 3490–3. acute promyelocytic leukemia treated with all-trans retinoic acid 50. Lengfelder E, Hanfstein B, Haferlach C, et al. Outcome of elderly and idarubicin. Blood. 2008;11:3395–4302. patients with acute promyelocytic leukemia: results of the German 35. Powell BL, Moser B, Stock W, et al. Arsenic trioxide improves Acute Myeloid Leukemia Cooperative Group. Ann Hematol. event-free and overall survival for adults with acute promyelocytic 2013;92:41–52. Outcome of older (≥70 years) APL patients frontline treated with or without arsenic trioxide—an. . . 2341 Affiliations Sabine Kayser1,2,3 Ramy Rahmé4 David Martínez-Cuadrón5,6 Gabriel Ghiaur7 Xavier Thomas8 Marta Sobas9 ● ● ● ● ● ● Agnes Guerci-Bresler10 Ana Garrido11 Arnaud Pigneux12 Cristina Gil13 Emmanuel Raffoux14 Mar Tormo15 ● ● ● ● ● ● Norbert Vey16 Javier de la Serna17 Olga Salamero18 Eva Lengfelder19 Mark J. Levis 7 Pierre Fenaux4 ● ● ● ● ● ● Miguel A. Sanz5,6 Uwe Platzbecker1 Richard F. Schlenk3,20 Lionel Adès4 Pau Montesinos 5,6 ● ● ● ● 1 11 Medical Clinic and Policlinic I, Hematology and Cellular Therapy, Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de University Hospital Leipzig, Leipzig, Germany Barcelona, Barcelona, Spain 2 12 German Cancer Research Center (DKFZ), Heidelberg, Germany Department of Hematology, Bordeaux University Hospital, 3 Bordeaux, France Department of Internal Medicine V, Heidelberg University 13 Hospital, Heidelberg, Germany Hospital General, Alicante, Spain 4 14 Hôpital Saint Louis, Université Paris Diderot, Paris, France Hôpital Saint Louis, Université Paris Diderot, Paris, France 5 15 Hematology Department, Hospital Universitari i Politècnic, La Fe, Hematology Department, Hospital Clínico Universitario, Avinguda Fernando Abril Martorell, 106, 46026 València, Spain INCLIVA Research Institute, University of Valencia, 6 Valencia, Spain CIBERONC, Instituto Carlos III, Madrid, Spain 16 7 Institut Paoli Calmette, Marseille, France Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins 17 University, Baltimore, MD, USA Hospital 12 de Octubre, Madrid, Spain 8 18 Hospices Civils de Lyon, Centre Hospitalier Lyon-Sud, Hospital Universitario Vall d´Hebron, Barcelona, Spain Pierre Bénite, Lyon, France 19 Department of Hematology and Oncology, University Hospital 9 Department of Hematology, Blood Neoplasms and Bone Marrow Mannheim, Mannheim, Germany Transplantation, Wroclaw Medical University, Wroclaw, Poland 20 NCT Trial Center, National Center for Tumor Diseases, German 10 Department of Hematology, Nancy University Hospital, Cancer Research Center and Heidelberg University Hospital, Nancy, France Heidelberg, Germany