Acute myeloid leukemia: challenges for diagnosis and treatment in Latin America

ABSTRACT Objective to review the current diagnostic and therapeutic landscape of AML in Latin America as a reflection of other low- and middle-income countries and regions of the world. Encompassing both acute promyelocytic and non-promyelocytic disease types. Methods We reviewed the literature and study registries concerning epidemiological features of patients with AML/APL treated in Latin America, as well as evaluated diagnostic and genetic stratification and patient fitness assessment challenges, the importance of early mortality and supportive care capacity, intensive and non-intensive chemotherapy alternatives, consolidation, and maintenance strategies including novel agents and hematopoietic stem cell transplantation. Results Although most of the current technologies and treatment options are available in the region, a significant fraction of patients have only limited access to them. In addition, mortality in the first weeks from diagnosis is higher in the region compared to developed countries. Conclusions Disparities in access to technologies, supportive care capacity, and availability of novel agents and HSCT hinder results in our region, reflecting barriers common to other LMICs. Recent developments in the diagnosis and treatment of this disease must be implemented through education, collaborative clinical research, and advocacy to improve outcomes.


Introduction
Acute myeloid leukemia (AML), a neoplasm derived from the malignant transformation and expansion of myeloid precursors, is common around the world.Across the age spectrum, older adults are the most frequent group affected, with a median age at diagnosis of 65 years in North America and Europe [1][2][3].Epidemiological differences in Latin-America (LA) compared to high-income countries (HICs) have been reported.For instance, a younger age at diagnosis ranging from 40-47 years due to our overall younger population, and a higher prevalence of acute promyelocytic leukemia (APL) in comparison to non-APL subtypes [4][5][6].Furthermore, differences in access to treatment alternatives for this disease changes the way AML should be approached according to geographic regions and socioeconomic contexts.In this review, we will summarize the diagnosis and treatment of AML through the lens of hematologists working in LA as a reflection of other low and middle-income countries (LMICs) and disadvantaged populations in HICs and focus on specific challenges and their potential solutions faced in these regions of the world.

Diagnosis
The diagnosis of AML is challenging considering the need for specialized diagnostic tools.Including flow cytometry, cytogenetic and molecular assessments which are required to properly classify, risk-stratify, and use targeted therapies [7].Current recommendations include the use of flow cytometry to confirm the diagnosis and establish the immunophenotype with at least 8 colors which are not always available in LA.The Brazilian Group of Flow Cytometry recognized this limitation and considering that most laboratories were working with 4-colour immunophenotyping, published locally adapted recommendations [8].For the molecular classification of AML, the European LeukemiaNet(ELN) 2017 guidelines recommended the analysis of at least 8 genes in addition to cytogenetics for a comprehensive assessment [7].In the 2022 update, another 8havebeen incorporated and required to establish the diagnosis with 20 additional genes recommended to be tested [9].While in HICs the availability of comprehensive genetic testing through high throughput sequencing has been implemented faster than the evidence to apply it, in LA a different reality is palpable.In Mexico, even conventional cytogenetic assessment has been difficult to standardize.In 3 different AML cohorts from 2010 to 2018, a successful karyotype was analyzed in 46-61% of cases [10][11][12].In the largest registry study in the public setting, karyotype data was available in 69% at baseline, withFLT3 and NPM1 status available only in 12% and 8% of cases, respectively [4].Lack of education, training, availability, and coverage limit the conduct of these expensive tests, which caneven surpass the costs of treatment.This issue raises the question of what therapy should we recommend a patient that has not been appropriately studied according to current standards?Today for any given AML patient in LA that is most likely the case.Locally performed analyzes, training and accessible reagents are urgently needed to improve our diagnostic capacity, avoid over and under-treatment and optimize existing resources.Despite these barriers, available evidence suggest most abnormalities are of a similar incidence in our population than others, including core binding factor [13], FLT3and NPM1 mutations [14][15][16].

Remission induction, supportive measures, and early mortality
The aim of an intensive induction regimen is to eliminate the large burden of leukemic blasts and to reestablish normal hematopoiesis.Most patients with AML in LA will be candidates for this treatment strategy.Traditionally 7 days of cytarabine plus 3 days of daunorubicin (7 + 3) have been an international standard for over 4 decades and remains so in most of LA [17].This strategy is successful in achieving complete responses (CR) in 40-60% of older adults and 60-80% of younger patients varying according to genetic risk categories.Higher doses of daunorubicin (90 mg/m2 for 3 days) have shown to improve survival outcomes in comparison to 45 mg/m 2 up to 65 years [18,19], whereas the dose of 60 mg/m 2 is also commonly used and potentially of equivalent benefit and efficacy, similarly to idarubicin at 12 mg/m 2 [20,21].An addition of a third agent to 7 + 3 or the use higher doses of cytarabine have been compared but not consistently proven to improve survival on all patients [22,23], although the latter are still favored for patients with core binding factor AML including inv16, t(16;16) and t(8;21) with some groups advocating for the use of fludarabine, higher doses of cytrabaine, and filgrastim (FLAG)in this population [24,25].Moreso than adjustments in 7 + 3, historical improvement in outcomes in intensively treated AML patients in HICs have occurred due to a steady improvement in the delivery of supportive care [26][27][28].Administered at the time with the highest disease burden, intensive chemotherapy comes with significant risks; severe neutropenia and thrombocytopenia are nearly universal, with transfusion dependence and a high incidence of chemotherapy-induced mucositis.Early mortality, ranges between 3 and 6% in modern clinical trials [20,24,26].However, when we look at the implementation of induction chemotherapy in LMICs, increased rates of induction mortality are common and reported to be15% in India up to 41% among older adults in Brazil, also higher than historical rates outside of the clinical trial setting in HICs [28][29][30][31].White blood cell counts as a measure of disease burden correlate with induction mortality [32], particularly patients with hyperleukocytosis 100 × 10 9 /L have early death rates that can range from 30 to 50% in patients with leukostasis [32][33][34][35].Therefore, establishing urgent treatment measures for patients with highly elevated white blood cell counts is imperative.Interventions reported include tumor lysis prophylaxis cytoreduction with hydroxyurea, an early initiation of chemotherapy and leukoapheresis for patients with leukostasis.There are no data to support leukapheresis over other measures and this procedure is complex, expensive, and less accessible than rapid pharmacologic treatment [36].Delaying treatment until genomic data is available is a new trend which should be taken with caution, and it is important to note that the use of cytoreduction does not necessarily limit the acquisition of genomic data [37].Chemotherapy shortages due to poor governmental leadership and administration is a complex issue faced in several countries in our region [38,39] which complicates patient management at the bedside and forces us to use be creative and substitute drugs for what is available with a negative impact in outcomes.Other non-modifiable patient-related factors associated with early mortality include age and functional status, comorbidities, cytogenetic risk, insurance access, socioeconomic status, and social support [27,29,33].
Infections are the most common cause of early death for AML patients with bacterial and invasive fungal infections common culprits.While fluoroquinolones are recommended, widespread resistance begs the question of whether we should continue to use antibacterial prophylaxis at all [40].New problems that have risen include the high rates of gram-negative carbapenem-resistant organisms in other regions of the globe [29].Early de-escalation after resolution of fever for 48-72 h can reduce antibiotic-related adverse events and selection for drug-resistant organisms without compromising outcomes [41].Central venous lines should be handled by trained nurses and access to cultures should be immediate.Using recommended mold-active agents such as posaconazole and isavuconazole can be expensive and inaccessible, and one must rely on earlier generation azoles such as voriconazole or itraconazole which can be limited by toxicities, absorption issues and pharmacologic interactions [42].
Transfusion support is key and the availability of plateletphereses can be challenging, as less than half of the blood supply in LA comes from altruistic donors [43].Optimization of blood products through patient blood management programs is highly important [44].Avoiding sibling and family directed donation becomes even more relevant, as the generation of anti-HLA donor-specific antibodies can become a barrier for haploidentical hematopoietic stem cell transplantation (HSCT) [45].Thus, characteristics of the treatment center itself are of key importance and can improve or limit the capacity to deliver supportive care; induction mortality is inversely correlated with center specialization and experience [27,30,46].ICU capacity and access, nursing, and quality management, establishment of outreach programs for timely referral to leukemia centers, and a multidisciplinary treatment team with continued education are all relevant factors that should be increasingly fostered and developed in our region [46].Implementation of practices known to be successful is challenging but can likely save more lives worldwide than any diagnostic study or novel agent addition to the current standard management [41].Furthermore, novel induction strategies such as early discharge and outpatient follow-up can decrease exposures, costs, and improve quality of life [47].This strategy has been adopted successfully in the context of allogeneic HSCT in Mexico even in centers without a conventional transplant unit [48].

Novel agents in intensive therapy
Several additions to 7 + 3 have demonstrated to improve outcomes.Midostaurin is a multikinase inhibitor that targets the FLT3pathway and was studied in the randomized, placebo controlled RATIFY CALGB 10603 trial where 717 patients <60 years were included.Patients in the experimental arm received midostaurin 50 mg PO BID on days 8-21 of induction and consolidation; a 7% improvement in 4-year OS and event-free survival (EFS) were obtained in comparison to placebo [49].It is recommended that midostaurin is started no later day 8, but this can be challenging where there is no access to FLT3 testing in-house.Midostaurin is associated with increased rates of anemia, rash, nausea, and QTc interval prolongation.It is available in several countries in LA (Figure 1).Sorafenib is another first-generation multi-kinaseFLT3 inhibitor that has been studied in phase 3 trials for AML.The SORMAIN study planned to accrue 200 patients but was terminated early due to poor enrollment; 83 patients with FLT3-ITD AML in CR after transplant were randomized to sorafenib maintenance at 200-400 mg BID for up to 2 years and compared to.placebo.In this study, the primary endpoint of relapsefree survival (RFS) favored the sorafenib arm with a 31.7%improvement at 2 years translating to an OS benefit of 24.3%, with similar results reproduced in an open-label trial [50].Sorafenib is the FLT3 inhibitor most widely available in LA (Figure 1) due to its original indication in hepatocellular, renal, and thyroid cell carcinomas, but its real-world use in AML in our region is unknown.It is limited by gastrointestinal adverse events, fatigue, and infections.The use of the second-generation FLT3 inhibitor quizartinib, a second-generation FLT3 inhibitor selective to the ITD mutation has been added to 7 + 3 and compared to placebo in the QuANTUM-First trial which included 539 patients 18-75 years.In this study an OS and RFS benefit has been reported with the use of 40 mg per day on days 8-21 for up to 3 years, although it is yet to be approved for this indication [51].The use of other second-generation FLT3 inhibitorsgilteritinib and crenolanib in combination with chemotherapy vs. midostaurin and in maintenance therapy are under study.Novel agents that have been incorporated to intensive therapy in AML but remain unavailable in most countries in Latin America including gemtuzumabozogamicin, quizartinib and CPX-351 are summarized in Table 1.

Consolidation, maintenance, and long-term outcomes in LA
For adults, higher doses of single agent cytarabine for consolidation remain the standard [62].There is debate regarding the exact dose and schedule, however doses of 2-3 g/m 2 twice daily can be quite toxic, leading to severe neutropenia and transfusion dependence, and with a risk of significant organ toxicities, thus it is usually limited for patients younger than 60.Intermediate doses of cytarabine (1-1.5 g/m 2 ) seem similarly effective and less toxic and can be a preferred strategy across the age spectrum, being our standard, particularly when supportive care capacity is limited by lack of resources which has also recently been favored in the 2022 ELN update as well [7,9,24,63].This lower intensity strategy may also improve the rate of treatment abandonment that occurs after achievement of hematopoietic recovery, as financial constraints and a lower educational level can limit patients' and families' understanding of their disease and the importance of post-remission care [64].Retrospective and registry studies report a consistent decrease of approximately 10-30% in the 5-year probability of OS in LA vs. HICs during the last decade, ranging from 22 to 55% in young adults [10,65,66].Similarly, across risk categories worse outcomes have been observed for favorable risk patients with 5-year OS rates ranging from 60 to 40% and intermediate-risk of 40-20% suggesting there is room for improvement in these subgroups [4,67].The COVID19 pandemic further impaired our capacity to deliver care and increased the risk of mortality from infection due to prolonged exposure to the medical environment [68,69].The addition of more chemotherapeutic agents and subcutaneous azacytidine (Aza) to consolidation has not shown to be effective in improving survival [24,[70][71][72].Oral azacytidine has shown survival benefit in a randomized trial in patients that are ineligible for HSCT, but it is still unavailable in LA.Its use after transplantation is under study (Figure 1, Table 1)

Non-intensive therapy
For patients who are not considered candidates for intensive therapy after the comprehensive evaluation of comorbidities, functional, social, and geriatric assessments, the prior standard was hypomethylating agents continued until progression which achieve a median OS of 10 months [73].There are not many studies that focus on this population in LA.However, most patients in our region have been treated either with low dose cytarabine (LDAC) and best supportive care achieving median OS of 6-8 months and <3 months, respectively, although the latter is no longer considered an appropriate therapy for AML [74].In Mexico, most older adults receive LDAC in the public setting with a 30-day mortality of 31.4% and median OS of only 1 month [4,75].On the other hand, several recent drug approvals for AML in HICs are focused on this patient group.In Figure 2 you can see and adapted algorithm from current NCCN guidelines [76].
Venetoclax-based combinations were the first change in the standard non-intensive treatment for AML after the publication of two randomized phase 3 trials combining the BLC2 inhibitor with cytarabine and azacytidine.With the cytarabine combination, the initial OS benefit was not significant until longer follow-up revealed a median of 8.4 vs. 4.1 months when compared to placebo [77].In contrast, the combination with azacytidine was associated with a clear OS benefit (14.7 vs. 9.6 months) and became the standard of care in many countries [78].A retrospective experience on the use of venetoclax combinations included 50 patients from Mexico and Peru was recently published.Interestingly, venetoclax-based combinations were used in both, newly diagnosed and relapsed/refractory patients, combined with cytarabine in 40% of cases, and frequently in lower doses due to its combination with azoles, reflecting a need for lower-cost regimens [79].CR/CRi rates were 78.6% in newly diagnosed patients with a median OS of 9.6 months, shorter than the phase 3 trial but improved over prior regional experiences showing this therapy can be implemented in LA [79].This drug combination is particularly effective in patients with IDH2 and NPM1 mutations with 2year OS exceeding 70% [80].On the other hand, responses for FLT3 mutated patients are usually not durable, lasting less than a year [80].The use of FLT3 inhibitors in addition to venetoclax and hypomethylating agents is under study.Other drugs recently available in high-income countries that have yet to reach most of LA are IDH inhibitors and glasdegib (Table 1).
Interestingly, some countries in LA participated in the study that led to the approval of ivosidenib(6 in Brazil and 2 in Mexico) but in our region the approval of a drug is not immediately associated to increased uptake and access for most of the population and is frequently limited to patients with private insurance, leading to local disparities [59].(Figure 2).

Hematopoietic stem cell transplantation
HSCT remains one of the most potent antileukemic treatments available and is a treatment goal for all fit patients with intermediate and high-risk disease, as well as those with positive MRD and relapsed/refractory disease [81,82].Unfortunately, most transplant candidates in our region still do not have access to this potentially life-saving therapy [4].The limitations and barriers for HSCT faced in our region are discussed in another review in this series, but several important developments in AML relevant to our region should be noted.The availability of reduced intensity conditioning regimens (RIC) has allowed for more patients to receive allogeneic transplantation; age should no longer be considered a limit for undergoing the procedure and arbitrary age limits to access HSCT programs should be eliminated.This is more relevant now as the probability to achieve remission and hematopoietic recovery is improved with venetoclax-based regimens.Patients who were not previously considered fit for intensive therapy may be considered candidates for transplant and achieve favorable outcomes [83].In the latest CIBMTR analysis including 1,321 patients ≥60 years in CR1 showed comparable OS at 3 years of 49.4% in ages 60-64, 42.3% in 65-60 and 44.7% in those ≥70 years [84].
A historical paradigm in AML was to administer a myeloablative conditioning regimen (MAC) if possible, based on predicted tolerability, similarly to the decision to receive intensive therapy due to an increased risk of relapse with RIC regimens, but this has been challenged by a recent study in the era of MRD.The phase III BMT CTN 0901 trial randomized patients with AML and myelodysplastic syndrome in CR to MAC or RIC; pre-transplant bone marrow samples for MRD with high throughput sequencing were taken and analyzed post-hoc in 190 patients.In this study, MRD-positive patients had a 3-year cumulative incidence of relapse (CIR) and survival favoring MAC (67% vs. 19% and 61% vs. 43%, respectively).In contrast, MRD-negative patients had similar 3-year OS regardless of conditioning intensity (MAC 56% vs. RIC 63%) balanced due to a lower transplant-related mortality (MAC 27% vs. RIC 9%) [85].This study allows us to conclude that RIC is a reasonable approach even in patients eligible for MAC if MRD is negative, with less toxicity and similar outcomes.This is highly relevant for our region as MAC demands high quality supportive care capacity, whereas RIC regimens can be implemented as outpatients and performed even without a conventional HSCT unit [48].
Intensive chemotherapy promptly followed by a RIC regimen and cell infusion without the need for cytopenia recovery (sequential or augmented conditioning) has been studied to reduce the risk of relapse and improve on efficacy.This strategy, performed most frequently with fludarabineamsacrine, cytarabine and busulfan (FLAMSA-Bu), was studied in 244 patients with high-risk AML or MDS in a randomized fashion and compared to RIC in the FIGARO trial.No differences in 2-year OS (58.8% vs. 60.9%) or CIR (26.7% vs. 29.5%)were revealed [86].In this study, up to approximately 50% of patients with pre-transplant MRD by flow cytometry did not relapse.Therefore, although pre-transplant MRD is a poor prognostic factor it should not discourage transplantation or become an impediment for referral to a transplant center.For patients who are not in CR, an alternative double-alkylator strategy with the addition of thiotepa to busulfan and fludarabine has not proven successful [87].Taken together, these data suggest that, as conditioning regimens go, we can go lower, but we probably should not go higher.
The availability of haploidentical donors in Latin America through post-transplant cyclophosphamide has greatly increased the number of procedures that can be performed in our region as most countries in our region do not have an unrelated donor registry and face great bureaucratic and economic challenges with international unrelated donors [88].So far, results with the haploidentical donor strategy have shown to be superior to umbilical cord blood grafts in a randomized trial and comparable to unrelated donors in single-arm studies with both bone marrow or peripheral blood and across conditioning regimen intensities [89,90].The Beijing protocol based on anti-thymocyte globulin, and a primed bone marrow graft has also shown comparable results to matched sibling grafts [91].In contrast with matched donors, the presence of donor-specific antibodies is a limiting barrier, and viral reactivations are more common than with calcineurin inhibitor and methotrexate prophylaxis.In our context, we recommend considering first-line allogeneic transplants regardless of a matched or haploidentical donor for patients without proper risk stratification, and those with high-risk and intermediate-risk disease to be performed in specialized and experienced centers.
Regarding autologous transplantation, a randomized study by the HOVON-SAKK group revealed a reduced relapse rate but higher non-relapse mortality resulting in similar OS vs. a third cycle of consolidation chemotherapy in intermediate-risk patients [92].Concerns regarding the presence of residual disease and the availability of alternative donors have led to declining rates of autologous grafts [93].However, in LA autologous transplants could be efficacious and costeffective compared to repeated doses of consolidation chemotherapy as it can lead to a reduced exposure to severe neutropenia, transfusions, and hospital admissions.This strategy has been studied in the prospective GIMEMA AML1310risk-adapted trial in which patients with favorable and MRD-negative intermediate-risk patients were allocated to autologous HSCT after a single consolidation, achieving two-year OS of 74% and 79% respectively, and disease-free survival of 61% in both groups [94].In APL, HSCT is a strategy that remains a standard for consolidating relapsed disease, although based on retrospective, uncontrolled studies.It should be avoided in the first line altogether given the high probability for cure even with a regimen that does not include arsenic trioxide (ATO) [7].In this context, autologous transplantation is viewed more favorably as non-randomized comparisons have shown superior EFS and OS to allogeneic transplants in large registry studies [95][96][97] with allo- HSCT recommended only for patients who fail to achieve a second molecular remission.

Relapsed/refractory AML
Patients with relapsed or refractory AML have dismal survival.Second-line chemotherapy regimens such as FLAG IDA, HAM or MEC followed by allogeneic hematopoietic stem cell transplantation are still recommended [81].We have had success treating patients with venetoclax-based combinations in relapsed/refractory patients with a 45.5% CR/Cri rate which agrees with other experiences worldwide as no RCT is available in this context [79].Patients who were not considered candidates for intensive therapy and have relapsed or are refractory to either Aza, LDAC or venetoclax-based combinations fair the worst, with supportive care often the only remaining alternative, although this has recently changed in HICs.The use of FLT3 inhibitor monotherapy has been compared with chemotherapy in two randomized phase 3 trials in patients with the ITD mutation.Quizartinib was evaluated in the randomized QUANTUM-R trial and compared with second-line intensive and non-intensive strategies documenting an improvement in median OS of 6.2 months vs. 4.7 months, albeit this drug was not granted approval in the United States and Europe due to concerns with the quality of the data [98].Similarly, gilteritinib, has been studied in the phase 3 trial ADMIRAL trialat 120 mg vs. a mixed choice of intensive and non-intensive second-line strategies where it achieved a 3.7month median OS advantage [99].Gilteritinib is currently only available in Brazil (Figure 1) and although sorafenib is available off-label there are only phase 2 studies to support its use in post-transplantation relapse [100].Lastly both ivosidenib and enasidenib are approved in the US for relapsed/refractory AML based on phase 2 studies with ORR of 41.6% and 40.3%, respectively, with median OS of 9months and neither are yet available in LA [101,102].

Acute promyelocytic leukemia
APL is a distinct subtype of AML characterized by gene rearrangements involving the Retinoic Acid Receptor α (RARA) locus on chromosome 17 in 98% of cases with the PML gene; bcr1 breakpoints are more common in our population than non-Latino patients [103,104] with variant translocations of similar frequency than in Europe and the US [105].

Diagnosis and early management
Death within the first hours or days from the diagnosis is more frequent in APL than in other subtypes of acute leukemia [106,107] and, therefore patients suspected to have APL should be managed as a medical emergency [108].The most frequent cause of death is bleeding which is associated with an APL-associated coagulopathy, and even before the genetic confirmation of the diagnosis, all trans retinoic acid (ATRA) and measures to counteract the coagulopathy should be initiated immediately based solely on the clinical suspicion of APL and the review of the peripheral blood smear [108].The identification of the APLspecific genetic lesion is mandatory and in LA most laboratories investigate initially for the PML/RARA rearrangement which can be detected by fluorescence in situ hybridization(FISH), reverse transcriptase polymerase chain reaction (RT-PCR; or real-time quantitative PCR [RQ-PCR]) [108].The analysis of PML nuclear staining in leukemic cells using anti-PML monoclonal antibodies can be a surrogate for genetic diagnosis and has been successfully used in the ICAL study as a rapid method diagnosis [106].Conventional karyotyping is an important diagnostic method but due to its long turnaround time, it is used for confirmation of results obtained by techniques of molecular biology, to identify variant translocations and additional abnormalities.Several additional genetic abnormalities have been shown to affect the outcome of APL patients and our group has proposed an integrative score (ISAPL) based on FLT3-ITD mutational status, ΔNp73/TAp73 expression ratio, ID1, BAALC, ERG and KMT2E gene expression levels [109].The ISAPL was developed in patients treated with ATRA and chemotherapy and there is data suggestive that the presence of genetic mutations other than PML/RARA inpatients receiving arsenic trioxide (ATO) plus ATRA do not change the outcome.
APL coagulopathy is characterized by consumptive coagulation as well as activation of primary and secondary fibrinolysis due to multiple mechanisms which are the basis for the high incidence of intracerebral and pulmonary hemorrhages reported as the most frequent causes of death, both prior to and shortly after treatment initiation [110,111].The ELN recommendations for supportive measures to counteract the coagulopathy are the administration of transfusions of fibrinogen and/or cryoprecipitate, platelets, and fresh-frozen plasma immediately upon suspicion of the diagnosis, to maintain the fibrinogen concentration above 100-150 mg/dL, the platelet count above 30 × 10 9 /L to50 × 10 9 /L, and the international normalized ratio (INR)below1.5 [108].

APL treatment
Studies combining ATRA and chemotherapy have reported complete remission rates of 90-95% and rates of long-term survival in >80% of newly diagnosed APL cases [112,113].The most common protocols used in LA that are based on the use of ATRA, anthracycline, and cytarabine in a risk-stratified therapy following the experience of the PETHEMA group [106,[114][115][116]. comparison of the PETHEMA/HOVONLPA2005 and the ICAPL2006 studies which used idarubicin and daunorubicin, respectively, showed that the 2 drugs were associated with similar rates of primary resistance, molecular persistence of disease, and molecular and hematological relapse rates [106,114].In the IC-APL2006, complete hematological remission was achieved in 85% and 15% of patients died during induction.After a median follow-up of 28 months, the 2-year CIR, OS, and DFS were 4.5%, 80%, and 91%, respectively, [106,114].Lower ATRA doses of 25 mg/m 2 in combination with chemotherapy have been studied based on similar pharmacodynamic data to the standard 45 mg/m 2 dose achieving similar initial results, this strategy can be useful if resources are limited, although concerns for long-term exist [117][118][119].Despite the dramatic progress achieved in frontline therapy of APL with ATRA plus anthracyclinebased regimens, relapses still occur in approximately 20% of patients.Moreover, these regimens are associated with significant toxicities due to severe myelosuppression frequently associated with life-threatening infections and potentially serious late effects including the development of secondary myelodysplastic syndrome and AML.
Results from two phases III studies comparing the efficacy and safety of ATRA plus ATO versus ATRA plus chemotherapy have led to ATRA-ATO becoming the new standard of care for treating patients with low/intermediate-risk APL (WBC ≤ 10x 10 9 /l) [108].However, ATO is not available in most of LA (Figure 1) due to the prohibitive costs and the requirement of intravenous administration in a hospital setting, a challenge in countries with limited infrastructures.There are two most used treatment regimens with ATRA-ATO for patients with low/intermediate-risk APL, based on the APL 0406 trial (GIMEMA-SAL-AMLSG) [120,121] and the AML17 trial (UK NCRI) [122,123] studies.Both use a similar total dose of ATO; however, they have differences in the duration of treatment and the distribution of the dose of ATO during therapy.In the APL 0406, ATO was administered at a lower dose daily, while in the AML17 it was used at a higher dose administered 2 or 3 days per week.APL 0406is a phase 3, randomized, multicenter, non-inferiority study comparing ATRA-chemotherapy versus ATRA-ATO in low-intermediate-risk APL patients.After a median follow-up of 40.6 months, EFS, CIR, and 50-month OS for patients in the ATRA-ATO versus ATRA-chemotherapy arms were 97.3% vs 80%, 1.9% vs 13.9% and 99.2% vs 92.6%, respectively with significant differences [120,121].ATRA-ATO had less neutropenia and prolonged thrombocytopenia and fewer episodes of febrile neutropenia.Adverse effects of the ATRA-ATO combination consisted mainly of frequent increase in liver enzymes, QTc interval prolongation, and hyperleukocytosis [120,121].The AML17 trial showed that the combination of ATO-ATRA was highly effective in all risk APL patients with a high cure rate and less relapse than AIDA treatment [122,123].This study confirms the applicability of this approach in low/intermediate-risk patients and suggested the feasibility of minimizing chemotherapy in high-risk patients, using 1-2 doses of GO [122,123].
For high-risk patients, the use of ATRA-ATO plus chemotherapy has not yet been shown to be superior to conventional treatment with ATRA plus chemotherapy in a randomized trial, although studies with historical control, such as APML4 [124], have shown a benefit in overall survival.A large, randomized trial involving most European cooperative groups (APOLLO trial, NCT02688140) is being conducted in this population (high-risk APL) comparing ATRA + ATO + 2 doses of idarubicin in induction versus ATRA + chemotherapy.Regardless of which first-line treatment is selected, CNS prophylaxis should be restricted to patients with WBC counts≥10 × 10 9 /L at presentation, or to those who have had a CNS hemorrhage.Regarding measurable residual disease (MRD) assessment, the most important MRD endpoint in APL is PCR negativity for PML-RARA at the end of consolidation [108].For patients with low/intermediate-risk APL, MRD monitoring is only recommended after completion of consolidation and may be discontinued once bone marrow MRD-negativity is achieved.For high-risk APL, MRD is recommended to be assessed by qPCR from BM every 3 months for 24 months, starting at the end of treatment or from peripheral blood every 4-6 weeks during follow-up.However, the availability of MRD assessment at the recommended frequency in LA is quite heterogeneous [125], even in the private system in some countries.Conversion of PML-RARA by PCR from undetectable to detectable, and/or a ≥ 1 log 10 increase in high-risk patients with previously stable PML/RARA levels should be regarded as imminent disease relapse in APL and must be confirmed in a second sample [108].Two independent retrospective studies demonstrated that early intervention of patients with molecular relapse provides a better outcome than treatment in hematologic relapse alone.Salvage therapy for molecular persistence after consolidation, molecular relapse, or hematologic relapse should be chosen considering the first-line treatment used previously and the duration of the first relapse.For patients relapsing after ATRA + chemotherapy is that they should be managed with ATRA + ATO-based approaches and patients relapsing after ATRA + ATO should be managed with ATRA + chemotherapy.A potential exception for crossing over to a different treatment of relapsed patients may be considered for those with late relapse.Patients who reach the second CR should receive intensification with HSCT as discussed.For patients in whom HSCT not feasible, available options include repeated cycles of ATO with or without ATRA with or without chemotherapy [108].

Future directions
Adapting existing and upcoming therapies for AML in our region should have its own development pathway different to that of HICs and be adapted to our reality.Establishment of collaborative efforts led by Latin American investigators with cooperative groups in HICs can help implement existing technologies and decision-making through clinical networking, exemplified by the experience of the ICAL [126].Opportunities for studying alternative dosing and scheduling of expensive drugs in clinical trials can increase access and decrease costs while improving outcomes.Similarly, alternative frameworks of care through early discharge and outpatient strategies can decrease the burden of AML on our health systems.Haploidentical transplants should continue to grow in regions where unrelated donors are not feasible.Timely reference, clinician education and advocacy for access to effective therapies through grassroots efforts including all stakeholders are critical to improving our limited treatment arsenal, which should be guided by costefficacy analyzes and critical thinking.

Conclusions
AML represents a great challenge to all stakeholders in LA.Disparities in access to technologies, supportive care capacity, and availability of novel agents and HSCT hinder results in our region, reflecting barriers common to other LMICs.Recent developments in the diagnosis and treatment of this disease must be implemented through education, collaborative clinical research, and advocacy to improve outcomes.

Figure 1 .
Figure 1.Availability of novel therapies for acute myeloid leukemia across selected countries in Latin America.Dark-shaded areas refer to partial availability.Light shaded areas refer to universal availability.White areas refer to unavailability.

Figure 2 .
Figure 2. Adapted treatment algorithm for patients with acute myeloid leukemia who judged not to be candidates for intensive chemotherapy.

Table 1 .
Novel agents approved for the treatment of AML that are unavailable in Latin America.