Estey 2020

Acute myeloid leukemia: 2021 update on risk-stratification and management
Elihu H Estey, MD1,2
1. Division of Hematology, University of Washington, Seattle WA USA

2. Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle WA USA
Correspondence Author:
Elihu H Estey, MD
Division of Hematology, University of Washington and Clinical Research
Division, Fred Hutchinson Cancer Research Center
825 Eastlake Ave E, #CE3-300, Seattle WA, USA
E eestey@uw.edu
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
10.1002/ajh.25975
This article is protected by copyright. All rights reserved.

Funding Information: None
Conflict of Interest: No conflict
When first seeing a newly- diagnosed patient physicians must decide whether to offer
conventional treatment or investigational therapy, the latter preferably in the context of a clinical
trial. As noted below, such trials have led to changes in what today is considered “conventional”
therapy compared to even 1-2 years ago. In older patients decision making has often included
inquiring whether specific anti-AML therapy should be offered at all, rather than focusing on a
purely palliative approach emphasizing transfusion and antibiotic support, with involvement of a
palliative care specialist.
In principle these decisions should be based on the comparative ratios of benefit to risk with
conventional, investigational, or purely palliative approaches. However, as implied by the word
“trial”, benefit/risk ratios of a given trial are generally unknown. Accordingly, a principal reason to
enroll on a trial is dissatisfaction with the benefit/ risk ratios associated with conventional and
purely palliative approaches.
Physicians might have varying opinions as what constitutes benefit and what constitutes risk. The
same is true of patients. Longer survival would likely qualify as a benefit but so might achievement
of a prolonged remission since it is plausibly associated with longer survival, reduced frequency
of transfusions, and better “quality of life” (QOL). The potential benefits, relative to conventional
CR, of newer less-stringently defined categories of remission, such as CRp, CRi, CRh, and of the
more-stringently defined CR without measurable residual disease (MRD) need to be accounted
for. Treatment-related mortality (TRM) resulting in death from treatment before expected death
from disease is an obvious risk but so are various toxicities of treatment and a decreased QOL.
Indeed, a major benefit of many of the recently- FDA approved therapies for AML is their putative
|Page

association with less TRM, fewer toxicities, and better QOL. In this presentation although referring
to QOL, I will emphasize survival, (including TRM), event-free and relapse-free survival and
remission because these have, previously at least, been easiest to quantify. I will also discuss the
recently the promulgated response categories CRp, CRi, CRh in relation to CR and will
emphasize the importance of MRD.
Purely palliative approach

In the United States this been the most common strategy in patients aged 65 or greater, thus
suggesting American physicians have often felt the risks of treatment outweigh its benefits.
Three large US studies[1-3] involving 4058,5480, and 8336 patients age 65 or greater (median age
75-80), and based on Surveillance, Epidemiology, and Endpoints Registry (SEER) data linked to
Medicare claims, have reported only 39-43% of such patients received induction chemotherapy.
In each study survival was longer in those receiving chemotherapy, with median survivals of 1-2
months for untreated vs. 5-7 months for treated patients. Conclusions remained unchanged after
accounting for age and co-morbidities[2]. Superior survival following intensive treatment (largely
conventional 7+3) in patients aged 70-79 regardless of performance status has also been
reported by the Swedish AML Registry[4], with a suggestion of less early death even in patients
with performance status 2-4 (figure 1). In Sweden the proportion of intensively treated older
patients varies by geographic region; the national rate is 50-55%, but in some regions this rate
approaches 75%. Since in such regions more relatively unfit patients might be expected to receive
intensive chemotherapy, survival following intense treatment might be expected to be less in
these regions. However, such has not been the case, with survival being longer in regions where
a higher proportion of patients receive intense induction.[4, 5].
Physicians may be able to identify factors that although difficult to quantify (“I know it when I see
it”) motivated them to recommend treatment rather than purely palliation and were themselves
responsible for the longer survival in treated older patients. Randomization is a means to address
this possibility. Dombret et al.[6] asked physicians of 488 patients ≥ age 65 with newly-diagnosed
AML, >30% blasts, and performance status 0-2 whether they would typically assign these
patients to best supportive care (BSC) only, low-dose cytarabine(LDAC), or 7+3, which could
|Page

include daunorubicin at a now generally regarded inadequate dose of 45m/m2 daily X3;
collectively these were referred to “conventional care regimens”(CCR). Patients were next
randomized 1:1 between the selected CCR and azacytidine at the standard dose of 75 mg/m2
daily X 7; 64% of the CCR patients received LDAC, and 18% BSC and 18% 7+3. Median follow-
up of patients remaining alive was 24 months. The risk of death in the azacytidine arm was 0.85
that in the CCR arm (95% CI 0.69-1.03, median survivals were 10.4 vs. 6.5 months) Several “post-
hoc” analyses accounting for pre-treatment covariates, e.g. age, cytogenetics, performance
status, subsequent treatment, and continent of treatment, confirmed the superiority of azacytidine
vs. CCR. Reflecting the number of patients in each CCR arm, this superiority was most obvious
for LDAC. 1-year survival probabilities were 30% for azacytidine vs. 19% for BSC (95% CI for the
difference -6 to +30%) with differences least obvious for azacytidine vs. 7+3 (1-year survival
probabilities 56% vs 51% ,95% CI for the difference -16 to +26%). Patients receiving red cell or
platelet transfusions at enrollment were more likely to become transfusion independent with
azacytidine than CCR. Toxicities, particularly those leading to hospitalization were similarly
frequent, with TRM rates for azacytidine of 7% at day 30 and 16% at day 60 vs. 10% and 18% for
CCR. Kantarjian et al. reported analogous results comparing decitabine 20 mg/m2 daily X5 to
BSC or LDAC, although the small number (29) of BSC patients precluded a better comparison
with decitabine[7].
Although these randomized trials can be criticized for insufficient patients in the purely palliative
(BSC) arm or exclusion of patients with poor performance status, the preponderance of evidence
suggests little reason to offer a purely palliative approach to as many patients age ≥ 65 with newly-
diagnosed AML as has been the practice in the past. This is even more so today given improved
supportive care, especially better antifungal therapy[8], whose importance is magnified following
administration of, particularly, intensive chemotherapy. These developments have led to declines
in TRM rates following such therapy[9]. They may play a role in improvements in survival observed
over the 1997-2016 period (8 years median follow-up) in Sweden despite the introduction of no
new therapies, although for unclear reasons these improvements were limited to men aged 50-
75 years[10] (figure 2). A Danish population based study (3825 patients) accounting for the
increasing age of AML patients over the 2000-2015 period found improved 2- year survival rates
|Page

in men and women aged 60-75 given intensive induction during this time[11] (figure 3). Reporting
on 1,493 patients age ≥ 60 the Spanish national group (PETHEMA) group also noted improved
survival following intensive therapy between 2007-2013 vs. 1999-2006[12]. The increasing
proportion of patients age >60-65 receiving allogeneic hematopoietic cell transplant and their
better outcomes also need to be considered as discussed below.
New induction therapies that should now be considered conventional/standard (table 1)

Acknowledging their importance, the improvements described above obtained with conventional
induction therapies such as 7+3, azacytidine or decitabine, have nonetheless been nominal,
prompting the introduction of new induction therapies. Principal among these are venetoclax,
CPX, midostaurin, gemtuzumab ozogamicin (GO) and the IDH 1 inhibitor ivosidenib. CPX and
potentially ivosidenib as a single agent and the others in combinations can now be considered
standard therapy in many newly-diagnosed patients.
Venetoclax – Despite remaining issues, this BCL-2 inhibitor, when combined with azacytidine (as
approved by FDA in 2018), should be considered, rather than purely palliative care, or single-
agent azacytidine or decitabine, conventional management in many newly-diagnosed older
patients and, accordingly, the standard to which future therapies should be compared in these
patients.
FDA approval of venetoclax together with either azacytidine or decitabine (“hypomethylating
agents”, HMA) or with low-dose cytarabine (LDAC) was granted for newly-diagnosed patients age
≥ 75 or with co-morbidities “precluding use of intensive induction therapy”. The trials leading to
approval[13, 14]
were single-arm and, given the infrequency with which results seen in such
“promising” trials are confirmed in randomized trials[15], led to randomized trials comparing
azacytidine +/- venetoclax and LDAC +/- venetoclax. As reported at the 2020 European
Hematology Association meeting[16], the former had enrolled 431 patients (median age 76, range
49-91), with similar eligibility criteria as in the approval study[13], and with 2:1 double-blind
randomization between azacytidine + venetoclax or azacytidine + placebo. Azacytidine was
given as 75mg/m2 daily X 7 every 28 days and venetoclax as 400 mg daily on days 1-28 after a
3-day “ramp-up” in cycle 1, leading to an incidence of tumor-lysis syndrome of only 1%. With a
|Page

median follow-up of 20.2 months the risk of death in the venetoclax arm was 2/3s that in the
placebo arm (HR 0.66, 95% CI 0.52-0.85, p < 0.001, medians 14.7 vs. 9.6 months, figure 4); 30-
day TRM rates were 7% and 6%.EFS was also longer in the venetoclax arm (p< 0.001, medians
of 9.8 vs. 7 months. CR + CRi rates were 66% vs. 28% (43% vs. 7% by initiation of cycle 2) and
their median durations were 17 vs. 13 months, while CR rates were 37% vs 18%. Transfusion
“independence” for red cells was noted in 60% vs. 35% and for platelets in 69% vs. 50%; their
durations were unspecified. Higher CR + CRi rates were noted in patients with IDH1/IDH2, FLT,
NPM1 or TP53 mutations (55% vs. 0% with the latter) when given venetoclax. Use of this drug
rather than placebo seemed more likely to benefit patients with “intermediate” than “adverse”
cytogenetics: CR + CRi rates in the former group were 74% vs. 32% (venetoclax vs. placebo) and
53% vs. 23% (venetoclax vs. placebo) in the latter group.
A parallel trial randomized 211 patients (median age 76, range 36-93) considered unfit for
intensive therapy, using similar criteria as in the azacytidine +/- venetoclax study, in 2:1 double
blind fashion between LDAC + venetoclax or + placebo[17]. The venetoclax dose was 600 mg daily
in 28 -day cycles after a ramp-up period. At the time of pre-planned analysis, median follow-up in
living patients was 12 months, and the risk of death in the venetoclax arm was three-quarters that
in the placebo arm (HR 0.75, 95% CI 0.52-1.07, p=0.11, with median survivals of 7.2 vs 4.1
months ). Two subsequent (post-hoc) analyses were performed : the first after accounting for
covariates influencing survival (performance status < 2 vs ≥ 2, age < vs > 75, intermediate vs.
adverse cytogenetics, secondary vs de novo AML) yielded hazard rate 0.67 (95% CI 0.47-0.96,
p = 0.03), and the second after an additional 6 months of follow-up yielded HR (0.70, 95%CI 0.50-
0.99, p = 0.04 and median survivals of 8.4 vs. 4.1 months). Various metrics of QOL were also
improved following treatment with venetoclax + LDAC.
The longer survival seen when venetoclax was combined with azacytidine rather than LDAC might
have reflected azacytidine’s superiority to LDAC, as noted by Dombret et al[6]. It also may have
reflected the eligibility of patients for the LDAC +/- venetoclax trial if they had not responded to
azacytidine (or decitabine) when given for MDS; such patients had worse outcomes following
venetoclax + LDAC than those who had never received azacytidine or decitabine. A comparison
|Page

of venetoclax + azacytidine with venetoclax + LDAC in patients never given HMA would be
interesting but unlikely to be done. Nonetheless, assuming confirmation in a formal paper, I
believe the azacytidine +/- venetoclax trial[16] establishes venetoclax + azacytidine as the standard
low intensity therapy for patients considered unfit for intensive therapy. A question for the future
is whether some patients benefit more (and some less) from the combination. For example
patients aged ≥ 75 seemed to derive considerably more survival benefit from the combination
than younger patients. But how this relates to performance status or co-morbidities etc. in addition
to age remains unclear (Abbvie Pharmaceuticals Advisory Board July 24,2020)
CPX 351
In 2017 FDA approved this liposomal formulation of daunorubicin and cytosine arabinoside for
patients with therapy-related AML (t-AML) or with AML and “myelodysplasia related changes
(AML-MRC). Approval followed a trial randomizing (1:1) 309 patients age 60-75 between CPX
and 7+3, with daunorubicin at the now accepted minimally effective dose of 60 mg/m2 daily X
3[18]. Patients had t-AML, AML arising after myelodysplasia or CMML, or de-novo AML but with
MRC as defined by MDS-associated cytogenetic changes. PS 0-2 was required, 88% had PS 0-
1. CR rates were 37% with CPX vs. 26% for 7+3 (p=0.04) and CR+ CRi rates were 48% vs
33%(p=0.02). Event-free survival was longer with CPX (p=0.02, medians 2.5 vs 1.3 months.
Improved survival was cited as the grounds for approval. As updated at the 2020 ASCO meeting,
with a median follow-up of 61 months administration of CPX reduced the risk of death by 30%
(HR 0.7, 95% CI 0.6-0.9, and thus p < 0.05, medians 9.3 months CPX, 6.0 months 7+3)[19]. CPX
appears superior to 7+3 in AML-MRC regardless of whether defined by a history of MDS ,
including patients who have and have not received HMAs, or only by MDS-cytogenetic changes[20]
although it is unclear whether the same is true in patients whom MRC is defined only by
morphologic dysplasia. The latter appear to have a better prognosis than other AML-MRC
patients[21], possibly suggesting AML with MRC defined only by dysplasia is not truly secondary
AML. Because the annual rate of relapse or death in CR does not decline to <5% until 3 years
have elapsed from remission[22] the CPX data might be viewed as more stable than the venetoclax
data (median follow-up 20 months), although it is unknown whether the failure rates will decline
|Page

at the same rate after less intense therapy such as venetoclax as after more intense therapy such
as CPX.
These results establish CPX, and not 7+3, as the standard “intensive” therapy for older adults
with secondary AML. The updated results presented at ASCO 2020[19] also noted the risk of death
in patients in whom survival was dated from allogeneic hematopoietic cell transplant (allo HCT)
date was less in recipients of CPX than of 7+3 (HR 0.51, 95% CI 0.3-0.9). This HR was also less
than that (0.7) favoring CPX considering all treated patients. This suggests that CPX may be
particularly useful in patients who subsequently receive allo HCT, perhaps by being more effective
at reducing MRD than 7+3. Of course, this assumes such reduction is particularly important in
recipients of allo HCT, which is unclear. Besides its potential disproportionate value in patients
who are allo HCT candidates, other questions about CPX are its effectiveness in patients age <
60 with secondary AML and in patients with secondary AML who met eligibility criteria for the trial
leading to approval but were not included. The possibility of such “selection bias” is an important
consideration in all trials.
Midostaurin
The randomized “RATIFY” trial of Stone et al.[23] established 7+3 + the FLT 3 inhibitor midostaurin
as conventional therapy for adults age < 60 with either a FLT3 tyrosine kinase domain (TKD)
mutation or a FLT3 internal tandem duplication (ITD) regardless of the ratio of affected to
unaffected alleles(allelic ratio, AR, with a ratio > 0.5 denoting a worse prognosis). The superiority
of midostaurin was independent of allo HCT with best results seen in patients receiving allo HCT
in CR1 after receiving midostaurin[24]. Questions remaining in the 2019 update[25] were the need
for midostaurin maintenance[26] and whether the benefit of midostaurin would also be seen in
adults age > 60. The German AMLSG addressed these questions by administering
7+3+midostaurin to 86 patients aged 61-70 and 198 patients aged 18-60 with a FLT3 ITD[27]. They
intended all patients entering CR or CRi to receive allo HCT followed by 12 months of midostaurin
maintenance. The CR/CRi rate was 76% and was similar in older and younger patients and
unaffected by FLT 3 AR < vs > 0.5. Allo HCT was performed in 58% of those entering CR with
the remaining patients receiving “high-dose” cytarabine(HDAC) post-remission therapy followed
|Page

by midostaurin maintenance.55% of the allo HCT patients and 56% of the HDAC patients began
maintenance which was continued for a median of 9-10 months in both older and younger patients
with toxicity/patient’s wish the primary reasons for discontinuation in the HDAC group and relapse
in the allo HCT group. As confirmed by multivariate analyses event-free survival was similar in
younger and older patients, although survival was slightly worse in the latter. EFS, was
considerably better than seen in propensity scoring matched historical controls, including older
patients, who had not received midostaurin; survival also appeared better after midostaurin.
(figure 5). Best results were seen in patients, including those age 61-70, who received HCT in
CR1, particularly if they received midostaurin maintenance. Hence, although not randomized, the
AMLSG results suggest patients age 61-70 benefit from midostaurin and that the best therapy for
all patients age < 70 includes allo HCT and midostaurin maintenance.
Gemtuzumab ozogamicin (GO)

As noted in the 2019 update[25], an individual-patient meta-analysis of five randomized trials
including 3325 adults median age 58 comparing 7+3 or FLAG-ida +/- GO, found, without inter
trial heterogeneity, that although CR rates were similar ±GO, significant reductions in risk of relapse
with GO contributed to longer survival with addition of this drug[28]. These benefits were confined
to those with UK Medical Research Council-defined “favorable” cytogenetics[29] (analogous to the
[30]
ELN2017 system described below ) in whom survival probability at 6 years was 55% without vs
76% with GO(odds ratio 0.47, 95% confidence interval 0.31-0.73) and with intermediate
cytogenetics (34% vs 39%, odds ratio 0.84, 95% CI 0.75-0.95). Administration of 3 mg/m2 GO on
days 1, 4, and 7 of chemotherapy was associated with the least toxicity and equal benefit. These
data prompted FDA approval of GO for adults with CD33 positive AML, GO consisting of an anti
CD33 antibody attached to a toxin(caleacheamycin). GO should routinely be combined with 7 + 3
or FLAG-ida in those with ELN favorable or intermediate risk assuming an acceptable risk of TRM,
as discussed below. In presumably less fit patients Amadori, et al compared GO used alone vs
“best supportive care” ‘only in patients over age 75, or age 61-75 with performance status 3-4, or
who declined’ intensive chemotherapy”, finding a statistically significant (P = .005) but medically less
significant survival advantage for GO (medians of 5.9 vs 3.6 months) and a CR + CRi rate of 27%
(30/111).[31]
|Page

Given the prognostic information added by incorporation of molecular data[30], Fournier et al. [32]
compared results in the 7+3+/- GO arms of their randomized study (one of the 5 included in the
above meta-analysis) according to mutational pattern as summarized in the ELN2017 guidelines
below[30] and according to individual mutations. They found addition of GO was beneficial in the
ELN2017 “favorable” group and the ELN 2017 intermediate group, but not in the ELN2017
adverse group (figure 6). They also found a benefit for addition of GO in patients with mutations
in NPM1, CEBPA, FLT3 ITD or TKD, NRAS, KRAS, and SRSF2. Many of these are considered
“signaling mutations” and their presence correlated with higher expression of CD33. The relation
between higher CD33 expression and better response to GO is not universally accepted
however[33, 34], nor is the positive predictive value for response to GO of the CC CD33 single
nucleotide polymorphism(SNP) which encodes a more complete CD33 binding site for GO[35, 36].
More recently, a SNP associated with increased efflux from AML cells of caleacheamycin , the
toxic moiety of GO has been proposed to predict resistance to GO[37].
Ivosidenib
FDA approved this IDH1 inhibitor as a single agent for IDH1 mutated AML occurring in a very
similar population for which it granted approval for venetoclax + HMA or LDAC. As recently
reported[38], at a dose of 500 mg daily given to 34 patients ,median age 77, who remained on
study for a median of 4.5 months (up to 40 months), CR rate was 30% (95% CI 16-49%) and CR
+ CRh (see below) rate was 42% (95% CI 26-61%) . 95% CIs overlapped but rates of CR (7/18
vs. 3/15) and of CR + CRh (10/18 vs. 4/15) were higher in patients who had not received HMA,
e.g. for MDS. Median time to response was 2.8 months for CR and for CRh (up to 4.6 months for
CR and 12.9 months for CRh). Median durations of response were not reached, with 62% of CR+
CRh and 78% of CR patients remaining in remission at 1 year. With a median follow-up time of 2
years, median survival was 12.6 months (95% CI 4.5-25.7). IDH1 mutations disappeared in 5/10
patients entering CR, 4/4 entering CRi, and 0/16 achieving neither. Durations of such molecular
remissions were not noted. 9 of 21 patients who were transfusion dependent pre-treatment
became transfusion independent for.at least 56 days, while 7 of 12 who were transfusion
independent at study entry remains so for at least 56 days. Consistent with the drug’s mechanism
of action a “differentiation syndrome”(DS) analogous to that which occurs following treatment of
|Page

acute promyelocytic leukemia with all-trans retinoic acid or arsenic trioxide was the major unique
toxicity (occurring in 18%, severe in 9%) but readily responsive to steroids and requiring
temporary discontinuation of only in the 3 with severe DS. Of the recent approvals listed in this
section ivosidenib is the only one yet to be evaluated in a randomized trial, for example compared
to venetoclax.
Although a randomized trial in “unfit” patients similar to those in the venetoclax and ivosidenib
studies showed longer survival patients randomized to the hedgehog pathway inhibitor glasdegib
+ LDAC vs LDAC alone[39], leading to FDA approval of glasdegib, the drug has yet to garner the
same traction, as venetoclax, and I will it omit from further discussion, noting the issues pertaining
to venetoclax discussed below are pertinent to glasdegib.
Table 1 summarizes scenarios in which the recently approved FDA drugs should be considered
“standard of care”. These conform to FDA’s bases for approval, supplemented by data published
since approval in the cases of midostaurin[27]and gemtuzumab[32]. The table also notes some
remaining issues. Primary among these is the definition of “unfitness” for intensive chemotherapy.
The issue of “fitness”

Criteria for unfitness in the venetoclax trials were either age ≥ 75 or co-morbidities, both
presumably precluding use of intensive induction. FDA approved ivosidenib and glasdegib for the
same population[40, 41]. However, assuming a principal consequence of unfitness would be TRM,
the effect of age alone on TRM by day 28 (after which TRM rates after intensive induction decline
precipitously) has been shown to be less than the effect of performance status (PS)[42]. The EHA
abstract reporting the azacytidine +/- venetoclax study [16] did not note the distribution of patients
according to PS. However, the paper motivating FDA approval of venetoclax + either azacytidine
or decitabine[13] noted only 16% of patients had PS 2 and none had PS 3-4. (4% of patients in the
LDAC +/- venetoclax trial had PS 3, although 44% had PS 2, with patients with PS 2-3 having
shorter survival than those with PS <2[17].) 3% of the 34 patients in the trial leading to approval of
ivosidenib had PS 3 and 15% had PS2[38] .Although formal “geriatric assessment” has been shown
to refine the prognostic effect of age[43], none of the venetoclax, ivosidenib, or glasdegib trials
|Page

included such assessments; the same applies to many other trials in older patients. Because the
[42, 43]
effect of age can be confounded with the effect of other covariates the 2017 European
LeukemiaNet AML guidelines[30] explicitly note “age alone should not be the only determinant of
fitness for intensive induction”. Furthermore, the proportion of patients age < 75 who had none,
1, 2 or more co-morbidities was unclear in the venetoclax and ivosidenib studies, as was the effect
of the number of co-morbidities, or their type, on TRM, survival, or toxicity. Bories et al. have
demonstrated assessment of “unfitness” is often subjective; given the same scenarios physicians
more willing to accept risk in their own lives were more likely to recommend more intense
induction[44]. 8/24 patients who did not respond to venetoclax + HMA went on to receive intensive
“salvage” therapy, suggesting they may have been fit to receive such therapy at initial diagnosis,
although median survival after intense therapy was only 2-3 months[45]. Analogous results (median
survival 3 months) were reported when venetoclax combinations were used as salvage after
failure of unspecified induction[46]
Fitness must be judged in the context of efficacy. “Favorable” cytogenetics are extraordinarily rare
in IDH-mutated AML[47]. However some patients otherwise meeting eligibility criteria but with
“favorable” risk cytogenetics[30] were likely excluded from the venetoclax and ivosidenib trials,
possibly in favor of intensive induction, implying a patient considered “unfit” for such therapy, can
be considered “fit” if the benefit/risk ratio is sufficiently high, as with favorable cytogenetics.
It cannot be overemphasized that, at least as published, the principal cause of failure in AML ,
regardless of age, and even in some patients with poor performance status 2-4, is not TRM, but
therapeutic resistance, manifested either as failure to enter initial CR despite not incurring TRM,
or relapse from CR[48, 49]. Above I noted declining TRM rates after intensive therapy and modest
improvements with such therapy over time in older patients[9-12]. If at least some patients given
venetoclax + azacytidine were fit for intensive induction, some may have benefited from same.
Observational data suggest some older patients with relatively poor AML- Comorbidity Indices
(AML-CI)[50] live longer when given more intense therapy than when given azacytidine or
decitabine, although without venetoclax[51]. Accordingly, a randomized trial comparing venetoclax
+ azacytidine with more intense induction might be of interest in patients whom objective criteria
suggest are fit for the latter. Multivariate models providing such criteria have been published and
are available online. It would be helpful to see the proportion of patients initially given HMA or
|Page

venetoclax + HMA whom would be judged unfit based on these models.[42, 50, 52, 53]. Maiti et al.
noted that 72% of 85 patients given venetoclax + 10 days of decitabine had TRM scores that
would have qualified them for intensive induction at the Fred Hutchinson Cancer Research
Center/University of Washington[42, 54]. The 28% of patients with higher TRM scores had superior
survival than a propensity-score matched historical control of 48 MD Anderson patients given
intensive induction, but 95% CI for survival in the 148 historical MD Anderson patients with lower
TRM scores who received intense induction overlapped the 95% CI in patients given venetoclax
+ 10-day decitabine[54]. Only a randomized trial can throw further light on this issue.
Although it is often assumed patients given less intense induction would have superior QOL, El-
Jawhari et al. analyzing QOL, anxiety, and depression in patients ≥ 60 during the first 24 weeks
following initiation of remission induction therapy found that QOL and anxiety generally improved
(depression stayed the same) regardless of whether intensive (anthracycline+ cytarabine) or less
intense (HMA) induction was given[55]. Of course, it could be argued the results with neither
standard intensive therapy nor venetoclax + azacytidine as a new non-intensive standard are
such that trials combining more and less intensive induction would be out of place in appropriate
patients, a topic I address below.
Conventional therapy vs. investigational therapies
Assuming the advent of therapies such as venetoclax + azacytidine or ivosidenib means fewer
patients should be offered purely palliative care, the question remains whether a given patient
should be offered conventional therapy, including those newly identified as such, or investigational
therapy on a clinical trial. The latter often requires referral to a large academic center and the time
spent there may eventually represent a significant proportion of a patient’s remaining life
expectancy. Furthermore, the trial’s outcome cannot be known a priori. It follows the main reason
to refer a patient for a trial is dissatisfaction with the results of conventional therapy. For example,
median survival in patients given azacytidine + venetoclax arm in the recently reported
randomized trial was 14.7 months vs. 9.6 months for azacytidine + placebo; at 2 years probability
of survival was 40% vs. 30%[16]. CPX afforded a 9.6 months median survival vs. 6 months for the
prior standard, 7+3[18]. The median does not give a complete account, but represents a convenient
picture of probable outcome, particularly given our limited ability to know who will do much better
|Page

or much worse than the median[56]. However, at least to date it appears that many conventional
prognostic factors, for example cytogenetics (“intermediate” vs. “adverse”) remain prognostic
when patients receive CPX[18], venetoclax + HMA[13, 16] or + LDAC[17], as does failure to respond
to HMA for prior MDS in the case of venetoclax + LDAC[17] and possibly ivosidenib[38]. Patients
with the prognostically unfavorable TP53 or ASXL1 mutations appear more likely than others to
be resistant to or have only short remissions following venetoclax + HMA (figure 7)[57].
For some patients the 5- 6- month improvement in median survival afforded by addition of
venetoclax to azacytidine or the 3-4 month improvement in median survival following CPX351
rather than 7+3 is very meaningful, allowing them for example to see a grandchild graduate
college. For other patients the improvement will be less meaningful particularly if informed that
US Social Security Administration tables[58] indicate the average 75 year old man can expect to
live another 11.4 years and the average 75 year old woman another 13.1 years in the absence of
AML, admitting that patients who develop AML may have been susceptible to other life-
threatening conditions if they had not developed AML. The point is the choice between
conventional and investigational therapy is largely subjective, particularly because patients may
vary in what they value as benefit (e.g. survival vs. QOL) and the amount of risk they are willing
to accept to achieve a benefit[59]. As noted by Schiffer[60], the “alternative treatment” section of
typical written informed consent documents for clinical trials, while listing alternatives to the trial
such as conventional treatment or purely palliation, usually give very little specific information
about what might be expected with these. Provision of such information is thus incumbent on the
treating physician.
ELN 2017 prognostic system to assess expected prognosis with conventional therapy:
general remarks about “prognostic systems”
ELN 2017 provides a useful guide to aid in the choice between conventional and investigational
therapy[30]. Patients are divided into “favorable”, “intermediate” and “adverse” groups. (table 2).
The German AMLSG recently validated the prognostic significance of ELN2017 for remission,
survival, and relapse-free survival (figure 8)[61]. They also noted the particularly poor prognosis of
patients with TP53 mutations (TP53 is located on the short (p) arm of chromosome 17) or complex
|Page

cytogenetics, which are highly associated with each other, and showed these should be
considered a “very adverse” group, as distinct from the “adverse” group. A further modification
indicated those patients in the ELN favorable group based on presence of an NPM1 mutation and
a FLT3 ITD AR < 0.5 have a favorable prognosis only in the presence of a normal karyotype;
adverse cytogenetics is associated with the same prognosis regardless of whether an NPM1
mutation is found[62].
Multivariate analyses indicate that even after accounting for its association with adverse
cytogenetics and thus disproportionate representation in the ELN adverse group, older age is
itself associated with poorer relapse-free, and overall, survival[61]. SWOG and UK MRC/NCRI data
indicate patients age > 65 with the presumed “favorable” constellation of an NPM1 mutation, no
FLT3 ITD, and normal cytogenetics have inferior survival and RFS survival than patients age 55-
65 and < 55 with the same findings[63].
Neither ELN2017[30] nor its modifications[61-64] incorporate data from patients treated with the new
therapies listed in table 1. However above I noted several examples (cytogenetics, TP53
mutations, prior HMA) where patients expected to do relatively poorly with previous conventional
therapies (e.g. 7+3) also did relatively poorly with newer conventional therapies (e.g. venetoclax
+ azacytidine, CPX351, ivosidenib). This suggests systems like ELN2017 are likely to be useful
in forecasting outcome even in patients given these newer therapies.
An important question is the accuracy of ELN 2017 and other prognostic systems. One way to
quantify accuracy is via areas under receiver operating characteristic curves (AUC). An AUC
value of 1.0 indicates every time a model predicts a given patient will fare better than another
patient the forecast is correct, although the amount of improvement is not specified. In contrast
an AUC value of 0.5 indicates no predictive ability (a coin flip). Although AUC values are
uncommonly reported, when reported the predictive value of most models has been found to be
approximately intermediate between certainty and a coin flip (AUC 0.70-0.75), which is a much
poorer performance than suggested by p-values often << 0.05[56].
|Page

Current prognostic systems are undoubtedly too simplistic. For example, it has been shown that
beyond the interaction between NPM1 mutations and FLT3ITDs, the prognostic significance of
one mutation can depend on the presence of other mutations[64]. Efforts to quantify the roles of
epigenetic (e.g. methylation), gene expression, or proteomic profiling in informing prognosis have
been ongoing, with further complexity introduced by the possibility study of single cells, for
example AML “stem cells”, rather than bulk populations, will be particularly informative[65-69]. Below
I discuss the value of MRD assessment in informing prognosis once response to initial induction
is known.
“Can I wait until cytogenetics and mutational data become available?”

AML has often been assumed to require urgent treatment, making it impossible to wait the
perhaps 10-14 days needed for this information to be in hand. However, examining 599 patients
with a median time from diagnosis to initiation of intense treatment (TDT) of 8 days, Bertoli et al.
reported TDT had no effect on survival even in patients presenting with WBC > 50,000 or age
>60[70].These results were confirmed by Röllig et al. in a cohort of 2200 patients with median TDT
of 3 days[71](figure 9).. Although these results could simply reflect inherently better prognoses in
patients with longer TDTs, they suggest the risk in awaiting results of cytogenetic and mutational
analyses is less than the risk in beginning therapy before results are available, thus potentially
administering conventional therapy to patients in whom it is unlikely to be effective. Although
details were not provided and the same biases apply as with analyses of TDT, the authors noted
administration of hydroxyurea was without effect on outcome[71].
Recommendations for initial induction therapy (table 3)

As discussed above, subjective considerations necessarily influence the choice between
conventional/standard therapy or a clinical trial. Although a clinical trial is not necessarily
inappropriate in any ELN2017 prognostic group since results are suboptimal even in the ELN
favorable group for patients age >65[63], the imperative for a trial becomes greater in the ELN
intermediate and, particularly, the ELN adverse group. The decision whether a patient should
receive more vs. less intense induction depends on several factors, probably most importantly
the risk of TRM. Several validated multivariate models are available to quantify this risk[42, 50, 52, 53],
|Page

which should not be based on age alone[30] and should be made while cognizant the primary
cause of treatment failure even in most older patients, at least those referred to academic centers,
has been resistance to therapy rather than TRM[48, 49]. A problem of course is many patients at
high risk of TRM are often excluded from trials (even those said to be testing therapies for unfit
patients), because of performance status or co-morbidities, with excluded patients having
demonstrably worse outcomes[72]. A trial whose eligibility criteria were essentially ineligibility for
standard trials, has demonstrated the potentially favorable benefit/risk ratio of such more inclusive
trials[73].
A still unsettled question is whether 7+3 or a regimen including higher doses of cytarabine, such
as FLAG-ida, is the preferable intense induction regimen. SWOG trial 1203 randomized 754
adults age < 60 among 7+3 (with a daunorubicin dose of 90 mg/m2), idarubicin (12 mg/m2 daily
X 3) + cytarabine 1.5 g/m2 daily by continuous infusion X 4 days (IA) and IA + vorinostat.CR rates
were 75-79% although 24% of 7+3 patients required 2 courses to enter CR vs 9-11% for the IA
treatments. Allogeneic hematopoietic cell transplant rates in CR were similar in each arm (about
48%) and there were no differences in EFS, RFS, or survival among the 3 arms, including in
patients with NPM1, FLT3, or CEBPA mutations or intermediate or adverse cytogenetics, with
patients with “favorable” cytogenetics having inferior EFS,RFS, or survival if in the IA arms[74]. The
study has been criticized because post CR doses of cytarabine were higher in the 7+3 arms
(18g/m2 vs. 2.25g/m2 on each cycle for 1-4 cycles depending on HCT availability). In contrast the
NCRI/MRC group in the United Kingdom found reduced relapse rates with FLAG-ida than with 10
days of standard cytarabine + daunorubicin.[75] However, there were no differences in survival
possibly because of higher treatment-related mortality rate in CR in the FLAG- ida arm.
Further complicating the issue is the probability a given result may not be applicable to different
subsets, recognizing a principal characteristic of AML is its heterogeneity. For example, although
a 1205 patient randomized NCRI trial found no difference, other than more TRM in the higher
dose arm, between 7 + 3 using daunorubicin doses of 90 mg/m2 vs the usual 60 mg/m2 each daily
X 3, patients with FLT3 ITD (n = 200) had less relapse leading to longer survival if given the 90
[76]
mg/m2 dose . The same may apply when deciding on intensity of induction. For example,
|Page

Montalban-Bravo et al have reported that in patients with MDS and NPM1 mutations induction
with higher doses of cytarabine, rather than HMA, was associated with longer survival and EFS[77].
Ongoing/ potential trials (table 4)
A search of the NCI website clinicaltrials.gov indicates there are currently 32 active trials for adults
with newly- diagnosed AML. Since a pharmaceutical company seems unlikely to sponsor a trial it
thought less likely to be successful than other companies’ trials, it can be inferred much
uncertainty surrounds which trial is “best”. Table 4 lists several ongoing trials in newly-diagnosed
patients; an obvious trend is to combine active agents[78-81]. The future will almost certainly see
more extensive use combinations, for example azacytidine + venetoclax + ivosidenib (or
enasidenib). This seems particularly relevant for therapies targeted at specific mutations. When
resistance to a single targeted agent becomes apparent, a mutation different than the one
predominating at initiation of therapy is often found[82], suggesting the potential benefit of
combining several targeted agents at diagnosis.
With the increasing use of targeted therapy an important question is the practicality of “precision
medicine” in AML, i.e. assigning individuals to therapy based on their mutational pattern. Above
we noted the lack of obvious influence of time from diagnosis to treatment on outcome[70, 71].
Building on these findings the ongoing “BEAT AML” trial enrolled, at 14 sites, 497 newly-
diagnosed patients age > 60, of whom 395 were eligible, with ineligibility generally reflecting an
incorrect diagnosis[83]. “Next generation sequencing” (NGS) was done with the goal of treatment
assignment (TA) to a specific sub-study, for example anti TP53 or to a “marker-negative” therapy
within 7 days of enrollment (figure 10). This goal was accomplished in 57% of the 395 patients
with minimal inter-center variation. The remaining patients either received standard therapy (7+3
or azacytidine, decitabine), a non-specific trial, or a purely palliative approach. Survival analyses
adjusted for standard prognostic covariates indicated survival was similar following receipt of a
BEAT AML or a non-specific trial (figure 10), although it was not clear how results would have
been affected had only patients on a specific BEAT AML sub-study been included.
A major source of drugs for treatment of newly-diagnosed AML are drugs initially tested in
relapsed/refractory (R/R) AML. I discuss several of the latter including bi-specific antibodies
(BiTEs) and newer FLT3 inhibitors in the section on R/R AML
|Page

Goal of induction therapy
For many years CR was considered the goal of induction therapy. This was based on work
suggesting CR was associated with longer survival with the difference in survival between patients
achieving and not achieving CR largely due to time spent in CR, suggesting CR per se contributed
significantly to longer survival[84]. The value of CR has been seen with more and less intense
induction[85].
Recent years have seen the adoption of responses less than CR, for example CRp or CRi in
which absolute neutrophil count (ANC) or platelet count respectively meet the criteria for CR, but
both ANC and platelet count do not. The category CRh in which platelet count is > 50,000 and
ANC > 500 but not meeting requirements for CR (ANC > 1,000, platelet count > 100,000) has
appeared most recently. FDA approved ivosidenib and enasidenib based on rates of CR + CRh[40,
86],
. Particularly with new drugs, CRh, and CRi/CRp are often combined into “overall response rate
(ORR)”. However, patients undoubtedly care more about the effect of a given effect on their
survival (or quality of life) than on the nomenclature affixed to the response. After accounting for
the longer time needed to achieve CRp than CR and differences in standard prognostic
covariates, Walter et al. reported CR was associated with longer relapse-free survival, and to a
less extent survival, than CRp[87]. However, CRp was associated with longer survival than
resistance (neither CR nor CRp but living at least 30, 60, or 90 days after initiation of therapy),
thus validating CRp as a clinically relevant response[87]. The same has yet to be done for CRh.
Although blood count criteria for CR are typically considered to include only ANC > 1000 and
platelet count > 100 000, Danish investigators have noted the association between continued receipt of
red cell transfusions at time of CR and increased risk of relapse[88].
It has long been known disease recurs in most patients who enter CR. This suggested such
patients still had disease present in CR. But no means were available to detect such disease.
These days various methods, discussed below, are available to discover “measurable residual
disease” (MRD). In general, CR without MRD is associated with longer remissions, RFS, and
survival than is CR with MRD. This seems to be the case in patients given more[89] or less[90]
|Page

intense induction and in patients who receive[91] or do not receive allogeneic
HCT[89] (allo HCT). The prognosis of patients who receive allo HCT with MRD, although in CR by
conventional morphologic criteria (< 5% blasts), appears approximately the same as those with
morphologic relapse and much worse than patients who are transplanted in morphologic CR but
without MRD[92]. These data prompted ELN2017 to recognize CR without MRD as a distinct
response category[30]. I believe this response should be the goal of initial induction therapy. The
ability of less vs. more intense induction approaches to produce CR without MRD is accordingly
an important, and unsettled, issue. Furthermore, responses less than CR seem more likely to be
associated with measurable residual disease (MRD) than is CR[93], emphasizing the importance
of investigating the relative contributions of CR and various lesser responses to relapse, RFS,
and survival. These contributions may differ according to intensity of initial therapy.
Post-Remission Therapy
It is broadly accepted post remission therapy is needed to prevent rapid occurrence of relapse.
Such therapy has usually involved hematopoietic cell transplant (HCT, allogeneic or autologous)
or continued chemotherapy. HCT, particularly allo HCT, is more likely to produce non-relapse
mortality and morbidity due, for example, to acute and chronic graft vs host disease (GVHD) and
development of secondary malignancies reflecting use of immunosuppressive to treat GVHD.
However, HCT, particularly allo HCT, typically reduces risk of relapse, with a graft vs. leukemia
(GVL) effect playing a major role in the allo case and leading to efforts to manipulate the
transplanted immune cells to minimize GVHD and maximize GVL.
Quantifying risks of relapse and non-relapse mortality (NRM) to inform the allo HCT
decision
The decision to recommend allo HCT is based on the presumption the reduced risk of relapse
will be greater than the increased risk of NRM and morbidities such as GVHD. Cornelissen et al.
proposed a projected > 10% improvement in RFS following allo HCT to justify the procedure[94],
although such a criterion likely varies according to individual preferences. The potential morbidity
from GVHD has been recognized by the introduction of a new endpoint (GVHD-free, relapse-free
survival [GRFS] [95].
|Page

Assessing risk of relapse entails integration of pre-induction data, particularly ELN 2017 risk
group, with assessment of post treatment MRD. MRD can be quantified by persistence of
abnormalities in (a) cytogenetics or fluorescent in situ hybridization (FISH) (b) multiparameter
flow cytometry (MFC), or (c) quantitative PCR or next generation sequencing (NGS) (table 5)[96,
97]
. Cytogenetics and FISH are the least sensitive. MFC can be measured relying on either
detection of leukemia-associated immunophenotypes (LAIP) or of cells that are different than
normal; the latter can be done without a pre-treatment specimen. The most sensitive means for
MRD detection rely on qPCR technology, which however is limited to mutations in NPM1, BCR-
ABL, or those associated with core-binding factor (CBF) AML (CBFB/MYH11 for inv 16,RUNX1-
RUNX1T1 for t(8;21)) or APL (PML-RARA. In contrast, next generation sequencing is applicable
to a wider range of mutations, e.g. FLT3 ITD or TKD, RUNX1, ASXL, or TP53 but is considerably
less sensitive. Data suggest MFC and NGS are complementary at least in adults age < 65 i.e.
those of such patients who are NGS positive although MFC negative, or NGS negative although
MFC positive, are more likely to relapse than patients who are negative for both MFC and NGS
and less likely to relapse than patients who are positive for both MFC and NGS(figure 11)[98].
Analogous findings have been reported in the post allo HCT situation[99] and for
FISH/cytogenetics vs. MFC[100]. Persistence by NGS of mutations (DNMT3a, TET2, and ASXL1,
”DTA”) associated with clonal hematopoiesis of aging are not prognostically significant[98].
Elimination of MRD based on NGS testing for TP53 may not be associated with the same
improved prognosis as elimination of MRD for other mutations[101]. Unsettled issues[96, 97] include
(a) the timing (e.g. after the first or second induction or, after “consolidation” therapy) for MRD
testing associated with the highest positive and negative predictive values for morphologic
relapse and whether this differs by MRD detection method, (b) the cut-point (e.g. > 0 vs. > 0.1)
used to distinguish a “positive” from a “negative” MFC test, (c) whether the effect of MRD is
independent of treatment, e.g. more vs. less intense induction, as suggested by some, albeit,
limited data[102], (d) whether after accounting for MRD the distinctions between CR and
responses less than CR noted above remain important and ( e) whether, unlike MFC[103], the
predictive value of NGS assessment for morphologic relapse in adults age > 65 , even those
with non DTA mutations, is similar to the value in younger patients; data presented at the 2020
|Page

AACR virtual meeting suggest this is not the case[101]. Crucially, even in patients age < 60 area
under receiver operating characteristic curves suggest the incremental prognostic ability
afforded by a single MRD determination at time of CR is limited[104], prompting efforts to develop
more sophisticated MRD technology[105, 106] and suggesting serial determinations of MRD will be
important. MRD testing lacks standardization, although the same can be said of morphologic
evaluation. Nonetheless, despite these issues MRD detection has unambiguously been shown to
have a high positive predictive value for subsequent morphologic relapse, which typically occurs
within 6-12 months of detection of MRD. Godwin et al.[107] have demonstrated high concordance
of MRD positivity as measured by MFC in peripheral blood vs. marrow, allowing more frequent
testing to detect impending morphologic relapse. Peripheral blood can also be used rather than
marrow for qPCR evaluations in NPM1- mutated[108] and inv 16 or t(8;21) AML[109].
Using Mantel-Byar methodology and landmark analyses to account for the “guarantee time” HCT
patients, but not chemotherapy only patients, must live before HCT,[110] 0stgård et al. used the
Danish Acute Leukemia Registry assuring complete data collection on virtually all 1031 adults with
intermediate—or adverse-risk cytogenetics age <70 who entered CR between 2000 and 2014, of
whom 19% received HCT in CR1[111]. HCT patients were younger with fewer co-morbidities but
more often had adverse cytogenetics; two-thirds of transplants used reduced intensity, rather
than myeloablative, conditioning. Median follow-up was 2.2 years. cumulative incidence of
relapse was lower, and survival (figure 12) and relapse-free survival longer, in HCT recipients with
either intermediate or adverse cytogenetics or age <60 or 60-69 (Figure 12). Relapses occurred
in 49% and deaths in 68% of chemotherapy patients vs 24% and 42% for HCT patients. Median
RFS was 293 days (interquartile range 142-910) for chemotherapy only vs 1067 (IQ range 397-
2696) for HCT with median survival times of 476 days (IQ 167-1589) and 1173 (IQ 474-2859).
These findings support the general recommendation to offer allo HCT to all ELN2017
intermediate and adverse risk patients. In contrast, the risk of relapse without allo HCT in the
ELN 2017 favorable subgroup is generally considered too low to justify the risk of allo HCT.
MRD assessments can refine these recommendations. Thus table 6 indicates the risk of relapse
without allo HCT in patients in the ELN 2017 favorable groups NPM1 mutated/FLT3 negative
and CBF varies substantially depending on whether there was a 3-log or greater reduction in the
|Page

relevant PCR transcripts after the first two cycles of therapy[108, 112]. Similar variation occurs in
the ELN intermediate group, but not in the ELN adverse group (table 6). Hence the minority of
ELN favorable patients who are MRD positive might be advised to assume a greater risk of NRM
or other complications post HCT than their ELN negative counterparts, assuming a reasonable
probability HCT would decrease their risk of relapse. In contrast, ELN intermediate patients who
are MRD negative should be advised to accept only a much lower risk of NRM than their MRD
positive counterparts or might be advised to postpone HCT until relapse. Even if MRD negative,
patients in the ELN adverse group should be advised to accept a relatively high risk of HCT-
related complications. It would be useful to have an equivalent of figure 12 but subdivided
according to MRD status as well as cytogenetics.
A GIMEMA group study in younger adults with newly- diagnosed AML showed the potential
value of using MRD data in deciding whether to recommend allo HCT[113]. After an induction and
a post-remission course of standard therapy patients with intermediate-risk cytogenetics were
divided into MRD (by MFC) positive and negative groups. Allo HCT was recommended in the
former, autologous HCT in the latter. The MRD positive and negative patients had similar
survival and RFS, suggesting the decision to offer the MRD positive patients allo HCT played a
role in these unexpected findings. Analogous results have been reported in favorable risk
cytogenetic patients[114].
The relative contribution of NRM to RFS is much less with chemotherapy than allo HCT[49].
Although there are alternatives, the hematopoietic cell transplantation comorbidity index (HCT-CI)[115]
effectively estimates NRM post- HCT (table 6).
Increasing use of allo HCT

Although as with induction therapy, age alone even up to 75-80 appears less important than
comorbidities in influencing post HCT NRM, [116] the proportion of adults with AML receiving HCT
(approximately 30% at any time in their course), while unaffected by age up to 70, falls
considerably once age 70 is reached.[117] Nonetheless Muffly et al. [118] have noted the increasing
use of HCT for AML patients in their 70s . Despite this increase McDonald et al. reported less
200-day NRM and less relapse-related mortality in patients receiving HCT between 2013-2017
|Page

(n=1131, 7% age 70 or above) than between 2003-2007 (n=1148, 1% age 70 or above)[119]. A
donor can now be found for “almost any subject for whom HCT is indicated”[120] and it is generally
accepted results with matched sibling and matched unrelated donors (MUD) are similar. If a
MUD donor is used a 526-patient randomized study suggest that although survival was similar
when peripheral blood (PB) or marrow was the cell source patients remaining alive at 5 years
(93 received PB, 102 marrow) had better psychological well-being, less burdensome GVHD and
were more likely to return to work if marrow was the cell source[121]. Although with declining
birthrates, the use of umbilical cords as a source of hematopoietic cells is likely to decline; when
a matched sibling donor cannot be found, some centers proceed immediately with a
haploidentical donor, as these can be found quickly, overcoming the underrepresentation of
racial minorities in donor registries. With the use of post HCT cyclophosphamide results with
haplo donors seem comparable to those using MUD donors.
With these developments the use of allo HCT may increase. In addition to potentially benefiting
patients, this might decrease the potential effect of selection bias in allo HCT studies, a problem
that of course is not unique to allo HCT. Although in the study of Ostgard et al[111], it was
recommended that all patients with intermediate or adverse cytogenetics in CR1 receive allo
HCT, only 19% did so, although this proportion increased from11% to 27% in the later years of
the study. HCT rates of 60% in patients with adverse cytogenetics in CR1 have been reported
in a cooperative group setting[122]. As in any study it remains unknown how representative these
60% were of the entire population, a problem that becomes worse when studies do not report
the number of patients receiving various types of HCT as a proportion of those who might have
done so.
Preventing post allo HCT relapse

Relapse remains the principal cause of failure after allo HCT. For years there has been debate
whether the reduction in NRM afforded by reduced intensity conditioning (RIC) regimens is
outweighed by their potential to increase relapse. Some randomized studies have found this to
be the case while other randomized studies have not[123] with the former finding a much higher
relapse rate despite seeming similarities in RIC regimens and known prognostic factors for
relapse, suggesting selection bias as an issue. Comparing myeloablative (MA) and RIC in patients
|Page

according to their pre HCT MRD status as assessed by NGS Hourigan et al.[124] found MA
decreased relapse risk and improved survival and RFS only in patients who were NGS positive.
Eapen et al. emphasized the value of examining regimens individually rather than as either “RIC”
or “MA” [125]. Comparing seven non- total body irradiation containing regimens they concluded
patients with “acceptable” HCT-CI scores should be conditioned with “Bu4/Cy”, “Flu/Bu4”, or
Flu/Mel, with younger patients possibly also benefitting from Flu/Mel. Treosulfan in combination
with fludarabine and relatively low dose TBI is another conditioning regimen that has attracted
attention[126] and several studies are targeting the myeloid surface antigens CD33 and CD45 with
alpha particles such as actinium-225 and astatinie-211[127].
Risk of relapse might also be reduced by post HCT interventions. Madiarz et al. randomized 60
adults age < 70 with FLT ITDs in remission 1-2 months after allo HCT to receive usual care +/-
midostaurin 50 mg twice daily; median exposure was 10 months. Relapse rates were 24% in the
control and 11% in the midostaurin arm, with 18-month RFS probabilities 76% (95% CI 54-88%)
in the control and 89% (69-96%)[128]. Analogous trials of newer FLT3 inhibitors (see below) are
ongoing in this setting. Despite the overlapping 95% CI in the +/- midostaurin trial, the proportion
of patients who will be willing to accept a 50% chance of receiving placebo rather than a 100%
chance of receiving midostaurin (off study) is uncertain, especially given the results of Schlenk et
al. noted above[27]. Because of azacitidine’s ability to upregulate potential AML- associated
antigens thus inducing a CD8+ T cell response augmenting the GVL effect, Craddock
administered the drug to 51 patients undergoing allo HCT, finding a decreased relapse risk (HR
0.30, 95% CI 0.10-0.85) with similar improvement in RFS in patients demonstrating a CD8+ T-
cell response to tumor antigens[129]. Although azacytidine has yet to be accepted as prophylaxis
against post HCT relapse, it is well-tolerated and thus is often given to patients at high risk of this
complication. The availability of the potentially more effective oral form of azacytidine (CC-486,
see below) may further increase the use of azacytidine as prophylaxis of post HCT relapse[130].
More sophisticated attempts have also been made to augment the GVL effect, for example by
generating T cells expressing receptors (TCR) that target selected AML antigens, for example
WT-1. Chapuis et al. identified a high-affinity, HLA-A2-restricted WT1-specific TCR (TCRC4) from
|Page

normal HLAA2 + donors and inserted TCRC4 into Epstein-Bar virus-specific donor CD8+ T cells
(TTCR-C4) to minimize graft-versus-host disease risk and enhance transferred T cell
survival[131].TTCR-C4 were infused prophylactically post-HCT into 12 patients. with no detectable
disease post HCT. Relapse-free survival was 100% at a median of 44 months following infusion,
while a concurrent comparative group of 88 patients with similar pre-HCT risk of relapse had 54%
relapse-free survival (P = 0.002). TTCR-C4 maintained TCRC4 expression and persisted long-term.
A future challenge is to dissect mechanisms of resistance (e.g. lack of T-cell persistence) in
patients with MRD post HCT, in whom clinical results were less impressive.
Autologous (auto) HCT

In contrast to the increasingly frequent use of allo HCT, the use of auto HCT has been in decline,
rarely being recommended in the U.S[132, 133]. Nonetheless there are indications that, despite lack
of an obvious GVL effect, auto HCT provides more anti AML effect than chemotherapy, at least
in “favorable” and intermediate risk patients[134, 135]. Survival after allo HCT can be comparable to
that seen after allo HCT, reflecting the use of allo HCT in 2nd CR(CR2) in up to 30% of patients
whose disease has recurred after auto HCT; these findings are reminiscent of those observed in
patients with intermediate risk disease, in whom allo HCT might be deferred, in lieu of
chemotherapy, until relapse[136].
MRD testing might prompt a re-evaluation of auto HCT. Specifically, patients in the ELN2017
intermediate (and possibly some in the ELN favorable group) who are MRD negative after initial
therapy might be considered for auto HCT rather than allo HCT.
Patients who are not plausible HCT candidates (table 6)

The first group of such patients (Group A) are those whose risk of relapse without allo HCT is
sufficiently low that allo HCT should not be recommended or be recommended only with
reservations These patients include those in the ELN2017 favorable group without MRD, in whom
even the less risky auto HCT need not be considered, and who thus become candidates for post-
remission chemotherapy. Patients in the ELN2017 intermediate group without MRD have a
somewhat higher risk of relapse without allo HCT and hence might be candidates for allo HCT if
|Page

their risk of post -allo HCT NRM is low and might also be candidates for auto HCT or post-
remission chemotherapy; a randomized trial comparing these options in this group would be of
interest. The second group of patients (Group B) are those in whom the risk of HCT is prohibitive,
for example a > 40% chance of NRM after allo HCT HCT-CI score of 4 or greater). While these
patients thus become candidates for post-remission chemotherapy, such therapy would optimally
be less intense than the chemotherapy recommended for Group A patients.
Chemotherapy for Group A emphasizes cytarabine in individual doses of 1-3 g/m2 daily or BID
for 3 days per cycle for 3 cycles, often given as 3 g/m2 every 12 hours on days 1, 3, and 5.
However, a German- Austrian AML Study Group found administration of this dose on days 1, 2,
and 3 to 392 patients was associated with a median 4-day reduction in time to ANC >500 from
start of chemotherapy and with fewer platelet transfusions than in 176 historical patients given
the 1, 3, and 5 schedule[137]. Cytarabine doses of 1 g/m2 twice daily for 3-4 days for three courses
after CR is achieved are very probably as effective as higher doses and/or more courses
irrespective of cytogenetic group.[138] Furthermore, after randomizing 221 adults age < 60
between cytarabine in the usual 3 g/m2 schedule and cytarabine 1 g/m2 daily X5 plus clofarabine
30 mg/m2 daily X5 (CLARA) both preceded by G-CSF, Thomas et al. reported a lower incidence
of relapse (34% vs 46% at 2 years P = .025) accompanied by longer RFS (but not survival) with
CLARA regardless of risk group, raising the question whether addition of a purine analogs such
as fludarabine, cladribine, or clofarabine should be added to cytarabine for standard post-
remission therapy in such patients.[139] Perhaps most importantly decisions about doses and
schedules for subsequent courses of post-remission therapy cannot be stereotyped and must be
based on tolerance of the previous course, bearing in mind that relapse rather than death in CR
is the predominant problem.
“Maintenance” has generally referred to the prolonged administration of relatively less intense
therapy to maintain CR. It had until very recently fallen out of favor[140]. However, azacitidine
appears to offer a new maintenance option for patients who are candidates for neither HCT nor
cytarabine in the doses described in the preceding paragraph (table 6). Huls et al. randomized
between observation and azacytidine 116 patients age 60 or greater (median 69) with
|Page

performance status < 3 who were in CR (71%) or CRi after 2 cycles of intensive therapy[141].
Azacytidine dose was 50 mg/m2 s.c. daily X 5 every 4 weeks for up to 12 cycles or until relapse.
82% received at least 4 cycles and 62% received 12 cycles. Median follow-up was 3.5 years.
After accounting for age, 1 vs. 2 cycles to best response (CR or CRi) , adverse risk cytogenetics
and platelet count < vs > 100,000 (as a surrogate for CR vs CRi, azacytidine reduced the risk of
relapse, or death in remission (HR 0.62, 95% CI 0.41-0.95, p = 0.03) (figure 13) . 12-month RFS
probability was 64% vs. 42%. The effect of azacytidine was similar regardless of age,
cytogenetics, cycles to best response, and performance status, but was greater in patients who
had attained platelet count > 100,000 after induction (presumably CR). There were no differences
in survival, likely because of the more frequent use of “salvage” therapy in the observation group,
for example 11 allo HCT vs 4. Salvage therapy improved survival in both observation and
treatment groups, without suggestion that relapse after azacytidine prejudiced chance of
responding to salvage.
The above study required 7.5 years to accrue 118 patients, a not uncommon occurrence in studies
of maintenance therapy in older patients[141]. A possible explanation is the need to receive
parenteral therapy for a long time. An oral preparation may not only be useful per se but the
[142]
prolonged exposure it affords may be therapeutically beneficial. Wei et al. randomized 472
patients age 55-86 (median 68) with performance status < 3), intermediate (86%) or adverse risk
(14%) cytogenetics between oral azacitidine (CC-486, 300 mg daily on days 1-14 of 28-day
cycles) and placebo. 81% were in CR and 19% in CRi. 45% had received one and 35% two
courses of “consolidation” therapy other than azacytidine. Azacytidine/placebo could continue
indefinitely barring excess toxicity or appearance of > 15% blasts. With a median follow-up of 3.5
years CC 486 reduced the risk of death (HR 0.69, 95% CI 0.55-0.86) with median survivals of
24.8 vs. 14.2 months and the risk of relapse and/or death (HR 0.65, 95% CI 0.52-0.81), with
median RFS of 10.2 vs. 4.8 months (figure 14). Nausea, vomiting, and diarrhea, generally mild-
moderate, were more common with CC486 occurring in 64%,59% and 49% of patients and were
the most common reasons for discontinuation. However, health- related QOL was similar in both
arms.
|Page

This study establishes CC-486, which will likely be approved expeditiously by FDA, as the
standard of care (and of comparison) for patients age 60-65 or above with AML in remission, who
are not candidates for HCT or chemotherapy using higher doses of cytarabine. However because
neither parenteral nor the oral azacytidine study presented results according to MRD status it is
less clear whether it replaces “clinical trial” as an option in patients who are either MRD positive
or with adverse cytogenetics, the latter comprising only 14% of the patients in the CC-486 trial
and 16% in the trial of Huls et al[141]. As always, the choice between standard therapy (e.g. CC486)
and trial should be made once patients are informed of the results of the former and the unknown
results with the latter. Examples of potential trials are provided in table 4 and in the section on
relapsed/refractory disease below under the assumption drugs found effective in relapsed or
newly- diagnosed disease might also be effective in the remission setting.
RELAPSED AND REFRACTORY AML

Detection of relapse – National Comprehensive Cancer Network (NCCN) guidelines recommend
routine follow-up bone marrows in patients in remission only should there be a deterioration in
blood counts. This recommendation was based on MD Anderson data indicating marrow relapse
(at least 5% blasts by morphologic enumeration) unaccompanied by blood count deterioration
occurred in only 16% of 375 patients[143]. Furthermore, although the data were sparse, patients
who were treated at time of such “marrow first” relapses did not appear to do better than marrow
first patients who were treated only at time of count deterioration, which generally occurred within
a few months of a marrow first relapse. The frequency of blood count monitoring should vary with
the risk of relapse, recalling the risk of relapse per unit time does not begin to appreciably
decrease until 2 years have elapsed from remission date[22].
There is plausible therapeutic value in detecting relapse by MRD before appearance of

morphologic relapse. The above cited paper[143] did not examine the value of serial examination
of marrows to detect MRD. However, data suggest peripheral blood can be substituted for marrow
for this purpose[107, 109]. At any rate Zhou et al. have shown that should marrow show no evidence
of MRD by MFC it will inevitably not detect relapse by morphology, questioning the need to
examine a marrow for morphology should it be negative by MFC[144].
|Page

Definition of primary refractory AML
ELN 2017 considers AML refractory if it does not respond to 2 courses of 7+3[30]. This was based
on SWOG data showing the rates of CR and TRM on a 2nd course of 7+3 (43% and 10%
respectively) were similar to those on a 1st course (48% and 9%), without evidence covariates
such as age, cytogenetics, results of a “day 14” marrow, or days between starting course 1 and
starting course 2 influenced 2nd course CR rate[145]. However, data from more recent SWOG
protocols suggest patients in CR only after two courses of 7+3 have shorter RFS and survival
than patients in CR after a first 7+3, even after accounting for prognostic covariates (figure 15)[146].
This would suggest that patients not in CR after a first 7+3 should also be considered refractory.
Because the CR rate on a 2nd induction course of therapy including cytarabine at doses of at least
1 g/m2 is considerably lower than the CR rate on a first such course(61% vs. 26%), AML might
be considered refractory if not in CR after a first course of either 7+3 or more intense induction
(e.g. FLAG-ida)[147].Whether this definition should lead to a change in therapy is a decision that
must rest on the unknown probability a change to one of the clinical trials described below would
lead to a better outcome.
Clinical distinction between relapsed and refractory AML-
Although duration of prior remission is often regarded as a discrete variable (e.g. < vs > 6 months),
in general the longer the duration of remission the higher the probability of achieving a CR
following salvage therapy[148]. It follows there is more difference between a CR duration of 7
months and one of 13 months (both “long”) than between a CR duration of 5 months (“short”) and
a CR duration of 7 months (“long”) and probably between a CR duration of 2 months and 0
months, the latter considered “primary refractory(PREF). Hence, I will consider PREF at the
lowest end of a continuum of prior CR durations.
Management of relapsed/refractory AML as defined by morphology (5% blasts or more)

Several systems which combine CR1 duration with other factors are available for assessing
prognosis.[149] A convenient one, derived from 667 adults aged <60 in first relapse, remains Breems et
al.[150]( also known as the European Prognostic Index, EPI, Table 7). Survival probabilities at 1-year
were 70% in the best group, (46% at 5 years), which however constituted only 9% of patients, with a
|Page

mean CR1 duration of 17 months), and were 46% in the intermediate group (25% of patients, mean
CR1 duration 16 months) and 16% in the worst group (2/3 s of patients, mean CR1 duration 7
months) .CR rates were 85%,60%, and 34% respectively among patients in the “favorable”,
intermediate, and adverse groups who received treatment. Results in the adverse group would undoubtedly
have been worse had PREF patients been included. Although reasons underlying initial therapeutic choices
were difficult to ascertain and therapy- specific CR rates were accordingly not provided, these results
suggest patients in the worst group (alternatively those with CR1 durations <1 year) are unlikely to
benefit from standard intensive therapy (e.g. FLAG-ida or MEC), thus becoming candidates for new
therapies/ “clinical trials.” In contrast, standard therapy might be more appropriate in the best and
intermediate groups. In general, however repetition at relapse of the same induction and post-remission
therapy used initially leads to worse results This suggests that even should 2nd CR be attained with
standard induction, post-remission therapy should, if feasible, include allo HCT, particularly if this has not
been done before; indeed, regardless of prognostic group patients lived longer if in CR2 they received
allo HCT, although, as usual, various selection biases may have affected the decision to perform allo
HCT. [150] The recommendation for new therapy/clinical trials in all patients in the intermediate and,
particularly worst, Breems et al. prognostic groups certainly extends to older patients, who all else
being equal would be expected to do even worse than the patients analyzed by Breems et al. [151]
and to all patients who are receiving a 2nd re-induction attempt after failing a first[152].
Table 8 summarizes recommendations for management of R/R AML. If feasible, allo HCT is favored
in CR2 even in cases with in which allo HCT was performed in CR1. A possible exception may be cases
in which allo HCT was done in CR1, but relapse occurred within 6 months of HCT. Here, unless some
very different approach is taken to 2nd HCT the morbidities of the procedure may not be commensurate
with the benefits. A possible alternative to a second allo HCT is donor lymphocyte infusions. A
retrospective comparison of 2nd allo HCT vs. DLI found, subject to the usual selection biases,
similar survival with the two procedures (26% vs 25% at 2 years, 19% vs. 15% at 5 years). With
either 2nd allo HCT or DLI a shorter time after 1st allo HCT and active disease at time of the
[153]
procedure were associated with poorer outcomes . Presence of MRD should militate against
DLI.
Newer therapies for morphologically relapsed AML
|Page

Gilteritinib – this inhibitor of the AML-promoting effects of FLT3 internal tandem duplications (ITD)
and tyrosine kinase domain (TKD) mutations was approved by FDA in 2018 for patients with
relapsed AML and FLT3 ITDs or TKDs. A trial randomizing 371 such patients 2:1 between
gilteritinib (120 mg daily) and salvage chemotherapy (FLAG-ida or mitoxantrone+ etoposide+
cytarabine at doses of 1g/m2 or greater (MEC) as intense, and azacitdine or low dose cytarabine
as less intense options, found gilteritinib reduced the risk of death (HR 0.64, 95% CI 0.49-0.83,
P < 0.001), medians 9.3 vs 5.6 months)[154]. Results were similar regardless of (a) age < vs > 65,
(b) intensity of salvage therapy, (c)whether patients had had allo HCT, and (d) duration of CR1
(< vs > 6 months, although patients with PREF AML had less obvious benefit (figure 16). Only
12% of patients had previously received a FLT3 inhibitor, making it difficult to know whether
relapse after receipt of midostaurin would prejudice response to gilteritinib (figure 16). A trial
comparing chemotherapy + midostaurin or + gilteritinib in newly- diagnosed patients would be of
interest. Although gilteritinib is now conventional care for patients with relapsed FLT3 positive
AML, the development of new FLT3 inhibitors and understanding development of resistance
remains a very active area[155].
E-selectin ligand inhibition- E-selectin ligand is expressed on AML blasts and mediates adherence
to E-selectin on bone marrow endothelium resulting in resistance to chemotherapy. Uproleselan
disrupts adherence, putatively re-sensitizing blasts to chemotherapy. A single-arm trial of
uproleselan added to MEC reported[156] a 45% CR/CRi rate and 12.8 month median survival in
the 22 patients with >10% E-selectin ligand expression vs. 29% CR/CRi rate with 5.4 month
median survival in the 14 patients with less expression; 79% of the former and 72% of the latter
patients were in the Breems et al. poor prognostic group or receiving at least 2nd salvage
suggesting results in the higher expression group were better than those otherwise expected[150,
152]
. A randomized trial comparing FAI or MEC +/- uproleselan is ongoing in relapsed/ refractory
and newly- diagnosed patients.
Immunotherapeutic approaches
a) Bispecfic T-cell Engagers antibodies(BiTEs)- these attempt to redirect T cells to hematologic
malignancies. Blinatumomab and inotuzumab, which engage CD3 present on almost all T cells
|Page

and CD19 (blinatumomab) and CD22 (inotuzumab) present on B-cell malignancies, are probably
the most successful examples, particularly as single agents in acute lymphocytic leukemia[157].
Several BiTEs (or analogously DARTs (dual-affinity receptor targeting) have been developed for
use in AML with CD33 and CD123 the most common target antigens; others are CLL1, FLT3,
WT1 and CD13. A significant potential toxicity is cytokine release syndrome (CRS), which was
observed in 15/30 (grade 3 in 4, grade 4 in none) relapsed/refractory patients given the CD33-
CD3 BiTE AMG673[158]. A > 50% decrease in marrow blasts occurred in 6/27 evaluable patients,
with 1/27 achieving CRi (85% decrease in blasts). Information was not provided to permit
estimation of expected response rates with more standard therapy, but, given CRS’s amenability
to treatment with tocilizumab[157] dose escalation was continuing.
b) Chimeric antigen receptor (CAR) T-cells – as with BiTEs, to date the success of CAR T cell
therapy in AML has been considerably less than in ALL. One likely reason is the absence of a
specific cell surface antigen, such as CD 19, that can be targeted with relatively little toxicity, such
as myelosuppression. Current approaches to address this problem include: 1) using transient or
removable CAR T-cells as a component of an allo HCT conditioning regimen, 2) gene-editing
(e.g. deleting CD33 from the donor’s cells) to allow safe and protracted anti-AML CAR T-cell
function[159].
c) immune checkpoint blockade -the success of this approach in solid tumors has motivated trials
in AML. Resistance to azacytidine or decitabine (“HMAs”) appears at least partially mediated by
these drugs’ ability to increase expression of PD-1 thus decreasing the immune response to AML
and suggesting the use of anti PD-1 antibodies, which by themselves are only minimally effective
in AML[160] This prompted Daver et al to administer azacytidine (75 mg/m2 daily X 7) and
nivolumab(3 mg/kg days 1 and 14) to 70 patients with relapsed/refractory AML[160].The ORR was
33%(6% CR) with median survival 6 months. Results appeared better in patients who had not
received HMAs than seen in analogous historical controls (ORR 52% vs. 22%), as did survival in
patients receiving azacytidine and nivolumab as “first salvage” (medians 10.6 vs. 5.3 months)
with similar observations for EFS, although no multivariate analysis was provided and follow-up
was naturally longer in the controls (medians 51 vs 13 months). Higher marrow CD3 and CD8
expression seemed to distinguish responders from non- responders and might be useful in
|Page

identifying patients for future immune checkpoint approaches[160].
Another example of this approach is magrolimab an antibody targeting CD47, a macrophage

immune checkpoint. CD47 blockade induces tumor phagocytosis and eliminates leukemia stem
cells (LSC) in AML models. Azacitidine (AZA) enhances phagocytosis thus presumably
synergizing with magrolimab. The azacytidine + magrolimab combination produced CR or CRi in
7/13 patients with newly-diagnosed MDS and in 8/16 patients with newly-diagnosed AML
considered unfit for intensive induction [161].75% of the patients had adverse risk cytogenetics and
28% a TP53 mutation and the median time to response of 1.9 months seemed quicker than
expected with azacitidine alone. The median follow-up time of 4.9 months hampers estimates of
duration of response and survival compared, for example, to azacytidine.
Allo HCT- like all therapies for AML allo HCT is much less effective when morphologic disease is
present. Improved conditioning regimens are one means to address this problem[127]. In a recent
example Gyurkocza et al. randomized 75 patients age > 55 not in remission, or in relapse and
with > 5% blasts, between investigator’s choice of chemotherapy including venetoclax, ivosidenib,
or enasidenib (“conventional care”) and [130]I apamistamab (IOMAB B)
followed 12-14 days later by reduced intensity conditioning (fludarabine (FLU)/TBI) and allo
HCT[162]. Patients in the conventional care arm could proceed to allo HCT in the event of CR or
crossover to the IOMAB-HCT arm if not in CR. Reflecting that 85% of patients had previously
failed at least 2 induction therapies, 31/38 conventional arm patients again failed chemotherapy,
including 11 receiving targeted therapy. However, patients were able to cross-over within a
median of 64 days (range 51-161) from randomization such that 50/74 patients (31 initially
randomized to HCT plus 19 cross overs) were able to receive HCT. The 100-day post HCT NRM
rate was 0% in the former and 7% in the latter, with a median of 2-3 weeks to neutrophil and
platelet engraftment in each arm. Although EFS, RFS, CR, and survival were not reported the trial
suggests the feasibility of allo HCT in patients typically not considered HCT candidates based on
older age, active AML, or number of prior salvage attempts.
Treatment of MRD
|Page

In 80- 90% of cases untreated MRD leads to morphologic relapse, typically in 6-12 months[96-99].
Observational studies suggest changing treatment to include allo HCT in patients in MRD positive
remission is beneficial[113, 114]
. These findings suggest the potential benefit of using MRD for
therapeutic decision making and, specifically, of treating patients to eliminate MRD before
morphologic relapse occurs. However, administering azacytidine or high-dose cytarabine-
containing chemotherapy to 49 patients in whom MRD assessed by MFC remained after an initial
[163]
induction course, Appelbaum et al. reported elimination of MRD in only 10/49 . It is unknown
if MRD elimination would improve outcome since MRD may simply be a surrogate for “bad AML”.
However, the effect of eliminating MRD on outcome can only be investigated if effective means
to eliminate MRD are available. But because the entity of “MRD-only relapse” has yet to be
recognized (although recognition of CR without MRD[30] would seem to imply a category of relapse
based solely on MRD), pharmaceutical companies have typically required 5% or more blasts as
determined by morphology as an eligibility criterion for trials of new drugs in relapsed/refractory
AML. However. it is plausible a drug may demonstrate activity in the MRD-only but not the
morphologic relapse scenario. If so, testing new drugs only in the latter situation might produce
false-negative conclusions. Effective MRD-directed therapy might also allow investigation, e.g. by
randomization, whether allo HCT eligible patients with MRD should proceed directly to HCT or
first receive MRD directed therapy.
OTHER TOPICS
Need for hospitalization
Conventionally, AML patients have been hospitalized for 5-7 days to receive initial induction,
remaining in hospital for another 3-4 weeks until neutrophil and platelet count recovery. The
increasing use of less intense induction will almost certainly reduce the need for hospitalization.
However, it has been shown that patients can be safely discharged from hospital after completion
of intense induction therapy (“early discharge”), assuming availability of caregivers and an
adequate outpatient infrastructure, with patients living relatively near (e.g. 30-50 miles) of the
center. Early discharge patients were shown to have similar TRM as controls who met the same
medical eligibility criteria (doing well without fever etc. when chemotherapy completed) but who
lacked caregivers or who lived at too great a distance from the outpatient center[164]. Selected
|Page

patients can also receive intense induction therapy as outpatients, again assuming adequate
infrastructure, residence relatively close to the center, and availability of caregivers[165]. In addition
to improving quality of life, the possibility that even patients given intensive induction can spend
less time in hospital has practical importance during the COVID-19 pandemic when hospital beds
may be at a premium.
Testing for germline mutations
Mutations present in hematopoietic cells at birth are germline mutations (i.e. present at birth in all
cells)[166, 167]
. Germline mutations often involve RUNX1, GATA2, and DDX41 and characterize
hereditary myeloid malignancy syndrome (HHMS). These frequently culminate in MDS/AML, which,
although usually presenting in childhood, can be seen in adults usually age <40 years but occasionally
older. Screening families for germline mutations is advisable if a family member is a potential donor for
allo- HCT, if a patient has a potentially germline mutation (e.g., RUNX1, GATA2, and DDX41), HHMS
organ manifestations, or a suggestive family history[166, 167]
.Skin biopsies or nail clippings are
commonly used to test for germline mutations.
Fertility preservation
Given the potential teratogenicity of AML therapy storage of sperm or eggs is an option that
should be offered every young patient, in consultation with appropriate specialists. Such storage
is eminently feasible given data suggesting the need to begin AML therapy as an emergency is
probably less than suspected (see above and references 70 and 71).
Distinction between AML and MDS

The 2016 World Health Organization's (WHO) criterion for AML is at least 20% myeloblasts in marrow
(or blood) with myeloid lineage established by multiparameter flow cytometry (MFC). [168] Exceptions to
the ≥20% criterion are cases of CBF AML (cytogenetic abnormalities t [8;21], inv [16], or t[16;16]), NPM1
mutated AML, or acute promyelocytic leukemia; in each the diagnosis of AML is independent of blast %.
This is clinically sensible because each of these entities does well with “AML-type” therapy[30, 73].
Biologically, patterns of mutations in AML arising after a preceding hematologic disorder such as
|Page

myelodysplasia often bear more resemblance to those seen in MDS than in primary (de novo) AML.[169,
170]
In any event the 20% blast cut-off used to distinguish “AML” from “MDS” is arbitrary. Based on the
possibility mutations and cytogenetics are more prognostically and therapeutically
relevant than the 20% cut-off, some centers treat patients with >10% blasts as AML, even if they
are considered as excess blasts 2 by the WHO.
Date-sharing: Data sharing is not applicable to this article as no new data were created or analyzed
in this study.
Funding: None
References
1. Meyers J, Yu Y, Kaye JA, et al. Medicare fee-for-service enrollees with primary acute myeloid
leukemia: an analysis of treatment patterns, survival, and healthcare resource utilization and costs. Appl
Health Econ Health Policy 2013;11:275-286.
2. Medeiros BC, Satram-Hoang S, Hurst D, et al. Big data analysis of treatment patterns and
outcomes among elderly acute myeloid leukemia patients in the United States. Ann Hematol
2015;94:1127-1138.
3. Oran B, Weisdorf DJ. Survival for older patients with acute myeloid leukemia: a population-
based study. Haematologica 2012;97:1916-1924.
4. Juliusson G, Lazarevic V, Hörstedt AS, et al. Acute myeloid leukemia in the real world: why
population-based registries are needed. Blood 2012;119:3890-3899.
5. Juliusson G, Antunovic P, Derolf A, et al. Age and acute myeloid leukemia: real world data on
decision to treat and outcomes from the Swedish Acute Leukemia Registry. Blood 2009;113:4179-4187.
6. Dombret H, Seymour JF, Butrym A, et al. International phase 3 study of azacitidine vs
conventional care regimens in older patients with newly diagnosed AML with >30% blasts. Blood
2015;126:291-299.
|Page

7. Kantarjian HM, Thomas XG, Dmoszynska A, et al. Multicenter, randomized, open-label, phase III
trial of decitabine versus patient choice, with physician advice, of either supportive care or low-dose
cytarabine for the treatment of older patients with newly diagnosed acute myeloid leukemia. J Clin
Oncol 2012;30:2670-2677.
8. Halpern AB, Lyman GH, Walsh TJ, et al. Primary antifungal prophylaxis during curative-intent
therapy for acute myeloid leukemia. Blood 2015;126:2790-2797.
9. Othus M, Kantarjian H, Petersdorf S, et al. Declining rates of treatment-related mortality in
patients with newly diagnosed AML given 'intense' induction regimens: a report from SWOG and MD
Anderson. Leukemia 2014;28:289-292.
10. Juliusson G, Hagberg O, Lazarevic VL, et al. Improved survival of men 50 to 75 years old with
acute myeloid leukemia over a 20-year period. Blood 2019;134:1558-1561.
11. Jakobsen LH, Roug AS, Øvlisen AK, et al. Temporal development of survival in adult AML patients
in the period 2000-2016: A Danish population-based study. 2020. In Press.
12. Martínez-Cuadrόn D, Serrano J, Gil C, et al. Evolving treatment patterns and outcomes in older
patients (≥ 60 years) with AML: changing everything to change nothing? . Leukemia 2020. In Press.
13. DiNardo CD, Pratz K, Pullarkat V, et al. Venetoclax combined with decitabine or azacitidine in
treatment-naive, elderly patients with acute myeloid leukemia. Blood 2019;133:7-17.
14. Wei AH, Strickland SA, Jr., Hou JZ, et al. Venetoclax Combined With Low-Dose Cytarabine for
Previously Untreated Patients With Acute Myeloid Leukemia: Results From a Phase Ib/II Study. J Clin
Oncol 2019;37:1277-1284.
15. Soni PD, Hartman HE, Dess RT, et al. Comparison of Population-Based Observational Studies
With Randomized Trials in Oncology. J Clin Oncol 2019;37:1209-1216.
16. Courtney DiNardo, Brian Jonas, Vinod Pullarkat, et al. A RANDOMIZED, DOUBLE-BLIND,
PLACEBO-CONTROLLED STUDY OF VENETOCLAX WITH AZACITIDINE VS AZACITIDINE IN TREATMENT-
NAÏVE PATIENTS WITH ACUTE MYELOID LEUKEMIA INELIGIBLE FOR INTENSIVE THERAPY-VIALE-A. EHA
Library 2020.
17. Wei AH, Montesinos P, Ivanov V, et al. Venetoclax plus LDAC for newly diagnosed AML ineligible
for intensive chemotherapy: a phase 3 randomized placebo-controlled trial. Blood 2020;135:2137-2145.
18. Lancet JE, Uy GL, Cortes JE, et al. CPX-351 (cytarabine and daunorubicin) Liposome for Injection
Versus Conventional Cytarabine Plus Daunorubicin in Older Patients With Newly Diagnosed Secondary
Acute Myeloid Leukemia. J Clin Oncol 2018;36:2684-2692.
19. Lancet JE, Uy GL, Newell LF, et al. Five-year final results of a phase III study of CPX-351 versus
7+3 in older adults with newly diagnosed high-risk/secondary AML. Journal of Clinical Oncology
2020;38:7510-7510.
20. Ryan DH, Newell LF, Ritchie EK, et al. Outcomes in Patients with Acute Myeloid Leukemia with
Myelodysplasia-Related Changes (AML-MRC) Who Achieved Remission with CPX-351 Versus 7+3: Phase
3 Exploratory Analysis. Blood 2019;134:3842-3842.
21. Montalban-Bravo G, Kanagal-Shamanna R, Class CA, et al. Outcomes of acute myeloid leukemia
with myelodysplasia related changes depend on diagnostic criteria and therapy. Am J Hematol
2020;95:612-622.
22. de Lima M, Strom SS, Keating M, et al. Implications of potential cure in acute myelogenous
leukemia: development of subsequent cancer and return to work. Blood 1997;90:4719-4724.
|Page

23. Stone RM, Mandrekar SJ, Sanford BL, et al. Midostaurin plus Chemotherapy for Acute Myeloid
Leukemia with a FLT3 Mutation. N Engl J Med 2017;377:454-464.
24. Stone RM, Mandrekar SJ, Sanford BL, et al. The Addition of Midostaurin to Standard
Chemotherapy Decreases Cumulative Incidence of Relapse (CIR) in the International Prospective
Randomized, Placebo-Controlled, Double-Blind Trial (CALGB 10603 / RATIFY [Alliance]) for Newly
Diagnosed Acute Myeloid Leukemia (AML) Patients with FLT3 Mutations. Blood 2017;130:2580-2580.
25. Estey EH. Acute myeloid leukemia: 2019 update on risk-stratification and management. Am J
Hematol 2018;93:1267-1291.
26. Larson RA, Mandrekar SJ, Sanford BL, et al. An Analysis of Maintenance Therapy and Post-
Midostaurin Outcomes in the International Prospective Randomized, Placebo-Controlled, Double-Blind
Trial (CALGB 10603/RATIFY [Alliance]) for Newly Diagnosed Acute Myeloid Leukemia (AML) Patients with
FLT3 Mutations. Blood 2017;130:145-145.
27. Schlenk RF, Weber D, Fiedler W, et al. Midostaurin added to chemotherapy and continued
single-agent maintenance therapy in acute myeloid leukemia with FLT3-ITD. Blood 2019;133:840-851.
28. Hills RK, Castaigne S, Appelbaum FR, et al. Addition of gemtuzumab ozogamicin to induction
chemotherapy in adult patients with acute myeloid leukaemia: a meta-analysis of individual patient data
from randomised controlled trials. Lancet Oncol 2014;15:986-996.
29. Grimwade D, Hills RK, Moorman AV, et al. Refinement of cytogenetic classification in acute
myeloid leukemia: determination of prognostic significance of rare recurring chromosomal
abnormalities among 5876 younger adult patients treated in the United Kingdom Medical Research
Council trials. Blood 2010;116:354-365.
30. Döhner H, Estey E, Grimwade D, et al. Diagnosis and management of AML in adults: 2017 ELN
recommendations from an international expert panel. Blood 2017;129:424-447.
31. Amadori S, Suciu S, Selleslag D, et al. Gemtuzumab Ozogamicin Versus Best Supportive Care in
Older Patients With Newly Diagnosed Acute Myeloid Leukemia Unsuitable for Intensive Chemotherapy:
Results of the Randomized Phase III EORTC-GIMEMA AML-19 Trial. J Clin Oncol 2016;34:972-979.
32. Fournier E, Duployez N, Ducourneau B, et al. Mutational profile and benefit of gemtuzumab
ozogamicin in acute myeloid leukemia. Blood 2020;135:542-546.
33. Olombel G, Guerin E, Guy J, et al. The level of blast CD33 expression positively impacts the effect
of gemtuzumab ozogamicin in patients with acute myeloid leukemia. Blood 2016;127:2157-2160.
34. Khan N, Hills RK, Virgo P, et al. Expression of CD33 is a predictive factor for effect of
gemtuzumab ozogamicin at different doses in adult acute myeloid leukaemia. Leukemia 2017;31:1059-
1068.
35. Lamba JK, Chauhan L, Shin M, et al. CD33 Splicing Polymorphism Determines Gemtuzumab
Ozogamicin Response in De Novo Acute Myeloid Leukemia: Report From Randomized Phase III
Children's Oncology Group Trial AAML0531. J Clin Oncol 2017;35:2674-2682.
36. Gale RE, Popa T, Wright M, et al. No evidence that CD33 splicing SNP impacts the response to
GO in younger adults with AML treated on UK MRC/NCRI trials. Blood 2018;131:468-471.
37. Rafiee R, Chauhan L, Alonzo TA, et al. ABCB1 SNP predicts outcome in patients with acute
myeloid leukemia treated with Gemtuzumab ozogamicin: a report from Children's Oncology Group
AAML0531 Trial. Blood cancer journal 2019;9:51.
|Page

38. Roboz GJ, DiNardo CD, Stein EM, et al. Ivosidenib induces deep durable remissions in patients
with newly diagnosed IDH1-mutant acute myeloid leukemia. Blood 2020;135:463-471.
39. Cortes JE, Heidel FH, Hellmann A, et al. Randomized comparison of low dose cytarabine with or
without glasdegib in patients with newly diagnosed acute myeloid leukemia or high-risk myelodysplastic
syndrome. Leukemia 2019;33:379-389.
40. Administration USFaD. FDA approves ivosidenib as first-line treatment for AML with IDH1
mutation. 2019.
41. Administration USFaD. FDA approves glasdegib for AML in adults age 75 or older or who have
comorbidities. 2018.
42. Walter RB, Othus M, Borthakur G, et al. Prediction of early death after induction therapy for
newly diagnosed acute myeloid leukemia with pretreatment risk scores: a novel paradigm for treatment
assignment. J Clin Oncol 2011;29:4417-4423.
43. Klepin HD. Geriatric perspective: how to assess fitness for chemotherapy in acute myeloid
leukemia. Hematology Am Soc Hematol Educ Program 2014;2014:8-13.
44. Bories P, Lamy S, Simand C, et al. Physician uncertainty aversion impacts medical decision
making for older patients with acute myeloid leukemia: results of a national survey. Haematologica
2018;103:2040-2048.
45. Maiti A, Rausch CR, Cortes JE, et al. Outcomes of relapsed or refractory acute myeloid leukemia
after frontline hypomethylating agent and venetoclax regimens. Haematologica 2020.
46. DiNardo CD, Rausch CR, Benton C, et al. Clinical experience with the BCL2-inhibitor venetoclax in
combination therapy for relapsed and refractory acute myeloid leukemia and related myeloid
malignancies. Am J Hematol 2018;93:401-407.
47. DiNardo CD, Ravandi F, Agresta S, et al. Characteristics, clinical outcome, and prognostic
significance of IDH mutations in AML. Am J Hematol 2015;90:732-736.
48. Appelbaum FR, Gundacker H, Head DR, et al. Age and acute myeloid leukemia. Blood
2006;107:3481-3485.
49. Yanada M, Garcia-Manero G, Borthakur G, et al. Relapse and death during first remission in
acute myeloid leukemia. Haematologica 2008;93:633-634.
50. Sorror ML, Storer BE, Fathi AT, et al. Development and Validation of a Novel Acute Myeloid
Leukemia-Composite Model to Estimate Risks of Mortality. JAMA Oncol 2017;3:1675-1682.
51. Sorror ML, Storer BE, Elsawy M, et al. Intensive Versus Non-Intensive Induction Therapy for
Patients (Pts) with Newly Diagnosed Acute Myeloid Leukemia (AML) Using Two Different Novel
Prognostic Models. Blood 2016;128:216-216.
52. Krug U, Röllig C, Koschmieder A, et al. Complete remission and early death after intensive
chemotherapy in patients aged 60 years or older with acute myeloid leukaemia: a web-based
application for prediction of outcomes. Lancet 2010;376:2000-2008.
53. Kantarjian H, O'Brien S, Cortes J, et al. Results of intensive chemotherapy in 998 patients age 65
years or older with acute myeloid leukemia or high-risk myelodysplastic syndrome: predictive prognostic
models for outcome. Cancer 2006;106:1090-1098.
54. Abhishek Maiti, Courtney DiNardo, Tapan Kadia, et al. 10-DAY DECITABINE AND VENETOCLAX
(DEC10-VEN) VS. INTENSIVE CHEMOTHERAPY (IC) IN ACUTE MYELOID LEUKEMIA (AML): A PROPENSITY
|Page

SCORE MATCHED ANALYSIS STRATIFIED BY RISK OF TREATMENT-RELATED MORTALITY. EHA Library
2020.
55. El-Jawahri A, Abel GA, Traeger L, et al. Quality of life and mood of older patients with acute
myeloid leukemia (AML) receiving intensive and non-intensive chemotherapy. Leukemia 2019;33:2393-
2402.
56. Estey E, Gale RP. How good are we at predicting the fate of someone with acute myeloid
leukaemia? Leukemia 2017;31:1255-1258.
57. Maiti A, Rausch CR, Cortes JE, et al. Outcomes in Molecular Subgroups and Resistance Patterns
with Ten-Day Decitabine and Venetoclax (DEC10-VEN) in Acute Myeloid Leukemia. Blood 2019;134:645-
645.
58. Social Security Administration, Retirement & Survivors Benefits: Life Expectancy Calculator.
https://www.ssa.gov/OACT/population/longevity.html.
59. Seo J, Smith BD, Estey E, et al. Developing an instrument to assess patient preferences for
benefits and risks of treating acute myeloid leukemia to promote patient-focused drug development.
Curr Med Res Opin 2018;34:2031-2039.
60. Schiffer CA. An Important Gap in Informed Consent Documents for Oncology Clinical Trials: Lack
of Quantitative Details About Expected Treatment Outcomes. JAMA Oncol 2019.
61. Herold T, Rothenberg-Thurley M, Grunwald VV, et al. Validation and refinement of the revised
2017 European LeukemiaNet genetic risk stratification of acute myeloid leukemia. Leukemia 2020.
62. Angenendt L, Röllig C, Montesinos P, et al. Chromosomal Abnormalities and Prognosis in NPM1-
Mutated Acute Myeloid Leukemia: A Pooled Analysis of Individual Patient Data From Nine International
Cohorts. J Clin Oncol 2019;37:2632-2642.
63. Ostronoff F, Othus M, Lazenby M, et al. Prognostic significance of NPM1 mutations in the
absence of FLT3-internal tandem duplication in older patients with acute myeloid leukemia: a SWOG
and UK National Cancer Research Institute/Medical Research Council report. J Clin Oncol 2015;33:1157-
1164.
64. Papaemmanuil E, Gerstung M, Bullinger L, et al. Genomic Classification and Prognosis in Acute
Myeloid Leukemia. N Engl J Med 2016;374:2209-2221.
65. Prada-Arismendy J, Arroyave JC, Röthlisberger S. Molecular biomarkers in acute myeloid
leukemia. Blood Rev 2017;31:63-76.
66. Lamba JK, Cao X, Raimondi SC, et al. Integrated epigenetic and genetic analysis identifies
markers of prognostic significance in pediatric acute myeloid leukemia. Oncotarget 2018;9:26711-
26723.
67. Tyner JW, Tognon CE, Bottomly D, et al. Functional genomic landscape of acute myeloid
leukaemia. Nature 2018;562:526-531.
68. Haubner S, Perna F, Köhnke T, et al. Coexpression profile of leukemic stem cell markers for
combinatorial targeted therapy in AML. Leukemia 2019;33:64-74.
69. van Galen P, Hovestadt V, Wadsworth Ii MH, et al. Single-Cell RNA-Seq Reveals AML Hierarchies
Relevant to Disease Progression and Immunity. Cell 2019;176:1265-1281.e1224.
70. Bertoli S, Bérard E, Huguet F, et al. Time from diagnosis to intensive chemotherapy initiation
does not adversely impact the outcome of patients with acute myeloid leukemia. Blood 2013;121:2618-
2626.
|Page

71. Röllig C, Kramer M, Schliemann C, et al. Time from Diagnosis to Treatment Does Not Affect
Outcome in Intensively Treated Patients with Newly Diagnosed Acute Myeloid Leukemia. Blood
2019;134:13-13.
72. Percival MM, Othus M, Mirahsani S, et al. Survival of patients with newly diagnosed high-grade
myeloid neoplasms who do not meet standard trial eligibility. Haematologica 2020.
73. Montalban-Bravo G, Huang X, Naqvi K, et al. A clinical trial for patients with acute myeloid
leukemia or myelodysplastic syndromes not eligible for standard clinical trials. Leukemia 2017;31:318-
324.
74. Garcia-Manero G, Othus M, Pagel JM, et al. SWOG S1203: A Randomized Phase III Study of
Standard Cytarabine Plus Daunorubicin (7+3) Therapy Versus Idarubicin with High Dose Cytarabine (IA)
with or without Vorinostat (IA+V) in Younger Patients with Previously Untreated Acute Myeloid
Leukemia (AML). Blood 2016;128:901-901.
75. Burnett AK, Russell NH, Hills RK, et al. Optimization of chemotherapy for younger patients with
acute myeloid leukemia: results of the medical research council AML15 trial. J Clin Oncol 2013;31:3360-
3368.
76. Burnett AK, Russell NH, Hills RK. Higher daunorubicin exposure benefits FLT3 mutated acute
myeloid leukemia. Blood 2016;128:449-452.
77. Montalban-Bravo G, Kanagal-Shamanna R, Sasaki K, et al. NPM1 mutations define a specific
subgroup of MDS and MDS/MPN patients with favorable outcomes with intensive chemotherapy. Blood
Adv 2019;3:922-933.
78. Andrew H. Wei, Ing S. Tiong, Andrew W. Roberts, et al. CHEMOTHERAPY AND VENETOCLAX IN
ELDERLY AML TRIAL (CAVEAT): A PHASE 1B DOSE ESCALATION STUDY EXAMINING MODIFIED INTENSIVE
CHEMOTHERAPY IN FIT ELDERLY PATIENTS. European Hematology Association Library; 2018.
79. Aboudalle I, Konopleva MY, Kadia TM, et al. A Phase Ib/II Study of the BCL-2 Inhibitor Venetoclax
in Combination with Standard Intensive AML Induction/Consolidation Therapy with FLAG-IDA in Patients
with Newly Diagnosed or Relapsed/Refractory AML. Blood 2019;134:176-176.
80. Daigle SR, Choe S, Quek L, et al. High Rate of IDH1 Mutation Clearance and Measurable Residual
Disease Negativity in Patients with IDH1-Mutant Newly Diagnosed Acute Myeloid Leukemia Treated with
Ivosidenib (AG-120) and Azacitidine. Blood 2019;134:2706-2706.
81. DiNardo CD, Schuh AC, Stein EM, et al. Enasidenib Plus Azacitidine Significantly Improves
Complete Remission and Overall Response Compared with Azacitidine Alone in Patients with Newly
Diagnosed Acute Myeloid Leukemia (AML) with Isocitrate Dehydrogenase 2 (IDH2) Mutations: Interim
Phase II Results from an Ongoing, Randomized Study. Blood 2019;134:643-643.
82. Choe S, Wang H, DiNardo CD, et al. Molecular mechanisms mediating relapse following
ivosidenib monotherapy in IDH1-mutant relapsed or refractory AML. Blood Advances 2020;4:1894-1905.
83. Burd A, Levine RL, Ruppert AS, et al. Precision Medicine Treatment in Older AML: Results of Beat
AML Master Trial. Blood 2019;134:175-175.
84. Freireich EJ, Gehan EA, Sulman D, et al. The effect of chemotherapy on acute leukemia in the
human. J Chronic Dis 1961;14:593-608.
85. Othus M, Sekeres MA, Nand S, et al. Relative survival following response to 7 + 3 versus
azacytidine is similar in acute myeloid leukemia and high-risk myelodysplastic syndromes: an analysis of
four SWOG studies. Leukemia 2019;33:371-378.
|Page

86. Administration USFaD. FDA approves new targeted treatment for relapsed or refractory acute
myeloid leukemia. 2017.
87. Walter RB, Kantarjian HM, Huang X, et al. Effect of complete remission and responses less than
complete remission on survival in acute myeloid leukemia: a combined Eastern Cooperative Oncology
Group, Southwest Oncology Group, and M. D. Anderson Cancer Center Study. J Clin Oncol 2010;28:1766-
1771.
88. Øvlisen AK, Oest A, Bendtsen MD, et al. Stringent or nonstringent complete remission and
prognosis in acute myeloid leukemia: a Danish population-based study. Blood Adv 2018;2:559-564.
89. Chen X, Xie H, Estey EH. Reply to D. Przepiorka et al. J Clin Oncol 2015;33:3676-3677.
90. Boddu P, Jorgensen J, Kantarjian H, et al. Achievement of a negative minimal residual disease
state after hypomethylating agent therapy in older patients with AML reduces the risk of relapse.
Leukemia 2018;32:241-244.
91. Buckley SA, Wood BL, Othus M, et al. Minimal residual disease prior to allogeneic hematopoietic
cell transplantation in acute myeloid leukemia: a meta-analysis. Haematologica 2017;102:865-873.
92. Araki D, Wood BL, Othus M, et al. Allogeneic Hematopoietic Cell Transplantation for Acute
Myeloid Leukemia: Time to Move Toward a Minimal Residual Disease-Based Definition of Complete
Remission? J Clin Oncol 2016;34:329-336.
93. Chen X, Xie H, Wood BL, et al. Relation of clinical response and minimal residual disease and
their prognostic impact on outcome in acute myeloid leukemia. J Clin Oncol 2015;33:1258-1264.
94. Cornelissen JJ, Gratwohl A, Schlenk RF, et al. The European LeukemiaNet AML Working Party
consensus statement on allogeneic HSCT for patients with AML in remission: an integrated-risk adapted
approach. Nat Rev Clin Oncol 2012;9:579-590.
95. Holtan SG, DeFor TE, Lazaryan A, et al. Composite end point of graft-versus-host disease-free,
relapse-free survival after allogeneic hematopoietic cell transplantation. Blood 2015;125:1333-1338.
96. Schuurhuis GJ, Heuser M, Freeman S, et al. Minimal/measurable residual disease in AML: a
consensus document from the European LeukemiaNet MRD Working Party. Blood 2018;131:1275-1291.
97. Freeman SD, Hourigan CS. MRD evaluation of AML in clinical practice: are we there yet?
Hematology Am Soc Hematol Educ Program 2019;2019:557-569.
98. Jongen-Lavrencic M, Grob T, Hanekamp D, et al. Molecular Minimal Residual Disease in Acute
Myeloid Leukemia. N Engl J Med 2018;378:1189-1199.
99. Getta BM, Devlin SM, Levine RL, et al. Multicolor Flow Cytometry and Multigene Next-
Generation Sequencing Are Complementary and Highly Predictive for Relapse in Acute Myeloid
Leukemia after Allogeneic Transplantation. Biol Blood Marrow Transplant 2017;23:1064-1071.
100. Fang M, Storer B, Wood B, et al. Prognostic impact of discordant results from cytogenetics and
flow cytometry in patients with acute myeloid leukemia undergoing hematopoietic cell transplantation.
Cancer 2012;118:2411-2419.
101. Peter V. Molecular MRD detection by next generation sequencing in AML. In Press 2020.
102. Hochman MJ, Othus M, Shaw C, et al. Does intensity of induction chemotherapy affect the
impact of measurable residual disease (MRD) on prognosis in acute myeloid leukemia (AML)? Journal of
Clinical Oncology 2019;37:7031-7031.
|Page

103. Freeman SD, Virgo P, Couzens S, et al. Prognostic relevance of treatment response measured by
flow cytometric residual disease detection in older patients with acute myeloid leukemia. J Clin Oncol
2013;31:4123-4131.
104. Othus M, Wood BL, Stirewalt DL, et al. Effect of measurable ('minimal') residual disease (MRD)
information on prediction of relapse and survival in adult acute myeloid leukemia. Leukemia
2016;30:2080-2083.
105. Salk JJ, Schmitt MW, Loeb LA. Enhancing the accuracy of next-generation sequencing for
detecting rare and subclonal mutations. Nature reviews Genetics 2018;19:269-285.
106. Terwijn M, Zeijlemaker W, Kelder A, et al. Leukemic stem cell frequency: a strong biomarker for
clinical outcome in acute myeloid leukemia. PloS one 2014;9:e107587.
107. Godwin CD, Zhou Y, Othus M, et al. Comparison of Acute Myeloid Leukemia Measurable
Residual Disease Detection By Flow Cytometry in Peripheral Blood and Bone Marrow. Blood
2019;134:2729-2729.
108. Ivey A, Hills RK, Simpson MA, et al. Assessment of Minimal Residual Disease in Standard-Risk
AML. N Engl J Med 2016;374:422-433.
109. Yin JA, O'Brien MA, Hills RK, et al. Minimal residual disease monitoring by quantitative RT-PCR in
core binding factor AML allows risk stratification and predicts relapse: results of the United Kingdom
MRC AML-15 trial. Blood 2012;120:2826-2835.
110. Wheatley K, Gray R. Commentary: Mendelian randomization--an update on its use to evaluate
allogeneic stem cell transplantation in leukaemia. Int J Epidemiol 2004;33:15-17.
111. Østgård LSG, Lund JL, Nørgaard JM, et al. Impact of Allogeneic Stem Cell Transplantation in First
Complete Remission in Acute Myeloid Leukemia: A National Population-Based Cohort Study. Biol Blood
Marrow Transplant 2018;24:314-323.
112. Jourdan E, Boissel N, Chevret S, et al. Prospective evaluation of gene mutations and minimal
residual disease in patients with core binding factor acute myeloid leukemia. Blood 2013;121:2213-
2223.
113. Venditti A, Piciocchi A, Candoni A, et al. GIMEMA AML1310 trial of risk-adapted, MRD-directed
therapy for young adults with newly diagnosed acute myeloid leukemia. Blood 2019;134:935-945.
114. Zhu HH, Zhang XH, Qin YZ, et al. MRD-directed risk stratification treatment may improve
outcomes of t(8;21) AML in the first complete remission: results from the AML05 multicenter trial. Blood
2013;121:4056-4062.
115. Elsawy M, Sorror ML. Up-to-date tools for risk assessment before allogeneic hematopoietic cell
transplantation. Bone marrow transplantation 2016;51:1283-1300.
116. Sorror ML, Storb RF, Sandmaier BM, et al. Comorbidity-age index: a clinical measure of biologic
age before allogeneic hematopoietic cell transplantation. J Clin Oncol 2014;32:3249-3256.
117. Appelbaum FR. Impact of allogeneic hematopoietic cell transplantation on the outcome of older
patients with acute myeloid leukemia. Best practice & research Clinical haematology 2017;30:320-326.
118. Muffly L, Pasquini MC, Martens M, et al. Increasing use of allogeneic hematopoietic cell
transplantation in patients aged 70 years and older in the United States. Blood 2017;130:1156-1164.
119. McDonald GB, Sandmaier BM, Mielcarek M, et al. Survival, Nonrelapse Mortality, and Relapse-
Related Mortality After Allogeneic Hematopoietic Cell Transplantation: Comparing 2003-2007 Versus
2013-2017 Cohorts. Annals of internal medicine 2020;172:229-239.
|Page

120. Appelbaum FR. Alternative donor transplantation for adults with acute leukemia. Best practice
& research Clinical haematology 2014;27:272-277.
121. Lee SJ, Logan B, Westervelt P, et al. Comparison of Patient-Reported Outcomes in 5-Year
Survivors Who Received Bone Marrow vs Peripheral Blood Unrelated Donor Transplantation: Long-term
Follow-up of a Randomized Clinical Trial. JAMA Oncol 2016;2:1583-1589.
122. Pagel JM, Othus M, Garcia-Manero G, et al. Feasibility of Allogeneic Hematopoietic Cell
Transplantation Among High-Risk AML Patients in First Complete Remission: Results of the Transplant
Objective from the SWOG (S1203) Randomized Phase III Study of Induction Therapy Using Standard 7+3
Therapy or Idarubicin with High-Dose Cytarabine (IA) Versus IA Plus Vorinostat. Blood 2016;128:1166-
1166.
123. Freeman S, Craddock C. Less Is Not Necessarily More: Toward a Rational Selection of the
Conditioning Regimen in Acute Myeloid Leukemia. J Clin Oncol 2020;38:1249-1251.
124. Hourigan CS, Dillon LW, Gui G, et al. Impact of Conditioning Intensity of Allogeneic
Transplantation for Acute Myeloid Leukemia With Genomic Evidence of Residual Disease. J Clin Oncol
2020;38:1273-1283.
125. Eapen M, Brazauskas R, Hemmer M, et al. Hematopoietic cell transplant for acute myeloid
leukemia and myelodysplastic syndrome: conditioning regimen intensity. Blood Adv 2018;2:2095-2103.
126. Deeg HJ, Stevens EA, Salit RB, et al. Transplant Conditioning with Treosulfan/Fludarabine with or
without Total Body Irradiation: A Randomized Phase II Trial in Patients with Myelodysplastic Syndrome
and Acute Myeloid Leukemia. Biol Blood Marrow Transplant 2018;24:956-963.
127. Jurcic JG. Targeted Alpha-Particle Therapy for Hematologic Malignancies. Seminars in nuclear
medicine 2020;50:152-161.
128. Maziarz RTT, Patnaik MM, Scott BL, et al. Radius: A Phase 2 Randomized Trial Investigating
Standard of Care ± Midostaurin after Allogeneic Stem Cell Transplant in FLT3-ITD-Mutated AML. Blood
2018;132:662-662.
129. Craddock C, Jilani N, Siddique S, et al. Tolerability and Clinical Activity of Post-Transplantation
Azacitidine in Patients Allografted for Acute Myeloid Leukemia Treated on the RICAZA Trial. Biol Blood
Marrow Transplant 2016;22:385-390.
130. de Lima M, Oran B, Champlin RE, et al. CC-486 Maintenance after Stem Cell Transplantation in
Patients with Acute Myeloid Leukemia or Myelodysplastic Syndromes. Biol Blood Marrow Transplant
2018;24:2017-2024.
131. Chapuis AG, Egan DN, Bar M, et al. T cell receptor gene therapy targeting WT1 prevents acute
myeloid leukemia relapse post-transplant. Nat Med 2019;25:1064-1072.
132. Lazarus HM, El Jurdi N. Autologous hematopoietic cell transplantation for adult acute myeloid
leukemia: An obsolete or resurfacing concept? Best practice & research Clinical haematology
2017;30:327-332.
133. Ganzel C, Rowe JM. Revisiting autologous transplantation in acute myeloid leukemia. Current
opinion in hematology 2018;25:95-102.
134. Cornelissen JJ, Blaise D. Hematopoietic stem cell transplantation for patients with AML in first
complete remission. Blood 2016;127:62-70.
|Page

135. Schlenk RF, Taskesen E, van Norden Y, et al. The value of allogeneic and autologous
hematopoietic stem cell transplantation in prognostically favorable acute myeloid leukemia with double
mutant CEBPA. Blood 2013;122:1576-1582.
136. Burnett AK, Goldstone A, Hills RK, et al. Curability of patients with acute myeloid leukemia who
did not undergo transplantation in first remission. J Clin Oncol 2013;31:1293-1301.
137. Jaramillo S, Benner A, Krauter J, et al. Condensed versus standard schedule of high-dose
cytarabine consolidation therapy with pegfilgrastim growth factor support in acute myeloid leukemia.
Blood cancer journal 2017;7:e564.
138. Löwenberg B. Sense and nonsense of high-dose cytarabine for acute myeloid leukemia. Blood
2013;121:26-28.
139. Thomas X, de Botton S, Chevret S, et al. Randomized Phase II Study of Clofarabine-Based
Consolidation for Younger Adults With Acute Myeloid Leukemia in First Remission. J Clin Oncol
2017;35:1223-1230.
140. Wei AH. Maintenance therapy for AML: are we there yet? Blood 2019;133:1390-1392.
141. Huls G, Chitu DA, Havelange V, et al. Azacitidine maintenance after intensive chemotherapy
improves DFS in older AML patients. Blood 2019;133:1457-1464.
142. Wei AH, Döhner H, Pocock C, et al. The QUAZAR AML-001 Maintenance Trial: Results of a Phase
III International, Randomized, Double-Blind, Placebo-Controlled Study of CC-486 (Oral Formulation of
Azacitidine) in Patients with Acute Myeloid Leukemia (AML) in First Remission. Blood 2019;134:LBA-3-
LBA-3.
143. Estey E, Pierce S. Routine bone marrow exam during first remission of acute myeloid leukemia.
Blood 1996;87:3899-3902.
144. Zhou Y, Wood BL, Walter RB, et al. Is there a need for morphologic exam to detect relapse in
AML if multi-parameter flow cytometry is employed? Leukemia 2017;31:2536-2537.
145. Othus M, Mukherjee S, Sekeres MA, et al. Prediction of CR following a second course of '7+3' in
patients with newly diagnosed acute myeloid leukemia not in CR after a first course. Leukemia
2016;30:1779-1780.
146. Othus M, Estey EH, Garcia-Manero G, et al. Second cycle remission achievement with 7+3 and
survival in adults with newly diagnosed acute myeloid leukemia: analysis of recent SWOG trials.
Leukemia 2019;33:554-558.
147. Ravandi F, Cortes J, Faderl S, et al. Characteristics and outcome of patients with acute myeloid
leukemia refractory to 1 cycle of high-dose cytarabine-based induction chemotherapy. Blood
2010;116:5818-5823; quiz 6153.
148. Walter RB, Othus M, Löwenberg B, et al. Empiric definition of eligibility criteria for clinical trials
in relapsed/refractory acute myeloid leukemia: analysis of 1,892 patients from HOVON/SAKK and
SWOG. Haematologica 2015;100:e409-411.
149. Thol F, Schlenk RF, Heuser M, et al. How I treat refractory and early relapsed acute myeloid
leukemia. Blood 2015;126:319-327.
150. Breems DA, Van Putten WL, Huijgens PC, et al. Prognostic index for adult patients with acute
myeloid leukemia in first relapse. J Clin Oncol 2005;23:1969-1978.
|Page

151. Roboz GJ, Rosenblat T, Arellano M, et al. International randomized phase III study of
elacytarabine versus investigator choice in patients with relapsed/refractory acute myeloid leukemia. J
Clin Oncol 2014;32:1919-1926.
152. Kantarjian HM, DiNardo CD, Nogueras-Gonzalez GM, et al. Results of second salvage therapy in
673 adults with acute myelogenous leukemia treated at a single institution since 2000. Cancer
2018;124:2534-2540.
153. Kharfan-Dabaja MA, Labopin M, Polge E, et al. Association of Second Allogeneic Hematopoietic
Cell Transplant vs Donor Lymphocyte Infusion With Overall Survival in Patients With Acute Myeloid
Leukemia Relapse. JAMA Oncol 2018;4:1245-1253.
154. Perl AE, Martinelli G, Cortes JE, et al. Gilteritinib or Chemotherapy for Relapsed or Refractory
FLT3-Mutated AML. N Engl J Med 2019;381:1728-1740.
155. Daver N, Schlenk RF, Russell NH, et al. Targeting FLT3 mutations in AML: review of current
knowledge and evidence. Leukemia 2019;33:299-312.
156. DeAngelo DJ, Jonas BA, Liesveld JL, et al. High E-Selectin Ligand Expression Contributes to
Chemotherapy-Resistance in Poor Risk Relapsed and Refractory (R/R) Acute Myeloid Leukemia (AML)
Patients and Can be Overcome with the Addition of Uproleselan. Blood 2019;134:2690-2690.
157. Velasquez MP, Bonifant CL, Gottschalk S. Redirecting T cells to hematological malignancies with
bispecific antibodies. Blood 2018;131:30-38.
158. Subklewe M, Stein A, Walter RB, et al. Preliminary Results from a Phase 1 First-in-Human Study
of AMG 673, a Novel Half-Life Extended (HLE) Anti-CD33/CD3 BiTE® (Bispecific T-Cell Engager) in Patients
with Relapsed/Refractory (R/R) Acute Myeloid Leukemia (AML). Blood 2019;134:833-833.
159. Gill SI. How close are we to CAR T-cell therapy for AML? Best practice & research Clinical
haematology 2019;32:101104.
160. Daver N, Garcia-Manero G, Basu S, et al. Efficacy, Safety, and Biomarkers of Response to
Azacitidine and Nivolumab in Relapsed/Refractory Acute Myeloid Leukemia: A Nonrandomized, Open-
Label, Phase II Study. Cancer Discov 2019;9:370-383.
161. Sallman DA, Asch AS, Al Malki MM, et al. The First-in-Class Anti-CD47 Antibody Magrolimab
(5F9) in Combination with Azacitidine Is Effective in MDS and AML Patients: Ongoing Phase 1b Results.
Blood 2019;134:569-569.
162. Gyurkocza B, Nath R, Stiff PJ, et al. Re-Induction and Targeted Conditioning with Anti-CD45
Iodine (131I) Apamistamab [Iomab-B] Leads to High Rates of Transplantation and Successful
Engraftment in Older Patients with Active, Relapsed or Refractory (rel/ref) AML after Failure of
Chemotherapy and Targeted Agents: Preliminary Midpoint Results from the Prospective, Randomized
Phase 3 Sierra Trial. Blood 2019;134:5642-5642.
163. Appelbaum J, Loeb A, Gardner K, et al. Additional Cytotoxic Chemotherapy Is Unlikely to
Eliminate Measurable Residual Acute Myeloid Leukemia (AML). Blood 2019;134:260-260.
164. Halpern AB, Howard NP, Othus M, et al. Early hospital discharge after intensive induction
chemotherapy for adults with acute myeloid leukemia or other high-grade myeloid neoplasm. Leukemia
2020;34:635-639.
165. Mabrey FL, Gardner KM, Shannon Dorcy K, et al. Outpatient intensive induction chemotherapy
for acute myeloid leukemia and high-risk myelodysplastic syndrome. Blood Adv 2020;4:611-616.
|Page

166. Team TUoCHMCR. How I diagnose and manage individuals at risk for inherited myeloid
malignancies. Blood 2016;128:1800-1813.
167. Godley LA, Shimamura A. Genetic predisposition to hematologic malignancies: management and
surveillance. Blood 2017;130:424-432.
168. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to the World Health Organization
classification of myeloid neoplasms and acute leukemia. Blood 2016;127:2391-2405.
169. Lindsley RC, Mar BG, Mazzola E, et al. Acute myeloid leukemia ontogeny is defined by distinct
somatic mutations. Blood 2015;125:1367-1376.
170. Walter MJ. What came first: MDS or AML? Blood 2015;125:1357-1358.
|Page

Accepted Article
254x190mm (200 x 200 DPI)

Accepted Article
254x190mm (200 x 200 DPI)

Accepted Article
338x190mm (140 x 119 DPI)

Accepted Article
338x190mm (77 x 77 DPI)

Accepted Article

Accepted Article

Accepted Article

Accepted Article
Fo
rP
ee
rR
ev
iew

Accepted Article
Fo
rP
ee
rR
ev
iew

Accepted Article
254x190mm (200 x 200 DPI)

Accepted Article

Accepted Article
Fo
rP
ee
rR
ev
iew

Accepted Article

Accepted Article
Fo
rP
ee
rR
ev
iew

Accepted Article

Accepted Article

Accepted Article

Table 1 Indications for and issues with recently FDA approved drugs
Drug Should be considered standard in Specific issues

Venetoclax + Patients considered unfit for Criteria for unfitness;
azacytidine intensive induction based on age ≥ Might some do better with more
75 or comorbidities intensive induction?
CPX-351 Fit patients with secondary AML age Value in younger patients with
60-75 secondary AML?
Midostaurin In combination with 7+3 in fit Value of maintenance with midostaurin
patients up to age 70 with FLT3 ITD seems likely but awaits confirmation in
Accepted Article
or TKD regardless of allelic ratio randomized trial

Gemtuzumab In combination with 7+3 or FLAG in Value in patients with mutations not
ozogamicin fit patients in ELN 2017 “favorable” considered in ELN 2017
or “intermediate” groups
Ivosidenib As with venetoclax + azacytdine but As with venetoclax + azacytidine;
with IDH1 mutation Lack of randomized trial, for example
vs. venetoclax + azacytidine; value of
CRh vs value of CR?

Table 2 2017 ELN Prognostic groups
Accepted Article
Frequencies, response rates, and outcome measures should be reported by risk category, and,
if sufficient numbers are available, by specific genetic lesions indicated.
*Prognostic impact of a marker is treatment-dependent and may change with new therapies.
†Low, low allelic ratio (,0.5); high, high allelic ratio ($0.5); semiquantitative assessment of FLT3-
ITD allelic ratio (using DNA fragment analysis) is determined as ratio of the area under the curve
“FLT3-ITD” divided by area under the curve “FLT3- wild type”; recent studies indicate that AML
with NPM1 mutation and FLT3-ITD low allelic ratio may also have a more favorable prognosis
and patients should not routinely be assigned to allogeneic HCT.57-59,77
‡The presence of t(9;11)(p21.3;q23.3) takes precedence over rare, concurrent adverse-risk gene
mutations.
§Three or more unrelated chromosome abnormalities in the absence of 1 of the WHO-designated
recurring translocations or inversions, that is, t(8;21), inv(16) or t(16;16), t(9;11), t(v;11)(v;q23.3),
t(6;9), inv(3) or t(3;3); AML with BCR-ABL1.
||Defined by the presence of 1 single monosomy (excluding loss of X or Y) in association with at
least 1 additional monosomy or structural chromosome abnormality (excluding core-binding factor
AML).116
{These markers should not be used as an adverse prognostic marker if they cooccur with
favorable-risk AML subtypes.
#TP53 mutations are significantly associated with AML with complex and monosomal
karyotype.37,66-69

Table 3 Recommendations for initial induction therapy
ELN2017 group Treatment-related mortality TRM risk higher a,c
(TRM) lower a,b
“Favorable” d 7+3 + GOe or FLAG-ida + Venetoclax + HMA +/- GO
GOe
Intermediate As above or CPX 351 if As above;
secondary AML Clinical trial, for example
Clinical trial, for example 5+2 venetoclax + HMA +
or FLAG-ida each + ivosidenib or enasidenib etc.
venetoclax
Accepted Article
Adverse Clinical trial, for example 5+2 Clinical trial, for example
or FLAG-ida each + “BEAT AML”
venetoclaxe
a consider using validated model to compute probability TRM

b most commonly age < 75, PS < 2, no cardiac, renal, or hepatic dysfunction, albumin > 3.5
c most commonly age >70 and PS 2-4, and/or one of above co-morbidities,
d patients with NPM1 mutation and FLTITD allelic ratio < 0.5 and adverse cytogenetics should
be considered adverse group

e consider adding midostaurin if accompanying FLT ITD or TKD

Table 4 Ongoing clinical trials in newly-diagnosed AML
Class of trial Examples Preliminary findings References
Combinations of Venetoclax + 5+2 37 pts, age > 60; venetoclax Wei et al.78
more intense and dose escalation ongoing;
less intense
induction in fit
patients
11 pts, median age 47(16-72);
Venetoclax + FLAG- 9 CR, 1 CRh, 10/11 MRD Aboudalle et al.79
ida negative
Accepted Article
Addition of targeted Ivosidenib + 23 pts, age > 60 ; CR 14/23, Daigle at al.80

therapy to azacytidine in IDH1- CRi 2/23, IDH1 mutation
azacytidine in unfit mutated AML cleared in blood in 12/16
patients CR/CRh vs. 1/7 others
68 pts randomized to DiNardo et al.81

azacytidine +/- combination, 33 to aza alone;
enasidenib in IDH2- CR rates 50% vs. 12%, “ORR”
mutated AML rates 68% vs. 42%; no
statistical difference in
response duration, no survival
data
“Precision medicine” “BEAT AML” 57% of patients assigned to a Burd83
specific sub-study (including
for “marker negative” patients
within 7 days; survival similar
in BEAT AML subgroups and
non-specific clinical trial group

Table 5 Means to quantify MRD
Test Sensitivity
Multiparameter flow cytometry (MFC) Between 1 in 1,000 [10(-3] and 1 in 10,000[10(-4)
Cytogenetics 1 in 20 (<5%) or poorer
Fluorescent in situ hybridization 1 in 500 {2 X 10(-3)]
“Molecular”
qPCR [NPM1,BCR-ABL, CBFB/MYH11, Between 1 in 10,000[10(-4)] and 1 in 1 million [10(-6)]
RUNX1-RUNX1T, PML-RARA]
1 in 100 [10(-2≤)]
NGS [others]
Accepted Article

Table 6 Deciding on allo HCT: ELN 2017 and MRD to assess risk of relapse and HCT-CI to
assess risk of non-relapse mortality (NRM) post HCT
Pre-treatment MRD MRD status Risk Risk of NRM Possible treatments if
ELN 2017 Risk detected after 1-2 relapse at 2 justifying allo HCT risk/benefit excessive with
by cycles years (HCT-CI score allo HCT
therapy without allo associated with
HCT that risk)
Favorable PCR Negative(70% 10-20% Too risky Cytarabine 1-2 g/m2 QD or
(CBF) of pts) regardless of HCT- BID for 3 days X 3 cycles;
CI score add midostaurin if FLT3 +,
Accepted Article
consider adding GO ;
azacitidine
Clinical trial (see section on

Positive 50% ≤20% relapsed/refractory AML);
Favorable (≤2) azacitidine
(NPM+/FLT3 -) PCR
Cytarabine 1-2 g/m2 QD or
Negative(90% 25% ≤10% BID days + 1-3 X 3 cycles;
of pts) (≤1) add GO;
azacitidine
Positive 70% ≤40% Clinical trial (see section on

(≤3) relapsed/refractory AML);
azacitidine
Intermediate MFC Negative 25% ≤10% Autologous HCT;

(80% of pts) (≤1) Cytarabine 1-2 g/m2 QD or
BID days 1-3 X 3 cycles;
azacitidine
Positive 80% ≤40% Clinical trial (see section on

(≤3) relapsed/refractory AML);
azacitidine
Adverse MFC Negative 70- 80% ≤40% Clinical trial (see section on
(66% of pts) (≤3) relapsed/ref AML )
CR azacitidine
Positive 90-100% ≤ 40%

(≤3) Clinical trial (see section on
relapsed/refractory AML);
azacitidine

Table 7 Prognostic scoring system for AML in first relapse “favorable” group 1-6 points,
intermediate group 7-9 points, adverse group 10-14 points (modified from reference 151 to
include primary refractory AML)
Covariate Value Points

CR 1 duration >18 mos. 0
6-18 mos. 3
< 6 mos., including PREF 5
Accepted Article
Initial cytogenetics Inv 16 or t(16;16) 0

t(8;21) 3
Other 5
Age < 36 0
35-45 1
45-60 2
HCT in CR 1 No 0
Yes 2

Table 8 Management of relapsed/ primary refractory AML
Number of prior re- Breems score Re-induction therapy If CR is achieved
induction attempts (alternatively, prior
CR duration)
0 (1st salvage) 1-6 points Similar to initial Allo HCT if feasible,
induction If not, DLI or auto
(>18 months) HCT if MRD negative
Accepted Article
0 (1st salvage) 7-9 points As above or new As above

(12-18 months) drugs/clinical trial
0 (1st salvage) 10-14 points New drugs/clinical As above

< 12 months trial, possibly
including allo HCT
for re-induction
1 or more (> 1st N/A New drugs/clinical As above
salvage) trial, possibly
including allo HCT
for re-induction

Estey 2020

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Estey 2020

Uploaded by

Copyright:

Available Formats

Acute myeloid leukemia: 2021 update on risk-stratification and management

Elihu H Estey, MD1,2

1. Division of Hematology, University of Washington, Seattle WA USA

This article is protected by copyright. All rights reserved.

Conflict of Interest: No conflict

This article is protected by copyright. All rights reserved.

Purely palliative approach

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

New induction therapies that should now be considered conventional/standard (table 1)

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

Gemtuzumab ozogamicin (GO)

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

The issue of “fitness”

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

“Can I wait until cytogenetics and mutational data become available?”

Recommendations for initial induction therapy (table 3)

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

Increasing use of allo HCT

This article is protected by copyright. All rights reserved.

Preventing post allo HCT relapse

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

Autologous (auto) HCT

Patients who are not plausible HCT candidates (table 6)

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

RELAPSED AND REFRACTORY AML

There is plausible therapeutic value in detecting relapse by MRD before appearance of

This article is protected by copyright. All rights reserved.

Management of relapsed/refractory AML as defined by morphology (5% blasts or more)

This article is protected by copyright. All rights reserved.

Newer therapies for morphologically relapsed AML

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

Another example of this approach is magrolimab an antibody targeting CD47, a macrophage

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

Testing for germline mutations

Distinction between AML and MDS

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.