Therapy-related myeloid neoplasms: Management and prognosis

INTRODUCTION — 

Therapy-related myeloid neoplasms (t-MNs) refer to cases of acute myeloid leukemia (AML), myelodysplastic syndromes/neoplasms (MDS), and MDS/myeloproliferative neoplasms (MDS/MPN) that arise following exposure to cytotoxic chemotherapy or radiation therapy for an unrelated disorder. The designation of t-MN does not identify a distinct disorder, but it is considered a "diagnostic qualifier" in contemporary diagnosis and classification schemes.

Most patients with a t-MN have inferior outcomes compared with the corresponding de novo AML, primary MDS, or MDS/MPN. Abnormal karyotype and/or gene mutations are found in most cases of t-MNs, and the frequent unfavorable genetic features contribute to their inferior prognosis.

Management and prognosis of t-MNs are discussed in this topic.

Related topics include:

●(See "Therapy-related myeloid neoplasms: Epidemiology, causes, evaluation, and diagnosis".)

●(See "Acute myeloid leukemia in adults: Overview".)

●(See "Myelodysplastic syndromes/neoplasms (MDS): Overview of diagnosis and management".)

OVERVIEW — 

t-MNs arise after prior exposure to cytotoxic chemotherapy or radiation therapy for an unrelated disorder, either a malignant or nonmalignant condition.

The term t-MN refers to a spectrum of disorders that includes therapy-related acute myeloid leukemia (AML), myelodysplastic syndromes/neoplasms (MDS), and MDS/myeloproliferative neoplasms (MDS/MPN). The incidence of t-MNs is rising as the number of cancer survivors increases.

An overview of the management of t-MNs is provided below. (See 'Management overview' below.)

Incidence — t-MNs account for an estimated 10 to 15 percent of cases of AML [1,2]. Most cases manifest adverse cytogenetic/molecular features (>90 percent have an abnormal karyotype), which contribute to unfavorable outcomes [3-5]. Although survival outcomes are generally worse for patients with t-MNs compared with the corresponding de novo disorders [6], this may not apply to patients with a t-MN that has favorable cytogenetic features [2].

Details of diagnosis, classification, causes, and epidemiology of t-MNs are presented separately. (See "Therapy-related myeloid neoplasms: Epidemiology, causes, evaluation, and diagnosis".)

Spectrum of disorders — t-MNs comprise a spectrum of myeloid malignancies that include t-AML, t-MDS, and t-MDS/MPN. Most cases of t-MNs have inferior outcomes in association with unfavorable genetic features.

The contemporary classification systems for myeloid malignancies differ modestly in the labels and criteria that they apply for various disease categories. Use of either the International Consensus Classification (ICC) [7] or the World Health Organization 5th edition (WHO5) [8] is acceptable. The ICC and WHO5 supersede earlier classification systems. (See "Classification of hematopoietic neoplasms".)

The ICC labels myeloid neoplasms with 10 to 19 percent blasts as MDS/AML, while WHO5 labels them as either MDS or AML according to defining genetic features and/or the degree of morphologic differentiation. Other differences in the labels and categories that ICC and WHO5 apply to myeloid neoplasms are discussed separately. (See "Acute myeloid leukemia: Classification".)

EMERGENCIES — 

Therapy-related acute promyelocytic leukemia (t-APL) is a medical emergency that requires prompt recognition and urgent treatment. Clinical presentation of a t-MN can be associated with various other emergencies.

●t-APL – t-APL is a distinct category of t-MN that constitutes a medical emergency.

t-APL is distinguished from other subtypes of t-MN by its unique clinical manifestations and complications, the characteristic t(15;17) karyotype, and PML::RARA gene rearrangement. t-APL is more likely in patients with bleeding or extensive bruising when the malignant cells have coarse or dense cytoplasmic granules and with a low white blood cell (WBC) count and/or few circulating leukemic cells. (See "Clinical manifestations, pathologic features, and diagnosis of acute promyelocytic leukemia in adults", section on 'Clinical features'.)

Treatment of t-APL is like that for de novo APL. Urgent management that includes all-trans retinoic acid (ATRA), with or without arsenic trioxide or chemotherapy, is required, as discussed separately. (See "Initial treatment of acute promyelocytic leukemia in adults".)

●Other emergencies – Patients with cardiac, respiratory, or other organ system decompensation; >50,000 leukocytes/microL; and tumor lysis syndrome or disseminated intravascular coagulation (DIC) require urgent evaluation and management, as discussed separately. (See "Acute myeloid leukemia: Overview of complications" and "Hyperleukocytosis and leukostasis in hematologic malignancies".)

PRETREATMENT — 

The pretreatment evaluation should assess fitness for intensive treatments based on comorbidities, adverse effects (AEs) of the prior illness, and previous treatments.

All patients with a t-MN had an illness that required prior treatment with a deoxyribonucleic acid (DNA)-damaging agent. It is especially important to assess comorbid conditions because management may be constrained by AEs from the prior illness and/or its treatment. Some patients have persistence of the earlier illness.

Evaluation — Evaluation and diagnosis of a patient with a suspected t-MN are discussed separately. (See "Therapy-related myeloid neoplasms: Epidemiology, causes, evaluation, and diagnosis".)

Pretreatment evaluation includes:

●Clinical

•History should evaluate fatigue, dyspnea, infections, bleeding, and other symptoms related to cytopenias or other aspects of the t-MN, the prior condition, or other comorbid conditions.

Previous chemotherapy and/or radiation therapy (RT) should be documented, including the cumulative dose of anthracyclines.

•Physical examination should evaluate patients for comorbidities, including heart, lung, kidney, and liver dysfunction, that may have arisen from prior treatments.

●Laboratory

•Hematology – Complete blood count (CBC) with leukocyte differential count, review of a blood smear, prothrombin time (PT), partial thromboplastin time (PTT), fibrinogen.

•Chemistries – Comprehensive metabolic panel, lactate dehydrogenase, uric acid, phosphorus.

High numbers of metabolically active circulating leukocytes can interfere with certain laboratory tests. Resulting laboratory anomalies and other aspects of the pretreatment evaluation are discussed separately. (See "Acute myeloid leukemia: Clinical manifestations, pathologic features, and diagnosis", section on 'Clinical and laboratory'.)

●Imaging – We assess cardiac function by echocardiogram or radionuclide ventriculogram.

Other imaging is not routinely required unless it is needed to evaluate neurologic abnormalities or evidence of extramedullary disease.

Medical fitness — Fitness for therapy is an important determinant of management.

We classify patients functionally as one of the following categories while recognizing that each includes a range of physical functions and medical conditions:

●Medically fit – The patient is suitable for treatment with intensive therapies, such as induction chemotherapy for acute myeloid leukemia (AML) and/or allogeneic hematopoietic cell transplantation (HCT).

Medically fit patients should be evaluated for eligibility to undergo allogeneic HCT. Criteria for transplant eligibility are discussed separately. (See "Allogeneic hematopoietic cell transplantation: Indications, eligibility, and prognosis".)

If the patient is eligible for transplantation, a search for a suitable donor should be undertaken promptly. (See "Donor selection for hematopoietic cell transplantation".)

●Less fit – The patient has moderate comorbidities but can tolerate less intensive treatments.

●Frail – The patient has significant debility due to severe comorbidities or advanced age and can tolerate only low-intensity treatments or supportive care.

Some patients have only modest, recent, or transient impairment of functional status, while others have substantial comorbidities, cognitive impairment, or other conditions that may affect their ability to tolerate treatment. The t-MN, itself, can impair fitness, so achieving disease control may improve fitness in some patients; conversely, treatment-related AEs may lead to a functional decline.

The following instruments can be useful adjuncts for assessing medical fitness:

●Performance status (PS) – Eastern Cooperative Oncology Group (ECOG) scale (table 1) or Karnofsky performance status (KPS) (table 2).

●Comorbid illnesses – We favor the hematopoietic cell transplantation-specific comorbidity index (HCT-CI) (table 3) to assess the impact of comorbid illnesses. Some clinicians use the modified Charlson comorbidity index (CCI) (table 4).

A scoring system for patients with acute leukemia can be used to estimate early mortality from intensive induction therapy [9]. Some experts apply the Ferrara consensus criteria for evaluating fitness in patients with AML [10]. However, no fitness model estimates whether an individual will benefit more from intensive therapy versus less intensive therapy.

In selected older patients, a comprehensive geriatric consultation can aid assessment of fitness for treatment. Pretreatment evaluation of an older adult with AML is described separately. (See "Pretreatment evaluation and prognosis of acute myeloid leukemia in older adults".)

Goals of care — Goals of care should be evaluated at diagnosis and reassessed periodically through the course of the disease.

Goals of care are informed by medical fitness and the patient's values and preferences. (See 'Medical fitness' above.):

●Medically fit – We generally seek to achieve long-term survival, while alleviating symptoms and improving quality of life (QOL). (See 'Management overview' below.)

●Less fit – We seek to alleviate symptoms and prolong survival with improved QOL. (See 'Less-fit patients' below.)

●Frail – For some patients with extreme debility, advanced age, or severe comorbid conditions, treatment aimed at modifying the disease course may be inadvisable. Patients can receive supportive care (eg, blood transfusions, antibiotics), with or without other means for providing symptom relief. (See 'Frail patients' below.)

Goals should be established using shared decision-making by clinicians and an informed patient. Mutual understanding facilitates goal setting and decisions regarding health care proxies, do-not-resuscitate directives, and end-of-life decisions.

The adverse prognosis of t-MNs should be discussed with the patient. The patient should understand that this is a life-ending disease for most while emphasizing the potential short- and long-term benefits of treatment. Many patients with hematologic malignancies overestimate their long-term prognosis and do not recall being offered more than one approach to treatment [11,12].

PROGNOSIS — 

Patient age and genetic features of the t-MN are important prognostic factors that should be considered prior to treatment.

Each patient is assigned a prognostic category based on the category of disease and the cytogenetic and molecular features of the malignant cells. (See 'Prognostic categories' below.)

Prognostic factors — Age, cytogenetics, and molecular features are important prognostic factors for patients with a t-MN.

Outcomes are influenced by patient characteristics (eg, age, comorbid illnesses) and by cytogenetic and/or molecular features of the malignant cells.

Most patients with a t-MN survive less than one year, and five-year overall survival (OS) is generally <10 percent [6,13-15]. According to leukemia registry studies, t-MNs account for 6 to 8 percent of cases of acute myeloid leukemia (AML); outcomes with t-MNs are generally inferior to those with de novo AML [6,16,17]. Analysis of data from a Swedish tumor registry suggests that outcomes for patients with t-MNs have improved over time [2].

●Age – Increased age is a negative prognostic factor in patients with a t-MN.

Age ≥60 years was associated with inferior survival (hazard ratio [HR] 0.526 [95% CI 0.328-0.842]) according to multivariable analysis of 244 patients with a t-MN [13]. In a study that included 281 patients with therapy-related myelodysplastic syndromes/neoplasms (t-MDS), age ≥65 years was associated with an increased risk for death (HR 1.63 [95% CI 9.7-15.3]) [18].

●Genetic features – Cytogenetic and molecular features (table 5) are the strongest predictors of outcomes with t-MNs. Aberrant genetic features of the malignant cells are discussed separately. (See "Therapy-related myeloid neoplasms: Epidemiology, causes, evaluation, and diagnosis", section on 'Genetic findings'.)

A retrospective study reported that outcomes for patients with intermediate-risk or adverse-risk t-MNs are inferior to intermediate- or adverse-risk de novo AML, but patients with favorable-risk features had comparable outcomes regardless of whether they had t-AML or de novo AML [19]. Compared with intermediate- and adverse-risk de novo AML, t-MNs in the same risk categories had inferior five-year OS (28 versus 44 percent); by contrast, patients with favorable-risk t-AML and de novo AML had comparable five-year OS (49 versus 56 percent). A Swedish registry also reported that outcomes were similar for favorable-risk t-AML and de novo AML, while patients with intermediate- or adverse-risk t-AML had inferior outcomes compared with the corresponding risk groups with de novo AML [2].

•Cytogenetics – Adverse cytogenetic features include abnormalities of chromosomes 5 and/or 7 and complex karyotype. In general, adverse cytogenetic findings in cases of t-MNs are associated with median OS <1 year [6,13,14,20,21].

-A report of 306 patients with a t-MN reported that median OS was best in patients with a normal karyotype or a balanced rearrangement (approximately 11 months each) and was worse with abnormalities of both chromosomes 5 and 7 (approximately five months) [14].

-Analysis of 511 cases of t-MNs also reported better outcomes in patients with favorable cytogenetic features [22]. Patients with inv(16) and 21q22 had 28 and 14 months median OS, respectively, while an unfavorable feature, 11q23, was associated with an 8-month median OS.

•Molecular – Mutations of TP53 are reported in one-quarter to one-half of cases of t-MNs and are associated with inferior survival. Complex karyotypes are common in patients with mutated TP53.

A study of 217 patients with a t-MN reported TP53 mutations in 23 percent of patients; other studies reported similar rates [23-25]. In a study of 95 patients, multivariate analysis demonstrated that mutant TP53 was an independent prognostic marker for inferior OS [21].

The prognostic significance of other mutations that are frequently seen in de novo myeloid malignancies is not well-defined for t-MNs. Examples include mutations that are routinely evaluated for risk stratification in de novo AML (ie, FLT3-ITD, CEBPA, NPM1), other mutations associated with de novo AML (eg, TET2, PTPN11, IDH1/2, N-RAS), and genes that are commonly mutated in de novo MDS (eg, SRSF2, SF3B1, U2AF1, ZRSR2, ASXL1, EZH2, BCOR, STAG2) [4,5,21].

●Other factors

•Disease classification – t-MNs comprise a continuum of diseases, and outcomes may vary according to the disease type.

A study of 155 patients with t-MNs reported that survival rates were not associated with the specific morphologic category of t-MN; rather, outcomes in this study were primarily affected by cytogenetic abnormalities [26].

The diagnostic distinction between t-MDS and t-AML is described separately. (See "Therapy-related myeloid neoplasms: Epidemiology, causes, evaluation, and diagnosis", section on 'Diagnosis and classification'.)

•Prior therapy – Compared with prior chemotherapy, t-MNs that arise after radiation therapy had a higher frequency of favorable cytogenetic abnormalities (65 versus 12 percent) and better survival rates [27].

Patients who received >1 prior cytotoxic therapy had inferior survival in a retrospective study [19].

•Prior malignancy – The status of the prior malignancy contributes to outcomes in this population.

In a retrospective study, patients with a t-MN and an active prior cancer had inferior survival [19]. In a study of 411 patients with a t-MN, a difference between t-MN-specific survival and OS (median 17 versus 10.3 months, respectively) appeared to reflect deaths associated with the earlier malignancies [28].

•Germline/heritable genetic variants – Patients with an inherited hematologic disorder or a germline pathogenic variant who develop a t-MN may have additional reasons for unfavorable outcomes.

Clinical features (eg, bone marrow failure) and/or pathobiologic factors (increased sensitivity to cytotoxic treatments) associated with the germline syndrome may further complicate management and worsen prognosis. (See "Down syndrome: Routine health care, management of comorbidities, and prognosis" and "Management and prognosis of Fanconi anemia".)

Prognostic categories — We stratify the prognosis of t-MNs using the same guidelines that are used for de novo MDS and AML.

Note that the International Consensus Classification (ICC) and the World Health Organization 5th edition (WHO5) apply different labels and diagnostic criteria to MNs, as discussed above. (See 'Spectrum of disorders' above.)

A t-MN with 10 to 19 percent blasts and no defining genetic abnormalities is classified as MDS/AML by the ICC [7]. Such cases are classified as either MDS or AML by WHO5, according to defining genetic abnormalities, the degree of differentiation, and the percentage of blasts [8]. (See "Acute myeloid leukemia: Classification", section on 'International Consensus Classification'.)

t-MDS prognosis — We use the International Prognostic Scoring System (IPSS)-Revised (IPSS-R) or a validated mutation-based model, such as the IPSS-Molecular (IPSS-M), for risk stratification in patients with t-MDS. Details of these and other prognostic models for MDS are discussed separately. (See "Prognosis of myelodysplastic syndromes/neoplasms (MDS) in adults".)

Using either of these models, we classify a patient as having either lower-risk MDS or higher-risk MDS.

●Lower-risk MDS

•If using IPSS-R (table 6) (calculator 1), the following categories correspond to lower-risk MDS:

-Very low risk (≤1.5 points)

-Low risk (>1.5 to 3.0 points)

-Intermediate risk (3.5 points to 4.5, without TP53 mutation)

•If using IPSS-M, the following categories correspond to lower-risk MDS:

-Very low risk

-Low risk

-Moderate low risk

-Moderate high risk

●Higher-risk MDS

•If using IPSS-R (table 6) (calculator 1), the following categories correspond to higher-risk MDS:

-Intermediate risk (3.5 to 4.5 points with TP53 mutation)

-High risk (5 to 6 points)

-Higher risk (>6 points)

•If using IPSS-M, the following categories correspond to higher-risk MDS:

-Moderate high risk

-High risk

-Very high risk

We consider the original IPSS model to be obsolete because it considers only limited clinical and cytogenetic findings, it does not incorporate molecular findings, and it lacks the prognostic accuracy of IPSS-R and molecular models. Nevertheless, patients classified as low risk or intermediate-1 according to the original IPSS can be considered to have lower-risk MDS, while patients classified as intermediate-2 or high risk can be considered to have higher-risk MDS.

t-AML prognosis — We classify prognosis in patients with t-AML according to European LeukemiaNet criteria for AML (table 5) [1]:

•Favorable

•Intermediate

•Unfavorable

Further discussion of AML risk categories is presented separately. (See "Acute myeloid leukemia: Risk factors and prognosis", section on 'European LeukemiaNet (ELN) 2022 classification'.)

MANAGEMENT OVERVIEW — 

The management of t-MNs is generally like that for the corresponding de novo acute myeloid leukemia (AML), myelodysplastic syndromes/neoplasms (MDS), and MDS/myeloproliferative neoplasms (MDS/MPN).

However, treatment differs from that of the corresponding myeloid malignancy when management is informed by studies that included substantial numbers of patients with t-MNs.

●Disease category – The category of t-MNs is a key aspect for selecting treatment.

As described above, there are some differences in disease classification according to the chosen classification system. (See 'Spectrum of disorders' above.)

Patients diagnosed with therapy-related acute promyelocytic leukemia (t-APL) require urgent and distinctive treatment. Features that suggest a diagnosis of t-APL are discussed above. (See 'Emergencies' above.)

●Fitness – Management is stratified by fitness for intensive treatments, including intensive remission induction therapy and eligibility for allogeneic hematopoietic cell transplantation (HCT).

Age, per se, does not determine treatment choice, but fitness and prognosis may decline with advancing age.

●Prognostic category – As discussed above, a prognostic category is assigned to each patient, according to the disease category (eg, t-MDS, t-AML). (See 'Prognostic categories' above.)

t-MDS — 

Management of medically fit patients with t-MDS is guided by the prognostic category, as discussed above. (See 't-MDS prognosis' above.)

t-MDS with 10 to 19 percent blasts is generally treated like t-AML, as discussed below. (See 't-AML' below.)

Management of t-MDS in patients who cannot tolerate intensive treatment is discussed below. (See 't-AML' below.)

Lower-risk t-MDS — Management of lower-risk MDS is informed by the severity of cytopenias and related symptoms.

●Asymptomatic – We provide supportive care, if needed, while closely monitoring for the development of symptoms or disease progression. Many asymptomatic patients will soon develop symptoms or other reasons for treatment.

Asymptomatic patients should receive supportive care, including antibiotics for infections. Patients can be given red blood cell (RBC) transfusions for dyspnea, fatigue, or weakness, and platelet transfusions for bleeding or <20,000 platelets/microL. The need for transfusions generally indicates that the patient is developing findings that warrant treatment.

Supportive care for patients with lower-risk MDS is discussed separately. (See "Treatment of lower-risk myelodysplastic syndromes/neoplasms (MDS)", section on 'Supportive care'.)

●Symptomatic – We treat most symptomatic patients with lower-intensity therapy informed by the nature of the symptoms. Treatment is individualized, and the strategy should be chosen jointly by an informed patient and the treating clinician.

Some fit patients with lower-risk t-MDS may select intensive treatments like those used for higher-risk t-MDS. (See 'Higher-risk t-MDS' below.)

Some patients will choose to receive supportive care alone, to avoid treatment-related toxicity. (See "Treatment of lower-risk myelodysplastic syndromes/neoplasms (MDS)", section on 'Supportive care'.)

The selection of a lower-risk treatment should consider the following factors (algorithm 1):

•Cytopenia(s) – The hematologic lineage(s) causing symptoms (eg, anemia, thrombocytopenia, neutropenia, or more than one lineage)

We consider symptomatic cytopenias to be any of the following:

-Anemia – Symptoms related to anemia (eg, dyspnea, fatigue, weakness)

-Thrombocytopenia – Platelets <20,000/microL or excessive bleeding or bruising with platelets <50,000/microL

-Neutropenia – Recurrent and/or severe infections with absolute neutrophil count (ANC) <500 neutrophils/microL or ANC <1000 neutrophils/microL with recurrent infections

•Clinicopathologic features – Examples include MDS with del(5q) or MDS with ring sideroblasts/mutated SF3B1

•Patient preference – Considerations include toxicity, comorbidities, and convenience

Examples of lower-intensity treatments include erythropoiesis-stimulating agents, thrombopoietin receptor agonists, luspatercept, lenalidomide, hypomethylating agents (HMAs; ie, azacitidine, decitabine), and others. The selection of a lower-intensity regimen is presented separately. (See "Treatment of lower-risk myelodysplastic syndromes/neoplasms (MDS)".)

For patients who are treated with an HMA, the addition of venetoclax is acceptable, as discussed below. (See 't-MDS' below.)

Higher-risk t-MDS — We manage most fit patients who have higher-risk t-MDS like t-AML. (See 't-AML' below.)

Treatment in this setting is individualized, and the treatment strategy should be chosen jointly by an informed patient and the treating clinician.

Some fit patients will choose lower-intensity therapy to avoid the greater toxicity of intensive treatments. (See 'Lower-risk t-MDS' above.)

Some patients will choose to receive supportive care alone. (See "Treatment of lower-risk myelodysplastic syndromes/neoplasms (MDS)", section on 'Supportive care'.)

t-AML — 

Treatment of fit patients with t-AML is guided by the European LeukemiaNet (ELN) prognostic category (table 5). (See 'Prognostic categories' above.)

t-APL is a medical emergency that requires prompt diagnosis and urgent treatment, as discussed above. (See 'Emergencies' above.)

Intensive treatment for other subtypes of t-AML involves the following:

●Remission induction therapy, as guided by the ELN prognostic category, TP53 status, and patient comorbidities. (See 'Remission induction' below.)

●Treatment response is assessed by clinical evaluation and bone marrow examination. (See 'Response assessment' below.)

●For patients with complete remission (CR), post-remission management is guided by the ELN prognostic category at diagnosis. (See 'Post-remission management' below.)

Patients who do not achieve CR after two rounds of induction therapy are managed as refractory t-AML. (See 'Relapsed/refractory disease' below.)

Remission induction — Induction therapy for t-AML is informed by the ELN prognostic category. (See 'Prognostic categories' above.)

Favorable or intermediate risk — For medically fit patients with favorable- or intermediate-risk t-AML, we suggest CPX-351, rather than other intensive remission induction regimens or lower-intensity approaches, based on the balance of efficacy and toxicity. This approach differs from the management of de novo AML, as discussed elsewhere. (See "Acute myeloid leukemia: Induction therapy in medically fit adults".)

The choice of therapy is individualized and must consider patient preference. CPX-351 offers the best likelihood of achieving remission and proceeding to allogeneic HCT, with the opportunity for long-term survival. However, this must be weighed against the substantial toxicity and the possible treatment-related early death.

Some fit patients may instead choose treatment with a lower-intensity regimen (see 'Less-fit patients' below) or with supportive care alone, as discussed separately. (See "Treatment of lower-risk myelodysplastic syndromes/neoplasms (MDS)", section on 'Supportive care'.)

CPX-351 is a liposomal formulation of cytarabine and doxorubicin in a fixed 5:1 molar ratio that is associated with better outcomes than other intensive regimens in patients with t-AML. No randomized trials have directly compared CPX-351 with other intensive treatments or with lower-intensity therapy for a t-MN.

Some patients cannot be treated with CPX-351 because of prior anthracycline therapy or cardiac comorbidities. We assess suitability for anthracycline therapy by cardiac clinical evaluation and imaging, rather than the cumulative dose of anthracyclines; cardiac evaluation is discussed above. (See 'Evaluation' above.)

Anthracycline-free regimens that have been used for treating t-AML include FLAG (fludarabine, cytarabine, granulocyte colony-stimulating factor [G-CSF]), clofarabine-cytarabine (CLARA), or topotecan-cytarabine (TA); however, there is limited experience with using them to treat t-MNs [29,30]. Alternative induction regimens for AML are discussed separately. (See "Acute myeloid leukemia: Induction therapy in medically fit adults", section on 'Alternative induction regimens'.)

Although there is limited experience in t-MNs, a targeted agent can be added to induction therapy for t-AML with mutated FLT3, IDH1, or IDH2. Targeted agents in induction therapy for AML are discussed separately. (See "Acute myeloid leukemia: Induction therapy in medically fit adults", section on 'AML with mutated FLT3'.)

●Administration – CPX-351 (daunorubicin 44 mg/m² plus cytarabine 100 mg/m²) is infused over 90 minutes on days 1, 3, and 5. If needed, the same dose is administered on days 1 and 3 of a second cycle of induction therapy.

The US Food and Drug Administration (FDA) approved CPX-351 for t-MNs in adults and children ≥1 year. The European Medicines Agency approved CPX-351 for adults with t-AML.

●Toxicity – The most common grade ≥3 adverse effect (AE) is hemorrhage (12 percent); other AEs include cardiotoxicity, tissue necrosis, and hypersensitivity reactions. Although data are limited, we do not consider CPX-351 less cardiotoxic than other daunorubicin formulations.

The packaging label includes a boxed warning to not substitute liposomal daunorubicin-cytarabine with other daunorubicin- or cytarabine-containing products.

●Outcomes – Most studies of intensive treatments for t-MN also included patients with secondary AML (AML arising from prior hematologic disorders) or AML with other adverse features.

•CPX-351 achieved better long-term survival than "7+3" therapy (infusional cytarabine plus daunorubicin) in a phase 3 trial of 309 patients (60 to 75 years) with various myeloid malignancies [31,32]. For the entire trial population, CPX-351 achieved better median OS (9 versus 6 months; hazard ratio [HR] 0.70 [95% CI 0.55-0.91]) and superior five-year OS (18 versus 8 percent). More patients who were treated with CPX-351 proceeded to allogeneic HCT (35 versus 25 percent). Post-hoc analysis of the 63 patients with a t-MN reported that CPX-351 was associated with better median OS (12 versus 6 months; HR 0.54 [95% CI 0.31-0.94]). Febrile neutropenia, pneumonia, and hypoxia were the most frequent grade ≥3 AEs, but the incidence was comparable in both trial arms; mortality at 60 days was 14 and 21 percent, respectively with CPX-351 and 7+3 therapy.

•Other studies that included patients with a t-MN reported that CPX-351 was associated with 50 to 70 percent CR plus CRi (CR with incomplete hematologic recovery) [33-35].

Comparisons of outcomes with t-MNs and de novo AML have yielded mixed results. While patients with t-MNs generally have inferior outcomes, not all studies matched patients by prognostic features. Some studies reported comparable outcomes for patients in the same risk category [6,20,36-38], while others reported inferior outcomes with t-MNs [16,39-41]. Patients with favorable-risk t-MNs appear to have similar outcomes to favorable-risk de novo AML, as discussed above. (See 'Prognostic factors' above.)

Adverse-risk — Management of adverse-risk t-AML is guided by the TP53 status.

t-AML with mutated TP53 — We consider CPX-351, other intensive remission induction regimens, or lower-intensity treatments acceptable for TP53-mutated t-AML. We encourage participation in a clinical trial when possible.

Treatment of TP53-mutated t-AML is individualized and made jointly according to comorbidities and patient preference. No studies have directly compared various treatments in this setting, but outcomes are generally poor. Many patients with AML overestimate their long-term prognosis and chance for cure, and they do not recall being offered more than one treatment [11,12].

●Some experts favor CPX-351 or other intensive regimens, despite the substantial toxicity and limited likelihood of achieving a robust sustained response. Treatment with CPX-351 and other intensive regimens is described above. (See 'Favorable or intermediate risk' above.)

As an example, a vigorous or younger patient (eg, <40 years) may favor intensive remission induction therapy, followed by allogeneic HCT. (See 'Post-remission management' below.)

●Some patients opt for lower-intensity approaches or supportive care alone because of comorbid conditions, age, or personal values and preferences.

Lower-intensity treatments are discussed below. (See 'Less-fit patients' below.)

Supportive care for patients with AML is discussed separately. (See "Acute myeloid leukemia: Management of medically unfit adults", section on 'Supportive care'.)

Some experts extrapolate the management of TP53-mutated t-AML from treatment with decitabine for AML or MDS with mutated TP53; however, that study did not include patients with t-MNs [42]. In that study, outcomes of treatment with decitabine in 116 patients with TP53-mutated AML or MDS were comparable to outcomes in patients who had intermediate- or favorable-risk prognosis disease [42].

Other adverse features — For t-AML with adverse-risk features other than mutated TP53, we treat like favorable- or intermediate-risk t-AML. (See 'Favorable or intermediate risk' above.)

Response assessment — The goal of intensive therapy is long-term disease control, which typically requires achieving CR.

Response assessment includes:

●Bone marrow examination to assess blast clearance is typically performed on day 14.

●Bone marrow examination is again repeated after recovery of blood counts (eg, four to six weeks after beginning treatment) and should include bone marrow morphology and an assessment of measurable residual disease (MRD). MRD status is associated with clinical outcomes in patients with AML or a t-MN [43], and the role of MRD in selecting post-remission therapy is evolving.

For patients with prolonged pancytopenia and no signs of hematologic recovery, bone marrow examination should be repeated to exclude regrowth of the t-MN.

Management is described below for fit patients who achieve CR. (See 'Post-remission management' below.)

Patients who do not achieve CR after two cycles of induction therapy are managed as a refractory t-MN. (See 'Relapsed/refractory disease' below.)

Details of response assessment, remission criteria, and assessment of MRD are described separately. (See "Acute myeloid leukemia: Induction therapy in medically fit adults", section on 'Remission status'.)

Post-remission management — Post-remission management is informed by the ELN prognostic category, MRD status, fitness, comorbidities, and patient preference. ELN risk groups are discussed above. (See 'Prognostic categories' above.)

Treatment decisions are informed by studies of patients with de novo AML, as there are only limited observational, retrospective, and population-based studies of t-AML [43].

Favorable risk and MRD negative — For MRD-negative favorable-risk t-AML, we suggest consolidation chemotherapy, rather than allogeneic HCT.

Outcomes are similar for patients with a favorable-risk t-MN compared with favorable-risk de novo AML [2,19]. Consolidation therapy for MRD-negative favorable-risk de novo AML is associated with good outcomes and this approach avoids transplant-associated AEs.

●For patients initially treated with CPX-351, we give CPX-351 consolidation therapy (daunorubicin 29 mg/m² and cytarabine 65 mg/m² intravenously over 90 minutes on days 1 and 3).

●For patients treated with other intensive regimens, we use high-dose or intermediate-dose cytarabine-based or comparable myeloid consolidation regimens. (See "Acute myeloid leukemia in younger adults: Post-remission therapy", section on 'Favorable-risk disease'.)

Consolidation therapy using CPX-351 has not been directly compared with other consolidation regimens in this setting.

Others — For transplant-eligible patients with MRD-positive favorable-risk t-AML, and for intermediate-risk or adverse-risk t-AML, we suggest allogeneic HCT.

Patients who are not transplant-eligible and decline transplantation are managed as discussed above. (See 'Favorable risk and MRD negative' above.)

No randomized trials have directly compared allogeneic HCT versus consolidation chemotherapy in the setting of t-AML. However, observational studies reported that, aside from patients with favorable-risk t-AML, outcomes are worse than the corresponding de novo AML [2,19].

Post-remission management of intermediate-risk and adverse-risk AML is discussed separately. (See "Acute myeloid leukemia in younger adults: Post-remission therapy".)

Retrospective and population-based studies of t-AML suggest that allogeneic HCT is beneficial for selected patients, but these studies lack appropriate control groups and are subject to biases in patient selection. Informative studies include:

●An international retrospective analysis reported outcomes of 868 persons who underwent allogeneic HCT for t-MNs [44]. After one and five years, OS was 37 and 22 percent, respectively; disease-free survival (DFS) was 32 and 21 percent; relapse occurred in 27 and 31 percent; and treatment-related mortality (TRM) was 41 and 48 percent. Grade II to IV acute graft-versus-host disease (GVHD) was reported in 39 percent, and one-third of patients had chronic GVHD at one, three, and five years. There was no difference in outcomes after reduced-intensity conditioning versus myeloablative conditioning. Risk factors associated with inferior OS and DFS included age >35 years, adverse-risk cytogenetics, transplantation without remission, and a graft other than a human leukocyte antigen (HLA)-matched sibling donor (MSD) or a partially/well-matched unrelated donor. At five years, OS for patients with zero, one, two, three, or four of these risk factors was 50, 26, 21, 10, and 4 percent respectively.

●An international registry study of 461 patients with t-MNs reported 35 percent three-year OS [45]. The cumulative incidence of relapse at three years was 31 percent, and nonrelapse mortality at one and three years was 32 and 37 percent, respectively.

●Other registry studies report superior outcomes with transplantation, but these analyses were not well-controlled for prognostic features or medical fitness. Examples include an Italian Network registry study that reported a 58.8-month median OS for transplant recipients, compared with 12.1 months for nontransplant approaches [13], and a German study that reported a 7.2-month median OS with nontransplant approaches, while median OS for transplantation was not reached after median follow-up of 41 months [46].

Selection of a conditioning regimen and the choice of a graft donor and graft source are discussed separately. (See "Acute myeloid leukemia in younger adults: Post-remission therapy", section on 'Unfavorable-risk disease'.)

t-MDS/MPN — 

t-MDS/MPN can be managed like MDS or AML. The treatment strategy should be made jointly, with consideration of clinical severity and patient preference.

LESS-FIT PATIENTS — 

Management of patients who cannot tolerate intensive remission induction therapy is informed by whether the patient has therapy-related acute myeloid leukemia (t-AML) versus therapy-related myelodysplastic syndromes/neoplasms (t-MDS).

t-AML — For patients with t-AML who cannot tolerate an intensive induction regimen, we suggest a hypomethylating agent (HMA; ie, azacitidine or decitabine) plus venetoclax (BCL2 inhibitor), based on extrapolation from studies of de novo AML.

The choice of treatment is individualized and informed by patient preference. No prospective studies have directly compared azacitidine versus decitabine or compared an HMA with or without venetoclax for patients with t-AML.

Some less-fit patients may choose alternative treatment approaches, as discussed below. (See 'Frail patients' below.)

●Administration – Venetoclax plus an HMA should be used with caution in patients who have liver disease or kidney dysfunction, and in patients who are receiving CYP3A inhibitors. Venetoclax and HMAs are associated with embryo-fetal toxicity.

Doses may need to be delayed or modified for persistent cytopenias. Details of treatment with an HMA plus venetoclax are presented separately. (See "Acute myeloid leukemia: Management of medically unfit adults", section on 'HMA plus venetoclax'.)

•Choice of HMA – Azacitidine and decitabine have comparable efficacy and toxicity for treating de novo AML or MDS, but outcomes are less well-defined for t-MNs.

The choice of azacitidine versus decitabine is guided by the route and schedule of administration and preferences of the patient and clinician. Some patients favor azacitidine because it is given subcutaneously and requires less time in an infusion center. Others favor decitabine because it is given intravenously over fewer days, and it can be given via an indwelling intravenous access device, if present. Decitabine is also available as an oral therapy in combination with cedazuridine, but its role for t-MNs is not well-defined.

The selection of an HMA, administration, and toxicity are discussed separately. (See "Treatment of lower-risk myelodysplastic syndromes/neoplasms (MDS)", section on 'Hypomethylating agents'.)

Azacitidine is approved by the US Food and Drug Administration (FDA) for treatment of MDS, and the European Medicines Agency (EMA) approved azacitidine for the treatment of higher-risk MDS. Oral azacitidine (CC-486) is only approved as a maintenance therapy for AML in remission.

Decitabine is approved by the FDA for adults with MDS and by the EMA for AML.

•Venetoclax – The dose of venetoclax is escalated in the first treatment cycle (but not in subsequent cycles) to reduce the risk of tumor lysis syndrome (TLS).

Venetoclax is given for 28 days per cycle. However, the duration of venetoclax administration can be shortened (eg, to 21 days or 14 days per cycle) for patients who achieve remission.

Venetoclax is approved by the FDA in combination with an HMA in patients who are not suitable for intensive induction therapy or for patients ≥75 years with newly diagnosed AML; venetoclax is not approved by the FDA as a single agent for the treatment of AML.

Venetoclax in combination with an HMA is approved for adults with newly diagnosed AML who are ineligible for intensive chemotherapy by the EMA.

●Toxicity – The most common grade ≥3 AEs are prolonged cytopenias, febrile neutropenia, and TLS. Other AEs include mild or moderate gastrointestinal effects (eg, nausea, vomiting, diarrhea), fatigue, and edema.

●Response assessment – Bone marrow examination is performed to document the response to treatment.

We generally perform a bone marrow examination between days 14 and 21 after beginning therapy. Patients with remission (ie, <5 percent blasts) may be candidates for shortening the duration of venetoclax in subsequent cycles, as discussed above.

Another bone marrow examination can be performed after three cycles and then repeated every three cycles for patients in remission.

●Duration of therapy – Treatment can continue indefinitely in the absence of intolerable adverse effects (AEs). However, doses may need to be delayed or modified for persistent cytopenias, especially in patients who achieve remission.

No studies have compared open-ended treatment versus therapy for a predefined period in patients with t-MNs.

●Outcomes – There are no studies that compared outcomes with an HMA, with or without venetoclax, in patients with t-AML.

The addition of venetoclax to an HMA is based on improved outcomes in de novo AML. A randomized trial (VIALE-A) in patients with de novo AML reported that, compared with an HMA alone, the addition of venetoclax achieved superior survival with acceptable toxicity [3]. In a study of an HMA plus venetoclax for de novo AML that included patients with adverse cytogenetic/molecular profiles, more than one-half of patients had a clinically meaningful response with acceptable toxicity [47].

●Other targeted agents – We do not add inhibitors of IDH1 (eg, ivosidenib, olutasidenib), IDH2 (enasidenib), or FLT3 (eg, midostaurin, quizartinib) to HMA-based or other induction regimens for patients with t-MNs.

There is no evidence in t-MNs that adding a targeted agent to an HMA improves outcomes.

The use of targeted agents in patients who are not suitable for other treatments is discussed below. (See 'Frail patients' below.)

t-MDS — For patients with t-MDS who cannot tolerate intensive therapy, we suggest an HMA, with or without venetoclax.

●Administration – The choice of an HMA and whether to add venetoclax is individualized and informed by patient preference. HMA-based treatment can relieve symptoms, improve quality of life, and/or prolong survival with modest toxicity, but no studies have directly compared HMAs with or without venetoclax for t-MNs.

HMAs should be used with caution in patients with pre-existing liver impairment, and they are associated with embryo-fetal risk.

●Toxicity – The most common AEs with HMAs are cytopenias and febrile neutropenia. Other AEs include mild/moderate diarrhea or constipation, fatigue, and edema.

●Response assessment – Response assessment for HMA treatment generally focuses on achieving a functional response (eg, reducing transfusion needs, avoiding hospitalization, controlling symptoms), rather than the documentation of remission with repeated bone marrow examinations.

Responses to single-agent azacitidine or decitabine are generally slow. If a bone marrow examination is performed to assess response, it should be done after four to six treatment cycles.

Response assessment for patients treated with an HMA plus venetoclax is described above. (See 't-AML' above.)

●Outcomes – Observational studies have reported that outcomes with HMAs for t-MDS were comparable to responses of de novo MDS [48-50]. Outcomes with HMAs for MDS are discussed separately. (See "Treatment of lower-risk myelodysplastic syndromes/neoplasms (MDS)", section on 'Hypomethylating agents'.)

FRAIL PATIENTS — 

Treatment of frail patients and less-fit patients who decline hypomethylating agent (HMA)-based therapy should focus on relieving symptoms and improving the quality of life. 

The selection of an approach should reflect the individual's personal values and goals of care:

●Some patients may favor HMA-based treatment, as discussed above. (See 'Less-fit patients' above.)

●Frail patients with a susceptible mutation may choose treatment with a targeted agent, although their efficacy in t-MNs has not been demonstrated.

Drugs that target mutated IDH1 (eg, ivosidenib, olutasidenib) and mutated IDH2 (eg, enasidenib) can be given as single agents for the treatment of de novo acute myeloid leukemia (AML) and are discussed separately. (See "Acute myeloid leukemia: Management of medically unfit adults", section on 'IDH inhibitors'.)

●Patients with symptoms related to disease proliferation (eg, hyperleukocytosis, splenomegaly) may be candidates for hydroxyurea, an oral agent with little toxicity.

●Other frail patients may select supportive care alone (eg, transfusions, antibiotics) for symptom relief. Supportive care in this setting is like that for patients with myelodysplastic syndromes/neoplasms (MDS), as discussed separately. (See "Treatment of lower-risk myelodysplastic syndromes/neoplasms (MDS)", section on 'Supportive care'.)

MONITORING — 

There is no consensus regarding the frequency or protocol for follow-up visits after treatment for t-MNs. Monitoring should include both surveillance for disease relapse and long-term complications of therapy.

The frequency of visits should be informed by clinical judgment. We generally schedule patient visits every one to two months for the first two years (when relapse is most likely), and then every three to six months for up to five years post-consolidation.

We evaluate the patient clinically and obtain a complete blood count/differential count and review of the blood smear. Bone marrow aspirate/biopsy is not routinely performed but may be informative if the peripheral smear demonstrates abnormalities or if unexplained cytopenias develop.

RELAPSED/REFRACTORY DISEASE — 

The prognosis for patients who relapse or have a refractory t-MN is generally grim. We encourage participation in a clinical trial when possible.

The approach is guided by medical fitness, prior therapies, and the presence of susceptible mutations in the malignant cells. Examples of potential treatments for patients with relapsed or refractory t-MNs include:

●Patients with mutated IDH1 or IDH2 may be candidates for a targeted agent if not previously given.

●Patients with favorable or intermediate risk who are eligible for allogeneic hematopoietic cell transplantation (HCT) and were not previously transplanted may be considered for transplantation.

●Patients who relapse after allogeneic HCT may be candidates for donor lymphocyte infusion.

These and other potential treatment approaches are discussed separately. (See "Treatment of relapsed or refractory acute myeloid leukemia".)

CLINICAL TRIALS — 

Often there is no better therapy to offer a patient than enrollment onto a well-designed, scientifically valid, peer-reviewed clinical trial. Additional information and instructions for referring a patient to an appropriate research center can be obtained from the United States National Institutes of Health [51].

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Acute myeloid leukemia" and "Society guideline links: Myelodysplastic syndromes".)

INFORMATION FOR PATIENTS — 

UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient education" and the keyword(s) of interest.)

●Beyond the Basics topics (see "Patient education: Acute myeloid leukemia (AML) treatment in adults (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Description – Therapy-related myeloid neoplasms (t-MNs) arise after prior treatment with DNA-damaging agents, including chemotherapy and radiation therapy.

●Diagnosis/classification – The spectrum of t-MNs includes acute myeloid leukemia (AML), myelodysplastic syndromes/neoplasms (MDS), and MDS/myeloproliferative neoplasms (MDS/MPN). (See 'Spectrum of disorders' above.)

●Emergencies – Therapy-related acute promyelocytic leukemia (t-APL) is an emergency that requires urgent diagnosis and treatment like that for de novo APL. (See 'Emergencies' above.)

●Evaluation – Assess comorbidities, prior treatments, and effects of previous illness. (See 'Pretreatment' above.)

●Prognostic category – Determined by disease category:

•MDS – Use International Prognostic Scoring System (IPSS)-Revised (table 6) (calculator 1) or IPSS-Molecular to assign a prognostic category (see 't-MDS prognosis' above):

-Lower risk

-Higher risk

•AML – Use European LeukemiaNet (ELN) criteria for assigning an AML prognostic category (table 5) (see 't-AML prognosis' above):

-Favorable

-Intermediate

-Unfavorable

●Management – Guided by fitness, disease type (ie, t-MDS, t-AML, t-MDS/MPN), and prognostic category, chosen jointly by the clinician and informed patient. For some t-MNs, management may be more aggressive than the corresponding de novo MN, given the inferior prognosis. (See 'Management overview' above.)

●t-MDS – Management of fit patients is guided by prognostic category. (See 't-MDS' above.)

•Lower risk – Informed by type/severity of cytopenia-related symptoms, like de novo lower-risk MDS (algorithm 1).

Examples include erythropoiesis-stimulating agents, thrombopoietin receptor agonists (not given with blasts ≥5 percent), luspatercept, lenalidomide, and hypomethylating agents (HMAs; ie, azacitidine or decitabine, with or without venetoclax), as described above. (See 'Lower-risk t-MDS' above.)

•Higher risk – We treat like t-AML. (See 'Higher-risk t-MDS' above.)

●t-AML – Remission induction and post-remission management are guided by ELN prognostic category (table 5) (see 't-AML' above):

•Induction therapy (see 'Remission induction' above)

-Favorable/intermediate risk – We suggest CPX-351 (liposomal cytarabine/doxorubicin) rather than other intensive or lower-intensity approaches (Grade 2B). This differs from the approach in de novo AML. (See 'Favorable or intermediate risk' above.)

-Adverse risk – Clinical trials are encouraged. For TP53-mutated t-AML, acceptable treatments include CPX-351, other intensive regimens, or lower-intensity therapy. (See 't-AML with mutated TP53' above.)

t-MNs with other adverse features are treated like favorable-/intermediate-risk t-AML. (See 'Other adverse features' above.)

•Response assessment – Determined by clinical response, bone marrow morphology, and measurable residual disease (MRD). (See 'Response assessment' above.)

•Post-remission management – For patients who achieve complete remission (CR):

-MRD negative and favorable risk – We suggest consolidation chemotherapy, rather than allogeneic hematopoietic cell transplantation (HCT) (Grade 2C). (See 'Favorable risk and MRD negative' above.)

-Others – We suggest allogeneic HCT for MRD-positive, favorable-risk t-AML and intermediate-risk or adverse-risk t-AML (Grade 2C). (See 'Others' above.)

Transplantation is used more often with t-MNs than corresponding de novo diseases because of the inferior prognosis.

●t-MDS/MPN – Individualized by clinical severity and patient preference, using treatments like either t-MDS or t-AML. (See 't-MDS/MPN' above.)

●Less-fit patients – Treatment of less-fit patients is stratified by t-MN subtype:

•t-MDS – Treat like de novo MDS using an HMA; venetoclax can be added due to inferior prognosis (whereas venetoclax is not typically added for de novo MDS). (See 't-MDS' above.)

•t-AML – Treat with an HMA plus venetoclax, like de novo AML. (See 't-AML' above.)

●Frail – Individualized treatment focused on symptom relief, such as HMA, targeted agent (for mutated IDH1 or IDH2), and/or supportive care. (See 'Frail patients' above.)

●Relapsed/refractory t-MN – Prognosis is generally grim. Management is individualized, but options include a targeted agent for susceptible mutations, allogeneic HCT (if not previously performed), or donor lymphocyte infusion for relapse after allogeneic HCT. (See 'Relapsed/refractory disease' above.)