Chronic myelomonocytic leukemia: Management and prognosis

INTRODUCTION — Chronic myelomonocytic leukemia (CMML) is a hematologic malignancy with clinical and pathologic features of both myeloproliferative neoplasms (MPN) and myelodysplastic syndromes/neoplasms (MDS). CMML is characterized by peripheral blood monocytosis and bone marrow dysplasia, often accompanied by cytopenias, constitutional symptoms, and/or splenomegaly. CMML is among the most aggressive chronic leukemias, with a propensity for progression to acute myeloid leukemia (AML).

Management and prognosis of CMML are discussed in this topic.

Pathogenesis, epidemiology, clinical presentation, diagnosis, and differential diagnosis of CMML are discussed separately. (See "Chronic myelomonocytic leukemia: Clinical features, evaluation, and diagnosis".)

PRETREATMENT EVALUATION — Pretreatment evaluation of the patient with chronic myelomonocytic leukemia (CMML) identifies disease-associated symptoms, assesses medical fitness, and evaluates comorbid illnesses to aid selection of a management strategy.

Clinical features, initial evaluation and diagnosis, and differential diagnosis of CMML are discussed separately. (See "Chronic myelomonocytic leukemia: Clinical features, evaluation, and diagnosis", section on 'Evaluation'.)

Clinical and laboratory — Evaluation seeks evidence of symptoms and other findings related to CMML and comorbid conditions.

●Clinical - The history and physical examination should evaluate the presence of disease-associated findings, including constitutional symptoms (ie, unexplained fevers, weight loss, night sweats), manifestations of splenomegaly (eg, early satiety, abdominal fullness or pain), lymphadenopathy, disease-related skin lesions, complaints associated with cytopenias (eg, fatigue, infections, bleeding/bruising), and document the history of transfusions.

●Laboratory - Studies include:

•Hematology - Complete blood count (CBC) with differential count and reticulocyte count

•Serum chemistries – Basic metabolic panel, kidney and liver tests, serum erythropoietin (EPO)

•Other – For selected patients:

-Monocyte subsets (repartitioning) - When there is a high index of suspicion for CMML, but the diagnosis is uncertain (eg, absolute monocyte count <1x10⁹/L with ≥10 percent monocytes in the differential count), flow cytometry should be performed to characterize the relative percentages of monocyte subtypes.

-HLA typing - Human leukocyte antigen (HLA) typing should be performed for patients who are candidates for allogeneic HCT and/or those who require HLA-matched platelets. (See 'Allogeneic transplantation' below.)

-Antibody screen - Lymphocytotoxic (anti-HLA) antibody screen is useful if there are concerns regarding alloimmunization, especially for multiparous women and multiply-transfused patients.

Diagnosis and classification — CMML is diagnosed and classified according to either the International Consensus Classification (ICC) [1] and the World Health Organization 5^th edition (WHO5) [2].

Diagnosis of CMML is based on peripheral blood monocytosis (≥0.5 x 10⁹/L; monocytes ≥10%) for at least three months, usually with accompanying bone marrow dysplasia. Clonal cytogenetic abnormalities occur in one-third of patients, while >90% have somatic gene mutations. Details of the diagnosis and differential diagnosis of CMML are presented separately. (See "Chronic myelomonocytic leukemia: Clinical features, evaluation, and diagnosis".)

Both ICC and WHO5 define two CMML subtypes:

●Proliferative (CMML-MPN) – WBC count ≥13,000/microL (≥13 x 10⁹/L)

●Dysplastic (CMML-MDS) – WBC count <13,000/microL

Some patients with CMML have systemic mastocytosis with associated hematologic neoplasm (SM-AHN), with KIT D816V mutation in the neoplastic monocytes and mast cells. These patients generally have marked hepatosplenomegaly, mast cell activation symptoms, or cutaneous lesions with elevated serum tryptase levels. Management of CMML with SM-AHN is described separately. (See "Advanced systemic mastocytosis: Management and prognosis".)

Disease stage — CMML stage is determined by the percentage of blasts in peripheral blood and bone marrow.

●CMML-1 – <4 percent blasts plus promonocytes in peripheral blood and <9 percent blasts in bone marrow

●CMML-2 – 5 to 19 percent blasts plus promonocytes in peripheral blood and 10 to 19 percent blasts in bone marrow and/or the presence of Auer rods

Both the ICC and WHO5 eliminated CMML-0 (<2 percent blasts in blood and <5 percent blasts in marrow), because it had little or no prognostic impact [3,4].

Clonal cytopenia of undetermined significance (CCUS) is characterized by monocytosis ≥10 percent, WBC count ≥0.5 x 10⁹/L, clonal myeloid neoplasm-associated mutation(s), but without bone marrow features that fulfill criteria for CMML; CCUS is associated with potential to progress to MDS/MPN [5]. Clonal cytopenia and monocytosis of undetermined significance (CCMUS) can be applied if cytopenia is present [1]. CCUS and CCMUS are described by the ICC [1], but not in WHO5 [2].

MEDICAL FITNESS — Patients should be assessed for fitness to receive intensive treatments, including hematopoietic cell transplantation (HCT), before choosing a treatment strategy.

Referral to transplantation specialists to judge eligibility for allogeneic HCT and initiation of a search for a suitable graft donor should be offered to medically-fit patients with higher-risk CMML.

Medical fitness guides the goals of care and management choices for CMML. Age, per se, does not determine medical fitness, but caution should be used when considering intensive therapy for patients ≥70 years old because of age-related comorbidities. Chronic comorbid conditions should be weighted more heavily than transient medical complications of the leukemia itself (eg, infection, heart failure exacerbated by anemia). The burden of CMML can contribute to a lack of fitness and, in some cases, treatment may alleviate consequences/complications of the leukemia and enhance the patient's ability to tolerate and benefit from subsequent treatments.

Our assessment of medical fitness includes evaluation of:

●Performance status (PS) - Based on the Eastern Cooperative Oncology Group (ECOG) performance scale (table 1).

●Physiologic fitness - Comorbid conditions, activities of daily living, physical performance tests, cognition) can be assessed with the Charlson comorbidity index (CCI) (table 2).

Other instruments for assessing performance status (eg, Karnofsky scale (table 3)) or physiologic fitness (eg, the short physical performance battery [SPPB] https://geriatrictoolkit.missouri.edu/SPPB-Score-Tool.pdf [6] or the Hematopoietic cell transplantation-specific comorbidity index [HCT-CI] (table 4)) are also acceptable.

There are no clear distinctions between fitness categories and some measures of PS or physiologic fitness can apply to different categories. In selecting a category of fitness, we seek to protect frail patients from treatment that they are unlikely to survive, while not depriving a medically-unfit but not frail patient of the opportunity to achieve remission and prolonged survival. We categorize medical fitness as follows:

●Medically-fit (both of the following):

•ECOG PS: 0 to 2

•CCI: 0 to 2

●Medically-unfit, but not frail (either of the following):

•ECOG PS: 3

•CCI: 3

●Frail (both of the following):

•ECOG PS: ≥3

•CCI: ≥3

PROGNOSIS

Prognostic factors — The prognosis of patients with CMML is generally poor, but there is marked clinical heterogeneity in natural history and outcomes. Some clinical and laboratory features that are associated with outcomes are incorporated into prognostic models of CMML. (See 'Prognostic models' below.)

Patients with CMML are at risk for disease complications (eg, cytopenias), transformation to acute myeloid leukemia (AML), and treatment-related toxicity. Median survival for patients with CMML is generally 20 to 40 months and one-third may progress to AML, but outcomes vary with an individual's prognostic features [7-15]. Compared with patients who have CMML-1, those with CMML-2 have higher rates of transformation to AML (42 percent versus 27 percent) and shorter time to transformation (7.5 months versus 9.5 months) [16].

Features most closely associated with survival in patients with CMML are [7-13,17-19]:

●White blood cell (WBC) count

●Blast percentage in blood and bone marrow

●Hemoglobin (Hb) level

●Red blood cell (RBC) transfusion-dependence

●Karyotype

●Mutation status

Mutations are present in nearly all cases of CMML and some mutations are independently associated with outcomes [20,21]. The most common mutations include genes that encode splicing factors (eg, SRSF2), epigenetic regulators (eg, ASLX1, TET2, DNMT3A, IDH2, EZH2, UTX), tyrosine kinases (eg, NRAS, KRAS, CBL, JAK2), and transcription factors (eg, RUNX1) [22-27]. Nonsense and frameshift mutations of ASXL1 (but not missense mutations) are an adverse prognostic feature [28,29]. Other mutations are less closely linked to outcomes, while some mutations are associated with better prognosis. As an example, in one study, loss-of-function mutations of TET2 were present in more than half of 1084 patients with CMML and those patients had more favorable outcomes compared to patients with no TET2 mutation (median survival 49 versus 30 months, respectively) [30]. Mutations also differ according to the CMML subtype; mutations in JAK2, NRAS, and SETBP1 are more common with the proliferative subtype, whereas TET2 and SF3B1 mutations are more closely associated with the dysplastic subtype [20].

Prognostic models — The preferred prognostic model for CMML varies among experts, but we consider any of the following models, which incorporate cytogenetic and molecular features, to be acceptable for clinical use. They have comparable performance, but some are more likely to upstage patients (ie, reclassify a patient from a lower-risk group in one model to higher-risk disease with another model).

Our preferred models are described below, but other models have been created [19,31-33].

MMM (Mayo molecular model) — The Mayo molecular model includes the monocyte count, circulating immature myeloid cells, hemoglobin (Hb) level, platelet count, and ASXL1 mutation status [28].

Points are assigned as follows:

●Absolute monocyte count (AMC) – >10,000/microL (10 x 10⁹/L) (1 point)

AMC = white blood cell (WBC) count x percentage of monocytes ÷ 100

●Circulating immature cells – Presence of myeloblasts, myelocytes, metamyelocytes, and/or promyelocytes in peripheral blood (1 point)

●Decreased Hb – <10 g/dL (<100 g/L) (1 point)

●Decreased platelets – <100,000/microL (1 point)

●ASXL1 frameshift or nonsense mutation (1 point)

The Mayo molecular model stratifies patients into four risk groups (based on the sum of the points above) with the following rates of median overall survival (OS):

●Low risk (0 points): OS 97 months

●Intermediate-1 risk (1 point): OS 59 months

●Intermediate-2 risk (2 points): OS 31 months

●High risk (≥3 points): OS 16 months

CPSS-Mol — The CMML-specific prognostic scoring system with molecular features (CPSS-Mol) includes the CMML subtype and stage, transfusion-dependence (or Hb level), cytogenetic features, and mutation status [20]. CMML subtype and stage are described above. (See 'Pretreatment evaluation' above.)

The CPSS-Mol score is calculated by adding points from clinical features plus points assigned to the genetic risk group:

●Clinical features:

•CMML-proliferative subtype – WBC count ≥13,000/microL (1 point)

•Blasts – Bone marrow blasts ≥5 percent (1 point)

•Transfusion-dependence/hemoglobin (Hb) level - ≥1 unit red blood cell (RBC) transfusions per 8 weeks over a 4-month period) or Hb <9 g/dL for men and <8 g/dL for women (1 point)

plus

●Genetic risk group score (the sum of cytogenetic risk plus mutation risk), as follows:

•Cytogenetic risk [31]:

-Low cytogenetic risk (0 points) – Normal karyotype or isolated -Y

-Intermediate cytogenetic risk (1 point) – Other cytogenetic abnormalities, exclusive of high cytogenetic risk features (below)

-High cytogenetic risk (2 points) – Trisomy 8, complex karyotype (≥3 abnormalities), or abnormalities of chromosome 7

plus

•Mutation risk

-ASXL1, NRAS, or SETBP1 (1 point each)

-RUNX1 (2 points)

●Risk groups - CPSS-Mol stratifies patients into four risk groups (based on the sum of the points above), with the following rates of median OS and 48-month cumulative incidence of leukemia (CIL):

•Low (0 points): OS not reached; CIL 0 percent

•Intermediate-1 (1 point): OS 68 months; CIL 8 percent

•Intermediate-2 (2-3 points): OS 30 months; CIL 24 percent

•High (≥4 points): OS 17 months; CIL 52 percent

Compared with other prognostic models, CPSS-Mol was superior for predicting survival and it upstaged the risk category for a substantial proportion of patients, but it has not yet been independently validated [20].

GFM (Groupe Francophone des Myélodysplasies) — The GFM model stratifies risk according to age, WBC count, Hb, platelet count, and ASXL1 mutation [34]. Points are assigned as follows:

●Features:

•Age – >65 years (2 points)

•WBC count – >15,000/microL (>15 × 10⁹/L) (3 points)

•Hb – <10 g/dL for women or <11 g/dL for men (2 points)

•Platelet count – <100,000/microL (2 points)

•ASXL1 mutation – (2 points)

●Risk groups - The GFM model stratifies patients into three risk groups (based on the sum of the points above) with the following rates of median OS among 312 patients in a training set (32 month median follow-up) and 165 patients in a validation cohort (27 month median follow-up):

•Low risk (0-4 points): OS not reached

•Intermediate risk (5-7 points): OS 22 to 39 months

•High risk (8-12 points): OS 14 to 18 months

Risk categories — Management of CMML is informed by the risk category, but these categories are not absolute.

Risk categories are derived from one of the prognostic models described above (see 'Prognostic models' above):

●Higher-risk:

•CPSS-Mol – High risk and Intermediate-2 risk

•MMM – High risk and Intermediate-2 risk

•GFM – High risk and selected patients with Intermediate risk

●Lower-risk:

•CPSS-Mol – Low risk and Intermediate-1 risk

•MMM – Low risk and Intermediate-1 risk

•GFM – Low risk and selected patients with Intermediate risk

CMML risk categories are not absolute, especially for individuals with intermediate risk. Assignment of risk may also be influenced by disease subtype, stage, and/or mutation status.

RISK-STRATIFIED MANAGEMENT — Management is guided by the CMML risk category (ie, higher-risk versus lower-risk), the nature of symptoms, suitability for transplantation, and patient preference. We use a risk-adapted approach for transplantation decisions and symptom-directed management for other patients. We encourage participation in a clinical trial, whenever possible.

Assignment of a CMML risk category is discussed above. (See 'Risk categories' above.)

Within risk categories, treatment choices are informed by age and medical fitness, the nature and severity of disease-related symptoms, and individual values and preferences. Age, per se, does not determine the choice of treatment, but older patients generally have more comorbid illnesses many institutions limit allogeneic HCT to patients ≤70 years. Determination of medical fitness and eligibility for allogeneic hematopoietic cell transplantation (HCT) are discussed above. (See 'Medical fitness' above.)

Our approach to management of CMML is consistent with the European Hematology Association [35], Italian Society of Hematology (SIE) [36], other expert panels [37].

Higher-risk — Management of higher-risk CMML is informed by eligibility for allogeneic HCT (based on age and medical fitness), availability of a suitable graft donor, and the patient’s values and preferences.

Younger, fit patients — For younger (eg, ≤70 years) medically-fit patients with higher-risk CMML, we suggest allogeneic HCT rather than symptom-directed therapy. Allogeneic HCT is the only approach that has the potential to cure CMML, but it is associated with substantial toxicity; while an HMA or hydroxyurea can provide symptom relief, they generally do not offer long-term disease control.

No randomized trials have directly compared transplantation with symptom-directed therapy in this setting, but retrospective reviews reported that allogeneic HCT was associated with improved survival in patients with higher-risk CMML. The impact of allogeneic HCT in patients with lower-risk CMML is discussed below. (See 'Symptomatic lower-risk CMML' below.)

●A retrospective study of 261 patients ≤70 years reported that allogeneic HCT in patients with higher-risk CMML was associated with better survival compared with non-transplant management [38]. Compared with 142 no-transplant patients, the 119 transplanted patients had superior median overall survival (OS; 4.3 years versus 2.0 years). Importantly, transplantation was associated with a survival benefit only for patients with higher-risk CMML (hazard ratio [HR] 0.63 [95% CI 0.41-0.96]). There was less advantage for patients with significant comorbid conditions, but regression analysis indicated that the benefit of transplantation was not associated with patient age, per se. The survival advantage of transplantation became more pronounced with longer follow-up, with a survival plateau after six years.

●Analysis of 1114 patients (18 to 70 years) in two large CMML databases reported that allogeneic HCT was associated with a trend toward improved OS in patients with higher-risk disease [39]. In patients with higher-risk CMML, allogeneic HCT was associated with increased risk of death in the first two years after transplant (hazard ratio [HR] 1.46), but not beyond that point.

●A single-center study of 70 transplanted patients reported an OS benefit with allogeneic HCT, compared with matched nontransplanted patients after adjustment for age and CMML risk [40].

Although transplant outcomes vary among studies, approximately one-third of patients with CMML are cured, while similar proportions of patients relapse or experience transplant-related mortality (TRM). As an example, a multicenter study that included 513 patients with CMML (median age 53 years) reported 33 percent four-year OS, 27 percent four-year disease-free survival (DFS), 41 percent non-relapse mortality (NRM), 32 percent relapse rate, and 33 percent and 24 percent acute and chronic graft-versus-host disease, respectively [41].

Bridging therapy prior to transplantation, conditioning regimens, graft sources, and outcomes from other studies of allogeneic HCT are described below. (See 'Allogeneic transplantation' below.)

Older or less-fit — For patients with higher-risk CMML who are not suitable for transplantation, we suggest symptom-directed treatment using either a hypomethylating agent (HMA; eg, azacitidine or decitabine) or hydroxyurea, rather than supportive care alone. These agents offer symptom relief with acceptable toxicity profiles, but they do not provide long-term disease control.

The choice of an HMA versus hydroxyurea should be individualized. We favor initial therapy with azacitidine or decitabine because these agents can both cytoreduce and improve cytopenias. For patients who do not experience adequate relief of proliferative symptoms with an HMA, hydroxyurea can later be added or substituted. The choice of an HMA versus hydroxyurea may be influenced by the nature of symptoms, ease of administration, and patient preferences. Some experts select an agent according to the predominant nature of symptoms; as examples:

●Cytopenic symptoms – Transfusion-dependence and other symptoms related to cytopenias (eg, fatigue, dyspnea, infections, bleeding/bruising) are more likely to benefit from an HMA. (See 'Hypomethylating agents' below.)

●Proliferative symptoms – Predominantly proliferative symptoms (eg, splenomegaly, soaking sweats, weight loss) or renal dysfunction (due to high lysozyme levels) may benefit from hydroxyurea. (See 'Hydroxyurea' below.)

Studies that compared an HMA versus hydroxyurea for treatment of CMML yielded mixed results, but only a subset of such studies distinguished patients with proliferative versus cytopenic symptoms:

●The phase 3 DACOTA trial, which randomly assigned 170 patients with advanced proliferative CMML (one-third with CMML-2) to decitabine versus hydroxyurea, reported similar survival with both treatments, but decitabine was associated with a lower risk of CMML progression or transformation to acute leukemia [42]. Median OS (18.4 months with decitabine versus 21.9 months with hydroxyurea) and median event-free survival (EFS; 12.1 months versus 10.3 months, respectively) were similar with both treatments. However, decitabine was associated with a lower risk of CMML progression or transformation to acute leukemia (cause-specific HR 0.62 [95% CI 0.41-0.94]). Grade ≥3 infections occurred in 33 treated with decitabine and in 18 percent with hydroxyurea; hospitalization occurred in 60 percent and 40 percent, respectively.

●A retrospective international study reported a survival benefit with HMA therapy compared with hydroxyurea in patients with proliferative CMML [43]. Among 949 patients with CMML, 43 percent received an HMA and 41 percent received hydroxyurea. Treatment with an HMA was associated with superior median OS compared with hydroxyurea (20.7 months versus 15.6 months; HR 1.39 [95% CI 1.17-1.65]).

●In a study of azacitidine for CMML, initial treatment with azacitidine was associated with better median OS compared with initial treatment using hydroxyurea (28 months versus 6 months) in a matched-pair analysis [44].

Other studies of HMAs or hydroxyurea for CMML are presented below. (See 'Hydroxyurea' below and 'Hypomethylating agents' below.)

Lower-risk — We stratify management of lower-risk CMML according to the nature and severity of symptoms.

Symptomatic lower-risk CMML — We treat patients with symptomatic lower-risk CMML using a hypomethylating agent (HMA; eg, azacitidine, decitabine) or hydroxyurea, rather than allogeneic HCT, based on inferior survival among patients with lower-risk CMML who are transplanted.

The choice of an HMA versus hydroxyurea is guided by the nature of symptoms (eg, proliferative versus cytopenic), as described above. (See 'Older or less-fit' above.)

●A study that included 1114 patients from two CMML registries reported that transplantation was detrimental in patients with lower-risk CMML [39]. For patients with lower-risk CMML, five-year OS was 20 percent with allogeneic HCT, compared with 42 percent without transplantation. Allogeneic HCT before transformation to acute leukemia was associated with increased risk of death within two years of transplantation (HR 3.19 [HR] [95% CI 2.30-4.42]), with no significant change in long-term survival beyond that time point.

●A retrospective study reported that for patients with lower-risk CMML, there was no survival advantage associated with allogeneic HCT compared with non-transplant approaches (HR 0.69 [95% CI 0.43-1.11]) [38]. By contrast, transplantation was associated with superior survival for patients with higher-risk CMML, as described above. (See 'Younger, fit patients' above.)

Outcomes of other studies with HMAs or hydroxyurea for CMML are presented below. (See 'Hydroxyurea' below and 'Hypomethylating agents' below.)

Some younger, medically-fit patients with lower-risk CMML may instead choose allogeneic HCT because they place greater weight on the potential for cure than on the adverse effects. (See 'Allogeneic transplantation' below.)

Asymptomatic lower-risk CMML — For asymptomatic patients who have no critical cytopenias, we generally observe while monitoring for evidence of disease progression, as described below. (See 'Follow-up' below.)

The choice of observation versus treatment in asymptomatic patients should be individualized and may be influenced by disease features or comorbid conditions. As an example, one study reported a higher risk for acute kidney injury in patients with leukocyte count ≥35 × 10⁹/L or absolute monocyte count ≥5.4 × 10⁹/L, and treatment with hydroxyurea was associated with improved OS (22.6 versus 12.4 months) [16].

Selection of therapy for such patients is described above. (See 'Symptomatic lower-risk CMML' above.)

Frail patients — For frail patients, we generally offer supportive care alone, as described below. (See 'Supportive care' below.)

Criteria for medical frailty are presented above. (See 'Medical fitness' above.)

TREATMENTS

Symptom-directed therapies

Hydroxyurea — Hydroxyurea (HU) is an oral agent that can relieve splenomegaly, constitutional symptoms, renal dysfunction associated with high lysozyme levels, and other proliferative-type symptoms of CMML.

●Administration - A typical starting dose of HU is 500 mg twice daily or 1000 mg once daily by mouth.

The dose should be adjusted to an absolute neutrophil count of 500 to 1000/microL, with a goal of achieving a balance between reducing symptoms and exacerbating neutropenia, anemia, and thrombocytopenia. The onset of action is usually within three to five days of treatment initiation and the effect is short-lived after hydroxyurea is stopped; accordingly, dose adjustments should not be made more frequently than once per week to prevent wide fluctuations in the platelet count.

Complete blood count (CBC) and liver function tests should be obtained frequently during the first three months of treatment to monitor for rapid changes in blood counts due to dose modifications and to assess effects on liver function tests.

●Adverse effects - Toxicity of HU is usually mild and may include oral ulcers, hyperpigmentation, skin rash, and nail changes [45]. A small percentage of patients develop leg ulcers (eg, over the medial or lateral malleolus), nausea, diarrhea, or alopecia. Rare complications of HU include fever and abnormal liver function tests. Varying degrees of megaloblastic anemia and elevated mean corpuscular volume (MCV) reflect appropriate drug action and is not a reason to modify the dose. HU should not be used in pregnancy, in women with child-bearing potential who are unwilling to use highly effective contraception, and in women who are breastfeeding.

Persistent symptoms in a patient who cannot tolerate HU may benefit from an alternative oral antimetabolite (eg, 6-thiopurine), ruxolitinib, or a hypomethylating agent (HMA). The combination of HU plus supportive measures (eg, transfusions of red cells and platelets, erythropoiesis-stimulating agents) was shown to be superior to etoposide [46-48].

A phase 3 trial that randomly assigned patients with CMML with proliferative features to hydroxyurea versus etoposide was closed early after hydroxyurea achieved clear superiority (20 months overall survival [OS] versus 9 months; 60 percent overall response rate [ORR] versus 36 percent, respectively) [46].

Hypomethylating agents — Azacitidine and decitabine can provide symptomatic relief in patients with CMML, especially for cytopenia-related symptoms.

Treatment with an HMA is generally associated with a response in 30 to 60 percent of patients with CMML and a complete remission (CR) in <15 percent, but responses are typically not durable; median survival is generally one to three years [49]. The regimens for azacitidine and decitabine are the same as those used for myelodysplastic syndromes (MDS). (See "Treatment of lower-risk myelodysplastic syndromes (MDS)", section on 'Hypomethylating agents'.)

●Outcomes with HMAs - A retrospective international study that included more than 400 patients treated with an HMA reported median OS was 21 months [43]. A retrospective series of 151 patients treated with HMA-based therapy reported 24 months median OS and 14 months event-free survival (EFS) [50]. ORR was 75 percent (including 41 percent CR) and the rate of CR was higher with decitabine compared with azacitidine (58 versus 21 percent, respectively). A population-based study reported 17 months median OS among 225 patients treated with an HMA [51].

●Comparisons of azacitidine versus decitabine - No randomized trial has directly tested azacitidine versus decitabine for CMML and few studies have compared them in this setting. In a retrospective analysis, there was no difference in ORR between 56 patients treated with azacitidine compared with 65 patients treated with decitabine [52].

Other informative studies of HMA therapy in CMML include:

•Azacitidine – A phase 2 study randomly assigned patients (median age 70 years) with higher-risk MDS or CMML to one of three HMA-based regimens: azacitidine alone (75 mg/m²/day on days 1 to 7), azacitidine plus lenalidomide (10 mg/day on days 1 to 21), or azacitidine plus vorinostat (300 mg twice daily on days 3 to 9) [53]. Among 53 patients with CMML, overall survival (OS) was similar across all trial arms and median OS was not reached for any cohort after nearly two years of follow-up. OS was inferior for patients with adverse cytogenetic features, but it was not affected by treatment arm. Overall response rate (ORR) was higher for azacitidine plus lenalidomide compared with azacitidine (68 percent versus 28 percent), but lenalidomide dose reductions were associated with inferior OS. Serious adverse effects (AEs) were similar across trial arms, but those receiving combination therapy were more likely to undergo non-protocol-defined treatment modifications and to be withdrawn from therapy because of toxicities.

Among 30 patients treated with azacitidine in a multicenter study, ORR was 43 percent (including 7 percent CR); one-third of patients who were red cell and/or platelet transfusion-dependent at trial entry became transfusion-independent during treatment [54].

•Decitabine – Treatment with decitabine of 42 patients with CMML patients reported median OS of 17 months and 48 percent ORR (including 17 percent CR) in a multicenter study [55]. More than three-quarters of patients had higher-risk disease and two-thirds had proliferative features. Grade ≥3 AEs included anemia (29 percent), neutropenia (50 percent), and thrombocytopenia (38 percent).

Other studies with decitabine included 48 percent ORR (including 14 percent CR) and median OS of 17 months among 43 patients with higher-risk CMML [50], 25 percent ORR in a study of 31 patients [56], 39 percent ORR (including 11 percent CR) in 39 patients [57], and 68 percent ORR, 58 percent CR, and 48 percent two-year OS with minimal toxicity in 19 patients [58].

•Decitabine-cedazuridine – Treatment with oral decitabine-cedazuridine (an inhibitor of cytidine deaminase in the gut and liver) emulated the pharmacokinetic, pharmacodynamic, and safety profiles of intravenous decitabine, based on a phase 1 study of 43 evaluable patients [59].

A phase 2 study that included 21 patients with CMML and 59 patients with MDS used a fixed-dose preparation (35 mg decitabine plus 100 mg cedazuridine by mouth, daily for 5 days every 28 days) alternating with cycles of intravenous decitabine (20 mg/m² for 5 days) [60]. With median follow-up of 24 months and median treatment duration of 7 months, among all patients in the study (ie, both CMML and MDS), 18 percent achieved CR (with 9 month median duration of response) and 49 percent converted to red blood cell- and/or platelet-transfusion independence for ≥56 days. Oral decitabine-cedazuridine was associated with similar pharmacokinetic effects to standard intravenous decitabine in a cross-over study of 80 patients that included 17 patients with CMML [61].

●Prognostic factors - Factors associated with disease response have been inconsistent in various studies.

•Clinical features - Retrospective analysis of 412 patients who were treated with HMA reported that benefits of HMA therapy were greatest for patients with higher-risk CMML or myeloproliferative disease [43]. Another reported that these drugs are particularly ineffective in proliferative subtypes [55].

A retrospective study reported improved OS in association with an absolute monocyte count (AMC) <10,000/microL and <5 percent blasts in peripheral blood at the start of HMA therapy [62].

•Molecular features – Studies that analyzed outcomes of HMA therapy with molecular features have yielded mixed results.

A multicenter study of 174 patients reported with CMML reported that ASXL1 mutations predicted lower ORR, while mutations of RUNX1 or CBL and high leukocyte count predicted worse OS [63]. However, another study reported no association of outcomes with mutation status [52,56].

HMAs can improve epigenetic features in a subset of CMML patients [64], but they do not alter the mutational allele burden (even in responders) or prevent transformation to acute leukemia [52,65]. Disease evolution is inevitable for most patients, but among patients with higher-risk CMML, HMA therapy was associated with a lower rate and longer time to acute myeloid leukemia (AML) compared to untreated patients (36 percent versus 66 percent; 12 months versus 3 months, respectively) [16].

The US Food and Drug Administration (FDA) has approved azacitidine, decitabine, and oral decitabine-cedazuridine for treatment of CMML [60]. The European Medicines Agency has approved azacitidine in this setting.

Allogeneic transplantation — Allogeneic hematopoietic cell transplantation (HCT) is the only potentially curative option for patients with CMML, but it is associated with graft-versus-host disease (GVHD) and other causes of non-relapse mortality (NRM). Although 20 to 50 percent of patients achieve long-term disease control/cure with transplantation, given a median age >70 years at presentation, <20 percent of patients are eligible and <5 percent actually undergo allogeneic HCT [66].

Transplantation is associated with improved survival only for patients with higher-risk CMML; outcomes are inferior for patients with lower-risk CMML when transplanted prior to transplantation to AML, as discussed above. (See 'Higher-risk' above and 'Lower-risk' above.)

All patients who may be candidates for allogeneic HCT should be referred for consultation with a transplantation physician soon after diagnosis.

A decision to proceed with allogeneic HCT should consider the CMML risk category, medical fitness, availability of a suitable donor, and individual values and preferences. The risk of transplantation should be assessed using the hematopoietic cell transplantation-specific comorbidity index (HCT-CI) (table 4) or a similar instrument. Determination of CMML risk category and eligibility for allogeneic HCT are discussed above and separately. (See 'Risk-stratified management' above and "Determining eligibility for allogeneic hematopoietic cell transplantation".)

●Timing of transplantation – The optimal timing of transplantation is uncertain, but most experts plan for transplantation soon after higher-risk CMML is diagnosed or after achieving the best possible outcomes.

Because outcomes are better when allogeneic HCT is undertaken prior to transformation to acute leukemia, we generally treat patients who have a high or rising blast percentage with an HMA to control the blast count as a bridge to transplantation. However, practices vary among institutions and some favor intensive chemotherapy in this setting.

In a large study, the only predictor of survival after transplantation was complete remission (CR) at the time of transplant [41]. Some studies reported that treatment with an HMA to achieve a CR prior to transplantation was associated with better OS [41,67,68], but there is no persuasive evidence that cytoreductive therapy can improve outcomes. In one study, compared with intensive chemotherapy, patients treated with an HMA prior to transplantation had superior responses, less treatment-related mortality, and better progression-free survival [67]. Another study reported that patients who received azacitidine-based treatment before transplantation had outcomes similar to historical controls who received intensive chemotherapy prior to transplantation [53]. A single-center study reported that pre-transplant HMA did not appear to improve outcomes [40]. (See 'Hypomethylating agents' above.)

Transplantation prior to AML transformation was associated with better survival (five-year OS 51 percent versus 19 percent, respectively), but GVHD-free, relapse-free survival was only seven months, indicating the significant morbidity associated with allogenic transplantation [40]. A study that reported a trend toward better survival with allogeneic HCT in patients with higher-risk CMML, compared with no transplant; deaths increased in the first two years after HCT (hazard ratio [HR] 1.46), but not beyond that point [39].

●Conditioning regimen - There is no preferred conditioning regimen for transplantation in CMML. Most studies used myeloablative conditioning (MAC), but historical comparisons suggest that nonmyeloablative (NMA) or reduced intensity conditioning (RIC) can achieve similar results. When available, a matched related donor is the preferred graft source, but a matched unrelated donor, haploidentical donor, or umbilical cord blood may be an acceptable graft source. (See "Preparative regimens for hematopoietic cell transplantation" and "Donor selection for hematopoietic cell transplantation".)

In a study in which 43 of 73 patients underwent NMA conditioning, three-year OS, NRM, and rate of relapse were 32, 36, and 35 percent, respectively [69].

●Outcomes with allogeneic HCT in CMML – Across numerous studies, allogeneic HCT was associated with approximately 30-40 percent five-year OS and 20-30 percent transplant-related mortality (TRM) [21,70].

•A multicenter study that included 513 patients with CMML (median age 53 years) reported 33 percent four-year OS, 27 percent disease-free survival (DFS), 41 percent non-relapse mortality (NRM), and 32 percent relapse rate [41]. Nearly equal numbers of patients in this study underwent MAC versus RIC.

•In a registry study of 209 patients, rates of OS for higher-risk patients were 38 percent, 32 percent, and 19 percent at one, three, and five years, respectively; corresponding rates for lower-risk patients were 61, 48, and 41 percent, respectively [71].

Most other studies of transplantation for CMML were small and included patients with heterogeneous disease features [69,72-78].

●Prognostic factors – CR at the time of transplantation is associated with improved outcomes [41]. Other factors associated with transplant outcomes include cytogenetics [72,74], blast percentage in marrow [74], age [72], comorbidities [71,72,75], and bone marrow as a graft source [71]. In one study, extensive chronic GVHD was associated with better survival and lower leukemia-associated mortality [79].

Supportive care — Supportive care should be provided to all patients with CMML, whether they receive another treatment or not.

Patients should be treated with antibiotics for infections and transfusions for symptomatic anemia or thrombocytopenia. Repeated red blood cell (RBC) transfusions carry the risk of iron overload; some patients may benefit from iron chelation therapy after numerous RBC transfusions. Erythropoietin-stimulating agents (ESAs) can improve hemoglobin levels and lessen transfusion-dependence in patients with symptomatic anemia, especially when the baseline erythropoietin (EPO) level is <500 IU/L [80,81]. In a study in 94 patients with CMML, ESA treatment was associated with an erythroid response rate of 64 percent and transfusion-independence in 31 percent [80].

Supportive care for CMML is like that for patients with myelodysplastic syndromes, as described separately. (See "Myelodysplastic syndromes/neoplasms (MDS): Management of hematologic complications in lower-risk MDS".)

MONITORING

Treatment response — Response to treatment should be judged by relief of disease-associated symptoms, hematologic response, achievement of transfusion-independence, and improved quality of life (QOL).

Response can be judged according to clinical benefit, such as relief of symptoms and decreased need for transfusions. Response criteria used in clinical trials (table 5A-B) include resolution of hepatosplenomegaly and other extramedullary disease, morphologic response in bone marrow, and QOL (reflected by the Myeloproliferative neoplasms symptom assessment total symptom score [MPN-SAF TSS]) [82].

Discontinuation of symptom-directed therapy is indicated for patients who cannot tolerate therapy or whose disease progresses while on treatment.

Disease progression can be defined as manifestation of two major criteria, one major and two minor criteria, or three minor criteria, as follows:

●Major criteria – Increase in blast count; cytogenetic evidence of clonal evolution; new extramedullary disease (either worsening splenomegaly or extramedullary disease outside of the spleen).

●Minor criteria – Transfusion dependence; decrease in granulocytes or platelets by ≥50 percent from maximum response; decrease in hemoglobin by 1.5 g/dL or more from best response or baseline; increasing symptoms; molecular evidence of clonal evolution.

MPN-SAF TSS has been validated in retrospective studies of hypomethylating agents in CMML [48,83].

Follow-up — Patients should be followed to assess response to therapy and to monitor for disease progression.

There is no consensus schedule and protocol for patient follow-up. We generally schedule follow-up visits every one to six months, which include clinical evaluation and complete blood count (CBC) and differential count. The schedule may be influenced by concerns of the clinician or patient.

Routine bone marrow examinations are not required, unless blood counts deteriorate unexpectedly or there is other evidence of possible progression to acute myeloid leukemia (eg, rising blast counts) [84]. Alternatively, some clinicians perform a bone marrow biopsy and aspirate after two to four cycles of hypomethylating agent therapy to guide further treatment decisions.

All patients with CMML should receive age-appropriate health monitoring and smoking should be discontinued. Other aspects of follow-up are presented separately. (See "Treatment of lower-risk myelodysplastic syndromes (MDS)", section on 'Monitoring'.)

Patients with CMML can receive killed or recombinant immunizations, but live vaccines should generally not be given to immunocompromised individuals. Specific recommendations regarding immunizations for patients with cancer and other immunocompromised conditions are summarized (figure 1) and discussed in greater detail separately. (See "Immunizations in adults with cancer".)

RELAPSED OR REFRACTORY DISEASE — We encourage participation in a clinical trial for patients with relapsed or refractory CMML.

There is no consensus regarding a preferred treatment for relapsed or refractory (r/r) CMML and management is informed by prior therapy.

●Medically-fit – For all medically-fit patients who were previously treated with symptom-directed therapy, we seek to proceed to allogeneic hematopoietic cell transplantation (HCT). Given the grave prognosis of r/r CMML, we broaden medical eligibility for transplantation and consider various alternative graft sources, including matched unrelated donors, haploidentical donors, and umbilical cord grafts.

●Less-fit – Management of r/r CMML is guided by prior therapy:

•Prior hydroxyurea treatment - For patients who progressed on or were intolerant of hydroxyurea (HU), we generally offer treatment with another oral antimetabolite (eg, 6-thiopurine), a hypomethylating agent (HMA), or ruxolitinib [85].

•Prior HMA therapy - For patients who progressed on or were intolerant of an HMA, we treat with HU if there are proliferative symptoms (eg, troublesome splenomegaly, constitutional symptoms) and the blood counts are adequate. There is no evidence that patients who did not respond to one HMA will respond to an alternative HMA.

We encourage patients with r/r CMML to participate in a clinical trial. Agents under investigation for treatment of r/r CMML include RAS/mitogen-activated protein kinase signaling pathway inhibitors (tipifarnib and trametinib), guadecitabine plus atezolizumab, lenzilumab, tagraxofusp, NMS-035290288 (multikinase inhibitor), AZD6738 (ATR inhibitor), MBG453 (TIM3 inhibitor), CPX351, CB839 (glutaminase inhibitor), H3B-8800 (spliceosome inhibitor) venetoclax, pevonedistat, and eltrombopag [86-93].

Information and instructions for referring a patient to an appropriate research center can be obtained from the US National Institutes of Health (www.clinicaltrials.gov). The Myelodysplastic Syndromes Foundation (www.mds-foundation.org/) and the Aplastic Anemia and MDS International Foundation (www.aamds.org) maintain websites that contain additional information as well as listings of clinical trials in this disorder [94].

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: Myeloproliferative neoplasms".)

SUMMARY AND RECOMMENDATIONS

●Description – Chronic myelomonocytic leukemia (CMML) is a malignant hematopoietic disorder characterized by monocytosis and bone marrow dysplasia, often accompanied by cytopenias and splenomegaly. CMML has features of both myeloproliferative neoplasms (MPN) and myelodysplastic syndromes/neoplasms (MDS).

●Pretreatment evaluation – Evaluation should identify disease-related symptoms and assess medical fitness, according to comorbid conditions (table 4) and performance status (table 6). (See 'Pretreatment evaluation' above and 'Medical fitness' above.)

●Risk stratification – A CMML-specific prognostic scoring system that incorporates cytogenetic and molecular features is used to classify CMML as higher-risk or lower-risk disease, using one of the following :

•MMM (Mayo molecular model) (See 'MMM (Mayo molecular model)' above.)

•CPSS-Mol (CMML-specific prognostic scoring system with molecular features) (See 'CPSS-Mol' above.)

•GFM (Groupe Francophone des Myélodysplasies) (See 'GFM (Groupe Francophone des Myélodysplasies)' above.)

●Higher risk – Treatment choice is informed by eligibility for allogeneic hematopoietic cell transplantation (HCT), including age (generally ≤70 years) and no major comorbid illnesses, an available graft donor, and patient preference (see 'Higher-risk' above):

•Fit, younger patients – For fit, younger patients with higher-risk CMML, we suggest allogeneic HCT, rather than treatment with a hypomethylating agent (HMA) or hydroxyurea, because transplantation is the only treatment with potential to cure CMML. (Grade 2C) (See 'Younger, fit patients' above.)

•Older or less fit patients – For patients with higher-risk CMML who are not suitable for transplantation, we suggest treatment with an HMA (eg, azacitidine, decitabine) or hydroxyurea, rather than supportive care alone. (Grade 2C) Choice of initial therapy with an HMA versus hydroxyurea is discussed above. (See 'Older or less-fit' above.)

●Lower risk – For patients with lower-risk CMML, management is guided by the nature and severity of symptoms (see 'Lower-risk' above):

•Symptomatic – For symptomatic patients with lower-risk CMML, we suggest treatment with an HMA or hydroxyurea, rather than allogeneic HCT or supportive care alone. (Grade 2C). (See 'Symptomatic lower-risk CMML' above.)

•Asymptomatic and no critical cytopenias – Most patients can be observed without treatment, while monitoring for disease progression. Some asymptomatic patients (eg, high leukocyte count monocyte count) may benefit from treatment. (See 'Asymptomatic lower-risk CMML' above.)

●Monitoring

•Treatment response – Response to therapy is based on relief of symptoms, hematologic response, transfusion-independence, and improved quality of life (table 5A). (See 'Treatment response' above.)

•Follow-up – Monitoring for disease progression and treatment toxicity is described above. (See 'Follow-up' above.)

●Relapsed/refractory disease – We encourage participation in a clinical trial. Management is informed by prior therapy and medical fitness, as described above. (See 'Relapsed or refractory disease' above.)

ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledge David P Steensma, MD, who contributed to an earlier version of this topic review.