Case Study: Predicting the Course of Treatment in an Elderly Patient with Refractory Anemia


Learning objectives

  • Summarize the diagnostic and prognostic tests that should be performed to appropriately diagnose MDS in an elderly patient and to accurately risk stratify that patient
  • Outline potential comorbidities that may be present in an elderly patient that may complicate diagnosis and/or prognosis in MDS
  • Identify questions commonly asked by patients and/or their caregivers concerning diagnosis and prognosis in MDS

Activity overview

Myelodysplastic syndromes (MDS) are clonal disorders primarily of the elderly that are characterized by ineffective hematopoiesis and increased risk of transformation to acute myeloid leukemia (AML). MDS are considered clinically heterogeneous conditions because they can present with widely varying degrees of severity and because patients with similar features at diagnosis can have large differences in how their disease evolves over time.

There have been significant advances across many aspects of MDS including diagnosis (development/investigation of new diagnostic techniques), classification (new classification criteria have been developed), prognostic scoring systems (developed and validated), and new treatments have been approved.

This MDS Quarterly e-Newsletter incorporates a case vignette that illustrates the use of current diagnostic tests and prognostic scoring systems in daily clinical practice as well as addresses the practical aspects of delivering prognostic information to patients and/or their caregivers.

Case presentation

Chief complaint

Unexplained fever, bruising, and recent onset of episodes of nontraumatic epistaxis.

History of present illness

The patient is a 74-year-old male who presented to his primary care physician complaining of increased fatigue and dyspnea approximately 12 months ago. Laboratory studies at the time revealed normal complete blood cell counts (CBCs) and hemoglobin levels. Fatigue and dyspnea were attributed to worsening heart failure and no further investigations were carried out.

Approximately 6 months ago, the patient developed an upper respiratory infection that resolved only after 3 courses of antibiotics. Laboratory studies at the time showed a white blood cell count (WBC) of 3.6 x 109 with 1.90 x 109 neutrophils, hemoglobin 11.2 g/dL, and mean corpuscular volume (MCV) of 110 fl. Since then, the patient has experienced two further respiratory infections – the first approximately 2 months ago and the second 6 weeks ago.

Presentation of the patient at this time prompted reassessment of warfarin dose, but the INR was within the therapeutic range. Given the history of recent infections as well as the unexplained bleeding events, the patient was referred to hematology for evaluation.

Medical history

Patient has New York Heart Association grade 1 heart failure. Patient has normal hepatic and renal function.

Current medications include losartan, warfarin, (INR: 2 – 2.5), spironolactone, simvastatin, and furosemide.

Initial evaluation

Current laboratory studies reveal a WBC of 2.9 x 109 with 1.80 x 109 neutrophils (absolute), and granulocytes 0.6 x 109, hemoglobin 9.5 g/dL, and platelet count of 95 x 109. The patient has normal vitamin B12 and folic acid levels, is sero-negative for human immunodeficiency virus, and iron stores and iron saturation are normal.

Bone marrow biopsy reveals 70% cellularity with megablastoid erythroid cells, 9% myeloblasts, and no Auer rods. Cytogenetics include abnormal karyotype; del(11q), in 5 of 20 metaphase spreads. The other 15 metaphase spreads show a normal karyotype.


Based on the initial evaluation, the diagnosis assigned to this patient is refractory anemia with excess blasts-1 (RAEB-1). The predicted clinical course for this patient is progressive pancytopenia with increased likelihood of repeated pneumonias.

Diagnostic evaluation

According to the current National Comprehensive Cancer Network (NCCN) guidelines,1 initial evaluation of patients with suspected MDS includes:

  • CBC, platelets, and differential reticulocyte count
  • Examination of a peripheral blood smear
  • Bone marrow aspiration with iron stain, biopsy, and cytogenetics with karyotyping
  • Serum erythropoietin (prior to RBC transfusion)
  • Red blood cell (RBC) folate and serum B12
  • Serum ferritin, iron, total iron-binding capacity
  • Documentation of transfusion history
  • Thyroid-stimulating hormone (TSH) to rule out hypothyroidism

Additionally, clinical history should capture the timing, severity, and tempo of abnormal cytopenias; incidence of prior infections or bleeding episodes; and number of transfusions. To support a diagnosis of MDS, cytopenias observed should have persisted for a period of 6 months. In addition, blood count stability should be used to distinguish MDS from evolving acute myeloid leukemia (AML).1

Peripheral blood smears are used to determine the degree of dysplasia; while bone marrow aspiration and a biopsy are necessary to evaluate the degree and proportions of hematopoietic cell maturation abnormalities, percentage of marrow blasts, marrow cellularity, occurrence of sideroblasts, and fibrosis.1 Estimates of blast percentage should be determined by histopathology, as flow cytometry does not provide the same prognostic information.1 Cytogenetics for bone marrow samples by standard karyotyping (at least 20 metaphase spreads) should be obtained because they are of major prognostic significance.1 Fluorescence in situ hybridization (FISH) is inadequate to accurately characterize patients.

For patients requiring platelet transfusions for severe thrombocytopenia, human leukocyte antigen (HLA) typing (A and B) may be helpful.1 Expanded flow cytometric assays may be considered when performed by experienced laboratories and should only be used in situations where diagnosis cannot be made using traditional approaches. Additional genetic screening can be considered in patients with familial cytopenias.1

Prognostic stratification

Tools for prognostication

Once an accurate diagnosis of MDS is achieved, it is important to determine the prognosis for each patient, as this will impact the treatment selection.1,2 Several prognostic systems have been developed to predict disease progression in MDS including the International Prognostic Scoring System (IPSS), the IPSS-Revised (IPSS-R), the WHO-based Prognostic Scoring System (WPSS); and specifically for low-risk (LR) patients, the LR-Prognostic Scoring System (LR-PSS).1 Until recently, the IPSS served as the gold standard for predicting prognosis. Despite the utility of the IPSS, heterogeneity within some of the risk categories remained. Greater understanding of the cytogenetics of MDS, as well as the disease pathophysiology, led to further analyses and subsequently the IPSS-R was published in 2012.3

The NCCN guidelines advocate use of the IPSS-R as the preferred prognostic tool given its greater accuracy; however, the guidelines also recognize that the other scoring systems are also valuable.1 Therefore, the IPSS-R should be performed as standard of care, but in addition, when evaluating the efficacy and safety of therapeutic options, it is strongly suggested that the patient also be evaluated according to the risk-assessment system that was used in clinical study for any specific therapeutic intervention that is being considered, including the IPSS as necessary.

The IPSS-R uses 5 main features and the model is shown in Table 1. Risk is calculated by adding together the scores for each individual feature and the risk groups are defined as: very good (less than 1.5), good (greater than 1.5 to 3), intermediate (greater than 3 to 4.5), poor (greater than 4.5 to 6), and very poor (greater than 6).3

Table 1. Definition of the Revised International Prognostic Scoring System (IPSS-R)









Cytogenetic risk group

Very good




Very Poor

Bone marrow blasts


>2% to <5%

5% to 10%


Hemoglobin g/dL


8 to <10




50 to <100





SOURCE: Greenberg, et al.3

Cytogenetic risk is determined as shown in Table 2.

Table 2. Cytogenetic risk groups

Cytogenetic risk group

Cytogenetic abnormalities

Very good

−Y, del(11q)


Normal, del(5q), del(12p), del(20q), double including del(5q)


del(7q), +8, +19, i(17q), any other single or double independent clones


−7, inv(3)/t(3q)/del(3q), double including −7/del(7q), complex: 3 abnormalities

Very poor

Complex: >3 abnormalities

 SOURCE: Greenberg, et al.3

Age is a another major prognostic parameter for overall survival, but not for AML evolution.3 In the IPSS-R prognostic model, the data were shown for age 70 years, which was the near median age of the patient cohort. Greenberg and colleagues3 provide a formula, which permits statistical adjustment of survival prognosis for patients of all ages.

Adjustment of IPSS-R for age3:
(years – 70) x [0.05 – (IPSS-R risk score x 0.005)]

Applying the IPSS-R to the patient case

Based on the IPSS-R prognostic model, the risk group of the current patient would be classified as intermediate.

Cytogenetic risk group
Very good
Bone marrow blasts
Hemoglobin g/dL
Total score

However, the age of the patient (74 years) requires that the risk score be adjusted, thus the age-adjusted risk score for the patient would be 4.12 and the patient's risk would still be classified as intermediate. Please see the MDS Foundation for the Revised International Prognostic Scoring System (IPSS-R) for Myelodysplastic Syndromes Risk Assessment Calculator, available at

Patients classified as intermediate are considered to be at higher risk according to the NCCN guidelines.1 Treatment for patients at higher risk is selected based on their ability to tolerate intensive therapy such as allogeneic stem cell transplant or intensive chemotherapy.1 Factors that determine a patient’s eligibility for intensive therapy include age, performance status, presence of comorbidities, psychosocial status, patient preference, as well as donor and caregiver availability.1 Thus, for this patient, it is important to consider the comorbidities in order to appropriately select treatment.

Impact of comorbidities on prognosis

The advanced age of patients with MDS poses a challenge for treatment tolerability and outcomes predominantly due to the presence of comorbidities. Approximate half of newly diagnosed patients with MDS present with one or more comorbidities with the most common being cardiac disease and diabetes.4 Several tools for assessing the presence and degree of comorbidities exist; however, only one, the MDS Comorbidity Index (MDS-CI)5 has been shown to improve the prognostic accuracy of the IPSS-R.6,7 The MDS-CI has been shown to identify MDS patients who can have better outcomes with 5-azacitidine, for instance.8 

The MDS-specific comorbidity index (MDS-CI) includes five comorbidities (Table 3). The five comorbidities listed were found to be independently associated with the risk of non-leukemic death in multivariable analysis, and each of them was assigned a score proportional to the regression coefficient of the multivariable Cox’s proportional hazards model. This score is taken into account if the specific comorbidity is present, and the MDS-CI is obtained as the sum of individual variable scores.

Table 3. Calculation of the MDS-specific comorbidity index (MDS-CI).


Hazard Ratio*

Variable Weighted Score

Cardiac disease

3.57 (P < .001)


Moderate-to-severe hepatic disease

2.55 (P = .01)


Severe pulmonary disease

2.44 (P = .005)


Renal disease

1.97 (P = .04)


Solid tumor

2.61 (P < .001)



Sum of individual variable scores

Low risk


Intermediate risk

1 – 2

High risk


SOURCE: Della Porta, et al. 20115
* Hazard ratio obtained from a Cox survival analysis with non-leukemic death as an outcome.

At this time, however, the NCCN guidelines do not provide specific recommendations regarding the optimal comorbidity index to be used, but the guidelines do specify that a thorough evaluation of the presence of comorbidities should be conducted.1

The impact of comorbid conditions affects prognosis differently in low- and high-risk patients with MDS.9 In low-risk patients, comorbidity affects the prognosis by directly increasing the risk of non-leukemic death.9 Conversely, in high-risk patients, the clinical relevance of mild or moderate comorbidity is overcome by the severity of the MDS.9 In these patients, however, comorbidity influences the outcome by reducing eligibility for, and tolerance of, treatments.

Communicating diagnosis and prognosis

Clear communication of diagnosis and prognosis between the health care team and the patient and/or caregivers is crucial. Unfortunately, few health care providers are trained to deliver bad news, and many may inadvertently say things that cause unnecessary emotional pain.10,11

People vary in their ability to cope with bad news, thus physicians need to individualize their delivery of diagnostic and prognostic news based on the patient’s desires and needs.12 Immediate reactions to diagnosis/prognosis may include crying, stunned silence, disbelief, anger, or acute distress. Doctors should let patients know that “it is quite normal to feel this way.” Importantly, physicians should allow for silence and tears and proceed at the patient’s pace. It is likely that the patient will not process the information when the diagnosis is first made and therefore, discussions surrounding prognosis may have to take place at several visits.

Patients want clear information, provided in plain language, about their diagnosis, prognosis, and treatment.13 Patients are unlikely to understand what MDS is and will require education so that they will be able to explain what their disorder is, not only to their friends and family, but also to other health care providers who may not care for many patients with these syndromes. Important information to convey to patients is that, unlike many cancers, MDS is a chronic disease that will require management for the rest of the patient's life. Patients will likely ask if there is a cure for MDS and how long are they likely to survive. Treatment outcomes and side effects must be clearly communicated to patients.

Practical tips for discussing diagnosis and prognosis include adopting a sensitive, honest, and straightforward approach.14 Start by confirming that a patient wants to receive prognostic information and the type (eg, risk score, chance of cure, survival estimates) and format of information (eg, words, numbers, graphs).14 When providing prognostic estimates, clearly define the limitations of prognostic formulations and if providing a time frame, emphasize a range and not a specific endpoint.14 Provide up-to-date information and be able to explain its source. Emphasize hope-giving aspects of information, for example extraordinary survivors.14 Offset negatives by providing positive information first. Lastly, broaden the discussion of prognosis to include the effect of MDS on the individual’s life.13