There is growing evidence to indicate that comorbidity has an unfavorable effect on life expectancy of patients with myelodysplastic syndromes (MDS), and significantly influences clinical decision making.1
Recently, Naqvi and colleagues evaluated the prognostic impact of comorbidity in MDS2 by using the Adult Comorbidity Evaluation-27 (ACE-27) scale, a measure of comorbidity developed for assessment of patients with solid tumor.3 They found that comorbidity significantly affects survival and improves the prognostic stratification of patients with MDS.
Despite these remarkable results, this study has potential weaknesses. ACE-27 failed to stratify the prognosis in subjects aged 65 years or older, the great majority of MDS patients, as well as in the IPSS low-risk subgroup,2 in whom comorbidity was reported to considerably increase the risk of death.1 In addition, the use of a general measure of comorbidity may have led to possible interactions between hematologic disease and comorbid conditions being overlooked.
The choice of the instrument to measure comorbidity is crucial to effectively capture and score the effect of comorbidity in the population of interest. Available measures can be divided into two groups: general measures, and disease-specific measures.4 Like “ready-to-wear” clothes, general comorbidity measures are intended for use in several different populations.3,5 These scores are widely used and have a good reliability. However, the weakness of these tools lies in the assumption that each comorbidity computed in the measure has the same impact on different diseases and patient populations.4 Conversely, disease-specific comorbidity measures are developed and tested in a single disease population and intended for use in that specific setting. Like “made-to-measure” garments, disease-specific comorbidity measures have a conceptual advantage in that they account for specific features unique to the population of interest.
Recent data suggest that in MDS cardiac comorbidity may have not just an additive detrimental effect per se but interacts with anemia in worsening the course of the disease.6 In this setting, a general comorbidity measure such as the Charlson-Comorbidity-Index failed to provide prognostic information, whereas the Hematopoietic Cell Transplantation-specific Comorbidity Index appears to be only partially effective in the general MDS patient population.7,8 We developed an MDS-specific comorbidity index (MDS-CI) in which the comorbid conditions to be included in the score were selected on the basis of a multivariable regression accounting for possible associations between clinical conditions and interactions with MDS-specific features, such as anemia. The score obtained is a combination of five clinical conditions, and identifies three risk groups with different probabilities of survival.8 The prognostic value of the MDS-CI was validated in two large independent cohorts of patients.8,9
To evaluate the effectiveness of a “ready-to-wear” versus a “made-to-measure” strategy in MDS, we compared the ability to capture prognostic information on comorbidity of ACE-27 and MDS-CI in 840 consecutive patients diagnosed with MDS at our institution between 1992 and 2007. When comparing MD Anderson and Pavia patient populations, no significant differences were found in age or IPSS risk distribution. A significant proportion of the MD Anderson patients (>50%) underwent treatments potentially modifying the natural history of the disease while only 17% of the Pavia patients were treated.
In our patient population, the four ACE-27 risk groups showed no significantly different probabilities of survival (P=0.11, Figure 1). In a multivariable analysis including demographic and disease-related factors, ACE-27 did not show a significant effect on survival (HR 1.01–1.35; P=0.9–0.06).
We then carried out a multivariable analysis to evaluate which comorbid conditions among those included in ACE-27 have a prognostic effect in MDS patients, and found that only myocardial infarction (HR 1.48; P=0.003), congestive heart failure (HR 1.32; P=0.05), respiratory system disease (HR 1.74; P=0.033), paralysis (HR 13.19; P<0.001), solid tumor (HR 1.47; P=0.002), leukemia/lymphoma (HR 1.894; P=0.041), and neuromuscular diseases (HR 6.24; P=0.039) showed an independent significant effect on survival.
As a further step, we focused on the clinical definition of these items and the grading of severity adopted in MDS-CI and ACE-27, and evaluated their goodness of fit in the MDS patient population. To this purpose, Akaike information criterion (AIC) was employed which allows the evaluation of a model by combining goodness of fit and complexity (among different models, a lower AIC indicates a better trade-off between fit and complexity, while models with an AIC difference of 4 or more with respect to the reference model have considerably less support).10 For all comorbid conditions considered in the analysis, clinical definition and grading of severity by MDS-CI had greater goodness of fit for capturing the prognostic information of these comorbid conditions in MDS patients compared to ACE-27 (AIC for ACE 27 vs. MDS-CI cardiac disease 2929 vs. 2925, hepatic disease 2932 vs. 2930, pulmonary disease 2931 vs. 2928, renal disease 2930 vs. 2926, and solid tumor 2929 vs. 2926). Taken together, these data strengthen the importance of not basing definition of the prognostic value of comorbidity in the clinical setting of MDS patients exclusively on the severity of the comorbid condition per se, but also taking MDS-specific features into account.
- The information provided by the authors about contributions from persons listed as authors and in acknowledgments is available with the full text of this paper at www.haematologica.org.
- Financial and other disclosures provided by the authors using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are also available at www.haematologica.org.
- Della Porta MG, Malcovati L. Clinical relevance of extra-hematologic comorbidity in the management of patients with myelodysplastic syndrome. Haematologica. 2009; 94(5):602-6. PubMedhttps://doi.org/10.3324/haematol.2009.005702Google Scholar
- Naqvi K, Garcia-Manero G, Sardesai S, Oh J, Vigil CE, Pierce S, Lei X, Shan J, Kantarjian HM, Suarez-Almazor ME. Association of comorbidities with overall survival in myelodysplastic syndrome: development of a prognostic model. J Clin Oncol. 2011; 29(16):2240-6. PubMedhttps://doi.org/10.1200/JCO.2010.31.3353Google Scholar
- Piccirillo JF, Tierney RM, Costas I, Grove L, Spitznagel EL. Prognostic importance of comorbidity in a hospital-based cancer registry. Jama. 2004; 291(20):2441-7. PubMedhttps://doi.org/10.1001/jama.291.20.2441Google Scholar
- Geraci JM, Escalante CP, Freeman JL, Goodwin JS. Comorbid disease and cancer: the need for more relevant conceptual models in health services research. J Clin Oncol. 2005; 23(30):7399-404. PubMedhttps://doi.org/10.1200/JCO.2004.00.9753Google Scholar
- Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987; 40(5):373-83. PubMedhttps://doi.org/10.1016/0021-9681(87)90171-8Google Scholar
- Malcovati L, Della Porta MG, Strupp C, Ambaglio I, Kuendgen A, Nachtkamp K, Travaglino E, Invernizzi R, Pascutto C, Lazzarino M. Impact of the degree of anemia on the outcome of patients with myelodysplastic syndrome and its integration into the WHO classification-based Prognostic Scoring System (WPSS). Haematologica. 2011; 96(10):1433-40. PubMedhttps://doi.org/10.3324/haematol.2011.044602Google Scholar
- Zipperer E, Pelz D, Nachtkamp K, Kuendgen A, Strupp C, Gattermann N, Haas R, Germing U. The hematopoietic stem cell transplantation comorbidity index is of prognostic relevance for patients with myelodysplastic syndrome. Haematologica. 2009; 94(5):729-32. PubMedhttps://doi.org/10.3324/haematol.2008.002063Google Scholar
- Della Porta MG, Malcovati L, Strupp C, Ambaglio I, Kuendgen A, Zipperer E. Risk stratification based on both disease status and extra-hematologic comorbidities in patients with myelodysplastic syndrome. Haematologica. 2011; 96(3):441-9. PubMedhttps://doi.org/10.3324/haematol.2010.033506Google Scholar
- Breccia M, Federico V, Latagliata R, Mercanti C, D’Elia GM, Cannella L, Loglisci G, Salaroli A, Santopietro M, Alimena G. Evaluation of comorbidities at diagnosis predicts outcome in myelodysplastic syndrome patients. Leuk Res. 2011; 35(2):159-62. PubMedhttps://doi.org/10.1016/j.leukres.2010.06.005Google Scholar
- Akaike H. A new look at the statistical model identification. IEEE Transactions on Automatic Control. 1974; 19(6):716-23. https://doi.org/10.1109/TAC.1974.1100705Google Scholar