JGO-00207; No. of pages: 8; 4C: J O U RN A L OF GE RI A T RI C O NC O L O G Y XX ( 20 1 4 ) XX X–XX X
Available online at www.sciencedirect.com
ScienceDirect
Clustering of comorbidities is related to age and sex and impacts clinical outcome in myelodysplastic syndromes C. Bammera , W.R. Sperrb , G. Kemmlerc , F. Wimazald , T. Nösslinger e , A. Schönmetzlere , O. Krieger f , M. Pfeilstöcker e,g , P. Valent e,g , R. Stauder a,⁎ a
Department of Internal Medicine V (Hematology and Oncology), Innsbruck Medical University, Innsbruck, Austria Department of Internal Medicine I, Division of Hematology and Hemostaseology, Vienna Medical University, Vienna, Austria c Department of Psychiatry and Psychotherapy, Innsbruck Medical University, Innsbruck, Austria d Department of Obstetrics and Gynecology, Vienna Medical University, Austria, e Third Medical Department for Hematology and Oncology, Hanusch Hospital, Vienna, Austria f Department of Internal Medicine I, Hospital of the Elisabethinen, Linz, Austria g Ludwig Boltzmann Cluster Oncology, Vienna, Austria b
AR TIC LE I N FO
ABS TR ACT
Article history:
Objectives: Myelodysplastic syndromes (MDS) are typical diseases of the elderly. The clinical
Received 19 September 2013
outcome of a well-characterized cohort of patients with MDS was analyzed for prevalence
Received in revised
and impact of comorbidities to establish the basis for tailored treatment algorithms. Focus
form 30 December 2013
was on age- and sex-related differences.
Accepted 14 February 2014
Material and Methods: The hematopoietic cell transplantation-comorbidity index (HCT-CI) was assessed in 616 well-defined patients from the Austrian MDS platform (median age: 71 years).
Keywords:
Results: Most patients displayed one (24.5%) or more (23.7%) comorbidities. The highest
Age
frequencies were observed for cardiovascular disease (28.4%), diabetes (12.2%), and prior
Comorbidities
tumors (9.9%). Comorbidities were more frequent (mean number: 0.92 vs. 0.74 [male vs.
MDS
female]; p = 0.030) and more severe in men than in women (mean HCT-CI score: 1.41 vs.
Myelodysplastic syndromes
1.09 [male vs. female]; p = 0.016). Elderly patients (65 + years) showed a higher prevalence
Prognosis
of comorbidities than younger patients (HCT-CI score: 1.52, mean in 65 +, vs. 0.24 and 0.76
Sex
in < 45 years and 46–65 years, respectively) (p < 0.001). These differences were most
Survival
pronounced for cardiovascular disease, diabetes, and prior tumors (p < 0.001). Presence of
Assessment
cardiac arrhythmia or prior solid tumor was significantly associated with shorter overall survival (p = 0.023, 0.024, respectively). Moreover, HCT-CI risk grouping remained an independent prognostic parameter for survival in multivariate analysis. Conclusions: Comorbidities impact clinical outcome in elderly patients with MDS. Distinct diseases cluster in an age- and sex-related manner, which may have clinical implications when designing individualized therapies. Comorbidities should be evaluated with established scores and integrated in decision making. © 2014 Published by Elsevier Ltd.
⁎ Corresponding author at: Department of Internal Medicine V (Hematology and Oncology), Innsbruck Medical University, Anichstrasse 35, 6020 Innsbruck, Austria. Tel.: + 43 512 504 23255. E-mail address: [email protected] (R. Stauder).
http://dx.doi.org/10.1016/j.jgo.2014.02.002 1879-4068/© 2014 Published by Elsevier Ltd.
Please cite this article as: Bammer C., et al, Clustering of comorbidities is related to age and sex and impacts clinical outcome in myelodysplastic syndromes, J Geriatr Oncol (2014), http://dx.doi.org/10.1016/j.jgo.2014.02.002
2
J O U RN A L OF GE RI A TR I C O NC OLO G Y XX ( 20 1 4 ) XX X–XXX
1. Introduction Myelodysplastic syndromes (MDS) encompass a broad spectrum of clonal hematological disorders characterized by dysplastic changes and ineffective production of one or more hematopoietic cell lineages, resulting in cytopenias and symptoms thereof.1–3 MDS are commonly classified according to French–American– British (FAB) Cooperative Group or World Health Organization (WHO) categories. These categories are typically based on blast cell counts, signs of dysplasia in peripheral blood or bone marrow cells, as well as evaluation of karyotype. Significant differences in survival among these groups, as well as varying likelihood of developing secondary acute myeloid leukemia (sAML) have been described.4 Highly individual courses of MDS, even within particular groups, and the need to individualize therapeutic decisions make prognostic scores a necessity. Actually, several scoring systems are applied in clinical practice. The gold standard in prognostication in MDS remains the International Prognostic Scoring System (IPSS),5 which was recently updated (IPSS-R). Based on the number and degree of cytopenias, blast counts, cytogenetic risk groups, and age the IPSS-R defines different risk groups for overall survival and leukemic evolution.6 The transfusion need of patients was integrated as a relevant parameter in the WHO classification-based prognostic scoring system (WPSS).4 In recent years a large armamentarium for treatment of MDS has been developed, ranging from best supportive care to new agents like hypomethylating or chelation therapy or hematopoietic stem-cell transplantation. As MDS is an illness primarily observed in the elderly, valid prediction of clinical outcome is vital for individualized therapeutic decisions. Due to the inhomogeneous character of the disease and the high median patient age, constant efforts have been made in recent years to improve and refine scoring systems. Risk scoring in MDS is so far based on evaluation of parameters representing the biology of the disease like blast counts, cytogenetics, or cytopenias.6 However, patient-associated parameters like age7 or performance status8,9 have been established as prognostic parameters in MDS. Moreover, while the impact of distinct comorbidities or possible cumulative effects on the survival of patients with MDS have been described by several groups,10–17 the implications were discussed controversially as different scoring systems, namely the Charlson Comorbidity Index (CCI),11–14 hematopoietic cell transplantation-comorbidity index (HCT-CI),12–14,17 Adult Comorbidity Evaluation-27 (ACE-27)16 and MDS-Specific Comorbidity Index (MDS-CI),16 were applied and multivariate analyses incorporating the IPSS were not included in all studies.11 Moreover, the distribution of comorbidities among distinct age and sex groups has not yet been analyzed. The aim of this study was thus to analyze the spectrum and the age- and sex-specific prevalence as well the clustering and clinical impact of comorbidities in a well-defined cohort of 616 patients with MDS from the Austrian MDS platform to establish the basis for individualized treatment planning.
1.1. Patients and Methods 1.1.1. Patient Characteristics Data on 616 patients with MDS from four centers of the Austrian MDS platform (Vienna Medical University, Hanusch
Krankenhaus Vienna, Innsbruck Medical University, Hospital of the Elisabethinen Linz, Austria) were analyzed retrospectively. To assess comorbidities, medical records (or charts) were retrospectively reviewed by a single trained observer. Both de-novo and therapy-related MDS, diagnosed between 1985 and 2007, were included. Informed consent and ethics committee approval were obtained according to local regulations.
1.1.2. Clinical Parameters and Scoring of Comorbidities The following parameters recorded in the dataset were considered in the analysis: age, sex, FAB subtype, IPSS and HCT-CI scores. Comorbidities were classified retrospectively by means of the HCT-CI score18 as described by Sperr et al.13 The HCT-CI scoring system consists of a list of 17 groups of diseases with assigned weight.18 The score assigns 1 point to arrhythmia, cardiovascular, inflammatory bowel, cerebrovascular, psychiatric and mild hepatic disease as well as diabetes, obesity or infection; 2 points to rheumatologic, renal and moderate pulmonary disease or peptic ulcer; 3 points to severe pulmonary, moderate to severe hepatic and valvular heart disease as well as prior solid tumor. The score results in a classification as low (0 points), intermediate (1 to 2 points), and high (≥ 3 points) risk.
1.1.3. Statistics Analysis was performed using SPSS software v. 20 by IBM. Statistical analyses applied included descriptive statistics, chi-square analysis, non-parametric tests, and multivariate regression models. Prevalence of particular comorbidities among various patient subgroups was analyzed using the chi-square, Fisher's exact and Mann–Whitney U tests. Overall survival (OS) was defined as the time from diagnosis to death or last follow-up. Leukemia-free survival (LFS) was defined as time from diagnosis to date of transformation to sAML. Patients lost for follow-up were censored at the date of last visit or start of therapy. Survival analysis was performed using Kaplan–Meier plots and compared by log rank test. In order to assess the joint effect of several potential predictors (in particular, age, sex, IPSS, HCT-CI) on OS or LFS, we used Cox regression analysis with forward stepwise variable selection. In a further Cox regression analysis, interactions between clinical variables and age and sex were added to the model in order to test for differential effects of age and sex. Throughout the analysis, p values < 0.05 were considered significant and p values from 0.05 to 0.1 were deemed a trend.
2. Results The patient cohort consisted of 616 individuals including 286 (46.4%) females and 330 (53.6%) males (f/m: 1/1.15). The overall median age was 71 (range: 24–99, mean: 69.3) years; the median age of female patients was 72 (range: 24–99, mean: 69.6) years; the median age of male patients was 70 (range: 25–91, mean: 69.1) years. Of the patients 99.5% were classified according to FAB criteria: refractory anemia (RA, n = 189, 30.7%), refractory anemia with ring sideroblasts (RARS, n = 133, 21.6%), refractory anemia with excess of blasts (RAEB, n = 152, 24.7%), RAEB in transformation (RAEB-T, n = 81, 13.1%), and chronic
Please cite this article as: Bammer C., et al, Clustering of comorbidities is related to age and sex and impacts clinical outcome in myelodysplastic syndromes, J Geriatr Oncol (2014), http://dx.doi.org/10.1016/j.jgo.2014.02.002
3
J O U RN A L OF GE RI A T RI C O NC O L O G Y XX ( 20 1 4 ) XX X–XX X
Table 1 – Prevalence of comorbidities in patients with MDS. Sex
Cardiac disease CHD, CI, MI Arrhythmia Valvular heart disease Cerebrovascular disease Cardiovascular disease Pulmonary disease Moderate Severe Hepatic disease Mild Moderate–severe Renal insufficiency Prior tumor Diabetes Rheumatologic disease Peptic ulcer Inflammatory bowel Obesity Infection Psychiatric disease
Total
Female
Male
n = 616 (%)
n = 286 (%)
n = 330 (%)
155 113 57 21 44 175 31 28 3 19 13 6 11 61 75 12 36 2 10 9 14
(25.2) (18.3) (9.3) (3.4) (7.1) (28.4) (5.0) (4.6) (0.5) (3.1) (2.1) (1.0) (1.8) (9.9) (12.2) (2.0) (5.8) (0.3) (1.6) (1.5) (2.3)
60 38 23 9 17 68 14 11 3 4 3 1 2 23 29 8 14 0 6 7 6
(21.0) (13.3) (8.0) (3.2) (5.9) (23.8) (4.9) (3.9) (1.1) (1.4) (1.1) (0.4) (0.7) (8.0) (10.4) (2.8) (4.9) (0) (2.1) (2.5) (2.1)
95 75 34 12 27 107 17 17 0 15 10 5 9 38 46 4 22 2 4 2 8
(28.8) (22.7) (10.3) (3.6) (8.2) (32.4) (5.2) (5.2) (0) (4.6) (3.0) (1.5) (2.7) (11.5) (13.9) (1.2) (6.7) (0.6) (1.2) (0.6) (2.4)
p-Value 0.037 0.009 n. s. n. s. n. s. 0.029 n. s. n. s. 0.062 0.073 n. s. n. s. n. s. n. s. n. s. n. s. n. s. n. s. n. s. n. s. n. s.
Classification of comorbidities according to HCT-CI score [10]. Significant p values (p < 0.05) are shown in bold; a trend towards significance (p = 0.05–0.1) is indicated in bold and italic. Abbreviations: CHD, coronary heart disease; CI, cardiac insufficiency; MI, myocardial infarction, n. s., not significant.
myelomonocytic leukemia (CMML, n = 58, 9.4%). Three (0.5%) cases were not able to be classified beyond doubt.
2.1. Prevalence of Comorbidities Half (297, 48.2%) of the patients displayed one or more comorbidities as defined by HCT-CI. In detail, 320 (51.8%) patients had a score of 0, 151 (24.5%) cases were found to have a score of 1, and 146 (23.7%) individuals showed a score of 2 or higher. The highest count found was 7. The mean HCT-CI score of all the patients was 1.26 (median 0.0). Of the patients 320 (52.0%) were assigned to the HCT-CI low-risk group (score 0), 168 (27.2%) to the intermediate-risk group (score 1 or 2) and 128 (20.8%) to the high-risk group (score ≥ 3), denoting a significantly higher prevalence of the low-risk group (p < 0.001). Cardiovascular diseases were the most frequent comorbidities (175, 28.4%). In detail, this group consisted of coronary heart disease/cardiac insufficiency/myocardial infarction (113, 18.3%), arrhythmia (57, 9.3%), cerebrovascular disease (44, 7.1%) and valvular heart disease (21, 3.4%). The second most frequent condition was diabetes (75, 12.2%). In a relevant proportion of patients prior tumors were observed (61, 9.9%) (Table 1).
2.2. Aspects of Sex in Comorbidities Clustering The mean number of comorbidities was significantly higher in men than in women (f/m: 0.74/0.92, p = 0.030). Similarly, males were characterized by significantly higher mean HCT-CI scores than were females (f/m: 1.09/1.41, p = 0.016) (Table 1). Male patients displayed a significantly higher prevalence of cardiovascular disease than did females (f/m: 23.8%/32.4%, p = 0.029).
This difference was also evident in the subgroups with cardiac disease (f/m: 21.0%/28.8%, p = 0.037) and coronary heart disease/ cardiac insufficiency/myocardial infarction (f/m: 13.3%/22.7%, p = 0.009). A trend towards significance was observed for a higher prevalence of hepatic disease (f/m: 1.4%/4.6%, p = 0.073) in males. Severe pulmonary disorders were more often observed in females (f/m: 1.1%/0%, p = 0.062); however, due to small case numbers in this comorbidity subgroup reasonable conclusions can hardly be drawn.
2.3. Impact of Age on Comorbidities in Patients with MDS Elderly persons (> 65 years) had a significantly higher mean HCT-CI score (1.52) than did younger patients (< 45 years and 46–65 years: 0.24 and 0.76, respectively) (p < 0.001). Several comorbidities were characterized by a higher prevalence in the elderly patients: cardiovascular disease (p < 0.001), prior tumor (p = 0.006), and diabetes (p = 0.019). Cardiac disease and prior tumor each showed a continuously growing increase with age (the majority of comorbidities were characterized by a continuous increase in prevalence). On the other hand, other comorbidities such as diabetes mellitus or cerebrovascular disease no longer show at advanced age (76 + years) an increase in prevalence as compared to younger patients (66–75 years) (Table 2).
2.4. Comorbidities Impact Clinical Outcome Overall survival (OS) in the cohort analyzed was 2.9 ± 0.41 years (median ± SD). The IPSS score was a very significant predictor of outcome, revealing a mean OS of 7.42 years in the IPSS low-risk
Please cite this article as: Bammer C., et al, Clustering of comorbidities is related to age and sex and impacts clinical outcome in myelodysplastic syndromes, J Geriatr Oncol (2014), http://dx.doi.org/10.1016/j.jgo.2014.02.002
4
J O U RN A L OF GE RI A TR I C O NC OLO G Y XX ( 20 1 4 ) XX X–XXX
Table 2 – Clustering of comorbidities in different age groups in patients with MDS (n = 616). Age groups
Cardiac disease CHD, CI, MI Arrhythmia Valvular heart disease Cerebrovascular disease Cardiovascular disease Pulmonary disease Moderate Severe Hepatic disease Mild Moderate–severe Renal insufficiency Prior tumor Diabetes Rheumatologic disease Peptic ulcer Inflammatory bowel Obesity Infection Psychiatric disease
Total
≤45 years
46–65 years
66–75 years
≥76 years
(n = 616)
(n = 29)
(n = 160)
n = (225)
n = (202)
n (%)
n (%)
n (%)
n (%)
n (%)
155 113 57 21 44 175 31 28 3 19 13 6 11 61 75 12 36 2 10 9 14
(25.2) (18.3) (9.3) (3.4) (7.1) (28.4) (5.0) (4.6) (0.5) (3.1) (2.1) (1.0) (1.8) (9.9) (12.2) (2.0) (5.8) (0.3) (1.6) (1.5) (2.3)
0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 1 0 0 0 0 0
(0) (0) (0) (0) (0) (0) (3.4) (3.4) (0) (0) (0) (0) (0) (3.4) (0) (3.4) (0) (0) (0) (0) (0)
16 10 6 0 7 22 8 8 0 7 6 1 0 8 12 5 10 1 5 2 3
(10.0) (6.3) (3.8) (0) (4.4) (13.8) (5.0) (5.0) (0) (4.4) (3.8) (0.6) (0) (5.0) (7.5) (3.1) (6.3) (0.6) (3.1) (1.3) (1.9)
62 47 21 9 18 72 13 11 2 5 2 3 6 21 33 3 15 1 3 5 5
(27.6) (20.9) (9.3) (4.5) (8.0) (32.0) (5.8) (4.9) (0.9) (2.2) (0.9) (1.3) (2.7) (9.3) (14,7) (1.3) (6.7) (0.4) (1.3) (2.2) (2.2)
77 (38.1) 56 (27.7) 30 (14.9) 12 (5.9) 19 (9.4) 81 (40.1) 9 (4.5) 8 (4.0) 1 (0.5) 7 (3.5) 5 (2.5) 2 (1.0) 5 (2.5) 31 (15.3) 30 (14.9) 3 (1.5) 11 (5.4) 0 (0) 2 (1.0) 2 (1.0) 6 (3.0)
p-Value
Available online at www.sciencedirect.com
ScienceDirect
Clustering of comorbidities is related to age and sex and impacts clinical outcome in myelodysplastic syndromes C. Bammera , W.R. Sperrb , G. Kemmlerc , F. Wimazald , T. Nösslinger e , A. Schönmetzlere , O. Krieger f , M. Pfeilstöcker e,g , P. Valent e,g , R. Stauder a,⁎ a
Department of Internal Medicine V (Hematology and Oncology), Innsbruck Medical University, Innsbruck, Austria Department of Internal Medicine I, Division of Hematology and Hemostaseology, Vienna Medical University, Vienna, Austria c Department of Psychiatry and Psychotherapy, Innsbruck Medical University, Innsbruck, Austria d Department of Obstetrics and Gynecology, Vienna Medical University, Austria, e Third Medical Department for Hematology and Oncology, Hanusch Hospital, Vienna, Austria f Department of Internal Medicine I, Hospital of the Elisabethinen, Linz, Austria g Ludwig Boltzmann Cluster Oncology, Vienna, Austria b
AR TIC LE I N FO
ABS TR ACT
Article history:
Objectives: Myelodysplastic syndromes (MDS) are typical diseases of the elderly. The clinical
Received 19 September 2013
outcome of a well-characterized cohort of patients with MDS was analyzed for prevalence
Received in revised
and impact of comorbidities to establish the basis for tailored treatment algorithms. Focus
form 30 December 2013
was on age- and sex-related differences.
Accepted 14 February 2014
Material and Methods: The hematopoietic cell transplantation-comorbidity index (HCT-CI) was assessed in 616 well-defined patients from the Austrian MDS platform (median age: 71 years).
Keywords:
Results: Most patients displayed one (24.5%) or more (23.7%) comorbidities. The highest
Age
frequencies were observed for cardiovascular disease (28.4%), diabetes (12.2%), and prior
Comorbidities
tumors (9.9%). Comorbidities were more frequent (mean number: 0.92 vs. 0.74 [male vs.
MDS
female]; p = 0.030) and more severe in men than in women (mean HCT-CI score: 1.41 vs.
Myelodysplastic syndromes
1.09 [male vs. female]; p = 0.016). Elderly patients (65 + years) showed a higher prevalence
Prognosis
of comorbidities than younger patients (HCT-CI score: 1.52, mean in 65 +, vs. 0.24 and 0.76
Sex
in < 45 years and 46–65 years, respectively) (p < 0.001). These differences were most
Survival
pronounced for cardiovascular disease, diabetes, and prior tumors (p < 0.001). Presence of
Assessment
cardiac arrhythmia or prior solid tumor was significantly associated with shorter overall survival (p = 0.023, 0.024, respectively). Moreover, HCT-CI risk grouping remained an independent prognostic parameter for survival in multivariate analysis. Conclusions: Comorbidities impact clinical outcome in elderly patients with MDS. Distinct diseases cluster in an age- and sex-related manner, which may have clinical implications when designing individualized therapies. Comorbidities should be evaluated with established scores and integrated in decision making. © 2014 Published by Elsevier Ltd.
⁎ Corresponding author at: Department of Internal Medicine V (Hematology and Oncology), Innsbruck Medical University, Anichstrasse 35, 6020 Innsbruck, Austria. Tel.: + 43 512 504 23255. E-mail address: [email protected] (R. Stauder).
http://dx.doi.org/10.1016/j.jgo.2014.02.002 1879-4068/© 2014 Published by Elsevier Ltd.
Please cite this article as: Bammer C., et al, Clustering of comorbidities is related to age and sex and impacts clinical outcome in myelodysplastic syndromes, J Geriatr Oncol (2014), http://dx.doi.org/10.1016/j.jgo.2014.02.002
2
J O U RN A L OF GE RI A TR I C O NC OLO G Y XX ( 20 1 4 ) XX X–XXX
1. Introduction Myelodysplastic syndromes (MDS) encompass a broad spectrum of clonal hematological disorders characterized by dysplastic changes and ineffective production of one or more hematopoietic cell lineages, resulting in cytopenias and symptoms thereof.1–3 MDS are commonly classified according to French–American– British (FAB) Cooperative Group or World Health Organization (WHO) categories. These categories are typically based on blast cell counts, signs of dysplasia in peripheral blood or bone marrow cells, as well as evaluation of karyotype. Significant differences in survival among these groups, as well as varying likelihood of developing secondary acute myeloid leukemia (sAML) have been described.4 Highly individual courses of MDS, even within particular groups, and the need to individualize therapeutic decisions make prognostic scores a necessity. Actually, several scoring systems are applied in clinical practice. The gold standard in prognostication in MDS remains the International Prognostic Scoring System (IPSS),5 which was recently updated (IPSS-R). Based on the number and degree of cytopenias, blast counts, cytogenetic risk groups, and age the IPSS-R defines different risk groups for overall survival and leukemic evolution.6 The transfusion need of patients was integrated as a relevant parameter in the WHO classification-based prognostic scoring system (WPSS).4 In recent years a large armamentarium for treatment of MDS has been developed, ranging from best supportive care to new agents like hypomethylating or chelation therapy or hematopoietic stem-cell transplantation. As MDS is an illness primarily observed in the elderly, valid prediction of clinical outcome is vital for individualized therapeutic decisions. Due to the inhomogeneous character of the disease and the high median patient age, constant efforts have been made in recent years to improve and refine scoring systems. Risk scoring in MDS is so far based on evaluation of parameters representing the biology of the disease like blast counts, cytogenetics, or cytopenias.6 However, patient-associated parameters like age7 or performance status8,9 have been established as prognostic parameters in MDS. Moreover, while the impact of distinct comorbidities or possible cumulative effects on the survival of patients with MDS have been described by several groups,10–17 the implications were discussed controversially as different scoring systems, namely the Charlson Comorbidity Index (CCI),11–14 hematopoietic cell transplantation-comorbidity index (HCT-CI),12–14,17 Adult Comorbidity Evaluation-27 (ACE-27)16 and MDS-Specific Comorbidity Index (MDS-CI),16 were applied and multivariate analyses incorporating the IPSS were not included in all studies.11 Moreover, the distribution of comorbidities among distinct age and sex groups has not yet been analyzed. The aim of this study was thus to analyze the spectrum and the age- and sex-specific prevalence as well the clustering and clinical impact of comorbidities in a well-defined cohort of 616 patients with MDS from the Austrian MDS platform to establish the basis for individualized treatment planning.
1.1. Patients and Methods 1.1.1. Patient Characteristics Data on 616 patients with MDS from four centers of the Austrian MDS platform (Vienna Medical University, Hanusch
Krankenhaus Vienna, Innsbruck Medical University, Hospital of the Elisabethinen Linz, Austria) were analyzed retrospectively. To assess comorbidities, medical records (or charts) were retrospectively reviewed by a single trained observer. Both de-novo and therapy-related MDS, diagnosed between 1985 and 2007, were included. Informed consent and ethics committee approval were obtained according to local regulations.
1.1.2. Clinical Parameters and Scoring of Comorbidities The following parameters recorded in the dataset were considered in the analysis: age, sex, FAB subtype, IPSS and HCT-CI scores. Comorbidities were classified retrospectively by means of the HCT-CI score18 as described by Sperr et al.13 The HCT-CI scoring system consists of a list of 17 groups of diseases with assigned weight.18 The score assigns 1 point to arrhythmia, cardiovascular, inflammatory bowel, cerebrovascular, psychiatric and mild hepatic disease as well as diabetes, obesity or infection; 2 points to rheumatologic, renal and moderate pulmonary disease or peptic ulcer; 3 points to severe pulmonary, moderate to severe hepatic and valvular heart disease as well as prior solid tumor. The score results in a classification as low (0 points), intermediate (1 to 2 points), and high (≥ 3 points) risk.
1.1.3. Statistics Analysis was performed using SPSS software v. 20 by IBM. Statistical analyses applied included descriptive statistics, chi-square analysis, non-parametric tests, and multivariate regression models. Prevalence of particular comorbidities among various patient subgroups was analyzed using the chi-square, Fisher's exact and Mann–Whitney U tests. Overall survival (OS) was defined as the time from diagnosis to death or last follow-up. Leukemia-free survival (LFS) was defined as time from diagnosis to date of transformation to sAML. Patients lost for follow-up were censored at the date of last visit or start of therapy. Survival analysis was performed using Kaplan–Meier plots and compared by log rank test. In order to assess the joint effect of several potential predictors (in particular, age, sex, IPSS, HCT-CI) on OS or LFS, we used Cox regression analysis with forward stepwise variable selection. In a further Cox regression analysis, interactions between clinical variables and age and sex were added to the model in order to test for differential effects of age and sex. Throughout the analysis, p values < 0.05 were considered significant and p values from 0.05 to 0.1 were deemed a trend.
2. Results The patient cohort consisted of 616 individuals including 286 (46.4%) females and 330 (53.6%) males (f/m: 1/1.15). The overall median age was 71 (range: 24–99, mean: 69.3) years; the median age of female patients was 72 (range: 24–99, mean: 69.6) years; the median age of male patients was 70 (range: 25–91, mean: 69.1) years. Of the patients 99.5% were classified according to FAB criteria: refractory anemia (RA, n = 189, 30.7%), refractory anemia with ring sideroblasts (RARS, n = 133, 21.6%), refractory anemia with excess of blasts (RAEB, n = 152, 24.7%), RAEB in transformation (RAEB-T, n = 81, 13.1%), and chronic
Please cite this article as: Bammer C., et al, Clustering of comorbidities is related to age and sex and impacts clinical outcome in myelodysplastic syndromes, J Geriatr Oncol (2014), http://dx.doi.org/10.1016/j.jgo.2014.02.002
3
J O U RN A L OF GE RI A T RI C O NC O L O G Y XX ( 20 1 4 ) XX X–XX X
Table 1 – Prevalence of comorbidities in patients with MDS. Sex
Cardiac disease CHD, CI, MI Arrhythmia Valvular heart disease Cerebrovascular disease Cardiovascular disease Pulmonary disease Moderate Severe Hepatic disease Mild Moderate–severe Renal insufficiency Prior tumor Diabetes Rheumatologic disease Peptic ulcer Inflammatory bowel Obesity Infection Psychiatric disease
Total
Female
Male
n = 616 (%)
n = 286 (%)
n = 330 (%)
155 113 57 21 44 175 31 28 3 19 13 6 11 61 75 12 36 2 10 9 14
(25.2) (18.3) (9.3) (3.4) (7.1) (28.4) (5.0) (4.6) (0.5) (3.1) (2.1) (1.0) (1.8) (9.9) (12.2) (2.0) (5.8) (0.3) (1.6) (1.5) (2.3)
60 38 23 9 17 68 14 11 3 4 3 1 2 23 29 8 14 0 6 7 6
(21.0) (13.3) (8.0) (3.2) (5.9) (23.8) (4.9) (3.9) (1.1) (1.4) (1.1) (0.4) (0.7) (8.0) (10.4) (2.8) (4.9) (0) (2.1) (2.5) (2.1)
95 75 34 12 27 107 17 17 0 15 10 5 9 38 46 4 22 2 4 2 8
(28.8) (22.7) (10.3) (3.6) (8.2) (32.4) (5.2) (5.2) (0) (4.6) (3.0) (1.5) (2.7) (11.5) (13.9) (1.2) (6.7) (0.6) (1.2) (0.6) (2.4)
p-Value 0.037 0.009 n. s. n. s. n. s. 0.029 n. s. n. s. 0.062 0.073 n. s. n. s. n. s. n. s. n. s. n. s. n. s. n. s. n. s. n. s. n. s.
Classification of comorbidities according to HCT-CI score [10]. Significant p values (p < 0.05) are shown in bold; a trend towards significance (p = 0.05–0.1) is indicated in bold and italic. Abbreviations: CHD, coronary heart disease; CI, cardiac insufficiency; MI, myocardial infarction, n. s., not significant.
myelomonocytic leukemia (CMML, n = 58, 9.4%). Three (0.5%) cases were not able to be classified beyond doubt.
2.1. Prevalence of Comorbidities Half (297, 48.2%) of the patients displayed one or more comorbidities as defined by HCT-CI. In detail, 320 (51.8%) patients had a score of 0, 151 (24.5%) cases were found to have a score of 1, and 146 (23.7%) individuals showed a score of 2 or higher. The highest count found was 7. The mean HCT-CI score of all the patients was 1.26 (median 0.0). Of the patients 320 (52.0%) were assigned to the HCT-CI low-risk group (score 0), 168 (27.2%) to the intermediate-risk group (score 1 or 2) and 128 (20.8%) to the high-risk group (score ≥ 3), denoting a significantly higher prevalence of the low-risk group (p < 0.001). Cardiovascular diseases were the most frequent comorbidities (175, 28.4%). In detail, this group consisted of coronary heart disease/cardiac insufficiency/myocardial infarction (113, 18.3%), arrhythmia (57, 9.3%), cerebrovascular disease (44, 7.1%) and valvular heart disease (21, 3.4%). The second most frequent condition was diabetes (75, 12.2%). In a relevant proportion of patients prior tumors were observed (61, 9.9%) (Table 1).
2.2. Aspects of Sex in Comorbidities Clustering The mean number of comorbidities was significantly higher in men than in women (f/m: 0.74/0.92, p = 0.030). Similarly, males were characterized by significantly higher mean HCT-CI scores than were females (f/m: 1.09/1.41, p = 0.016) (Table 1). Male patients displayed a significantly higher prevalence of cardiovascular disease than did females (f/m: 23.8%/32.4%, p = 0.029).
This difference was also evident in the subgroups with cardiac disease (f/m: 21.0%/28.8%, p = 0.037) and coronary heart disease/ cardiac insufficiency/myocardial infarction (f/m: 13.3%/22.7%, p = 0.009). A trend towards significance was observed for a higher prevalence of hepatic disease (f/m: 1.4%/4.6%, p = 0.073) in males. Severe pulmonary disorders were more often observed in females (f/m: 1.1%/0%, p = 0.062); however, due to small case numbers in this comorbidity subgroup reasonable conclusions can hardly be drawn.
2.3. Impact of Age on Comorbidities in Patients with MDS Elderly persons (> 65 years) had a significantly higher mean HCT-CI score (1.52) than did younger patients (< 45 years and 46–65 years: 0.24 and 0.76, respectively) (p < 0.001). Several comorbidities were characterized by a higher prevalence in the elderly patients: cardiovascular disease (p < 0.001), prior tumor (p = 0.006), and diabetes (p = 0.019). Cardiac disease and prior tumor each showed a continuously growing increase with age (the majority of comorbidities were characterized by a continuous increase in prevalence). On the other hand, other comorbidities such as diabetes mellitus or cerebrovascular disease no longer show at advanced age (76 + years) an increase in prevalence as compared to younger patients (66–75 years) (Table 2).
2.4. Comorbidities Impact Clinical Outcome Overall survival (OS) in the cohort analyzed was 2.9 ± 0.41 years (median ± SD). The IPSS score was a very significant predictor of outcome, revealing a mean OS of 7.42 years in the IPSS low-risk
Please cite this article as: Bammer C., et al, Clustering of comorbidities is related to age and sex and impacts clinical outcome in myelodysplastic syndromes, J Geriatr Oncol (2014), http://dx.doi.org/10.1016/j.jgo.2014.02.002
4
J O U RN A L OF GE RI A TR I C O NC OLO G Y XX ( 20 1 4 ) XX X–XXX
Table 2 – Clustering of comorbidities in different age groups in patients with MDS (n = 616). Age groups
Cardiac disease CHD, CI, MI Arrhythmia Valvular heart disease Cerebrovascular disease Cardiovascular disease Pulmonary disease Moderate Severe Hepatic disease Mild Moderate–severe Renal insufficiency Prior tumor Diabetes Rheumatologic disease Peptic ulcer Inflammatory bowel Obesity Infection Psychiatric disease
Total
≤45 years
46–65 years
66–75 years
≥76 years
(n = 616)
(n = 29)
(n = 160)
n = (225)
n = (202)
n (%)
n (%)
n (%)
n (%)
n (%)
155 113 57 21 44 175 31 28 3 19 13 6 11 61 75 12 36 2 10 9 14
(25.2) (18.3) (9.3) (3.4) (7.1) (28.4) (5.0) (4.6) (0.5) (3.1) (2.1) (1.0) (1.8) (9.9) (12.2) (2.0) (5.8) (0.3) (1.6) (1.5) (2.3)
0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 1 0 0 0 0 0
(0) (0) (0) (0) (0) (0) (3.4) (3.4) (0) (0) (0) (0) (0) (3.4) (0) (3.4) (0) (0) (0) (0) (0)
16 10 6 0 7 22 8 8 0 7 6 1 0 8 12 5 10 1 5 2 3
(10.0) (6.3) (3.8) (0) (4.4) (13.8) (5.0) (5.0) (0) (4.4) (3.8) (0.6) (0) (5.0) (7.5) (3.1) (6.3) (0.6) (3.1) (1.3) (1.9)
62 47 21 9 18 72 13 11 2 5 2 3 6 21 33 3 15 1 3 5 5
(27.6) (20.9) (9.3) (4.5) (8.0) (32.0) (5.8) (4.9) (0.9) (2.2) (0.9) (1.3) (2.7) (9.3) (14,7) (1.3) (6.7) (0.4) (1.3) (2.2) (2.2)
77 (38.1) 56 (27.7) 30 (14.9) 12 (5.9) 19 (9.4) 81 (40.1) 9 (4.5) 8 (4.0) 1 (0.5) 7 (3.5) 5 (2.5) 2 (1.0) 5 (2.5) 31 (15.3) 30 (14.9) 3 (1.5) 11 (5.4) 0 (0) 2 (1.0) 2 (1.0) 6 (3.0)
p-Value
