European Journal of Haematology 95 (461–466)
ORIGINAL ARTICLE
Predictors of survival and cause of death in patients with essential thrombocythemia Danijela Lekovic1, Mirjana Gotic1,2, Dijana Sefer1, Olivera Mitrovic-Ajtic3, Vladan Cokic3, Natasa Milic2,4 1
Clinic for Hematology, Clinical Center of Serbia, Belgrade; 2Medical Faculty, University of Belgrade, Belgrade; 3Institute for Medical Research, University of Belgrade, Belgrade; 4Institute for Medical Statistics, University of Belgrade, Belgrade, Serbia
Abstract Objectives: Standard risk stratification for overall survival (OS) in patients with essential thrombocythemia (ET) is based on advanced age and history of thrombotic events. Recently, International Prognostic Score for ET (IPSET) incorporated also leukocytosis in prognostic model. The aim of this study was to establish additional risk factors for OS in ET patients. Methods: After the median follow-up of 7 yr, in 244 consecutive ET patients, 32 deaths were documented (13.2%). The 5- and 10-yr OS was 95.9% and 79.7%, respectively. Considered additional risk factors at diagnosis of ET were the presence of arterial hypertension, diabetes, hyperlipidemia, and smoking attitude. Results: The main cause of death in 75% of patients was cardiovascular (CV) comorbidity. Patients with CV risk factors had increased risk of death (HR = 2.33). Cox regression model identified age, leukocytosis, presence of CV risk factors, and previous thrombosis as unfavorable predictors of survival. Based on these parameters, four risk groups were defined, with significantly different survivals (P < 0.001). Improved prognostic model displayed a better hazard ratio profile compared to the standard risk stratification and IPSET. Conclusion: The addition of CV risk factors allows better prognostic assessment by delineating the intermediate-risk category and improved identification of the high-risk patients. Key words myeloproliferative neoplasms; essential thrombocythemia; cardiovascular risk factors; prognosis; survival Correspondence Danijela Lekovic, MD, MSc, Clinic for Hematology, Clinical Center of Serbia, Koste Todorovica 2, 11000 Belgrade, Serbia. Tel: +381 62 217 333, +381 11 366 3718; Fax: +381 11 366 3718; e-mail: [email protected] Accepted for publication 7 December 2014
Essential thrombocythemia (ET) is a clonal stem cell disorder characterized by persistent thrombocytosis (1). The new classification of World Health Organization (WHO) in 2008 enabled morphological distinguishing true ET from primary myelofibrosis (PMF) in prefibrotic phase, entity which also might be presented with thrombocytosis (2). This separation had a clinical impact with subsequent improved survival and lower incidence of disease complications in morphologically accurate ET (3). The life expectancy of these patients is almost normal at least in the first 10 yr from diagnosis, and the risk of leukemic and fibrotic progression is 140 mmHg and/or diastolic blood pressure >90 mmHg (9). Hyperlipidemia was designated as elevated concentrations either any or all of the lipids in blood, including total cholesterol >4.9 mmol/L, low-density lipoprotein (LDL) cholesterol >3.0 mmol/L, and triglycerides >1.7 mmol/L (9). Diabetes was determinated as fasting plasma glucose ≥7.0 mmol/L (≥126 mg/dL) or 2-h postload plasma glucose ≥11.1 mmol/L (≥200 mg/dL) after a 75-g oral glucose tolerance test (OGTT) (10). Study included current smokers defined as adults who had smoked 100 cigarettes in their lifetime and currently smoke cigarettes everyday (daily) or some days (non-daily) at the time of ET diagnosis (11). All variables were available in 98% of patients except JAK2 mutational status. Statistical analysis
Data were expressed as mean values with standard deviations (for normally distributed data) or as medians with interquartile ranges (for skewed data). Categorical data were presented by absolute numbers with percentages. Overall survival was calculated from the date of diagnosis to death due to any cause. Survival curves were estimated using the Kaplan–Meier method, while differences between curves were evaluated with the Log rank test. Cox proportional
462
Lekovic et al.
hazards regression was adopted to carry out univariate and multivariate survival analyses and to determine Hazard ratios (HRs) with 95% confidence intervals. The ROC analysis was applied to set the best cutoff level for leukocyte count to predict survival. A multivariate Cox proportional hazards regression was used to assess the most important risk factors for prediction of OS in patients with ET. In all tests, P value 60 and previous thrombosis discriminated two patient groups: low (106 patients) and high risk (138 patients) with median OS not reached in both groups (log rank = 23.262, P < 0.001) (Fig. 1). Evaluation of IPSET prediction model on our group of patients was as follows: 82 patients in low, 113 patients
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Lekovic et al.
Predictors of survival and cause of death in ET
Table 1 Difference between the group of alive and group of dead patients according to clinical data Feature
Alive patients (%)
Dead patients (%)
P
Median age (yr)1 Age ≥ 60 yr Gender (M/F) ET symptoms1 Median hemoglobin level (g/L)1 Median leukocyte count (9109/L)1 Leukocyte count ≥ 10 9 109/L Median platelet count (9109/L)1 Platelet count ≥ 1000 9 109/L Median lactate dehydrogenase level (U/L)1 Splenomegaly Cardiovascular (CV) risk factors2 CV risk factors + smoking Previous thrombosis ≥2 previous thrombosis Arterial thrombosis during follow-up
57 85/212 62/148 104/208 139 8.7 82/212 1020 90/212 399 21/212 131/212 44/212 33/212 6/212 8/212
71 28/32 15/17 18/32 136 11.9 22/32 1119 19/32 417 5/32 25/32 13/32 9/32 5/32 9/32
ORIGINAL ARTICLE
Predictors of survival and cause of death in patients with essential thrombocythemia Danijela Lekovic1, Mirjana Gotic1,2, Dijana Sefer1, Olivera Mitrovic-Ajtic3, Vladan Cokic3, Natasa Milic2,4 1
Clinic for Hematology, Clinical Center of Serbia, Belgrade; 2Medical Faculty, University of Belgrade, Belgrade; 3Institute for Medical Research, University of Belgrade, Belgrade; 4Institute for Medical Statistics, University of Belgrade, Belgrade, Serbia
Abstract Objectives: Standard risk stratification for overall survival (OS) in patients with essential thrombocythemia (ET) is based on advanced age and history of thrombotic events. Recently, International Prognostic Score for ET (IPSET) incorporated also leukocytosis in prognostic model. The aim of this study was to establish additional risk factors for OS in ET patients. Methods: After the median follow-up of 7 yr, in 244 consecutive ET patients, 32 deaths were documented (13.2%). The 5- and 10-yr OS was 95.9% and 79.7%, respectively. Considered additional risk factors at diagnosis of ET were the presence of arterial hypertension, diabetes, hyperlipidemia, and smoking attitude. Results: The main cause of death in 75% of patients was cardiovascular (CV) comorbidity. Patients with CV risk factors had increased risk of death (HR = 2.33). Cox regression model identified age, leukocytosis, presence of CV risk factors, and previous thrombosis as unfavorable predictors of survival. Based on these parameters, four risk groups were defined, with significantly different survivals (P < 0.001). Improved prognostic model displayed a better hazard ratio profile compared to the standard risk stratification and IPSET. Conclusion: The addition of CV risk factors allows better prognostic assessment by delineating the intermediate-risk category and improved identification of the high-risk patients. Key words myeloproliferative neoplasms; essential thrombocythemia; cardiovascular risk factors; prognosis; survival Correspondence Danijela Lekovic, MD, MSc, Clinic for Hematology, Clinical Center of Serbia, Koste Todorovica 2, 11000 Belgrade, Serbia. Tel: +381 62 217 333, +381 11 366 3718; Fax: +381 11 366 3718; e-mail: [email protected] Accepted for publication 7 December 2014
Essential thrombocythemia (ET) is a clonal stem cell disorder characterized by persistent thrombocytosis (1). The new classification of World Health Organization (WHO) in 2008 enabled morphological distinguishing true ET from primary myelofibrosis (PMF) in prefibrotic phase, entity which also might be presented with thrombocytosis (2). This separation had a clinical impact with subsequent improved survival and lower incidence of disease complications in morphologically accurate ET (3). The life expectancy of these patients is almost normal at least in the first 10 yr from diagnosis, and the risk of leukemic and fibrotic progression is 140 mmHg and/or diastolic blood pressure >90 mmHg (9). Hyperlipidemia was designated as elevated concentrations either any or all of the lipids in blood, including total cholesterol >4.9 mmol/L, low-density lipoprotein (LDL) cholesterol >3.0 mmol/L, and triglycerides >1.7 mmol/L (9). Diabetes was determinated as fasting plasma glucose ≥7.0 mmol/L (≥126 mg/dL) or 2-h postload plasma glucose ≥11.1 mmol/L (≥200 mg/dL) after a 75-g oral glucose tolerance test (OGTT) (10). Study included current smokers defined as adults who had smoked 100 cigarettes in their lifetime and currently smoke cigarettes everyday (daily) or some days (non-daily) at the time of ET diagnosis (11). All variables were available in 98% of patients except JAK2 mutational status. Statistical analysis
Data were expressed as mean values with standard deviations (for normally distributed data) or as medians with interquartile ranges (for skewed data). Categorical data were presented by absolute numbers with percentages. Overall survival was calculated from the date of diagnosis to death due to any cause. Survival curves were estimated using the Kaplan–Meier method, while differences between curves were evaluated with the Log rank test. Cox proportional
462
Lekovic et al.
hazards regression was adopted to carry out univariate and multivariate survival analyses and to determine Hazard ratios (HRs) with 95% confidence intervals. The ROC analysis was applied to set the best cutoff level for leukocyte count to predict survival. A multivariate Cox proportional hazards regression was used to assess the most important risk factors for prediction of OS in patients with ET. In all tests, P value 60 and previous thrombosis discriminated two patient groups: low (106 patients) and high risk (138 patients) with median OS not reached in both groups (log rank = 23.262, P < 0.001) (Fig. 1). Evaluation of IPSET prediction model on our group of patients was as follows: 82 patients in low, 113 patients
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Lekovic et al.
Predictors of survival and cause of death in ET
Table 1 Difference between the group of alive and group of dead patients according to clinical data Feature
Alive patients (%)
Dead patients (%)
P
Median age (yr)1 Age ≥ 60 yr Gender (M/F) ET symptoms1 Median hemoglobin level (g/L)1 Median leukocyte count (9109/L)1 Leukocyte count ≥ 10 9 109/L Median platelet count (9109/L)1 Platelet count ≥ 1000 9 109/L Median lactate dehydrogenase level (U/L)1 Splenomegaly Cardiovascular (CV) risk factors2 CV risk factors + smoking Previous thrombosis ≥2 previous thrombosis Arterial thrombosis during follow-up
57 85/212 62/148 104/208 139 8.7 82/212 1020 90/212 399 21/212 131/212 44/212 33/212 6/212 8/212
71 28/32 15/17 18/32 136 11.9 22/32 1119 19/32 417 5/32 25/32 13/32 9/32 5/32 9/32
