Original Article |
1 Department of Medical Informatics, Biometrics, and Epidemiology; University of Munich;
2 Munich Cancer Registry of the Munich Comprehensive Cancer Center;
3 Department of Internal Medicine I, University of Dresden;
4 Department of Medical Informatics and Biomathematics, University of Münster;
5 Department of Internal Medicine A, University of Münster;
6 Department of Hematology/Oncology, University of Leipzig;
7 Department of Hematology, Hemostasis, and Oncology; Hannover Medical School;
8 Department of Hematology and Pediatric Oncology", Childrens Hospital, University of Münster and
9 Department of Internal Medicine III, University of Ulm, Germany
Correspondence: Dorle Messerer, Institut für Med. Informationsverarbeitung, Biometrie und Epidemiologie der Universität München (IBE); Marchioninistr. 15; D-81377 München, Germany. E-mail:dorle{at}messerer.info
|
|
|---|
Design and Methods: The EORTC Quality of Life Core Questionnaire (QLQ-C30), supplemented by information on self-assessed concomitant diseases, late treatment effects, and demographics was used. The questionnaire was returned by 419 of 818 patients (51.2%) identified by six study groups. The patients median age at diagnosis was 42 years, and the median follow-up period was 8 years. One hundred and seventy patients were treated with stem cell transplantation (121 allogenic, 49 autologous) in first complete remission; the other 249 patients were treated with conventional chemotherapy.
Results: The ECOG activity index revealed normal activity in 45% vs. 60% of the patients in the allogeneic stem cell transplantation vs. conventional chemotherapy groups, respectively and disabled person status in 60% vs. 35%. All QLQ-C30 functions, except physical functioning and pain, were poorer in allogeneic stem cell transplantation patients. Problems in leisure-time activities, social life, and financial management, sexual limitations and adverse effects were significantly more frequent in patients after allogeneic stem cell transplantation than after conventional chemotherapy. Multivariate logistic regression models on global health status revealed concomitant disease, age >45 years, and allogeneic stem cell transplantation as significant risk factors.
Conclusions: These results indicate that, compared to conventional chemotherapy, allogeneic stem cell transplantation has a significantly worse long-term impact on quality of life. This needs to be considered when treatment options are discussed.
Key words: long-term quality of life, allogeneic stem cell transplantation, acute myeloid leukemia.
|
|
|---|
Assessment of quality of life (QOL) and late effects after completion of anti-leukemic treatment has increasingly become a focus of research.4–17 Measurement of QOL shortly after diagnosis and during the course of conventional chemotherapy (CCT) showed that QOL improved steadily over time, paralleling the normalization of bone marrow function.11 After completion of treatment, physical, psychological, and emotional well-being appeared to recover to almost normal levels.15 However, the different post-remission strategies, CCT, autologous SCT and allogeneic SCT, seem to affect QOL differently.9,17 Significantly higher rates of somatic symptoms, repeated acute medical problems, physical impairment, role impairment, leisure restriction, and sexual impairment have been reported to occur during the first years after allogeneic SCT.9,16,17 The studies yielding this information were, however, limited by small sample size, heterogeneous cohorts, or an inadequately short time interval between completion of treatment and QOL assessment.
The German AML-Intergroup, therefore,18 initiated a survey of QOL in patients who were relapse-free for over 5 years after first-line treatment. The main objectives of this cross-sectional study were to assess QOL in a large cohort of AML patients late after completion of treatment and to compare the influence of different post-remission strategies on QOL.
|
|
|---|
Procedure and measures
Each long-term survivor received a self-administered questionnaire from the central study offices. The patients were asked to return the questionnaires either personally to the clinic or by mail.
The questionnaire consisted of three parts. In the first section, patients were asked about any disease, their employment situation before having leukemia and at the time of filiing in the questionnaire, and their current ECOG performance status.28 The questions about current symptoms were developed by the Munich Cancer Registry (MCR) of the Munich Comprehensive Cancer Center (MCCC) and had already been used in various surveys of breast cancer and rectal cancer patients.29–31 The second part of the questionnaire was the EORTC QLQ-C30, a validated QOL evaluation tool.32–34 This questionnaire contains 30 questions covering five functional scales (physical, emotional, cognitive, social, and role), a global QOL measure, and symptoms including pain, fatigue, diarrhea, and constipation. Patients responses were combined and converted to a score on a 0 to 100 scale according to the guidelines.35 On this scale, high functional scores indicate good function, and high symptom scores significant problems. However, the symptom scores were reversed, as recommended,29 so that high scores represent positive outcomes for all variables.
In the third part of the survey, patients were asked to indicate leukemia-specific late effects and whether there had been any changes in their lives since the onset of leukemia. The study co-ordinating centers provided data from the time of diagnosis on age, sex, lactate dehydrogenase level, white blood cell count, karyotype, date of treatment onset, and major treatment strategy.
Statistical analysis
Pearsons
2 tests were used to analyze treatment and age differences. Stratified Mantel-Haenszel tests were used to examine a possible confounding effect of patients age. The patients gender was not correlated with either age or treatment, so tests were not stratified for this variable. Non-parametric tests were employed for QOL variables. The main functioning and global QOL scores from the EORTC QLQ-C30 were dichotomized according to the median for logistic regression analyses assessing clinical and demographic predictors of QOL. QOL scores below the median were coded "0", and all others "1". Each median- dichotomized QOL score was entered separately as a dependent variable in the logistic regression analysis. The independent variables used in the analyses were selected from the literature and our own univariate analyses. The following variables were entered simultaneously as independent variables: AML study group cohort, age (<45/
45 years), sex, post-remission treatment (allogeneic SCT/CCT), and current concomitant disease (absent/present). An a priori error probability of 0.01 was assumed, therefore p values below 0.01 should be considered statistically significant.
|
|
|---|
|
View this table: [in a new window] [Download PPT slide] |
Table 1. Patients characteristics according to post-remission treatment.
|
Concomitant disease/symptoms
Of all responding patients, 88% indicated that they suffered from at least one concomitant disease/symptom. The most frequent disease/symptom was lower back pain (41%), followed by impaired vision (29%) and hypertension (27%). Table 2 gives the incidences and relative frequencies of concomitant problems according to the different post-remission therapies. Impaired vision, cataract surgery, and treatment of hormonal disorders were significantly more frequent in the allogeneic SCT group than in the CCT group. There was also a slightly high rate of chronic skin disorders in the allogeneic SCT group. The ECOG activity index revealed normal activity in 60% and 45% of the CCT group and allogeneic SCT group, respectively (Table 3). Higher age was significantly associated with a lower rate of employment, mainly due to retirement. There was no significant difference in employment when comparing patients under 50 years who had been treated with CCT (77%) or allogeneic SCT (68%). Disabled person cards were held by 60% of the patients in the allogeneic SCT group and by 35% of those treated with CCT.
|
View this table: [in a new window] [Download PPT slide] |
Table 2. Concomitant disease/symptoms.
|
|
View this table: [in a new window] [Download PPT slide] |
Table 3. ECOG performance, employment and disability.
|
Personal outcome
Long-term-survivors complained about leukemia-related problems (5–31% after CCT and 12–48% after allogeneic SCT) in various areas, with an emphasis on physical problems. Problems related to adverse effects, leisure activities, sexual limitations, and daily habits were significantly more frequent in patients after allogeneic SCT than in patients after CCT (Table 4). The general feeling/ perception of a positive attitude in life was not different between the two groups: 64% for CCT patients vs. 63% for allogeneic SCT patients.
|
View this table: [in a new window] [Download PPT slide] |
Table 4. Negative and positive personal outcome.
|
|
View this table: [in a new window] [Download PPT slide] |
Table 5. EORTC QLQ-C30 subscales.
|
|
|
|---|
The assessment of QOL and late effects after completion of anti-leukemic treatment have increasingly become a focus of research. Unfortunately, most previous studies had limitations. The study by Zittoun et al. was based on only 98 patients, of whom 35 had undergone allogeneic SCT, and the time of assessment was heterogeneous, ranging from 19 to 79 months.9 The study by Watson et al.16,17 was adequately powered being based on 481 patients of whom 98 had received an allogeneic transplant, but QOL assessment was performed only 1 year after completion of treatment. Taking into account that less than half of the patients had a recovery of peripheral lymphocyte counts during the first year,36 and that consequently the incidence of infectious complications was highest during the first years after allogeneic SCT, a QOL assessment 1 year after completion of consolidation therapy seems to be inappropriate for determining the impact of different treatment strategies on the long-term outcome.
The overall response rate to the questionnaires of 51% in our survey is moderate and we, therefore, examined the selection of patients. There was no identifiable selection bias, based on known prognostic factors in AML such as cytogenetics and white cell count. The karyotype risk profile of our long-term survivors (Table 1) is comparable to that published for younger AML patients at diagnosis. 26 Seriously ill patients may prefer not to answer questionnaires, or long-term survivors in excellent shape might no longer want to be confronted with their disease again.
In our study the response rate to the questionnaires correlated with the duration of follow-up since diagnosis. The response rate in the QOL study of patients in the MRC AML 10 trial was 78%, but the study was performed only 1 year after completion of consolidation treatment16,17 and, therefore, cannot be compared to our cross-sectional study with a median follow-up period of 8 years. Other cross-sectional studies addressing the issue of QOL in AML-patients did not explicitly state potential selection biases.4–10
There might be a selection bias towards allogeneic SCT in younger patients with unfavorable cytogenetics or residual disease. However, these differences between the two treatment groups were taken into account as far as concerns patients characteristics available for analysis. It is difficult to compare our results with those of previous studies because the external validity of these studies is rarely equivalent. In addition, there was considerable patient heterogeneity concerning diagnosis (chronic myeloid leukemia at least 10 years after SCT)37, age (mainly children),38 time of evaluation (at least 1 year after treatment)39,40 and sample size.4–12 Furthermore, a variety of different scales and self-made questionnaires were used.4,6,7,10 The cross-sectional QOL study of the MRC AML 10 trial16,17 used the EORTC QLQ-C30 score and further questions concerning sexual health and infertility in 479 patients, comparing the effects of allogeneic SCT, autologous SCT, and CCT, at least 1 year after consolidation treatment. The median age of the patients was lower than in our study (allogeneic SCT group 33 vs. 38 years; CCT 43 group vs. 47 years). That study found that significantly higher proportions of post-SCT patients than CCT patients suffered from mouth dryness and from worse sexual relationships, social relationships, professional activities, and leisure time activities. Hormonal disorders and infertility were also more common among the allogeneic SCT patients than among the CCT patients. Similar results were found in our study, except for mouth dryness which was not investigated.
Watson et al. did not publish scores but percentages of patients with problems in the EORTC QLQ-C30 functional and symptom scales and items, so the results of this study need to be converted for comparison to our results. The overall rates of our patients with problems were higher than in the MRC study: physical function 54% vs. 41%, role function 53% vs. 35%, cognitive function 59% vs. 53%, emotional function 84% vs. 76%, social function 61% vs. 56%, pain 49% vs. 34%, sleep disturbance 55% vs. 45%, and financial difficulties 53% vs. 46%. The results were nearly identical for global health, being 81% vs. 80%, and fatigue, 75% vs. 79%, in our study and in the MRC study, respectively. The adverse impact of allogeneic SCT on role, social function, fatigue, nausea/vomiting, and financial difficulties was evident in both study cohorts. A different effect on cognitive or emotional function was seen only in our analysis. The main difference between the two studies was the time of assessment, which was 1 year compared to a median of 8 years in the MRC-QOL study and our study, respectively. However, the results of both studies suggest that problems in allogeneic SCT patients occur early after transplant and seem to persist.
Fatigue, feeling tired, and lacking energy are the most common symptoms reported by cancer patients,41 but the exact causes are not known. There is a correlation between fatigue and depression, and even after adjusting for multiple factors, major depression was associated with mortality among cancer patients after SCT.42 The symptom of fatigue was rather severe in our study, especially in the allogeneic SCT group. The challenge of future trials will be to evaluate prophylactic interventions in patients at high risk.
Fifty percent of the long-term survivors lead an as active a life as they did before the onset of leukemia; 72% of the respondents under the age of 50 years are employed, and more than 60% consider that their attitude in life has become more positive since the diagnosis and treatment of their leukemia. Significant effects of study group or duration of follow-up since diagnosis on QOL could not be shown.
Karyotype and response to induction therapy are among the most important prognostic factors in adult AML.2 Initial results from studies with risk-adapted treatment strategies suggest that high-risk patients, in particular, benefit from allogeneic SCT; other risk groups show equivalent results.43–46 The main result of our cross-sectional study – the negative impact of allogeneic SCT on QOL in long-term survivors – must, therefore, be taken into account in those risk groups without a clear survival benefit from either post-remission strategy.
In conclusion, these results indicate that a considerable number of long-term survivors live without major, restricting problems. However, QOL was significantly reduced in patients after allogeneic SCT, and this must be taken into account when choosing the treatment for AML patients in whom the benefit from such an intensive strategy is not obvious.
DM: initiative and main author of the paper; JE: scientific co-worker in the same area; JH: methodological consultant; MS, GE, CS, TB, RK, JK, AG, UC, HD: contributor of data; AS: scientific consultant on score methodogy; DN: co-writer and contributor of data; RFS: main author of the paper. The authors reported no potential conflicts of interest.
Received for publication July 13, 2007. Revision received January 26, 2008. Accepted for publication January 30, 2008.
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
J. Pidala, C. Anasetti, and H. Jim Quality of life after allogeneic hematopoietic cell transplantation Blood, July 2, 2009; 114(1): 7 - 19. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Blum Post-remission therapy in acute myeloid leukemia: what should I do now? Haematologica, June 1, 2008; 93(6): 801 - 805. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||