Annals of Oncology Advance Access originally published online on December 3, 2007
Annals of Oncology 2008 19(3):570-576; doi:10.1093/annonc/mdm543
quality of life |
The impact of follicular lymphoma on health-related quality of life
1 Haematology and Oncology, St George's Hospital, London
2 Oxford Outcomes Ltd, Oxford
3 Mount Vernon Cancer Centre, Northwood, Middlesex
4 Department of Haematology, University Hospital of Wales, Heath Park, Cardiff, Wales
5 Belfast City Hospital
6 Western Park Hospital, Sheffield Hospital, Sheffield
7 Department of Haematology, Taunton and Somerset Hospital, Musgrove Park, Taunton
8 Haemato-Oncology, MRC Toxicology Unit / Leicester University, Leicester
9 Derriford Hospital, Plymouth
10 Healthcare Management, Roche Products Ltd, Welwyn Garden City, Hertfordshire, UK
* Correspondence to: Dr R. Pettengell, Department of Haematology, St George's University of London, Cranmer Terrace, London SW170RE, UK. Tel: +44-20-87255454; Fax: +44-20-87252859; E-mail: rpetteng{at}sghms.ac.uk
 |
Abstract |
|---|
 Top Abstract introductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Background: The purpose of this study was to determine whether there was a relationship between disease activity and health functioning, as measured by a range of patient-reported outcome (PRO) measures in patients with follicular lymphoma (FL).
Patients and methods: A total of 222 patients with FL were recruited from eight sites across the UK and they completed a number of PRO measures. The participants were analyzed across five disease states: ‘active disease—newly diagnosed’, ‘active disease—relapsed’, ‘partial response’, ‘complete response’ and ‘disease free’. The relationship between these disease states and their level of health functioning was assessed as well as the relationship between being ‘on’ or ‘off’ chemotherapy and disease state.
Results: In terms of health-related quality of life (HRQoL), participants in the relapsed category had the lowest mean physical well-being, emotional well-being, functional well-being and social well-being score. In a regression analysis, the ‘active disease–relapsed’ group acted as a significant predictor for each PRO variable. In addition, the remission group acted as a significant predictor of high anxiety scores as measured by the Hospital Anxiety and Depression Scale.
Conclusion: The results of this study demonstrate that various aspects of patient-reported health outcomes differ according to disease state in patients with FL. For those patients who have relapsed, they are more likely to experience worse HRQoL and other patient-reported health outcomes than patients newly diagnosed, in partial or complete remission or when completely disease free.
Key words: FACT-LYM, follicular lymphoma, HRQoL, PRO, relapse, remission
|
introduction |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Follicular lymphoma (FL) is the second most common form of non-Hodgkin's lymphomas (NHLs) comprising 35% of adult NHL in the United States and 22% worldwide [1]. In 2002, there were 9443 people diagnosed with NHL lymphoma in the UK. The incidence of NHL is equally common in men and women and increases with age with around two-thirds (69%) of all cases diagnosed in people aged >60 years [2–6].
At presentation, most patients with FL present with multiple sites of lymphadenopathy and/or bone marrow disease [advanced-stage disease (III/IV)]. This may manifest itself with disease-related symptoms such as fatigue, weight loss, fever and night sweats. Restricted movement, cosmetic disfigurement and pain may also occur due to enlarged lymph nodes. Treatment generally attempts to control rather than cure the disease and the natural history is typically characterized by multiple remissions and relapses [7]. Regardless of the treatment chosen, most patients eventually die of this disease. One large-scale systematic review found that the median life expectancy of a patient diagnosed with FL was between 8 and 12 years [7].
The aim of the treatment of FL is to maximize overall survival, maintain health-related quality of life (HRQoL) and minimize treatment-related morbidity. In FL, the treatment can sometimes have more of a negative impact than the condition itself. For this reason, it is particularly important to consider HRQoL and other patient-reported outcomes (PROs) in establishing the suitability of a treatment. Chemotherapy can result in nausea, vomiting, hair loss and cause skin irritation, sore mouth, dysphagia and gastrointestinal problems depending upon the site treated [8]. Webster and Cella [9] discussed the impact of treatment on the patient's HRQoL. They reported that although aggressive therapy may have little impact on overall survival, adopting a ‘watch and wait’ strategy may have adverse psychological effects. ‘Watch and wait’ or ‘watchful waiting’ is a term used by the medical profession to describe a period from diagnosis when the patient is living comfortably with few side-effects from the disease during which they are closely observed but not actively treated. This approach to management does not adversely affect outcome, however, for patients feeling that they are not actively fighting their disease and may increase the likelihood of negative psychological outcomes, including anxiety or depression. Individual differences among patients in areas such as personality, illness adaptation, preferred involvement in treatment decisions, patient information levels, social support and the relationship with their physician act as moderating factors in treatment choice [9].
Issues such as uncertainty (especially in relation to relapse), perceived lack of control, feelings of dependency, anxiety and depression are important in a recurrent cancer such as FL [10]. Patients may have a heightened concern about issues of death, dependence, disfigurement and disability [11]. For these reasons, it is important to consider the impact of FL and its treatment in areas such as functional ability, self-image, social activities, relationships and work.
In considering HRQoL or other aspects of patient-reported health outcomes, it is important to consider that patients may be willing to endure the adverse side-effects of a treatment if it results in a longer period of remission, a chance of cure or increased survival [12]. Thus, even though survival may carry with it a negative psychological impact (i.e. a fear of recurrence), many survivors of cancer typically accept life after cancer and cope and adjust effectively [13].
There is a paucity of research considering HRQoL and other aspects of patient-reported health outcomes in FL, or NHL more generally. Many factors including a lack of standardization regarding treatment choice, the relatively indolent nature of the disease and the adverse psychological impact of a chronic condition characterized by the uncertainty of remission and relapse emphasize the importance of considering these issues.
In the present paper, we examined three PRO measures that covered HRQoL, depression and anxiety and overall activity impairment in patients with FL. We investigated these PRO factors according to five disease states: newly diagnosed, relapsed, partial response to treatment, complete response to treatment and disease free; patients were also divided into those receiving chemotherapy and those who were not.
|
patients and methods |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
A total of 222 patients aged
18 years, with histologically confirmed FL and a performance status of zero to two according to the Eastern Cooperative Oncology Group (ECOG) criteria, participated in this study. Treatment of the FL patients was given according to best clinical practice, which may have included chemotherapy, radiotherapy, biologic therapy and watch and wait. Specific assessment of the treatment plan for patients was not the subject of this manuscript.
The mean age of participants was 60.4 years (range 36–85 years), 59.7% of whom were female and 47.7% were in paid employment. Table 1 presents the patient's clinical and demographic information by disease state, with each patient placed in one of the five categories according to the stage of their FL: ‘active disease—newly diagnosed’, ‘active disease—relapsed’, ‘partial response’, ‘complete response’ (or remission) (at first follow-up) and ‘disease free’ (no detectable disease beyond first follow-up). Table 1 reports the break down of those on and off chemotherapy across the five categories. Across all these categories, where patients were not receiving chemotherapy, 125 were on a watch and wait strategy (or no therapy). For those not on chemotherapy and in the active disease—newly diagnosed group, 24 patients were on a watch and wait strategy. Table 1 also reports the number of FL patients in each disease state at the various stages of their disease. No further analysis was undertaken by stage of disease as the focus of this paper was on HRQoL for the different disease states previously defined. It should be noted from Table 1 that there is a relatively high percentage of stage I/II in this study, but on average, one would expect 
15%–20% of patients to be at stage I. This study recruited 21%, which is at the higher end. Given the study design and the nature of recruitment (cross-sectional, with nurses ringing patients to see if they would like to participate), it is more than likely that patients will be alive to participate and have a better prognosis.
|
The study was approved by the Thames Valley Research Ethics Committee (COREC) and conducted in eight UK sites. All participants who completed the study provided written informed consent. Letters were sent to all FL patients who were due to come in for an outpatient appointment in the next week asking whether they would like to participate in the study and if so, informed consent was obtained. Once consent was obtained, the nurse completed a clinical evaluation and asked the patient to complete the various PRO measures described below.
clinical measure
The research nurse completed a clinical evaluation. This included questions on the patient's date of diagnosis, the stage of disease at diagnosis, their current ECOG status and chemotherapy status. This was completed using the patient's medical records.
demographic questions
Each patient completed a demographic measure; this included details on age, gender, ethnic origin and education level.
the function assessment of chronic illness therapy–lymphoma
This measure was used to assess health status. It consists of the Function Assessment of Chronic Illness Therapy–general (FACT-G) and a 15-item lymphoma-specific additional concerns subscale (LYM) [14]. The FACT-G consists of three 7-item subscales, physical well-being (PWB) (scored 0–28), social well-being (SWB) (scored 0–28) and functional well-being (FWB) (scored 0–28), and a six-item subscale emotional well-being (EWB) (scored 0–21). The Function Assessment of Chronic Illness Therapy–Lymphoma (FACT-LYM) total score is calculated by adding the score obtained on the FACT-G to the score obtained on the LYM subscale (total possible score is 168, which indicates the highest possible HRQoL). Provisional testing shows that the FACT-LYM has good internal consistency (
= 0.70) across five different languages [14]. Validation work has also been done recently by Webster et al. [15], confirming the reliability, validity and sensitivity of the FACT-LYM. Scoring of the FACT-LYM followed the FACIT scoring guidelines including the calculation of both the FACT-LYM total mean score and the FACT-LYM trial outcome index (TOI) score. The TOI score sums the PWB, FWB and the specific LYM subscale to give a total. It does not include the EWB or the SWB subscales. The higher the score, the better the HRQoL.
work productivity and activity impairment scale
The Work Productivity and Activity Impairment Scale (WPAI) is a six-item questionnaire, which elicits hours worked (for pay), hours missed due to the condition and hours missed for any other reasons [16]. Hours missed for ‘other reasons’ is not used in the scoring, but is included as a prompt to the respondent to exclude those hours from the count of actual hours worked. The WPAI yields four types of scores: ‘activity impairment’, ‘absenteeism’ (work time missed), ‘presenteeism’ (impairment at work/reduced on-the-job effectiveness) and ‘work productivity loss’ (overall work impairment/absenteeism plus presenteeism). Scores were transformed into percentages. Higher scores indicate greater impairment.
hospital anxiety and depression scale
The Hospital Anxiety and Depression Scale (HADS) was developed to detect anxiety and depression in a nonpsychiatric hospital setting [17, 18]. It is widely used to measure psychological morbidity in cancer patients. It is a 14-item scale with seven items measuring anxiety and seven items measuring depression. Each item is measured on a four-point scale (0–3) with a higher score indicating higher levels of anxiety or depression. For each scale, the total score ranges from 0 to 21. Normal anxiety is considered to be in the range of 0–7; mild anxiety, 8–10; moderate anxiety, 11–14; and severe anxiety, 15–21. Normal levels of depressive symptoms are considered to be in the range of 0–7; for mild depression, 8–10; moderate depression, 11–14; and severe depression, 15–21. These scales are intended for separate consideration and the calculation of a total score is not advised.
statistical methods
Group differences for each of the PRO measures on the basis of the five disease groupings were investigated using the Kruskal–Wallis H-test (or Mann–Whitney for on or off chemotherapy) to test whether several independent samples are from the same population. The Kruskal–Wallis H-test assumes that there is no a priori ordering of the people comprising the disease stage groupings from which the samples are drawn. The ability of the disease states to predict the various PRO variables outlined above was investigated using ordinary least-squares linear regression. Regression analysis allows us to test the hypothesis that there is a difference between the five disease states. The regression approach also estimates the proportion of the variance of the dependent variable (DV), which is accounted for by the independent variables and which indicates the strength of the proposed predictive model. The ‘active disease—newly diagnosed’ group acted as the reference (or constant) variable. A probability level of 0.05 was used to define statistical significance and the analyses were carried out using SPSS (V12) for windows.
|
results |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
The scores presented in Table 2 indicate that participants in the ‘relapsed’ category have the lowest mean score for each of the FACT-LYM domains, the TOI score and the total score. The Kruskal–Wallis test indicates that for all domains except SWB, there is a statistically significant difference between the five disease status groupings.
|
Table 2 also shows the mean HADS scores for each disease state. The majority of the participants in this study reported anxiety or depression levels within the normal range. A total of 81.7% of participants had anxiety scores in the mild to normal range and 94.5% had depression scores in the mild to normal range. Participants in the ‘disease-free’ category had the lowest mean anxiety and depression scores (3.78 ± 3.27 and 2.30 ± 2.45, respectively). Table 2 reports a statistically significant difference between the disease state and the mean total anxiety and depression scores on the HADS.
Just over 47% of participants were in paid employment. Those who were in ‘remission’ (15%) or who were ‘active disease—newly diagnosed’ (11.5%) represented the largest proportion of those in paid employment. Participants in both these disease states also represented the greatest proportion of those retired (13.7% and 11.5%, respectively). Participants were asked the degree to which health problems affected their ability to carry out their regular daily activities (regardless of the current employment status). Table 2 shows that participants in the ‘relapsed’ category had the highest percentage of activity impairment (46.2%), while participants in the disease-free category reported the lowest percentage (14.8%) of activity impairment. The Kruskal–Wallis test indicates a statistically significant difference in the level of activity impairment for participants across the five different disease states.
The Mann–Whitney U-test was used to assess whether total mean scores on the PRO measures outlined above were different if participants were currently receiving chemotherapy. Table 3 presents the results for each of the PRO measures. There were 44 participants currently receiving chemotherapy. From Table 3, it is reported that, other than the HADS anxiety subscale score, those participants receiving chemotherapy were reporting overall, statistically significant worse health functioning (P = 0.004), depressive symptoms (P = 0.005) and activity impairment (P = 0.009) than those participants who were not having chemotherapy.
|
An analysis was also conducted to assess whether there was a difference in HRQoL across the disease categories depending on the type of treatment being received. For those who were in the active disease—newly diagnosed group, there was no significant difference in the FACT-LYM total score irrespective of whether they were on watch and wait (24 of 51 patients), rituximab, other antibody therapy, radiotherapy, other medication or therapy (P = 0.256). (For the remaining categories, the number of patients on watch and wait was too small for analysis as the study was not primarily powered for this analysis.)
regression analysis
the five disease states
The Kruskal–Wallis tests indicate that differences exist across the five disease states for all FACT-LYM subscales (with the exception of SWB) and also for the total scores (i.e. the FACT-LYM total and the FACT-LYM TOI). Similarly, there was a significant difference for the HADS anxiety and depression subscales and the WPAI—overall activity impairment. These analyses, however, do not provide information on the statistical significance of differences within each of the five disease states. For this reason, it was decided to further examine these hypotheses that stated that there would be difference in FACT-LYM total scores and TOI scores, the HADS anxiety and depression subscale scores and the WPAI total scores (for overall activity impairment) for each disease state. Each of the disease states was made into a dummy variable. This model was entered to allow for the influence of the four variables (the fifth acted as a reference category). Table 4 reports the unstandardized beta values for each disease state across those HRQoL variables where the Kruskal–Wallis tests indicated a significant difference between the different disease states. For each PRO variable (the DV), the significant predictor in each of the models was the ‘active disease—relapsed’ group. In addition, for the HADS anxiety scale the ‘remission/full response’ group also acted as a significant predictor of HADS anxiety scores.
|
|
discussion |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
There is a statistically significant difference in the impact of disease state on HRQoL (FACT-LYM) for patients with FL. This was the case for the FACT-LYM and FACT-LYM TOI scores and each of the domains on the FACT-LYM, other than social functioning (i.e. for PWB, EWB, FWB and lymphoma additional concerns). The social functioning scale of the FACT-LYM addresses issues around family and friend support and remains fairly constant despite the patient's disease state. All remaining subscales or domains of the FACT-LYM address specific issues about their health, which has a direct impact on the patient's ability to function. Clearly, the disease state will impact on the level of their PWB, their ability to carry out various daily activities, the psychological impact of their condition and how intensely they experience the additional lymphoma-specific concerns.
Of particular interest in this study was the influence of disease state on the various PRO scores. To assess where these differences are located for the different disease states of patients, a regression analysis was used. Although the data themselves are nonnormal, it is common to assume a normal sampling distribution for the mean, relying on the ‘central limit theorem’, which states that the sampling distribution of the mean approaches normality with sufficient sample size. The advantages of using a regression framework are two-fold. First, the regression framework allows the examination of other potentially important factors that might affect the PRO scores in a multivariate context (i.e. allows the assessment of the importance of a given factor controlling for other potentially confounding factors). Secondly, if the assumptions underlying the regression hold, then the power of the estimation procedure is enhanced, with all observations contributing to a joint estimate of variance.
This demonstrated the significant impact on patient-reported health outcomes for a patient who had relapsed. The mean difference for the FACT-LYM total mean scores between the ‘active disease—newly diagnosed’ and the ‘active disease—relapsed’ groups was 26.34. This represents a statistically significant decrement in HRQoL for those who have relapsed. Interestingly, those in the newly diagnosed group had the highest total mean score for the FACT-LYM. There was, however, no significant difference in the newly diagnosed patients who were on watch and wait and those who were being actively treated. It could be hypothesized that this is because they have yet to experience the full impact of the symptoms of the disease, the treatment prescribed or a relapse after a period of successful treatment. The FACT-LYM total score includes the EWB subscale and it is possible that the high scores of the ‘active disease—newly diagnosed’ group reflect a degree of optimism (i.e. items in this subscale include ‘I feel sad’, ‘I am satisfied with how I am coping with my illness’, ‘I am losing hope in the fight against my illness’, ‘I worry about dying’, ‘I worry that my condition will get worse’, etc.) regarding the outcome of any treatment. Indeed, by comparison, the FACT-LYM TOI score (which does not include EWB) demonstrates that the ‘disease-free’ group had the highest level of health functioning and, significantly, the ‘relapsed’ group had the worst level of health functioning. Excluding scores from the EWB subscale perhaps removes a biased ‘early optimism’ of treatment effect. While the regression model using the FACT-LYM total score as the DV accounts for slightly more of the variance (11%) than if the FACT-LYM TOI (9%) were the DV, these models have not taken into account the potential of other predictors in the model affecting the overall significance on FACT-LYM or FACT-LYM TOI scores. This needs further exploration.
For the HADS scores, the ‘relapsed’ group is statistically significantly more likely to report higher levels of anxiety or depressive symptoms than any other disease state group, though it should be noted that for these two models (i.e. HADS anxiety and depression subscales), the increase of the HADS score for the ‘relapsed’ group is only predicting mild (to perhaps moderate) levels of anxiety or depression (see Table 2). The HADS anxiety subscale mean total score is also related to those participants in the ‘remission/complete response’ group, but who are not yet considered ‘disease free’. Previous studies have found that many survivors of cancer fear recurrence and death [19].
In Maher's exploratory study of recovered cancer patients, he found that there was a combination of both positive and negative elements associated with survival and anxiety about recurrence of the disease. The results of this study are consistent with Maher's finding and additionally would account for the fact that only anxiety (and not depressive symptoms) is a significant feature in the full remission group. Not surprisingly, the depression subscale of the HADS is statistically significant for the relapsed patients.
Finally, it is the ‘relapsed’ group that is significantly predictive of work productivity loss (i.e. overall work impairment). This regression model accounted for 8.8% of the variance in total WPAI scores. Consistent with the assessment of anxiety, depressive symptoms and HRQoL, work productivity is most affected by patients in the ‘relapsed’ group.
The results of the regression analysis, which place the ‘relapsed’ group as being the most sensitive to changes in HRQoL and health/psychological impairment, are consistent with the previous research. It is therefore arguable that if patients do enter the ‘disease-free’ period, then to maintain, or possibly increase, their health functioning, treatment should be aimed at keeping them in remission (or disease free) for as long as possible. Cella et al. [10] previously reported that people who had recently undergone a recurrence or relapse of their cancer demonstrated higher levels of adjustment difficulty when compared with a normative sample of cancer patients. In this study, the ‘relapsed’ group was experiencing a significant decline of their health compared with any other disease stage. In addition, those who were currently receiving chemotherapy were experiencing worse HRQoL, depressive symptoms and activity impairment than those who were not on chemotherapy. Again, this is consistent with the findings of Cella et al.
Where this study differs from Cella et al. [10] is, however, with respect to their conclusion that given the questionable survival benefit of most treatment options, and given the impact of chemotherapy on quality of life, less aggressive approaches are the best strategy. First, the study of Cella et al. considered chemotherapy only and available treatment options now include biological therapy and stem-cell transplant therapy. Secondly, advances in the treatment of FL mean that the chances of prolonged survival following treatment are increasing substantially. In recent research, rituximab in combination with anthracycline-based chemotherapy in patients with chemotherapy naive and relapsed refractory NHL has been shown to improve progression-free survival by 19%–40% with little additional toxicity [20–22]. Moreover, maintenance rituximab following chemotherapy for up to 2 years may also have a role in prolonging disease-free survival in patients with FL. Similarly, recent research has found that high-dose therapy and radioimmunoconjugate therapy significantly improve progression-free survival and overall survival in patients with FL [23–25].
Essentially, the prospects for prolonged survival for patients with lymphoma are now far greater and recommending lower toxicity chemotherapy regimens, which may lead to earlier relapses and the need for retreatment, may now not be the most suitable approach. Some of the items in the FACT-LYM (e.g. ‘Because of my illness, I have difficulty planning for the future’, ‘I get nervous about making decisions regarding treatment’) would indicate that patients with FL do worry about the potential of relapse and the treatment options that may be recommended. This study has demonstrated that keeping patients in remission for as long as possible and preventing subsequent relapse would assist in optimizing the HRQoL of patients with FL beyond the duration of the actual treatment and for the balance of their life.
This study presents new research with respect to the impact on patients' reported health outcomes when in a particular stage of the disease: newly diagnosed, relapsed, partial remission, complete remission and disease free. The limitation in extrapolating these results is that the subgroups are relatively small and the number of male patients is less than the current UK incidence rates would indicate with an overall male–female incidence ratio of 1.4 : 1.0 [4]. Another limitation of the study is its cross-sectional design, such that a more representative cohort across the stages of the disease would be obtained from a prospective longitudinal design.
In conclusion, FL patients who have relapsed are more likely to experience worse HRQoL and other patient-reported health outcomes than patients newly diagnosed, in partial or complete remission or when completely disease free. For those patients who have relapsed, increasing the amount of cancer-specific social support may also be associated with better patient-reported health outcomes [26]. The patient-reported health outcomes for an individual with FL differ according to his or her disease state. Those with the best patient-reported health outcomes had a complete response to treatment or were categorized as ‘disease free’.
Thus, it is the patient's disease status that had the greatest impact on these scores. It is arguable that prolonging a patient's status quo as disease free (i.e. prolonging the time to treatment failure is beneficial either by more intense induction or by maintenance treatment) is important in terms of improving HRQoL.
|
funding |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Roche Products Ltd.
|
Acknowledgements |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
We would also like to thank the following people for their invaluable input into the study through the recruitment of patients: Helen Beedham, St George's Hospital, London; Ruth Hall, Belfast City Hospital, Belfast; Sandra Laretta, University Hospital of Wales, Cardiff; Jackie Whyte and Emma Buckby, Leicester Royal Infirmary, Leicester; Jonathan Bullard, Mount Vernon Cancer Centre, Northwood, Middlesex; Janet Hutchinson, Weston Park Hospital, Sheffield; Rebecca Reddell-Denton and Nicky Crosbie, Derriford Hospital Plymouth; Sheelagh Price, Taunton and Somerset Hospital, Taunton.
Received for publication May 3, 2007. Revision received October 22, 2007. Accepted for publication October 25, 2007.
|
References |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
1. Ganti AK, Bociek RG, Bierman PJ, et al. Follicular lymphoma: expanding therapeutic options. Oncology (2005) 19:213–228.[CrossRef][Web of Science][Medline]
2. ISD Online. Cancer Incidence, Mortality and Survival Data (2005) http://www.isdscotland.org/.
3. Northern Ireland Cancer Registry. Cancer Registrations in Northern Ireland 2002. http://www.qub.ac.uk/research-centres/nicr/Data/OnlineStatistics/Non-HodgkinsLymphoma/. (21 November 2007, date last accessed).
4. Office for National Statistics. Cancer Statistics Registrations of Cancer Diagnosed in 2002. (Series MB1 no. 33). http://www.statistics.gov.uk/ (21 November 2007, date last accessed).
5. Welsh Cancer Intelligence and Surveillance Unit. Cancer Registrations in Wales 2002. http://www.wales.nhs.uk/ (21 November 2007, date last accessed).
6. Parkin DM, Pisani P, Ferlay J. Global cancer statistics. CA Cancer J Clin (1999) 49:33–64.
7. Wake B, Hyde C, Bryan S, et al. Rituximab as third-line treatment for refractory or recurrent stage III or IV follicular non-Hodgkin's lymphoma: a systematic review. Health Technol Assess (2002) 6:1–85.[Medline]
8. National Comprehensive Cancer Network. Non-Hodgkin's Lymphoma: Treatment Guidelines for Patients. Version I/October 2003. http://www.cancer.org/docroot/STT/content/STT_1x_Cancer_Facts__Figures_2007.asp/ (21 November 2007, date last accessed).
9. Webster K, Cella D. Quality of life in patients with low-grade non-Hodgkin's lymphoma. Oncology (1998) 12:697–714.[Web of Science][Medline]
10. Cella D, Mahon S, Donavan M. Cancer recurrence as a traumatic event. Behav Med (1990) 16:15–22.[Web of Science][Medline]
11. McLaughlin P. Progress and promise in the treatment of indolent lymphomas. Oncologist (2002) 7:217–225.
12. Basen-Engquist K, Cohen L, Caramack C. The Webster/Cella article reviewed. Oncology (2002) 12:714–717.
13. Wallwork L, Richardson A. Beyond cancer: changes, problems and needs expressed by adult lymphoma survivors attending an out-patients clinic. Eur J Cancer Care (1994) 3:122–132.[CrossRef]
14. Eremenco S, Webster K, Kutikova L, et al. Development and multilingual validation of the FACT-LYM. Presented at ISOQOL Annual Conference. 17th October 2004. Qual Life Res 2004; 13 (9): 1495–1603.
15. Webster K, Cashy D, Cella D, et al. Measuring quality of life (QOL) in patients with non-hodgkin's lymphoma (NHL): the Functional Assessment of Cancer Therapy-Lymphoma (FACT-LYM). Qual Life Res (2005) 14:2103.
16. Reilly M, Zbrozek A, Dukes EM. The validity and reproducibility of a work productivity and activity impairment instrument. Pharmacoeconomics (1993) 4:353–365.[Web of Science][Medline]
17. Snaith R, Zigmond A. The Hospital Anxiety and Depression Scale: Manual London, UK: (1994) NFER-Nelson.
18. Zigmond A, Snaith R. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand (1983) 67:361–370.[Web of Science][Medline]
19. Maher E. Anomic aspects of recovery from cancer. Soc Sci Med (1982) 16:907–912.[CrossRef][Web of Science][Medline]
20. Herold M, Sacchi S, Hieke K. The cost of treating relapsed indolent non-Hodgkin's lymphoma in an international setting; retrospective analysis of resource use. Haematologica (2002) 87:719–729.
21. Hiddemann W, Kneba M, Dreyling M, et al. Frontline therapy with rituximab added to the combination of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) significantly improves the outcome for patients with advanced-stage follicular lymphoma compared with therapy with CHOP alone: results of a prospective randomized study of the German Low-Grade Lymphoma Study Group. Blood (2005) 106:3725–3732.
22. van Oers MH, Klasa R, Marcus RE, et al. Rituximab maintenance improves clinical outcome of relapsed/resistant follicular non-Hodgkin's lymphoma, both in patients with and without rituximab during induction: results of a prospective randomized phase III intergroup trial. Blood (2006) 108:3295–3301.
23. Gopal AK, Gooley TA, Maloney DG, et al. High-dose radioimmunotherapy versus conventional high-dose therapy and autologous hematopoietic stem cell transplantation for relapsed follicular non-Hodgkin lymphoma: a multivariable cohort analysis. Blood (2003) 102:2351–2357.
24. Schilder RJ, Witzig TE, Flinn I, et al. Yttrium 90 (90Y) Ibritumomab Tiuxetan (Zevalin(R)) induces long-term responses in patients with relapsed or refractory follicular lymphoma (FL) (2004) 104. ASH Annual Meeting Abstracts.
25. Schouten HC, Qian W, Kvaloy S, et al. High-dose therapy improves progression-free survival and survival in relapsed follicular non-Hodgkin's lymphoma: results from the randomized European CUP trial. J Clin Oncol (2003) 21:3918–3927.
26. Oken M, Creech R, Tormey D, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol (1982) 5:655.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||