Annals of Oncology Advance Access originally published online on December 15, 2005
Annals of Oncology 2006 17(2):304-312; doi:10.1093/annonc/mdj072
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2005 European Society for Medical Oncology
Clinically meaningful changes in health-related quality of life in patients diagnosed with hepatobiliary carcinoma
1 University of Pittsburgh School of Medicine, Starzl Transplantation Institute, Liver Cancer Center, Pittsburgh, PA; 2 Evanston Northwestern Healthcare and Northwestern University Feinberg School of Medicine, Chicago, IL, USA
* Correspondence to: Dr J. L. Steel, University of Pittsburgh, School of Medicine, Starzl Transplantation Institute, Liver Cancer Center, 3459 Fifth Avenue, Montefiore 7 South, Pittsburgh, PA 15213, USA. Tel: +1-412-692-2041; Fax: +1-412-692-2002; E-mail: steeljl{at}msx.upmc.edu
 |
Abstract |
|---|
 Top Abstract introductionmethodsresultsdiscussionReferences |
|---|
Background: To test the reliability, sensitivity to change in biomarkers associated with disease progression and response to treatment, and clinical meaningfulness of the Functional Assessment of Cancer Therapy-Hepatobiliary (FACT-Hep) in patients with hepatobiliary carcinoma.
Patients and methods: One hundred and fifty-eight patients diagnosed with hepatobiliary carcinoma were prospectively studied. Health-related quality of life (HRQL) was assessed at baseline (prior to treatment), 3-month follow-up (n = 55) and 6-month follow-up (n = 27).
Results: The internal consistency of all the scales of the FACT-Hep were adequate at all time points (>0.75). The FACT-Hep was found to be sensitive to changes in clinical indicators (alkaline phosphate, alpha-fetoprotein, hemoglobin and survival) that reflect disease progression and response to treatment. Combined results from distribution-based and cross-sectional anchor-based analyses provide the following minimally important difference (MID) estimates: FACT-General (FACT-G) subscales = 2–3; FACT-G = 6–7; Hepatobiliary Cancer Subscale = 5–6; FACT-Hep = 8–9; Trial Outcome Index = 7–8; and FACT-Hepatobiliary Symptom Index = 2–3 points.
Conclusions: The FACT-Hep is a reliable instrument that is responsive to clinical indicators of disease progression and response to treatment. The MID estimates can aid interpretation of HRQL data and facilitate sample size calculation in clinical trials.
Key words: clinical significance, Functional Assessment of Cancer Therapy-Hepatobiliary, health-related quality of life, hepatobiliary carcinoma, minimally important difference
|
introduction |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionReferences |
|---|
Hepatocellular carcinoma (HCC) accounts for more than 250 000 deaths a year, and is the leading cause of death from cancer in eastern Asia and sub-Saharan Africa and the sixth leading cause of cancer death worldwide [1
]. With the estimated 3.9 million people who are chronically infected with hepatitis C in the USA, the rates of HCC are expected to increase in the next decade in North America and Europe, where hepatobiliary cancers were previously considered relatively rare [2].
Approximately 80% of patients diagnosed with HCC present with unresectable lesions and are unable to undergo transplantation [2]. Non-surgical treatment of this cancer has been demonstrated to have, at best, only modest improvements in survival [3–6], and as a result, health-related quality of life (HRQL) becomes paramount. The measurement of HRQL as an end point has become routine in clinical trials testing the safety and efficacy of new chemotherapeutic agents for nearly all types of cancer, including hepatobiliary carcinoma.
A variety of instruments designed to assess HRQL are available to clinicians and researchers [7, 8]. One of these instruments, the Functional Assessment of Cancer Therapy (FACT) [9], is used extensively in oncology clinical trials especially in North America. The FACT is a general cancer HRQL instrument; however, disease-specific modules are also available to assess additional concerns related to a specific malignancy and/or its treatment. A version of the FACT, the FACT-Hepatobiliary (FACT-Hep), has recently been developed and demonstrated to be valid and reliable in a sample of patients with hepatobiliary carcinoma [9, 10]. The sample of patients who were recruited for the validation study consisted of patients diagnosed primarily with colorectal cancer with liver metastases (40%). Therefore, further testing of the reliability of this instrument in a sample of patients primarily diagnosed with hepatocellular carcinoma may offer an important confirmation of the instrument's generalizability.
The FACT-Hep has begun to be employed as an end point in non-randomized [11] and randomized controlled trials testing new treatments for hepatobiliary carcinoma [12–14]. The instrument's sensitivity to change associated with disease progression and response to treatment has not yet been tested. The present study will begin to evaluate the sensitivity of FACT-Hep to clinical indicators that reflect disease progression and response to treatment. In addition, the clinical meaningfulness of FACT-Hep changes and score differences have not been evaluated. A second aim of this study will be to estimate minimally important differences (MIDs) for the FACT-Hep.
A MID on a patient-reported outcome measure has been defined as the ‘smallest difference in score in the domain of interest that patients perceive as important, either beneficial or harmful, and which would lead a clinician to consider a change in the patient's management’ [15]. Prior studies of instruments within the FACIT measurement system have used a combination of distribution and anchor-based analyses to estimate scale and subscale MIDs [16, 17].
Distribution-based methods are based on statistical properties of the scale and the distribution of HRQL scores. We have used one-third to one-half of a standard deviation (SD) to estimate MIDs on FACT scales. Score differences of this magnitude are associated with effect sizes of 0.33 and 0.50, respectively. This range falls within Cohen's recommended cutoffs for small (0.20) to moderate (0.50) effect sizes [18]. The use of the standard deviation alone to estimate MIDs is problematic since results can vary across samples. Hence, we also calculate the standard error of measurement (SEM). A single SEM has been used to estimate the MID on health status measures because it is less sample dependent [19, 20]. MID estimates are further refined using anchor-based methods.
Anchor-based methods ‘anchor’ or ‘map’ score differences onto differences in clinical indicators. Clinical anchors can be objective (e.g. response to treatment) or subjective (e.g. performance status), and can be determined using cross-sectional or longitudinal analyses. Since there is no single best method for estimating MIDs, the use of multiple strategies simultaneously has been recommended [16]. Hence, we combined the distribution- and anchor-based methods to determine scale MIDs for the FACT-Hep.
The aims of the present study were to: (i) confirm the reliability of the FACT-Hep in a heterogeneous sample of patients with hepatobiliary carcinoma; (ii) test the sensitivity of FACT-Hep to changes of biomarkers that reflect disease progression and treatment response; and (iii) estimate MIDs on the FACT-Hep using distribution- and anchor-based methods.
|
methods |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionReferences |
|---|
patients
One hundred and fifty-eight patients diagnosed with hepatobiliary carcinoma were prospectively followed and HRQL was assessed at baseline (prior to treatment), and to 3 months (n = 55) and 6 months (n = 27) follow-up. The data come from three separate studies [11
, 21]. Inclusion criteria for each of the participants for all three samples included: (i) biopsy proven hepatobiliary carcinoma; (ii) treatment with TACE or 90-yttrium microspheres; (iii) 18 years of age or older; and (iv) ability to read and speak English fluently. Exclusion criteria included: (i) current suicidal or homicidal ideation; (ii) current psychosis; or (iii) health too poor to complete questionnaires. Of the patients in the study, 69% received treatment with TACE and 31% with 90-yttrium microspheres. Patients treated with TACE were treated on average every 8 weeks. 90-Yttrium microspheres were administered once after diagnosis and subsequently with progression of lesion(s).
health-related quality of life
HRQL data were collected prospectively. Version 4 of the FACT-Hep [9, 10] was administered at baseline (prior to treatment), and at 3 and 6 months follow-up. The 45-item FACT-Hep consists of five subscales: (1) physical well-being (PWB); (2) social and family well-being (SFWB); (3) emotional well-being (EWB); (4) functional well-being (FWB); and the hepatobiliary cancer subscale (HepCS). The HepCS includes 18 items that assess specific symptoms of hepatobiliary carcinoma and side-effects of its treatment. Aggregate scores can also be formed. The PWB, FWB, SFWB and EWB are summed to form the FACT-General total score (FACT-G). The FACT-G and HepCS score are summed to form the FACT-Hep total score. The Trial Outcome Index (TOI) consists of the summation of the PWB, FWB and HepCS subscales. The TOI has been demonstrated to be a sensitive indicator of clinical outcome in other disease types [9]. Finally, the FACT-Hepatobiliary Symptom Index (FHSI) includes eight items from the FACT-Hep that measure specific symptoms and side-effects of hepatobiliary carcinoma [22]. All FACT items are rated on 5-point scales ranging from 0 = not at all to 4 = very much. Higher scores on all scales of the FACT-Hep reflect better quality of life or fewer symptoms [10].
clinical anchors
The clinical anchors for this study were determined by an experienced clinician and researcher who has treated patients with hepatobiliary carcinoma for nearly two decades (B.C.). The clinical anchors included: (i) survival; (ii) alpha-fetoprotein (AFP); (iii) alkaline phosphate (ALK); and (iv) hemoglobin. The clinical anchors were obtained through results from routine computed tomography scans, pathology and laboratory tests. Survival was measured from the date of diagnosis of hepatobiliary cancer to the date of death. The clinical anchors that were chosen are based on previous research concerning predictors of HRQL and disease progression in patients with hepatocellular carcinoma [23–33]. Although HRQL would be expected to be positively associated with survival, the results regarding this association are not consistent [23–27]. AFP [27–29] and ALK [30] have been demonstrated to be associated with tumor progression and poor prognosis in patients with HCC. Anemia has been found to be associated with lower levels of HRQL [31] and is believed to be associated with disease progression and survival [32, 33]; however, this is still under investigation.
data analyses
Descriptive statistics were used to obtain information regarding the demographic and disease-specific characteristics of this sample. Cronbach's alphas were calculated to determine the internal consistency of each of the scales of the FACT-Hep at all three time points. One-way analyses of variance (ANOVAs) and independent samples t-tests were used to compare FACT-Hep scores across clinically distinct groups. Paired samples t-tests and repeated measures ANOVAs were used to determine the instrument's responsiveness to change. Neuman–Keuls post-hoc tests (P < 0.05) were performed to specify any significant omnibus effects. Spearman correlations were used to determine associations between FACT-Hep scores and any continuous variables.
Two distribution-based analyses were used to estimate MIDs for the FACT-Hep. Standard deviations of FACT-Hep scores were divided by 3 and 2 to establish 1/3 and 1/2 SD estimates. The SEM for the FACT-Hep scores were also calculated using the following formula:
 |
x = the SD of the scale or subscale and relx = the reliability of the scale or subscale (test–retest reliability or internal consistency). These parameters have frequently been used to estimate MIDs on other FACIT measures [16]. For anchor-based analyses, FACT-Hep score differences between clinically distinct groups were computed along with corresponding effect sizes (group means divided by the pooled within-group SD). |
results |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionReferences |
|---|
demographic and disease-specific characteristics
Demographic and disease-specific characteristics of the patient sample can be found in Tables 1–3
. Significant differences in gender [F(2,154) = 3.0; P = 0.05], age [F(2,150) = 9.0; P <0.001], gamma glutamyl transpeptidase (GGTP) [F(2,108) = 4.5; P = 0.01] and survival [F(2,90) = 4.1; P = 0.02] were observed across the three studies (see Tables 1–3). Neither gender nor GGTP was associated with FACT-Hep scores (data not shown). Although age was positively correlated with a few FACT-Hep scores (e.g. PWB, FHSI), correlation magnitudes were small (rhos <0.28). These rather modest differences led us to combine the three samples to maximize power and generalizability. Survival, different across the three samples, was strongly associated with FACT-Hep scores and was therefore used as an anchor in later analyses.
|
|
|
With the three samples combined, 75% of the sample was male, which is consistent with the gender ratio for this disease (2:1). The mean age was 64 years (range 22–90). The majority of the sample was Caucasian (90%), followed by African-American (6%), Asian-American (3%) and Hispanic (1%).
The primary diagnosis was HCC (85%) and the primary etiology of the cancer was hepatitis B and/or C (45%). Of the 158 patients in this study, 73% had cirrhosis. Please refer to Tables 2 and 3 for further information regarding disease-specific factors.
reliability of the FACT-Hep
Internal consistency for each of the subscales at baseline (prior to treatment), and at 3 and 6 months' follow-up was found to be adequate (all alphas >0.76). At baseline the Cronbach alphas ranged from 0.76 to 0.92. At 3-months follow-up the Cronbach alphas ranged from 0.85 to 0.97 and at 6-months follow-up the alphas ranged from 0.82 to 0.87.
sensitivity to change and effect size of the FACT-Hep over time
Paired sample t-tests were performed from baseline to 3 months' follow-up (n = 55) for all FACT-Hep subscales. Significant decrements in all subscales were found from baseline to 3 months' follow-up (see Table 4).
|
Repeated measures ANOVA was performed to test changes over time from baseline to 6 months. Significant differences were found on all the subscales of the FACT-Hep with the exception of FWB subscale, which approached significance: PWB [F(2,54) = 8.6; P = 0.01]; SFWB [F(2,52) = 15.0; P = 0.001]; EWB [F(2,54) = 5.4; P = 0.01]; FWB [F(2,52) = 3.2; P < 0.06]; FACT-G [F(2,52) = 11.1; P = 0.001]; HepCS [F(2,55) = 16.1; P = 0.001]; FACT-Hep [F(2,52) = 15.0; P = 0.001]; TOI [F(2,52) = 13.3; P = 0.001]; and FHSI [F(2,54) = 11.0; P = 0.001]. Pairwise comparisons between baseline, and 3 and 6 months' follow-up were performed, and for all subscales the baseline means were found to be higher than 3 and 6 months' follow-up. For the HepCS, FACT-Hep and TOI, the mean scores at 6 months' follow-up were found to be significantly higher than at 3 months' follow-up (Figure 1).
|
clinically meaningful changes in the FACT-Hep
distribution-based analyses.
Distribution-based estimates of the MID are shown in Table 5. Since sample sizes varied considerably over the two time points, we computed a weighted mean of each parameter to summarize the data for each scale. Weighted means for the 1/3 SD, 1/2 SD and 1 SEM criteria are: PWB = 2.53, 3.81 and 2.30; SFWB = 2.32, 3.49 and 2.38; EWB = 2.06, 3.09 and 2.23; FWB = 2.22, 3.33 and 2.56; FACT-G = 7.79, 10.95 and 5.62; HepCS = 5.30, 7.94 and 4.53; FACT-Hep = 11.72, 17.59 and 7.74; TOI = 8.82, 13.23 and 6.27; and FHSI = 2.36, 3.55 and 3.08. These results provide initial estimates for MIDs for each of the scales of the FACT-Hep.
|
anchor-based analyses.
Baseline ALK and hemoglobin were used as cross-sectional anchors. Critical values of these biomarkers were used to compare clinically distinct groups. For ALK, 44–147 U/l was considered to be normal, whereas >147 U/l was considered high. For hemoglobin, >12 g/dl was considered normal and
12 g/dl was considered low. Critical value comparisons were found to be significant for ALK on the EWB, FWB and the FACT-G subscales, in which higher subscale scores on the FACT-Hep were associated with normal levels of ALK. For hemoglobin, significant differences were found on the EWB subscale, in which higher levels of hemoglobin were associated with higher reported EWB (Tables 6 and 7).
|
|
prospective anchor-based evidence.
Prospective analyses of changes in clinical anchors were also performed. Significant relationships were found between changes in AFP from baseline to 3-month on the PWB, HepCS, TOI and FHSI subscales (Table 8). For each of the subscales, an improvement in AFP (reduction in AFP) was associated with a less negative change in HRQL when compared to worsening of AFP (increase).
|
Short- and long-term survival was also analyzed in relation to changes in scores on the FACT-Hep scales. Significant differences were found on the PWB, SFWB, FACT-G, FACT-Hep, HepCS, TOI and the FHSI subscales (Table 9). Patients who had a survival of >5 months had a greater decline from baseline to 3 months on the above FACT-Hep subscales compared with patients who had a survival of <5 months.
|
A summary of the cross-sectional and prospective anchor-based analyses can be found in Table 10. Based on the results of the cross-sectional anchor-based analyses, <1% of the analyses involving the relationship between HRQL and clinical anchors had small effect sizes (<0.2). Thirty-nine per cent (14/39) of the analyses resulted in a small to moderate effect size (0.2–0.5), 25% (9/36) of the analyses resulted in a moderate to large effect size (0.5–0.8), and 28% (10/36) resulted in a large effect size (>0.8). For the purpose of this paper, 31% (11/36) of the analyses had an effect size that would be categorized as an MID (0.3–0.5). The overall mean differences found in the first and third columns of Table 10 fall within the distribution based estimates provided in Table 5. Based on a combination of distribution-based criteria, cross-sectional and prospective anchor-based methods, the last column of Table 10 provides estimates for MIDs for the FACT-Hep.
|
|
discussion |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionReferences |
|---|
Although previous validation studies have been conducted with the FACT-Hep [9
, 10], the present study provides additional data regarding the reliability of this instrument. Furthermore, the results of this study add novel information regarding the sensitivity of FACT-Hep to changes in biomarkers that reflect disease progression and response to treatment, and as a result estimates for MIDs. The reliability for the FACT-Hep, with this sample of patients with hepatobiliary carcinoma, was adequate for all scales of the instrument at all time points. These results further confirm the reliability of the FACT-Hep in a new heterogeneous sample of patients, primarily with a diagnosis of HCC. Significant decrements in HRQL, with large effect sizes, were observed from baseline to 3 months' follow-up on nearly all scales of the FACT-Hep. The decrease in HRQL from baseline to 3 months' follow-up is likely a consequence of the side-effects associated with TACE. Sixty-nine per cent of the patients in this study were treated with TACE. At the 3 month assessment, most of the patients would have been only 2–3 weeks post-treatment and likely still experiencing side-effects of treatment.
At 6 months' follow-up, albeit in few patients (n = 27), scores improved from the 3- to 6-month follow-up on the HepCS, FACT-Hep and TOI subscales. None of the scale scores, including the scales that improved, returned to baseline levels, suggesting that disease progression may have attenuated improvements in HRQL subsequent to treatment. The timing of assessment may explain the improvement on some of the FACT-Hep subscales at 6 months, as the majority of patients were likely to be 6–10 weeks post-treatment.
The present study also differentiated groups of patients based on clinical indicators at baseline as well as prospectively. The FACT-Hep was found to be related to baseline measures of ALK, hemoglobin and vascularity of the lesion. Higher ALK may be associated with bile obstruction and increased ascites and/or malnutrition that may be associated with alcohol-related cirrhosis and/or HCC. In either case, lower HRQL would be expected.
Prospectively, we also found that the FACT-Hep subscales were related to changes in AFP and survival. When AFP increased from baseline to 3 months (i.e. worsened), patients reported greater decrements in HRQL. Unexpectedly, patients who had lived longer than the median survival for this sample (>5 months) reported lower HRQL on several of the subscales of the FACT-Hep (PWB, SFWB, FACT-G, HepCS, FACT-Hep, TOI, FHSI) than patients who had lived less than the median survival (<5 months) for this sample. One explanation of this finding may be that those patients who responded to treatment also experienced greater side-effects, and as a result poorer HRQL, than patients who did not respond to treatment and thus had shorter survival.
The distribution-based methods were initially employed to provide guidelines to estimate MIDs. The SEM was employed as it is theoretically less sample dependent and has greater generalizability. The cross-sectional and longitudinal anchor-based analyses facilitate the ability to estimate MIDs by narrowing the range of initial estimates calculated using distribution-based methods. Recommendations for MIDs were based on the distribution- and anchor-based analyses, and can be found in Table 10.
A significant limitation to this investigation is the attrition of patients over the course of the study and the small sample size at 6 months' follow-up. Owing to illness and death, many patients were not able to complete the 6-month assessment, thus resulting in small sample size and reduced power. Prior research has demonstrated that younger age as well as more severe impairment may result in differential MIDs [15]; however, this may be more likely with the use of distribution-based methods to obtain MIDs. Because both distributional- and anchor-based methods were used, the present estimates are likely to be more generalizable to other samples of patients with hepatobiliary carcinoma, independent of age and functional ability.
The present study included a sample that was relatively homogeneous with regard to diagnosis but heterogeneous on many other demographic and disease-specific variables. The SEM was employed in addition to the SD, as the SEM is considered independent of the sample. Although low internal consistency can affect the SEM, the Cronbach alphas for each of the subscales of the FACT-Hep were in the acceptable range (0.76–0.98). Some argue that test–retest reliability is more important when using the SEM with the distribution-based approach to calculating MIDs [19, 20]. Wyrwich et al. suggested that a single SEM (i.e. 1x SEM) corresponds with anchor-based approaches [19, 20], while others have suggested that 2x SEM or 2.77x SEM represent an important change (95–99% confidence interval [34, 35]).
Although the sample size was small and the sample disease-specific characteristics rather heterogeneous, the results of this study add vital information regarding the measurement and interpretation of HRQL data in patients with hepatobiliary carcinoma. At this time, although there is one other disease-specific HRQL instrument for patients with hepatobiliary carcinoma, the European Organization on Research and Treatment of Cancer's QLQ-HCC18 [36], there is no method for interpreting the clinical meaningfulness of the data obtained using this instrument.
The MIDs for the FACT-Hep have several clinical and research applications including (i) interpretation of individual or group differences on measures of HRQL, (ii) estimation of sample size or power for future studies and (iii) determining whether a trial or treatment should be stopped based on clinically relevant decrements in HRQL. The use of MIDs will be important for testing the efficacy of new treatments for hepatobiliary carcinoma as well as psychosocial interventions.
Received for publication August 8, 2005. Revision received October 5, 2005. Accepted for publication October 10, 2005.
|
References |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionReferences |
|---|
1. Harrison TJ. Viral hepatitis and primary liver cancer. In Arrand JR, Harper DR (eds): Viruses and Human Cancer. Oxford: BIOS Scientific Publishers Limited 1998.
2. Edwards BK, Brown ML, Wingo PA et al. Annual report to the nation on the status of cancer, 1975–2002, featuring population-based trends in cancer treatment. J Natl Cancer Inst 2005; 97: 1407–1427.
3. Llovet JM, Bruix J. Systematic review of randomized trial for unresectable hepatocellular carcinoma: Chemoembolization improves survival. Hepatology 2003; 37: 429–442.[CrossRef][Web of Science][Medline]
4. Tinchet JC, Ganne-Carrie N, Beaurant M. Review article: Intra-arterial treatments in patients with hepatocellular carcinoma. Aliment Pharmacol Ther 1996; 7 (Suppl 2): 111–118.
5. Camma C, Schepis F, Orlando A et al. Transarterial chemoembolization for unresectable hepatocellular carcinoma: Meta-analysis of randomized controlled trials. Radiology 2000; 224: 47–54.[CrossRef]
6. Geschwind JF, Ramsey D, Choti M et al. Chemoembolization of hepatocellular carcinoma: Results of a meta-analysis. Am J Clin Oncol 2003; 26: 344–349.[CrossRef][Web of Science][Medline]
7. Kantz ME, Harris WJ, Levitsky K et al. Medical Outcomes Study Short Form–36—"condition specific". Methods for assessing condition-specific and generic functional status outcomes after total knee replacement. Med Care 1992; 30: MS240–MS252.[CrossRef][Web of Science][Medline]
8. Bergman B, Aaronson NK, Ahmedzai S et al. EORTC Core Quality of Life Questionnaire—Lung Cancer-13. The EORTC QLQ-LC13: A modular supplement to the EORTC Core Quality of Life Questionnaire (QLQ-C30) for use in lung cancer clinical trials. Eur J Cancer 1994; 30A: 635–642.[CrossRef]
9. Cella DF, Tulsky DS, Gray G et al. The Functional Assessment of Cancer Therapy scale: Development and validation of the general measure. J Clin Oncol 1993; 11: 570–579.
10. Heffernan N, Cella D, Webster K et al. Measuring health-related quality of life in patients with hepatobiliary cancers: The Functional Assessment of Cancer Therapy-Hepatobiliary questionnaire. J Clin Oncol 2002; 20: 2229–2239.
11. Steel JL, Carr BI, Baum A. Quality of life in patients diagnosed with primary hepatocellular carcinoma: Hepatic arterial infusion of Cisplatin versus 90-Yttrium microspheres (Therasphere®). Psychooncology 2004; 13: 73–79.[CrossRef][Medline]
12. Carr BI. Phase II study of arsenic trioxide in patients with unrectable metastatic hepatocellular carcinoma. National Cancer Institute Clinical Trials Protocol ID PCI-04–06060. http://www.cancer.gov/search/ResultsClinicalTrials.aspx?protocolsearchid=1640831&sort=4&page=2&batchsize=10 (November 2005, date last accessed).
13. Carr BI. Phase II Study of hepatic arterial infusion of yttrium 90 glass microspheres (Therasphere) with unresectable hepatocellular carcinoma. National Cancer Institute Protocol ID PCI-04128. http://www.cancer.gov/search/ResultsClinicalTrials.aspx?protocolsearchid=1640831&sort=4&page=2&batchsize=10 (November 2005, date last accessed).
14. Carr BI. A research study to treat patients with advanced hepatocellular carcinoma. National Cancer Institute Protocol ID 11546. http://www.cancer.gov/search/ResultsClinicalTrials.aspx?protocolsearchid=1860746 (November 2005, date last accessed).
15. Guyatt GH, Osoba D, Wu AW et al. Methods to explain the clinical significance of health status measures. Mayo Clin Proc 2002; 77: 371–383.
16. Yost KJ, Eton DT. Combining distribution- and anchor-based approaches to determine minimally important differences: The FACIT experience. Eval Health Prof 2005; 28: 172–191.
17. Eton DT, Cella D, Yost KJ et al. A combination of distribution- and anchor-based approaches determined minimally important differences (MIDs) for four endpoints in a breast cancer scale. J Clin Epidemiol 2004; 57: 898–910.[CrossRef][Web of Science][Medline]
18. Cohen J. Statistical Power Analysis for Behavioral Sciences. Hillsdale, NH: Lawrence Erlbaum 1988.
19. Wyrwich KW, Nienaber NA, Tierney WM, Wolinsky FD. Linking clinical relevance and statistical significance in evaluating intraindividual changes in health-related quality of life. Med Care 1999; 37: 469–478.[CrossRef][Web of Science][Medline]
20. Wyrwich KW, Tierney WM, Wolinsky FD. Further evidence supporting an SEM-based criterion for identifying meaningful intraindividual changes in health-related quality of life. J Clin Epidemiol 1999; 52: 861–873.[CrossRef][Web of Science][Medline]
21. Steel JL, Geller DA, Carr BI. Proxy ratings of health related quality of life in patients with hepatocellular carcinoma. Qual Life Res 2005; 14: 1025–1033.[CrossRef][Web of Science][Medline]
22. Yount S, Cella D, Webster K et al. Assessment of patient-reported clinical hepatobiliary cancers: The FACT Hepatobiliary Symptom Index. J Pain Symptom Manage 2002; 24: 32–44.[CrossRef][Web of Science][Medline]
23. Dancey J, Zee B, Osoba D et al. Quality of life scores: An independent prognostic variable in a general population of cancer patients receiving chemotherapy. The National Cancer Institute of Canada Clinical Trials Group. Qual Life Res 1997; 6: 151–158.[Web of Science][Medline]
24. Hwang SS, Scott CB, Chang VT et al. Prediction of survival for advanced cancer patients by recursive portioning analysis: role of Karnofsky performance status, quality of life, and symptoms distress. Cancer Invest 2004; 22: 678–687.[CrossRef][Web of Science][Medline]
25. Llobera J, Eseva M, Rifa J et al. Terminal cancer duration and prediction of survival time. Eur J Cancer 2000; 36: 2036–2043.[CrossRef][Web of Science][Medline]
26. Toscani P, Brunelli C, Miccinesi G et al. Predicting survival in terminal cancer patients: Clinical observation or quality of life evaluation. Palliat Med 2005; 19: 220–227.
27. Tamburini M, Brunelli C, Rosso S, Ventafridda V. Prognostic value of quality of life scores in terminal cancer patients. J Pain Symptom Manage 1996; 11: 32–41.[CrossRef][Web of Science][Medline]
28. Peng SY, Ou YH, Chen WY et al. Abberrant expressions of annexin A10 isoform, osteopontin and alpha-fetoprotein at chromosome 4q cooperatively contribute to progression and poor prognosis of hepatocellular carcinoma. Int J Oncol 2005; 26: 1053–1061.[Web of Science][Medline]
29. Yoon SK, Lim NK, Ha SA et al. The human cervical cancer oncogene protein is a biomarker for human hepatocellular carcinoma. Cancer Res 2004; 64: 5424–5441.
30. Farinati F, Salvagnini M, de Maria M et al. Unresectable hepatocellular carcinoma: A prospective controlled trial with tamoxifen. J Hepatol 1990; 11: 297–301.[CrossRef][Web of Science][Medline]
31. Cella D, Dobrez D, Glaspy J. Control of cancer-related anemia with erythropietic agents: A review of evidence for improved quality of life and clinical outcomes. Ann Oncol 2003; 14: 511–519.
32. Vaupel P, Mayer A. Hypoxia and anemia: Effects on tumor biology and treatment resistance, Transfus Clin Biol 2005; 12: 5–10.[CrossRef][Web of Science][Medline]
33. Knight K, Wade S, Balducci L. Prevalence of outcomes of anemia in cancer: A systematic review of the literature. Am J Med 2004; 116: 11S–26S.
34. McHorney CA, Tarlov AR. Individual-patient monitoring in clinical practice: Are available health status surveys adequate? Qual Life Res 1995; 4: 293–307.[CrossRef][Web of Science][Medline]
35. Beckerman H, Roebroeck ME, Lankhorst GJ et al. Smallest real difference, a link between reproducibility and responsiveness. Qual Life Res 2001; 10: 571–578.[CrossRef][Web of Science][Medline]
36. Blazeby JM, Currie E, Zee BCY et al. Development of a questionnaire module to supplement the EORTC QLQ-30 to assess quality of life in patients with hepatocellular carcinoma, the EORTC QLQ-HCC18.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. J. Hlubocky, M. J. Ratain, M. Wen, and C. K. Daugherty Complementary and Alternative Medicine Among Advanced Cancer Patients Enrolled on Phase I Trials: A Study of Prognosis, Quality of Life, and Preferences for Decision Making J. Clin. Oncol., February 10, 2007; 25(5): 548 - 554. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||