Journals | Policy | Permission

World Journal of Oncology

Original Article

Volume 2, Number 4, August 2011, pages 166-174

What QLQ-C15-PAL Symptoms Matter Most for Overall Quality of Life in Patients With Advanced Cancer?

Traditional endpoints of overall survival are no longer the primary endpoint of most palliative oncology clinical trials and have been replaced with more patient-specific determinants measuring symptom control and quality of life (QOL) [1]. Assessing treatment effect on QOL is particularly challenging given that there has yet to be a consensus with respect to a unified definition of QOL, which generally includes factors beyond health in the context of an individual’s culture and value system [2]. As a component of overall QOL, health-related quality of life (HRQOL) is a subjective multidimensional construct that includes physical and psychosocial factors related to patient health care [3]. To address the need for standardized assessment of HRQOL in oncology patients, various HRQOL questionnaires have been developed including the well established QLQ-C30 questionnaire developed by the European Organization for Research and Treatment of Cancer (EORTC) [4].

In studies that have assessed the prevalence of distressful symptoms in patients with advanced cancer, up to 85% of patients have reported at least one moderate to severe symptom [5]. With such a significant symptom burden, it becomes important to identify which symptoms matter most to the patient in terms of overall QOL, so that palliative treatments and supportive care may be tailored accordingly. In the lung cancer literature, this concept has been explored using the EORTC QLQ-C30 to investigate the relative importance of various HRQOL symptoms and functional domains in predicting overall QOL [6, 7].

An abbreviated version of the QLQ-C30, known as the QLQ-C15-PAL, has been developed to decrease the burden of the longer more time-consuming parent questionnaire for patients with advanced cancer [8]. Although abbreviated, the QLQ-C15-PAL still provides a single, convenient platform to measure HRQOL symptoms and functional domains in addition to overall globally-rated QOL. However, few studies have actually integrated or evaluated the QLQ-C15-PAL in its intended target population. Using the QLQ-C15-PAL in a cohort of patients with advanced cancer, the present study assessed the relative importance of different functional/symptom domains in predicting overall QOL ratings prior to palliative radiation treatment (RT) of symptomatic bone, brain or lung lesions.

The Rapid Response Radiotherapy Program (RRRP) is a rapid-access outpatient palliative RT clinic running daily in the Odette Cancer Centre at the Sunnybrook Health Sciences Centre, Toronto, Ontario Canada. All research was therefore conducted following approval from the Sunnybrook Health Science Centre research ethics board. This study included patients seen in RRRP consultation between October 2007 and July 2010, with completed QLQ-C15-PAL questionnaires prior to palliative RT for bone metastases, brain metastases or advanced lung cancer. The term advanced lung cancer will be used throughout the manuscript referring to patients who were seen in consultation for palliative RT of primary lung cancer or lung metastasis. Baseline patient data collected included age, Karnofsky performance status (KPS), gender, primary cancer site and presence of other metastases, including visceral or bone metastases outside of the RT site.

Prior to RT, general HRQOL and overall QOL was assessed using the QLQ-C15-PAL. This QOL measurement tool consists of 15 questions; two multi-item functional scales (physical and emotional functioning), two multi-item symptom scales (fatigue and pain) along with five single item symptom scales (nausea/vomiting, dyspnoea, insomnia, appetite loss, constipation) and one final question referring to overall QOL. The physical functioning scale is based on three questions that ask about walking, activities of daily living and time spent in bed or in a chair. The emotional functioning scale is based on two questions that ask about feeling tense or depressed. Patients rated each question/item on a numeric scale from 1 (not at all) to 4 (very much), with the exception of global QOL which was rated from 1 (very poor) to 7 (excellent). The EORTC QLQ-C30 scoring manual [9] was used to generate the QLQ-C15-PAL scores (0-100) for the unchanged pain scale and the four single items unchanged from the QLQ-C30 (dyspnoea, insomnia, appetite loss, constipation). Scores (0-100) for the remaining scales were generated using the QLQ-C15-PAL scoring addendum available from the EORTC Quality of Life Unit.

Results were initially analyzed for the entire group of patients with advanced cancer. Demographics were expressed as proportions for categorical variables but means, standard deviations (SD), inter-quartiles, medians and ranges (minimum to maximum) for continuous variables. Pearson correlations were computed between all functional/symptom scales and the overall QOL scale. The strength of the relationship was indicated by the correlation coefficient (r). The larger the correlation, the stronger the relationship. The following rules were used for categorizing r: ≤ 0.50 very low; 0.51-0.79 low; 0.80-0.89 moderate, ≥ 0.90 high. The significance of the relationship was expressed in probability levels of p-value. The smaller the p-value, the more significant the relationship. When sample size is large, a small r-value may be significant but with a weaker relationship [10].

A multiple linear regression analysis was then used to evaluate the relative importance of different functional/symptoms scales in predicting overall QOL. Two separate models were computed. In the first model, the 2 functional scale variables were entered simultaneously with the regression equation. In the second model, the 7 symptom scale variables were entered simultaneously with the regression equation. Finally, variables found to be statistically significant in previous regression analyses were entered together as independent variables in predicting overall QOL. The above process of statistical analysis was performed for each subgroup of patients with symptomatic bone, brain and lung disease. Two-tailed p-values less than 0.05 were considered statistically significant. All analyses were conducted by Statistical Analysis Software (SAS for Windows, version 9.2).

A total of 369 patients with advanced cancer completed the QLQ-C15-PAL questionnaires prior to palliative RT for bone (190 patients), brain (150 patients) or lung disease (29 patients). Baseline patient characteristics are presented in Table 1.

Click to view

Table 1. Patient Characteristics (n = 369)

Pearson correlations in all patients showed that all functional and symptom scales were significantly correlated with overall QOL (Table 2), although the small r-value may have indicated very low or low relationships. In the initial multiple linear regression analysis (Table 3), both physical and emotional functioning significantly predicted the overall QOL (p < 0.0001). Fatigue (p < 0.0001), pain (p = 0.0009) and appetite loss (p = 0.0041) were significant symptoms predictive of overall QOL. The functional and symptom scales accounted for 35% and 28% of the variance in overall QOL, respectively. Positive coefficients of the two functional scales indicated that higher functional scores (better physical and emotional functioning) predicted higher overall QOL. Negative coefficients of the three significant symptom scales indicated that higher symptom scales (more symptom burden) predicted lower overall QOL.

Click to view

Table 2. Pearson Correlations Between all Functional/Symptom Scale Variables and Overall QOL in All Patients With Advanced Cancer

Click to view

Table 3. Multiple Linear Regression Analysis of All Functional/Symptom Scale Variables With Overall QOLa in All Patients With Advanced Cancer

When the significant functional scale and significant symptom scale variables were put into one model (Table 4), both functional scale variables (p < 0.0001 for emotional; p = 0.0026 for physical) along with symptoms of pain and appetite loss (p = 0.025 for both) were significantly related to overall QOL. The functional and symptom scale variables together accounted for 39% of the variance in overall QOL.

Click to view

Table 4. Multiple Linear Regression Analysis With Physical and Emotional Functioning, Pain and Appetite Loss as Predictors of Overall QOLa in all Patients With Advanced Cancer

Baseline characteristics are presented in Table 5 for the three subgroups of patients with advanced cancer of bone, brain and lung.

Click to view

Table 5. Patient Characteristics of the Three Patient Subgroups

With the exception of dyspnoea, all functional and symptom scale variables were significantly correlated with overall QOL according to Pearson correlations (Table 6), although the small r-value indicated very low or low relationships. Through regression analysis, both physical (p = 0.0038) and emotional (p < 0.0001) functional scales as well as the symptom scales of pain (p = 0.003) and appetite loss (p = 0.007) were initially identified as significant predictors of overall QOL. The functional and symptom scales accounted for 32% and 21% of the variance in overall QOL, respectively. When significant variables were put into one model, only emotional functioning (p < 0.0001) and appetite loss (p = 0.024) were significantly predictive of overall QOL (Table 7). The functional and symptom scales together explained 35% of the variance in overall QOL.

Click to view

Table 6. Pearson Correlations Between all Functional/Symptom Scale Variables and Overall QOL in the Three Patient Subgroups

Click to view

Table 7. Multiple Linear Regression Analysis With Emotional Functioning and Appetite Loss Associated With Overall QOLa in Patients With Bone Metastases

Pearson correlations showed that all functional and symptom scales were significantly correlated with overall QOL (Table 6). Through regression analysis, both physical and emotional functional scales (p < 0.0001 for both) as well as the symptom scale of fatigue (p < 0.0001) were initially identified as significant predictors of overall QOL. Both functional scales and fatigue accounted for 31% and 24% of the variance in overall QOL, respectively. All three initially significant variables were predictive of overall QOL when put into one model (p < 0.0001 for emotional; p = 0.02 for physical; p = 0.001 for fatigue) (Table 8). The functional and symptom scales together explained 37% of the variance in overall QOL.

Click to view

Table 8. Multiple Linear Regression Analysis With Physical and Emotional Functioning and Fatigue Associated With Overall QOLa in Patients With Brain Metastases

With the exception of pain and constipation, all functional and symptom scales were significantly correlated with overall QOL according to Pearson correlations (Table 6). Through regression analysis, both physical (p = 0.026) and emotional (p = 0.001) functional scales as well as the symptom scale of appetite loss (p = 0.001) were initially identified as significant predictors of overall QOL. Both functional scales and appetite loss accounted for 54% and 58% of the variance in overall QOL, respectively. Only appetite loss (p = 0.035) was found to be predictive of overall QOL, when initially significant variables were put into one model (Table 9). The functional and symptom scales together explained 61% of the variance in overall QOL.

Click to view

Table 9. Multiple Linear Regression Analysis With Physical and Emotional Functioning and Fatigue Associated With Overall QOLa in Patients With Advanced Lung Cancer

In the palliative setting when facing treatment decisions for patients with poor prognosis, symptom control and QOL become arguably the most important goals of care rather than traditional oncologic endpoints of survival [1]. While objective endpoints such as overall survival are easily measured, standardization of subjective endpoints is possible with various HRQOL instruments such as the cancer-specific EORTC QLQ-C30 questionnaire [4] allowing for reproducibility across studies. Although the QLQ-C30 is one of the most frequently used measures of HRQOL in oncology clinical trials, it is lengthy. Certain studies have had issues with patient completion of questionnaires, particularly when combining the QLQ-C30 with disease-specific questionnaires such as the brain cancer-specific BN20 tool [11, 12]. Since the QLQ-C30, progress has been made in the field of HRQOL research with the development of an abbreviated QLQ-C15-PAL for patients with advanced cancer [8].

It has previously been proposed that HRQOL is antecedent to overall QOL and that it is important to differentiate between the two [6]. Using the QLQ-C30, fatigue and social functioning have been identified as independent predictors of overall QOL in 98 oncology patients with an estimated life expectancy of 1-6 months [13]. The QLQ-C15-PAL is an abbreviated HRQOL questionnaire consisting of two multi-item functional scales (physical and emotional functioning) along with seven symptom scales. In contrast to the QLQ-C30, there is only one question referring to overall QOL in the QLQ-C15-PAL. There have been several editorials discussing the use of QLQ-C15-PAL [14, 15] along with several papers discussing the planned use of QLQ-C15-PAL in upcoming studies [16-18]. However, to our knowledge only two other studies to date have assessed QOL using the QLQ-C15-PAL since its development in 2006 [19, 20]. Suarez-Del-Real et al. validated the Mexican-Spanish version of the QLQ-C15-PAL questionnaire for the evaluation of HRQOL in patients with advanced cancer receiving palliative care [19]. In a study of patients with brain metastases, Steinmann et al. found better compliance and practicality with the abbreviated QLQ-C15-PAL as compared to the QLQ-C30, and the group reported current use of the QLQ-C15-PAL in their ongoing larger scale study [20]. In the pilot study, the QLQ-C15-PAL identified pre-RT symptoms ranging from little nausea to more prominent fatigue and insomnia [20].

To our knowledge, this is the first study using the QLQ-C15-PAL to explore dimensions of HRQOL which may predict overall QOL in patients with advanced cancer. In the initial analysis of all 369 patients with advanced cancer, multiple regressions identified both physical and emotional functional scales as well as pain and appetite loss symptoms as significant predictors of overall QOL. This initial analysis included a broad group of patients with respect to their individual characteristics and tumor types. As the symptom burden is often very different depending on the disease site, sub-group analysis was performed. It is acknowledged that not all patients fit specifically into discrete categories as patients with advanced cancer often have multiple sites of disease. In this study, other metastases were present in 23%, 47% and 69% of patients with bone metastases, brain metastases and advanced lung cancer, respectively. By classifying patients according to the main reason for referral, this study has shown that different functional and symptom scales have varying significance to overall QOL in specific subpopulations of patients with advanced cancer.

In the relatively small group of patients with advanced lung cancer (n = 29), appetite loss was the only significant predictor of overall QOL. Emotional and physical functioning were initially significant but were not identified as independent predictors in the final model. These results are in contrast to a previously published study in lung cancer, identifying emotional functioning and fatigue as independent predictors of overall QOL, using the core QLQ-C30 along with the lung cancer-specific QLQ-LC13 questionnaire [7]. The authors concluded that emotional concerns may be of more importance to this lung cancer population with a potentially life-threatening disease as compared to a population of patients with chronic lung disease excluding cancer, in which physical functioning had been found to predict overall QOL [21]. Likewise, it may be proposed that overall QOL may be driven by different factors in patients with advanced versus localized and potentially curable cancer. The present study focused on patients with advanced lung cancer. Therefore, results may not reflect those found by Ostlund et al. [7] who reported on a more heterogeneous patient population, in which only 29 of 52 patients were identified as having stage 3b or 4 lung cancer. The present study’s interpretation of results may also be limited by the small sample size of patients with advanced lung cancer.

Our cohort consisted of a large number of patients with bone metastases (n = 190). In this patient population, emotional functioning and appetite loss were found to be predictive of overall QOL. Although pain is often the primary reason for referral of patients with bone metastases to a palliative RT clinic, pain was not predictive of overall QOL in this assessment. These findings are consistent with previous work using a variety of questionnaires, identifying psychosocial issues such as depression, and not pain, to be most predictive of overall QOL in patients with bone metastases [22]. While it has been shown that health care providers consider symptom issues such as pain to be more important for their patients with bone metastases, patients themselves emphasize the importance of psychosocial issues [23].

In the present study’s assessment of patients with brain metastases, predictors of overall QOL included both emotional and physical functioning as well as fatigue. The emotional burden of a brain metastases diagnosis may be multi-factorial, including the extremely poor prognosis associated with it and the current or anticipated cognitive/physical impairments which lead to loss of independence [24]. As most patients with brain metastases are taking dexamethasone during RT, it is not surprising that nausea and vomiting were not predictors of overall QOL. Steroid use may also explain why loss of appetite was predictive of overall QOL in patients with bone metastases or advanced lung cancer but not brain metastases.

The data from this study, like those from studies using the QLQ-C30 [7, 13], does not support a previous proposal that symptoms affect patient functioning but not overall QOL directly [25]. The current study has in fact highlighted the relative importance that patients place on certain symptoms such as appetite loss or fatigue, which are extremely common as cancer progresses [26] yet often neglected with respect to assessment or management [27, 28]. One recent study of advanced cancer patients has shown a statistically significant incremental decline in patient satisfaction with QOL for every 10 unit increase in appetite loss on the QLQ-C30 [29].

In conclusion, symptom control and QOL preservation are crucial goals in oncology. Standardized assessments of such subjective endpoints are now possible through a wide range of tools such as the QLQ-C15-PAL. Using the QLQ-C15-PAL, this study has shown that different functional and symptom scale variables predict overall QOL in patients with bone metastases, brain metastases or advanced lung cancer prior to palliative RT. It is acknowledged that such an assessment may be affected by various features including the assessment tool used and the patient population under investigation. While this is a study of patients referred for RT, results may vary in those patients receiving different treatment modalities. Future studies may wish to investigate whether the treatments themselves subsequently affect overall QOL and predictors thereof. Simply assessing the single variable of overall QOL may be important when an outcome variable is required or a global impression of QOL is desired [30]. On the other hand, multi-variable assessment, as is possible with the QLQ-C15-PAL, provides a broader, richer view of the situation [31]. In the present study, the QLQ-C15-PAL was used in its intended patient population with advanced cancer to identify HRQOL functional domains and symptoms which matter most for overall QOL.

Acknowledgments

This project is generously supported by the Michael and Karyn Goldstein Cancer Research Fund. We thank Stacy Yuen for secretarial assistance.

Conflict of Interest Statement

None

1. Kirkbride P, Tannock IF. Trials in palliative treatment—have the goal posts been moved?. Lancet Oncol. 2008;9(3):186-187.
   pubmed doi
2. Study protocol for the World Health Organization project to develop a Quality of Life assessment instrument (WHOQOL). Qual Life Res. 1993;2(2):153-159.
   pubmed
3. Movsas B. Quality of life in oncology trials: a clinical guide. Semin Radiat Oncol. 2003;13(3):235-247.
   pubmed doi
4. Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, Filiberti A, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993;85(5):365-376.
   pubmed doi
5. Riechelmann RP, Krzyzanowska MK, O'Carroll A, Zimmermann C. Symptom and medication profiles among cancer patients attending a palliative care clinic. Support Care Cancer. 2007;15(12):1407-1412.
   pubmed
6. Cooley ME. Quality of life in persons with non-small cell lung cancer: a concept analysis. Cancer Nurs. 1998;21(3):151-161.
   pubmed doi
7. Ostlund U, Wennman-Larsen A, Gustavsson P, Wengstrom Y. What symptom and functional dimensions can be predictors for global ratings of overall quality of life in lung cancer patients?. Support Care Cancer. 2007;15(10):1199-1205.
   pubmed
8. Groenvold M, Petersen MA, Aaronson NK, Arraras JI, Blazeby JM, Bottomley A, Fayers PM, et al. The development of the EORTC QLQ-C15-PAL: a shortened questionnaire for cancer patients in palliative care. Eur J Cancer. 2006;42(1):55-64.
   pubmed doi
9. Fayers PM, Aaronson N, Bjordal K, et al. The EORTC QLQ-C30 scoring manual. 3rd ed. Brussels: European Organization for Research and Treatment of Cancer; 2001.
10. Cohen, J. Statistical power analysis for the behavioral sciences. 2nd ed. New Jersey: Lawrence Erlbaum Associates, Inc., Publishers; 1988.
11. Gerrard GE, Prestwich RJ, Edwards A, Russon LJ, Richards F, Johnston CF, Kwok-Williams MC. Investigating the palliative efficacy of whole-brain radiotherapy for patients with multiple-brain metastases and poor prognostic features. Clin Oncol. 2003;15(7):422-428.
    pubmed doi
12. Roos DE, Wirth A, Burmeister BH, Spry NA, Drummond KJ, Beresford JA, McClure BE. Whole brain irradiation following surgery or radiosurgery for solitary brain metastases: mature results of a prematurely closed randomized Trans-Tasman Radiation Oncology Group trial (TROG 98.05). Radiother Oncol. 2006;80(3):318-322.
    pubmed
13. Beijer S, Kempen GI, Pijls-Johannesma MC, de Graeff A, Dagnelie PC. Determinants of overall quality of life in preterminal cancer patients. Int J Cancer. 2008;123(1):232-235.
    pubmed
14. Echteld MA, Deliens L, Onwuteaka-Philipsen B, Klein M, van der Wal G. EORTC QLQ-C15-PAL: the new standard in the assessment of health-related quality of life in advanced cancer?. Palliat Med. 2006;20(1):1-2.
    pubmed doi
15. Groenvold M, Petersen MA, Aaronson NK, Arraras JI, Blazeby JM, Bottomley A, Fayers PM, et al. EORTC QLQ-C15-PAL: the new standard in the assessment of health-related quality of life in advanced cancer?. Palliat Med. 2006;20(2):59-61.
    pubmed doi
16. Chow E, Bottomley A. Understanding the EORTC QLQ-BM22, the module for patients with bone metastases. Expert Rev Pharmacoecon Outcomes Res. 2009;9(5):461-465.
    pubmed doi
17. Chow E, Hird A, Velikova G, Johnson C, Dewolf L, Bezjak A, Wu J, et al. The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire for patients with bone metastases: the EORTC QLQ-BM22. Eur J Cancer. 2009;45(7):1146-1152.
    pubmed doi
18. Rosemann T, Hermann K, Miksch A, Engeser P, Szecsenyi J. The PAMINO-project: evaluating a primary care-based educational program to improve the quality of life of palliative patients. BMC Palliat Care. 2007;6:5.
    pubmed
19. Suarez-Del-Real Y, Allende-Perez S, Alferez-Mancera A, Rodriguez RB, Jimenez-Toxtle S, Mohar A, Onate-Ocana LF. Validation of the Mexican-Spanish version of the EORTC QLQ-C15-PAL questionnaire for the evaluation of health-related quality of life in patients on palliative care. Psychooncology. 2010;20(8):889-896.
    pubmed
20. Steinmann D, Schafer C, van Oorschot B, Wypior HJ, Bruns F, Bolling T, Sehlen S, et al. Effects of radiotherapy for brain metastases on quality of life (QoL). Prospective pilot study of the DEGRO QoL working party. Strahlenther Onkol. 2009;185(3):190-197.
    pubmed
21. Arnold R, Ranchor AV, Sanderman R, Kempen GI, Ormel J, Suurmeijer TP. The relative contribution of domains of quality of life to overall quality of life for different chronic diseases. Qual Life Res. 2004;13(5):883-896.
    pubmed
22. Rustoen T, Moum T, Padilla G, Paul S, Miaskowski C. Predictors of quality of life in oncology outpatients with pain from bone metastasis. J Pain Symptom Manage. 2005;30(3):234-242.
    pubmed doi
23. Harris K, Chow E, Zhang L, Velikova G, Bezjak A, Wu J, Barton M, et al. Patients' and health care professionals' evaluation of health-related quality of life issues in bone metastases. Eur J Cancer. 2009;45(14):2510-2518.
    pubmed doi
24. Larson DA, Rubenstein JL, McDermott MW. Metastatic Cancer to the Brain. In Lawrence TS, editor. DeVita VT, Lawrence TS, Rosenberg SA, DePinho RA, Weinberg RA, authors. Cancer: Principles & Practice of Oncology. Philadelphia: Lippincott Williams & Wilkins. 2008:p.2462-75.
25. Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life. A conceptual model of patient outcomes. JAMA. 1995;273(1):59-65.
    pubmed
26. Elmqvist MA, Jordhoy MS, Bjordal K, Kaasa S, Jannert M. Health-related quality of life during the last three months of life in patients with advanced cancer. Support Care Cancer. 2009;17(2):191-198.
    pubmed
27. Vogelzang NJ, Breitbart W, Cella D, Curt GA, Groopman JE, Horning SJ, Itri LM, et al. Patient, caregiver, and oncologist perceptions of cancer-related fatigue: results of a tripart assessment survey. The Fatigue Coalition. Semin Hematol. 1997;34(3 Suppl 2):4-12.
    pubmed
28. Churm D, Andrew IM, Holden K, Hildreth AJ, Hawkins C. A questionnaire study of the approach to the anorexia-cachexia syndrome in patients with cancer by staff in a district general hospital. Support Care Cancer. 2009;17(5):503-507.
    pubmed
29. Lis CG, Gupta D, Grutsch JF. Can anorexia predict patient satisfaction with quality of life in advanced cancer?. Support Care Cancer. 2009;17(2):129-135.
    pubmed
30. Bowling A. Just one question: If one question works, why ask several?. J Epidemiol Community Health. 2005;59(5):342-345.
    pubmed doi
31. Sloan JA, Aaronson N, Cappelleri JC, Fairclough DL, Varricchio C. Clinical Significance Consensus Meeting Group. Symposium on quality of life in cancer patients. Assessing the clinical significance of single items relative to summated scores. Mayo Clin Proc. 2002;77:479-87.

This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

World Journal of Oncology is published by Elmer Press Inc.