Quality of life measurement in cancer patients receiving palliative radiotherapy for symptomatic lung cancer: a literature review.

Approximately 27% of North American cancer deaths are attributable to cancer of the lung. Many lung cancers are found at an advanced stage, rendering the tumours inoperable and the patients palliative. Common symptoms associated with palliative lung cancer include cough, hemoptysis, and dyspnea, all of which can significantly debilitate and diminish quality of life (QOL). In studies of the effects of cancer therapies, the frequent evaluative endpoints are survival and local control; however, it is imperative that clinical trials with palliative patients also have a QOL focus when a cure is unattainable. We conducted a literature review to investigate the use of QOL instrument tools in trials studying QOL or symptom palliation of primary lung cancer or lung metastases through the use of radiotherapy. We identified forty-three studies: nineteen used a QOL tool, and twenty-four examined symptom palliation without the use of a QOL instrument. The European Organization for Research and Treatment of Cancer (eortc) QLQ-C30 survey was the most commonly used QOL questionnaire (in thirteen of twenty trials). Of those thirteen studies, eight also incorporated the lung-specific QOL survey eortc QLQ-LC13 (or the eortc QLQ-LC17). A second lung-specific survey, the Functional Assessment of Cancer Therapy-Lung (fact-L) was used in only two of the twenty trials. In total, only ten of forty-three trials (23%) used a lung-specific QOL tool, suggesting that QOL was of low priority as an endpoint and that measures created for lung cancer patients are underused. We encourage investigators in future trials to include specific QOL instruments such as the eortc QLQ-LC13 or the fact-L for studies in palliative thoracic radiotherapy because those instruments provide a measure of QOL specific to patients with lung cancer or lung metastases.


INTRODUCTION
Lung cancer is a rising epidemic and remains the leading cause of cancer death in both men and women in Canada 1 . In general, 500 Canadians are diagnosed with and 400 Canadians die of lung cancer every week 1 . Such high morbidity and mortality in patients with primary lung cancer emphasizes the need for palliative treatment intent.
Morbidity from lung cancer or lung metastases often presents as troublesome thoracic symptoms such as hemoptysis, cough, chest pain, and dyspnea. Palliative radiotherapy has been effective in ameliorating these symptoms [2][3][4] and improves or preserves the quality of life (qol) remaining in approximately one third of affected patients 5 .
In the past, clinical trials in patients with lung cancer have focused on traditional endpoints such as overall survival, disease-free survival, or local control 6 . Given the relatively poor prognosis of patients with locally advanced lung cancer or lung metastases, the inclusion of qol as a primary endpoint of treatment becomes increasingly important. Quality of life encompasses the minimization of risks and maximization of benefits of a treatment, including physical and psychosocial effects on the well-being of patients 7 . Studying qol is particularly relevant in the field of palliative radiotherapy because of known treatment-related side effects and toxicities.
Few studies focus on qol and symptom palliation as primary endpoints. The purpose of the present review was to accurately assess the recent use of qol tools in trials that evaluated the efficacy of palliative radiotherapy in patients with lung cancer or lung metastases.

Search Strategy
We conducted a literature review using the medline (Ovid) database for 1950 to February 2008. Key terms such as "lung cancer," "lung neoplasms," or "lung metastases" were combined with the terms ABSTRACT Approximately 27% of North American cancer deaths are attributable to cancer of the lung. Many lung cancers are found at an advanced stage, rendering the tumours inoperable and the patients palliative. Common symptoms associated with palliative lung cancer include cough, hemoptysis, and dyspnea, all of which can significantly debilitate and diminish quality of life (qol). In studies of the effects of cancer therapies, the frequent evaluative endpoints are survival and local control; however, it is imperative that clinical trials with palliative patients also have a qol focus when a cure is unattainable. We conducted a literature review to investigate the use of qol instrument tools in trials studying qol or symptom palliation of primary lung cancer or lung metastases through the use of radiotherapy. We identified forty-three studies: nineteen used a qol tool, and twenty-four examined symptom palliation without the use of a qol instrument. The European Organization for Research and Treatment of Cancer (eortc) qlq-C30 survey was the most commonly used qol questionnaire (in thirteen of twenty trials). Of those thirteen studies, eight also incorporated the lung-specific qol survey eortc qlq-LC13 (or the eortc qlq-LC17). A second lung-specific survey, the Functional Assessment of Cancer Therapy-Lung (fact-L) was used in only two of the twenty trials. In total, only ten of forty-three trials (23%) used a lung-specific qol tool, suggesting that qol was of low priority as an endpoint and that measures created for lung cancer patients are underused. We encourage investigators in future trials to include specific qol instruments such as the eortc qlq-LC13 or the fact-L for studies in palliative thoracic radiotherapy because those instruments provide a measure of qol specific to patients with lung cancer or lung metastases.

KEY WORDS
Lung cancer, quality of life, qol instrument, review, fact-L, eortc qlq-LC13, eortc qlq-C30 Current OnCOlOgy-VOlume 16, number 2 "radiotherapy," "radiation," "external-beam irradiation," or "palliative radiotherapy." This search was then combined with "quality of life" or "qol" and also "symptom palliation." Relevant articles and abstracts were reviewed, and references from those sources were also manually searched for additional relevant publications.

Inclusion Criteria
To be included in the present literature review, articles had to meet these criteria: • Population: patients with a histologic, cytologic, or radiologic diagnosis of primary lung cancer or lung metastases • Intervention: external beam radiotherapy or endobronchial brachytherapy in at least one study arm, with palliative intent • Types of studies: randomized trials, prospective or retrospective cohort studies • Endpoints: qol or symptom palliation as a primary or secondary endpoint or measured outcome

Exclusion Criteria
Articles were excluded if they met any of these criteria: • Article type: individual case report or review article • Language: publication in a language other than English • Intervention: no evaluation, in at least one arm, of external beam irradiation to the thorax or endobronchial brachytherapy; or studies of interventions with curative intent • Types of studies: focus on populations other than those with primary lung cancer or lung metastases • Endpoints: use of the Karnofsky performance status (kps) or other similar prognostic tools, correlation of qol with cost-utility, or test of the reliability or validity of a qol instrument

Data Extraction
We extracted the following information from the studies: • Primary and secondary outcomes • Radiotherapy treatment details • Type and number of qol, symptom palliation, and additional tools, if any, used • Number of patients in each study arm • Median age and male:female ratio of the patients enrolled in the study • Median survival in each study arm

RESULTS
We identified a total of forty-three trials that evaluated, in at least one study arm, the use of palliative radiotherapy to the thorax, and that assessed qol or symptom palliation as a primary or secondary endpoint. Thirty studies (Table i) evaluated the treatment of patients with non-small-cell lung cancer (nsclc). Four studies (Table ii) involved patients who were treated with endobronchial brachytherapy alone or in addition to external-beam radiation. Brachytherapy differs from external-beam radiation in that it is a more localized form of radiation that limits toxicity in healthy tissue to the immediate vicinity of the radiated region 5 . Another nine trials (Table iii) evaluated the use of palliative radiotherapy in patients with lung cancer of a histologic type other than nsclc. The four identified studies that measured the difference in efficacy between endobronchial brachytherapy and external beam radiation 37-40 used both symptom palliation and qol scores as a primary outcome.

QOL and Symptom Palliation Tools Used
A total of 11 tools were used to assess either qol or palliation of lung cancer-related symptoms; the frequency of use of each tool is presented in Table iv. The most common qol tool used was the European Organization for Research and Treatment of Cancer (eortc) qlq-C30, a questionnaire that was created and validated to assess qol in individuals with any form of cancer. It has been translated into 81 languages and consists of 30 questions that encompass 5 functional scales: physical, role, cognitive, emotional, and social functioning 49 . The eortc qlq-C30 also incorporates 3 symptom scales: fatigue, pain, and nausea and vomiting. The remaining items on the questionnaire cover other symptom-related events that are often described by cancer patients, including dyspnea, diarrhea, and loss of appetite, among others 48 .
The eortc qlq-C30 was used in fourteen of the forty-three studies identified in the search (32%), eight of which also used the lung cancer supplement, eortc qlq-LC13. The eortc qlq-LC13 is the latest version of a lung cancer-specific questionnaire that consists of questions concerning lung cancer symptoms  Current OnCOlOgy-VOlume 16, number 2 and the side effects of conventional treatments used for lung cancer 49 . One trial used an older version of the lung-specific module, the eortc qlq-LC17, in addition to the general questionnaire 48 . The Functional Assessment of Cancer Therapy (fact) qol tools constituted a second group used in the identified studies. Both the general questionnaire (fact-G) and the lung-specific questionnaire (fact-L) were used. Like the eortc qlq-C30, the fact-G is a general questionnaire that was developed for patients with any type of cancer. The fact-G covers 4 dimensions of qol: physical, social, emotional, and functional well-being 50 . The fact-L is similar to the eortc qlq-LC13 because it includes additional questions that relate specifically to qol in patients with lung cancer. The fact-L was used in two studies, and the fact-G in one.
A third validated qol tool was used in one trial: the Spitzer qol Index. The Spitzer Index covers 5 dimensions of qol: activity, daily living, health, support of family and friends, and outlook 51 . It is not a lung cancer-specific questionnaire, however; and thus it does not incorporate questions directly related to the lung-cancer-specific patient population.
Study-designed questionnaires were the most prevalent tool used in the forty-three identified studies. A study-specific method of determining qol was used in three trials, and nineteen trials attempted to evaluate symptom palliation using a study-designed questionnaire. Table v shows a breakdown of the proportion of studies using a validated qol or symptom palliation tool as compared with a study-designed tool. Study-designed instruments present a difficulty: drawing comparisons across studies is harder because the methods of measurement vary.
In five studies, a validated symptom palliation tool was used (the frequency of use can be seen in Table iv). The two general symptom tools used were the Hospital Anxiety and Depression Scale and the Rotterdam Symptom Checklist. The Rotterdam Symptom Checklist measures psychological and physical distress in cancer patients through the use of 38 items 52 . The Hospital Anxiety and Depression Scale is a tool used to measure anxiety and depression levels using 14 statements based on a patient's experience over the preceding week 53 . One lungspecific symptom tool-the Lung Cancer Symptom Scale-was used. The Lung Cancer Symptom Scale is a tool designed to measure 6 lung-specific symptoms and their effects on symptomatic distress, functional burden, and global quality of life 54,55 . Figure 1 outlines the overall picture of questionnaire use in the identified trials. Most of the trials (54%) measured symptom palliation alone; some measured both symptom palliation and qol (14%). The remaining trials measured qol only.

Performance Assessment
In forty studies (91%), the performance status of the subjects was measured in addition to qol or symptom palliation. Performance status was measured primarily as a prognostic factor (twenty of forty trials, 50%) or as part of the exclusion criteria (fourteen of forty trials, 35%). Only six studies used a performance

DISCUSSION
In patients with terminal cancer, qol is a significant concept, and it is influenced by many factors, including symptoms, functional level, coping strategies, and support systems 51 . Common symptoms that influence a lung cancer patient's qol include anxiety, depression, pain, fatigue, dyspnea, and cough 52 . Because lung cancer is the leading cause of cancer death in men and the second-leading cause in women globally 2 , it is important that qol is considered when caring for these patients.
Meaningful palliation refers to symptom relief and prolongation of good-quality survival in lung cancer patients 26 . When treating a patient with palliative intent, it is necessary to use tools that measure the intent of the treatment. For 86% of doctors from the United Kingdom, the United States, and Canada, the treatment of choice for patients with inoperable lung cancer is palliative radiotherapy 33 . It is therefore important that, when considering the side effects of palliative radiotherapy as compared with the side effects of the lung cancer itself, trials investigating the use of palliative radiotherapy use a qol measure to determine the benefit of the treatment.
A total of twenty identified trials considering palliative radiotherapy for lung cancer included an evaluation of qol. Of these trials, eleven used a tool that was specific to patients with lung cancer; the remaining nine used general qol questionnaires for cancer patients or a study-designed questionnaire. In thirty-one identified studies, the level of symptom palliation, one aspect that contributes to a qol measure, was assessed. This finding suggests that more trials should use a validated lung-specific tool when evaluating the outcome of palliative thoracic radiotherapy. Use of a validated, lung-specific tool will allow for comparisons between trials and will also increase the internal validity of individual studies. Two recommended lung-specific validated tools that would be beneficial for the measurement of qol in trials evaluating palliative thoracic radiotherapy are the fact-L and the eortc qlq-LC13.

ACKNOWLEDGMENT
This project was generously supported by the Michael and Karyn Goldstein Cancer Research Fund. We thank Ms. Stacy Lue for secretarial service.