Psychometric validation of the functional assessment of cancer therapy brain (FACT-Br) for assessing quality of life in patients with brain metastases

Transcription

1 Support Care Cancer (2014) 22: DOI /s ORIGINAL ARTICLE Psychometric validation of the functional assessment of cancer therapy brain (FACT-Br) for assessing quality of life in patients with brain metastases Nemica Thavarajah & Gillian Bedard & Liying Zhang & David Cella & Jennifer L. Beaumont & May Tsao & Elizabeth Barnes & Cyril Danjoux & Arjun Sahgal & Hany Soliman & Edward Chow Received: 15 May 2013 /Accepted: 15 November 2013 /Published online: 28 November 2013 # Springer-Verlag Berlin Heidelberg 2013 Abstract Objective This study aimed to test the reliability, psychometric, and clinical validity of the use of the Functional Assessment of Cancer Therapy Brain (FACT-Br) in patients with brain metastases. Methods Patients with brain metastases were interviewed using the FACT-Br (including the FACT-general) 1 week prior to treatment. All patients completed a follow-up assessment 1 month post-treatment. Patients with a good performance status and receiving stereotactic radiosurgery completed an additional 1 week follow-up assessment after the initial baseline interview to assess test retest reliability. Results Forty patients had complete 1 month follow-up data. Ten of these patients also completed the 1 week follow-up assessment from baseline. The median Karnofsky performance status of patients was 80 and the median age was 64 years. All subscales of the FACT-Br were found to be conceptually related (except for two correlations) using the following subscales: physical well-being (PWB), social/ family well-being (SWB), emotional well-being (EWB), N. Thavarajah: G. Bedard : L. Zhang : M. Tsao : E. Barnes : C. Danjoux: A. Sahgal : H. Soliman : E. Chow Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada D. Cella: J. L. Beaumont Northwestern University, Chicago, IL, USA E. Chow (*) Department of Radiation Oncology, Odette Cancer Centre, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, Toronto, ON, Canada M4N3M5 Edward.Chow@sunnybrook.ca functional well-being (FWB), FACT-G total score, brain cancer subscale (BrC), and the FACT-Br total score. All FACT-Br scores demonstrated excellent reliability, except for the SWB scale which revealed good reliability. The FACT-Br scores showed no significant change in the quality of life (QoL) of patients from baseline to 1 month follow-up. Conclusion The use of the combined FACT-G and FACT-BrSubscaletoassessQoLspecificallyinpatients with brain metastases has successfully undergone psychometric validation. Future clinical trials should use the FACT-G and FACT-Br Subscale to assess QoL in this patient population. Keywords Quality of life. Brain metastases. Questionnaire. Psychometric validation. FACT-Br. FACT-G. Field-testing Introduction Brain metastases are a significant complication of advanced cancer, affecting approximately % of cancer patients during the course of their disease [1]. The most common primary sites of cancer from which brain metastases arise from include lung, breast, skin, kidney, and colon [2]. With the improvement in imaging modalities and screening techniques available, the detection and incidence of brain metastases continues to increase [3]. Newly diagnosed brain metastases patients often present with some degree of neuro-cognitive dysfunction, including symptoms such as speech impediment, seizures, weakness, headaches, and lack of muscle coordination, to name a few [2, 4, 5]. Various treatment modalities exist for the treatment of brain metastases and are ultimately targeted at improving quality of

2 1018 Support Care Cancer (2014) 22: life (QoL) and overall survival in patients. The treatment modality selected is often dependent on patient factors including the extent of extra-cranial disease, age, and functional status [3]. Such treatment modalities include therapeutic and symptomatic strategies. For instance, therapeutic strategies include stereotactic radiosurgery (SRS), whole brain radiation therapy (WBRT), neurosurgery, and chemotherapy [3, 6]. Symptomatic strategies include corticosteroids to reduce edema, and anticonvulsants to prevent seizures [3]. QoL is typically described as a subjective and multidimensional construct encompassing both physical and psychosocial factors [7]. Evaluating QoL in brain metastases patients using a comprehensive assessment tool can be valuable in determining a patient s current state of disease, and assessing the effectiveness of treatment over time. The Functional Assessment of Cancer Therapy Brain (FACT-Br) is a QoL instrument that was originally validated and intended for patients with primary brain tumors (Appendix I) [8]. The tumor-specific subscale is often administered alongside the FACT-General (FACT-G), with a combined total of 50 items for the full FACT-Br [9, 10]. Although the FACT-Br Subscale was originally developed for patients with primary brain tumors, it has been widely used in studies involving patients with brain metastases. Content validation of the FACT-Br Subscale with 50 patients with brain metastases and 37 healthcare professionals was recently conducted by Chen et al. The study revealed that both patients and health care professionals (HCPs) agreed that the FACT-Br Subscale items are relevant to patients with brain metastases, and required no further modification to the items on the questionnaire [9]. The purpose of this study was to test the reliability, clinical, and psychometric validity of the use of the FACT-Br Subscale (accompanied by the FACT-G) in patients with brain metastases through field-testing. A secondary objective included examining the change in the FACT-Br scores at 1 month after treatment to determine whether QoL in patients significantly changed over time. Methods Patient sample Patients over the age of 18 years with histologically proven cancer, had documented single or multiple brain metastases, and able to provide written informed consent were eligible for the study. Eligible patients were undergoing neurological resection (group A), SRS with or without WBRT (group B), or WBRT alone (group C). Patient demographic information was collected at baseline including gender, age, primary cancer site, number of brain metastases, Eastern Cooperative Oncology Group performance status (ECOG), Karnofsky Performance Score (KPS), and other variables listed in Table 1. Table 1 Patient characteristics (n =40 patients) Age (years) n 40 Mean±SD 61.8±10.5 Median (range) 64 (33 86) KPS n 36 Mean±SD 75.8±13.4 Median (range) 80 (40 90) KPS distribution 40 1 (2.78 %) 50 3 (8.33 %) 60 2 (5.56 %) 70 8 (22.22 %) (33.33 %) (27.78 %) Years from primary cancer to brain metastases n 37 Mean±SD 3.7±5.8 Median (range) 2 (0 28) Patients group A neurosurgery 2 (5 %) B radiosurgery w/wo WBRT 24 (60 %) C WBRT alone 14 (35 %) ECOG n 36 Mean 1 Median (range) 1 (0 3) 0 10 (27.78 %) 1 20 (55.56 %) 2 5 (13.89 %) 3 1 (2.78 %) 4 0 (0 %) Gender Female 31 (77.50 %) Male 9 (22.50 %) Primary cancer site Lung 23 (57.50 %) Breast 9 (22.50 %) Kidney 2 (5.00 %) Colon 1 (2.50 %) Melanoma 1 (2.50 %) Others 3 (7.50 %) Unknown 1 (2.50 %) Accrual Radiotherapy clinic 40 (100 %)

3 Support Care Cancer (2014) 22: Table 1 (continued) Out/in-patients Outpatient 40 (100%) Inpatient 0 (0%) Education level Elementary school 1 (2.63%) High school 19 (50.00%) College 3 (7.89%) University 12 (31.58%) Masters 2 (5.26%) PhD 1 (2.63%) Employment status Retired 20 (50.00%) Employed 13 (32.50%) Unemployed 7 (17.50%) Number of brain metastases 1 20 (51.28%) (17.94%) >3 12 (30.77%) Other site of metastases Bone 6 (16.22%) Lung 4 (10.81%) Liver 4 (10.81%) Lymph 5 (13.51%) Adrenal 1 (2.70%) None 24 (60.00%) Previous chemotherapy No 11 (28.95%) Yes 27 (71.05%) Previous hormone therapy No 33 (86.84%) Yes 5 (13.16%) Previous other therapies No 24 (60%) Yes 16 (40%) Baseline dexamethasone dose 16mg 1 (3%) 12mg 3 (8%) 8mg 5 (13%) 4mg 3 (8%) 2mg 1 (3%) None 26 (67%) Psychometric validation procedure Patients were interviewed in person by a trained research assistant using the FACT-Br within 1 week prior to their given treatment. Patients rated each of the 50 items on a scale of 0 ('not at all') to 4 ('very much') based on the past 7 days. All patients also completed a follow-up assessment approximately 1 month (4 to 6 weeks) following the initiation of the brainmetastases-specific treatment. The follow-up assessment was conducted using the FACT-Br over the phone. A timeframe of 1 month follow-up was selected because we acknowledged that a longer follow-up may be difficult to achieve in this population due to the potential rapid deterioration of this patient population. An additional assessment approximately 1 week after the initial baseline interview was completed by patients receiving SRS with a good performance status to assess the test retest reproducibility of the FACT-Br. Statistical methods Data were analyzed using Statistical Analysis Software (SAS version 9.3 for Windows). Descriptive statistics were used to convey demographic data; proportions for categorical variables; and mean, median, range, and standard deviation (SD) for continuous variables. FACT-Br scores were analyzed using the following subscales: physical well-being (PWB), social/family well-being (SWB), emotional wellbeing (EWB), functional well-being (FWB), FACT-G total score, brain cancer subscale (BrC), and the FACT-Br total score. Higher scores on each subscale indicate better quality of life. Multi-trait scaling analysis was used to examine whether the individual items of the FACT-Br can be aggregated into the aforementioned subscales. Evidence of item convergent validity was defined as a correlation of Reliability of the multi-item scales was assessed with Cronbach s alpha coefficient. Internal consistency estimates of 0.70 were considered acceptable for group comparisons [11]. Test retest reliability was assessed using intra-class correlation coefficients (ICC) where a value of 0.75 indicates excellent agreement, good agreement, fair to moderate agreement, and <0.4 poor agreement [11]. Convergent and divergent validity was examined by evaluating Pearson s product moment correlations between the FACT-Br scores where r 0.40 indicates that scales are likely to be conceptually related with one another [12]. Clinical validity was assessed as well to determine whether the questionnaire scores can discriminate between subgroups of patients based on clinical conditions such as KPS, prior treatments, number of brain metastases, and primary cancer site. These subgroups were selected because they present various patient outcomes. For instance, patients with advanced disease may present with numerous brain metastases. Primary cancer site may also play a factor in the development of brain metastases. Consequently, these measures are important in discriminating the results between these different subgroups. The Wilcoxon rank-sum or Kruskal Wallis (only for primary cancer site) nonparametric test was applied to explore which scale scores can discriminate between these subgroups of patients, where p <0.05 is considered statistically significant. Differences between baseline and 1 month follow-up assessments were also examined for significance (p <0.05) using the Wilcoxon signed rank test. Effect size estimates were also used to measure the magnitude of subgroup differences or association between baseline and 1 month follow-up. It is especially important because the effect size allow us to compare the magnitude of subgroup differences. The effect size (Cohen s d-statistic) was calculated as the subgroup s mean difference divided by the pooled standard deviation. Cohen proposed rules of thumb for interpreting effect size: a small effect size is 0.20, a medium effect size is 0.50, and a large effect size is 0.80 [13].

4 1020 Support Care Cancer (2014) 22: Results A total of 62 patients were initially enrolled in the study with complete baseline assessment. Of these patients, 40 completed the 1 month follow-up assessment and were included in the final analyses. Ten of these 40 patients also participated in the 1 week follow-up assessment after the initial interview. Of the 22 patients without 1 month follow-up assessment, six patients deceased within the study period, six patients withdrew from the study, and 10 patients were unable to be reached for follow-up. Patient demographics Patient characteristics are shown in Table 1. All patients, 78 % of whom were female, were outpatients accrued in a radiotherapy clinic. The median age of patients was 64 years old (range, 33 to 86 years), the median KPS was 80 (range, 40 to 90), and the median ECOG performance status was 1 (range, 0 to 3). In regards to treatment groups, 5 % of patients underwent neurological resection (group A), 60 % of patients received SRS with or without WBRT (group B), and 35 % of patients received WBRT alone (group C). The most common primary cancer sites included lung (58 %), breast (23 %), and kidney (5 %). Fifty-one percent of patients had only one brain metastasis and 60 % of patients had no other sites of metastasis. Patients with sites of metastases included the bone (16 %), lymph node (14 %), liver (11 %), lung (11 %), and adrenal gland (3 %). With regards to previous systemic treatment, 71 % of patients had prior chemotherapy, 13 % had prior hormone therapy, and 40 % had other prior therapies (e.g., radiation therapy). Sixty-seven percent of patients did not receive dexamethasone 24 h prior to baseline assessment. Forty-seven percent of patients had a level of education higher than high school. Multi-trait scaling analysis Multi-trait scaling analysis, as shown in Table 2,revealed that all items within a subscale correlated higher with items within its own subscale compared to other subscales at both baseline and at 1 month follow-up (all correlations 0.40 except for the BrC subscale). Cronbach s alpha coefficients were above 0.70 for all scales, except for EWB and BrC subscale scores at baseline or at 1 month. The analysis showed that the FACT-Br items could appropriately be aggregated into their corresponding subscales. FACT-Br scores Figures 1 and 2 show the mean FACT-G and FACT-Br subscale scores for patients at baseline and 1 month follow-up. Test retest reliability Test retest reliability was analyzed using data from the 10 patients with 1 week follow-up data (Table 3). All FACT-Br scores revealed excellent reliability (ICC 0.75), except for the SWB scale which revealed good reliability (ICC=0.67). Validity Correlations among the subscales of the FACT-Br were examined using Pearson s product moment correlation. In Table 4, all subscales appeared to be conceptually correlated with one Table 2 Multi-trait scaling analysis of FACT-G and FACT-Br subscale scores at baseline and 1 month follow-up (n =40 patients) a The absolute values are displayed for scale correlation Scale Item-own scale correlation a Item-other scale correlation a Baseline Physical well-being subscale score 0.48 to to Social/family well-being subscale score 0.63 to to Emotional well-being subscale score 0.37 to to Functional well-being subscale score 0.41 to to FACT-G total score 0.63 to to FACT-Br total score 0.56 to to BrC subscale score 0.07 to to One month follow-up Physical well-being subscale score 0.50 to to Social/family well-being subscale score 0.30 to to Emotional well-being subscale score 0.27 to to Functional well-being subscale score 0.37 to to FACT-G total score 0.62 to to FACT-Br total score 0.60 to to BrC subscale score 0.14 to to Cronbach s alpha

5 Support Care Cancer (2014) 22: Fig. 1 Mean FACT-G subscale scores for past 7 days at baseline and 1 month follow-up (n =40 patients) Baseline 1 Month Follow-up 5 0 Physical wellbeing Social/Family well-being Emotional wellbeing Functional wellbeing another (r 0.40) for baseline and 1 month follow-up except for two correlations at baseline: FWB with SWB (r =0.39) and FWB with EWB (r =0.38). Clinical validity was assessed at baseline and 1 month follow-up to explore which subscale scores can discriminate between subgroups of patients. In Table 5, the analyses revealed no significant difference in subscale scores when comparing patients with a good performance status (KPS>80) to patients with a poor performance status (KPS 80), except for PWB at 1 month follow-up (p = 0.034). More specifically, patients with a good performance status were more likely to have a higher state of PWB compared to patients with a poor performance status. The effect sizes were determined to compare the relative magnitude of the subgroup differences. The effect size for all subscales at baseline was small (absolute values from 0.05 to 0.34), and the effect size was medium at follow-up for all subscales (from 0.54 to 0.82) except SWB and EWB (effect size=0.25 and 0.36). Table 5 showed that patients without previous chemotherapy were more likely to have higher mean scores on PWB, EWB, and FACT-G total score at baseline (p <0.05). The effect size was large for PWB (0.85), EWB (1.12), and FACT-G total scores (0.90), medium for BrC (0.37), and small for FWB (0.34). Wilcoxon rank-sum test determined there were no significant differences among all subscales in patients with or without prior hormone therapy (p 0.05).The effect size was FACT-G Total FACT-Br Total BrC Baseline 1 Month Follow-up Fig. 2 Mean FACT-G total, FACT-Br total, and BrC subscale scores for past 7 days at baseline and 1 month follow-up (n =40 patients) large for almost all subscales (0.74 to 1.50) except for PWB (small; 0.39), and FWB (medium; 0.53). Having had other cancer treatments prior to baseline showed to have a significant difference for FWB, the FACT-G total score, BrC subscale, and the FACT-Br total score (p <0.05). The effect size was large for the FWB (1.07), FACT-G total (0.85), FACT-Br total (1.12), and BrC subscale (1.21). Patients with one brain metastasis were more likely to have higher mean PWB score at baseline (p =0.04) compared to patients with more than one metastasis. The effect size was medium for all subscales (0.41 to 0.74) except SWB (absolute value of 0.11) and EWB (small; 0.11). In regards to primary cancer site, patients with breast or lung cancer were more likely to have lower average EWB or BrC scores at baseline (p =0.02) compared to other primary cancer sites. The effect size was small for all subscales (0.02 to 0.27) except FACT-G total (0.47), SWB (0.62), and EWB (1.05). Change in scores over time The Wilcoxon signed rank test revealed no significant changes in the FACT-Br subscales over time from baseline to 1 month follow-up (p 0.05). Discussion This is the first study to conduct a psychometric validation of the FACT-Br (including the FACT-G) to assess QoL specifically in patients with brain metastases. Our study revealed that the combined FACT-G and FACT-Br Subscale can be administered to this specific patient population, despite the fact that the instrument was originally designed for patients with a primary brain tumor. Other QoL instruments such as the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Brain Neoplasm (EORTC QLQ- BN20) used in conjunction with the QLQ-C30 [9, 10], were also originally validated for patients with a primary brain tumor. There is currently no single widely accepted QoL instrument

6 1022 Support Care Cancer (2014) 22: Table 3 Test retest reliability of the FACT-G and FACT-Br subscales using intra-class correlation (ICC) between baseline and retest assessments 1 week after the initial baseline assessment (n =10 patients) FACT-Br scale at post 7 days ICC 95 % confidence interval Interpretation Physical well-being subscale score Excellent Social/family well-being subscale score Good Emotional well-being subscale score Excellent Functional well-being subscale score Excellent FACT-G total score Excellent FACT-Br total score Excellent BrC subscale score Excellent for patients with brain metastases; however, the FACT-Br Subscale has been frequently used in past clinical studies [9]. A recent literature review identified a total of 13 studies that utilized the FACT-Br to assess QoL in patients with brain metastases. The types of treatments patients were receiving in these studies varied from study to study, but were fairly consistent with the types of treatments that patients received within our study (i.e., SRS, WBRT, and neurosurgery among others). Furthermore, follow-up intervals ranged from 1 month after treatment to up to over 18 months. Overall the studies showed statistically significant improvements in QoL among certain scales, while others showed no improvement, which is similar to the results of our study as well. The overall findings of this review were favorable towards the use of the FACT-Br to assess QoL in patients with brain metastases and recommended formal validation of the tool [14]. QoL is generally expected to decline over time as a patient s disease progresses and causes changes in neuro-cognitive function [15, 16]. The 13 studies included in the recent FACT-Br literature review revealed varied results among subscales over varying timeframes, making it difficult to confidently assess whether the FACT-Br Subscale accurately detects change in patient QoL over time. Our study in particular revealed no significant changes in the FACT-G and FACT-Br subscale scores over time from baseline to 1 month follow-up. Similar results were found in a study conducted by Wong et al. involving the Spitzer Quality of Life Index (SQLI), which is a validated QoL instrument used in the same patient population. The authors used the SQLI to assess QoL in brain metastases patients for up to 3 months post radiation therapy treatment. The results of the study showed that 47 % of patients had overall improved or stable neuro-cognitive function over time. The patient cohort involved in the study had a primary lung or breast cancer for the majority of patients, a median KPS of 70, and a median age of 64, all of which reflect the demographics of our patient sample [17]. Considering the already validated SQLI shows similar results to our study, it suggests these patients did not change in a meaningful way over the course of the 1 month follow-up. The characteristics of our patient sample could have potentially contributed to the reason why patients did not experience a significant change in QoL over time. For instance, the median Table 4 Pearson correlation among FACT-G and FACT-Br subscales at baseline and 1 month follow-up (n =40 patients) Subscale PWB SWB EWB FWB FACT-G FACT-Br BrCS At baseline Physical well-being subscale score (PWB) 1.00 Social/family well-being subscale score (SWB) Emotional well-being subscale score (EWB) 0.54 a 0.46 a 1.00 Functional well-being subscale score (FWB) 0.65 a FACT-G total score (FACT-G) 0.85 a 0.63 a 0.76 a 0.83 a 1.00 FACT-Br total score (FACT-Br) 0.83 a 0.56 a 0.64 a 0.85 a 0.95 a 1.00 BrC subscale score (BrCS) 0.68 a 0.40 a 0.41 a 0.75 a 0.76 a 0.93 a 1.00 At month 1 Physical well-being subscale score (PWB) 1.00 Social/family well-being subscale score (SWB) Emotional well-being subscale score (EWB) 0.45 a 0.52 a 1.00 Functional well-being subscale score (FWB) 0.53 a 0.42 a 0.50 a 1.00 FACT-G total score (FACT-G) 0.76 a 0.62 a 0.80 a 0.84 a 1.00 FACT-Br total score (TOTAL) 0.75 a 0.59 a 0.74 a 0.80 a 0.96 a 1.00 BrC subscale score (BrCS) 0.66 a 0.52 a 0.61 a 0.69 a 0.83 a 0.95 a 1.00 a The absolute Pearson correlation r 0.40

7 Support Care Cancer (2014) 22: Table 5 Subgroup comparisons on FACT-G and FACT-Br subscales at baseline (n =40 patients) Subscale Subgroup mean (SD) p value a KPS 80 at baseline KPS>80 at baseline (n =26) (n =10) Physical well-being subscale score 21.0 (4.9) 21.9 (5.2) Social/family well-being subscale score 26.4 (3.3) 25.2 (4.5) Emotional well-being subscale score 16.6 (5.2) 17.1 (5.0) Functional well-being subscale score 19.9 (5.2) 18.6 (8.0) FACT-G total score 83.5 (14.0) 82.7 (20.6) FACT-Br total score (23.4) (34.9) BrC subscale score 72.7 (10.9) 74.8 (15.3) KPS 80 at month 1 KPS>80 at month 1 (n =26) (n =10) Physical well-being subscale score 17.8 (6.6) 22.8 (4.4) Social/family well-being subscale score 25.7 (5.0) 26.8 (2.3) Emotional well-being subscale score 16.2 (5.7) 18.2 (5.9) Functional well-being subscale score 16.4 (7.3) 21.0 (6.7) FACT-G total score 76.1 (18.5) 88.8 (16.7) FACT-Br total score (33.8) (31.0) BrC subscale score 68.2 (16.7) 77.1 (16.1) Without previous chemotherapy With previous chemotherapy (n =11) (n =27) Physical well-being subscale score 23.9 (3.9) 19.4 (5.8) Social/family well-being subscale score 27.5 (1.1) 25.6 (4.2) Emotional well-being subscale score 20.5 (3.1) 15.4 (5.0) Functional well-being subscale score 21.0 (5.4) 18.9 (6.6) FACT-G total score 92.9 (10.8) 79.0 (16.9) FACT-Br total score (16.8) (31.0) BrC subscale score 75.2 (6.7) 70.2 (15.5) Without previous hormotherapy With previous hormotherapy (n =33) (n =5) Physical well-being subscale score 21.0 (5.9) 18.8 (4.5) Social/family well-being subscale score 26.8 (2.0) 22.0 (7.8) Emotional well-being subscale score 17.3 (4.8) 13.6 (6.3) Functional well-being subscale score 20.0 (6.5) 16.6 (4.8) FACT-G total score 84.8 (15.6) 70.9 (19.3) FACT-Br total score (27.6) (30.1) BrC subscale score 73.2 (13.2) 61.6 (13.5) Without other therapy With other therapy (n =24) (n =16) Physical well-being subscale score 19.7 (6.4) 22.7 (3.5) Social/family well-being subscale score 25.7 (4.2) 27.2 (1.9) Emotional well-being subscale score 16.2 (5.2) 17.7 (4.6) Functional well-being subscale score 16.9 (6.3) 23.0 (4.5) FACT-G total score 78.0 (17.3) 90.8 (10.8) FACT-Br total score (28.5) (17.7) BrC subscale score 66.1 (13.2) 80.2 (8.7) <0.001 One brain metastasis >1 brain metastasis (n =20) (n =19) Physical well-being subscale score 22.9 (4.4) 19.0 (6.2) Social/family well-being subscale score 26.4 (3.7) 26.8 (2.0) Emotional well-being subscale score 17.2 (5.5) 16.6 (4.4) 0.588

8 1024 Support Care Cancer (2014) 22: Table 5 (continued) Subscale Subgroup mean (SD) p value a Functional well-being subscale score 21.2 (6.2) 17.4 (6.1) FACT-G total score 87.7 (16.6) 79.8 (14.8) FACT-Br total score (27.8) (27.9) BrC subscale score 74.4 (11.5) 68.9 (15.3) Primary cancer site Breast Lung Kidney Other (n =9) (n =23) (n =2) (n =6) Physical well-being subscale score 18.6 (5.5) 21.5 (5.3) 22.6 (0.6) 21.4 (7.9) Social/family well-being subscale score 23.5 (6.1) 26.8 (1.9) 28.0 (0.1) 28.0 (0.1) Emotional well-being subscale score 13.8 (4.8) 16.6 (4.7) 24.0 (0.1) 19.5 (3.1) Functional well-being subscale score 17.1 (5.4) 20.4 (6.0) 18.1 (5.8) 18.9 (9.2) FACT-G total score 73.0 (17.1) 85.1 (15.0) 92.7 (5.2) 87.7 (17.4) FACT-Br total score (28.9) (23.4) (15.5) (35.7) BrC subscale score 61.4 (13.8) 75.7 (9.5) 75.3 (10.3) 70.9 (20.3) a p value was obtained by Wilcoxon rank-sum test or Kruskal-Wallis test (only for primary cancer site). p <0.05 was considered as statistical significance KPS of our patient sample was 80, which indicates the patient shows some signs of symptoms or disease and can carry out normal activity with effort [18]. In addition, approximately 67 % of patients were not receiving dexamethasone at baseline, which could also indicate better overall performance status. Dexamethasone is commonly prescribed in patients with signs of neurological decline, cerebral edema, and poor overall disease status [19]. Furthermore, 60 % of patients were receiving SRS, which often does not adversely affect overall patient performance status and preserves neuro-cognitive function [20]. All patients accrued were outpatients, which could also indicate a better performance status. Consequently, the characteristics of our patient sample could be another possible explanation as to why patients showed no significant change in QoL over time. Patients with a better performance status at baseline may not show as much progression in their disease after 1 month in comparison to those who started with a poor performance status at baseline. The FACT-G and FACT-Br Subscale were well-received overall by patients; however, limitations still exist within this research design which could potentially be improved for future efforts. For instance, our patient sample consisted only of outpatients with relatively good performance status. Future studies may benefit from including a more diverse patient population including inpatients and patients with poorer performance status to confirm that the FACT-Br accurately assesses QoL in these patients. Another limitation within our study design was that KPS was not assessed at 1 month follow-up since the follow-up assessments were conducted over the phone. Assessing the KPS of patients in person at follow-up could allow for the analysis of patient groups defined by changes in KPS over time. For instance, if the analysis revealed that the KPS of patients did not significantly change over time, this could further confirm the lack of change over time seen here. In the results of our statistical analyses, the high correlation of the FACT-Br total score may have been influenced by the FACT-G subscale being included in the FACT-Br total score subscale. Future studies may explore the possibility of having patients rate the items of the FACT-Br, based on a timeframe shorter than the past 7 days. For instance, patients with poorer performance status or signs of neuro-cognitive decline may have difficulty recalling the past 7 days, and may benefit from rating the items based on a shorter timeframe (i.e., past 24 h). Furthermore, future studies could also explore whether the FACT-Br can be used to accurately detect changes in QoL beyond 1 month post-treatment. Conclusion Evaluating QoL in patients with brain metastases continues to be an important aspect of palliative care in terms of assessing the efficacy of treatment and overall disease state. The FACT- Br Subscale was originally validated to assess QoL in patients with a primary brain tumor. In this study, we have successfully completed the psychometric validation of the FACT-Br Subscale (and the FACT-G) to be used for patients with brain metastases. Future clinical trials can continue to use the FACT-G and FACT-Br Subscale to assess QoL in this patient population. Acknowledgments We thank the generous support of the Bratty Family Fund, Michael and Karyn Goldstein Cancer Research Fund, Joseph and Silvana Melara Cancer Research Fund, and the Ofelia Cancer Research Fund. Conflicts of interest None.

9 Support Care Cancer (2014) 22: Appendix I FACT-Br VERSION 4 Below is a list of statements that other people with your illness have said are important. Please circle or mark one number per line to indicate your response as it applies to the past 7 days. Br1 Br2 Br3 Br4 Br5 Br6 Br7 NTX6 Br8 Br9 Br10 Br11 Br12 Br13 Br14 Br15 Br16 Br17 Br18 Br19 Br20 Br21 An10 ADDITIONAL CONCERNS Not at all A little bit Somewhat Quite a bit I am able to concentrate I have had seizures (convulsions) I can remember new things I get frustrated that I cannot do things I used to I am afraid of having a seizure (convulsion) I have trouble with my eyesight I feel independent I have trouble hearing I am able to find the right word(s) to say what I mean I have difficulty expressing my thoughts I am bothered by the change in my personality I am able to make decisions and take responsibility I am bothered by the drop in my contribution to the family I am able to put my thoughts together I need help in caring for myself (bathing, dressing, eating, etc.) I am able to put my thoughts into action I am able to read like I used to I am able to write like I used to I am able to drive a vehicle (my car, truck, etc.) I have trouble feeling sensations in my arms, hands, or legs I have weakness in my arms or legs I have trouble with coordination I get headaches Very much

10 1026 Support Care Cancer (2014) 22: FACT-G VERSION 4 Below is a list of statements that other people with your illness have said are important. Please circle or mark one number per line to indicate your response as it applies to the past 7 days. PHYSICAL WELL-BEING Not at all A little bit Somewhat Quite a bit Very much GP1 GP2 GP3 GP4 GP5 GP6 GP7 I have a lack of energy I have nausea Because of my physical condition, I have trouble meeting the needs of my family I have pain I am bothered by side effects of treatment I feel ill I am forced to spend time in bed SOCIAL/FAMILY WELL-BEING Not at all A little bit Somewhat Quite a bit Very much GS1 GS2 GS3 GS4 GS5 GS6 Q1 I feel close to my friends I get emotional support from my family I get support from my friends My family has accepted my illness I am satisfied with family communication about my illness I feel close to my partner (or the person who is my main support) Regardless of your current level of sexual activity, please answer the following question. If you prefer not to answer it, please mark this box and go to the next section. GS7 I am satisfied with my sex life

11 Support Care Cancer (2014) 22: Please circle or mark one number per line to indicate your response as it applies to the past 7 days. EMOTIONAL WELL-BEING Not at all A little bit Somewhat Quite a bit Very much GE1 GE2 GE3 GE4 GE5 GE6 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 feel nervous I worry about dying I worry that my condition will get worse GF1 GF2 GF3 GF4 GF5 GF6 GF7 FUNCTIONAL WELL-BEING Not at all A little bit Somewhat Quite a bit I am able to work (include work at home) My work (include work at home) is fulfilling I am able to enjoy life I have accepted my illness I am sleeping well I am enjoying the things I usually do for fun I am content with the quality of my life right now Very much References 1. Doyle M, Bradley NM, Li K, Sinclair E, Lam K, Chan G et al (2007) Quality of life in patients with brain metastases treated with a palliative course of whole-brain radiotherapy. J Palliat Med 10(2): Thomas SS, Dunbar EM (2010) Modern multidisciplinary management of brain metastases. Curr Oncol Rep 12(1): Eichler AF, Loeffler JS (2007) Multidisciplinary management of brain metastases. Oncologist 12(7): Chang EL, Wefel JS, Maor MH, Hassenbusch SJ III, Mahajan A, Lang FF et al (2007) A pilot study of neurocognitive function in patients with one to three new brain metastases initially treated with stereotactic radiosurgery alone. Neurosurgery 60(2): , discussion Posner JB (1977) Management of central nervous system metastases. Semin Oncol 4(1): Tsao MN, Lloyd N, Wong R, Chow E, Rakovitch E, Laperriere N (2006) Whole brain radiotherapy for the treatment of multiple brain metastases. Cochrane Database Syst Rev 3: Movsas B (2003) Quality of life in oncology trials: a clinical guide. Semin Radiat Oncol 13(3): Weitzner MA, Meyers CA, Gelke CK, Byrne KS, Cella DF, Levin VA (1995) The Functional Assessment of Cancer Therapy (FACT) scale. Development of a brain subscale and revalidation of the general version (FACT-G) in patients with primary brain tumors. Cancer 75(5): Chen E, Cella D, Zeng L, Thavarajah N, Zhang L, Chang E, et al. Content validation of the FACT-Br with patients and health-care professionals to assess quality of life in patients with brain metastases. Journal of Radiation Oncology Caissie A, Nguyen J, Chen E, Zhang L, Sahgal A, Clemons M et al (2012) Quality of life in patients with brain metastases using the EORTC QLQ-BN20+2 and QLQ-C15-PAL. Int J Radiat Oncol Biol Phys 83(4): Cicchetti DV, Sparrow SA (1981) Developing criteria for establishing interrater reliability of specific items: applications to assessment of adaptive behavior. Am J Ment Defic 86(2): Nunnally JC, Bernstein IH (1994) Psychometric theory, 3rd edn. Mc Graw-Hill, New York 13. Cohen J (1988) Statistical power analysis for the behavioral science. Lawrence Erlbaum Associates, New Jersey 14. Lien K, Zeng L, Nguyen J, Cramarossa G, Cella D, Chang E et al (2011) FACT-Br for assessment of quality of life in patients receiving treatment for brain metastases: a literature review. Expert Rev Pharmacoecon Outcome Res 11(6): Li J, Bentzen SM, Li J, Renschler M, Mehta MP (2008) Relationship between neurocognitive function and quality of life after whole-brain radiotherapy in patients with brain metastasis. Int J Radiat Oncol Biol Phys 71(1): Bezjak A, Adam J, Barton R, Panzarella T, Laperriere N, Wong CS et al (2002) Symptom response after palliative radiotherapy

12 1028 Support Care Cancer (2014) 22: for patients with brain metastases. Eur J Cancer 38(4): Wong J, Hird A, Zhang L, Tsao M, Sinclair E, Barnes E et al (2009) Symptoms and quality of life in cancer patients with brain metastases following palliative radiotherapy. Int J Radiat Oncol Biol Phys 75(4): Friendlander AH, Ettinger RL (2009) Karnofsky performance status scale. Spec Care Dent 29(4): Hempen C, Weiss E, Hess CF (2002) Dexamethasone treatment in patients with brain metastases and primary brain tumors: do the benefits outweigh the side-effects? Support Care Cancer 10(4): Tsao M, Xu W, Sahgal A (2012) A meta-analysis evaluating stereotactic radiosurgery, whole-brain radiotherapy, or both for patients presenting with a limited number of brain metastases. Cancer 118(9):