Marc Kerba

Neuropathic Pain Features in Patients With Bone Metastases Referred for Palliative Radiotherapy

PURPOSE To estimate the prevalence of pain with neuropathic features among patients with metastatic bone pain and to assess differences between patients with and without neuropathic features by pain severity, functional interference, and quality-of-life (QOL) measures. PATIENTS AND METHODS A prospective cross-sectional survey of consecutive patients with symptomatic bone metastases was conducted between December 2006 and March 2008 at a comprehensive cancer center. Patients completed the Brief Pain Inventory (BPI), the Self-Reported Leeds Assessment of Neuropathic Symptoms and Signs (S-LANSS), and the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire C30 (EORTC QLQ-C30). Statistical associations between pain with neuropathic features and other measures were explored. RESULTS Ninety-eight patients were enrolled. Seventeen percent of patients (95% CI, 10% to 24%) had positive S-LANSS scores suggesting pain with neuropathic features. Mean worst pain and mean interference scores were 7.2 (standard deviation [SD], 2.0) and 5.8 (SD, 2.5), respectively. EORTC QLQ-C30 global QOL, function, and symptom scores were 42 (SD, 24), 52 (SD, 20), and 46 (SD, 17), respectively. Patients with neuropathic features had a higher BPI worst pain score than patients without neuropathic features (8.3 v 7.0, respectively; P = .016). Corticosteroid use, oral morphine equivalent dosing, and site of bone pain were not associated with neuropathic features. CONCLUSION Some patients with bone metastases manifest bone pain with distinguishable neuropathic features, and these patients reported greater pain intensity. Additional work is required to validate the S-LANSS against clinical criteria for neuropathic pain in this context and to explore the unmet pain management needs in this population.

Quality of Life in Patients With Brain Metastases Using the EORTC QLQ-BN20+2 and QLQ-C15-PAL

PURPOSE The 20-item European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Brain Neoplasm (QLQ-BN20) is a validated quality-of-life (QOL) questionnaire for patients with primary brain tumors. The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 15 Palliative (QLQ-C15-PAL) core palliative questionnaire is a 15-item version of the core 30-item QLQ-C30 and was developed to decrease the burden on patients with advanced cancer. The combination of the QLQ-BN20 and QLQ-C30 to assess QOL may be too burdensome for patients. The primary aim of this study was to assess QOL in patients before and after treatment for brain metastases using the QLQ-BN20+2 and QLQ-C15-PAL, a version of the QLQ-BN20 questionnaire with 2 additional questions assessing cognitive functioning that were not addressed in the QLQ-C15-PAL. METHODS AND MATERIALS Patients with brain metastases completed the QLQ-C15-PAL and QLQ-BN20+2 questionnaires to assess QOL before and 1 month after radiation. Linear regression analysis was used to assess changes in QOL scores over time, as well as to explore associations between the QLQ-BN20+2 and QLQ-C15-PAL scales, patient demographics, and clinical variables. Spearman correlation assessed associations between the QLQ-BN20+2 and QLQ-C15-PAL scales. RESULTS Among 108 patients, the majority (55%) received whole-brain radiotherapy only, with 65% of patients completing follow-up at 1 month after treatment. The most prominent symptoms at baseline were future uncertainty (QLQ-BN20+2) and fatigue (QLQ-C15-PAL). After treatment, significant improvement was seen for the QLQ-C15-PAL insomnia scale, as well as the QLQ-BN20+2 scales of future uncertainty, visual disorder, and concentration difficulty. Baseline Karnofsky Performance Status was negatively correlated to QLQ-BN20+2 motor dysfunction but positively related to QLQ-C15-PAL physical functioning and QLQ-BN20+2 cognitive functioning at baseline and follow-up. QLQ-BN20+2 scales of future uncertainty and motor dysfunction correlated with the most QLQ-C15-PAL scales, including overall QOL (negative association) at baseline and follow-up. CONCLUSION After radiation, the questionnaires showed maintenance of QOL and improvement of QOL scores such as future uncertainty, which featured prominently in this patient population. It is proposed that the 37-item QLQ-BN20+2 and QLQ-C15-PAL, as opposed to the 50-item QLQ-BN20 and QLQ-C30, may be used together as a universal QOL assessment tool in this setting.

Estimating the Need for Radiotherapy for Patients With Prostate, Breast, and Lung Cancers: Verification of Model Estimates of Need With Radiotherapy Utilization Data From British Columbia

PURPOSE Estimates of the need for radiotherapy (RT) using different methods (criterion based benchmarking [CBB] and the Canadian [C-EBEST] and Australian [A-EBEST] epidemiologically based estimates) exist for various cancer sites. We compared these model estimates to actual RT rates for lung, breast, and prostate cancers in British Columbia (BC). METHODS AND MATERIALS All cases of lung, breast, and prostate cancers in BC from 1997 to 2004 and all patients receiving RT within 1 year (RT(1Y)) and within 5 years (RT(5Y)) of diagnosis were identified. The RT(1Y) and RT(5Y) proportions in health regions with a cancer center for the most recent year were then calculated. RT rates were compared with CBB and EBEST estimates of RT needs. Variation was assessed by time and region. RESULTS The RT(1Y) in regions with a cancer center for lung, breast, and prostate cancers were 51%, 58%, and 33% compared with 45%, 57%, and 32% for C-EBEST and 41%, 61%, and 37% for CBB models. The RT(5Y) rates in regions with a cancer center for lung, breast, and prostate cancers were 59%, 61%, and 40% compared with 61%, 66%, and 61% for C-EBEST and 75%, 83%, and 60% for A-EBEST models. The RT(1Y) rates increased for breast and prostate cancers. CONCLUSIONS C-EBEST and CBB model estimates are closer to the actual RT rates than the A-EBEST estimates. Application of these model estimates by health care decision makers should be undertaken with an understanding of the methods used and the assumptions on which they were based.

Patterns of Practice in Palliative Radiotherapy for Painful Bone Metastases: Impact of a Regional Rapid Access Clinic on Access to Care

PURPOSE External beam radiotherapy (RT) is commonly indicated for the palliation of symptomatic bone metastases, but there is evidence of underutilization of this treatment modality in palliative care for cancer populations. This study was conducted to investigate factors that influenced the use of palliative RT services at a regional comprehensive cancer center. METHODS AND MATERIALS A cohort of patients with radiographically confirmed bone metastases and first-time users of palliative RT between 2003 and 2005 was retrospectively reviewed from the time of initial diagnosis of bone metastases to death or last follow-up. Type of radiation treatment service provider used (rapid access or routine access) and patient-, tumor-, and treatment-related factors were analyzed for their influences on the number of treatment courses given over the duration of disease. RESULTS A total of 887 patients received 1,354 courses of palliative RT for bone metastases at a median interval of 4.0 months between courses. Thirty-three percent of patients required more than one RT course. Increased age and travel distance reduced the likelihood and number of treatment courses, while service through a rapid access clinic was independently associated with an increase in subsequent use of palliative RT. CONCLUSIONS A rapid access service model for palliative RT facilitated access to RT. Travel distance and other factors remained substantial barriers to use of palliative RT services. The pattern of practice suggests an unmet need for symptom control in patients with bone metastases.

Effect of re-irradiation for painful bone metastases on urinary markers of osteoclast activity (NCIC CTG SC.20U)

PURPOSE The NCIC CTG Symptom Control.20 randomized trial (SC.20) confirmed the effectiveness of re-irradiation to painful bone metastases. This companion study correlates urinary markers of osteoclast activity with response to re-irradiation, survival and skeletal related events (SREs). METHODS Pain response was assessed using the International Consensus Endpoints. Urinary markers of bone turnover-pyridinoline (PYD), deoxypyridinoline (DPD), N-telopeptide (NTX), Alpha and Beta cross-laps of C-telopeptide (CTX)-before and 1month after re-irradiation were correlated to response to re-irradiation and then to both, either or none of the initial and re-irradiation: frequent responders (response to both); eventual responders (response to re-irradiation only); eventual non-responders (response to initial radiation only), and absolute non-responders (no response to both). RESULTS Significant differences between 40 responders and 69 non-responders to re-irradiation existed for PYD (p=0.03) and DPD (p=0.04) at baseline. When patients were categorized as frequent responders (N=34), eventual responders (6), eventual non-responders (59) and absolute non-responders (10), the mean values of all markers in the absolute non-responders at baseline and the follow-up were about double those for the other three groups with statistically significant difference for DPD (p=0.03) at baseline. Absolute non-responders had the worst survival. The few occurrences of the SREs did not allow meaningful comparisons among the groups. CONCLUSION There were significant differences between responders and non-responders to re-irradiation for PYD and DPD at baseline. The urinary markers in the absolute non-responders were markedly elevated at both baseline and follow-up with a statistically significant difference for DPD at baseline.

A population-based study examining the impact of a multidisciplinary rapid access clinic on utilization of initial treatment options for patients with localized prostate cancer.

15 Background: Treatment decisions in localized prostate cancer (LPCa) are complicated by the variety of available options. A rapid access cancer clinic (RAC) has been unique to Calgary, Alberta (AB) since 2007. RAC offers multidisciplinary prostate cancer care by a urologist, medical oncologist, and radiation oncologist. It is hypothesized that treatment utilization data from decisions taken at RAC may serve to benchmark the appropriateness of treatment decisions on a population level. OBJECTIVES To compare utilization rates for initial treatment of LPCa between AB and RAC. METHODS Records of patients with clinically LPCa in AB between 2007-9 were reviewed with ethics approval. Records were linked to the AB cancer registry database. Clinical, treatment and health services characteristics pertaining to patients attending RAC were compared to those managed elsewhere in AB. The primary endpoints were utilization rates by initial treatment; prostatectomy (P), radiotherapy (RT), hormone therapy (H), active surveillance (A). A logistics regression model was constructed to examine the influence of RAC on initial treatment decisions, while controlling for interactions and factors of interest. RESULTS 2,660 patients were diagnosed with LPCa. 375 presented to RAC. Utilization rates among RAC patients: P-60.3% (95%CI: 55.3-65.2), A-16%(12.3-19.7), RT-11.7%(8.5-15.0) and H-8.0%(CI:5.2-10.8). This compares to AB rates of P-47.2%(45.9-48.3), A-6.1%(15.2-17.0), RT-18.8%(17.9-19.7), and H-14.5%(13.6-15.4). On multivariate analysis, RAC was associated with a trend towards receiving RT (OR 1.6, p=0.097). CONCLUSIONS A specialized clinic for LPCa may be associated with a higher likelihood of receiving radiotherapy as initial treatment compared to the prostate cancer population in Alberta.

Outcomes in stage I non-small cell lung cancer following the introduction of stereotactic body radiotherapy in Alberta, Canada.

306 Background: Stereotactic body radiotherapy (SBRT) is a treatment option for patients with Stage I (T1-T2N0M0) NSCLC (non-small cell lung cancer) who decline surgery or are medically inoperable. Case series suggest superior outcomes with lung SBRT compared to conventional radiotherapy and high local control rates. The primary study objective was to describe population outcomes in the initial management of stage 1 NSCLC patients including treatment utilization rates and survival. METHODS Clinical records of patients diagnosed with Stage 1 NSCLC who attended any SBRT clinic in Alberta between 2005 and 2011 were examined with REB approval. These cases were linked to the Alberta Cancer Registry and Provincial Health Administration databases. Clinical, treatment, pre-diagnosis 3M Aggregate Clinical Risk Grouping (ACRG3) scores as a proxy for prevailing patient comorbidity and health services characteristics pertaining to all cases of stage 1 NSCLC between 2005 and 2011 were determined. A cox regression model was constructed to examine the influence of these factors and their interactions on cancer outcomes. RESULTS 2,146 patients were diagnosed with stage 1 NSCLC. Median patient age was 72 and overall 43.0% of cases had a 1 year pre-treatment ACRG3 score of 10-49. Observed treatment utilization rates; surgery 63.4% (95%CI: 61.4-65.4), conventional RT 9.0% (7.9-10.3%), SBRT 3.6% (2.8-4.4%), chemotherapy 0.8% (0.5-1.4%) and no treatment 23.2% (21.4-25.0%). Median survival ranged from 69.6 months for surgery to 17.0 months if no active treatment was delivered. SBRT had survival outcomes superior to conventional radiotherapy: median survival of 39.4 vs. 23.5 months (p<0.001), despite more patients 69.9% vs 63.0% having an ACRG3-1y of 50 or higher. On multivariate analysis, while accounting for all modeled variables including ACRG3 pre-treatment and compared to patients receiving no active treatment, primary surgical intervention HR= 0.23 (95%C.I.:0.18-0.28) and SBRT HR=0.33 (0.21-0.51) remained most strongly associated with survival.No significant variable interactions (p<0.05) between treatment modality, patient age, pathology diagnosis and ACRG3 scores were demonstrated to impact on survival outcomes. CONCLUSIONS Surgery for Stage I NSCLC patients is associated with the best overall survival. SBRT patients have improved survival over conventional radiotherapy. Improvements in population level outcomes may result from an increased utilization of SBRT in the non-surgical management of stage I lung cancer.

A population-based study examining the influence of a specialized rapid-access cancer clinic on initial treatment choice in localized prostate cancer

INTRODUCTION Treatment decisions in localized prostate cancer are complicated by the available choices. A rapid-access cancer clinic (RAC) has been unique to Calgary, AB, since 2007. This RAC offers multidisciplinary prostate cancer education by a urologist, medical oncologist, and radiation oncologist. It is hypothesized that treatment utilization data from decisions taken at RAC may serve to benchmark the appropriateness of treatment decisions on a population level. METHODS Records of patients with clinically localized prostate cancer in Alberta between October 1, 2007 and September 30, 2009 were reviewed with ethics approval. Records were linked to the Alberta Cancer Registry database. Clinical, treatment, and health services characteristics pertaining to patients attending RAC were compared to the general population. The primary endpoint was utilization rates of each initial treatment. RESULTS During this two-year period, 2838 patients were diagnosed with localized prostate cancer; 375 attended RAC. The utilization rates among RAC patients vs. the whole Alberta population were: prostatectomy 60.3% (95% confidence interval [CI] 55.3-65.2) vs. 48.0% (95% CI 47.1-50.7; χ2 p<0.001); active surveillance 16.0% (95% CI 12.3-19.7%) vs. 13.5% (95% CI 12.2-15.8; χ2 p=0.214); radiotherapy 11.7% (95% CI 8.5-15.0) vs. 18.0% (95% CI 16.9-20.5; χ2 p=0.002); and hormone therapy 8.0% (95% CI 5.2-10.8) vs. 17.4% (95% CI 16.1-18.9; χ2 p<0.001). CONCLUSIONS A specialized clinic for localized prostate cancer may be associated with a higher likelihood of receiving surgery or active surveillance as initial treatment compared to the prostate cancer population in Alberta.

End-of-life cancer care: Health service delivery in the last 12 months of life in Calgary, Alberta, Canada.

171 Background: In Calgary, Alberta, the Calgary Zone Palliative Care Collaborative (CZPCC) undertook a study to examine the current state of cancer and palliative services and to recommend steps to address gaps in service delivery. We hypothesized that early access to palliative care services would reduce utilization of active cancer treatments and services for individuals nearing the end of life. Our study objectives were to determine the utilization and timing of acute, palliative and oncology-related services in Calgary. METHODS This retrospective study examined cancer registry and administrative data for patients > 18 years, who died in 2012. Measures of aggressive end of life care (EOL) were also collected. A combination of descriptive statistics, tests of association and multivariate regression analysis were conducted. RESULTS N = 1909 died of cancer in 2012: median age 73 years (IQR: 62-82 years) and median disease duration 364 days (IQR: 92-1114 days). 40.6% of patients received systemic treatment in last 12 months of life. 29.9% received radiotherapy and 13.0% received psychosocial/spiritual care. Palliative care contact was 80.7%, inclusive of 20.6% who had an intensive palliative care unit admission. 5.2% had EOL chemotherapy and 3.8% received EOL radiotherapy. Up to 10.4% of patients had one or more hospital admission. There was no significant effect of age on those who received aggressive EOL care. Men had an increased probability to receive aggressive EOL care (p = 0.015). Tumor group was also associated with receiving aggressive EOL (p < 0.001), with the highest utilization in Head and Neck and hematological malignancies. In patients with a disease duration of > 4 months those who received palliative care at least 2-3 months prior to death were less likely to receive aggressive EOL care (P < 0.001). Patients whose disease duration was < 1 month were less likely to receive aggressive EOL care if they received palliative care services (p = 0.02). CONCLUSIONS The provision of palliative care services at the end of life is most needed among men and certain tumor groups who are the highest users of aggressive EOL care.

Determining the need and utilization of radiotherapy in cancers of the breast, cervix, lung, prostate, and rectum in Alberta, Canada.

116 Background: Determining the appropriate rate of RT is important for health care planning and resource allocation. Establishing RT shortfalls (difference between observed and estimates of RT need) could provide an estimate of the capacity expansion that would be required to address them. Our primary objective was to determine the utilization of RT for cancers of the breast, cervix, lung, prostate and rectum in Alberta (AB), Canada. To determine the burden of RT shortfalls in AB, the secondary objective was to compare the observed AB RT rates to estimates of need derived from criterion-based benchmarking (CBB) and evidence-based estimates (EBEST). METHODS All incident cases of breast (B), cervix (C), lung (L), prostate (P) and rectal (R) cancers diagnosed in 2004-8 in AB were identified from the provincial cancer registry (ACR). Ethics board approval was obtained. Patients receiving RT within one year (RT-1y) of diagnosis were identified and grouped by cancer site. The proportion of cases receiving RT-1y was then calculated. Rates were compared using a Z statistic of the normal approximation for a difference in proportions. Estimates of the appropriate RT rate were derived from CBB and EBEST methods described in the literature. RESULTS A total of 68,164 cancer cases of interest were identified from the ACR. RT-1y rates for AB (95%CI) were: B: 50.5%(49.5-51.4), C: 45.7%(42.2-49.3), L: 36.5%(35.5-37.3), P:26.4%(25.6-27.3) and R:38.8%(37.1-40.6). Observed rates of RT in AB were lower than estimates derived using CBB and EBEST of RT-1y for B: 60.7%(59.3-62.1) and 57.1%(52.6-62.0), C: 48.6%(39.1-58.1) and 63.4%(61.1-65.7), L: 41.3%(39.9-42.7) and 44.6%(41.0-48.2), P: 37.2%(35.8-38.7) and 32.0%(28.4-36.0), and R: 43.4%(39.1-47.6) and 69.6%(68.7-70.5). Shortfalls varied across cancer sites according to whether CBB or EBEST estimates were referenced, ranging from 4.8% in lung cancer to 30.8% in rectal cancer. CONCLUSIONS Important shortfalls exist in the utilization of RT in Alberta, Canada despite centralized cancer care and a publically funded health care system. The magnitude of the shortfall varied according to whether a CBB or EBEST estimate of RT was applied.