BACKGROUND:Health utility values are required for cost-utility analyses in breast cancer, yet EQ-5D-5L-based health utility estimates and validity evidence across clinically relevant health states remain limited. The objectives of this study were to estimate health utility values for pre-defined breast cancer health states using the EQ-5D-5L (EuroQoL 5-level) instrument and to investigate its construct validity in breast cancer. METHODS:This cross-sectional study included women with invasive breast cancer, who completed both EQ-5D-5L and the Edmonton Symptom Assessment System. Participants were classified into five pre-defined health states, considered relevant both to clinical practice and economic modeling. We used the Canadian EQ-5D-5L value set to calculate community-valued health utility scores for each health state. Additionally, we evaluated aspects of construct validity (known-group and convergent validity) of EQ-5D-5L in breast cancer. RESULTS:549 women were included; the mean age was 57 (SD 12) years. The mean EQ-5D-5L index score was 0.83 (SD 0.13; range 0.13 to 0.95), with a distribution skewed towards full health and a ceiling effect of 20%. The mean health utility value for early-stage breast cancer was 0.84 (95% CI 0.83-0.86) and for metastatic breast cancer was 0.78 (95% CI 0.76-0.81). This difference was 0.060 (lower 95% CI bound 0.036, slightly lower than the pre-specified minimum important difference of 0.037). Health utility values and ESAS scores met almost all pre-specified criteria for convergent validity. CONCLUSION:We generated a list of health utility values for five pre-defined breast cancer health states using EQ-5D-5L. Additional research is required to confirm the validity of EQ-5D-5L as an outcome measure in breast cancer.

Background: Symptom burden and functional impairment are common in women with breast cancer, yet their prevalence and clinical significance across the disease spectrum remain underexplored. We sought to describe symptom burden and performance status using patient-reported outcome measures and to identify patient characteristics associated with symptoms requiring clinical intervention. Methods: In this cross-sectional study, women with stage I–IV breast cancer completed the Edmonton Symptom Assessment System (ESAS) and the Patient-Reported Functional Status tool. We assessed the prevalence and severity of symptoms and calculated summary distress scores. Multivariable logistic regression was used to identify patient characteristics associated with clinically significant symptoms (ESAS ≥ 4). Results: Among 381 women (mean age 56.8 years; 27% metastatic; 72% with no comorbidities), 70% reported at least one moderate to severe symptom. The most common were tiredness (31%), lack of well-being (30%), and anxiety (21%). Mean summary distress scores were low overall. Most patients reported functional status scores of 0 or 1, and 43% of those with scores ≥2 had metastatic disease. Compared with metastatic patients, women within the first year after diagnosis were less likely to report a symptom requiring intervention (OR 0.49, 95% CI 0.24–0.90). Conclusions: Clinically significant symptoms are common among women with breast cancer, including those with potentially curable disease. Threshold-based use of ESAS, rather than reliance on mean scores, provides a more accurate assessment of patient needs. These findings support the routine integration of patient-reported outcomes into oncology care and underscore the importance of targeted multidisciplinary interventions.

Purpose A cross-sectional study was conducted to investigate the feasibility of implementing routine collection of the Euro-Qol 5 dimensions (EQ-5D) questionnaire, to inform drug and health technology reimbursement decision making.Methods Women with breast cancer were recruited during scheduled clinic visits to an academic cancer centre. EQ-5D-5L was self-administered using electronic tablets. Diagnostic and treatment data were abstracted from patient charts. Feasibility was assessed primarily by the proportion of patients who fully completed EQ-5D-5L and by their willingness to complete the instrument at each clinic visit.Results 588 women were approached for study participation, 341 were enrolled. Fully completed EQ-5D-5L questionnaires were obtained in 323 participants (95% of participants, 95% CI 92-97%). Median time for EQ-5D-5L completion was 1.5 minutes (range:0.35 to 14.7). Mean age of participants was 58 years old. Most women who completed EQ-5D were White, born outside Canada and presented a high education level; one-quarter had metastatic disease. Most participants reported "No problems" in all EQ-5D-5L dimensions. Mean EQ-5D-5L index and mean EQ-5D-5L VAS values for all participants were respectively 0.83 (SD 0.13) and 75.7 (SD 17.45), with patients with metastatic disease scoring the lowest values. Seventy-eight percent of participants were willing to complete EQ-5D-5L at each clinic visit; lower Charlson comorbidity index and higher education level were predictors of willingness to continue to answer EQ-5D-5L.Conclusions Tablet-based collection of EQ-5D-5L in the context of routine clinical practice proved to be feasible. However, many patients declined study participation or reported being in full health, raising concerns about whether this method of collecting EQ-5D adequately represents the health status of all breast cancer patients.

BACKGROUND:There is no widely accepted framework to guide the development of condition-specific preference-based instruments (CSPBIs) that includes both de novo and from existing non-preference-based instruments. The purpose of this study was to address this gap by reviewing the published literature on CSPBIs, with particular attention to the application of item response theory (IRT) and Rasch analysis in their development.METHODS:A scoping review of the literature covering the concepts of all phases of CSPBI development and evaluation was performed from MEDLINE, Embase, PsychInfo, CINAHL, and the Cochrane Library, from inception to December 30, 2022.RESULTS:The titles and abstracts of 1,967 unique references were reviewed. After retrieving and reviewing 154 full-text articles, data were extracted from 109 articles, representing 41 CSPBIs covering 21 diseases or conditions. The development of CSPBIs was conceptualized as a 15-step framework, covering four phases: 1) develop initial questionnaire items (when no suitable non-preference-based instrument exists), 2) establish the dimensional structure, 3) reduce items per dimension, 4) value and model health state utilities. Thirty-nine instruments used a type of Rasch model and two instruments used IRT models in phase 3.CONCLUSION:We present an expanded framework that outlines the development of CSPBIs, both from existing non-preference-based instruments and de novo when no suitable non-preference-based instrument exists, using IRT and Rasch analysis. For items that fit the Rasch model, developers selected one item per dimension and explored item response level reduction. This framework will guide researchers who are developing or assessing CSPBIs.

Background: The EuroQol- 5 Dimension (EQ-5D) is a generic patient-reported outcome measure widely used to capture meaningful change in health-related quality of life between treatments to inform drug and health technology reimbursement decision making. We investigated the construct validity of EQ-5D-5L in women with breast cancer. Methods: This study included adult women diagnosed with stage I to IV breast cancer, who completed the EQ-5D-5L and the Edmonton Symptom Assessment System (ESAS) during outpatient clinic visits at two academic cancer centres in Toronto. We evaluated construct validity through assessing known-group validity and convergent/divergent validity. For known-group validity, the primary analysis tested the hypothesis that EQ-5D-5L could adequately discriminate between patients with metastatic disease and early-stage disease; secondary analyses addressed utility values between women in breast-cancer associated health states. A suggested minimally important difference (MID) for the Canadian scoring of the EQ-5D-5L is 0.037; we evaluated whether the lower bound of the 95% confidence interval (95%CI) exceeded this value. In terms of convergent/divergent validity, the primary analysis tested the hypothesis that EQ-5D-5L mean utility values for each health state (HS) would be at least moderately correlated with ESAS total symptom distress score (SDS) (|r|>0.30) using Wilcoxon rank-sum tests and Spearman’s correlation tests. Construct validity was considered as acceptable if the hypotheses of the primary analysis are satisfied. Results: We recruited 549 women, 406 (74%) with early-stage disease and 143 (26%) with metastatic disease (HS5), with a mean age of 57 (SD 12); 412 (75%) had been diagnosed with breast cancer in the 7 years prior to recruitment and were receiving active treatment for their cancer. The mean utility value for early-stage breast cancer was 0.84 (95% CI 0.83-0.86) and for metastatic breast cancer (0.78, 95% CI 0.76-0.81). This difference was 0.060 (95% CI 0.036 to 0.085, p< 0.001) with the lower bound of the confidence interval slightly less than the prespecified MID (0.037). There was no significant difference between the mean utility value for women in the first year after primary breast cancer diagnosis (HS1), and women in their second to fifth year after a primary breast cancer treated with curative intent (HS3) or between women in HS1 and women in their sixth and following years after a primary breast cancer treated with curative intent (HS4). EQ-5D-5L also did not discriminate between women in HS3 and HS4. For convergent/divergent validity, there was a negative correlation between utility values and ESAS physical, emotional and total SDSs. EQ-5D-5L and ESAS total SDSs correlation coefficients were higher than 0.30 for all health states. Conclusion: EQ-5D-5L met criteria for convergent/divergent validity in women with breast cancer. The tests for discriminant validity were equivocal, suggesting more research is needed for assessing construct validity with a larger sample size. Table 1. EQ-5D-5L Utility Values. N= number; Std Dev= standard deviation; IQR= interquartile range; CI= confidence interval Citation Format: Sofia Torres, Maureen Trudeau, Geoffrey Liu, Nicholas Mitsakakis, Ahmed Bayoumi. Validity of EQ-5D-5L for women with breast cancer [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P6-05-43.

Background: The EuroQol- 5 Dimension (EQ-5D) is a generic patient-reported outcome measure widely used to capture meaningful change in health-related quality of life between treatments to inform drug and health technology reimbursement decision making. The EQ-5D-5L (5 level) consists of 5 dimensions (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression) and a visual analogue scale (VAS). Each dimension has 5 levels, ranging from 1 “no problems” to 5 “extreme problems”, generating 3125 distinct health states (HS). The VAS score ranges from 0 (worst health imaginable) to 100 (best health imaginable). We used Canadian Index values to calculate community-weighted utility scores for 5 pre-defined breast cancer (BC) disease states.Methods: This cross-sectional study included adult women diagnosed with stage I to IV BC, who completed EQ-5D-5L, during outpatient clinic visits at two academic cancer centres in Toronto. Participants were classified into 5 disease states, considered relevant both to clinical practice and economic modeling, ranging from primary breast cancer to metastatic disease. EQ-5D-5L scores were described for each disease state, and mean utility scores and VAS scores were calculated.Results: 549 women were included; the mean age was 57 (SD 12) years; 72% had a Charlson Comorbidity Index of 0. A majority reported “Slight” to “Moderate” problems for the “Pain/Discomfort” (60%) and “Anxiety/Depression” (55%) dimensions of EQ-5D-5L; in all other dimensions, most patients reported “No problems”. As expected, patients with metastatic BC had the highest proportion of reported problems. The mean EQ-5D-5L index score was 0.83 (SD 0.13; range 0.36 to 0.95), with a distribution skewed towards full health and a ceiling effect of 20% (n=110; Table 1). Only 126 different self-reported health states were observed in our patient population. Mean VAS was 75 (SD 17.5; range 10 to 100). For early BC, mean index scores were: 0.87 (n=49; SD 0.07) for patients on chemotherapy (+/-targeted therapy); 0.85 (n=212; SD 0.12) for patients on endocrine treatment (+/-targeted therapy); 0.79 (n=16; SD 16) for patients on targeted therapy; and 0.84 (n=129; SD 0.14) for patients not receiving systemic therapy. For metastatic BC, mean index scores were: 0.79 (n=36, SD 14) for patients on chemotherapy (+/-targeted therapy); 0.80 (n=79, SD 0.14) for patients on endocrine treatment (+/-targeted therapy); 0.69 (n=8; SD 0.19) for patients on targeted therapy; and 0.77 (n=20; SD 0.14) for patients not receiving systemic therapy.Conclusions: BC had an impact mainly in the “Pain/Discomfort” and “Anxiety/Depression” dimensions of EQ-5D-5L. A high ceiling effect was observed in this patient population, despite using the 5L version of EQ-5D. EQ-5D-5L health utilities were significantly lower for metastatic BC, but no significant differences were found among various early-stage BC states. Table 1.- Mean EQ-5D-5L Index and VAS scoresEQ-5D-5LIndex ScoresVASDisease StateN (%)Mean (SD)Min-MaxMean (SD)Min-Max1st year after primary BC146 (27)0.85 (0.12)*0.23-0.9576.5(16.6)10-1001st year after recurrence or primary BC13 (2)0.78 (0.21)0.13-0.9568.9(23.3)20-1002nd-5th year after a primary BC or recurrence treated with curative intent185 (34)0.84 (0.11)*0.21-0.9578.3(14.7)36-1006th and following years after a primary BC or recurrence treated with curative intent62 (11)0.86 (0.14)*0.36-0.9581.0(15.6)20-100Metastatic BC143 (26)0.78 (0.14)0.29-0.9567.3(19.2)21-100Total549 (100)0.83 (0.13)0.13-0.9575.0(17.5)10-100*significant difference was found between these disease states and metastatic diseases (P < 0.05). Citation Format: Sofia Torres, Ahmed Bayoumi, Maureen Trudeau, Ana Beatriz Kinupe Abrahao, Nicholas Mitsakakis, Murray Krahn, Geoffrey Liu. Eq-5d-5l utility scores for patients with breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P4-12-11.

OBJECTIVES:Breast cancer (BrC) and its treatments impair health-related quality of life (HRQoL). Utility is a measure of HRQoL that includes preferences for health outcomes, used in treatment decision-making. Generic preference-based instruments lack BrC-specific concerns, indicating the need for a BrC-specific preference-based instrument. Our objective was to determine dimensions of the European Organisation for Research and Treatment of Cancer (EORTC) general cancer (QLQ-C30) and breast module (BR45) instruments, the first step in our development of the novel Breast Utility Instrument (BUI). METHODS:Patients (n = 408) attending outpatient BrC clinics at an urban cancer centre, and representing a spectrum of BrC health states, completed the QLQ-C30 and BR45. We performed confirmatory factor analysis of the combined QLQ-C30 and BR45 using mean-and variance-adjusted unweighted least squares estimation. The hypothesized factor model was based on clinical relevance, item distributions, missing data, item-importance, and internal reliability of dimensions. Models were evaluated based on global and item fit, local areas of strain, and likelihood ratio tests of nested models. RESULTS:Our final model had 10 dimensions: physical and role functioning, emotional functioning, social functioning, body image, pain, fatigue, systemic therapy side effects, sexual functioning and enjoyment, arm and breast symptoms, and endocrine therapy symptoms. Good overall model fit was achieved: χ2/df: 1.45, Tucker-Lewis index: 0.946, comparative fit index: 0.951, standardized root-mean-square residual: 0.069, root-mean-square error of approximation: 0.033 (0.030-0.037). All items had salient factor loadings (λ>0.4, p<0.001). CONCLUSIONS:We identified important BrC HRQoL dimensions to develop the BUI, a BrC-specific preference-based instrument.

Background: To improve symptom management in cancer patients, the province of Ontario (Canada) implemented the collection of the Edmonton Symptom Assessment System (ESAS) and Patient-Reported Functional Status (PRFS) tool at cancer centre outpatient visits. ESAS measures 9 symptoms on a scale from 0 (no symptoms) to 10 (worst possible symptom); scores > 4 (moderate to severe) require intervention. The PRFS tool is a version of the Eastern Cooperative Oncology Group performance status tool designed to be completed by the patient; scores range from 0 (“normal with no limitation”) to 4 (“pretty much bedridden, rarely out of bed”). Our main objective was to describe the symptom burden and functional status of breast cancer (BC) patients.Methods: This cross-sectional study included women with stage I-IV BC who completed the ESAS and the PRFS tool during outpatient clinic visits at 2 Toronto academic centres. ESAS and PRFS scores were described for 3 participant groups: women diagnosed with early-stage BC (ESBC= stages I-III) in the year prior to enrolment; women with ESBC diagnosed > 1 year prior to enrolment; and women with metastatic BC. Multivariable logistic regression models were used to identify factors associated with ESAS scores > 4 for each ESAS symptom. Covariates included age, comorbidity, BC subtype, disease group, previous surgery, and current systemic therapy. Results: Of 381 women, 107 (28%) were diagnosed with ESBC ≤ 1 year prior to enrolment; 171 (45%) were diagnosed with ESBC > 1 year prior to enrolment; 103 (27%) had metastatic BC. Mean age was 57 (SD17) years. For ESBC, tiredness had the highest incidence of moderate to severe scores, followed by impaired well-being and anxiety (Table 1). For metastatic BC, impaired well-being had the highest incidence of moderate to severe symptoms, followed by tiredness and anxiety. More than 20% of metastatic BC patients had scores > 4 for pain, drowsiness, shortness of breath and depression. Most reported PRFS scores of 0 (“Normal with no limitations”) or 1 (“Not my normal self, but able to be up and about with fairly normal activities”), but 17% of metastatic BC patients reported a PRFS of 2 (“Not feeling up to most things, but in bed or chair less than half the day”). A comorbidity score ≥1 was associated with moderate to severe depression (OR 2.0; 95%CI 1.1-3.7), lack of appetite (OR 3.1; 95%CI 1.4-6.6), shortness of breath (OR 2.6; 95%CI 1.3-5.2), tiredness (OR 2.0; 95%CI 1.2-3.3) and impaired well-being (OR 2.0; 95%CI 1.2-3.4). Metastatic BC was associated with moderate to severe scores for anxiety (OR 2.3; 95%CI 1.2-4.5), pain (OR 3.0; 95%CI 1.4-6.4) and impaired well-being (OR 2.1; 95%CI 1.1-4.0). Conclusion: A high prevalence of symptom scores > 4 was found both in ESBC and metastatic BC, highlighting the need of symptom screening for all BC patients. Particular attention should be paid to patients with comorbidity scores ≥ 1, who have increased odds of presenting higher symptom scores. Table 1.- Prevalence of ESAS scores ≥ 4 and PRFS scores for each disease group.ESAS SymptomsPercent with ESAS scores ≥ 4 (moderate to severe)Early-Stage Breast Cancer – 1st year (N=107)Early-Stage Breast Cancer – > 1st year (N=171)Metastatic Breast Cancer (N=103)Total (N=381)Pain12.1%14.7%30.1%18.2%Tiredness25.2%30.1%39.8%31.3%Drowsiness1.8%12.9%25.2%16.6%Nausea9.4%3.5%7.8%4.2%Lack of Appetite10.3%7.1%13.6%9.5%Shortness of Breath14%7.6%22.3%12.4%Depression17.7%14.6%22.6%16.5%Anxiety24.3%17.2%35.3%21.3%Well-being1.8%25.7%44.1%30.3%PRFS scoresPercent with PRFS ScoreNormal with no limitations (0)30.8%49.7%28.2%38.6%Not my normal self, but able to be up and about with fairly normal activities (1)52.3%42.1%50.5%47.2%Not feeling up to most things, but in bed or chair less than half the day (2)12.2%4.7%16.5%10%Able to do little activity and spend most of the day in bed or chair (3)3.7%2.3%4.9%3.4%Pretty much bedridden, rarely out of bed (4)---- Citation Format: Sofia Torres, Ahmed Bayoumi, Geoffrey Liu, Ana Beatriz Kinupe Abrahao, Nicholas Mitsakakis, Murray Krahn, Maureen Trudeau. Patient reported symptoms and functional status in breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P4-10-22.

197 Background: Recently, our group reported Canadian-derived EQ-5D-3L-derived health utility scores for over 20 different cancer sites (PMID: 27567613). However, more recently, a Canadian valuation set for EQ-5D-5L derived health utilities has been reported. As we switch from the 3L to 5L version, we need to evaluate whether the data can be analyzed together across versions, or whether the older cancer valuations need to be replicated. Methods: Breast cancer patients in three health states (primary breast cancer, locoregional recurrence of breast cancer, and metastatic breast cancer) were evaluated using EQ-5D-3L from 2014-2015 and EQ-5D-5L from 2016-2017. Mean (SD) values were compared. We opted not to compare the two versions in the same patients because of recall bias, so we compared two different sets of patients. Results: Of 387 breast cancer patients, 259 (67%) had completed the EQ-5D-5L, while 128 (33%) completed the EQ-5D-3L. The two groups had similar distributions for clinico-demographic data except for ethnicity, where there were more Asian patients in the 5L (22% vs 9%; p < 0.0074). When comparing EQ-5D-5L and EQ-5L-3L within health states, the EQ-5D-5L values were numerically higher for all three health states: for primary breast cancer (mean HUS(SD)/version: 0.85(0.11)/5L vs. 0.81(0.16)/3L) with a p-value of 0.028; locoregional recurrence (0.87(0.08)/5L vs. 0.85(0.20)/5L) with no available p-value due to the small sample size; metastatic disease (0.79(0.15)/5L vs. 0.78(0.13)/3L) with a p-value of 0.78. Histogram distributions appeared different between the two versions. There was greater separation and statistical significance between health states when using the 5L version when compared to the 3L version. The 5L version had a greater variation of values than the 3L, utilizing the entire spectrum of values from 0.0 to 1.0. Conclusions: Though the trends across health states were similar in EQ-5D-5L and EQ-5D-3L health states, values using the 5L were higher and distributions appeared different. There should be extreme caution if considering combining health utility values from 5L and 3L in Canadian breast cancer datasets.

We undertake an analysis of ongoing BC targeted therapy trials registered to CT.gov to describe patterns of ongoing clinical research, highlight gaps in current research programs and identify ways of optimizing ongoing initiatives. A search of clinicaltrials.gov was conducted on September 4, 2013 to identify ongoing randomized phase II and III trials of targeted therapies in BC. A total of 280 trials were analyzed, the majority conducted in either human epidermal growth factor receptor 2 (HER2)-positive (n = 79, 28.2 %) or hormone receptor (HR)-positive (n = 104, 37.1 %) populations. Less than half of all trials were conducted in populations selected to match the agent under investigation (n = 126, 45 %). HER2-directed therapy is the single most investigated class of targeted agents (n = 73, 26.1 %), but trials investigating anti-angiogenic agents are also common (n = 49, 17.5 %). The most common new classes of agents under investigation in HR-positive and triple negative (TN)/BRCA-positive disease, are non-receptor protein kinase-inhibitors (n = 12; 11.5 %) and poly (ADP-ribose) polymerase inhibitors (n = 6; 30 %), respectively. The majority of regimens combine new targeted agents with either chemotherapy (n = 164, 58.6 %) or endocrine therapy (n = 113, 40.4 %); a total of 8 trials (2.8 %) investigated peptide-drug conjugates. The most frequently utilized end-points were pathological complete response in the neo-adjuvant setting (n = 36, 52.9 %) and time-to-event end-points in the adjuvant and advanced settings (77.3 and 72.6 %, respectively). Our findings suggest a need for more target-matched agent development, maintenance of a value-based focus in research and a need for the clinical development of agents to treat TN/BRCA-positive and HR-positive BC.

e12026 Background: The 21-gene Recurrence Score (RS) assay has been shown to be both prognostic and predictive of chemotherapy (CT) benefit in patients with node negative and node positive (N+), estrogen-receptor positive (ER+) early-stage breast cancer (EBC). Currently, the assay is only reimbursed by the provincial single payer Ontario Health Ministry for node-negative EBC. Methods: The primary objective of this prospective study is to characterize how the results of the RS assay impact the decision making processes of medical oncologists in an academic centre in Ontario by evaluating recommendations for adjuvant therapy for 70 patients with ER+, HER2-negative EBC and 1-3 positive lymph nodes. Secondary objectives are to characterize changes in the physician’s level of confidence in their recommendation, to determine whether the results of the RS affect patients’ treatment preferences and their level of confidence and to evaluate the actual treatment administered. Recruitment will be completed by the time of the ASCO meeting. Here we report the results for the first 50 patients, enrolled from October 2014 to December 2015. Results: Mean patients’ age was 61 (range: 41-84). Tumor size was ≤ 2 cm in 46% of patients, > 2-5 cm in 44% and > 5cm in 10%. Tumors were grade 1 in 24% of patients, grade 2 in 50% and grade 3 in 26%. RS was low ( < 18) in 52% of cases, intermediate (RS 18-30) in 38% and high ( ≥ 31) in 10%. Treatment recommendations changed in 35% of all evaluable patients (17 out of 48). The most significant change was in the group with a low RS ( < 18), with 48% of the recommendations changing from upfront CT pre-assay, to endocrine therapy only post-assay. Physicians’ confidence in the recommendations increased in 46% of cases and decreased in 12.5%. Patients’ confidence in their treatment choice increased in 65% of cases and decreased in 13%. Upfront CT was recommended to 75% of patients pre-assay, 43% ultimately received CT. Conclusions: The RS assay resulted in a substantial decrease in the number of node positive patients who would receive CT and in an increase of the physicians’ and patients’ confidence in the adjuvant treatment recommendations.

6628 Background: Radiation is a standard option in the management of prostate cancer (PC). Multiple options are available with similar outcomes, yet substantially different costs. SBRT is non-invasive and convenient but has never been compared to LDR. Herein we conduct a cost-utility analysis comparing SBRT, LDR and standard EBRT. Methods: A decision-analytic Markov model was developed from the Ontario payer perspective to simulate the history of a 66 year old man with low risk PC. The model followed patients yearly over their lifetimes. Three radiation options were considered: SBRT 35 Gy, LDR 145 Gy and EBRT 76 Gy. Health states included ‘Recurrence free’, ‘biochemical recurrence (BR)’, ‘metastatic’ and ‘death’. Transition probabilities were based on a retrospective cohort analysis undertaken at our institution. Utilities were derived from the literature. Costs were assigned in 2015 Canadian dollars ($) and reflected Ontario’s health system and departmental costs. Outcomes included quality adjusted life-years (QALYs), costs and incremental cost-effectiveness ratios. All outcomes were discounted at 5%. A willingness-to-pay threshold of 50,000$/QALY was used. Results: LDR and SBRT were less expensive and more effective than EBRT. These findings were robust to changes in input variables. SBRT was the dominant strategy with 0.008 LYs and 0.029 QALYs gained and a reduction in cost of $2,615 when compared to LDR. However, these results were sensitive to the BR probability. LDR becomes the preferred strategy if the BR with SBRT is 1.3*[baseline BR_SBRT] or if the BR with LDR is 0.76*[baseline BR_LDR]. When assuming the same BR for both strategies, LDR becomes more effective with 0.009 QALYs gained at a cost of 272,848 $/QALY. Conclusions: In Ontario and comparable jurisdictions, SBRT might represent an economically attractive radiation strategy; however longer-term follow-up and higher quality evidence are needed. Effectiveness Cost Life-Years Incremental Costs QALYs Life-Years CAD QALYs SBRT 11.174 9.926 72,378 - - - Dominant LDR 11.166 9.897 74,993 0.008 0.029 2,615 Weakly Dominated EBRT 10.936 9.404 77,644 0.23 0.493 2,651 Dominated

The purpose of this study was to compare survival and risk of adverse events in women with early stage breast cancer (BC) treated with (1) doxorubicin (A), cyclophosphamide (C) + paclitaxel (P), (2) fluorouracil (F), epirubicin (E), cyclophosphamide (C) + docetaxel (D), or (3) dose-dense AC-P. Retrospective cohort study including 8462 women aged ≥18 years, with resected stage I–III BC, diagnosed between 2003 and 2009 in Ontario, identified through linkage of administrative databases. Primary outcome is overall survival (OS). Secondary outcomes are emergency room (ER) visits/hospitalizations, heart failure (HF), and leukemia. 4710 women were treated with FEC-D, 2065 with AC-P, and 1687 with dd AC-P. Adjusted 5-year OS was 92.1, 87.7, and 90.3 %, for each regimen, respectively (p = 0.0006). There was no difference in OS for FEC-D and dd AC-P in the propensity score-matched analyses (HR 1.24, 95 % CI 0.99–1.55). Five-year risk of HF was also similar (HR 1.09; 0.66–1.791.4 % for dd AC-P and 1.3 % for FEC-D and, p = 0.72). Treatment with FEC-D was significantly associated with ER visits and hospital admissions (p < 0.0001). The risks of leukemia were low and similar among the 3 groups (AC-P: 0.34 %, FEC-D: 0.08 %, dd AC-P: 0.12 %; p = 0.09). Although the efficacy of the three regimens was similar to that observed in randomized trials, we report higher toxicity with the use of these regimens in clinical practice. This was especially concerning for the docetaxel-containing regimen.

A 55-year-old woman with metastatic human epidermal growth factor receptor 2 (HER-2) positive breast cancer (BC) to the lungs and bones was diagnosed with central nervous system (CNS) metastases in November 2011. The MRI showed a right parietal lobe mass with adjacent leptomeningeal disease and several small bilateral cerebellar metastases. She was treated with whole brain irradiation (WBI), followed by capecitabine and lapatinib (December 2011-March 2013) and trastuzumab and lapatinib (May 2013-August 2013). Then, the brain MRI showed progression. In the absence of significant neurological symptoms, we postponed WBI and closely monitored for the development of neurological symptoms. Systemic treatment with trastuzumab emtansine (T-DM1), an antibody-drug conjugate composed of the cytotoxic agent DM1 conjugated to trastuzumab, was initiated in September 2013 to control systemic disease. Unexpectedly, after two cycles of treatment the brain MRI showed a decrease in size of CNS metastases. This case report suggests possible activity of T-DM1 in HER-2 positive BC with CNS metastases.