Allicia C. Girvan

Treatment rationale and study design for the pointbreak study: a randomized, open-label phase III study of pemetrexed/carboplatin/bevacizumab followed by maintenance pemetrexed/bevacizumab versus paclitaxel/carboplatin/bevacizumab followed by maintenance bevacizumab in patients with stage IIIB or IV nonsquamous non-small-cell lung cancer.

We present the treatment rationale and study design of the PointBreak study, a phase III study of pemetrexed/ carboplatin/bevacizumab induction followed by pemetrexed/bevacizumab maintenance (arm A) compared with paclitaxel/carboplatin/bevacizumab induction followed by bevacizumab maintenance (arm B) in patients with advanced nonsquamous non-small-cell lung cancer (NSCLC). Treatment consists of up to 4 cycles of induction therapy followed by maintenance therapy until disease progression or treatment discontinuation in approximately 900 patients (450 per treatment arm). The efficacy objectives of this study are to compare overall survival (OS), response rates, disease control rates, progression-free survival, and time to progressive disease between the 2 treatment arms. In addition, safety, quality of life, pharmacokinetics, and translational research will be investigated for both treatment arms. If the primary objective (OS) is achieved, this study will provide robust results on an alternative treatment option, pemetrexed/carboplatin/bevacizumab followed by maintenance therapy with pemetrexed/bevacizumab, for patients with nonsquamous NSCLC.

Non-small cell lung cancer patients with brain metastases treated with first-line platinum-doublet chemotherapy: Analysis from the European FRAME study.

OBJECTIVES We report on a post-hoc analysis of patients with brain metastases from a large prospective observational study of first-line treatment of non-small cell lung cancer (NSCLC). The aim was to describe baseline characteristics of NSCLC patients with brain metastases, understand their first-line treatment and report outcomes attained in real-world settings. MATERIALS AND METHODS This post-hoc analysis included all patients in the European observational FRAME study who had brain metastases at initiation of first-line treatment. Descriptive statistics were used for continuous and categorical variables and survival outcomes were assessed using the Kaplan-Meier approach. RESULTS Our data showed that 17% of patients (263/1564) had spread of the disease to the brain at initiation of first-line treatment. Patients with brain metastases were slightly younger, and more likely to have NSCLC of non-squamous histology than the overall study sample. 34% had received prior palliative radiotherapy to the brain. Our analysis showed a median overall survival (OS) of 7.2 months [95% confidence interval (CI) 6.1-8.2] for all patients with brain metastases treated with first-line platinum-based chemotherapy, ranging from 5.6 months for those treated with gemcitabine plus platinum up to 9.3 months for those treated with pemetrexed plus platinum. Further analysis showed that patients with brain metastases were more frequently treated with pemetrexed platinum-doublet therapy than with any other regimen. CONCLUSIONS Our analysis provides a unique set of real-world data which adds to current understanding about treatment decisions and outcomes for NSCLC patients with brain metastases for whom there is little clinical trial data available.

Glioblastoma treatment patterns, survival, and healthcare resource use in real-world clinical practice in the USA.

Background: Glioblastoma (GB) treatment remains challenging because of recurrence and poorly defined treatment options after first-line therapy. To better understand real-world application of treatment paradigms and their impact on outcomes, we describe patterns of treatment, outcomes, and use of cancer-related healthcare resource for glioblastoma in the USA. Methods: A retrospective, online chart-abstraction study was conducted; each participating oncologist contributed ≤5 charts. Patients were ≥18 years with biopsy-confirmed primary or secondary newly diagnosed GB on or after 1 January 2010, had received first- and second-line therapies, and had information collected for ≥3 months after initiation of second-line therapy or until death. Assessments were descriptive and included Kaplan– Meier analyses from initiation to end of second-line therapy, disease progression, or death. Results: One hundred sixty physicians contributed information on 503 patient charts. During first-line therapy, patients most commonly underwent temozolomide monotherapy (76.5%). During second-line therapy, patients most commonly underwent bevacizumab monotherapy (58.1%). Median duration of second-line therapy was 130 days; median time to disease progression was 113 days. Median survival was 153 days. Use of supportive care was observed to be numerically higher in first- compared with second-line therapy except for anti-depressants, growth factors, and stimulants. Frequently used resources included corticosteroids (78.8% of patients in first-line and 62.6% in second-line therapies), anti-epileptics (45.8% and 41.5%) and narcotic opioids (45.3% and 41.4%). Conclusions: Most GB patients received temozolomide during first-line therapy and bevacizumab monotherapy or combination therapy during second-line therapy. Use of supportive care appeared to be higher in first- compared with second-line therapy for some agents.

Cisplatin and carboplatin-based chemotherapy in the first-line treatment of non-small cell lung cancer: Analysis from the European FRAME study.

OBJECTIVES To explore patient and disease factors, and reasons behind the physicians choice of platinum backbone for the first-line treatment of non-small cell lung cancer (NSCLC), as observed in a European prospective observational study of patients receiving platinum-based chemotherapy as first-line treatment for advanced or metastatic NSCLC (the FRAME study). Additionally, overall survival (OS) for patients who received cisplatin or carboplatin was evaluated. MATERIALS AND METHODS A post-hoc analysis of the prospective study population was conducted. Baseline characteristics of patients receiving cisplatin versus carboplatin were compared and summarized by propensity score. Survival for matched patients was summarized using the Kaplan-Meier approach. RESULTS Of the 1564 patients who were included in the prospective study, 1520 received either cisplatin (54%) or carboplatin (46%) in combination with pemetrexed, gemcitabine, taxanes or vinorelbine. Patients treated with carboplatin were older than patients receiving cisplatin (mean age 67 versus 61 years; p<0.001), had poorer performance status (p<0.001), and more comorbidities (p<0.001). Cisplatin was most frequently combined with pemetrexed (47%), and carboplatin most frequently with taxanes (31%). Unadjusted median OS estimates for patients from the total prospective study sample were 11.5 months (95% confidence interval [CI] 10.1-12.9) for cisplatin recipients and 9.0 months (95% CI 8.1-10.6) for carboplatin recipients. Median (95% CI) overall survival for the matched cohorts was 10.8 months (8.8-14.3) for cisplatin versus 9.5 months (8.2-11.3) for carboplatin; p=0.086. CONCLUSION This post-hoc analysis illustrated real-life differences in patients with NSCLC prescribed platinum-based first-line treatment, and suggested that baseline patient and disease characteristics were associated with physicians choice of platinum agent, with cisplatin being more frequently prescribed to younger and fitter patients.

1252PNSCLC PATIENTS WITH BRAIN METASTASES TREATED WITH PLATINUM-BASED DOUBLETS AS FIRST-LINE THERAPY: ANALYSES FROM THE EUROPEAN FRAME OBSERVATIONAL STUDY

ABSTRACT Aim: Patients (pts) with brain metastases (mets) are often excluded from clinical trials and limited data exists. FRAME (2009-2012) was a prospective, non-interventional observational study of pts with advanced NSCLC receiving platinum (plt)-doublet chemotherapy as first-line treatment (FLT) across Europe. Thus, in FRAME unique real-life data could be collected on baseline characteristics, FLT and survival for NSCLC pts with brain mets. Methods: FLT decisions were at the discretion of the physicians under routine clinical practice. The primary objective of overall survival (OS) from FRAME was previously reported. Here, we describe baseline characteristics and survival for pts with brain mets at FLT initiation. Survival was estimated using Kaplan-Meier methods and unadjusted estimates are presented. Results: In FRAME overall, 1564 pts received plt in combination with: pemetrexed (pem; n = 569), gemcitabine (gem; n = 360), taxanes (tax; n = 295), vinorelbine (vin; n = 300), or other (n = 40); 263 (17%) of these pts had brain mets at FLT initiation. Pts with brain mets had a median age of 58 years [range: 33-84] (overall 64 years [range: 33-87], n = 1564), 64% were male (overall 72%), 84% had non-squamous histology (overall 72%), 21% had an ECOG performance status (PS) 2-3 (overall 17%), 58% received cisplatin as FLT plt backbone (overall 55%) and 34% received prior radiotherapy to the brain (overall 6%). Several baseline characteristics for pts with brain mets numerically varied among cohorts (Table 1, “other” cohort [n = 5] is not shown). For the overall study, median OS was 10.3 months (95% CI: 9.5-11.2). For pts with brain mets, median OS was 7.2 months (95% CI: 6.1-8.2); OS estimates for cohorts are shown in Table 1. Pts with brain mets FLT Baseline characteristics Pem n = 117 Gem n = 49 Tax n = 54 Vin n = 38 ≥70 yrs, % 12 25 20 21 ECOG PS 2-3, % 18 14 30 21 Never smoker, % 14 6 6 5 Non-squamous, % 97 69 70 82 Median OS, mos. (95% CI) 9.3 (6.2-11.9) 5.6 (4.1-8.4) 6.6 (3.7-7.8) 6.7 (5.2-9.3) Conclusions: Real-world data from the European FRAME study provide relevant information on NSCLC pts with brain mets, and may have implications on FLT management decisions for these pts. Disclosure: D. Moro-Sibilot: Consulting fees from Eli Lilly and Company, Roche, Astra Zeneca, Boehringer Ingelheim France, Amgen; J. De Castro Carpeno: Advisory board for Eli Lilly and Company, Roche and Pfizer; speakers bureau for Roche; K. Lesniewski-Kmak: Invited speaker for Eli Lilly and Company, Roche, Astra Zeneca, GSK, Amgen; J.G. Aerts: Research grant from and consultant for Eli Lilly and Company; advisory board for Eli Lilly and Company, Genentech, BMS; R. Villatoro: Speaker fee from Jansen, Pfizer; K. Kraaij: Employee and stockholder of Eli Lilly and Company; K. Nacerddine: Employee and stockholder of Eli Lilly and Company; Y. Dyachkova: Employee of Eli Lilly and Company; K. Smith: Employee and stockholder of Eli Lilly and Company; A. Girvan: Employee and stockholder of Eli Lilly and Company; C.M. Visseren-Grul: Employee and stockholder of Eli Lilly and Company; P.A. Schnabel: Research funding from Eli Lilly and Company. Speakers bureau for AstraZeneca, InterMune, Eli Lilly and Company, Novartis, Pfizer and Roche. Advisory board for Eli Lilly and Company, AstraZeneca, Novartis and Roche. All other authors have declared no conflicts of interest.

1247PANALYSIS OF PATIENTS TREATED WITH CISPLATIN OR CARBOPLATIN-BASED DOUBLET CHEMOTHERAPY IN THE EUROPEAN FRAME OBSERVATIONAL STUDY

ABSTRACT Aim: FRAME was a prospective observational study of patients (pts) receiving platinum (plt)-based chemotherapy as first-line treatment (FLT) for advanced or metastatic non-small cell lung cancer (NSCLC) across Europe between 2009 and 2012. Methods: The study primary objective of overall survival was previously reported. Here, we describe the association between baseline characteristics and the physicians choice of a cisplatin (cis) or carboplatin (cb) backbone for FLT observed in FRAME. The distribution of baseline characteristics was compared by t-test or k2-test and summarized by propensity of receiving cb estimated by logistic regression. Cohorts were matched by propensity score; survival for matched pts was compared using Kaplan-Meier approach. Results: In total, 1520 pts received either cis (n = 825) or cb (n = 695) in combination with either: pemetrexed, gemcitabine, taxanes or vinorelbine, and were analyzed here. Pts prescribed cb were on average older, had worse ECOG PS and were more likely to have squamous histology than pts who received cis (Table 1); they also had more pre-existing conditions (cardiovascular p Cis (n = 825) Cb (n = 695) p-value ≥70 yrs (%) 15 47 ECOG PS (%) Gr 0-1 Gr 2-3 87 12 77 22 FLT (+plt) (%) pemetrexed (n = 567) gemcitabine (n = 360) taxanes (n = 293) vinorelbine (n = 300) 47 23 9 21 26 24 31 19 Histology (%) Squamous (sq) Non-sq 23 75 26 70 .031 Conclusions: FRAME observed that age, ECOG PS, pre-existing conditions and pathological diagnosis have strong associations with the choice of cis or cb in FLT for advanced or metastatic NSCLC across Europe. Disclosure: D. Moro-Sibilot: Consulting fees from Eli Lilly and Company, Roche, Astra Zeneca, Boehringer Ingelheim France, Amgen; J. De Castro Carpeno: Speakers Bureau for Roche; Advisory board for Eli Lilly and Company, Roche and Pfizer; K. Kmak-Lesniewski: Invited speaker for Eli Lilly, Roche, Astra Zeneca, GSK, Amgen; J.G. Aerts: Research grant from and Consultant for Eli Lilly and Company. Advisory board for Eli Lilly and Company, Roche Genentech and BMS; R. Villatoro: Speaker fee from Jansen, Pfizer; K. Kraaij: Employee and stockholder of Eli Lilly and Company; K. Nacerddine: Employee and stockholder of Eli Lilly and Company; Y. Dyachkova: Employee and stockholder of Eli Lilly and Company; K. Smith: Employee of Eli Lilly and Company; A. Girvan: Employee and stockholder of Eli Lilly and Company; C.M. Visseren-Grul: Employee and stockholder of Eli Lilly and Company; P.A. Schnabel: Speakers Bureau for AstraZeneca, InterMune, Eli Lilly and Company, Novartis, Pfizer, Roche. Advisory board for AstraZeneca, Eli Lilly and Company, Novartis, Roche. Prior research funding from Eli Lilly and Company. All other authors have declared no conflicts of interest.

Prospective observational comparison of clinical outcomes between african-american and caucasian patients receiving second-line treatment with pemetrexed for advanced non-small-cell lung cancer.

INTRODUCTION This prospective observational study evaluated the effect of race on disease control rate (DCR), progression-free survival (PFS), and overall survival (OS) in patients with NSCLC treated with second-line pemetrexed. PATIENTS AND METHODS Eligibility criteria included stage IIIB or IV NSCLC patients receiving single-agent pemetrexed for second-line therapy in routine clinical practice. Noninferiority was evaluated using logistic regression analysis of DCR, controlling for predefined covariates. Noninferiority was considered if the upper 95% confidence bound on the adjusted odds ratio (OR) for Caucasian vs. African-American individuals was less than 1.78, corresponding to a difference in proportion of 14% assuming Caucasian individuals to have a DCR of approximately 50%. The bound was chosen to be half of the anticipated difference between treatment and no second-line treatment. PFS and OS were estimated using the Kaplan-Meier method. Tools were used to measure functional status and symptom burden. RESULTS The unadjusted DCR was 43.7% (117/268) for Caucasian and 45.0% (27/60) for African-American individuals (unadjusted OR, 0.95; 95% confidence interval [CI], 0.54-1.66). The adjusted OR in the final logistic regression model was 0.82 (95% CI, 0.43-1.58). This upper 95% confidence bound was within the prespecified acceptable bound of 1.78. Median PFS times (months) were 2.7 (95% CI, 2.4-3.4) for Caucasian and 3.0 (95% CI, 2.3-4.7) for African-American individuals (P = .91). Median OS times (months) were 6.7 (95% CI, 5.7-7.9) for Caucasian and 6.9 (95% CI, 4.5-8.9) for African-American individuals (P = .92). Baseline and functional status after baseline assessment and mean symptom burden did not differ substantially among races. CONCLUSION African-American race was not considered to be a significant predictor of disease control after second-line treatment with pemetrexed.

Abstract A66: Recommendations to address common barriers to diversity in oncology clinical trials: An industry perspective

Introduction: Racial and ethnic minority enrollment in clinical trials is still a challenge and current recruitment planning has failed to improve diverse representation in clinical trials. The following literature review identified common barriers to diverse trial enrollment resulting in suggested tactics to address these barriers for breast cancer clinical trials. Methods: A literature review was conducted to identify barriers to oncology clinical trial enrollment for African American, Hispanic, Asian American, and Native American populations. A targeted review of barriers for patients in the lower socioeconomic status (SES) bracket was also performed. An advisory board of study coordinators was consulted to assess site–level awareness, barriers, and resources needed from the pharmaceutical industry. Additionally, ongoing initiatives were assessed including use of a Latino toolkit, patient navigators, and community outreach strategies. Results: The number of evidence based recruitment strategies targeting ethnic minorities was found to be lacking. There are four key aspects that influence diverse patients9 decisions to participate in trials: family/friends, finances, faith, and physicians/staff. Barriers identified affected a majority of the analyzed patient populations. These included: protocol inclusion/exclusion criteria; logistical concerns; few incentives for participating; and cultural factors. The informed consent form and process were cumbersome and not patient-centric; costs related to travel and co-pays, and mistrust of medical research and the pharmaceutical industry were also identified as common barriers. The primary barriers were lack of invitations and lack of awareness at multiple levels within medical institutions, health care professionals, communities, and the pharmaceutical industry. Suggestions to address enrollment barriers include education to increase awareness, a patient-centric approach to consent and trial discussions, and tailoring site selection strategies. Other suggestions are immersion into communities, patient reimbursement, cultural competency training, pro-active translations, early recruitment planning, and identifying processes for approaching non-English speaking patients. Epidemiology data should also be incorporated in planning to establish baseline and enrollment goals. Conclusion: Although each of the analyzed populations is unique, common themes emerged. A patient-centric approach to increasing diversity in clinical trials should include considering a patient9s key influences, addressing common concerns, conducting a properly informed consent process, and empowering a patient9s decision making process. Gaps in current recruitment practices do still exist and further research on evidence based diverse recruitment strategies is also needed. Citation Format: Kelly R. Kirsch, Gebra C. Carter, Jacqueline M. Cole, Allicia C. Girvan, Coleman K. Obasaju. Recommendations to address common barriers to diversity in oncology clinical trials: An industry perspective. [abstract]. In: Proceedings of the Sixth AACR Conference: The Science of Cancer Health Disparities; Dec 6–9, 2013; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2014;23(11 Suppl):Abstract nr A66. doi:10.1158/1538-7755.DISP13-A66

Impact of FDA guideline changes on treatment patterns and cost among cancer patients with anemia who do not respond to erythropoietin stimulating agents (ESAs).

97 Background: Due to emerging evidence in 2007, the FDA issued warnings in the labeling of ESAs regarding use among cancer patients. Subsequently, ASCO/ASH clinical practice guidelines were also updated to recommend discontinuation of ESAs in patients who fail to respond within 6-8 weeks. METHODS Patients with breast, head and neck, cervical, non-small cell lung cancer, or multiple myeloma initiating ESAs during 1/1/2003 - 7/1/2011 were identified from EMRs in the ACORN Data Warehouse (ACORN, Memphis TN). The study included those with at least 6-8 weeks of ESA use with documented hemoglobin (Hb) values. Following the ASCO/ASH guidelines, patients were defined as non-responders if Hb increased < 1g/dL between baseline (≤ 28 days before ESA initiation) and response determination date (Hb assessment closest to day 56). Non-responders were further classified as continuers or discontinuers based on the presence of ESA treatment following determination of non-response. ESA use, adverse events, and costs were assessed using Chi-square test for comparisons. RESULTS A total of 1,198 patients were classified as non-responders (591 in pre-2007 and 607 in post-2007). ESA continuation rate was lower for post-2007 compared to pre-2007 (75% vs. 90%, p < 0.0001). CONCLUSIONS The percentage of patients continuing ESA treatment in the absence of Hb response has decreased since 2007, yet still remains high despite guideline recommendations. It is not clear from these data whether the increased risks and cost burden in these patients were offset by other positive outcomes. [Table: see text].

Abstract B77: Asian-American response to pemetrexed: Results from an observational study of second-line treatment of non-small cell lung cancer.

Background: The association of race with clinical outcomes has been increasingly investigated in non-small cell lung cancer (NSCLC). This prospective observational study evaluated the impact of race on disease control rate (DCR) in patients with NSCLC treated with second-line pemetrexed. Our previous report of this study showed the DCR of African-Americans was non-inferior to that of Caucasians (C). This report compares outcomes for Asian-Americans (AS) and C. Methods: Patients with stage IIIB/IV NSCLC who received one prior chemotherapy regimen and second-line therapy with pemetrexed were eligible. The primary endpoint was DCR (complete/partial response or stable disease). Secondary endpoints were overall survival (OS), progression-free survival (PFS), and adverse events. Logistic regression was used to analyze the DCR after therapy, controlling for 15 predefined covariates including prognostic and socioeconomic factors. Survival was estimated using Kaplan-Meier and compared with a log-rank test. Analyses were not adjusted for multiple race comparisons. Results: Challenges in enrolling minorities led to a lower than anticipated enrollment (AS=37 vs. 200 planned; C=304 vs. 400 planned), leaving a small sample size from which to draw conclusions. The mean age was 65 years (range, 34-90) for AS and 66 years (range, 37-86) for C. There were 17 males/20 females in the AS group and 161 males/143 females in the C group. Most patients had stage IV disease (AS=78.4%; C=80.3%). A higher percentage of AS had adenocarcinoma (AS=89.2%; C=62.8%; p=0.004) and a higher percentage of C were current smokers (AS=0.0%; C=27.7%; p=0.0002). The mean number of cycles received was 5.8 for AS and 4.1 for C. For the evaluable patients (AS=30; C=268), the DCR was numerically higher for AS (53.3% vs. 43.7%, OR=0.609, 95% CI: 0.253-1.469), but this difference was not statistically significant (p=0.270). AS had a significantly longer unadjusted median OS and PFS than C: OS of 16.0 (95% CI: 5.75-16.0) vs. 6.73 months (95% CI: 5.65-7.92), p=0.041; PFS of 5.08 (95% CI: 2.50-11.1) vs. 2.66 months (95% CI: 2.40-3.35), p=0.045. Adverse events, regardless of causality, occurred in <15% of patients, except for fatigue (AS=60%; C=68%). Conclusions: Despite the small sample size, this study provides insight into characteristics and outcomes of AS treated with second-line pemetrexed. Citation Format: A A. Adjei, E Pennella, A C. Girvan, G Peltz, G Pohl, C Obasaju, K Winfree, M S. Walker, E J. Stepanski, L S. Schwartzberg. Asian-American response to pemetrexed: Results from an observational study of second-line treatment of non-small cell lung cancer. [abstract]. In: Proceedings of the Fifth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2012 Oct 27-30; San Diego, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2012;21(10 Suppl):Abstract nr B77.