Budget Impact Analysis Of Darbepoetin Alfa Every 3 Weeks Versus Epoetin Alfa Every Week For Cancer Patients Receiving Chemotherapy From A Us Payer's Perspective

VALUE IN HEALTH 16 (2013) A1-A298

treatment and survival. RESULTS: A total of 2484 patients in the linked database met the study criteria. Mean age was 56.9 (SD=14.4) and 63.7% were male. The mean Deyo Charlson Comorbidity Index score was 0.59 (SD=1.11) over the six month pre-index period (modified to exclude malignancy). Overall, median survival time was 976 days (95% confidence interval: 847, 1194). A greater proportion of women (75%) than men (67%) were alive 365 days after their index surgery. Survival time decreased steadily with age as 98% of patients age 19-34 survived for at least 365 days after surgery compared to 95% for patients 35-44, 84% for patients 45-54, 70% for patients 55-64 and 43% for patients over 65. Most patients (52.1%) received external beam radiation, while 42.2% received temozolomide, 5.4% received a carmustine wafer implant, 2.0% received chemotherapy (including bevacizumab) and 1.3% received stereotactic radiosurgery within 90 days of their index surgery. CONCLUSIONS: Linking claims and public death records provided results in large populations that were similar to those in clinical trials and observational literature, both in terms of survival and treatment patterns. PCN20 RISK OF DEATH AT ONE YEAR IN PATIENTS WITH NON-SMALL CELL LUNG CANCER (NSCLC) TREATED WITH CISPLATIN REGIMENS: AN INDIRECT COMPARISON META-ANALYSIS BASED ON RENAL ELIGIBILITY CRITERIA Bellows BK1, Dahal A1, Agarwal N2 1 University of Utah, Salt Lake City, UT, USA, 2Huntsman Cancer Institute University of Utah, Salt Lake City, UT, USA

OBJECTIVES: The risk of nephrotoxicity with cisplatin may be lower when glomerular filtration rate (GFR) is used as eligibility criteria compared to serum creatinine (SCr), but the impact on survival remains unknown. The objective of this meta-analysis was to indirectly compare the relative risk (RR) of death at one year in trials using cisplatin for NSCLC when renal function was assessed using either SCr or GFR for eligibility criteria. METHODS: Randomized trials comparing cisplatin to non-cisplatin regimens from 1990-2010 were identified in PubMed. Included studies used SCr or GFR as inclusion criteria, reported incidence of WHO or NCI grade ≥3 nephrotoxicity, and reported survival rate. Review articles, observational and non-randomized studies, phase 1 trials, studies not reported in English or without a comparator group were excluded. The number of patients surviving at one year was calculated from one-year survival rates. RR of death for cisplatin versus non-cisplatin regimens was estimated using random effects methods with sub-group analyses performed on studies using SCr, GFR, or either SCr or GFR as eligibility criteria. RESULTS: The literature search identified 2,359 studies of which 17 met inclusion/exclusion criteria (N=4,177). Of these, 9 studies used SCr (N=2,685), 1 used GFR (N=111), and 7 used SCr or GFR (N=1,381) for screening. Overall, RR of death at one year with cisplatin versus non-cisplatin treatment was 1.05 (95%CI 0.96-1.15, p=0.28). In sub-group analyses, RR was 1.15 (95%CI 0.98-1.34, p=0.08) for SCr, 0.90 (95%CI 0.75-1.08, p=0.27) for GFR, and 1.00 (95%CI 0.92-1.09, p=0.95) for either SCr or GFR. RR of death when GFR was indirectly compared to SCr was 0.78 (95%CI 0.62-0.99, p=0.04). CONCLUSIONS: Using GFR as eligibility criteria for cisplatin treatment appeared to reduce the risk of death at one year compared to SCr. However, results should be interpreted cautiously as only one study using GFR was included. PCN21 IMPACT OF HORMONE RECEPTOR STATUS AND HER2 ON SURVIVAL OF TRIPLE NEGATIVE BREAST CANCER PATIENTS Xiao H1, Tan F2, Huang Y3, Feldman J3, Koniaris L4, Gummadi S4, Adunlin GB1, Ali AA1 1 Florida A&M University, Tallahassee, FL, USA, 2Indiana University-Purdue University, Indianapolis, IN, USA, 3Florida Department of Health, Tallahassee, FL, USA, 4Thomas Jefferson University, Philadelphia, PA, USA

OBJECTIVES: To determine the impact of hormone receptor status and HER2 on survival for patients with triple negative breast cancer. METHODS: Electronic medical records from a network of 9 hospitals were linked to female breast cancer patients who were diagnosed between 2007 and 2010 in Florida. Cox proportional hazards model was used to assess cause-specific survival. RESULTS: Two-year survival of the study population was 95.1%. Median followup time for those who died due to breast cancer was 569 days. Log-rank test indicated survival functions between blacks and non-blacks were significantly different over time (p< 0.0001). KM survival curves revealed that survival probability among black was lower than that of non-blacks. Hazard of breast cancer death among patients with triple negative and patients with unknown triple negative status were 4.34 and 2.35 times that of non-triple negative patients. Death rate among blacks was 1.6 times than non-blacks over time. Other factors associated with an increased hazard over time were being diagnosed in regional stage or having unknown diagnosis stage, having poorly or un-differentiated tumor, being Medicare user, being single, with larger tumor size, and with more positive nodes detected. Immediately after diagnosis distant stage and more comorbidity conditions were associated with an elevated risk, whereas more lymph nodes examined was associated with reduced risk of breast cancer death. Effects of distant stage, total comorbidity, and total lymph nodes examined gradually attenuated over time. CONCLUSIONS: This study highlights the importance of improvement of care for patients with triple negative biomarkers. CANCER – Cost Studies PCN22 DUAL-ACTING OSMOTIC AND STIMULANT LAXATIVE FOR BOWEL CLEANSING IN AN ELDERLY POPULATION: A US PAYER BUDGET IMPACT ANALYSIS

A131

Cyr PL1, Sidhu M2, Hussain R3, Hromin T3, Jensen I1, Kuan R4 1 PriceSpective LLC, Cambridge, MA, USA, 2Oxford Outcomes, Morristown, NJ, USA, 3Ferring Pharmaceuticals, Parsippany, NJ, USA, 4PriceSpective LLC, El Segundo, CA, USA

OBJECTIVES: Joint guidelines recommend colorectal cancer (CRC) screening every 10 years in average-risk adults beginning at age 50 years including colonoscopy (CSPY) where proper bowel preparation is critical for quality screening. The aim of our analysis was to quantify the budget impact on US payers of introducing a dual-acting osmotic and stimulant laxative for bowel cleansing, sodium picosulfate/magnesium citrate ( P/MC) in individuals 65 years and older. METHODS: A decision analytic model was developed to estimate the impact on direct medical costs of P/MC utilization in CRC screening by CSPY (2% and 12% in years one and three, respectively). Standard clinical practice was represented through a decision tree based on clinical guidelines and included utilization of currently prescribed bowel cleansing products (MoviPrep, HalfLytely, SuPrep, 4L PEG). Data from RCTs were used to quantify the adequacy of bowel cleansing. Prep costs were based on 2012 wholesale acquisition costs. Costs of complete, incomplete and repeat colonoscopies were obtained from Medicare claims analyses. RESULTS: For every 100,000 individuals 65 years of age and older who undergo colonoscopy, the use of P/MC demonstrated cost neutrality when used by 2% of subjects, yielding annual incremental savings of $86,555 ($213,016,329 before introduction vs. $212,929,775 after introduction). If P/MC use increases to 12% in year three, the annual estimated incremental savings per 100,000 cases increased to $333,846. Cost savings are mainly due to a reduction in repeat colonoscopies (-$439,904 year one and -$572,792 year three). One-way sensitivity analysis demonstrated the model to be most sensitive to P/MC drug cost and adequacy of cleansing when using generic 4L PEG. CONCLUSIONS: The introduction of P/MC into CRC screening practice in a 65 year and older population is cost neutral from the US payer perspective with moderate cost savings which becomes greater with increased utilization. PCN23 EGFR TEST-GUIDED ERLOTINIB VERSUS CHEMOTHERAPY FOR FIRST-LINE TREATMENT OF ADVANCED NON-SMALL CELL LUNG CANCER: A BUDGET IMPACT ANALYSIS Bajaj PS1, Carlson JJ1, Goertz HP2, Veenstra DL1 1 University of Washington, Seattle, WA, USA, 2Genentech, Inc., South San Francisco, CA, USA

OBJECTIVES: A recent phase III clinical trial, EURTAC, demonstrated that firstline treatment with erlotinib significantly improved progression-free survival compared with standard chemotherapy in advanced non-small cell lung cancer (NSCLC) patients whose tumors harbored epidermal growth factor receptor (EGFR) mutations. We sought to estimate the budgetary impact of adding coverage for erlotinib in this patient population. METHODS: We developed a budget impact model from the U.S. health plan perspective to compare EGFR test guided treatment with erlotinib (erlotinib for EGFR-positive patients and standard chemotherapy for EGFR-negative patients), compared to treatment with the standard chemotherapy without testing. Standard chemotherapy included recommended chemotherapy regimens as defined by the National Comprehensive Care Network guidelines. The target population was estimated using data extracted from the SEER 17 registries. Treatment duration, dosage, and adverse event data were derived from clinical trial data and manufacturer prescribing information. Cost data were obtained from the Centers for Medicare and Medicaid Services payment rates, and drug costs were based on wholesale acquisition costs. Sensitivity analyses were conducted to assess uncertainty. RESULTS: Overall health plan expenditures increased by $0.031 per member per month (PMPM). This increase was largely attributable to increased drug costs for patients receiving erlotinib who experienced longer progression-free survival and treatment duration. EGFR test cost also contributed to the increased expenditures, whereas adverse event costs reduced the incremental expenditures. The most influential model parameters were drug cost and treatment duration, however the budget impact did not exceed $0.045 PMPM in the sensitivity analyses. CONCLUSIONS: EGFR testing and treatment with erlotinib in 1st line NSCLC likely results in a relatively small budget impact for U.S. health plans, while providing a personalized medicine approach to lung cancer treatment. PCN24 BUDGET IMPACT ANALYSIS OF DARBEPOETIN ALFA EVERY 3 WEEKS VERSUS EPOETIN ALFA EVERY WEEK FOR CANCER PATIENTS RECEIVING CHEMOTHERAPY FROM A US PAYER'S PERSPECTIVE McGarvey N1, Shreay S2, Xu H2, Corey-Lisle P2 1 UCLA Fielding School of Public Health, Los Angeles, CA, USA, 2Amgen, Thousand Oaks, CA, USA

OBJECTIVES: To estimate the annual budget impact of drug treatment associated with treating cancer patients with anemia due to the effect of concomitant myelosuppressive chemotherapy (CIA) with erythropoiesis stimulating agents (ESAs), either darbepoetin alfa (DA) once every three weeks (Q3W) or epoetin alfa (EA) once every week (QW), for a large US health plan in 2010. METHODS: A retrospective database analysis of administrative claims data for US commercially-insured individuals from January 1, 2009 to December 31, 2010 was conducted to estimate prevalence of CIA among patients receiving chemotherapy treatment. Using this information, a patient database from a large US health plan was adjusted to estimate budget impact of ESA treatment on this patient population (1,755 patients each per DA and EA) in 2010. The analysis assumed: 1) a minimum of two additional months of chemotherapy and 2) that administration costs equal the sum of office visit and injection costs. The 2010 Centers for Medicare and Medicaid Services reimbursement rates used were: average sales price +12% of $3.06/mcg (DA) and $10.23/1,000 IU (EA), office-based

A132

VALUE IN HEALTH 16 (2013) A1-A298

injection cost of $23.10, and office visit cost of $42.45. RESULTS: The annual average drug costs per patient were $4,590 (DA) and $5,244 (EA). Average drug costs for administrations were $92 (DA) and $277 (EA). Per member per year (PMPY) costs were $4,682 (DA) and $5,521(EA). The annual total costs per population (n=1,755) were $8,217,612 (DA) and $9,688,952 (EA). CONCLUSIONS: DA Q3W has the potential to provide cost savings over EA QW in terms of annual average drug cost per patient ($654 savings), per member per year ($839 savings), and total cost per population ($1,471,340 savings). DA Q3W may offer a cost advantage over EA QW as it allows for synchronizing of anemia management with ongoing cancer treatments, which may reduce required patient visits and blood tests. PCN25 BUDGET IMPACT ANALYSIS (BIA) OF CABAZITAXEL INTRODUCTION IN TREATMENT OF METASTATIC HORMONE-REFRACTORY PROSTATE CANCER (MHRPC) IN RUSSIAN FEDERATION Yagudina R1, Kulikov A1, Kogon L2 1 I.M. Sechenov First Moscow State Medical University, Moscow, Russia, 2FGBU NCESMP Ministry of Health Russia, Moscow, Moscow, Russia

OBJECTIVES: The purpose of this study was to evaluate financial impact on Russian National Healthcare Service (RNHS) budget of the introduction of Cabazitaxel in second-line treatment of patients with mHRPC previously treated with a Docetaxel-containing regimen in first-line. METHODS: The numbers of mHRPC patients eligible for first-line and second-line treatment were estimated based on local demographic figures, incidence data and experts opinion. A BIA model combined the data of official algorithms (standard of treatment approved by Ministry of Health of Russian Federation) and guidelines of mPC management, as well as local expert opinion and published data on resource use and unit costs from published sources. The BIA model stimulates the impact of mHRPC secondline management with or without Cabazitaxel. Resources to manage first-line and second-line treatment included standard chemotherapy, premedication, G-CSF prophylaxis, concomitant medications, management of adverse events, visits and diagnostics. RESULTS: It was estimated that in the first three years following Cabazitaxel introduction in second-line mHRPC patients treatment regimen, just 104, 259 and 311 out of 1036 eligible patients with mHRPC previously treated with a Docetaxel-containing regimen in first-line would be placed on Cabazitaxel therapy respectively. It would imply moderate incremental cost of 121.8, 304.6 and 365.5 mln RUB for second-line treatment of mHRPC respectively. From the other hand cost for first-line treatment would significantly decline due generic erosion of Docetaxel, resulting in following cost savings 248.8, 289.0 and 325.2 mln RUB, respectively. As the result the total incremental budget increase will be 15,6 and 40,3 mln RUB on second and third years of Cabazitaxel introduction, respectively. CONCLUSIONS: Because of small size of estimated candidate population the incremental financial budget impact on RHNS following the introduction of Cabazitaxel would be very small and offset by savings from first-line treatment due to generic erosion of Docetaxel price. PCN26 ESTIMATING THE POTENTIAL BUDGET IMPACT OF ZELBORAF (VEMURAFENIB) FOR ADVANCED MELANOMA TREATMENT IN BRAZILIAN PRIVATE HEALTH CARE SYSTEM Tsuchiya CT, Buschinelli CT, Maximo MFM, Tobaruella FS, Borges LG Roche Brazil, São Paulo, Brazil

OBJECTIVES: Melanoma accounts for less than 5% of skin cancers, but is responsible for 90% of skin cancer-related deaths worldwide. According to Brazilian Institute of Cancer 2012 statistics, it was estimated 6,230 cases of melanoma. BRAF mutations occur in approximately 50% of cases, being associated with poorer patient prognosis in advanced melanoma. Historically, there have been limited treatment options for advanced melanoma, resulting in a critical unmet clinical need for more effective therapies. Vemurafenib is a novel targeted therapy, effective for BRAF-V600 mutation-positive unresectable or metastatic melanoma treatment, orally administrated and available in Brazilian market since 2012. Therefore, the aim of this study was estimate the economic impact of vemurafenib reimbursement in Brazilian Private Healthcare System Budget. METHODS: Based on an epidemiologic approach, the potential number of patients for vemurafenib was estimated. Only the private market was considered, accounting for 40% of all patients and only drug costs were evaluated. The ex-factory price and labeled dose of 960mg b.i.d. were used. Average therapy duration of 6 months was assumed. Costs were reported in Brazilian Reais (BRL1.00~USD0.48 Dec. 2012). A total health assistance budget of BRL67.9 billion was considered, according to Brazilian National Agency for Supplementary Health data from 2012. RESULTS: A total of 756 cases of advanced melanoma are expected in 2013 in the private system, corresponding potentially to 378 patients harboring BRAF-V600E mutation. Treating all potential patients with vemurafenib would yield a total drug cost of BRL59.101.395, corresponding to a potential budget impact of 0.087%, considering health assistance budget. Cost savings owing to oral administration was not considered. CONCLUSIONS: By identifying the patients with BRAF-V600E mutation, therapy can be targeted to those who present a higher chance to respond to treatment, resulting in a potential low impact of vemurafenib in private health care system budget, mainly because of its very selected and specific eligible population. PCN27 PHARMACOECONOMIC ANALYSIS OF DOCETAXEL IN THE ADJUVANT THERAPY OF BREAST CANCER

Zinchuk I1, Yagudina R2, Kulikov A2 1 First MGMU named after I.M. Sechenov, Moscow, Russia, 2First Moscow State Medical University named after I. M. Sechenov, Moscow, Russia

OBJECTIVES: To identify the dominant scheme of adjuvant breast cancer therapy (comparing the two chemotherapy regimens - 6FAÑ (6 cycles: 5-Fluorouracil + Doxorubicin + Cyclophosphamide) vs 6DAC (6 cycles: Docetaxel + Doxorubicin + Cyclophosphamide)). METHODS: Based on the Markov model, the cost of treatment for breast cancer was evaluated; an â òî æå âgåìÿ analysis of direct and indirect costs was conducted. Direct costs included cost of adjuvant chemotherapy (6FAC and 6DAC), medical services, adjuvant endocrine therapy, cardiac monitoring, current therapy cost and cost of treatment complications. Indirect costs included cost of patient’s disability, maintenance of orphans, GDP losses caused by mortality and disability and sick-pay. RESULTS: Based on a clinical study BCIRG 001 (Martin M, Mackey J, Pienkowski T) and as an outcome of Markov simulation it was determined that the use of 6DAC for adjuvant therapy in 10 year horizon of research allows to increase the disease-free survival by 6.09%, and overall survival by 5,88% comparing to 6FAC. Use of 6DAC instead of 6FAC allows to reduce the number of patients with local relapse, regional relapse and metastasis insignificantly as well. In the result of direct costs analysis it was determined that for 10-years horizon cost rate for the treatment of one patient with breast cancer will be $66,026 for 6DAC and $60,845 for 6FAC. Direct and indirect costs rate of the breast cancer treatment were $84,818 and $85,991 for the 6DAC and 6FAC schemes respectively. CONCLUSIONS: It was determined that the scheme of adjuvant therapy that includes Docetaxel besides of addition efficiency, allows to reduce the cost rate to $1173 in 10 years. PCN28 BUDGET IMPACT OF CRIZOTINIB FOR ALK-POSITIVE ADVANCED NON-SMALL CELL LUNG CANCER (NSCLC) IN COLOMBIA Rosado-Buzzo A1, Garcia-Mollinedo L1, Luna-Casas G1, Mould-Quevedo JF2, Gutierrez-Ardila MV3, Vargas Zea N3 1 Links & Links S.A, de C.V., Mexico City, Mexico, 2Pfizer, Inc., New York, NY, USA, 3Pfizer S.A.S., Bogota, Colombia

OBJECTIVES: Lung cancer (LC) is one of the major health problems due its high mortality. The most common type of LC is non-small cell lung cancer (NSCLC), which accounts for approximately 80% of all LC cases; approximately 3%-5% of these have a gene defect called ALK (anaplastic lymphoma kinase). The aim of this study was to evaluate the budgetary impact of adding crizotinib for patients with ALK-positive advanced NSCLC in Colombia. METHODS: A budget-impact model was developed to evaluate crizotinib from the payer perspective. Comparators were: crizotinib (250mg bid), erlotinib (150mg/day), bevacizumab/platinum doublet (15mg/kg/Carboplatin AUC6+Paclitaxel 200mg/m2 per cycle [28 days]), paclitaxel/platinum (Carboplatin AUC6+Paclitaxel 200mg/m2 per cycle), docetaxel/platinum (Cisplatin 75mg/m2 + Docetaxel 75mg/m2 per cycle), pemetrexed/platinum (Carboplatin AUC6+Pemetrexed 500mg/m2 per cycle), gemcitabine/platinum (Cisplatin 75mg/m2 day1+Gemcitabine 2,500mg/m2per cycle). Two scenarios were compared: (1) no tested patients, excluding crizotinib (2) all patients tested with FISH (diagnostic test) to identify ALK-positive NSCLC patients; it was taken into account a positive match for ALK of 4.2% and only 3% receiving crizotinib. Epidemiology and acquisition, administration and adverse events costs were estimated from Colombian sources, values are expressed in 2012 US$. RESULTS: Considering Colombian population >18 years (incident rate of LC 0.018%, proportion NSCLC 87.86% and advanced NSCLC 52.29%), it was estimated a cohort of 4768 advanced NSCLC patients, overall 1-year costs of treating patients with NSCLC would be US$197,929,584 with crizotinib, compared with US$190,725,844 without crizotinib. Acquisition costs of crizotinib were offset by reductions in adverse events (AE) costs (anemia, anorexia, asthenia, diarrhea, dyspnea, fatigue, neutropenia, neutropenic infection, nausea, vomiting, pulmonary events,). Administration costs accounted between 37.77%-38.16% of total costs, while AE management costs made up 1.19% to 1.27% of total costs. CONCLUSIONS: Treating ALK-positive advanced NSCLC patients with crizotinib, leads to a decrease in total costs of managing adverse events (US$65,042) and progression costs. PCN29 BUDGET IMPACT MODEL FOR CUTANEOUS T-CELL LYMPHOMA Aggarwal S, Topaloglu H Novel Health Strategies, Bethesda, MD, USA

OBJECTIVES: Develop budget impact model to forecast total cost of treatment for cutaneous T-cell lymphoma (CTCL) for U.S. public and private payer. METHODS: The clinical efficacy and safety data were obtained from the published pivotal study results. Costs of adverse events were estimated based on claims database analysis, AHRQ’s HCUP, and CMS Medicare 2009 databases. Drug cost was estimated based on 2011 AWP price. Epidemiology data was obtained from NCISEER and CDC databases. A budget impact model was implemented over a period of five years, based on a stable population and on different penetration and substitution rates of newly approved therapy. Model was developed in excel based format. Blinded Model design and outputs were tested with payers and KOLs. RESULTS: For rare cancers such as CTCL, the budget impact of treatment with targeted cancer therapies is in the range of $460,000-$530,000 per 1 million covered lives. The per patient per member (PPPM) budget impact of this treatment is 46-53 cents. U.S. payers rated PPPM output as the one of the most important relevant outputs of model. CONCLUSIONS: This budget impact model shows that new treatments for rare forms of cancer are likely to have minimal budget impact on payers. PPPM based outputs are more relevant to payers, than per patient treatment costs. However, an emerging concern is the total budget