The implementation of value-based frameworks, clinical care pathways, and alternative payment models for cancer care in the United States

BACKGROUND: Cancer treatment is a significant driver of rising health care costs in the United States, where the annual cost of cancer care is estimated to reach $246 billion in 2030. As a result, cancer centers are considering moving away from fee-for-service models and transitioning to value-based care models, including value-based frameworks (VBFs), clinical care pathways (CCPs), and alternative payment models (APMs). OBJECTIVE: To assess the barriers and motivations for using value-based care models from the perspectives of physicians and quality officers (QOs) at US cancer centers. METHODS: Sites were recruited from cancer centers in the Midwest, Northeast, South, and West regions in a 15/15/20/10 relative distribution. Cancer centers were identified based on prior research relationships and known participation in the Oncology Care Model or other APMs. Based on a literature search, multiple choice and open-ended questions were developed for the survey. A link to the survey was emailed to hematologists/oncologists and QOs at academic and community cancer centers from August to November 2020. Results were summarized using descriptive statistics. RESULTS: A total of 136 sites were contacted; 28 (21%) centers returned completed surveys, which were included in the final analysis. 45 surveys (23 from community centers, 22 from academic centers) were completed: 59% (26/44), 76% (34/45), and 67% (30/45) of physicians/QOs respondents had used or implemented a VBF, CCP, and APM, respectively. The top motivator for VBF use was "producing real-world data for providers, payers, and patients" (50% [13/26]). Among those not using CCPs, the most common barrier was a "lack of consensus on pathway choices" (64% [7/11]). For APMs, the most common difficulty was that "innovations in health care services and therapies must be adopted at the site's own financial risk" (27% [8/30]). CONCLUSIONS: The ability to measure improvements in cancer health outcomes was a large motivator for implementing value-based models. However, heterogeneity in practice size, limited resources, and potential increase in costs were possible barriers to implementation. Payers need to be willing to negotiate with cancer centers and providers to implement the payment model that will most benefit patients. The future integration of VBFs, CCPs, and APMs will depend on reducing the complexity and burden of implementation. DISCLOSURES :Dr Panchal was affiliated with the University of Utah at the time this study was conducted and discloses current employment with ZS. Dr McBride discloses employment with Bristol Myers Squibb. Dr Huggar and Dr Copher report employment, stock, and other ownership interests in Bristol Myers Squibb. The other authors have no competing interests to disclose. This study was funded by an unrestricted research grant from Bristol Myers Squibb to the University of Utah.

Economic burden during remission and after relapse among older patients with newly diagnosed acute myeloid leukemia without hematopoietic stem cell transplant: A retrospective study using the SEER-Medicare database

Acute myeloid leukemia (AML) is associated with a substantial clinical and economic burden. This study characterized the magnitude of this burden following initial treatment with standard or less intensive therapies (hypomethylating agents [HMAs]) and throughout different treatment phases post-remission. The Surveillance, Epidemiology, and End Results (SEER) cancer registry (2007-2016) linked with Medicare beneficiary claims (2007-2015) was analyzed. Patients were ≥ 65 years old with AML who initiated chemotherapy or HMAs and achieved remission. Outcomes included baseline characteristics, treatment patterns, clinical outcomes, healthcare resource utilization (HRU), and costs (2019 United States dollar). Economic impacts were stratified by treatment phase (initial treatment, early post-remission, late post-remission, and post-relapse). Early and late post-remission were defined as treatment initiated ≤ 60 days and > 60 days following initial treatment, respectively. A subgroup analysis of patients receiving only HMAs as initial treatment was also conducted. Overall, 530 patients were included (mean age: 74.1 years; 53.6 % male). In the overall analysis, 68.1 % of patients received post-remission treatment; 31.9% had no post-remission treatment. Mean monthly per patient healthcare costs by treatment phase were $45,747 (initial treatment), $30,248 (early post-remission), $23,173 (late post-remission), and $37,736 (post-relapse), driven predominantly by inpatient visits. The HMA subgroup analysis comprised 71 patients (mean age: 78.8 years; 50.7 % male); mean monthly per patient healthcare costs were highest post-relapse. The economic burden of AML among older patients is substantial across all treatment phases. AML treatments that induce and prolong remission may reduce HRU and the economic burden of disease.

Real-world clinical outcomes with enasidenib in relapsed or refractory acute myeloid leukemia

Enasidenib was approved by the Food and Drug Administration in 2017 for the treatment of patients with relapsed or refractory (RR) acute myeloid leukemia (AML) with an isocitrate dehydrogenase-2 (IDH2) mutation. Given limited data in clinical practice, this study assessed real-world clinical outcomes and healthcare resource use in patients with RR AML. Physicians performed chart abstraction of patients with RR IDH2-mutated AML treated with enasidenib (between 1/2018 and 6/2019) or other first-line (1 L) RR therapy (between 1/2016 and 7/2017). Progression-free survival (PFS) and overall survival (OS) were estimated by the Kaplan-Meier method, and adjusted risk of progression and death were estimated by multivariable Cox proportional hazard models. Among 124 patients treated with enasidenib and 76 patients treated with other 1 L RR therapy, overall response rate was higher among patients treated with enasidenib vs. other 1 L RR therapies (77% vs. 52%, p < 0.01). After a median follow-up of 9 and 6 months, median PFS was 8 months in enasidenib-treated patients and 5 months in patients receiving other 1 L RR therapy, respectively (adjusted HR=0.36, 95% CI: 0.23-0.57, p < 0.01). Median OS was 11 and 6 months in enasidenib-treated patients and patients receiving other 1 L RR therapy, respectively (adjusted HR=0.37, 95% CI: 0.22-0.60, p < 0.01). Fewer enasidenib-treated patients were hospitalized during 1 L RR therapy vs. those receiving other therapies (14% vs. 46%, p < 0.01). Results from this real-world study confirm the effectiveness of enasidenib among patients with IDH2-mutated RR AML and demonstrate that hospitalizations were significantly lower vs. other 1 L RR treatment in clinical practice.

Economic burden in US patients with newly diagnosed acute myeloid leukemia receiving intensive induction chemotherapy

Aim: This retrospective, observational study assessed healthcare resource utilization (HCRU) and costs for newly diagnosed acute myeloid leukemia (AML) patients receiving intensive induction chemotherapy. Materials & methods: Adult AML patients with inpatient hospitalization or hospital-based outpatient visit receiving intensive induction chemotherapy (CPX-351 or 7 + 3 treatments) were identified from the Premier Healthcare Database (US). Results: All 642 patients had inpatient hospitalizations (median number = 2; median length of stay = 16 days); 22.4% had an ICU admission. Median total outpatient hospital cost was US$2904 per patient, inpatient hospital cost was $83,440 per patient, and ICU cost was $16,550 per patient. Discussion: In the US hospital setting, substantial HCRU and costs associated with intensive induction chemotherapy for AML were driven by inpatient hospitalizations.

The impact of remission duration on the long-term economic burden of acute myeloid leukemia among patients without hematopoietic stem cell transplant in the United States

BACKGROUND AND AIMS

Acute myeloid leukemia (AML) prognosis is poor, with sustained remission occurring in <35% of young adults and <15% of older adults. This descriptive study examined the potential benefit of prolonged remission on the economic burden of AML.

METHODS

Using the IBM MarketScan Commercial and Medicare Supplemental databases, we identified newly diagnosed patients with AML without hematopoietic stem cell transplantation from January 1, 2012 to December 31, 2018; AML diagnosis was the index date. Patients had 6 months of pre-index eligibility and were followed until the end of continuous eligibility, study data, or death. Active treatment and supportive care cohorts were defined; duration-of-remission subgroups (0 to <3, 3 to <6, 6 to <12, and ≥12 months) were established among active treatment patients with remission. Healthcare service utilization and costs were reported over follow-up and mutually exclusive treatment, remission, and post-relapse periods.

RESULTS

This study included 1,558 active treatment and 1,127 supportive care patients who were followed for a median of 232 and 62 days, respectively. Over follow-up, active treatment and supportive care patients incurred mean ± standard deviation all-cause healthcare costs of $55,723 ± $61,994 and $68,596 ± $100,375 per-patient-per-month (PPPM), respectively. Decreasing PPPM costs were observed with increased remission duration (0 to <3 months: $71,823 ± $62,635; 3 to <6 months: $54,262 ± $44,734; 6 to <12 months: $35,287 ± $23,699; and ≥12 months: $15,615 ± $10,560). Although median follow-up varied by up to 5-fold, total costs were largely similar across duration-of-remission subgroups (0 to <3 months: $438,569 ± $332,675; 3 to <6 months: $590,411 ± $598,245; 6 to <12 months: $482,902 ± $369,115; and ≥12 months: $448,867 ± $316,133).

CONCLUSIONS

The economic burden of AML is substantial, even among untreated patients. Further, among patients with remission, longer durations in remission are associated with reduced PPPM healthcare costs, suggesting that remission-prolonging treatments could help mitigate healthcare costs.

OUP accepted manuscript

Background

This study analyses treatment patterns, health care resource utilization (HCRU), and costs in patients with myelofibrosis (MF) and a subgroup treated with ruxolitinib (RUX).

Materials and Methods

Treatment patterns, all-cause and MF-related HCRU, and costs were analyzed in adults with MF with continuous enrollment in a commercial or the Medicare Advantage health plan in the pre-index period, defined as the 12 months immediately prior to the index date (date of primary or secondary MF diagnosis), and the post-index period, defined as ≥6 months following the index date. In a subgroup analysis, outcomes were analyzed in patients treated with optimal RUX (OPT RUX, ≥30 mg) and suboptimal RUX (SUB RUX, <30 mg) in the pre-index RUX period, defined as the 3 months immediately prior to the index RUX date (first date for an RUX claim), and the post-index RUX period, defined as ≥6 months following the index RUX date.

Results

Of 2830 patients with an MF diagnosis, 1191 met eligibility requirements. The median age of patients was 72 years, 54% were male, and comorbidities were frequent. Sixty percent of patients received ≥1 line of therapy (LOT), of which 46% (n = 331) had ≥2 LOTs during the post-index MF period. Costs increased considerably 6-month pre-index to 6-month post-index (all-cause: cause ($24,216 to $48,966) and MF-related ($16,502 to $39,383), driven by inpatient stays and pharmacy costs. In the subgroup analysis, patients treated with RUX (n = 495) experienced significant disease burden and high costs, regardless of dose. A shorter duration of therapy and a higher rate of discontinuation were observed in patients treated with SUB RUX (n = 191) versus OPT RUX (n = 304).

Conclusion

These findings suggest a significant disease and economic impacts associated with MF patients that persists with RUX therapy, highlighting the need for additional therapeutic options for MF.

Challenges and opportunities of alternative payment models for physicians/quality officers in community/academic oncology practice sites via a national prospective survey

46

Background: Alternative payment models (APMs) aim to address the rising cost of cancer care ($173 billion in 2020) by promoting high-value care. This study explored the challenges and opportunities associated with implementation of APMs in oncology practice sites across the U.S. Methods: A novel survey was created through literature review and a focus group of content experts and researchers. This survey was distributed electronically (Qualtrics) to oncology/hematology physicians (MDs) and quality officers (QOs) at U.S academic (AC) and community (CO) cancer centers from August 2020 to November 2020. Each participant reported on their center’s experience with APMs. A descriptive analysis was conducted. Results: A total of 136 sites were contacted and 28 sites (13AC/15CO) participated (21% response rate). There were 30 (67%) sites that implemented an APM, the most common was the CMS Oncology Care Models (CMS OCM) (87%, n=26), which typically took 6 months or less for implementation (66%, n=20) and was reported to improve benchmark performance (63%, n=10). The most common reasons APMs were not implemented were administrative complexity (53%, n=8); difficulty with alignment of multiple payers to APM (47%, n=7); and need for a median of 5 full-time hires. A larger number of QOs indicated participation in CMS OCM compared to MDs [94% (n=15) vs. 76% (n=11)]. The biggest difficulty experienced by MDs was adoption of innovation at site’s financial risk (43%, n=6) compared to QOs who focused on administrative and financial burden (43%, n=7). Community cancer sites were more likely to use APMs (100%, n=23) compared to AC sites (31%, n=7). The largest difficulty experienced by CO sites was adoption of innovations at their own financial risk (26%, n=6) compared to high costs, administrative burden and lack of payer engagement (29%, n=2 each) for AC sites. The most common challenge with incentives among CO sites was concerns with risk adjustment of quality measures (57%, n=13) compared to increased complexity of incentive compensation (57%, n=4) for AC sites. CO sites indicated that quality measure performance reviews were primarily conducted by administrative staff compared to clinical leaders for AC sites. Conclusions: Administrative burden and access to appropriate patient data inhibit the impact of APMs. MDs are primarily concerned about the penalties and financial risk, whereas QOs focus on administrative complexities. For CO and AC sites, the adoption and implementation of APMs to improve cost and care delivery was not without significant challenges. This study demonstrates the multifaceted impact APMs have on care delivery.

Real-world outcomes of pembrolizumab plus carboplatin plus paclitaxel or nab-paclitaxel in non-small cell lung cancer (NSCLC)

e21717

Background: Pem+carbo+pac or nab-pac are approved first-line (1L) treatments for metastatic squamous NSCLC since 10/30/2018. Difference in real-world outcomes of these triplet combinations are unknown. Methods: Providers from community oncology practices in the USA reviewed the charts of consecutive NSCLC patients initiating 1L treatment with either pem+carbo+pac (“pac”) or pem+carbor+nab-pac (“nab-pac”) from 06/01/18-12/31/18; data were collected 11/22/2019-12/23/2019. Patient characteristics, treatment patterns, grade 3/4 toxicities, disease response, date of progression/death were retrospectively abstracted into an electronic case report form (eCRF). Univariate statistics were used to compare demographics, clinical characteristics and treatment patterns between cohorts. Overall survival (OS) from 1L was calculated using the Kaplan-Meier method. A Cox proportional hazards model was used to compare the risk of death between the two cohorts adjusted for multiple variables. Results: eCRFs were completed for 254 patients: nab-pac = 115, pac = 139. Median follow-up was 13.1 mos in both cohorts. No differences in demographics/clinical characteristics were noted (table). Median duration of therapy was 3.9 mos (nab-pac) and 3.8 mos (pac) (p = 0.58). Objective response rate: nab-pac = 69.6%, pac = 69.8% (p = 0.96). Receipt of maintenance therapy: nab-pac = 44.4%, pac = 48.9% (p = 0.47) . Median time to progression was 6.4 mos for nab-pac (n = 32) and 3.5 mos for pac (n = 29) (p = 0.10). 13.9% of nab-pac and 7.9% of pac patients had grade 3/4 toxicities (p = 0.16). At data cut-off, 18 nab-pac (15.7%) and 30 pac (21.6%) patients were deceased. Median OS from initiation of 1L was not reached. 12-month OS: nab-pac = 87.8% (95% CI: 81.8-93.8), pac = 79.3% (95% CI: 72.5-86.2). Adjusting for sex, age, race, PD-1 expression level, charlson comorbidity index (CCI) and ECOG-PS patients treated with nab-pac had a 40% reduction in mortality risk (HR = 0.55, 95% CI: 0.30-1.01, p = 0.05). Conclusions: No significant differences in demographics and clinical characteristics between patients treated with nab-pac or pac were observed. Median duration of treatment was equivalent with a marginally longer 12-month survival rate, and a lower adjusted mortality risk, among patients treated with nab-pac. [Table: see text]

A real-world comparative analysis of pomalidomide (POM) and other antimyeloma treatments following lenalidomide (LEN) discontinuation among patients with multiple myeloma

e19337

Background: In clinical trials, POM has demonstrated favorable clinical outcomes in patients with multiple myeloma (MM) who received prior LEN. Few studies, however, have examined POM treatment for MM in the community oncology setting. This retrospective cohort study compared treatment patterns and outcomes between patients who received a post-LEN treatment, either POM or another antimyeloma regimen. Methods: Adult patients with MM in the US Oncology Network (USON) who initiated a post-LEN treatment within 60 days of LEN discontinuation between Jan 1, 2016 and May 1, 2018, were not clinical trial participants, and had ≥ 2 subsequent clinic visits, were eligible. Data were sourced from USON’s iKnowMed electronic health records. Among patients observed to have discontinued treatment, time to treatment discontinuation (TTTD) was estimated from date of initiation of post-LEN treatment (index treatment) to date of discontinuation. Among patients who started a new treatment after the index treatment, time to next treatment (TTNT) was estimated from date of initiation of index treatment until date of initiation of the next treatment. TTTD and TTNT were analyzed using the Kaplan–Meier (KM) method across the whole study sample; patients who did not discontinue or start a next treatment were censored. Results: Of 547 eligible patients, 155 (28.3%) initiated POM and 392 (71.7%) initiated another antimyeloma regimen. Demographic characteristics were similar between the groups (for all patients, median age was 68 years, 54.5% patients were male and 71.7% were white). In total, 74.2% and 83.7% of patients discontinued the index treatment in the POM and other-treatment groups, respectively. Among the entire study population, KM estimates of median TTTD were 3.5 months (95% CI 2.8–4.6) and 1.9 months (95% CI 1.6–2.4) in the POM and other-treatment group, respectively (log-rank P < 0.001). In total, 65.2% and 71.2% of patients initiated subsequent treatment in the POM and other-treatment groups, respectively. KM estimates of median TTNT were 6.2 months (95% CI 4.5–7.8) and 4.5 months (95% CI 3.9–5.3) in the POM and other-treatment groups, respectively (log-rank P = 0.38). Conclusions: For patients with MM the use of POM following LEN treatment resulted in longer TTNT and TTTD compared with those who received other antimyeloma therapy. These findings support the use of POM treatment after LEN as an option for patients with relapsed/refractory MM.

Real-world disease burden for patients (pts) with myelofibrosis (MF) treated with ruxolitinib (RUX)

e19539

Background: MF is an aggressive myeloid neoplasm and until recently RUX was the only treatment available. Real-world, long-term outcomes in pts with MF have not been described. Methods: Providers from multiple community oncology clinics in the USA reviewed the charts of consecutive pts with MF (primary or post-polycythemia vera/essential thombocythemia MF) who received RUX at any time from Jan 1, 2015 to Dec 31, 2017; data were collected Nov 15–26, 2019. Pt characteristics, treatment patterns, and outcomes (symptoms, blood counts, disease transformation) were retrospectively abstracted into an electronic case report form (eCRF) at each visit during the first 6 months, at any RUX dose modification; and at 12, 24, and 36 months post-RUX initiation. Pt characteristics, treatment patterns, rationale for dose modification, and treatment duration are described for pts during the first 6 months of therapy. Results: eCRFs were completed for 104 pts with MF treated with RUX. 42 pts received a RUX dose modification and 62 did not. Median follow-up from RUX initiation was 23.5 months. In the first 6 months of therapy, 40.4% of pts had a RUX dose modification (76.2% dose reduced; 23.8% dose increased). The most common reason (70%) for dose increase was titration to therapeutic dose. The most common reason (75%) for dose reduction was hemoglobin 8–10 g/dL or platelet count 50–75 × 109. The starting dose of RUX was higher for no RUX dose modification pts (mean total daily dose 34.9 mg vs 27.0 mg, P < 0.01). RUX dose-modified pts were older vs pts with no RUX dose modification ( P = 0.01); International Prognostic Scoring System (IPSS) risk score ( P = 0.89) and the proportion of pts with primary MF ( P = 0.06) were similar (Table). Overall, 64.4% of RUX treated pts discontinued therapy (RUX dose-modified 78.6%; no RUX dose modification 54.8%, P = 0.01). Median duration of RUX treatment was 11.0 months (RUX dose modified 10.6 months; no RUX dose modification 12.2 months; log rank P = 0.12). Conclusions: In this real-world study, 64.4% of pts discontinued RUX during follow-up. Higher rates of discontinuation and shorter time to discontinuation were noted for pts with a dose modification during the first 6 months of therapy. Given the short duration of therapy observed in this study as compared with published clinical trials, additional MF treatments are needed that offer pt benefit. [Table: see text]

Real-world examination of remission patterns in patients (pts) with acute myeloid leukemia (AML)

e19336

Background: For pts with AML, prognosis is poor as long-term remission is elusive and ≥ 60% of pts relapse. The effect of remission status on healthcare resource utilization (HCRU) and costs is unclear. This study assessed administrative claims data from pts with AML to understand the potential benefit of novel, remission-prolonging therapies on HCRU and costs. Methods: Pts with newly diagnosed AML who received frontline therapy and did not undergo hematopoietic stem cell transplantation were identified in the MarketScan Commercial and Medicare Supplemental database from Jan 1, 2012–June 6, 2018. Pts were followed over a fixed 6-month (mos) pre- and variable post-diagnosis period; remission and relapse events were identified by diagnosis codes. Pts were analyzed by duration of remission ( < 3, 3– < 6, 6– < 9, and ≥ 9 mos), based on time from first remission claim to first relapse claim or end of follow up, whichever occurred first. Pt characteristics, relapse, HCRU, and costs were assessed. Results: 459/1003 eligible pts (45.8%) had evidence of remission. Most pts were in remission for < 3 mos (n = 161; 35.1%) or ≥ 9 mos (n = 165; 35.9%); 81 (17.6%) and 52 (11.3%) pts were in remission for 3– < 6 and 6– < 9 mos, respectively. Median follow-up for all pts was 236 days. Across remission cohorts, mean age at diagnosis was 55–59 years and median time to remission (from first AML diagnosis to first remission claim) was 82–90 days. Median time from AML diagnosis to relapse was 265.5 days. Of pts in remission, 30.5% relapsed (41.0%, 28.4%, 40.4%, and 18.2% for < 3, 3– < 6, 6– < 9, and ≥ 9 mos, respectively). From AML diagnosis to the end of follow-up, mean all-cause per pt per mos (PPPM) healthcare costs for all pts were $44,588. Longer durations of remission were associated with reduced mean PPPM costs ($64,188, $53,260, $30,219, and $16,654 for < 3, 3– < 6, 6– < 9, and ≥ 9 mos, respectively); similar trends were observed for AML-related costs. Mean all-cause PPPM costs were also reduced during remission, measured from the first remission claim to the first relapse claim or end of follow-up. A significant decrease was observed for pts in remission for ≥ 6 mos ($35,229, $36,193, $17,486, and $8,933 for < 3, 3– < 6, 6– < 9, and ≥ 9 mos, respectively). Conclusions: In this real-world study, 46% of pts with AML achieved remission after frontline therapy, although durations of remission varied. A longer duration of remission was associated with reduced PPPM costs over the study period indicating a potential economic benefit of remission-prolonging therapies in AML, including maintenance treatments.

Abstract P2-08-09: Treatment patterns and costs of metastatic triple negative breast cancer (mTNBC) in US women: A retrospective cohort study of first-line chemotherapy

Background: TNBC disproportionally affects younger women and is the leading cause of cancer-related deaths in women under 40. Treatment burden likely has a greater impact in younger women given employment and/or parental status. Little is known about treatment-related healthcare utilization, costs, or absenteeism or disability in women with mTNBC. This study describes real-world treatment patterns and the economic burden of mTNBC among patients treated with a first line (1L) chemotherapy.

Methods: A retrospective cohort analysis of IBM MarketScan(R) Commercial and Medicare Supplemental Claims Databases was conducted. The analysis included adult female patients diagnosed with mTNBC who initiated 1L therapy from 1/1/2011 - 9/30/2017. Patients were continuously enrolled with medical and pharmacy benefits &gt; 6 months prior to mTNBC diagnosis date and the start of 1L chemotherapy (index date) and for 3 mos after the index date. End of 1L was defined by treatment change, discontinuation (&gt;60-day gap), or inpatient death or hospice. Patient follow-up was from the index date to the end of continuous enrollment or the end of the study period, 12/31/2017. Baseline demographic, clinical characteristics and per patient per month (PPPM) healthcare utilization and costs were analyzed descriptively. Indirect productivity burden of mTNBC using employed patient’s absenteeism (ABS), short-term disability (STD), and long-term disability (LTD) are reported. All costs are 2017 US dollars.

Results: 1,027 patients with mTNBC meeting eligibility criteria were identified. Median age was 55 years; 69.5% of women were ages 45 - 64. Mean Charlson Comorbidity Index score was 8.8. Of the 27 1L treatments observed, the 3 most common were cyclophosphamide/doxorubicin (44.7%), cyclophosphamide/docetaxel (10.1%), and paclitaxel (7.7%). Median time to treatment post diagnosis was 27 days. Observed median follow-up time was 271 days; 13.2% of patients remained on 1L chemotherapy to end of follow-up. Mean duration of 1L treatment was 86 days (SD, 67 days); 69.2% of patients had a change in 1L treatment. For all-cause healthcare utilization during 1L, 11.6% patients had an inpatient admission, 16.9% had an ED visit, and 98.0% had an outpatient visit, of which 80.7%, 52.7%, and 98.8% were breast cancer-related, respectively. Mean all-cause total healthcare costs and outpatient prescription costs were $17,727 (SD $13,701) and $10,861 (SD $10,136) PPPM, respectively. Of breast cancer-related treatments, all patients received chemotherapy, 4.9% received radiation therapy, 1.2% underwent breast cancer surgery, and 96.5% received supportive care. The mean total breast cancer-related treatment cost was $10,322 (SD $9,512) PPPM.

At 6 months post-index, 56 patients (5.5%) had either ABS, STD, or LTD data eligibility. Of this subset, 4 (7.1%) patients had an ABS claim and missed an average of 245 work-hours and had a mean productivity loss of $6,472. STD and LTD claims were available for 14 (25.0%) and 4 (7.1%) patients who experienced an average of 63 and 76 work-days lost and a mean loss of $9,265 and $11,192, respectively.

Conclusion: Patients with mTNBC continue to experience significant treatment burden. Moreover, though many therapies were observed, limited long term treatment options are available suggesting a need for a durable treatment options that also minimize the burden for patients.

Citation Format: Ashley Tabah, Ronda Copher, David Huggar, Marc Tian, Sarah S. Mougalian. Treatment patterns and costs of metastatic triple negative breast cancer (mTNBC) in US women: A retrospective cohort study of first-line chemotherapy [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P2-08-09.

Real-World Treatment Patterns Among Patients Initiating Small Molecule Kinase Inhibitor Therapies for Thyroid Cancer in the United States

Introduction

Little is known about real-world use of small molecule kinase inhibitors (SMKI) for advanced thyroid cancer in the United States. This study examined prescribing patterns of SMKI agents recommended by the National Comprehensive Cancer Center (NCCN).

Methods

This retrospective study used a national health insurance database to identify patients diagnosed with thyroid cancer during 1/1/2006–6/30/2016 and with prescription claims for NCCN-recommended SMKI during 1/1/2010–5/31/2016 whose first claim date was the index date. Inclusion also required continuous enrollment in a health plan for 3 months pre-index (baseline) and ≥ 1 month post-index (follow-up) with no claims for SMKI during baseline. Lines of therapy (LOT) were defined by the date of SMKI claims and days of drug supply. Median time to SMKI discontinuation in each LOT was estimated by Kaplan–Meier method.

Results

The study included 217 patients. During follow-up (mean duration 499.0 days), 35.5% of patients (n = 77) received a second or later LOT; among patients with ≥ 12 months follow-up after first LOT (LOT1) initiation, 53.1% (n = 60) received a second or later LOT. Median treatment duration was 5.0 months for LOT1 and 5.1 months for LOT2. Over the entire follow-up period (2010–2016), sorafenib was the most common regimen in LOT1 (36.9% of patients) and LOT2 (24.7%) followed by sunitinib and levantinib (13.4% each) in LOT1 and sunitinib (19.5%) in LOT2. Starting in 2015, the year lenvatinib was approved for differentiated thyroid cancer, lenvatinib was the most common first-line regimen among patients initiating LOT1 in 2015 (43.4%) and 2016 (66.7%).

Conclusion

Sorafenib was the most common first-line agent during 2010–2014 but was supplanted by lenvatinib starting in 2015. Approximately 36–53% of patients received a second-line treatment. Median treatment duration results suggested potential benefit of SMKI in second-line therapy. SMKI treatment after first-line failure may be considered for appropriately selected patients.

Funding

Eisai, Inc. (Woodcliff Lake, NJ).

Abstract P6-17-28: Outcomes of real-world use of eribulin plus trastuzumab for HER2-positive metastatic breast cancer

Background

Eribulin mesylate is approved for the treatment of metastatic breast cancer (mBC) after two prior chemotherapy regimens including an anthracycline or a taxane in either the metastatic or adjuvant setting. Eribulin in combination with trastuzumab (E+T) has demonstrated tolerability and anti-tumor activity in phase I and II trials but is not FDA-approved for the treatment of HER2-positive mBC. Case series and retrospective research have noted the use of E+T in clinical practice. We sought to describe patient characteristics and long-term outcomes of treatment with E+T for HER2-positive mBC patients treated outside of clinical trials in the US.

Methods

US-based community oncologists from an open network of over 7,000 oncologists, hematologists, and urologists were invited to participate in identifying HER2-positive mBC patients treated with E+T between 01/01/11 and 12/31/13 outside of clinical trials. Data were collected from 03/18/2016 until 09/01/2016. Providers completed an electronic case report form (CRF) by abstracting data on disease characteristics, treatment patterns, disease response (per provider assessment), adverse events (Aes), and date of death. Duration of treatment and overall survival (OS) were calculated from the initiation of the E+T regimen. The target sample size was 60 patients and patients were selected according to resource available for chart data abstraction.

Results

Twenty-three providers submitted CRFs for 62 total patients. After data collection, 59 of 62 submitted records were validated for analysis. At mBC diagnosis, 69.4% of patients were ER/PR negative and 42.4% of patient had de novo stage IV disease. At initiation of E+T, the median age was 57 years and 81.4% were ECOG-OS 0/1. Mean length of follow-up from the initiation of any therapy was 33.6 months. Twenty-two (37.3%) patients initiated E+T as their first- or second-line of treatment; those remaining were in third-line or greater. At initiation of E+T, 72.8% of patients had prior treatment with trastuzumab in combination with chemotherapy, 25.4% had prior trastuzumab in combination with pertuzumab and chemotherapy, and 16.9% had received TDM-1. Mean duration of E+T treatment was 5.2 months (SD=2.4). A response (complete [CR] or partial [PR]) was recorded by the providers for 64.4% of patients (not independently verified). The most common Aes reported were fatigue (67.8%), weakness (50.8%), decreased appetite (28.8%), decreased hemoglobin (27.1%), peripheral neuropathy (25.4%), and neutropenia (18.6%). At the end of the study period, 34 patients (57.6%) were deceased; the median OS from the initiation of E+T was 23.9 months (95% CI: 17.8-30.4).

Conclusions

In a small cohort of patients treated with E+T in the community setting, the observed response rate of 64.4% (CR+PR) was comparable to that of a prior phase II trial of E+T which reported an ORR with first-line E+T of 71.2% overall, 77.4% among T-naïve and 61.9% in T-pretreated patients. Further research is warranted to examine the tolerability and efficacy of E+T for metastatic HER2-positive breast cancer patients in different treatment settings.

Citation Format: Mougalian SS, Copher R, McAllister L, Radtchenko J, Wang EC, Broscious M, Yu H-T, Kish J. Outcomes of real-world use of eribulin plus trastuzumab for HER2-positive metastatic breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-17-28.

Cost-effectiveness analysis of lenvatinib treatment for patients with unresectable hepatocellular carcinoma (uHCC) compared with sorafenib in Japan

BACKGROUND

Lenvatinib demonstrated a treatment effect on overall survival by the statistical confirmation of non-inferiority to sorafenib for the first-line treatment of uHCC. The objective of this study was to evaluate the cost-effectiveness of lenvatinib compared with sorafenib for patients with uHCC in Japan.

METHODS

A partitioned-survival model was developed to estimate the cost-effectiveness of lenvatinib versus sorafenib when treating uHCC patients over a lifetime horizon and considering total public healthcare expenditure. Efficacy and safety data were extracted from the REFLECT trial. Utility values were derived from the European Quality-of-Life 5-Dimension Questionnaire, conducted with patients enrolled in the REFLECT trial. Direct medical costs, such as primary drug therapy, outpatient visits, diagnostic tests, hospitalization, post-progression therapy, and adverse-event treatments, were included. Cost parameters unavailable in the clinical trial or publications were obtained based on the consolidated clinical standards from a Delphi panel of four Japanese medical experts.

RESULTS

For lenvatinib versus sorafenib, the incremental cost was - 406,307 Japanese Yen (JPY), and the incremental life years and quality-adjusted life years (QALYs) were 0.27 and 0.23, respectively. Thus, lenvatinib dominated sorafenib, due to the mean incremental cost-effectiveness ratio falling in the fourth quadrant, conferring more benefit at lower costs compared with sorafenib. The probabilistic sensitivity analysis showed that 81.3% of the simulations were favorable to lenvatinib compared with sorafenib, with a payer's willingness-to-pay-per-QALY of 5 million JPY.

CONCLUSIONS

Lenvatinib was cost-effective compared with sorafenib for the first-line treatment of uHCC in Japan.

Clinical benefit of treatment with eribulin mesylate for metastatic triple-negative breast cancer: Long-term outcomes of patients treated in the US community oncology setting

Introduction

Real‐world data are critical to demonstrate the consistency of evidence and external generalizability of randomized controlled trials (RCTs). This study examined characteristics and outcomes of metastatic triple‐negative breast cancer (mTNBC) patients treated with eribulin mesylate at community oncology practices in the United States.

Methods

Physicians identified mTNBC patients initiating eribulin between 1 January 2011 and 1 January 2014 and abstracted data into an electronic case report form (eCRF). Eribulin treatment in the metastatic setting was categorized as early use (EU, first‐/second‐line) and late use (LU, third‐line or later). Patient characteristics, overall survival (OS), disease response (per treating physician), and adverse events (AEs) rates in each group, respectively, are reported.

Results

Overall 252 eCRFs were completed: 125 (49.6%) EU and 127 (50.4%) LU. The median age at metastatic diagnosis was 53 years and 42.1% were stage IV at their initial diagnosis. The median duration of follow‐up from the initiation of first‐line treatment was 24 months. Rates of disease response (complete or partial per treating physician) were 69.9% in the EU group and 48.8% in the LU group. The five most commonly reported adverse events during eribulin were as follows: fatigue (65.1%), weakness (40.1%), decreased appetite (32.5%), neutropenia (31.0%), and leukopenia (27.4%). Discontinuation of eribulin due to AE was observed in 4.0% of patients. Median OS from initiation of eribulin was 23.0 months (95% CI: 18.7‐27.3) among EU and 14.7 (95% CI: 12.6‐16.9) among LU.

Conclusion

In the real‐world eribulin‐treated mTNBC, patients have more sites of metastatic disease and exposure to greater numbers of prior therapies compared to RCTs. The median OS of 14.7 months among LU patients is consistent with, and slightly longer than the 13.1 months and 14.4 months reported in the EMBRACE and Study 301 clinical trials, respectively.

Number needed to treat in indirect treatment comparison

AIM

For dichotomous outcomes, odds ratio (OR) is one of the usual summary measures of indirect treatment comparison. A corresponding number needed to treat (NNT) estimate may facilitate understanding of the treatment effect.

METHODS

We show how to estimate NNT based on OR results of a matching adjusted indirect comparison. We also have derived the explicit formula of its 95% CIs by applying the delta method, and as an alternative, a simulation-based method.

RESULTS

The method was applied in a case study example in radioiodine-refractory differentiated thyroid cancer (RR-DTC) patients, comparing lenvatinib to sorafenib. For every two RR-DTC patients treated with lenvatinib instead of sorafenib, one fewer would have progressed and for every eight RR-DTC patients treated with lenvatinib instead of sorafenib, one fewer would have died.

CONCLUSION

Using NNT to summarize the results of a matching adjusted indirect comparison can help the clinicians to better understand the results in addition to OR.

Measuring the Value of New Drugs: Validity and Reliability of 4 Value Assessment Frameworks in the Oncology Setting

BACKGROUND

Several organizations have developed frameworks to systematically assess the value of new drugs.

OBJECTIVE

To evaluate the convergent validity and interrater reliability of 4 value frameworks to understand the extent to which these tools can facilitate value-based treatment decisions in oncology.

METHODS

Eight panelists used the American Society of Clinical Oncology (ASCO), European Society for Medical Oncology (ESMO), Institute for Clinical and Economic Review (ICER), and National Comprehensive Cancer Network (NCCN) frameworks to conduct value assessments of 15 drugs for advanced lung and breast cancers and castration-refractory prostate cancer. Panelists received instructions and published clinical data required to complete the assessments, assigning each drug a numeric or letter score. Kendall's Coefficient of Concordance for Ranks (Kendall's W) was used to measure convergent validity by cancer type among the 4 frameworks. Intraclass correlation coefficients (ICCs) were used to measure interrater reliability for each framework across cancers. Panelists were surveyed on their experiences.

RESULTS

Kendall's W across all 4 frameworks for breast, lung, and prostate cancer drugs was 0.560 (P= 0.010), 0.562 (P = 0.010), and 0.920 (P < 0.001), respectively. Pairwise, Kendall's W for breast cancer drugs was highest for ESMO-ICER and ICER-NCCN (W = 0.950, P = 0.019 for both pairs) and lowest for ASCO-NCCN (W = 0.300, P = 0.748). For lung cancer drugs, W was highest pairwise for ESMO-ICER (W = 0.974, P = 0.007) and lowest for ASCO-NCCN (W = 0.218, P = 0.839); for prostate cancer drugs, pairwise W was highest for ICER-NCCN (W = 1.000, P < 0.001) and lowest for ESMO-ICER and ESMO-NCCN (W = 0.900, P = 0.052 for both pairs). When ranking drugs on distinct framework subdomains, Kendall's W among breast cancer drugs was highest for certainty (ICER, NCCN: W = 0.908, P = 0.046) and lowest for clinical benefit (ASCO, ESMO, NCCN: W = 0.345, P = 0.436). Among lung cancer drugs, W was highest for toxicity (ASCO, ESMO, NCCN: W = 0. 944, P < 0.001) and lowest for certainty (ICER, NCCN: W = 0.230, P = 0.827); and among prostate cancer drugs, it was highest for quality of life (ASCO, ESMO: W = 0.986, P = 0.003) and lowest for toxicity (ASCO, ESMO, NCCN: W = 0.200, P = 0.711). ICC (95% CI) for ASCO, ESMO, ICER, and NCCN were 0.800 (0.660-0.913), 0.818 (0.686-0.921), 0.652 (0.466-0.834), and 0.153 (0.045-0.371), respectively. When scores were rescaled to 0-100, NCCN provided the narrowest band of scores. When asked about their experiences using the ASCO, ESMO, ICER, and NCCN frameworks, panelists generally agreed that the frameworks were logically organized and reasonably easy to use, with NCCN rated somewhat easier.

CONCLUSIONS

Convergent validity among the ASCO, ESMO, ICER, and NCCN frameworks was fair to excellent, increasing with clinical benefit subdomain concordance and simplicity of drug trial data. Interrater reliability, highest for ASCO and ESMO, improved with clarity of instructions and specificity of score definitions. Continued use, analyses, and refinements of these frameworks will bring us closer to the ultimate goal of using value-based treatment decisions to improve patient care and outcomes.

DISCLOSURES

This work was funded by Eisai Inc. Copher and Knoth are employees of Eisai Inc. Bentley, Lee, Zambrano, and Broder are employees of Partnership for Health Analytic Research, a health services research company paid by Eisai Inc. to conduct this research. For this study, Cohen, Huynh, and Neville report fees from Partnership for Health Analytic Research. Outside of this study, Cohen receives grants and direct consulting fees from various companies that manufacture and market pharmaceuticals. Mei reports a grant from Eisai Inc. during this study. The other authors have no disclosures to report. Study concept and design were contributed by Bentley and Broder, with assistance from Elkin and Cohen. Bentley took the lead in data collection, along with Elkin, Huynh, Mukherjea, Neville, Mei, Popescu, Lee, and Zambrano. Data interpretation was performed by Bentley and Broder, along with Elkin, Cohen, Copher, and Knoth. The manuscript was written primarily by Bentley, along with Elkin and Broder, and revised by Bentley, Broder, Elkin, Cohen, Copher, and Knoth. Select components of this work's methods were presented at ISPOR 19th Annual European Congress held in Vienna, Austria, October 29-November 2, 2016, and Society for Medical Decision Making 38th Annual North American Meeting held in Vancouver, Canada, October 23-26, 2016.

Validity and reliability of four value frameworks for cancer drugs

6603

Background: Little is known about the validity and reliability of value assessment frameworks. Methods: Eight panelists used the ASCO, ESMO, ICER, and NCCN frameworks to conduct value assessments of 15 drugs for advanced lung and breast cancers and castration refractory prostate cancer. Panelists received instructions and published clinical data to complete the assessments, assigning each drug a numeric or letter score. We used Kendall’s W coefficient to measure convergent validity by cancer type among frameworks and intraclass correlation coefficients (ICC) to measure framework inter-rater reliability across cancers. Panelists were surveyed on their experiences. Results: Kendall’s W for breast, lung, and prostate cancer drugs were 0.560 ( p= 0.010), 0.562 ( p= 0.010), and 0.920 ( p< 0.001), respectively. Pairwise and subdomain W are shown in the table. ICC (95% CI) for ASCO, ESMO, ICER, and NCCN were 0.800 (0.660-0.913), 0.818 (0.686-0.921), 0.652 (0.466-0.834), and 0.153 (0.045-0.371), respectively. Panelists generally agreed the frameworks were logically organized and easy to use. Conclusions: Convergent validity among the frameworks was fair to excellent, increasing with clinical benefit subdomain concordance and simplicity of drug trial data. Inter-rater reliability, highest for ASCO and ESMO, improved with clarity of instructions and specificity of score definitions. Continued use, analyses, and refinements of the frameworks will bring us closer to using value-based treatment decisions to improve patient care and outcomes. [Table: see text]

Chemotherapy induced nausea and vomiting in breast cancer treated with antiemetic prophylaxis as recommended by the ASCO antiemesis guidelines

10109

Background: Current ASCO Antiemesis Guidelines recommend triple antiemetic therapy (a 5HT3RA, an NK1, and dexamethasone) to prevent chemotherapy (CT) induced nausea and vomiting (CINV) in patients undergoing highly emetogenic chemotherapy (HEC). This study evaluated whether this regimen resulted in reduced rates of CINV in patients diagnosed with breast cancer (BC) and initiated on HEC. The primary outcomes of interest were rates of acute and delayed CINV in patients whose antiemesis prophylaxis was or was not in accordance with the ASCO guideline (i.e., Per-guideline vs. Non-Guideline). Costs of treating CINV were also calculated. Methods: Patients were identified in the Premier Healthcare database, a complete geographically diverse census of inpatients and hospital-based outpatients. Adults treated for BC with HEC during the years 2012-14 were identified and stratified based on their antiemesis prophylaxis. Rates of acute (day of CT) and delayed CINV (days 2-7 post CT) were calculated following initiation of HEC. CINV was defined by ICD9 codes for nausea and vomiting or volume depletion/dehydration or use of a rescue antiemetic. Rates of CINV and health care costs were then compared between the two cohorts. Results: A total of 8,388 patients were included in the analysis. Of these, 5,447 (65%) had treatment Per-Guideline and 2,941 (35%) were Non-Guideline. For acute CINV, Per-Guideline patients had a significantly lower rate of CINV when compared to Non-Guideline patients (1.7% vs. 3.2%, respectively, p < .001). Similarly, in delayed CINV Per-Guideline patients had significantly lower rates of CINV when compared to Non-Guideline patients (15.4% vs. 19.1%, p < .001). Patients who experienced CINV also had significantly greater total health care costs versus those without CINV ($32,199 vs. $20,163, respectively, p < .001). Conclusions: The results showed adherence to the ASCO Antiemesis Guidelines led to lower rates of CINV and lower costs. Although defining CINV by claims may tell an incomplete story, this study suggests that following the ASCO Antiemesis Guidelines may help both patients and payers of health care costs.

Second or none: Many patients treated for refractory differentiated thyroid cancer with small molecular kinase inhibitors do not receive a second line of therapy

e17589

Background: Describe the treatment patterns of patients initiated on NCCN-recommended small molecular kinase inhibitors (SMKIs) for radioiodine-refractory differentiated thyroid cancer (DTC) approved in the United States. Methods: A large national US claims database was used to identify adult patients diagnosed with thyroid cancer (≥2 non-DX medical claims, ≥ 30 days apart) from 1/1/2006 - 6/30/2016 (study period) with claims for SMKIs from 1/1/2010 - 5/31/2016. Continuous enrollment required participation in a commercial or Medicare Advantage health plan ≥3 months before and ≥1 month following index date (date of first pharmacy claim for SMKI). Line of therapy (LOT) periods were defined by receipt and timing of SMKIs. Patient follow up was earliest disenrollment, death or end of the study period. Patient characteristics and SMKI treatment patterns were described. Results: A total of 217 DTC patients were identified; 63% commercially insured and 37% Medicare Advantage. Almost half were male (48%); mean age was 61.2 years (standard deviation SD 12.5 years) and mean follow-up period was 499 days (SD 414 days). In the study period, 35% (n = 77) of patients had ≥2 LOTs and 18% (n = 39) had ≥3 LOTs. Mean treatment duration was 5.4 months (SD 6.7 mos) for LOT1, 4.9 months (SD 3.8 mos) for LOT2, and 4.2 months (SD 4.9 mo) for LOT3. During the full study period, the most used regimens were Sorafenib for both LOT1 (37%) and LOT2 (25%), pazopanib (18%) and sunitinib (18%) in LOT3. Also, in the study period, 33 patients had sorafenib in LOT1 of which 16 were treated with sorafenib again (48%) in LOT2. Post FDA approval in 2015, Lenvatinib became the predominant first-line regimen (47%, n = 29) during study period. Across all first line therapies, for those patients with ≥12 months of follow-up, 53% (n = 60) initiated LOT2. Conclusions: Sorafenib was the most common first line of therapy for DTC, with Lenvatinib adoption increasing as first-line therapy since the drug’s approval in 2015. Depending on the period evaluated, almost half to 2/3 of patients are not receiving a second line of treatment, efficacious and patient appropriate therapy is of importance in treating this rare cancer.

Abstract P5-15-16: Utilization and outcomes of eribulin in triple negative metastatic breast cancer: Real-world findings

Background

Triple-negative breast cancer (TNBC) accounts for 10-20% of all breast cancers (BCs) and a significant proportion of all BC deaths. Eribulin is approved for the treatment of metastatic BC (MBC) after treatment with two prior regimens. A pooled analysis of two phase III studies of eribulin in women with TNBC patients found a 26% reduction in the risk of death vs. controls. Treatment patterns of eribulin and clinical outcomes associated with early vs. late use among TNBC patients treated in community oncology practices have not been evaluated.

Methods

Physicians from the Cardinal Health Oncology Research Network completed an electronic case report form (CRF) on up to 7 TNBC patients treated with eribulin between 01/01/11 and 12/31/13. Adult female patients with pathologically confirmed metastatic disease and not participating in any interventional clinical trial were included. Providers indicated the usage of chemotherapy, either alone or in combination, by line of therapy (LOT) up to the LOT of eribulin initiation. Reported data points include: clinical parameters (eg, site of metastases, ECOG performance status, and comorbidities), treatment events (eg, LOT start/end date and rationale for discontinuation), and outcomes (eg, clinical response and date of death). Dosing, adverse events, use of supportive care medications, and hospitalization were also captured during eribulin treatment. Use of eribulin in LOT 1/LOT 2 was considered early; LOT 3+ was considered late. All comparisons are univariate.

Results

An interim analysis was performed on 123 TNBC patients (planned sample size of 250) collected from 26 providers. Patient mean age at eribulin treatment initiation was 55.0 years. Mean follow-up duration was 27 mo (SD = 11.9) from initiation of first line metastatic treatment until date of last visit, death, or loss to follow-up. Overall, 74.0% were deceased, 85.4% had received at least 3 LOTs in the metastatic setting, and 45.4% were stage IV at diagnosis. Most women were prescribed eribulin in a later LOT (61.8%), 3 (2.4%) patients received eribulin in LOT1 and 44 in LOT2 (36.7%). Among patients with known treatment start and end dates (87.0%), mean duration of treatment (DOT) was 6.2 mo (SD = 3.3), median 5.8 mo among early recipients and 5.5 mo (SD = 5.7), median 4.1 mo, among later recipients (p = 0.39). Early users were more likely (p = 0.05) to have a complete/partial response (71.1% vs. 47.7%) and less likely to have progressive disease (7.1% vs. 12.3%). In comparing eribulin users to all other therapies, eribulin users had a significantly longer DOT in LOT2 (5.9 vs. 4.7 mo, p = 0.01) and LOT3 (5.8 vs. 3.6 mo, p = 0.03). In LOT3, eribulin users were significantly more likely to have complete/partial response (54.2% vs. 18.8%) and less likely to have to have progressive disease (4.2% vs. 37.5%) compared to all other observed LOT3 therapies.

Conclusions

This interim analysis indicates longer DOT for patients treated with eribulin for TNBC in LOT2 and LOT3 and a more favorable response rate compared to all other agents used in each LOT, respectively, among patients treated in community oncology practices. Full results will be available at the conference.

Citation Format: Kish JK, Mougalian SS, Copher R, McAllister L, Zhixiao W, Broscious M. Utilization and outcomes of eribulin in triple negative metastatic breast cancer: Real-world findings [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P5-15-16.

Validity and Reliability of Value Assessment Frameworks for New Cancer Drugs

BACKGROUND

Several organizations have developed frameworks to systematically assess the value of new drugs. These organizations include the American Society of Clinical Oncology (ASCO), the European Society for Medical Oncology (ESMO), the Institute for Clinical and Economic Review (ICER), and the National Comprehensive Cancer Network (NCCN).

OBJECTIVES

To understand the extent to which these four tools can facilitate value-based treatment decisions in oncology.

METHODS

In this pilot study, eight panelists conducted value assessments of five advanced lung cancer drugs using the ASCO, ESMO, and ICER frameworks. The panelists received instructions and published clinical data required to complete the assessments. Published NCCN framework scores were abstracted. The Kendall's W coefficient was used to measure convergent validity among the four frameworks. Intraclass correlation coefficients were used to measure inter-rater reliability among the ASCO, ESMO, and ICER frameworks. Sensitivity analyses were conducted.

RESULTS

Drugs were ranked similarly by the four frameworks, with Kendall's W of 0.703 (P = 0.006) across all the four frameworks. Pairwise, Kendall's W was the highest for ESMO-ICER (W = 0.974; P = 0.007) and ASCO-NCCN (W = 0.944; P = 0.022) and the lowest for ICER-NCCN (W = 0.647; P = 0.315) and ESMO-NCCN (W = 0.611; P = 0.360). Intraclass correlation coefficients (confidence interval [CI]) for the ASCO, ESMO, and ICER frameworks were 0.786 (95% CI 0.517-0.970), 0.804 (95% CI 0.545-0.973), and 0.281 (95% CI 0.055-0.799), respectively. When scores were rescaled to 0 to 100, the ICER framework provided the narrowest band of scores.

CONCLUSIONS

The ASCO, ESMO, ICER, and NCCN frameworks demonstrated convergent validity, despite differences in conceptual approaches used. The ASCO inter-rater reliability was high, although potentially at the cost of user burden. The ICER inter-rater reliability was poor, possibly because of its failure to distinguish differential value among the sample of drugs tested. Refinements of all frameworks should continue on the basis of further testing and stakeholder feedback.

Health-care costs and utilization related to long- or short-acting antiepileptic monotherapy use

PURPOSE

This study aimed to compare health-care utilization and costs in patients treated with long-acting (LA) vs. short-acting (SA) antiepileptic drug (AED) monotherapy.

METHODS

We conducted a cross-sectional study of claims from the OptumInsight™ database. Our analysis was restricted to adults diagnosed with epilepsy and who used AED monotherapy. Patients were excluded if they used >1 type of AED, had <9months of treatment, or had a treatment gap of >60days. Antiepileptic drugs were classified as LA or SA based on published data and expert opinion. Medical and pharmacy claims were used to estimate health-care utilization and costs, and baseline group differences were adjusted using multivariate analyses.

RESULTS

There were 4058 (49.6%) LA AED users and 4122 (50.4%) SA AED users. Medication possession ratios (MPRs) were not significantly different between LA AED users and SA AED users (P=0.125). Long-acting AED users had lower mean overall health-care costs ($9757 vs. $12,689), lower epilepsy-related costs ($3539 vs. $5279), and lower rate of overall (8.8% vs. 10.9%) and epilepsy-related hospitalizations (5.7% vs. 7.6%) compared with SA AED users (all P<0.01). After adjusting for demographics and clinical characteristics, mean overall costs were lower by $686 and the mean epilepsy-related costs were lower by $894 in LA AED users.

CONCLUSION

Although MPRs were similar in LA AED and SA AED groups, patients treated with LA monotherapy had a lower economic burden compared with those treated with SA monotherapy, indicating that using AEDs with extended duration of action is associated with decreased health-care use and lower health-care costs.

Effectiveness of antiepileptic drug combination therapy for partial-onset seizures based on mechanisms of action

IMPORTANCE

To our knowledge, the current study is the first to describe antiepileptic drug (AED) combination therapy patterns according to their mechanism of action (MOA) in a real-world setting and to evaluate the differences in outcomes comparing different-MOA combination therapy with same-MOA combination therapy for patients with partial-onset seizure.

OBJECTIVE

To compare treatment persistence and health care use with AED combinations categorized by MOA in patients with partial-onset seizures.

DESIGN, SETTING, AND PARTICIPANTS

Using the Truven Health MarketScan Commercial Claims Database containing 96 million covered lives from July 1, 2004, through March 31, 2011, adults with concomitant use of 2 different AEDs and a recent partial-onset seizure diagnosis were selected. Antiepileptic drugs were categorized by MOA: sodium channel blockers (SC), gamma-aminobutyric acid analogs (G), synaptic vesicle protein 2A binding (SV2), and multiple mechanisms (M). Patients were assigned a combination category based on their concomitant AED use.

MAIN OUTCOMES AND MEASURES

Treatment persistence was measured from the start of AED combination therapy until the end of the combination. Health care resource use was measured during the combination treatment duration. Multivariate analyses evaluated AED discontinuation risk and health care use according to MOA combinations.

RESULTS

Distribution of 8615 selected patients by combination was 3.3% for G+G, 7.5% for G+SV2, 8.6% for G+M, 13.9% for SC+SC, 19.0% for G+SC, 21.5% for SC+M, and 26.3% for SC+SV2. The same-MOA (G+G and SC+SC) combinations had the shortest persistence (mean [SD], 344 [345] days and 513 [530] days, respectively) and greater hazard of discontinuation compared with different-MOA combinations. Patients with different-MOA G combinations had a significantly lower risk for inpatient admission (odds ratio, 0.716; 95% CI, 0.539-0.952; P = .02) compared with G+G combinations. Patients with different-MOA SC combinations had significantly lower risks for emergency department visits (odds ratio, 0.853; 95% CI, 0.742-0.980; P = .03) compared with SC+SC combinations.

CONCLUSIONS AND RELEVANCE

The findings suggest that AED combinations with different MOAs have greater effectiveness as measured by treatment persistence and lower risks for hospitalization and emergency department visits. Further research is needed to more fully understand the role of the MOA in achieving optimal outcomes.

Patient preferences and treatment adherence among women diagnosed with metastatic breast cancer

BACKGROUND

Given the various profiles (eg, oral vs intravenous administration, risk of hot flashes vs fatigue) of treatment options (eg, endocrine therapy, chemotherapy) for metastatic breast cancer (mBC), how patients value these attributes of their medications has implications on making treatment decisions and on adherence.

OBJECTIVES

To understand how patients trade off medication side effects with improved effectiveness and/or quality of life, to provide estimates of nonadherence among women with mBC, and to quantify the association of medication nonadherence with health outcomes.

METHODS

The study was a cross-sectional, Internet-based survey of 181 women diagnosed with mBC who were recruited from cancer-specific online panels (response rate, 7%). Treatment information, demographics, nonadherent behaviors, and quality of life assessed by the Functional Assessment of Cancer Therapy-Breast (FACT-B) were collected in the survey, and each respondent completed a choice-based conjoint exercise to assess patient preferences. The patients' preferences were analyzed using hierarchical Bayesian logistic regression models, and the association between the number of nonadherent behaviors and the health outcomes was analyzed using general linear models.

RESULTS

The mean age of the patient sample was 52.2 years (standard deviation, ±9.1), with 93.9% of participants being non-Hispanic white. Results from the conjoint model indicated that effectiveness (overall survival) was of primary importance to patients, followed by side effects-notably alopecia, fatigue, neutropenia, motor neuropathy, and nausea/vomiting-and finally, dosing regimen. In all, 34.8% of survey respondents either discontinued their treatment or were nonadherent to their treatment regimen. Among those who have ever used oral chemotherapy (N = 95; 52.5%) and those currently using oral chemotherapy (N = 44; 24.3%), the number of nonadherent behaviors was significantly associated with a decrease in functional well-being (b [unstandardized regression coefficient] = -2.01 for patients who had ever used a targeted therapy and b = -3.14 for current users of a targeted therapy), FACT-General total score (b = -4.30 and b = -7.37, respectively), FACT-B total score (b = -3.93 and b = -6.11, respectively), and FACT trial outcome index (b = -5.22 and b = -8.63, respectively; all P <.05).

CONCLUSIONS

Patients were willing to accept substantial additional risks from side effects for gains in overall survival. Approximately 33% of women with mBC reported engaging in nonadherent behaviors. Because forgetfulness and adverse events were among the most frequent reasons for nonadherence, these results suggest that less complex treatment regimens, as well as regimens with less toxic profiles, may be associated with improvements in adherence and, subsequently, could correspond to perceptible patient benefits.

Retrospective analysis of variation in heavy menstrual bleeding treatments by age and underlying cause

OBJECTIVE

To describe treatment patterns associated with heavy menstrual bleeding (HMB) in US practice.

STUDY DESIGN

A retrospective claims-based analysis of organic (ICD-9 codes 218.x, 621.0, 622.7, 219.x, and bleeding disorders) or idiopathic (no underlying condition identified) HMB treatment patterns among newly diagnosed, commercially insured women who were enrolled in a large US health plan. First HMB claim (index date; ICD-9-CM 626.2 and 627.0), second HMB claim within 180 days of index date, and continuous enrollment ≥6 months prior to (pre-index period) and 18 months following (post-index period) index date were required.

RESULTS

The database included 13,579 organic and 21,362 idiopathic HMB patients. More organic HMB patients received only one treatment type (64% vs 58%; p < 0.001) or two treatments types (14% vs 11%; p < 0.001) compared to idiopathic HMB patients. During the 18 month post-index period, fewer organic HMB patients had no observed treatment compared to idiopathic HMB patients (21% vs 31%; p < 0.001). The idiopathic cohort had significantly higher rates (p < 0.001) of medication use and endometrial ablation, whereas the organic HMB cohort had a higher rate of hysterectomy (p < 0.001). Women <35 years were more frequently prescribed medical treatments (p ≤ 0.037), while women aged >35 years utilized significantly more surgical approaches (p < 0.001).

CONCLUSIONS

Among organic and idiopathic HMB patients, considerable variation was observed in the medications and procedures used to treat HMB. Current treatment pattern awareness may improve HMB management. Future research is needed to understand factors that influence women's treatment choices (including newer medications LNG-IUS and tranexamic acid) and age in relation to child-bearing preference.

Treatment patterns and healthcare system burden of managed care patients with suspected pulmonary arterial hypertension in the United States

OBJECTIVES

To describe treatment patterns and healthcare burden among individuals with suspected pulmonary arterial hypertension (PAH), as identified through a practice guideline-based healthcare claims algorithm.

METHODS

Adults with evidence of PAH from 1 January 2004 (commercial and Medicaid) or 1 July 2006 (Medicare Advantage) through 30 June 2008 were identified. Given the lack of an ICD-9 code for PAH, an algorithm was developed requiring: (1) ≥ 1 claim for PAH medication (index date); (2) ≥ 1 claim with a pulmonary hypertension diagnosis code in the 6-month pre-index period (baseline) or within 90 days post-index; (3) a right heart catheterization or pulmonary hypertension-related inpatient stay during baseline or within 90 days post-index; and (4) continuous health plan enrollment for 6 months pre-index and ≥ 6 months post-index. Patients with PAH-specific medications during baseline were excluded. Treatment patterns, healthcare utilization, and costs were assessed during the period ending with the earlier of health plan disenrollment or 31 December 2008.

RESULTS

Among the 521 included patients, 69% were female. Most patients (94%) initiated treatment with monotherapy (most commonly sildenafil or bosentan), and 12.7% of all patients augmented their therapy by the end of the observation period. The medication possession ratio was 0.96 each for ambrisentan (SD=0.04), bosentan (SD=0.04), and sildenafil (SD=0.05). Overall, 72.6% of patients discontinued therapy with a mean of 149 (SD=170) days until discontinuation. A mean (SD) of 2.14 (1.82) all-cause office and 1.64 (1.98) outpatient visits occurred per patient per month. Mean PAH-related healthcare costs were $6617 per patient per month, comprising 71% of all-cause costs. The guideline-based algorithm may not have perfectly captured patients with PAH.

CONCLUSIONS

Patients with suspected PAH were likely to initiate treatment with oral monotherapy, had high compliance rates, and received close ambulatory follow-up. PAH-related costs constituted the majority of all-cause healthcare costs.

Physician perception of patient adherence compared to patient adherence of osteoporosis medications from pharmacy claims

OBJECTIVE

This study explored physicians' perceptions of patient adherence to medications compared with patient adherence derived by administrative data in the treatment of osteoporosis.

RESEARCH DESIGN AND METHODS

A study involving written questionnaires from prescribers treating patients with postmenopausal osteoporosis (PMO) compared the questionnaire responses to pharmacy claims of these prescribers' patients' refill patterns. Approximately 2000 physicians from a large US health plan were faxed or mailed a survey. Data from the physician survey were merged with administrative claims data of the participating physicians' patients.

RESULTS

A total of 412 physicians (21.8%) responded. Although a low response rate, there were no significant demographic differences between participating and non-participating physicians. Surveyed physicians reported that 66% of their patients had private/commercial coverage and over 60% reported seeing their PMO patients annually. Overall, physicians estimated that 69.2% of patients were adherent 80% of the time after 12 months of therapy. Yet, pharmacy claims data for those physicians' patients indicated 48.7% of patients were adherent (defined as having an MPR of >or=80%) after 12 months of therapy. Physicians overestimated their patients' adherence regardless of medication class and across physician specialties. Regression modeling revealed that physicians who have been in practice longer estimated fewer patients as adherent, whereas those who prescribe more PMO treatments estimate a greater number of patients as adherent. Providers cited side effects and affordability of medication as the most frequent reasons for non-adherence.

CONCLUSIONS

Physicians overestimate patient adherence to PMO therapies. Improving physician awareness of medication non-adherence to PMO therapies may facilitate physician-patient dialogue, with the aim of identifying patient-centered reasons for non-adherence. These discussions are important because patients with poorer adherence have a higher risk of fracture. Future research should focus on reasons for patient non-adherence to osteoporosis regimens and intervention strategies that improve communication between the provider and patient. Findings must be considered within the limitations of this claims database analysis. Some degree of incomplete or incorrect coding may exist, and the presence of a claim for a filled prescription does not indicate that the medication was consumed or taken as prescribed. Patients included in the study are not necessarily representative of all patients being treated for osteoporosis.