Search

e20614 Background: Globally, small cell lung cancer (SCLC) accounts for approximately 10% of lung cancers and is typically diagnosed at an advanced/extensive stage in elderly current or former smokers. The objective of this study was to describe patient characteristics (demographic and clinical) and outcomes (survival) associated with SCLC in the Medicare population by systemic anticancer treatment status (treated vs untreated). Methods: A retrospective analysis was conducted using the Surveillance, Epidemiology, and End Results (SEER)–Medicare linked database. Patients were included if they had a record with a diagnosis of SCLC in the database from 1/1/2007 – 12/31/2017, were ≥65 years of age at diagnosis, and had continuous enrollment for ≥180 days pre-SCLC diagnosis. Patient characteristics and overall survival (OS) were stratified by treatment status. Median OS was estimated using the Kaplan Meier method. Results: 13,516 patients met the study criteria. Overall, 57.7% were female, mean age was 74 years, and 89.5% were white. Nearly three-fourths (73.1%) were diagnosed with stage 4 disease. Just over half (55.4%, N = 6,022) initiated 1L treatment. Older patients were less likely to be treated: 36.3%, 51.1%, and 78.5% of patients 65-74 yrs, 75-84 yrs, and 85+ yrs, respectively, were untreated. Patients diagnosed at stage I-III were more likely to be treated than those diagnosed with stage IV disease (68.2% vs 51.4% treated). Patients with metastasis in the brain and/or liver were also less likely to be treated than patients without metastasis in those sites. On average untreated patients had more comorbid conditions than treated patients (average Charlson Comorbidity Index [CCI] 9.2 vs 9.5). In particular, congestive heart failure (49.3% vs 32.6%), myocardial infarction (25.7% vs 18.9%), renal disease (31.5% vs 18.9%), dementia (13.3% vs 3.5%), hemiplegia/paraplegia (8.0% vs 3.2%), and diabetes with complications (21.1% vs 16.4%) were more common among untreated patients. Median survival among untreated patients was 1.1 months (95% CI: not estimable) versus 9.2 months (95% CI:9.0-9.5 months) among treated patients. Approximately 5% of untreated patients were alive 12 months after diagnosis whereas 40% of treated patients were alive 12 months after initiating 1L treatment. Conclusions: Nearly half of SCLC patients did not initiate systemic anticancer treatment. Those who were untreated were on average older, had more comorbid conditions, were more likely to be diagnosed with stage IV disease, and had shorter survival, affirming the continued need for safe and effective therapies for SCLC.

8584 Background: Characterizing changes in the incident small cell lung cancer (SCLC) patient population and SCLC treatment landscape is essential for understanding drivers of outcomes among patients diagnosed with SCLC. The objective of this study was to examine trends in patient characteristics and treatment patterns for SCLC in a Medicare population. Methods: A retrospective analysis was conducted using the Surveillance, Epidemiology, and End Results (SEER)–Medicare linked data. Patients were included if they had a record with a diagnosis of SCLC in the database from 1/1/2007 – 12/31/2017 and were ≥65 years of age at diagnosis. Patients also had continuous enrollment for ≥180 days pre-SCLC diagnosis. Characteristics and treatment patterns were stratified by year of SCLC diagnosis. Results: 13,516 patients met the study criteria. Overall, 57.7% were female, mean age was 74 years, 89.5% were white, and 45.6% had a history of tobacco use. The majority (73.1%) were diagnosed with stage IV disease. Just over half (55.4%) of patients initiated first line (1L) treatment and mean time from diagnosis to 1L treatment was 6 weeks. Only 36.7% of those initiating 1L went on to second line (2L). Platinum-based chemotherapy (carboplatin or cisplatin + etoposide) was by far the most common 1L regimen (91.3%). 44.7% of patients received monotherapy in 2L, with topotecan being the most common (28.4%) regimen. Carboplatin + etoposide was used for approximately 20% of 2L patients. Over time, the demographic characteristics of patients diagnosed with SCLC was fairly stable with the exception of patients reporting a history of tobacco use, which more than tripled, increasing from 19.3% to 67.7% between 2007 and 2017. Between 2007 and 2017 the share of SCLC patients initiating 1L treatment increased 8.2% from 53.8% to 58.2% and the share initiating 2L fluctuated between 35% and 39%. Platinum-based chemotherapy was consistently used by the vast majority (88.7%–96.6%) of patients in 1L during the study period. Other chemo monotherapy (e.g., topotecan) and platinum-based chemotherapy with or without irinotecan were the most common 2L regimens from 2007-2014. Use of checkpoint inhibitors (CPIs) in 2L began in 2015 and became the most common 2L treatment following platinum-based chemotherapy by 2017 with 41.5% of patients receiving a CPI in 2L. Conclusions: While an increasing share of SCLC patients are pursuing treatment, less than 60% of SCLC patients receive any anticancer therapy. Fewer still receive more than one line of treatment, highlighting the ongoing need for effective therapies for SCLC.

e20601 Background: There is limited research evaluating real-world platinum sensitivity and treatment patterns among patients with limited vs. extensive stage small cell lung cancer (SCLC). Methods: We identified patients with SCLC in 2017 using ICD-10 diagnosis (Dx) and drug codes in the 2016-20 Medicare Parts A/B/D 100% Research Identifiable Files. Patients had 1+ lung cancer Dx and continuous enrollment in Parts A/B/D from the month prior to first Dx in 2017 (index) through 2020 or death. Patients with evidence of end-stage renal disease, prior lung or any metastatic cancer Dx within 12 months of index, lung resection/lobectomy, or use of a drug indicated only for non-SCLC or lung neuroendocrine cancer per NCCN guidelines were excluded. Use of systemic therapies cisplatin (ci), carboplatin (ca), etoposide intravenous (etIV), atezolizumab (at), durvalumab (du) and irinotecan (ir) were observed. Patients included in our study were treated with first-line NCCN Category 1 preferred regimens: ca/etIV, ca/etIV/at, ca/etIV/du, ca/ir, ci/etIV, ci/etIV/du, or ci/ir. First line of therapy (LoT1) was defined based on the first claim for a first-line regimen administered within 90 days following or up to 14 days before index. The end of LoT1 was defined as the earliest of death, 32 days after the last administration, or the start of a second line of therapy (LoT2). Patients initially treated with ca/etIV/at, ca/etIV/du, ca/ir, ci/etIV/du, or ci/ir regimens were flagged with extensive stage disease. Patients not receiving these regimens were classified as limited stage disease if they had 15+ days of radiation therapy over a 6-week period during or within 60 days of LoT1. Patients were deemed platinum sensitive (PS) if we did not observe a LoT2, hospice, or death within 6 months of LoT1. Uncategorizable patients were flagged Status Unknown (SU). All other patients were deemed platinum refractory (PR). Results: We identified 3,772 SCLC patients receiving NCCN Category 1 preferred regimens: 43% (1,607) had limited stage, and 57% (2,165) had extensive stage at index. Among patients with extensive stage, 71% (1,541) were PR compared with 34% (548) of patients with limited stage. On average, patients with extensive stage had fewer platinum cycles in LoT1 than those with limited stage. Conclusions: More platinum refractory patterns were observed among Medicare FFS patients with extensive stage SCLC than among patients with limited stage SCLC. Patients with extensive stage SCLC also underwent fewer platinum cycles in their first line of therapy.[Table: see text]

e20617 Background: Platinum-based chemotherapy plus irinotecan or etoposide is the standard of care for first line (1L) treatment of small cell lung cancer (SCLC). Sensitivity to platinum-based chemotherapy has implications for regimen choice in subsequent lines of treatment. The goal of this study was to examine differences in patient characteristics, treatment patterns, and survival associated with the management of SCLC in a Medicare population by platinum sensitivity status, i.e., platinum sensitive (PS) or platinum refractory (PR). Methods: A retrospective analysis was conducted using the Surveillance, Epidemiology, and End Results (SEER)–Medicare linked data. Patients were included if they had a record with a diagnosis of SCLC in the database from 1/1/2007 – 12/31/2017 and were ≥65 years of age at diagnosis. Patients also had continuous enrollment for ≥180 days pre-SCLC diagnosis and platinum-based 1L treatment. Patient characteristics, treatment patterns, and outcomes were stratified by PS status, which was defined as initiating 2L therapy more than 6 months after the last administration of platinum-based chemotherapy in 1L. Median overall survival (OS) was estimated using the Kaplan Meier method. Results: 7,208 patients met the study criteria and one in three (33.6%, N = 2,424) SCLC patients were PS. The PS population was majority female (64.4% vs 35.6% males), younger on average (aged 72.4 yrs vs 72.9 yrs), and more likely to be diagnosed at stage I-III than the PR patient population (50.6% vs 21.1%). Among PS patients, mean time from diagnosis to 1L treatment was 6.1 weeks versus 5.7 weeks among PR patients. Only 29.7% of PS patients initiating 1L went on to second line (2L) versus 40.4% of PR patients. 68.6% of PR patients received monotherapy in 2L, with topotecan (31.7%), paclitaxel (12.1%), and irinotecan (9.9%) being the most common monotherapy regimens. 51.7% of PS patients were retreated with platinum-based chemotherapy in 2L. Median OS from initiation of 1L was 20.2 months (95%CI: 19.3-21.1 months) among PS patients versus 6.4 months (95%CI: 6.2-6.6 months) among PR patients. 5-year survival was 20.0% among PS patients and 1.5% among PR patients. Conclusions: Approximately two-thirds of SCLC patients are refractory to platinum-based chemotherapy. These patients have fewer treatment options and shorter OS than PS patients, underscoring the ongoing need for effective therapies for SCLC.

580 Background: Subgroup analyses of the NAPOLI-1 study identified that among patients who were IRI naïve prior to entering the clinical trial, a survival benefit was observed between the study arm and control arm (overall survival (OS): 6.7 months vs 4.2 months). This treatment benefit was not observed among those previously exposed to IRI (OS: 4.6 months in study arm vs 4.8 months in control arm). This study sought to understand the impact of prior exposure to IRI on clinical outcomes among patients treated with liposomal irinotecan in the real-world setting. Methods: This retrospective observational study utilized the Flatiron Health EHR database. Data were analyzed for adult patients with mPDAC treated with liposomal irinotecan -based regimens between January 2016 and October 2020. Patient characteristics, OS and progression-free survival (PFS) were assessed. Prior IRI was defined as IRI given in a prior regimen in the metastatic setting. Cox proportional hazard (PH) methods were used to calculate hazard ratios (HRs). HRs were adjusted to account for demographics and relevant clinical covariates. Patients without prior exposure to IRI were used as the reference population for the Cox PH model (an HR < 1 represents worse survival for unexposed patients relative to the exposed). Results: 675 patients with mPDAC treated with a liposomal irinotecan-based regimen were included. Median age at treatment initiation was 69 (IQR: 62 – 75) years and among patients with available ECOG performance status (PS), 77.4% had a PS of 0-1. 181 (27%) patients were previously exposed to IRI in the metastatic setting (Table). The unadjusted OS HR was 1.3 (95% CI: 1.1 – 1.6, p < 0.001) and the unadjusted PFS HR was 1.4 (95%CI: 1.2 – 1.7, p < 0.001). After adjustment for baseline characteristics the adjusted OS HR was 1.0 (95% CI: 0.8 – 1.3, p = 0.8836) and the adjusted PFS HR was 1.1 (95%: 0.8 – 1.4, p = 0.5626). Conclusions: The results of this study suggest prior exposure to IRI is not a predictor of worse clinical outcomes for patients treated with liposomal irinotecan-based treatment when accounting for key clinical characteristics in a multivariable model. The results from this real-world study can be used to support treatment sequencing decisions for patients with mPDAC following first line therapy. This is the largest real-world evidence study to date of patients with mPDAC treated with liposomal irinotecan.[Table: see text]

531 Background: Medicare Advantage accounts for almost 40% of Medicare beneficiaries in 2021. There is limited research evaluating utilization and cost for Medicare Advantage patients with metastatic pancreatic cancer (m-PANC) receiving various NCCN Category 1 preferred regimens. Methods: We used ICD-10 diagnosis codes to identify patients with m-PANC without end-stage renal disease in the 2016-2019 Milliman Consolidated Health Cost Guidelines Sources Database (CHSD) claims files. Study patients had 2+ claims with a pancreatic cancer diagnosis and Medicare Advantage coverage for 3 months pre- and 1 month post-metastasis diagnosis. Patients with stand-alone Part D plan coverage or aged

579 Background: Few analyses have identified characteristics associated with extended DOT (eDOT) among a large group of patients (pts) treated with liposomal irinotecan. This study examines the characteristics associated with eDOT among pts with mPDAC treated with liposomal irinotecan in a real-world setting. Methods: This retrospective observational study utilized the Flatiron Health EHR database from over 280 cancer clinics in the US. Data were analyzed for adult pts with mPDAC treated with liposomal irinotecan-based regimens between January 2016 and October 2020. eDOT was defined as a DOT > 2 standard deviations (SD) above the mean. Demographic and clinical characteristics evaluated included age, sex, stage at diagnosis, ECOG performance score (PS), and the number of prior lines of therapy at the time of treatment initiation. Dose delays and dose reductions during treatment were assessed. Additionally, changes in lymphocyte counts during treatment were described. Chi-squared test and Fisher’s exact test were used to test for significance for categorical variables and the t-test was used for continuous variables. Results: 675 pts with mPDAC treated with a liposomal irinotecan-based regimen were included. The overall mean DOT was 14.6 weeks (SD: 22.1). There were 30 pts (4.4%) that met the definition for eDOT. Patient sex and median age at treatment initiation were similar between those with eDOT and non-eDOT: male, 46.7% vs. 51.9% (p = 0.582); age, median 70 years vs 69 years (p = 0.923), respectively. A higher proportion of pts with eDOT had ECOG PS of 0-1 compared to those with non-eDOT, 73.3% vs 57.7%, respectively (p = 0.397). At treatment initiation 80.0% of pts who had eDOT had a serum albumin level of ≥ 35g/L versus 55.2% of pts with non-eDOT (p = 0.013). On average, those with eDOT experienced a 55.1% reduction in lymphocyte count (baseline to nadir) during treatment compared to a 34.7% reduction among pts with non-eDOT (p < 0.0001). Pts who experienced eDOT were more likely to be treated in earlier lines of therapy than those with non-eDOT: 33.3% vs 14.1% received 0 prior lines and 46.7% vs 47.3% received 1 prior line, respectively (p = 0.008). 53.3% of pts with eDOT experienced a dose reduction (of at least 7mg/m2) compared to 24.7% pts with non-eDOT (p = 0.0005). All pts with eDOT experienced a dose delay of > 16 days during treatment versus 45.3% of pts with non-eDOT (p < 0.0001). Conclusions: This real-world study identified pts treated earlier in their disease course with a liposomal irinotecan-based regimen and normal baseline serum albumin were more likely experience extended DOT. Treatment modifications and delays appear to have enabled longer treatment duration with liposomal irinotecan-based regimens. Further prospective studies should be considered to characterize factors that may predict extended DOT.

297 Background: There is limited research evaluating the share of patients (pts) with metastatic pancreatic cancer (m-PANC) treated according to NCCN guidelines. Methods: We identified pts with m-PANC using ICD-10 diagnosis codes in the 2016-2019 Medicare Parts A/B/D 100% Research Identifiable Files. Study pts had 2+ claims with a pancreatic cancer diagnosis and Medicare FFS coverage for 6 months pre- and 3 months post-metastatic disease diagnosis. A line of therapy (LOT) was assigned based on the order and number of therapies used. Pts with one, two, or three LOTs were defined as treated according to NCCN Category 1 guidelines if, in each LOT, pts used one of the following regimens: FOLFIRINOX (FFX), gemcitabine/nab-paclitaxel (gem/nab), gemcitabine + erlotinib, gemcitabine monotherapy, or 5-FU + leucovorin + liposomal irinotecan. Multi-drug LOTs were excluded from the analysis. Results: We identified 31,782 pts with m-PANC. 21,304 received one LOT, 7,352 received two LOTs, and 3,126 received three LOTs between 2016 and 2019. Among pts who received one or two LOTs, a higher portion were treated according to NCCN Category 1 guidelines in 2019 (72% and 43%, respectively) than in 2016 (64% and 33%, respectively). Among pts who received three LOTs, a higher portion were treated according to NCCN Category 1 guidelines in 2019 (17%) than in 2017 (12%); too few pts were treated in 2016 to make a comparison. From 2016 to 2019, FFX had the largest increase in share of pts receiving only one NCCN Category 1 LOT (11% to 27%) and gem-mono had the largest decrease (30% to 17%). Among pts receiving two NCCN Category 1 LOTs, gem/nab to liposomal irinotecan sequences had the largest increase in share of pts (18% to 32%) and gem/nab to FFX had the largest decrease (17% to 10%). Among pts receiving three NCCN Category 1 LOTs, patient share for FFX to gem/nab to Liposomal irinotecan was 35% in 2019, while gem/nab to FFX to Liposomal was 8%; pt counts in earlier years were too small to calculate patient share. Conclusions: The use of NCCN Category 1 therapies increased consistently from 2016 to 2019 among pts that received one, two, and three lines of therapy. FFX drove increases in NCCN Category 1 utilization among patients receiving one line of therapy, and gem/nab to liposomal irinotecan sequences were the primary drivers of the increase among patients receiving two lines of therapy. FFX to gem/nab to liposomal irinotecan was the primary driver of increase among patients receiving three lines of therapy.

302 Background: Helping patients manage cancer-related pain is an important aspect of supportive care. 80% of patients with mPDAC suffer from abdominal and/or back pain. Pancreatic cancer-related pain is typically treated pharmaceutically using pain medicines, which include opioids or narcotics. Previous studies have found that the presence of pain was associated with impaired survival in pancreatic cancer patients. The goal of this study was to assess the treatment patterns of pain medication usage and their association with duration of therapy (DOT) among patients with mPDAC treated in the real-world setting. Methods: This retrospective observational study utilized the IBM MarketScan Commercial and Medicare supplemental claims data. Data were analyzed for adult patients who were diagnosed with mPDAC and received treatment between January 2015 and March 2020. Patient characteristics and DOT were assessed. Opioid use during treatment including the number of prescriptions filled and average daily dose in morphine equivalents were summarized. Yearly and regional trends of opioid use were assessed. Results: There were 2,841 patients (median age: 61 years, IQR: 56-66) included in the study treated with first line (1L) therapies of interest. 55.6% (n = 1,579) of patients were male and 72.0% (2,046) had commercial insurance. The overall DOT in 1L was 78 days (IQR: 38 – 157). 54.3% of patients filled at least one prescription for opioids during treatment. The mean number of prescriptions filled during treatment was 2.2 (SD: 4.0) and the mean daily dose was 39.5mg morphine equivalents (SD: 134.6) The median DOT among patients who received at least one pain prescription was 92 days (IQR: 44 – 176) while those who did not receive pain medication experienced a DOT of 64 days (IQR: 29 – 135). Among the 1,248 patients who received second line treatment the median DOT was 60.5 days (IQR: 29 – 127). 50.6% (n = 632) filled at least one opioid prescription during treatment. Median DOT among patients who received pain medication was 73.5 days (IQR: 42 – 141) and 47.5 days (IQR: 15 – 103.5) among those who did not. Across lines of therapy opioid use remained stable during the study period ranging from 54.2% among patients treated in 2015 to 51.1% among those who initiated treatment in 2019. No regional differences were observed in prescription patterns within the four US census regions (range: 50.2 – 54.4%). Conclusions: In this real-world cohort of patients with mPDAC nearly half were prescribed at least one opioid during treatment. Patients who received opioids experienced a longer duration of therapy compared to those who did not. Further studies are needed to understand the association of pain control with improved clinical outcomes among patients with mPDAC.

e16252 Background: Treatment options remain limited for pts newly diagnosed with mPDAC. NCCN and ASCO guidelines both recommend treatment with FOLFIRINOX or gemcitabine plus nab-paclitaxel in the first line (1L) setting for pts with mPDAC and good performance status. Over 80% of randomized clinical trials (RCTs) studying treatments for mPDAC have failed to meet their primary endpoints. Recently published analyses have utilized real-world (RW) data to compare outcomes between pts enrolled in clinical trials and RW pts. This study sought to identify the eligible population of pts with mPDAC treated with FOLFIRINOX in the 1L setting who would meet RCT eligibility criteria for the phase 1/2 study (NCT02551991) evaluating NALIRIFOX for pts with previously untreated locally advanced or mPDAC, and to assess their survival outcomes. Methods: This retrospective observational study utilized the Flatiron Health EHR database. Data were analyzed for adult pts diagnosed with mPDAC between January 2016 and February 2020 who initiated treatment with FOLFIRNOX in 1L within 90 days of their diagnosis for metastatic disease. Eligibility criteria from the phase 1/2 trial were applied to select a population of RW pts who may have been eligible to enter the RCT. Pts meeting the following criteria were included: good performance scores (ECOG 0-1), adequate hematological, hepatic, and renal function, were recovered from the effects of surgery, were untreated in the year prior to initiating 1L, and had no evidence of a different cancer in the last three years. Kaplan-Meier analyses were used to assess the median overall survival (mOS) from the start of 1L FOLFIRNOX treatment. Results: Of the 1,210 pts treated with 1L FOLFIRINOX, 652 pts (53.8%) met less stringent versions of the RCT eligibility criteria in which missing values were deemed to indicate normal function/performance; 244 pts (20.2%) met the more stringent criteria and had complete data. The most restrictive selection criteria were the requirements for adequate hematological, hepatic, and renal function and having received prior therapy. The median age at treatment initiation among the 244 pts was 64 years (IQR: 58 – 70). 153 pts were male (62.7%), 158 were White (64.8%), and ECOG scores of 0 and 1 were split among the cohort 50%/50%. The mOS observed for the 244 pts was 10.1 months (95% CI: 9.1 – 11.3). The reported mOS from the phase 1/2 trial of NALIRIFOX was 12.6 months (8.7 – 18.7). Conclusions: This study demonstrates that RW data may be used to select a comparator cohort for a clinical trial. Initial estimates suggest NALIRIFOX pts from the RCT experienced longer survival than those receiving 1L FOLFIRINOX in the RW setting. Further analysis is necessary to minimize the effects of confounding and the differences in data collection between the RW and the RCT settings.

e16251 Background: Pancreatic cancer has an aggressive disease course, mandating close clinical monitoring while on treatment. Evaluation of treatment response in patients with mPDAC necessitates scheduled clinical and radiographic assessments along with monitoring serum CA 19-9 levels. We investigated the impact of serum CA 19-9 monitoring and its association with clinical outcomes in patients with mPDAC who received second- (2L) and third line (3L) in a population-based setting. Methods: Data were extracted from the nationwide Flatiron Health electronic health record (EHR)-derived de-identified database for patients diagnosed with mPDAC and subsequently treated in the 2L or 3L setting between January 1, 2014 and June 30, 2020. Serum CA 19-9 levels at baseline were extracted – defined as the values obtained ≤ 60 days of treatment initiation and during treatment. CA 19-9 levels > 40 IU/mL were considered elevated. Survival analysis was performed using Kaplan-Meier methods. Categorical measures were compared with a chi-square test and survival outcomes with a log-rank test. Results: There were 2,402 patients who received 2L and 790 patients who received 3L treatment included in the study. Among the 2L cohort, median age at treatment initiation was 67 years (IQR: 60 – 73), 54% were male, 57% had an ECOG score of 0-1, and 82% had a baseline serum CA 19-9 level available. Among the 3L cohort, median age at treatment initiation was 67 years (IQR: 60 – 73), 53% were male, 58% had an ECOG score of 0-1, and 84% had a baseline CA 19-9 level available. Most patients in the 2L and 3L cohorts had an elevated CA 19-9 at baseline, 85.2% and 82.0%, respectively. Among the 2L and 3L cohorts, 38.5% and 31.9% had CA 19-9 levels decrease/remain stable during treatment and 27.6% and 31.4% had levels increase during treatment, respectively. Patients with normal baseline CA 19-9 experienced longer survival than patients with elevated levels [2L: 7.2 months (95% CI: 6.1 – 9.2) vs 5.2 months (4.9 – 5.6), p < 0.001; 3L: 6.1 months (5.4 – 9.1) vs 3.9 months (3.4 – 4.3), p < 0.001]. Similarly, patients with decreasing/stable CA 19-9 during treatment had longer survival than patients who had their CA 19-9 levels increase [2L: 8.2 months (7.7 – 8.5) vs 4.3 months (4.1 – 4.7), p < 0.001; 3L: 7.5 months (6.6 – 9.2) vs 3.7 months (3.4 – 4.3), p < 0.001]. Conclusions: In this large, contemporary, real-world study of patients with mPDAC, CA 19-9 levels at treatment initiation had a prognostic value across later lines of therapy. Routine serial monitoring of CA 19-9 levels during treatment, in addition to clinical and radiographic assessment, may help with timely additional diagnostic testing and treatment intervention. Further validation studies are needed to understand the generalizability of these results.

e16248 Background: Chemotherapy-related adverse events (AEs) can negatively impact the treatment of pts. Managing and preventing these toxicities often require additional medications. This study examined the proportion of pts with mPDAC treated with 5-FU–based regimens in 2L who experienced AEs during treatment, and the proportion who received medication to manage those AEs. Methods: Data were extracted for pts diagnosed with mPDAC who initiated 2L treatment Jan 2016 to Aug 2020 from the Flatiron Health database. Pts included were treated with FOLFIRINOX (FFX), FOLFOX, FOLFIRI, or liposomal irinotecan (nal-IRI). The occurrence of grade 3 (G3) and grade 4 (G4) neutropenia and G3/G4 thrombocytopenia were determined using lab results and the grading criteria from the CTCAE v4.03. The occurrence of diarrhea and nausea/vomiting (N/V) were identified via ICD-10-CM codes. The use of atropine and granulocyte colony stimulating factor (G-CSF) during treatment was assessed. Duration of therapy (DOT) was assessed for each regimen. Descriptive statistics for AEs, medication use, and DOT were reported. Results: Of the 825 eligible pts, 29.0% (n=239) received FFX, 40.2% (n=332) received regimens containing nal-IRI, 24.0% (n=198) received FOLFOX, and 6.8% (n=56) received FOLFIRI. The median DOT (IQR) was 15.1 weeks (wks) (7.1 – 30.1), 12.9 wks (6.0 – 24.8), 11 wks (6.4 – 24.0), and 12.7 wks (8.3 – 24.6) for pts who received FFX, nal-IRI, FOLFOX, and FOLFIRI, respectively. G3/G4 thrombocytopenia (3) presented in 9.6%, 2.4%, 8.1%, and 14.3% of pts treated with FFX, nal-IRI, FOLFOX, and FOLFIRI, respectively. G3/G4 neutropenia (3) presented in 22.6%, 12.3%, 14.6%, and 19.6% of pts treated with FFX, nal-IRI, FOLFOX, and FOLFIRI, respectively. The median (Q1-Q3) cumulative dose of the G-CSF pegfilgrastim was 18mg (12 – 45), 18mg (12 – 48), 24mg (12 – 36), and 30mg (13 – 48) for pts treated with FFX, nal-IRI, FOLFOX, and FOLFIRI, respectively. Atropine use was higher in pts treated with FFX and FOLFIRI (90.8% and 94.6%, respectively) than for pts treated with nal-IRI (75.6%). The full study results are reported in the table. Conclusions: In this study of AEs among pts with mPDAC treated in 2L, pts who received liposomal irinotecan had the lowest proportion of thrombocytopenia and neutropenia. Diarrhea was similar across the cohorts despite the high proportion of use in atropine for pts treated with FFX and FOLFIRI.[Table: see text]

e18843 Background: The Department of Health and Human Services (HHS) designed the 340B drug pricing program to allow institutions that service specialty populations to acquire drugs at lower prices. Objective: To analyze the dispersion in total cost of care (TCOC) for Medicare FFS patients (pts) with metastatic pancreatic cancer (m-PANC) treated at 340B or non-340B institutions, by NCCN Category 1 regimen. Methods: We identified pts with m-PANC using ICD-10 diagnosis codes in the 2016-18 Medicare Parts A/B/D 100% Research Identifiable Files. Study pts had 2+ claims with a pancreatic cancer diagnosis and Medicare FFS coverage for 6 months pre- and 3 months post-metastasis diagnosis. Study pts were treated with NCCN Category 1 regimens: 1L gemcitabine monotherapy (gem-mono), 1L gemcitabine/nab-paclitaxel (gem-nab), 1L FOLFIRINOX (FFX), and 2L liposomal irinotecan-based regimen (nal-IRI). Pts were attributed to 340B or non-340B institutions based on plurality of chemotherapy claims. TCOC reflects insurer-paid services per line of therapy (LOT) for 3 categories: chemotherapy/supportive drugs (chemo/Rx), inpatient care (IP), and other outpatient care (OP). We grouped pts by quartile (qrt) and evaluated drivers of TCOC and mean rates of admissions (admits/pt). Results: We identified 2,697 (340B) and 3,839 (non-340B) pts taking NCCN Category 1 regimens. Gem-mono represented 1% and 4% of all pts in 340B and non-340B institutions, respectively. Gem-nab accounted for 72% of pts in both cohorts. For gem-nab, FFX, and nal-IRI pts, median TCOC was similar in both cohorts, although mean TCOC by qrt was lower at 340B institutions than non-340B institutions, except for gem-nab in the 1st qrt. The components of TCOC were similar between 340B and non-340B institutions in all qrts. In both cohorts, % IP costs increased between the 1st and 4th qrt (340B:15% to 23%, non-340B:14% to 25%). From the 1st to the 4th qrt, admits/pt increased in both cohorts. In the 340B cohort, nal-IRI pts had the lowest admits/pt while gem-nab pts had the highest in all qrts. In the non-340B cohort, nal-IRI pts had the lowest admits/pt except for in the 1st qrt. Conclusions: Median TCOC was lower at 340B institutions than non-340B institutions for all regimens, and the range of TCOC dispersion was also smaller at 340B institutions. Across qrts, chemotherapy accounted for approximately half the TCOC; however, IP costs were proportionally higher in the 4th qrt. Comparing regimens, despite 2L nal-IRI pts being more heavily pretreated, median costs in each cohort were similar to 1L gem-nab and 1L FFX, while admits/pt were generally lower than 1L gem-nab and 1L FFX across qrts and cohorts.

e16250 Background: PDAC is a lethal malignancy which accounted for the third most cancer related deaths in 2020. Patients (pts) who are initially diagnosed with stage I-III PDAC have a 5-year relative survival of 13.3 – 39.4%; those with metastatic disease at diagnosis have a 5-year relative survival of 2.9%. Limited data are published comparing the outcomes of pts with stage I-III who develop metastases (recurrent) compared to pts with de novo mPDAC (de novo). This analysis seeks to compare demographic, clinical characteristics, and survival outcomes of pts with recurrent versus de novo mPDAC in a community oncology setting. Methods: Using the Flatiron Health database, a retrospective observational study was conducted abstracting deidentified data from ≥280 US cancer clinics. Pts with mPDAC diagnosed from 01/2016 to 08/2020 with a known stage at initial diagnosis were included. Pts were stratified based on initial stage at diagnosis. Median overall survival (OS) from time of metastasis was derived using Kaplan-Meier analysis. Unadjusted and multivariable Cox proportional hazards models were used to compare survival between recurrent and de novo cohorts. Results: N = 6,543 pts analyzed; 70.1% (n = 4,586) had de novo mPDAC and 29.9% (n = 1,957) had recurrent mPDAC. Median age at time of metastasis was similar for both cohorts: 69 years (IQR: 62 – 76). The most common site of primary tumor location was head for both cohorts (recurrent mPDAC: 69.8%; de novo mPDAC: 40.3%). Approximately 45% of pts with recurrent mPDAC underwent a Whipple procedure (pre diagnosis of metastasis). A similar proportion of pts in both cohorts received treatment in the metastatic setting (recurrent mPDAC: 74.3%; de novo mPDAC: 77.3%). Pts with recurrent mPDAC had a longer median OS compared to the de novo cohort: 8.0 months (95% CI: 7.5 – 8.6) versus 6.1 (95% CI: 5.7 – 6.4) [unadjusted hazard ratio (HR): 0.79 (95% CI: 0.74 – 0.84); adjusted HR: 0.73 (0.68 – 0.78), p < 0.0001]. Conclusions: The results of this real-world study indicate that pts with recurrent mPDAC are more likely to have a head primary and to experience longer OS from time of metastasis than those with de novo mPDAC. These data suggest stratification for clinical trial enrollment for recurrent vs de novo is necessitated.

4137 Background: The coronavirus disease 2019 (COVID-19) pandemic has caused abrupt changes to the US health system and disruption in cancer care delivery. Little has been reported on the impact of COVID-19 on patients with mPDAC and the care delivery. Our study aimed to characterize the impact of COVID-19 on healthcare utilization and outcomes for patients with mPDAC in the US in the community oncology setting. Methods: We performed a retrospective cohort study of adult patients diagnosed with mPDAC between March – September 2019 and March – September 2020 using the nationwide Flatiron Health EHR database, comprising data from over 280 (largely community based) cancer clinics. Patients were stratified into two cohorts based on the year of diagnosis (2019 vs. 2020). Clinical characteristics were summarized including age at metastatic diagnosis, stage at initial diagnosis, and ECOG performance score (PS). Overall survival (OS) from metastatic diagnosis was estimated using Kaplan-Meier methods. A sensitivity analysis limiting the follow-up time to November of each year was conducted. Results: Overall, 1,719 patients were included in the study (2019: n = 923, 2020: n = 796); both cohorts had similar demographic compositions in terms of age and sex (2019: median age = 70 (IQR: 62 – 76), 52.2% male; 2020: median age = 70 [IQR: 62 – 76], 53.5% male). In 2020, the number of newly diagnosed mPDAC patients decreased by 13.8% compared to 2019. A slightly higher proportion of patients were initially diagnosed with stage IV disease in 2020 (69.7%) vs 2019 (62.3%). A similar proportion of patients with ECOG PS 0-1 was observed between the two cohorts (2019: 48.5%; 2020: 47.9%). The number of visits recorded within the first 90 days after metastatic diagnosis was similar between the two cohorts (2019: median 8 [IQR: 3 – 14]; 2020: median 9 [IQR: 4 – 14]). For both cohorts, the proportion of patients who received 1L treatment was similar (2019: 75.8%; 2020: 76.5%), and the most common 1L treatment regimen was gemcitabine plus nab-paclitaxel (2019: 37.6%; 2020: 40.9%). Of the 1L treated populations, 37.6% of patients diagnosed in 2019 received second line (2L) compared to 17.9% of the 2020 cohort; 16.9% of 1L treated patients in 2019 received 2L in the sensitivity analysis. Patients diagnosed in 2020 had a significantly lower median OS of 6.1 months (95% CI: 5.4 – 6.9) compared to patients diagnosed in 2019: 8.4 months (7.5 – 9.0) (p < 0.001). Conclusions: During the COVID-19 pandemic era, the diagnoses of de novo mPDAC appears to have been impacted, with a higher number of patients diagnosed with advanced stage at presentation. Our analysis suggests that while patients diagnosed in 2020 received a similar level of care as those in 2019, their survival outcomes were adversely affected. Further research is necessary to characterize the impact of the COVID-19 pandemic on cancer care and outcomes.

389 Background: Pancreatic cancer is expected to be the third deadliest cancer in the US in 2020. Many real-world studies of pts with mPDAC are restricted to single centers, limiting the generalizability of the insights they generate. There is a need to understand prognostic factors of survival in a broader setting to aid in tailoring treatment strategies for pts. This study aimed to identify important population-based predictors related to survival among pts diagnosed with mPDAC. Methods: Data were extracted for pts diagnosed with mPDAC between Jan 2017 and Dec 2019 from the Flatiron Health database. Predictive models for overall survival from the start of each treatment were developed using multivariable Cox proportional hazards regression. Treatment specific predictive models were generated for pts treated with first line (1L) gemcitabine + nab-paclitaxel (GNP), 1L FOLFIRINOX, 1L gemcitabine monotherapy (gem-mono), and 2L liposomal irinotecan-based regimens. The holdout method was used for cross-validation, splitting the data into 70% training / 30% validation. Age at diagnosis, sex, body mass index, smoking status, and ECOG performance score (PS) were included in all models due to clinical importance. Demographic, clinical characteristics, hematological labs, liver function tests (LFTs), and serum bilirubin levels were assessed for inclusion into the models. Uno’s concordance statistic (c-statistic) was used to assess the predictive accuracy of the models. Results: Of the 3,572 pts included in the study, 44% (n = 1,557) received 1L GNP, 27% (n = 954) received 1L FOLFIRINOX, 7% (n = 265) received gem-mono, and 22% (n = 796) received other regimens. 38% (n = 1,345) pts received 2L and of those, 17% (n = 222) received liposomal irinotecan-based regimens. Among all 1L pts, the following were included in the final model: prior surgery, white blood cell (WBC) counts, serum albumin, LFTs (ALP and ALT), serum bilirubin, and ascites (c-statistic: 0.65). The model for pts treated with GNP differed from the overall model via the addition of neutrophil counts and removal of serum bilirubin and ascites (c-statistic: 0.67). Stage at initial diagnosis was included in the model only for pts treated with 1L FOLFIRINOX (c-statistic: 0.68). Among pts treated with gem-mono the LFTs were not included in the model (c-statistic: 0.78). ALP, serum albumin, and WBC counts were important predictors of survival among pts treated with 2L liposomal irinotecan-based regimens (c-statistic: 0.70). Across all regimens the strongest predictors of survival were ECOG PS, serum albumin, and ALP. Conclusions: In one of the largest contemporary real-world studies of patients with mPDAC to date, important population predictors of survival in pts receiving systemic treatment were identified. Further validation studies are needed to understand the generalizability of these results.

392 Background: Pancreatic cancer remains one of the most lethal cancers in the United States (US) with a 5-year relative survival of 9%. Among patients who received liposomal irinotecan + 5-fluorouracil (5-FU) and leucovorin in the NAPOLI-1 study, a randomized phase 3 study in patients with mPDAC previously treated with gemcitabine-based therapy, 25% (n = 29) were alive at ≥ 1 year. This study examines the real world one-year survival of patients with mPDAC treated with liposomal irinotecan as a doublet with 5-FU in the NAPOLI-1 based regimen. Methods: This retrospective observational study utilized the Flatiron Health EHR database from over 280 cancer clinics in the US. Data were analyzed for adult patients with mPDAC treated with liposomal irinotecan-based regimens between November 2015 and July 2020. Patient characteristics and one-year overall survival (OS) based on Kaplan-Meier estimates were assessed. Cycles were defined as unique days with an administration of liposomal irinotecan. Results: There were 669 patients (median age: 69y, IQR: 62-75) included in the study. ECOG performance status (PS) of 0-1 and 2+ were reported for 78.3% (n = 396) and 21.7% (n = 110) of patients, respectively. ECOG PS was not captured for 24.4% (n = 163) of patients. 16.3% (n = 109) of patients initiated liposomal irinotecan-based in the first line (1L) metastatic setting, 47.5% (n = 318) in second line (2L), and 36.2% (n = 242) in the third line or later (3L+). The median number of cycles received was 4 (IQR: 2 – 8). Among all patients, one-year OS was 17.2% (95% CI: 14.3% - 20.7%). One-year OS was 31.5% (22.1% – 41.3%) for patients treated in 1L, 16.4% (12.2% - 21.1%) for patients treated in 2L, and 12.2% (7.5% - 18.0%) for patients treated in 3L. The median number of cycles for 1L, 2L, and 3L were 5, 4, and 3, respectively. One-year OS increased as patients were able to receive more cycles of liposomal irinotecan. Patients who received at least 2 cycles of liposomal irinotecan (n = 551) had a one-year OS of 20.4% (16.8% - 24.2%). Among patients who received at least 4 cycles (n = 359) and at least 8 cycles (n = 170), the one-year OS estimates were 29.1% (24.0% - 34.3%) and 47.9% (39.7% - 55.7%), respectively. Conclusions: In this real-world cohort of patients with mPDAC treated with liposomal irinotecan, as expected, one-year OS increased as patients remained on therapy. Patients in this cohort were older, had more prior lines of therapy, worse ECOG PS, and similar exposure to treatment compared with patients in the registrational phase 3 NAPOLI-1 study. Among patients who received at least 4 cycles of liposomal irinotecan, one-year OS (29%) was similar to both the intent-to-treat (25%) and per protocol treated patient populations in the NAPOLI-1 trial (34%). Further studies are needed to understand the predictors of long-term survival among patients with mPDAC.

390 Background: Chemotherapy related adverse events (AEs) can impact the treatment of patients, reducing quality of life and leading to dose delays and treatment discontinuation. This study examined the proportion of patients (pts) with mPDAC treated with 5-FU-based regimens in the 2L setting who experienced AEs during treatment. Methods: Data were extracted for pts diagnosed with mPDAC who initiated 2L treatment between January 2016 and July 2020 from the Flatiron Health electronic health database. Pts included in the study were treated with FOLFIRINOX (FFX), FOLFOX, FOLFIRI, or a regimen containing liposomal irinotecan. The occurrence of grade 3 (G3) and grade 4 (G4) neutropenia, G3/G4 elevated alanine transaminase (ALT) and anemia where transfusion was indicated were determined using lab results and the grading criteria from the Common Terminology Criteria for Adverse Events v4.03. The occurrence of diarrhea, fatigue, nausea and vomiting (N/V), and neuropathy were identified from structured diagnosis records through ICD-10-CM codes. Duration of therapy (DOT) was assessed for each regimen. Descriptive statistics for AEs and DOT were reported. Results: Of the 804 pts included in the study, 28.4% (n=228) received FFX, 39.8% (n=320) received regimens containing liposomal irinotecan, 24.8% (n=199) received FOLFOX, and 7.1% (n=57) received FOLFIRI. The median DOT (IQR) was 86 days (d) (43 – 206), 79d (41 – 169), 72d (43 – 166), and 84d (46 – 148) for pts who received FFX, liposomal irinotecan, FOLFOX, and FOLFIRI, respectively. G3/G4 neutropenia (

388 Background: Pancreatic cancer is mostly diagnosed in patients (pts) aged ≥ 65 years and the mortality rate is the highest among older adults. As the population ages, it is expected that there will be a significant rise in the number of older pts with mPDAC but guidance regarding their management is limited as these pts are under-represented in clinical trials. The oncological care of older adults in daily practice is challenged by various age-related conditions. Therefore, a better understanding of the real-world population will help in the development of more effective tx strategies. This study describes the proportion of pts with mPDAC who were treated, the types of regimens received, and the associated survival outcomes by age at diagnosis. Methods: Data were extracted for pts diagnosed with mPDAC between Jan 2015 and Mar 2020 from the Flatiron Health database. Pts were stratified into three age groups at diagnosis: 70% of regimens prescribed to pts aged 80y+. The median OS (mOS) for patients treated with GNP, gem-mono, FOLFIRINOX, and liposomal irinotecan-based regimens are presented in the table. Conclusions: This study of treatment patterns among pts with mPDAC found that older pts were more likely to have a decreased performance status and are less likely to receive treatment. However, older adults who received systemic therapy had comparable survival outcomes to younger pts treated with a similar regimen. The results of this large descriptive analysis suggest that the treatment strategy of mPDAC should not be based on age but rather on an overall assessment of the performance/functional status and geriatric profile of older pts. [Table: see text]

e16740 Background: The goals of randomized control trials (RCTs) are to make causal inferences and precise treatment comparisons, not to describe large heterogeneous pt populations. RWE allows population-based healthcare decision makers to assess and manage therapeutic and economic options for their pts, including those who would and would not have met inclusion/exclusion criteria of a given RCT and are instead managed under usual care, irrespective of clinical trial protocols. In the pivotal phase 3 trial, NAPOLI-1, 117 pts were treated with liposomal irinotecan + 5-fluorouracil/folinic acid, median age 63 years; 66% were treated first- (1L) or second line (2L), and 91% had performance score ECOG 0 or 1. Pts in the trial had overall survival (OS) of 6.2 months (mos), time to treatment failure (TTF) 2.3 mos, and 27% experienced grades 3-4 neutropenia. The present study describes the patient characteristics and outcomes of pts with mPDAC treated with liposomal irinotecan in the US. Methods: This retrospective observational study used data from Flatiron Health EHR-derived de-identified database from over 280 cancer clinics. Patient characteristics, OS, TTF, and rates of neutropenia during treatment (tx) were assessed in adult pts diagnosed with mPDAC who received liposomal irinotecan based tx between November 1, 2015 and October 31, 2019. Results: 600 pts with mPDAC treated with a liposomal irinotecan based regimen were identified. Of these, 56% were initially diagnosed with stage IV disease, 53% were male, 21% had undergone a previous Whipple procedure, and 61% initiated liposomal irinotecan in the 1L or 2L metastatic setting. Median age at tx initiation was 68 (IQR: 62 – 75) years. 92% of pts were treated in the community setting. Among pts with available ECOG (n = 440), 77.5% were score 0-1. Grade 3/4 neutropenia was observed in 11% (n = 66). Overall, median OS was 5.0 mos [95%CI: 4.2–5.6]. mOS among pts treated in 1L (n = 88), 2L (n = 280), and third line plus (3L+, n = 232) were 6.9 mos [5.3–9.2], 5.4 mos [4.6–6.4], and 4.0 mos [3.4–4.5], respectively. Overall, median TTF was 1.9 mos [1.6–2.1]. TTF by line was 3.5 mos [2.3–4.8] in 1L, 2.1 mos [1.7–2.8] in 2L, and 1.4 mos [1.2–1.6] in 3L. Conclusions: This real-world cohort of pts with mPDAC were older, had worse performance status, and had more prior lines compared to the pivotal trial for liposomal irinotecan. Median OS, TTF, and neutropenia were similar to those previously reported. As expected, pts receiving liposomal irinotecan in earlier lines had higher median OS and TTF.

e16733 Background: Real-world data allows healthcare decision-makers to assess and manage therapeutic and economic options for patients, including those who would and would not have met eligibility criteria for randomized control trials (RCT) and are instead managed under usual care. This retrospective multi-academic center chart review study describes real-world characteristics and outcomes of US patients receiving liposomal irinotecan for the management of metastatic pancreatic ductal adenocarcinoma (mPDAC). Methods: Patients with mPDAC treated with liposomal irinotecan were eligible. Initiation of liposomal irinotecan defined index date; covariates assessed included clinical characteristics and treatment patterns; real-world overall survival (rwOS) was assessed via Kaplan-Meier methodology. The target enrollment is 300 patients. The study centers included were Memorial Sloan Kettering Cancer Center, Cedars-Sinai Medical Center, Emory Winship Cancer Institute, Houston Methodist Cancer Center, Henry Ford Cancer Institute, and University of Pittsburgh Medical Center. Results: Data on 26 patients were available for initial analyses. Mean age was 68 years; 58% were female and 65% Caucasian. 54% of patients had stage IV disease at first diagnosis, and 17%, 65%, and 17% had index ECOG score of 0, 1, and 2, respectively. Common genetic mutations include KRAS (40%) and TP53 (40%). Prior to liposomal irinotecan, treatments received for metastatic disease include gemcitabine+nab-paclitaxel (77%) and fluorouracil (5-FU)/leucovorin (LV)+irinotecan+oxaliplatin (19%). Patients had received 0 (12%), 1 (23%), and ≥2 (65%) lines of therapy in the metastatic setting prior to liposomal irinotecan. Mean duration of liposomal irinotecan use was 3.0 months; liposomal irinotecan was mostly received with 5-FU (23%) or 5-FU/LV (69%). Median rwOS was 4.9 months (95% CI: 3.0, 6.3). Conclusions: Real-world data of the first 26 patients in this study show patients treated with liposomal irinotecan are older, sicker, and have had more lines of therapy than previously reported in RCT data.

e16780 Background: Pancreatic cancer remains one of the most lethal cancers in the United States (US) with a 5-year relative survival of 10%. Liposomal irinotecan is a topoisomerase inhibitor indicated, in combination with 5-fluorouracil and leucovorin, for pts with mPDAC following disease progression on gemcitabine-based therapy. This study examines the real-world use and therapeutic management of pts with mPDAC treated with liposomal irinotecan. Methods: This retrospective observational study utilized the Flatiron Health EHR-derived de-identified database from over 280 cancer clinics in the US. Data were analyzed for adult pts with mPDAC treated with liposomal irinotecan based regimens between Nov 2015 and Oct 2019. Pts were categorized into two starting dose groups: 70mg/m2 and < 65mg/m2. Pt characteristics, overall survival (OS), duration of treatment (DOT), and impact of dose reductions (DR, reduction ≥ 7mg/m2) were assessed among pts who received ≥3 cycles of treatment (tx). Results: Of the 532 pts (median age: 69y, IQR: 62-75) included in the study, 95 (18%) had an ECOG score of 2+ at tx initiation. Of the 184 pts (69y, 42 – 84) that did not receive 3 cycles of tx, 47 (25%) had an ECOG score of 2+ and 83 (45%) had two or more prior lines of tx. 348 pts (69y, 43 – 85) received ≥3 cycles of tx. 116 (33%) pts had two or more prior lines of tx, 209 (60%) had an ECOG score of 0-1, 48 (14%) had an ECOG score of 2+, and 91 (26%) had missing scores. 220 (63%) initiated tx at 70mg/m2 and 128 (37%) initiated at < 65mg/m2. 83 (38%) 70mg/m2 and 26 (20%) < 65mg/m2 pts experienced a DR during tx. 43 (52%) and 14 (54%) of the DRs occurred within the first 6 wks of tx in the 70mg/m2 and < 65mg/m2 cohorts, respectively. Median DOT was 12.6 weeks for 70mg/mg2 and 9.1 wks for < 65mg/m2 pts; DOT was longer among pts with a DR: 19.0 wks and 16.1 wks, respectively. Median OS (mOS) was 7.2 months (95% CI: 6.2 – 8.1) and 6.2 mos (5.0 – 7.4) for pts receiving 70mg/m2 and < 65mg/m2, respectively. mOS for pts with a DR was 8.9 mos (7.3 – 10.8) and 7.7 mos (5.0 – 14.9) for pts receiving 70mg/m2 and < 65mg/m2, respectively. mOS for pts with no DR was 6.0 mos (4.8 – 7.2) and 6.0 mos (4.7 - 7.2) for those receiving 70mg/m2 and < 65mg/m2, respectively. Conclusions: In this descriptive study among pts who were able to receive ≥3 cycles of liposomal irinotecan and remain on tx for ≥4 wks, DRs were effective in extending DOT and OS, independent of starting dose. The longest DOT and OS were observed in the pts who received 70mg/m2 with DRs. Pts who received 70mg/m2 and < 65mg/m2 had similar OS in the absence of DRs.

e16739 Background: There are currently four NCCN category 1 systemic regimens approved in the United States for the treatment of mPDAC: FOLFIRINOX (FFX), gemcitabine+nab-paclitaxel (gem+nab-P), gemcitabine monotherapy (gem), and liposomal irinotecan + 5-fluorouracil/leucovorin (5-FU/LV) following progression with gem-based therapy. There is limited real-world research on the IP admissions and ER visit healthcare resource utilization (HRU) of patients receiving these treatments. Methods: Using the IQVIA PharMetrics Plus administrative claims database, data were analyzed for adult patients with mPDAC treated with NCCN category 1 regimens in first through fourth line of therapy between January 1, 2014 and May 31, 2019. For each line of therapy, continuous treatment was defined as the time from first administration of a therapy until the last administration. Mean all-cause and mPDAC-related IP admissions, ER visits, inpatient length of stay (LOS) during treatment were assessed. Results: Of the 2,731 patients with mPDAC included in the study, 101 (3.7%) were treated with a liposomal irinotecan based regimen, 1,316 (48.2%) were treated with gem+nab-P, 612 (22.4%) with FFX, and 624 (22.8%) with gem in any treatment line. The mean number of IP admissions was 1.2 for liposomal irinotecan treated patients, 1.5 for gem+nab-P, 1.5 for FFX, and 1.2 for gem. Among patients with at least one IP admission the mean LOS was 4.5 days for liposomal irinotecan, 5.4 days for gem+nab-P, 3.8 for FFX, and 5.1 for gem treated patients. Patients treated with liposomal irinotecan had a mean of 1.3 ER visits during treatment. Gem+nab-P, FFX, and gem-treated patients experienced 1.7, 1.4, and 1.8 mean ER visits, respectively. Mean mPDAC-related IP admissions ranged from 1.1 – 1.5, ER visits ranged from 1.1 – 1.7, and mean LOS ranged from 3.8 – 5.5 days. Conclusions: In this descriptive retrospective study patients receiving liposomal irinotecan, across all treatment episodes, generally experienced numerically lower mean IP admissions and ER visits. LOS was similar across all regimens. Further studies are necessary to characterize the IP and ER HRU burden among mPDAC patients treated with approved regimens.

e16751 Background: The NAPOLI-1 study, a randomized phase 3 study in pts with mPDAC previously treated with gemcitabine-based therapy, demonstrated an improvement in overall survival (OS) with liposomal irinotecan + 5-fluorouracil/ leucovorin (5-FU/LV) vs. 5-FU/LV. In this analysis we describe the trends in pt characteristics, real-world OS (rwOS), and real-world time to-treatment failure (rwTTF) among pts with mPDAC treated with liposomal Irinotecan based regimens overall and stratified by tx initiation prior to 2018 (pre-2018) or after 2018 (post-2018). Methods: This retrospective observational study used de-identified data from Flatiron Health EHR database from over 280 cancer clinics in the US. Pt characteristics, rwOS, and rwTTF were assessed in adult pts diagnosed with mPDAC who received liposomal irinotecan treatment (tx) between January 1, 2016 and October 31, 2019. Results: Of the 590 pts treated with liposomal irinotecan based regimens, 53% were male, 56% were initially diagnosed with Stage IV disease, 92% were treated in the community setting, and median age at tx initiation was 69 (IQR: 62 – 75) years. Among pts with available ECOG scores (N = 435), 77% had a score of 0 or 1. 43% (n = 254) initiated tx pre-2018 and 57% (n = 336) post-2018. Pre-2018, 106 (42%) pts initiated liposomal irinotecan in the third line metastatic setting or later (3rd line+), 125 (49%) had ECOG score of 0-1, and median age was 68 (62–74) years. Post-2018, 36% of pts initiated tx in 3rd line+, 211 (63%) had ECOG score of 0-1, and median age was 70 (63 – 75) years. Median rwOS was 4.4 months [95% CI: 4.3–6.2] pre-2018 and 5.2 mos [4.3–6.2] post-2018. rwTTF was 1.6 mos [1.4–1.9] pre-2018 and 2.1 mos [1.6–2.5] among pts post-2018. Among pts treated in first- or second-line, pre-2018 rwOS was 5.3 mos [3.9–6.4] and post-2018 rwOS was 6.3 mos [5.0– 7.6]. Conclusions: In this descriptive real-world study of pts with mPDAC receiving liposomal irinotecan based regimens, pts initiating treatment post-2018 appear to be less pre-treated, older, and have better performance status than pts pre-2018. Pts treated post-2018 experienced numerically longer rwTTF and rwOS than pts treated pre-2018.

666 Background: Pancreatic cancer is the third deadliest cancer in the US and mPC has a 2.9% 5-year survival. The analyses herein describe treatment patterns, trends in usage, and overall survival (OS) in mPC. Methods: Using the Flatiron Health EHR-derived database, data were extracted and analyzed for patients with mPC (pts) between Jan 1, 2014 and Jun 30, 2019. The database includes de-identified data from over 280 cancer clinics (~800 sites of care) representing more than 2.2 million U.S. cancer patients available for analysis, with 80% of pts from community centers and 20% from academic centers. Lines of therapy in the metastatic setting are derived from structured medication records. OS from metastatic diagnosis was reported using the Kaplan-Meier method. Results: 7,666 pts with mPC were identified. 5,687 (74.2%) received therapy in the metastatic setting. Pts who didn’t receive therapy in the metastatic setting were more likely to be older (p < 0.0001) and less likely to have been diagnosed initially with stage IV disease (p < 0.0001) than pts who were treated. The frequency of (1L) regimens were gemcitabine plus nab-paclitaxel (GnP) 46.8%, FOLFIRINOX (FFX) 24.1%, gemcitabine monotherapy 9.3%, and FOLFOX 3.8%. Gemcitabine monotherapy use was 12.9% in 2014 and 7.3% in 2018. GnP (31.4%), FFX (12.3%), FOLFOX (11.4%), and liposomal irinotecan (nal-IRI) + 5-FU/LV (10.2%) were the most frequent second line (2L) regimens. Between 2015 and 2018 nal-IRI based regimens increased from 6% to 17.6% in 2L. In the third line (3L) setting nal-IRI + 5FU/LV (19.3%), GnP (12.1%), FOLFOX (11.4%), and FFX (9.1%) were the most common treatments. Aggregate median OS (mOS) for treated pts was 8.1 mos (95% CI 7.8 – 8.4), and mOS for untreated pts was 2.8 mos (2.6 – 3.0), p < 0.0001. Conclusions: Survival for mPC is improving and practice patterns are changing. GnP is the most commonly used 1L regimen, followed increasingly by nal-IRI + 5-FU/LV in 2L and 3L. Further studies are necessary to understand the treatment gaps for pts with mPC. [Table: see text]

670 Background: Low hematologic counts are a common, costly side effect of chemotherapy. This observational study examines rates and associated cost to treat. Methods: Data on adverse events (AEs) were extracted from the clinical Flatiron Health database for mPC patients (pts) from 01/2014-01/2019. Anemia, neutropenia, and lymphopenia occurrence was assessed via diagnosis codes and structured lab data. Costs due to AEs were derived from a claims analysis of mPC pts from 2013-2017 in Medicare Limited Data Set claims. Mean adjusted incremental costs (ICs) were estimated by comparing 30-day costs of pts with and without AEs, with controls selected among pts without AEs with the same regimen and line of therapy, when at least 80 cases/controls were identified. Results: 4592 treated mPC pts were identified (median age at diagnosis: 68y, IQR: 61 – 75). 1138 pts were treated with FOLFIRINOX (FFX) in first-line (1L), 2295 pts with 1L gemcitabine plus nab-paclitaxel (gem-nab), 218 pts with second-line (2L) FOLFOX, 56 pts with 2L FOLFIRI, and 178 pts with 2L liposomal irinotecan (nal-IRI) based therapy. Observed rates of anemia and neutropenia are shown in the table below. Lymphopenia rates were similar across regimens and ICs were not statistically significant. ICs for patients with any grade anemia were $3864, $3818, $3536, $3978, and $2963 for FFX, gem-nab, FOLFOX, FOLFIRI, and nal-IRI treated pts, respectively. ICs for pts with any grade neutropenia were $2382 for FFX, $2440 for gem-nab, $2688 for FOLFOX, $3551 for FOLFIRI and $2307 for nal-IRI. Conclusions: Any grade anemia ICs ranged from $2963 [$1544, $4400] (nal-IRI) to $3978 [$2241, $5817] (FOLFIRI), and any grade neutropenia ICs ranged from $2307 [$703, $4313] (nal-IRI) to $3551 [$1227, $6039] (FOLFIRI). Pts treated with nal-IRI had similar any grade AE rates but lower ICs, which suggest lower severity of AEs. These results are consistent with Flatiron Health’s lower rates of grades 3+ neutropenia and anemia. [Table: see text]

667 Background: Published data suggests prior exposure to irinotecan infers a lower likelihood of benefit to liposomal irinotecan. This analysis seeks to expand this hypothesis by evaluating U.S. patterns of care to understand how prior irinotecan therapy impacts outcomes in mPC. Methods: Using the Flatiron Health database, data were extracted and analyzed for treated mPC patients (pts) in the 2L+ setting between Jan 1, 2014 and Jun 30,2019. Therapies of interest included: gemcitabine/ nab-paclitaxel (GnP), FOLFOX, FOLFIRI, FOLFIRINOX (FFX), and liposomal irinotecan/5-FU/LV (nal-IRI). The reference date for each treatment group was the date of treatment initiation. Prior irinotecan was defined as any irinotecan given in a prior regimen in mPC diagnosis. Cox proportional hazard (PH) methods were used to calculate mortality hazard ratios (HRs). HRs were adjusted to account for demographics and relevant covariates. Pts with prior exposure to irinotecan were used as the reference population for the Cox PH model (an HR < 1 represents worse survival for exposed pts relative to the unexposed). Results: N = 1,978 were included in this analysis. The median age at treatment initiation, and the proportion of pts previously treated with irinotecan are reported in table. Crude mortality was: GnP pts, HR 0.93 [95% CI: 0.77 – 1.11, adjusted HR, 0.94, 0.76 – 1.15]; nal-IRI pts, HR 0.81 [0.64 – 1.02, adjusted HR: 0.89, 0.67 – 1.19]; HR for FOLFOX was 0.55 [0.38 – 0.78, adjusted HR: 0.51, 0.33 – 0.79]. HRs are not reported for FFX and FOLFIRI due to the small numbers with prior irinotecan exposure. Conclusions: In mPC, prior irinotecan treatment may not preclude benefit from later treatment with nal-IRI or GnP as can be seen from the adjusted and unadjusted HRs. These findings are hypothesis-generating and need to be considered in the context of wide CI’s, retrospective nature and the limitations of such data. Further study is required to understand the less-favorable signal observed with FOLFOX and prior irinotecan.[Table: see text]

665 Background: The recommended starting dose for nal-IRI is 70mg/m2 (free base, equivalent to 80 mg/m2 salt-based dosing). This study evaluates the impact of nal-IRI dose reductions on clinical outcomes. Methods: Using the nationwide Flatiron Health electronic health record-derived database, de-identified data were extracted and analyzed for adult mPC pts treated with nal-IRI Jan 2014-Jan 2019 and who initiated treatment at approximately the recommended dose (RD), 70mg/m2 +/- 5mg. Initial dose was derived from structured medication records, prioritizing administrations. The cumulative dose (CD) of nal-IRI over the first six weeks of treatment, the presence of dose reductions (DR) – (a decrease ≥ 7mg/m2), overall survival (OS) from treatment initiation, and duration of treatment (DoT) were assessed. Results: 257 mPC pts treated with nal-IRI (median age: 68y, IQR: 61 - 73) were identified initiating therapy at approximately the RD. 26.5% (N = 68) of pts experienced a DR during treatment. Mean 6-week CD was 175.8 mg/m2 (SD: 77.9) among pts with no DR. For pts with DR, mean CD was 191.8 mg/m2 (53.2). Median DoT was 6.1 wks (IQR: 2.1 – 15.3). Pts that experienced a DR had a longer median DoT: 15.1 wks (7.1 – 23.0) vs 4.3. wks (2.1 – 12.1) for pts with no DR. Overall Median OS (mOS) was 4.2 months (95% CI: 3.7 – 5.4). mOS for DR pts was 7.2 mos (95% CI: 5.5 – 9.7) and 3.7 mos (3.0 – 4.1) for pts who did not experience a DR. Conclusions: This real-world analysis suggests that reducing the dose of subsequent administrations of nal-IRI during treatment is associated with pts remaining on therapy longer, experiencing a larger CD, and a with longer OS. Additional real-world prospective studies are necessary to characterize the impact of nal-IRI dosing on clinical outcomes.

e18365 Background: Real-world evidence is lacking regarding costs for FDA-approved/NCCN Category 1 treatments for patients with metastatic pancreatic cancer (m-PANC). We analyzed costs by service in the Medicare fee-for-service (FFS) population by chemotherapy regimen and line of therapy (LOT). Methods: Patients with m-PANC were identified using ICD-9/10 diagnosis codes in the 2013-2017 Medicare 100% Limited Data Set claims, which include all Medicare paid FFS claims, except professional services, for 45 million Medicare FFS beneficiaries. We studied mean monthly costs by service category, regimen, and LOT. Patients in our study had two or more claims with a pancreatic cancer (PANC) diagnosis more than 30 days apart and one or more claims with a secondary malignancy (metastasis) diagnosis on or after the first PANC diagnosis date. We defined index date as the earliest metastasis diagnosis date. We excluded patients with pre-index non-PANC malignancies and those without 6 months pre-index and 3 months (or until death, if earlier) post-index Medicare FFS enrollment. LOTs were assigned based on therapies used. LOTs ended the day before a new chemotherapy began, 28 days after the last chemotherapy (if no new chemotherapy), or upon death. We analyzed the FDA-approved/NCCN Category 1 treatments used most commonly in first line (1L): gemcitabine monotherapy, gemcitabine/nab-paclitaxel, and FOLFIRINOX; and in second or third line (2L, 3L): liposomal irinotecan. Results: Mean monthly Parts A and B (excluding professional) costs for 1L gemcitabine monotherapy were lower than gemcitabine/nab-paclitaxel or FOLFIRINOX ($5,267, $9,116, and $8,046, respectively). Part B drug costs other than chemotherapy were higher for FOLFIRINOX than gemcitabine/nab-paclitaxel or gemcitabine monotherapy ($3,881, $1,155, and $827, respectively). Inpatient services were similar across 1L regimens ($2,721-$3,303). Despite disease progression, mean monthly 2L and 3L costs for liposomal irinotecan were $10,809 and $12,225, respectively. Part B drugs other than chemotherapy ($2,133-$2,509) were comparable to 1L regimens, but inpatient services ($2,306-$2,405) were lower. Conclusions: The mean monthly cost increased by LOT for m-PANC FDA-approved/NCCN category 1 regimens. Interestingly, Part A inpatient costs decreased in 2L and 3L, while Part B drug costs other than chemotherapy were comparable.

e19509 Background: Lenalidomide maintenance therapy after autologous stem cell transplant (ASCT) in the first-line treatment has been shown to improve progression-free survival (PFS) and overall survival (OS) in multiple myeloma (MM) patients (pts). This study assessed the impact of FDA approval of lenalidomide maintenance therapy on total healthcare costs of a US health plan. Methods: A model was developed to estimate the incremental (additional) total plan costs (in 2016 USD) for maintenance therapy in each year for the first 3 years after lenalidomide monotherapy (R) maintenance therapy approval. The number of post-ASCT adult MM pts eligible for initiating maintenance therapy was estimated from published epidemiological data and an analysis of Connect MM Registry data. Clinical endpoints for R-maintenance, including time on treatment, PFS and OS, were obtained from a meta-analysis of published clinical trials (CALGB, IFM, and GIMEMA). The use of common off-label maintenance therapies was considered. Types of costs included in the model were drug, drug administration, adverse events (AE), AE monitoring, one-time progression and terminal care costs. Results: In a hypothetical health plan with 1 million members, the number of adult MM pts eligible to initiate post-ASCT maintenance therapy was estimated to be 28. Among them, 14.8 pts initiated R-maintenance in Year 1, 15.2 in Year 2, and 15.3 in Year 3, representing an incremental increase of 2.9%, 4.2% and 4.4% after R-maintenance therapy approval, respectively. After considering additional costs of maintenance, as well as potential offsets resulting from delayed progression the incremental total healthcare costs by year are listed in the table. Conclusions: Approval of lenalidomide monotherapy for maintenance after ASCT in the first-line treatment of MM has minimal impact on total plan costs, primarily due to the small incident population and the already common use of lenalidomide in post-ASCT maintenance. [Table: see text]