Post-index Period Research Articles

The real-world prevalence of esophagogastric varices, bleeding, emergency room visits, and hospitalization among patients with advanced hepatocellular carcinoma in the US.

519 Background: Upper gastrointestinal bleeding associated with esophagogastric varices (EGV) is a morbid and potentially deadly complication of advanced hepatocellular carcinoma (aHCC). However, the presence of EGV among the aHCC population receiving esophagogastroduodenoscopy (EGD) in the US is not well understood, nor are the predictors of EGV- or bleeding-related ER or hospitalization (the outcome) among those who are newly initiated on systemic treatment. Methods: A retrospective cohort analysis was conducted utilizing IQVIA’s PharMetrics Plus Health Plans Claims database (January 1, 2016, to July 31, 2021). Patients ≥18 years of age with ≥1 prescription for an aHCC systemic treatment were included; the date of the first prescription was the index date. At least 12 months of continuous enrollment pre-index and 6 months post-index (unless patients were deceased) were required. Patients with pre-index diagnosis codes for renal cell carcinoma, differentiated thyroid carcinoma, colorectal cancer, gastric cancer, non-small cell lung carcinoma, and liver transplant were excluded. Logistic regression was conducted to identify associations between baseline characteristics and the outcome. Results: A total of 904 patients with aHCC were included (mean age: 61.3 years; 75.3% male); 39.4% of patients died within 6 months following the index date. During the pre-index period, 688 (76.1%) patients had portal hypertension-related comorbidity, 327 (36.2%) had EGD, 122 (13.5%) had EGV, and 63 (7.0%) had bleeding. During the observational period, 458 (50.7%) patients received EGD, among whom 209 (45.6%) also had EGV. 141 (15.6%) patients had ≥1 outcome event during the post-index period. In the adjusted analysis, patients with pre-index bleeding and with EGV but no bleeding had 4.5- and 3.1-times higher odds of having post-index outcome, respectively, as compared with those lacking pre-index bleeding and EGV. Moreover, the pre-index presence of portal hypertension-related comorbidities was associated with 3.1 times greater odds of having a post-index outcome. Conclusions: EGV and bleeding events are common in patients with aHCC receiving systemic therapies. The EGV prevalence of 45.6% among those receiving EGD is consistent with prior similar studies outside of the US (Giannini EG, et al. Clin Gastroenterol and Hepatol. 2006; Iavarone M, et al. United European Gastroenterol J. 2016; Hsieh WY, et al. Sci Rep. 2017). The presence of bleeding, EGV, and portal hypertension-related comorbidities prior to treatment initiation were associated with the increased post-treatment risk of EGV- or bleeding-related ER or hospitalization in these patients. To our knowledge, this is the first study to assess and report the presence of EGV among those receiving EGD in a US aHCC population.

Impact of Vagus Nerve Stimulation for the Treatment of Drug-resistant Epilepsy on Patterns of Use and Cost of Health Care Services and Pharmacotherapy: Comparisons of the 24-Month Periods Before and After Implantation

PurposeThis study examines the impact of vagus nerve stimulation (VNS) as treatment for drug-resistant epilepsy (DRE) on the use and cost of health care services and pharmacotherapy. MethodsUsing a large US health care claims database, we identified all patients with DRE who underwent VNS between January 1, 2012 and December 31, 2019. VNS implantation date was designated as the index date, and patients had to be continuously enrolled for the 24-month period before this date (preindex period). Outcomes included all-cause and epilepsy-related hospitalization, emergency department (ED) visits, and health care costs; health care claims resulting in an epilepsy diagnosis and all claims for antiseizure medications were deemed epilepsy related. Preindex data, except care related to preoperative medical clearance for VNS, were used to estimate multivariate regression models predicting outcomes during the 24-month postindex period (follow-up period). Predicted outcomes during follow-up were then compared with observed values. As a sensitivity analysis, we also replicated all analyses among subgroups defined by comorbid depression. FindingsA total of 659 patients underwent VNS for DRE and met the selection criteria. For the composite outcome of all-cause hospitalizations and ED visits, observed values were 42% lower than expected during the 24-month follow-up period; for the composite outcome of epilepsy-related hospitalizations and ED visits, observed values were 49% lower (P < 0.001 for both). Observed mean total all-cause costs, inclusive of costs of the procedure, were not significantly different than expected costs by month 19 of follow-up; mean total epilepsy-related costs were comparable by month 18. Findings were similar in subgroups with and without depression, although nominally greater differences (observed – expected) were seen in those with comorbid depression. ImplicationsOur findings suggest that VNS is associated with decreased risk of hospitalization or ED visits (all cause and epilepsy related) during the 2-year period subsequent to implantation and may become cost-neutral within 2 years of implantation (vs continued medical management of DRE without VNS). Although expected outcomes were estimated based on the 24-month period before implantation, the degree to which they approximated what would have happened in the absence of VNS is unknowable. Further research is needed to better understand the extend and duration of the impact of VNS on seizure frequency and severity and health-related quality of life, including its performance among those with and without comorbid depression.

Clinical Outcomes, Resource Utilization, and Treatment Over the Disease Course of Symptomatic Obstructive Hypertrophic Cardiomyopathy in the United States

We sought to describe the clinical outcomes, resource utilization, and treatment options for patients with symptomatic obstructive hypertrophic cardiomyopathy (HCM) over the course of their disease. Adults with obstructive HCM who were symptomatic were identified from the IBM MarketScan Commercial and Medicare supplemental database (January 2009 to March 2019). The index date was the initial obstructive HCM diagnosis date. Patients were required to have ≥12-month continuous eligibility before and after the index date. Incidence rates (IRs) and cumulative risk of cardiovascular events, healthcare resource utilization, and pharmacotherapy were assessed after index and compared with matched controls. Among 4,617 eligible patients with obstructive HCM, 2,917 (63.2%, mean age 60, 47.2% women) were symptomatic at index date. The most common cardiovascular events were atrial fibrillation/flutter (IR:1.421 per person-year [PPY], heart failure (IR: 0.895 PPY), and dyspnea (IR:0.797 PPY). Patients incurred higher resource use with frequent tests and monitoring, hospitalizations (0.454 PPY), and emergency room visits (0.611 PPY). The use of pharmacotherapy increased from 61.2% in the 6-month preindex period to 83.9% in the 6-month postindex period and remained stable after diagnosis. These events ranged from 3 to over 60-fold higher compared with controls, with the largest difference observed in arrhythmic events. The majority of patients were symptomatic at the time of obstructive HCM diagnosis, resulting in significantly increased cardiovascular complications and frequent disease monitoring after diagnosis versus controls. In conclusion, healthcare resource utilization was substantial, and these findings suggest a higher clinical and economic burden over the disease course among patients with symptomatic obstructive HCM, despite current treatment.

Real-World Evaluation of Demographics, Treatment Pattern, and Economic Burden of Heart Failure and Kidney Disease in Type 2 Diabetes Mellitus Patient Population in Dubai, United Arab Emirates

Aims: The current study evaluated the demographics, clinical characteristics, treatment patterns, and economic burden of patients with type 2 diabetes mellitus (T2DM) with comorbidities (heart failure [HF], chronic kidney disease [CKD], and cardiovascular disease [CVD] without HF) in Dubai, United Arab Emirates (UAE). Methods: This observational, retrospective study collected data from January 01, 2014, to December 31, 2019, from the Dubai Real-World Claims Database (adults ≥18 years; at least 1 T2DM diagnosis claim). Patients were stratified into 5 cohorts: T2DM alone (cohort 1), T2DM and CKD (cohort 2), T2DM and CVD without CKD and HF (cohort 3), T2DM and HF (cohort 4), and T2DM with HF and CKD (cohort 5). An evaluation of demographics and clinical characteristics during pre-index period, as well as treatment patterns, healthcare resource utilization, and costs during the post-index period was conducted. Results: The sample had 374,271 patients with T2DM (age 43–56 years; male [72–84%]). Patients in cohorts 4 and 5 had Deyo-Charlson Comorbidity Index scores of 4.4 and 5.8, respectively. General practitioners (GPs) routinely prescribed biguanides for patients in cohorts 1–4 (24–38%), and insulin to patients in cohort 5 (27.7%). Prescription rates of novel antihyperglycemic drugs, such as glucagon-like peptide-1 (GLP-1 RA), were very low (∼2–8%) even in cohorts with cardiovascular and renal comorbidities (cohorts 2–5). A similar observation was noted with prescribing rates (0.6–4.4%) of sodium-glucose cotransporter-2 inhibitors (SGLT-2i) in cohorts 2–5. Endocrinologists preferred to prescribe GLP-1 RA and SGLT2i to T2DM patients with comorbidities. During the 5-year study period, median outpatient claims were the highest in cohort 5 (8.0 [range, 1.0–168.0]), followed by cohort 2 (5.5 [range, 1.0–52.0]). The median cost for inpatient claims was higher in cohort 5 (16,429 [range, 3,732–29,126] AED) compared to other cohorts. The median cost for drugs and procedures was highest in cohort 5 (4,525 [range, 38–31,546] AED and 2,297 [range, 56–105,074] AED, respectively). Conclusion: Continued and increased usage of drugs such as SGLT2i and GLP-1 RA with proven cardiorenal benefits could improve long-term outcomes and reduce associated healthcare costs in patients with T2DM and comorbidities in Dubai, UAE.

Retrospective Analysis of Healthcare Resource Use, Treatment Patterns, and Treatment-related Events in Patients with Huntington’s Disease–associated Chorea Initiated on Tetrabenazine

Background: Huntington’s disease (HD) is a multifaceted neurodegenerative disorder characterized by involuntary movements, specifically chorea, as well as behavioral and psychiatric disturbance, and cognitive dysfunction. Tetrabenazine was the first approved treatment for chorea, although tolerability concerns exist. Objectives: To characterize demographic and clinical characteristics of HD patients with chorea based on tetrabenazine use and examine treatment persistence with tetrabenazine in a real-world setting. Methods: Patients with a claim for HD-associated chorea (ICD-9-CM code 333.4) between 1/1/08 and 9/30/15 were selected from the MarketScan® Commercial and Medicare Supplemental databases. The first diagnosis date during the study period was considered the index date, with ≥6 months of continuous medical and prescription coverage before and after the index date. Treatment persistence was defined as the number of days from initiation to discontinuation or end of follow-up period. Discontinuation was defined as a gap in therapy of ≥60 days. Results: 1644 patients met selection criteria (mean age ± standard deviation: 54.5 ± 15.5), of which 151 (9.2%) were treated with tetrabenazine during the study period. The average (median) daily dose of tetrabenazine during the treatment period was 45.5 (42.3) mg/day. A total of 41.8% (59/141) of HD patients who initiated tetrabenazine experienced a ≥60-day gap in tetrabenazine therapy, with a median time to discontinuation of 293.5 days. During the 6-month post-index period after HD diagnosis, HD patients incurred higher all-cause healthcare costs ($20 204) vs the 6-month pre-index period ($6057), driven by higher hospitalization and pharmacy costs. Conclusions: A small percentage of HD patients with chorea were treated with tetrabenazine and discontinuation rates were high among those receiving treatment, with a median time to discontinuation of 9 months.

1460. Respiratory Syncytial Virus (RSV)-Related Clinical Events among a Medicare-Insured Population in the United States

Abstract Background RSV is a contagious pathogen that is often underrecognized in older adults. While the general burden of RSV has previously been assessed, little is known on the frequency and factors associated with RSV-related clinical events in older adults. Methods Patients ≥60 years old with a diagnosis code for RSV (index date) and ≥6 months of enrollment pre-index (baseline) were analyzed using claims data from the 100% Medicare database (2007-2019). The following RSV-related clinical events were assessed during the up-to-6-month post-index period: acute respiratory failure, chronic respiratory disease (asthma or chronic obstructive pulmonary disease), congestive heart failure, dyspnea, hypoxia, non-RSV lower/upper respiratory tract infection, and pneumonia. Patients were at risk of developing each clinical event if they did not already have the event at baseline. A stepwise Poisson regression model was used to identify baseline predictors of having ≥1 clinical event. Results A total of 175,392 patients were included (mean age: 79.0 years, 64.8% female, 78.4% white, 76.9% had ≥1 RSV-related clinical event at baseline). During the up-to-6-month period following RSV infection, 47.9% had ≥1 incident RSV-related clinical event (Figure). The mean (median) time to a clinical event was 1.0 (0.1) month. Having an event was more likely for patients with baseline conditions (coronary artery disease, diabetes, or one of the above RSV-related clinical events [except pneumonia and asthma]) or with chemotherapy, chest x-ray, organ transplant, anti-asthmatic use, or bronchodilator use at baseline (incidence rate ratio [IRR] range=1.06 [bronchodilators] to 1.80 [chest x-ray], all P&amp;lt; .05); having an event was less likely for patients with one of the following: lower age, female gender, baseline influenza or RSV test, baseline use of antibiotics, or baseline use of influenza agents (IRR range=0.63 [antibiotics] to 0.92 [baseline RSV test], all P&amp;lt; .05). FigureRSV-related clinical events among Medicare beneficiaries ≥60 years old Conclusion Almost half of patients ≥60 years old had an RSV-related clinical event within 1 month of RSV infection; patients with pre-existing conditions (≥75% of patients) were at higher risk of an event. These findings highlight that many older adults with RSV experience significant clinical events that burden both the patient and the healthcare system. Disclosures Jessica K. K. DeMartino, PhD, Janssen Scientific Affairs, LLC: Employee of Janssen Scientific Affairs, LLC Marie-Hélène Lafeuille, MSc, Janssen Scientific Affairs, LLC: Advisor/Consultant Bruno Emond, MSc, Janssen Scientific Affairs, LLC: Advisor/Consultant Carmine Rossi, PhD, Janssen Scientific Affairs, LLC: Advisor/Consultant Jingru Wang, BA, Janssen Scientific Affairs, LLC: Advisor/Consultant Stephanie Liu, MS, Janssen Scientific Affairs, LLC: Advisor/Consultant Patrick Lefebvre, MSc, Janssen Scientific Affairs, LLC: Advisor/Consultant Girishanthy Krishnarajah, MBA, Janssen Scientific Affairs, LLC: Employee of Janssen Scientific Affairs, LLC.

1255. Single-tablet Regimens (STR) Offer Better Persistence and Adherence, with Lower Costs by Adherence Status, than Multiple-tablet Regimens (MTR) for People Living with HIV (PLWH) Enrolled in Medicaid

Abstract Background Pill burden associated with antiretroviral multiple-tablet regimens (MTR) can impact adherence. The shift to single-tablet regimens (STR) has lagged for people living with HIV (PLWH) covered by Medicaid. This study examines persistence, adherence, healthcare resource utilization (HCRU), and costs by STR or MTR use for new initiators and treatment-experienced PLWH over a 1-year study period. Methods A linked patient population was applied using data from IQVIA’s Prescription Claims (Rx), Professional Fee Claims (Dx), and Hospital Charge Data Master (CDM). A 6-month pre-index period was used to assess study eligibility and baseline characteristics. A 12-month post-index period was used to descriptively evaluate treatment patterns and HCRU/costs. Two mutually exclusive cohorts were created based on STR or MTR use during the selection window (01/2018-07/2019). For the STR cohort, date of the first STR claim during the selection window was termed the index date. For the MTR cohort, the date of the first MTR drug during the selection window was termed the index date. Results The final sample comprised 4,603 PLWH in the STR cohort and 2,728 in the MTR cohort (Table 1). The proportion persistent over the 1-year follow-up was higher among treatment experienced compared to new initiators, and higher for STR compared to MTR (Figure 1A). The proportion adherent was higher among treatment experienced compared to new initiators, and higher for STR compared to MTR (Figure 1B). HIV-specific per member per month (PMPM) pharmacy costs were higher among treatment experienced compared to new initiators, and higher for MTR compared to STR (Figure 2). Adherent PLWH had a lower proportion with ≥ 1 all-cause emergency room visit compared to non-adherent PLWH within a cohort/treatment status category; minimal differences in ≥ 1 all-cause hospitalization. Adherent PLWH had higher mean all-cause costs than non-adherent PLWH, driven by pharmacy costs. STR PLWH had lower mean all-cause total costs compared to MTR PLWH with the same adherence/treatment experience status (Figure 3). Conclusion PLWH enrolled in Medicaid are more persistent and adherent to STR than MTR. Among PLWH adherent to antiretroviral therapy, STR offer potential cost savings over MTR for appropriate patients. Disclosures Andrew P. Brogan, PhD, ViiV Healthcare: Employee, Salary|ViiV Healthcare: Stocks/Bonds Cindy Garris, MS, ViiV Healthcare: Employee|ViiV Healthcare: Stocks/Bonds Julie Priest, MSPH, ViiV Healthcare: Employee, Salary|ViiV Healthcare: Stocks/Bonds Victoria Divino, BA, IQVIA: Employee, Salary|ViiV Healthcare: Grant/Research Support Jing He, PhD, IQVIA: Employee, Salary|ViiV Healthcare: Grant/Research Support Justin Chen, MHS, IQVIA: Employee, Salary|ViiV Healthcare: Grant/Research Support Mitch DeKoven, MHSA, IQVIA: Employee, Salary|ViiV Healthcare: Grant/Research Support.

Real-World Resource Utilization and Productivity Loss Among Patients With Myasthenia Gravis in Sweden: A Nationwide Population-Based Study

Objective To assess annual healthcare resource utilization including inpatient admission and outpatient visits, employment status, and sickness absence associated with myasthenia gravis (MG). Background MG is a rare, chronic and debilitating autoimmune neuromuscular disease characterized by muscle weakness and fatigue that leads to hallmark symptoms including ptosis, dysphagia, dyspnea and limb weakness. Nearly 10% of patients are estimated to have treatment-refractory MG. Design/Methods Data were linked from four longitudinal nationwide population-based registries in Sweden. Patients with = 1 diagnosis of MG (ICD-10 G70.0) from 01/01/2001 to 12/30/2017 were selected. Date of 1st MG diagnosis in the national patient register was designated as index date. The healthcare resource use, employment status, and sickness absence for all cause and associated with MG within 1-year post-index period were evaluated. Results A total of 4,339 patients with newly diagnosed MG were identified from 2001 and 2017. Mean (±SD) age at index date was 59.8 (±19.5) years; 54% were female. During the first year post-MG diagnosis, 50.6% of patients) had = 1 MG-related inpatient admission and 23.6% spent &gt;1 month as an inpatient. Most patients (89.3%, n = 3,875) had = 1 specialist visits for MG and 16.1% had &gt;5 visits during 1-year post-index period. 58.9% of patients had = 1 all-cause inpatient admission and 97.5% of patients used = 1 outpatient specialist services in the same period. Among patients of working age with =1-year follow-up (n = 2,006), 37.3% of them were not employed; among those in employment (n = 1,250), 44.6% reported = 1 sickness absences within 1-year post-index period. Conclusions Patients with MG require considerable care both for MG and comorbidities over a period of years. An important sub-population of patients (e.g., those with MG crisis) may be the intensive users of both inpatient and outpatient care. Future research needs to detail treatment pattern and outcomes in this population.

IMPROVING DEMENTIA CARE THROUGH DIGITAL CAREGIVER EMPOWERMENT: DURABLE CLAIMS-BASED OUTCOMES DEMONSTRATED IN A PRAGMATIC REAL-WORLD TRIAL.

The Ceresti Digital Caregiver Empowerment Program (DCEP) was deployed to 164 family caregivers of Medicare Advantage patients with ADOD. The primary objective of this pragmatic trial was to demonstrate that increasing the knowledge, skills and confidence of family caregivers (e.g., spouses, adult children) via personalized education, proactive coaching and remote monitoring-delivered via an app on a dedicated tablet computer-reduces avoidable patient hospitalizations. Durability of claims-based outcomes was also studied. Caregiver engagement, caregiver satisfaction and caregiver mental health were also tracked METHODS: Patients were included in the claims analysis if their caregiver was enrolled in the DCEP for at least 45 days. Patient outcomes were compared to outcomes from a 3:1 propensity matched comparison group, post-index, from the DCEP start date to the end of claims. Comparison group members were matched to enrollees on demographics, pre-index utilization and costs, and other factors, including the patient's frailty index. The first 30-days post-index were deemed a "transition" period. The impact of the DCEP was determined by comparing the 6-month pre-index averages to outcomes from the 30-day post-index period using the "difference-in-differences" approach (Figure 1). The index date for program enrollees was their DCEP start date. For matched comparison group patients, the index date was the program start date of the enrollee to whom the comparison group member is matched. The DCEP was personalized for each caregiver/patient dyad using predictive models applied to claims data. Table 1 summarizes statistically significant (p-values < 0.05) monthly reductions in utilization and costs for patients enrolled > 45 days, and enrolled high utilizers enrolled > 45 days (PMPM = per member/patient per month). High utilizers are patients with at least one hospitalization or two ED visits in the prior 24 months. Results were durable through 9 months post-index (Figure 3). Empowering family caregivers of patients with ADOD in a personalized DCEP provides durable reductions in patient hospitalizations and medical costs.

Design, test, validate: development of a novel & robust analytic framework and econometric model using administrative claims data.

As health care expenditures continue to rise, policy makers and payers are assessing whether the cost of virtual care programs are justified by their outcomes. Ceresti has developed a novel, robust analytic approach and econometric model to assess the efficacy and overall value of virtual care programs. The methodology uses an administrative claims-based, analytic and statistical framework that accurately identifies and measures reduction in costs and services (including potential savings from avoiding other costly downstream events) attributable to deployed virtual care programs. The analytic framework and econometric model were developed for and validated in a pragmatic trial in a Medicare-aged population. A total of 164 subjects were enrolled in the trial sponsored by a leading Medicare Advantage plan and Ceresti Health. The pragmatic trial compared changes in patient outcomes between program enrollees (DCEP "Care Recipients") and a matched comparison group using a "difference-in-differences" approach. General Matching criteria included: Age Gender Geographic region Aligned index dates between Care Recipients and the comparison group During the 6 month pre-index period matching criteria included: Charlson Comorbidity Index number of condition-specific medical claims presence of a skilled nursing facility claim emergency department (ED) visits and costs inpatient hospital admissions and costs costs for nursing or assisted living, office visits and other outpatient services Costs and service utilization of Care Recipients and the matched comparator group were extracted from an administrative claims database. Total medical and inpatient costs were captured during the pre- and post-index periods for the active and comparator arms. The frequency of ED visits, inpatient admissions, and 30-day readmissions were calculated per patient per month. The change in costs and visit frequency that Care Recipients incurred > 30 days post-index to the end of the study was compared to the change experienced by the matched comparator controls. The analytic model detected statistically significant pre-, post-efficacy across health-related and economic outcomes. The analytic and statistical framework validated in this trial overcomes the challenge of attributing claims-based outcomes to virtual care interventions, and contribute to the methods available to accurately measure the value of virtual care programs in Medicare-aged populations.

Healthcare Resource Utilization and Associated Costs of German Patients with Diffuse Large B-Cell Lymphoma: A Retrospective Health Claims Data Analysis.

Diffuse large B-cell lymphoma (DLBCL) is the most common form of non-Hodgkin's lymphoma with increasing prevalence. Although the disease burden associated with DLBCL is high, only limited data on healthcare resource utilization (HCRU) and associated costs of German patients with DLBCL is available. Using a large claims database of the German statutory health insurance with 6.7 million enrollees, we identified patients who were newly diagnosed with DLBCL between 2011 and 2018 (index date). Treatment lines were identified based on a predefined set of medication. HCRU and related costs were collected for the entire post index period and per treatment line. A total of 2495 incident DLBCL patients were eligible for the analysis. The average follow-up time after index was 41.7months. During follow-up, 1991 patients started a first-line treatment, 868 a second-line treatment, and 354 a third-line treatment. Overall, patients spent on average (SD) 5.24 (6.17) days per month in hospital after index. While on anti-cancer treatment, this number increased to nine (10.9) in first-line, 8.7 (13.7) in second-line, and 9.4 (15.8) in third-line treatments. Overall costs per patient per month (PPPM) increased from €421 (875.70) before to €3695 (4652) after index. While on a treatment line, PPPM costs were €17,170 (10,246) in first-line, €13,362 (12,685) in second-line, and €12,112 (16,173) in third-line treatments. Time-unadjusted absolute costs sum up to €59,868 (43,331), €35,870 (37,387), and €28,832 (40,540) during first-line, second-line, and third-line treatments, respectively. The main cost drivers were hospitalizations (71% of total costs) and drug acquisition costs (18% of total costs). The financial burden of DLBCL in Germany is high, mainly due to hospitalization and drug costs. Therefore, there is a high medical need for new cost-effective therapeutic options that can lower the disease burden and remain financially viable to support the growing number of patients with this aggressive disease.

Comorbidities, Healthcare Utilization, and Costs Associated With Alopecia Totalis and Alopecia Universalis in the United States

Background: Alopecia areata (AA) is an autoimmune disease characterized by nonscarring hair loss. Extensive forms of AA include alopecia totalis (AT; complete scalp hair loss) or alopecia universalis (AU; complete scalp, face, and body hair loss). Limited information exists about the cost and healthcare burden of AT and AU. Methods: Using a large US administrative healthcare claims database, two mutually exclusive patient cohorts were identified: AA cohort: ≥1 diagnosis of AA; AT/AU cohort: ≥1 diagnosis of AT/AU between January 1 and December 31, 2017. Baseline characteristics measured at first AA diagnosis (index date) and all-cause healthcare utilization and costs identified in the 1-year post-index period were compared between AT/AU and non-AT/AU AA cohorts. Results: 14,340 patients with AA were identified, including 1,224 patients with AT or AU. Compared with patients with non-AT/AU AA (n=13,116), patients with AT/AU were older (mean age 43.1 vs 40.6 years, P&lt;0.0001) and more were female (68.1% vs 62.9%, P&lt;0.0001). More patients with AT/AU had baseline comorbidities (atopic disease [28.3% vs 24.1%], anemia [10.4% vs 7.4%], autoimmune disorders [9.6% vs 5.5%]; P≤0.001 for all). Post-index patients with AT/AU had higher per person per policy year healthcare resource utilization and costs (P≤0.001) and total adjusted annual mean costs (P&lt;0.05). Over 50% more patients with AT/AU received immunosuppressive agents than patients with non-AT/AU AA. Conclusion: Patients with AT/AU had higher rates of comorbidities and greater healthcare resource utilization medical costs than patients with non-AT/AU AA, suggesting that improved insight into the patterns of specific comorbid conditions is needed.

Belantamab Mafodotin (Belamaf) for Relapsed/Refractory Multiple Myeloma (RRMM): A Real-World Observational Study

Introduction: Belamaf, a first-in-class B-cell maturation antigen-targeting antibody-drug conjugate (BCMA ADC) approved for the treatment of RRMM, demonstrated rapid, deep, and durable responses as monotherapy in the DREAMM-2 study (NCT03525678). We analyzed electronic health records (EHRs) of US patients with RRMM to characterize those treated with belamaf, including the occurrence and management of ocular events, and the effectiveness of belamaf in the real-world (RW) setting. Methods: This retrospective, longitudinal, observational study used de-identified data sourced from the US EHR-derived Flatiron Health Database from 01/01/2011 to 12/31/2021. Eligible patients had a confirmed MM diagnosis, ≥1 record for belamaf administration (first administration=index date), were ≥18 years of age on the index date, had available chart abstraction data for ocular safety and effectiveness outcomes of interest, and did not participate in an interventional clinical trial between MM diagnosis date and index date. Patient characteristics and treatment history were described during the time from MM diagnosis to the index date. The post-index follow-up period spanned from index date to the earliest of start of participation in a clinical trial, end of clinical activity, end of data availability, or death. Prespecified terms for belamaf-related ocular adverse events of special interest (AESI; including keratopathy, blurred vision, dry eye, and keratitis), and physician-reported severity information, if available, were abstracted from patient charts. Additionally, frequency/timing of ophthalmic monitoring visits, persistence with belamaf treatment, reasons for belamaf discontinuation, and clinical effectiveness of belamaf (i.e., overall response rate [ORR], overall survival [OS], and progression-free survival [PFS]) were assessed during the post-index period. Kaplan-Meier methods were used for time-to-event analyses. Ocular AESI information for patients with ≥4 months of follow-up is presented separately, as follow-up time was too short to evaluate the incidence of ocular AESI and associated treatment mitigations for patients receiving their first dose of belamaf close to the data cutoff. Results: Of the 137 patients included in the study, mean age was 67.9 years, 63.5% were White, and 50.4% were female (Table 1). During a median (interquartile range [IQR]) time between MM diagnosis and index date of 4.8 (3.0-6.7) years, 64.2% had received at least 5 lines of therapy prior to belamaf initiation, and 76.6% were triple-class refractory. The median (IQR) post-index period was 3.1 (1.4-6.0) months. Among patients with ≥4 months of follow-up (57 patients), 71.9% had at least 1 ocular AESI, with keratopathy and blurred vision being the most frequent types (Table 2). Mean (SD) time to first ocular AESI was 44.7 (40.2) days. Most ocular AESIs were managed by therapy hold (58.5%). Among those with a keratopathy event and severity information available (n=25), severity was split between mild (48.0%) and moderate/severe (52.0%) keratopathy, and the predominant action taken was a therapy hold. ORR was 30.2% at 6 months and median PFS was 5.4 months. At 3- and 6-months after starting therapy, 48.1% and 31.3% of patients, respectively, were still on treatment with belamaf (Table 2). Among the 76 (55.5%) patients who eventually discontinued belamaf, reason for discontinuation was reported for 71 patients; the most common reasons were disease progression (33.8%), treatment toxicity (19.7%), and a combination of both (15.5%). Conclusions: These findings provide important insights into the RW use of belamaf in patients with RRMM. Compared with DREAMM-2, patients in this study were slightly older, had fewer prior lines of therapy, and had a shorter time between MM diagnosis and belamaf treatment initiation. As in DREAMM-2, keratopathy was one of the most common ocular AESIs reported in patients receiving belamaf treatment. ORR in this RW study was also consistent with the DREAMM-2 study (30.2% vs 31%).1 Overall, patients with RRMM treated with belamaf in the RW setting experienced similar outcomes, including therapy holds to address ocular AESI, consistent with the DREAMM-2 study.

Retrospective Assessment of Total Cost of Care in Non-Dialysis Dependent Chronic Kidney Disease Patients Receiving Parenteral Iron Therapy for Iron Deficiency Anemia

Introduction Iron deficiency anemia (IDA) is a common comorbidity among patients with chronic kidney disease. Iron deficiency occurs in 15% to 48% of non-dialysis dependent chronic kidney disease (ND-CKD) patients with anemia. Parenteral iron therapy is often required in ND-CKD patients with IDA who cannot tolerate or are nonresponsive to oral iron. Available IV iron treatments vary in the amount of iron dose per administration, the number of infusions required to achieve complete iron repletion and drug acquisition cost. The objective of this study was to compare healthcare costs between ferric carboxymaltose (FCM) and low dose IV iron (LDI) for the treatment of IDA in ND-CKD patients. Methods Medical and pharmacy claims from IQVIA PharMetrics® Plus data were analyzed. Adult patients who received first (index) dose of FCM or LDI (i.e., iron sucrose, iron dextran, sodium ferric gluconate complex in sucrose) from 2017 to 2019 were included. Eligible patients were required to be continuously eligible 6 months before (baseline) and 12 months after index IV iron infusion and had medical claims associated with diagnosis of chronic kidney disease (ICD-10: N18.1 to N18.5) and IDA (ICD-10: D50.x) anytime during the 18-month study observation period. In addition, eligible FCM patients were required to receive a second dose of FCM within 21 days of index date. Patients who had claims associated with end stage renal disease (ICD-10: N18.5 and N18.6) or dialysis (HCPCS CPT code: 90935-90999) and patients who had received other IV iron products in the 6-month baseline period were excluded. Unadjusted mean number of IV iron infusions, outpatient visits and hospital admissions in the 6 months before and 12 months following index IV iron dose were summarized on a per-patient-per-month basis. Post-index monthly inpatient, outpatient, and total healthcare costs for FCM and LDI were compared using generalized linear model with gamma log-link adjusting for pre-index monthly cost, age, gender, year of index treatment and the Charlson Comorbidity Index (CCI). Sensitivity analyses comparing FCM versus iron sucrose were performed. Results Data from 3,727 FCM patients [mean (SD) age= 67.9 (13.8) years, 61% females, CCI (SD)=1.9 (2.5)] and 5,068 LDI patients [mean (SD) age=64.8 (15.0) years, 60% females, CCI (SD)=1.6 (2.4)] were analyzed. Within the LDI cohort, 3,489 patients received index iron sucrose treatment [mean (SD) age=64.6 (15.1) years, 60% females, CCI (SD)=1.5(2.4)]. Mean (SD) number of IV iron infusions was 0.24 (0.26) for FCM, 0.43 (0.52) for LDI and 0.46 (0.52) for iron sucrose during the post-index period. During the post-index period, FCM has a smaller increase in the number of outpatient visits (mean change from baseline=0.13) than LDI (mean change from baseline=0.64) and iron sucrose (mean change from baseline=0.70). The number of hospital admissions in the post-index period were numerically lower for FCM (mean change from baseline=-0.05) but numerically higher for LDI (mean change from baseline=0.04) and iron sucrose (mean change from baseline=0.05). Mean total health care cost increased during the 12-month period after IV iron treatment compared to baseline in all treatment groups. After adjusting for pre-index cost and other covariates, FCM treatment was associated with significantly lower post-index inpatient costs [adjusted cost ratio (ACR)= 0.73, p<0.0001], ER costs (ACR=0.76, p<0.0001), outpatient costs (ACR=0.73, p<0.0001) and total costs (including IV iron) (ACR = 0.81, p<0.0001) than LDI. Similar results were observed in the comparisons between FCM and iron sucrose. Conclusion Higher drug acquisition cost of FCM compared to LDI was primarily offset by lower inpatient and outpatient costs during the 12 months after FCM treatment for IDA in ND-CKD patients than LDI, resulting in significantly lower overall total health care cost for FCM compared to LDI. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Economic Burden of Hematopoietic Cell Transplantation (HCT) Among Commercially Insured Patients with Hematological Malignancies in the United States (US)

INTRODUCTION: HCT is a potentially curative treatment option for high-risk hematologic malignancies. Approximately 20,000 HCTs are performed in the US annually. A wide range of costs for both autologous HCT (auto-HCT) and allogeneic HCT (allo-HCT) are reported. For commercially insured patients with hematologic malignancies, the economic burden of an auto-HCT was estimated at $390,159 and allo-HCT at $745,341 in 2011-2014 over a 6-month pre- and 12-month post-HCT period (Bonafede 2017), indirect and caregiver costs may further increase burden (Applebaum 2017; Denzen 2017). We characterize all-cause healthcare resource utilization (HRU) and costs among commercially insured patients with hematologic malignancies who received an auto- or allo-HCT in a more contemporary era. METHODS: Adults with hematologic malignancies and ≥1 inpatient admission for an auto- or allo-HCT (Jan 2015-Mar 2020) were identified from the MarketScan Commercial and Medicare Supplemental Databases. Patients were indexed on the date of first HCT inpatient admission. Included patients had 12 months of continuous enrollment before and after the index date and ≥2 medical claims with a diagnosis code for an eligible hematologic malignancy (acute myeloid leukemia, myelodysplastic syndrome, myelofibrosis, myeloproliferative disease, acute lymphoblastic leukemia, chronic myeloid leukemia, multiple myeloma, Hodgkin disease, or lymphoma) with ≥1 claim occurring on or before the index date. Patients with an HCT prior to the index date were excluded. Controls were selected from patients meeting study criteria with the exception of an HCT claim requirement. The index date for patients in the control group was randomly assigned based on the time between diagnosis and HCT for the HCT cohorts. Controls and HCT recipients were matched 3:1 based on age, sex, insurance type, Deyo-Charlson Comorbidity Index categories, and hematologic malignancy. Total all-cause HRU and inpatient, outpatient (ie, ED [emergency department], physician office, other visits), and pharmacy costs were compared between the HCT cohorts and controls over the 6-month pre- and 12-month post-index periods. Healthcare costs (2021 US$) were based on paid amounts of adjudicated claims, including insurer/health plan payments and patient cost-sharing (copayment, deductible, and coinsurance). RESULTS: A total of 1653 auto-HCT recipients were matched to 4959 controls and 411 allo-HCT recipients matched to 1233 controls. After matching, HCT cohorts and controls were similar in age, sex, insurance type (Table 1), Deyo-Charlson Comorbidity Index categories, and hematologic malignancy. Pre-index comorbidities differed among the HCT and control cohorts, with psychiatric disturbances, infections, and heart valve disease significantly more common in both the auto- and allo-HCT recipients vs matched controls (P<0.001 for all). Multiple myeloma was the most common eligible indication for auto-HCT whereas acute myeloid leukemia/myelodysplastic syndrome/myelofibrosis/myeloproliferative disease were the most common indications for allo-HCT. Over the 18-month observation period, total all-cause costs were significantly higher for auto-HCT recipients vs matched controls ($540,194 vs $182,846; P<0.001; Figure 1), with auto-HCT recipients having significantly higher costs than controls for all categories evaluated except ED visits during the post-index period ($817 vs $879; P<0.4). Auto-HCT costs were driven by inpatient and outpatient costs. For allo-HCT recipients vs matched controls, total all-cause costs were significantly higher ($929,799 vs $190,776; P<0.001) during the 18-month observation period, with allo-HCT recipients having significantly higher costs than controls for all categories evaluated except ED visits during the pre- and post-index periods ($568 vs $533; P=0.11, $824 vs $862; P>0.9). Inpatient costs were the primary cost driver for allo-HCT recipients. CONCLUSIONS: The use of HCT, a potentially curative treatment for some patients with hematologic malignancies, is associated with considerable HRU and economic burden. This analysis did not include indirect costs, caregiver burden, or donor costs, and therefore likely underestimates the true burden of HCT for patients with hematologic malignancies. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Healthcare Resource Utilization and Associated Costs of German Patients with Diffuse Large B-Cell Lymphoma (DLBCL) - a Retrospective Health Claims Data Analysis

Diffuse large B-cell lymphoma (DLBCL) is the most common form of non-Hodgkin lymphoma with increasing prevalence. Although the disease burden associated with DLBCL is high, only limited data on healthcare resource utilization (HCRU) and associated costs of German patients with DLBCL is available. Using a large claims database of the German statutory health insurance with 6.7 million enrollees, we identified patients who were newly diagnosed with DLBCL (index date) and had no other cancer co-morbidity. Identification period was between 2011 and 2018 and post index period had to be at least one year except for death. Treatment lines were identified based on a predefined set of medications categorized by established DLBCL treatment recommendations. HCRU and related costs were collected for the entire post index period and per treatment line. 2,495 incident DLBCL patients were eligible for the study. The average follow up time after index was 41.7 months. During follow up 1,991 patients started a 1st line treatment, 868 a 2nd line treatment, and 354 a 3rd line treatment. Overall, patients spent on average 5.24 days per month in hospital after index. While on anti-cancer treatment this number increase to 9 in 1st line, 8.7 in 2nd line, and 9.4 in 3rd line treatment. Overall costs per patient per month (PPPM) were €3,695 after index. While on a treatment line, PPPM costs were €17,170 in 1st line, €13,362 in 2nd line, and €12,112 in 3rd line. Absolute costs sum up to is €59,868, €35,870, and €28,832 during 1st line, 2nd line, and 3rd line, respectively. Main cost drivers were hospitalizations (71% of total costs) and drug acquisition costs (18% of total costs). The financial burden of DLBCL in Germany is high and there is a high medical need for new cost effective therapeutic options that can lower the disease burden and remain financially viable to support the growing number of patients with this aggressive disease.

MRD Assessment Using Next-Generation Sequencing (clonoSEQ® Assay) Among Fee-for-Service Medicare Beneficiaries with Newly Diagnosed Multiple Myeloma Who Received Transplant

Introduction The assessment of minimal residual disease (MRD) for multiple myeloma (MM) has been increasingly incorporated into both clinical trials and patient management. MRD assessment, using an assay with a minimum sensitivity of 10-5 or better, is now recommended within NCCN guidelines following the completion of treatment phases. The clonoSEQ Assay (Adaptive Biotechnologies; Seattle, WA) is currently the only FDA cleared test available for bone-marrow MRD assessment in MM patients, and its use in routine clinical practice across the US facilitates an important new ability to evaluate MRD patterns and outcomes in the real-world setting. This study examined a cohort of Medicare MM patients who underwent transplant, and described characteristics of the treatments and MRD assessment these patients received. Claims-based analyses exploring MM treatments in the real-world have been published (Song 2016, Fonseca 2020), however to our knowledge this is the first study to match distinct claims datasets with corresponding MRD testing patterns and results. Methods The study population includes MM patients who had been assessed for MRD with clonoSEQ, who were successfully matched with 100% Medicare claims data. Newly diagnosed MM (NDMM) patients were identified as adult fee-for-service (FFS) beneficiaries with a diagnosis of MM between January 2017 to December 2021 (the first observed MM diagnosis is Index Diagnosis), with continuous coverage in Medicare FFS Parts A/B/D of at least 6 months during pre-index and 12 months during post-index period. Patients were further required to have initiated MM treatment within 60 days of Index Diagnosis, received a transplant, had at least one MRD assessment with clonoSEQ, and had no enrollment in clinical trials nor prior MM treatment pre-index. Phases of treatment within frontline therapy and line of therapy (LOT) were identified following established rules from the literature. Charlson Comorbidity Index (CCI) was calculated during the 6 months pre-index. We describe patient characteristics, treatment journey, and assessment of MRD test results at a sensitivity of 10-5 in accordance with IMWG guidance. Results The study identified 102 NDMM transplant patients meeting the above criteria, among which 99 received transplant as part of frontline therapy. The mean age of the cohort was 69 years, 53% were male, and 89% were white. The mean CCI was 0.97. One third of patients were diagnosed between 2017-2018. Both the mean and median follow-up time for this cohort are 32 months. Among the 102 patients, approximately 20% received consolidation and 50% received maintenance, as part of their frontline regimen; 44% received second or later LOT. The most common induction therapies used in this cohort were a PI+IMiD+Corticosteroid triplet (53%) or a PI+IMiD+CD38+Corticosteroid quadruplet (16%). On average, transplant patients started treatment 20 days from Index Diagnosis and received transplant 249 days from Index Diagnosis. A total of 171 MRD assessment were performed among this population with an average of 1.7 assessments and a maximum of 4 assessments per patient. Approximately 45% of transplant patients had >=2 MRD assessments and the median interval between adjacent MRD assessment is 196 days. Most of the patients (80%) were being assessed for MRD during frontline therapy, while 29% also received MRD assessment during later lines. Of the 99 patients who received transplant during frontline therapy, 66 (67%) patients had an MRD assessment within 6 months of transplant. Among them, 28 (42%) were MRD negative. Conclusions This is the first study to map treatment phases and line of therapy with Medicare claims data and examine the characteristics of MRD assessment in NDMM transplant patients. The study provides insight into real-world MRD testing patterns and outcomes within the Medicare population using next-generation sequencing. Further applications of this methodology to evaluate therapeutic response and understand MRD-supported decision making in the real-world setting are planned. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Hospital Costs and Healthcare Resource Utilization (HRU) for Chimeric Antigen (CAR) T-Cell Therapy and Stem Cell Transplant (SCT) in Patients with Large B-Cell Lymphoma (LBCL) in the United States (US)

Introduction: The efficacy of CAR T-cell therapy is well-established. However, the comparative hospital system costs and HRU for CAR T-cell therapy and SCT are largely unknown or based on rough estimates of cost-charge ratios. This study describes and compares the actuarial hospital costs and HRU of CAR T-cell therapy, administered in authorized treatment centers, and SCT for the procedure, pre-procedure, and follow-up periods. Methods: Adult patients in the PINC AI™ Healthcare Database were included if they had an inpatient or outpatient visit between 01/01/ 2017-04/30/2021 with CAR T-cell infusion or SCT transfusion for treating LBCL (principal or secondary discharge ICD-10-CM diagnosis codes: C83.3x, C85.1x, C85.2x, C85.8x, C83.8x, and C83.9x). Patients were excluded if they 1) had ≥2 CAR T-cell and/or SCT infusions, or 2) the treating hospital did not have continuous data for the patients during the 365-day pre-procedure and 180-day follow-up periods. The inpatient or outpatient infusion visit was considered the index event. Based on the index procedure, patients were classified as being in a CAR T-cell, autologous SCT (auto-SCT), or allogenic (allo-SCT) group. Patient characteristics (pre-index through index) and treating hospital characteristics (index) were summarized. Heath resource utilization (HRU) and costs (2021 US dollars) for the index, pre-procedure, and follow-up periods were assessed and reported by procedure type. Results: A total of 852 patients (208 CAR T; 595 auto-SCT; 49 allo-SCT) were treated at 37 hospital systems across the US. The mean patient age was 60 years (SD=12; median=62; 39.6% ≥ 65 yrs). 38.6% of patients were female, and 76.8% were white. Healthcare coverage varied (commercial: 43.9%; Medicare:36.3%; Medicaid:11.6%; other: 8.2%). The most common comorbidities other than malignancy were diabetes (17.6%), chronic pulmonary disease (9.2%), and renal disease (6.6%). Key patient characteristics (age, sex, race, health coverage, comorbidity) were similar across procedure types (Table 1). There was no noticeable difference in treating hospitals' teaching status nor hospital size between CAR T-cell therapy and SCT cohorts. Most patients were treated at large (n=608; 71.4%), teaching (n=720; 84.5%) hospitals; 93.4% (n=796) received their procedures inpatient, with slightly more auto-SCT in the inpatient than CAR T-cell infusions (94.8% vs 89.9%; p=0.01). Total mean cost of index procedure was higher for CAR T-cell therapy than for SCT ($371,136 CAR T, $96,515 auto-SCT, $169,269 allo-SCT; p<0.001). However, the mean index non-pharmacy cost was lower for CAR T-cell therapy than SCT (mean $41,375 CAR T, $51,778 auto-SCT, $111,594 allo-SCT; p<0.001). Among inpatient-treated patients, the average length of stay (LOS) was shorter for CAR T-cell therapy than for auto-SCT or allo-SCT (mean [median]: 18[15] days CAR T; 21[20] days auto-SCT; 28[26] days allo-SCT; p<0.001). Total ICU cost was lower for CAR T-cell therapy than for allo-SCT (mean $86,755 CAR T, $86,497 auto-SCT, $191,980 allo-SCT; p<0.001). Details of the index procedure hospital costs and HRU outcomes for CAR T, auto-SCT, and allo-SCT are in Table 2. Assessment of other outcomes, including adverse events, costs and HRU during the pre- and post-index periods, and sensitivity analysis are ongoing. Conclusions: This study is the first to examine actuarial cost and HRU with most recent real-world use of CAR-T and SCT using the hospital system perspective. The total and pharmacy cost to hospital providers for the index procedure were higher for patients who received CAR T-cell therapy than for SCT patients. However, non-pharmacy cost incurred during index procedure was lower for CAR T-cell therapy. Shorter hospital stays for CAR T-cell therapy patients contributed to this cost differential. Even though CAR T-cell treated patients incurred higher upfront pharmacy costs to hospitals compared to SCT, there were non-pharmacy cost offsets and lower HRU burdens overall to hospital system. Results should be interpreted with caution because they may not reflect HRU and costs to hospitals not included in the PINC AI™ Healthcare Database. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Clinical and Economic Outcomes in Patients with Transfusion-Dependent β-Thalassemia and Patients with Sickle Cell Disease with Recurrent Vaso-Occlusive Crises Receiving Hematopoietic Stem Cell Transplants in the United States

Background: Transfusion-dependent β-thalassemia (TDT) and sickle cell disease (SCD) are inherited blood disorders caused by mutations in the β-globin gene, resulting in abnormal, reduced, or absent expression of hemoglobin. TDT is characterized by ineffective red blood cell production that leads to chronic anemia, need for regular red blood cell transfusions (RBCTs), and organ complications. SCD is characterized by sickle-shaped red blood cells that cause vaso-occlusive crises (VOCs), hemolysis, and end-organ damage. Hematopoietic stem cell transplant (HSCT) is currently the only curative option for patients with TDT and patients with SCD. There is limited research on the frequency and outcomes of HSCT in these two patient populations. This study describes the economic and clinical outcomes for a combined population of patients with TDT and patients with SCD with recurrent VOCs who received HSCTs in the United States. Methods: This retrospective cohort study used administrative claims from the MerativeTM Commercial, Medicare, and Multi-State Medicaid Databases to identify patients with ≥1 inpatient or ≥2 outpatient claims for β-thalassemia or SCD between March 1, 2010, and March 1, 2019. Eligible patients with TDT were required to have ≥8 discrete RBCTs in any 12-month period on or after the date of the earliest qualifying β-thalassemia claim. At least 3 days between service dates of RBCT claims were required to be considered discrete RBCTs. Patients with TDT were required to have 1 year of continuous enrollment after their first transfusion to be included. Eligible patients with SCD with recurrent VOCs were required to have ≥2 diagnosed VOCs per year in any 2 consecutive years after and including the first qualifying SCD claim. A VOC was defined as inpatient or outpatient claim of acute chest syndrome, priapism, SCD crisis, or splenic sequestration; at least 3 days between VOC claims were required to be considered discrete VOCs. Eligible patients with SCD with recurrent VOCs were required to have 24 months of continuous enrollment before the second VOC in the second consecutive year and 12 months of continuous enrollment after. Patients with TDT and patients with SCD who met the above criteria and received an HSCT were included in the analysis set. Patients receiving HSCT were identified utilizing relevant procedure codes. The index date was the date of transplant. Patients were followed from the index date to the earliest of end of enrollment, death, or end of study period (March 1, 2020). Demographics were assessed at the index date, and clinical complications, healthcare resource utilization, and costs were descriptively summarized in the post-index period. Costs were based on paid amounts of adjudicated claims and patient cost-sharing and included costs in the 10 days before the index date to account for pre-transplant healthcare services associated with the transplant. Results: Two hundred thirty-four and 3500 patients met the criteria for TDT and SCD with recurrent VOCs, respectively, for a total of 3734 patients. Of that total, 62 patients (1.66%) underwent a transplant and were therefore included in the analysis cohort. Mean patient age at the index date was 12.6 years, and 46.8% were female. A proportion of transplant patients experienced graft versus host disease (GvHD) (37/62; 59.7%), unspecified HSCT complications (33/62; 53.2%), or bone marrow transplant rejection or failure (6/62; 9.7%) post-transplant. Average length of stay for the index HSCT admission was 42 days; in the first year post-transplant, patients averaged 58.3 days in the hospital. Total costs in the first 100 days and 1 year post-transplant were $442,752 and $623,301, respectively. Inpatient costs accounted for 93.7% and 88.7% of costs at 100 days and 1 year, respectively. Conclusions: This retrospective cohort study suggests that only a small proportion of patients with TDT and patients with SCD with recurrent VOCs receive HSCTs. Many patients who undergo transplants experience clinical complications, including graft failure and GvHD, and substantial economic burden. Given the limited utilization and potential safety risks of HSCT, additional curative treatment approaches are needed for patients with TDT and patients with SCD with recurrent VOCs.

Burden of Hospitalization before and after a Disease Progression Following Triple-Class Exposure in Patients with Relapsed and Refractory Multiple Myeloma

Introduction: Multiple myeloma (MM) is a plasma cell malignancy with various clinical complications. Patients with relapsed and refractory multiple myeloma (RRMM) often suffer from disease progression and cycle through multiple treatments, including those from the same drug classes that they have failed before. Disease progression is usually associated with significant burden, including high health care resource utilization (HCRU) and costs. However, no current known literature quantifies HCRU and costs before and after disease progression in RRMM patients who have been exposed to multiple classes of therapies. This study aims to describe and compare hospitalization use and associated costs before and after disease progression in patients with RRMM who are triple-class exposed (TCE; defined as exposure to a proteosome inhibitor, an immunomodulatory agent, and an anti-CD38 antibody). Methods: This retrospective cohort study used the Optum Clinformatics® Data Mart database from 1/1/2012 to 12/31/2021. Adults with ≥2 diagnosis codes for MM (ICD-10-CM: C90.0x) ≥30 days apart and who received a subsequent line of therapy (LOT) after becoming TCE were included. The start of a subsequent LOT after TCE was used as a surrogate to define disease progression, and the initiation date of the latest of such LOTs was defined as the index date if multiple eligible LOTs were observed for the same patient. Patients were also required to have ≥6-month continuous enrollment (CE) in medical and pharmacy plans before and after index. Patient characteristics were described at index and during the 6-month baseline period. A 2-month pre-post analysis was performed to compare all-cause inpatient admissions, intensive care unit (ICU) stays, hospitalization length of stay, and direct inpatient costs. Categorical variables were compared using McNemar's test, and counts and costs were compared using paired generalized linear models. Subgroup analyses were performed in patients with commercial insurance or Medicare Advantage. Results: A total of 523 patients met the study criteria. The mean age was 70.2 years, 63.1% were White, 51.2% were female, 72.5% had Medicare Advantage, the mean Quan-Charlson comorbidity index was 4.2, and 34.8% had ≥4 prior LOTs within the CE period containing the index date. The proportion of patients who had an all-cause inpatient admission (20.1% vs 12.8%, P = 0.0008) and an ICU stay (6.5% vs 1.9%, P = 0.0004) were significantly higher during the 2-month post- versus pre-index period. During the 2-month post-index period, patients on average had significantly more all-cause inpatient admissions (0.29 vs 0.16, P <0.0001). The mean inpatient length of stay for the cohort was numerically higher during the 2-month post- versus pre-index period (1.67 vs 1.07 days, P = 0.0751). The mean per-month cost of all-cause hospitalization was significantly higher during the 2-month post- versus pre-index period ($3,590.81 vs $1,267.18, P = 0.0008). The majority of the all-cause hospitalization use and costs (>90%) were MM-related. Similar trends were observed among commercial and Medicare Advantage patients. Conclusions: This study found a pronounced increase in hospitalization and costs in the 2 months immediately following disease progression in RRMM patients who are TCE. The data may support economic assessment and value framework for innovative treatments with significant prolonged progression-free survival. In addition, there is a need to focus on multifaceted efforts to improve overall well-being and the total cost of care for patients with RRMM.