Vutrisiran Health Care Resource Utilization, Costs, Discontinuation, and Mortality: A Real-World Claims Analysis

Abstract

Introduction

Vutrisiran was approved by the US Food and Drug Administration for hereditary transthyretin amyloid polyneuropathy (ATTR-PN) in 2022 and for transthyretin amyloid cardiomyopathy in 2025. Real-world data on vutrisiran health care resource utilization (HCRU), costs, discontinuation, and mortality in the USA are limited.

Methods

We conducted a retrospective analysis of patients initiating vutrisiran from 2022 to 2024 (when only approved for hereditary ATTR-PN) using Optum’s de-identified Clinformatics^® Data Mart (CDM) Database. Patients were followed until death, disenrollment, or last date of available data. Hospitalizations, emergency department (ED) visits, and associated costs (excluding vutrisiran itself) were assessed. Discontinuation analyses were also performed in the Komodo Healthcare Map^® to serve as comparator.

Results

In Optum^® CDM (n = 111), mean (standard deviation [SD]) age was 70.9 years (9.9), 62% were male, 41% were Black, 27% were White, and 70% had heart failure. Mean (SD) follow-up was 1.5 years (0.6). After vutrisiran initiation, 41 (37%) patients had hospitalizations with a median of 9.3 days hospitalized annually and mean (SD) cost of $80,536 ($83,907) per patient; 55 (50%) patients had ED visits not leading to hospitalization, with a mean of 2.4 visits and mean (SD) ED costs of $6339 ($7484) annually per patient. Within 12 months of vutrisiran initiation, 13 (12%) patients died and an additional 13 (12%) discontinued therapy, of which 3 (23%) restarted vutrisiran and 10 (77%) switched ATTR therapy. Results were generally similar in the Komodo database comparator analyses (n = 718); within 12 months of vutrisiran initiation, 22 (3%) patients died and 142 (20%) discontinued therapy, of which 53 (37%) restarted vutrisiran, 40 (28%) switched ATTR therapy, and 47 (33%) discontinued all treatments.

Conclusion

Patients treated with vutrisiran continued to experience significant HCRU and costs related to disease progression despite treatment. New therapies are needed to further reduce burden and costs.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s12325-026-03664-7.

Key Summary Points

Why carry out this study?

Vutrisiran is approved for the treatment of transthyretin amyloid cardiomyopathy (ATTR-CM) (2025) and polyneuropathy (ATTR-PN) (2022), but real-world data on health care resource utilization (HCRU), cost, discontinuation, and mortality in the USA are limited.

The objective of this study was to report real-world vutrisiran HCRU, costs, discontinuation, and mortality in the USA, during a time period when vutrisiran was only approved for hereditary ATTR-PN and adequate follow-up data were available.

What was learned from the study?

Between 2022 and 2024, 50% of patients presented to the hospital and 37% were hospitalized within 1.5 years of initiating vutrisiran treatment, resulting in average annual patient costs exceeding $6000 and $80,000, respectively.

Within 12 months of initiating vutrisiran treatment, 24% of patients died or discontinued therapy, with many switching to other ATTR treatments.

These findings suggest that patients treated with vutrisiran continue to experience significant HCRU and costs related to disease progression despite treatment.

Introduction

Transthyretin (TTR) amyloidosis (ATTR) is a progressive, life-threatening condition caused by misfolded TTR protein aggregation into amyloid fibrils that deposit in various organs and tissues [1, 2]. ATTR presents as two primary phenotypes, transthyretin amyloid polyneuropathy (ATTR-PN), with predominant peripheral nervous system manifestations, and transthyretin amyloid cardiomyopathy (ATTR-CM), with predominant cardiac involvement, as well as a mixed phenotype that presents with a combination of these manifestations [1, 2]. While patients with wild-type ATTR predominantly have features consistent with cardiomyopathy and those with hereditary ATTR typically have neuropathic symptoms (referred to as ATTRv-PN) [3], recent studies suggest that there may be overlap in the initial manifestations, thus necessitating a multidisciplinary approach to ATTR diagnosis and patient management [4]. Median survival in patients with ATTRv-PN ranges from 14 to 25 years [5], while median survival in patients with ATTR-CM is 3–4 years after disease onset [2]. The survival rate in patients with a mixed phenotype was reported to be 75% at 30 months in the Transthyretin Amyloidosis Outcome Survey [6]. It should be noted that increased disease awareness and earlier diagnosis and treatment initiation have translated into improved survival probability over time [7].

ATTR is associated with a high cost of care. Within the USA, hospitalizations and related costs associated with ATTR-CM were found to exceed those of non-amyloid heart failure, and a survey conducted by the Amyloidosis Research Consortium found that 55% of patients experienced a high financial burden associated with the cost of ATTR care irrespective of phenotype [8, 9]. Additionally, substantial health care system burden associated with ATTRv-PN has been demonstrated in Europe and Asia [10–14]. Further, in patients with ATTRv-PN, treatment with a subcutaneous or intravenous ATTR therapy (e.g., inotersen, patisiran), which requires travel to a hospital or infusion center for administration, was shown to be more costly than the only approved oral treatment at that time, tafamidis [10].

Vutrisiran, a TTR-directed small interfering RNA, was approved in the USA for subcutaneous administration by a health care professional to treat hereditary ATTRv-PN in 2022, and wild-type or hereditary ATTR-CM in 2025 [15, 16]. The cost of vutrisiran is approximately $477,404 per year for four subcutaneous doses [17]. In the phase 3 HELIOS-A trial of vutrisiran in patients with hereditary ATTRv-PN, by month 18, 3 (2.5%) patients receiving vutrisiran discontinued treatment due to an adverse event (AE); two of these were due to death [18]. In the phase 3 HELIOS-B trial of vutrisiran in patients with ATTR-CM, by month 36, 10 (3%) patients receiving vutrisiran discontinued due to an AE [19].

Currently, there are no available real-world data on vutrisiran persistence, defined as the length of time from treatment initiation to discontinuation [20], though it is typical for persistence to be higher in clinical trials compared with real-world practice [21]. Because vutrisiran was only recently approved for ATTR-CM in 2025, conducting rigorous real-world analyses in this population is challenging due to limited patient follow-up and a small—though growing—sample size. However, there is substantial real-world experience with vutrisiran in ATTRv-PN. The objective of this study was to report real-world vutrisiran health care resource utilization (HCRU), costs, discontinuation, and mortality in the USA, during a time period when vutrisiran was approved for hereditary ATTRv-PN.

Methods

Data Sources and Study Design

This was a retrospective analysis of patients initiating vutrisiran from June 22, 2022, to April 30, 2024, from Optum’s de-identified Clinformatics^® Data Mart (CDM) Database (Fig. 1). Optum^® CDM is a large de-identified administrative medical and pharmacy claims database [22]. This nationwide database includes medical and pharmacy claims data on over 84 million members enrolled in commercial or Medicare Advantage health plans and captures all inpatient, outpatient, pharmacy, emergency department (ED), office visit, and other ancillary service utilization throughout a member’s period of enrollment. Patients were included if they were ≥ 18 years of age, had ≥ 1 vutrisiran claim, and had ≥ 1 year of continuous enrollment. Patients were followed from initiation of vutrisiran until death, disenrollment (≥ 1 year of continuous enrollment after initiation for patients who did not die), or last date of available data (March 31, 2025), whichever came first.

Fig. 1.

Patient flow. ^aPatients were required to have continuous enrollment from first vutrisiran instance to end of enrollment and at least 1 year of continuous enrollment in the database. CDM Clinformatics Data Mart

Discontinuation analyses were conducted first in Optum^® CDM, then replicated in the Komodo Healthcare Map^® to serve as a comparator (Fig. 1). The Komodo database comprised of medical and prescription claims data from several sources [23]. It is characterized by proprietary partnership with >150 key national payers and consortiums, representing over 150 million payer complete lives. The database is representative of the US commercial-, Medicare-, and Medicaid-insured populations, and provides insights into the complete patient journey over time. Cost analyses were not replicated, as the Komodo database only has imputed cost data. Patients were included if they had initiated vutrisiran and had a minimum of four consecutive quarters of continuous medication and medical service claims activity after the index date, or until death if death occurred sooner.

The Optum^® CDM and Komodo databases comply with the Health Insurance Portability and Accountability Act. Given the anonymized, de-identified nature of the data, patient consent was not possible or required, and institutional review board approval to conduct this study was not necessary. Only aggregate cohort-level results were reported, without identification of any health care providers or health care facilities.

Outcomes

Outcomes included the number of patients with inpatient hospitalizations and associated HCRU (not including events related to COVID-19) and costs, the number of patients with ED visits that did not lead to inpatient hospitalizations, and the number of patients with outpatient visits (not including those related to vutrisiran therapy); outpatient visits by setting were also assessed. Total per patient per year (PPPY) costs, with the exclusion of vutrisiran therapy, were captured regardless of the setting in which treatment was administered. Mortality was captured if death occurred within 12 months of treatment initiation. Outpatient care in a hospital facility was defined as acute care with no admission to an inpatient setting; ED visits were captured separately and were not included in the category of outpatient care in a hospital setting. COVID hospitalizations were defined as patients with a COVID diagnosis on or within 1 day of inpatient admission, or patients with a COVID diagnosis during an outpatient visit followed by an inpatient admission within 1 day.

Persistence was defined as a claim for vutrisiran that was ≤ 180 days from the previous claim (an interval that included 90 days of supply and a 90-day grace period). Discontinuation was defined as no claim for vutrisiran that was ≤ 180 days from the previous claim, excluding death or insurance change. Events occurring after discontinuation, including vutrisiran restart, switching to a different treatment, and any other changes in treatment, were also reported. Sensitivity analyses were conducted for patient discontinuations by evaluating different grace periods (110-day gap [90 + 20-day grace], 135-day gap [90 + 45-day grace]), and follow-up lengths (1 year, 1.5 years, 1.75 years, and 2 years).

Statistical Analysis

Demographic and all outcome variables were assessed descriptively. Differences in cardiovascular comorbidities between Optum^® CDM and the Komodo database were analyzed by standardized mean differences (SMD).

Results

Patients

Of the 211 patients identified in Optum^® CDM, 110 patients were excluded due to enrollment criteria (Fig. 1) and 111 patients were included in the analysis (Table 1 and Supplementary Table 1). Mean age was 70.9 years, 62% were male, 27% were White, and 41% were Black. Loop diuretics (57%), tafamidis (40%; the only approved oral TTR stabilizer at the time of the study), and beta-blockers (44%) were the most common medications used in the 12 months after index. Heart failure (70%), essential primary hypertension (67%), and cardiomyopathy disease classified elsewhere (62%) were the most common cardiovascular comorbidities in the 12 months after index. Patients in Optum^® CDM had higher rates of cardiovascular comorbidities compared with patients in the Komodo database, including cardiomyopathy disease classified elsewhere (62% vs 28%; SMD = 0.70), other pulmonary heart disease (25% vs 4%; SMD = 0.60), hypertensive heart disease (39% vs 13%; SMD = 0.59), and hypertensive heart disease and chronic kidney disease (33% vs 10%; SMD = 0.56) (Supplementary Table 2).

Table 1.

Baseline demographics and patient characteristics

Characteristic	Optum® CDM	Komodo
	(N = 111)	(N = 718)
Age, mean (SD), years	70.9 (9.9)	71.0 (11.0)
Age distribution, n (%), years
19–44	< 5 (3)	20 (3)
45–64	19 (17)	134 (19)
65–80	76 (69)	429 (60)
≥ 80	13 (12)	131 (18)
Unknown	0	4 (1)
Sex, n (%)
Male	69 (62)	441 (61)
Female	42 (38)	264 (37)
Unknown	0	13 (2)
Race, n (%)
Black	46 (41)	305 (43)
White	30 (27)	279 (39)
Asian	< 5 (2)	17 (2)
Unknown	33 (30)	117 (16)
Ethnicity, n (%)
Hispanic	7 (6)	50 (7)
Non-Hispanic	59 (53)	601 (84)
Unknown	45 (41)	67 (9)
Medication use within the past 12 months, n (%)a
Tafamidis	34 (40)	229 (35)
Loop diuretics	48 (57)	353 (54)
SGLT2 inhibitors	20 (24)	172 (27)
Beta-blockers	37 (44)	254 (39)
Mineralocorticoid receptor antagonists	28 (33)	189 (29)
CV comorbidities (≥ 25% in either database) within the past 12 months, n (%)
Heart failure	78 (70)	351 (49)
Essential primary hypertension	74 (67)	332 (46)
Cardiomyopathy disease classified elsewhere	69 (62)	198 (28)
Atrial fibrillation and flutter	45 (41)	190 (27)
Hypertensive heart disease	43 (39)	92 (13)
Cardiomyopathy	43 (39)	150 (21)
Chronic ischemic heart disease	39 (35)	132 (18)
Hypertensive heart disease and chronic kidney disease	37 (33)	73 (10)
Complications and ill-defined descriptions of heart disease	35 (32)	99 (14)
Other pulmonary heart diseases	28 (25)	28 (4)
Hypertensive chronic kidney disease	28 (25)	57 (8)

CDM Clinformatics^® Data Mart, CV cardiovascular, SD standard deviation, SGLT2 sodium-glucose cotransporter 2

^an = 85 for Optum^® CDM; n = 650 for Komodo

Of the 1594 patients identified in the Komodo database, 876 patients were excluded due to enrollment criteria (Fig. 1) and 718 patients were included in the analysis (Table 1). Mean age was 71.0 years, 61% were male, 39% were White, and 43% were Black. Loop diuretics (54%), tafamidis (35%), and beta-blockers (39%) were the most common medications used in the 12 months after index. Heart failure (49%), essential primary hypertension (46%), and cardiomyopathy disease classified elsewhere (28%) were most common cardiovascular comorbidities in the 12 months after index.

HCRU and Costs in Optum^® CDM

Overall mean (standard deviation [SD]) follow-up was 1.5 (0.6) years (Table 2). During the follow-up period, 41 (37%) patients experienced inpatient hospitalization (90 total interactions; 2.2 per patient). Median length of stay was 6 days per hospitalization and median number of days hospitalized annually was 9.3. Total PPPY costs for inpatient hospitalization (SD) were $80,536 ($83,907). A total of 55 (50%) patients had ED visits not requiring hospitalization, with 2.4 interactions per patient (132 interactions total) and a median of 1.3 days of ED visits annually. Total ED PPPY costs were $6339 ($7484). Together, all patients had a total of 9111 other health care visits over the follow-up period, inclusive of all non-inpatient acute non-ED places of service (82.1 mean interactions per patient). Total PPPY outpatient hospitalization costs (SD) were $32,724 ($50,875). Cumulative costs for inpatient hospitalization, ED visits, and outpatient hospitalization were $3,814,891, $455,831, and $5,556,981, which accounted for 38.8%, 4.6%, and 56.5% of total costs overall, respectively.

Table 2.

HCRU and costs by case setting in Optum^® CDM

Metric	Inpatient hospitalization (± ED visit)	ED visits not leading to hospitalization	Outpatient visits (excluding vutrisiran claims)
Patients with events, n (%)	41 (37)	55 (50)	111 (100)
Follow-up, mean (SD), years	1.45 (0.57)	1.50 (0.55)	1.52 (0.52)
Total interactions, n	90	132	9111
Mean interactions per patient	2.2	2.4	82.1
Length of hospital/ED visit, median (range), days	6 (1–24)	1 (1–1)	NA
Annual days hospitalized per patient, median (range)	9.3 (0.4–106.9)	1.3 (0.4–9.4)	NA
Total cost of all interactions in care setting, USD	$3,814,891	$455,831	$5,556,982
Total cost per interaction, USD
Median (range)	$35,885 ($5670–138,516)	$2968 ($16–14,669)	$206 ($0.01–282,880)
Mean (SD)	$42,388 ($27,767)	$3453 ($2594)	$610 ($3826)
Total cost PPPY, USD
Median (range)	$51,644 ($6730–396,532)	$3913 ($339–41,154)	$18,346 ($363–312,556)
Mean (SD)	$80,536 ($83,907)	$6339 ($7484)	$32,724 ($50,875)

CDM Clinformatics Data Mart, ED emergency department, HCRU health care resource utilization, PPPY per patient per year, SD standard deviation, USD United States dollars

The most influential drivers of total costs were outpatient care at hospital facilities (cumulative total, $2,852,541) and office visits (cumulative total, $1,102,085) (not including the cost of vutrisiran) (Table 3). Mean PPPY costs were $19,026 for outpatient care at hospital facilities and $6075 for office visits. A disproportionate cost burden was observed for end-stage renal disease treatment facilities, which accounted for only 10% of outpatient visits yet incurred the highest mean cost at $64,849 PPPY; ambulance services (4% of visits and outpatient care at hospital facilities (25% of visits) followed with mean costs of $22,142 PPPY and $19,026 PPPY, respectively.

Table 3.

Outpatient costs excluding vutrisiran claims by place of service in Optum^® CDM

Place of service	Patients, n (N = 111)	Visits, n (%) (n = 9111)	Total cost (all visits), USD	Mean cost per patient (all visits), USD	Mean cost per visit, USD	Mean cost PPPY, USD	Mean cost per visit PPPY, USD
Outpatient hospital facilitya	100	2278 (25)	$2,852,541	$28,525	$1252	$19,026	$15
Office	108	2836 (31)	$1,102,085	$10,204	$389	$6075	$17
Home	64	1720 (19)	$622,683	$9729	$362	$7232	$25
End-stage renal disease treatment facility	< 5	865 (10)	$424,119	$106,030	$490	$64,849	$141
Ambulance	< 5	316 (4)	$188,571	$47,143	$597	$22,142	$37
Ambulatory surgical center	20	61 (1)	$157,284	$7864	$2578	$5076	$2
Independent laboratory	91	628 (7)	$113,746	$1250	$181	$981	$5
Off campus—outpatient hospital facility	20	111 (1)	$19,065	$953	$172	$568	$3
Skilled nursing facility	< 5	92 (1)	$13,937	$3484	$151	$4364	$29
In-home telehealth	27	78 (< 1)	$12,305	$456	$158	$288	$2
Comprehensive outpatient rehabilitation facility	< 5	52 (< 1)	$12,099	$12,099	$233	$8509	$37
Other unlisted facility	15	70 (< 1)	$8684	$579	$124	$374	$3
Independent clinic	< 5	28 (< 1)	$7214	$3607	$258	$1604	$6
Out-of-home telehealth	20	48 (< 1)	$6962	$348	$145	$257	$2
Urgent care facility	15	22 (< 1)	$6359	$424	$289	$290	$1
Nursing facility	< 5	17 (< 1)	$3850	$1925	$226	$2620	$12
Residential substance abuse treatment facility	6	25 (< 1)	$3111	$519	$124	$308	$3
Mass immunization center	6	7 (< 1)	$1374	$229	$196	$163	$1
Mobile unit	< 5	< 5 (< 1)	$766	$383	$255	$288	$1
Federally qualified health center	< 5	< 5 (< 1)	$195	$195	$97	$170	$2

CDM clinformatics data mart, PPPY per patient per year, USD United States dollars

^aExcludes ED and inpatient hospital visits

The most common reasons for inpatient hospitalizations, ED visits not requiring hospitalization, and outpatient visits varied by care setting (Table 4). Inpatient hospitalizations were most frequently associated with genitourinary (renal and reproductive) conditions (14 patients; 21 events), endocrine/metabolic issues (12 patients; 13 events), and infectious/parasitic events (5 patients; 13 events). Most (12/14) patients hospitalized for genitourinary conditions were hospitalized for renal failure, including 1 patient with co-occurring infection. Two patients were hospitalized for genitourinary conditions other than renal failure, including one due to infection and another due to kidney stones. ED visits not leading to hospitalization were most commonly related to musculoskeletal conditions (18 patients; 33 events), non-specific symptoms (18 patients; 23 events), and genitourinary conditions (7 patients; 12 events). For outpatient visits, genitourinary (68 patients; 1887 events), endocrine/metabolic (109 patients; 1612 events), and circulatory conditions (95 patients; 1393 events) were the most frequent concerns.

Table 4.

Reasons for outpatient visits, ED visits, and hospitalizations by body system in Optum^® CDM

Patients, n (events)	Inpatient hospitalizations (± ED visit)	ED visits not leading to hospitalization	Outpatient visits (excluding vutrisiran claims)
Total	41 (90)	55 (132)	111 (9111)
Endocrine/metabolic	12 (13)	< 5 (2)	109 (1612)
Health status/encounters	< 5 (1)	6 (6)	96 (980)
Circulatory (cardiology)	9 (11)	5 (8)	95 (1393)
Symptoms/non-specific findings	< 5 (3)	18 (23)	79 (360)
Musculoskeletal	5 (6)	18 (33)	78 (885)
Genitourinary (renal and reproductive)	14 (21)	7 (12)	68 (1887)
Digestive (GI and liver)	7 (9)	7 (9)	49 (182)
Nervous system	< 5 (1)	0	49 (383)
Eye	0	0	44 (172)
Respiratory	< 5 (2)	6 (9)	44 (302)
Neoplasms (oncology)	< 5 (1)	< 5 (2)	41 (289)
Skin and subcutaneous	< 5 (4)	6 (6)	40 (228)
Neuropathy	0	< 5 (1)	34 (258)
Blood and immune (hematology)	< 5 (2)	< 5 (3)	30 (151)
Infectious and parasitic	5 (13)	< 5 (6)	15 (61)
Injury/poisoning	< 5 (1)	8 (9)	14 (157)
Mental health	0	< 5 (1)	13 (72)
Ear and mastoid	0	< 5 (1)	11 (28)
No diagnosis	< 5 (2)	0	11 (20)
External causes	0	0	< 5 (6)
Congenital	0	< 5 (1)	< 5 (5)

Data represent all patient activity except for the administration of vutrisiran. Patients may have events in multiple body system categories. Body systems are ordered by outpatient patient count (descending)

CDM Clinformatics Data Mart, ED emergency department, GI gastrointestinal

Vutrisiran Mortality, Persistence, and Discontinuation in the Optum^® CDM and Komodo Databases

Persistence at 1 year was 77% in Optum^® CDM; the same rate of persistence was observed in the Komodo database (77%), which served as the comparator (Table 5). Within the first year of treatment, 13 (12%) patients in Optum^® CDM and 22 (3%) patients in the Komodo database died (Table 5 and Fig. 2). In Optum^® CDM, 13 (12%) patients discontinued vutrisiran within the first year compared with 142 (20%) patients in the Komodo database. Among the 13 patients who discontinued vutrisiran in Optum^® CDM, no patients restarted vutrisiran plus tafamidis and 62% switched to tafamidis monotherapy. Among the 142 patients who discontinued vutrisiran in the Komodo database, 14% of patients restarted vutrisiran plus tafamidis and 15% switched to tafamidis monotherapy. In Optum^® CDM, no patients discontinued all treatment, while 33% of patients discontinued all treatment in the Komodo database; information for why patients discontinued treatment was not available.

Table 5.

Vutrisiran persistence and post-discontinuation treatment patterns in the Optum^® CDM and Komodo databases

Patient category	Optum® CDM	Komodo
	(N = 111)	(N = 718)
Death within the first year of vutrisiran, n (%)	13 (12)	22 (3)
Vutrisiran persistence by quarter over 1 year, n (%)
Q1	111 (100)	718 (100)
Q2	98 (88)	633 (88)
Q3	93 (84)	614 (86)
Q4	85 (77)	554 (77)
Vutrisiran discontinuation by disposition, n (%)	n = 13	n = 142
Restarted vutrisiran	< 5 (23)	53 (37)
Restarted vutrisiran monotherapy	< 5 (15)	32 (23)
Restarted vutrisiran as a combination therapy	< 5 (8)	21 (15)
Vutrisiran + tafamidis	0	20 (14)
Vutrisiran + other	< 5 (8)a	1 (< 1)b
Switched therapies	10 (77)	40 (28)
Switched or dropped to tafamidis monotherapy	8 (62)	21 (15)
Switched to tafamidis combination therapy	< 5 (8)	5 (4)
Switched to other monotherapy or combination therapy	< 5 (8)	14 (10)
Diflunisal	0	5 (4)
Eplontersen	0	3 (2)
Patisiran	< 5 (8)	5 (4)
Diflunisal + patisiran	0	1 (< 1)
Discontinued all treatment	0	47 (33)
Insufficient data	0	2 (1)

CDM Clinformatics Data Mart, Q quarter

^aOne patient received vutrisiran + eplontersen

^bOne patient received vutrisiran + eplontersen + patisiran

Fig. 2.

Vutrisiran persistence, discontinuation, and post-discontinuation treatment patterns in the a Optum^® CDM and b Komodo databases. Persistence at 1 year was 77% across both databases; 3–12% of patients died and 12–20% discontinued vutrisiran within the first year of treatment. Totals may not add up to 100% due to rounding. CDM Clinformatics Data Mart

Discussion

In this retrospective analysis of patients treated with vutrisiran, during a time when it was approved for hereditary ATTRv-PN, 50% of patients presented to the ED and 37% were hospitalized within 1.5 years of initiating vutrisiran. Excluding the cost for vutrisiran itself and COVID-related costs, the total costs for all patients associated with inpatient hospitalization and ED visits not leading to hospitalization were $3,814,891 and $455,831, respectively. This high cost of care averaged $64,400 PPPY irrespective of treatment setting. Outpatient care at hospital facilities and office visits were the main cost drivers. Within 12 months of starting vutrisiran, 3% and 12% of patients died and an additional 12% and 20% of patients discontinued therapy across the Optum^® CDM and Komodo databases, respectively, with many switching to other ATTR treatments.

Results from this study reveal a notable burden of HCRU despite treatment with vutrisiran. In Optum^® CDM, 80% of patients were 65 years of age or older and 50% of patients had ED visits within 1.5 years of vutrisiran initiation. Data from the National Center for Health Statistics showed that 35% of individuals 65 years of age or older had ED visits between 2020 and 2022, and the rate increased with age [24]. This suggests that patients with ATTR in our study present to the ED 15% more often than individuals in the general population matched for age, reflecting disease-specific burden of ATTR even with vutrisiran treatment. Vutrisiran must be administered in a health care facility, which requires travel and may pose an additional burden to patients. Notably, outpatient visits at hospital facilities and office visits also accounted for a major proportion of total outpatient costs ($2,852,541 and $1,102,085, respectively), highlighting additional burden of chronic care for patients with ATTR outside the ED. The variation in reasons for health care encounters across settings reflects the multisystem nature of ATTR and the diverse care needs of this patient population.

Beyond HCRU, findings from this analysis show substantial cost burden, averaging $64,400 PPPY, related to overall disease burden and progression in patients with ATTRv-PN, even when drug costs are omitted (the wholesale acquisition cost is $477,404 annually). Cost analyses conducted in Italy and Spain corroborate the results shown in the current analysis. A retrospective analysis of burden on the Italian health care service of patients with ATTRv-PN demonstrated that mean annual cost per patient was €122,017 and drug costs were 95% of total expenditure [12]. A retrospective study conducted in Spain showed that total costs PPPY for patients with ATTRv-PN were €3105 [13]. While results from these studies cannot be compared directly to the current study, they demonstrate the high cost of care for patients with ATTRv-PN. By indirect comparison—given that 70% of patients within Optum^® CDM had reported heart failure—the annual cost of caring for a patient with heart failure in the USA is reported to be approximately $28,000–30,000 [25, 26]. Importantly, treatments that extend survival will result in higher cumulative health care costs as patients live longer and continue to require care—a consequence of therapeutic success rather than treatment burden.

Post-discontinuation treatment patterns provide insights into real-world therapeutic management. Of the patients who discontinued vutrisiran, 77% and 28% switched therapies, in the Optum^® CDM and Komodo databases, respectively. Of these switches, within Optum^® CDM, 62% switched to tafamidis monotherapy, 8% to vutrisiran and tafamidis combination therapy, and 8% to other monotherapy or combination therapy with patisiran. This high proportion of therapy switching may indicate perceived limitations in treatment effectiveness among some patients or providers and highlights the need for further investigation. Despite discontinuation differences between databases, treatment persistence patterns were remarkably similar in both cohorts; vutrisiran persistence was 77% at 12 months across the Optum^® CDM and Komodo databases.

To contextualize the 77% annual persistence rate observed with vutrisiran, it is helpful to consider persistence data from other injectable therapies that have similar dosing schedules. Proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors indicated for hypercholesterolemia provide a relevant comparator from the cardiovascular therapeutic area. Among the PCSK9 inhibitors, inclisiran is administered by a health care professional approximately every 6 months after an initial 3-month loading period, similar to the quarterly schedule of vutrisiran [27]. Inclisiran administered by health care professionals on a schedule comparable to vutrisiran demonstrated 77% persistence at 12 months in the same Komodo database population [28]. The identical 77% persistence rates for vutrisiran and inclisiran suggest that the similar routes of administration and dosing schedules may contribute to the observed persistence rates. However, it is important to note studies have shown that some patients prefer at-home vs clinic-based treatment [29].

Of note, at least half of patients had cardiovascular comorbidities in the 12 months after initiating vutrisiran (e.g., heart failure, 70% vs 49%; hypertensive heart disease, 39% vs 13%; cardiomyopathy, 39% vs 21%, in the Optum^® CDM and Komodo databases, respectively). The combination of higher mortality, greater tafamidis switching, and higher cardiac comorbidity burden strongly suggests a greater proportion of patients prescribed vutrisiran for ATTRv-PN in the Optum^® CDM database versus the Komodo database had comorbid ATTR-CM or a mixed phenotype (defined as a combination of ATTRv-PN and ATTR-CM manifestations).

As with all administrative claims databases, both Optum^® CDM and the Komodo database capture data as submitted for reimbursement rather than from direct clinical observation. This introduces several important limitations. First, diagnoses are assigned based on International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes recorded for billing purposes, which may not fully reflect clinical phenotype, disease severity, or diagnostic certainty. Rare diseases such as ATTR are particularly susceptible to underreporting or miscoding, as providers may default to broader diagnostic categories (e.g., unspecified cardiomyopathy or heart failure) rather than ATTR-specific codes, leading to potential misclassification of phenotype [30, 31]. Second, clinical measures routinely used to characterize disease severity, such as New York Heart Association Functional Classification, 6 min walk distance, TTR levels, neurological disability scores, or echocardiographic parameters, are not captured in claims data. Consequently, the degree of cardiac versus neurological involvement cannot be quantified beyond what is inferred from diagnostic coding patterns. Third, the reason for treatment discontinuation is not recorded in either database, limiting interpretation of persistence data. These constraints are inherent to claims-based analyses and underscore the need for complementary studies using registry or electronic health record data with richer clinical phenotyping.

This study was associated with several additional limitations. While Optum^® CDM is a closed database that is well suited to assess system costs, the Komodo database only has imputed cost data; thus, we were unable to replicate the cost analyses. Differences in treatment-switching patterns between Optum^® CDM and the Komodo database may be due to a coverage issue, as Optum^® CDM is a single payer and may not cover combination therapy. One benefit of the Komodo database is that it includes multiple payers and Medicare; therefore, it is a more reliable source than Optum^® CDM for broader generalization. Patients in Optum^® CDM appeared to have greater cardiac involvement and disease severity compared with the broader population of patients treated with vutrisiran; therefore, HCRU and cost estimates may represent an upper bound of what might be observed in a less severely ill cohort. Exclusion rates (52% in Optum^® CDM and 55% in the Komodo database) were similar to those routinely observed in real-world claims database analyses [32, 33], but the included population may not fully represent the overall population of patients that initiated vutrisiran during the study period. Due to differences in the databases, there were some differences in which outcomes were reported (i.e., the proportion of patients who switched treatment, mortality rates, and rates of cardiovascular comorbidities). While mean age was similar between the Optum^® CDM and Komodo databases, the Komodo database had a greater proportion of patients ≥ 80 years of age (18% vs 12%). It should be noted that Optum^® CDM includes Medicare Advantage, while the Komodo database includes Medicare Advantage and Fee-for-Service, potentially explaining the higher proportions of patients ≥ 80 years of age in the Komodo database. While the cost of medical care varies considerably by state [34], an analysis of HCRU and costs by region was not performed due to limited sample size. Data for vutrisiran were analyzed during the period when it was only approved for ATTRv-PN (before ATTR-CM approval), so future studies are needed to evaluate HCRU burden associated with vutrisiran in patients with ATTR-CM when adequate real-world exposure has accumulated for this patient population. We are unable to definitively differentiate between patients with pure ATTR polyneuropathy, pure ATTR cardiomyopathy, and mixed phenotype, as ICD-10-CM coding for ATTR subtypes was not used for patient identification. Consequently, our cohort may have included some patients with cardiac involvement or had mixed phenotype disease, which may involve different disease trajectories. Because ATTR-CM and mixed-phenotype disease are associated with higher rates of hospitalization and cardiac-related resource use than ATTRv-PN, the HCRU and cost estimates reported here from Optum^® CDM may not be generalizable to patients with predominantly neurological diseases; these data likely represent an upper bound of costs in the broader ATTRv-PN population. An analysis of costs by gender was not performed due to limited sample size. Finally, reasons for treatment discontinuation were not captured in either database. In Komodo, one-third of patients who discontinued did not switch to alternative ATTR therapies, highlighting the need for future research to understand discontinuation drivers and inform treatment optimization.

Conclusion

Within 1.5 years of vutrisiran initiation during a time period when it was approved for ATTRv-PN, 50% of patients presented to the ED and 37% of patients were hospitalized, highlighting continued disease burden and progression. Despite vutrisiran persistence of 77% at 12 months, within that year, roughly one-fourth of patients died or discontinued therapy, with many switching to other ATTR treatments. Vutrisiran treatment was associated with a high cost of care, excluding the cost of treatment itself. The total costs (all patients) included those associated with inpatient hospitalization (39%) and ED visits not leading to hospitalization (5%), and were driven by a significant portion of patients experiencing these events post-treatment initiation. Outpatient care at hospital facilities and office visits were the main cost drivers of total costs. These findings suggest that patients treated with vutrisiran continue to experience significant HCRU and costs related to disease progression despite treatment. New therapies are needed to further reduce ATTR burden and treatment costs.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Nicole K. Bart, Daniel P. Judge, Margarita Udall, Liana Hennum, Melissa Allison, Avinash Dash, Shradha Sanghi, Rakesh Ramesh, and Richard Wright contributed to and approved of the final manuscript.

Funding

Funding for this study and the journal’s Rapid Service and Open Access fees were provided by BridgeBio Pharma, Inc. (San Francisco, CA, USA).

Data Availability

The datasets generated and/or analyzed during the current study are not publicly available because data that support the findings of this study were used under license from the Optum^® CDM Database and from the Komodo Healthcare Map. Due to data use agreements and its proprietary nature, restrictions apply regarding the availability of the data. Further information is available from the sponsor of the study, BridgeBio Pharma, Inc. (San Francisco, CA, USA).

Declarations

Conflict of Interest

Nicole K. Bart received research funding from Pfizer; reports speaking fees from Pfizer; and reports advisory board fees from BridgeBio Pharma, Inc., Bristol Myers Squibb, Novo Nordisk, and Pfizer. Daniel P. Judge reports consultancy fees from Alexion Pharmaceuticals, Alnylam Pharmaceuticals, Attralus, Bayer, BridgeBio Pharma, Inc., Cytokinetics, Lexeo Therapeutics, Novo Nordisk, Rocket Pharmaceuticals, and Tenaya Therapeutics. Margarita Udall is an employee and stockholder of BridgeBio Pharma, Inc.; and stockholder of Pfizer. Liana Hennum is an employee and stockholder of BridgeBio Pharma, Inc. Melissa Allison, Avinash Dash, Shradha Sanghi, and Rakesh Ramesh are employees of Definitive Healthcare, which received financial compensation from BridgeBio Pharma, Inc. for performing research. Richard Wright reports research funding from BridgeBio Pharma, Inc. and Cytokinetics; speaking fees from Alnylam Pharmaceuticals, American Heart Association, Amgen, AstraZeneca, Bayer, Bristol Myers Squibb, Boehringer Ingelheim, BridgeBio Pharma, Inc., Lexicon, Lilly, and Novartis; consultancy fees from Alnylam Pharmaceuticals, American College of Cardiology, American Heart Association, Amgen, AstraZeneca, Bristol Myers Squibb, Boehringer Ingelheim, BridgeBio Pharma, Inc., Cytokinetics, MyoKardia, Novartis, and Providence Health; advisory board fees from Alnylam Pharmaceuticals, Amgen, AstraZeneca, Bristol Myers Squibb, Boehringer Ingelheim, BridgeBio Pharma, Inc., Cytokinetics, and Novartis.

Ethical Approval

The Optum^® CDM and Komodo databases comply with the Health Insurance Portability and Accountability Act. Given the anonymized, de-identified nature of the data, patient consent was not possible or required, and institutional review board approval to conduct this study was not necessary. Only aggregate cohort-level results were reported, without identification of any health care providers or healthcare facilities.