Direct-acting antiviral retreatment patterns for hepatitis C

Abstract

BACKGROUND:

Despite the strong efficacy of direct-acting antivirals (DAAs) against the hepatitis C virus, many patients require a second regimen of DAA treatment. However, limited research exists to characterize rates of retreatment across different DAA agents or potential factors that may increase retreatment risk.

OBJECTIVE:

To characterize patterns and predictors of DAA retreatment among a large, generalizable, commercially insured US population of patients.

METHODS:

Using the IBM MarketScan Commercial Claims and Encounters data source, this retrospective cohort study examined retreatment patterns among patients receiving DAAs between 2013 and 2019. Descriptive statistics were used to compare patient characteristics predictive of retreatment risk and to examine rates of retreatment in patients initiating different DAA treatments.

RESULTS:

Among 31,553 DAA users, a total of 1,017 (3.2%) required DAA retreatment. Among the 1,017 patients re-treated, 44 (4.3%) received a third treatment regimen and 2 patients received a fourth treatment regimen. The average total cost for a retreatment regimen was $109,683, with patient out-of-pocket costs totaling $1,287 Patients requiring retreatment had higher rates of hypertension (32.0% vs 26.7%; P < 0.001), diabetes (16.9% vs 11.9%; P < 0.001), coagulopathy (9.9% vs 4.5%; P < 0.001), deficiency anemia (11.1% vs 7.4%; P < 0.001), alcohol abuse (3.3% vs 2.3%; P = 0.038), prior liver transplantation (3.4% vs 2.3%; P = 0.024), and hepatocellular carcinoma (6.1% vs 1.9%; P < 0.001) compared with patients not requiring retreatment.

CONCLUSIONS:

Although uncommon, some patients receiving DAAs require a second regimen of DAA treatment at substantial cost to both health plans and patients. These patients tend to have more comorbidities and markers of hepatic disease severity. Patients with high retreatment risk may benefit from careful monitoring for occurrences of retreatment.

Plain language summary

Retreatment occurred in 3.2% of direct-acting antiviral users at an average cost of $109,683 to health plans and $1,287 to patients. Retreatment was more common in patients with comorbid health conditions, such as high blood pressure and diabetes. Patients at risk of retreatment may benefit from careful monitoring by physicians to ensure they are adherent to treatment once started and have sustained viral suppression once treatment is complete.

Implications for managed care pharmacy

Although uncommon, our study demonstrated that approximately 3% of direct-acting antiviral users had retreatment with a secondary direct-acting antiviral. This retreatment carried a significant cost to health plans and patients. Managed care efforts to reduce the risk of retreatment may hold benefit to health plans and patients. Managed care providers and payers should emphasize improving adherence to decrease reinfection with hepatitis C virus, particularly in patients with conditions associated with retreatment, such as diabetes and alcohol use.

The approval of Sovaldi (sofosbuvir [SOF]; Gilead Sciences) by the US Food and Drug Administration (FDA) in late 2013 represented a significant shift in the treatment of the more than 3.2 million patients in the United States with hepatitis C virus (HCV) infection.¹ This direct-acting antiviral (DAA) represented the first product with truly curative potential for HCV. The clinical trial evidence supporting the approval of SOF showed achievement of sustained virologic response (SVR) from 84 days of treatment in at least 90% of patients 12 weeks after the end of treatment.² This approval quickly led to additional FDA approvals of DAA treatments, including SOF and ledipasvir combination (SOF/LDV) and ombitasvir, paritaprevir, ritonavir, and dasbuvir combination (OBV/PTV/RTV and DSV) in 2014; SOF and velpatasvir combination (SOF/VEL) and elbasvir and grazoprevir combination (EBR/GRZ) in 2016; and SOF, VEL, and voxilaprevir combination (SOF/VEL/VOX) and glecaprevir and pibrentasvir combination (GLE/PIB) in 2017.² These newer treatments offer clinical advantages to SOF mono-treatment^3-6 resulting in a shift of DAA treatments toward newer products.

With the introduction of highly effective DAA treatments, the overall annual treatment rates among patients with HCV increased from 4.8% in 2014 to 16.9% in 2017,⁷ with the majority of this increase attributable to a greater prescribing of newer DAA treatments.⁸ At a median cost of $84,000 per DAA treatment regimen,⁹ this increased utilization has had a significant impact on US health expenditures. It is estimated that the United States spent $59.7 billion on HCV medications from 2014 to 2018 alone.¹⁰ This increased spending has been shown to result in a reduction in total health care spending for HCV because of its effectiveness in reducing the need for other health services, such as inpatient, outpatient, and physician services.¹¹ In addition, patient benefits from achieving SVR include a reduced risk of all-cause mortality, mortality due to liver disease, cirrhosis, and hepatocellular carcinoma.¹²

Although the benefits of DAA management are apparent, not all patients obtain SVR following their initial treatment. In addition to variations in the effectiveness of different DAA treatments, additional factors that have been shown to influence rates of SVR following DAA initiation include patient factors (eg, demographic factors, comorbid health conditions, or medication burden), treatment-related factors (eg, adherence or duration of treatment), and the presentation of the virus itself.¹³ Patients that fail to respond to initial treatment are recommended to receive a secondary DAA treatment. The American Association for the Study of Liver Diseases and the Infectious Diseases Society of America guidance recommends that the selection of a second treatment take into account HCV genotype, the patient’s treatment experience, and the extent of a patient’s liver damage.^14,15

Limited studies have examined patterns of retreatment with DAA medications.^16-21 Of the studies that have been conducted, most emphasize retreatment rates among patients initiating or being re-treated with a specific DAA agent.^16-19,21 The majority were also conducted in non-US clinical settings, which vary considerably from the US health care system^18-21 and contained small samples of patients from individual health systems.^16-19 The objective of this study was to characterize rates of retreatment among a large, generalizable, commercially insured US population of patients initiating DAA treatments. The specific primary aims of this study were to (1) examine variations in rates of retreatment among patients initiating different DAA treatments, (2) examine the retreatment options used as secondary treatments in patients requiring secondary treatment following initial DAA treatments, and (3) examine the patient characteristics that are predictive of the need for retreatment. Secondary aims were to (1) determine the average patient out-of-pocket (OOP) payment for the initial treatment and retreatment periods by patients taking each initial treatment regimen and (2) determine the average net payment for the initial treatment and retreatment periods by patients taking each initial treatment regimen. Given the burden of prolonged infection on patients with HCV as well as the considerable expense associated with the need for retreatment, the results of this study have importance to both clinicians and the managed care community.

Methods

DATA SOURCE

This study was conducted using the IBM MarketScan Commercial Claims and Encounters data source from calendar years 2013-2019. MarketScan Commercial Claims and Encounters data represent insurance claims information for inpatient, outpatient, physician, prescription drug, and carved-out services. In addition, health insurance information pertaining to plan type, periods of enrollment, and patient demographic information are also captured. The data come from a selection of more than 100 different large health plans and employers representing more than 30 million patients in the United States annually.²² The large size of the MarketScan database helps to maintain sufficient cohort sizes for data analysis and broad representation of health service utilization from different regions of the United States.²²

STUDY DESIGN

This study (used a retrospective cohort study design to examine retreatment patterns among patients initiating DAA treatment. To examine predictors of retreatment, we required at least 6 months of continuous insurance enrollment after the first DAA prescription fill (the index date) to examine patient characteristics that may be predictive of retreatment risk and followed patients for as long as their insurance enrollment allowed to examine the potential for retreatment over time (Supplementary Figure 1 ^{(476.4KB, pdf)} , available in online article). To capture the most generalizable description of DAA retreatment possible, no continuous enrollment restrictions prior to DAA initiation were applied. This study uses a prevalent use design to examine retreatment from the first DAA prescription noted in the data.

POPULATION INCLUSION CRITERIA

Patients were included in this study if they initiated a DAA agent between January 1, 2013, and December 31, 2019 (the most recent years for which data were available at the time the study was conducted) (Figure 1). Patients using DAAs were identified using national drug codes. Nine different DAA treatment regimens were included for analysis: SOF, SOF with daclatasvir (SOF + DCV), SOF with simeprevir (SOF + SMV), SOF/LDV, OBV/PTV/RTV and DSV, EBR/GRZ, SOF/VEL, SOF/VEL/VOX, and GLE/PIB. These patients were further grouped into 3 cohorts representing (1) SOF-based regimens (SOF, SOF/SMV, SOF/LDV, SOF/DCV, SOF/VEL and SOF/VEL/VOX), (2) NS5A plus NS3/4A inhibitors (GLE/PIB and EBR/GRZ), and (3) NS5A plus NS5B plus NS3/4A inhibitors (OBV/PTV/RTV and DSV).

Figure 1.

Study Patient Selection Process

Although most patients filled combination products as a single prescription, many patients filled medication separately as 2 separate prescriptions (eg, SOF + DCV). Patients using 2 or more products together during an initial treatment regimen were labeled as combination users regardless, so long as they were filled together with no gap longer than 84 days between prescription fills. Combination treatment was identified to ensure secondary treatments were not erroneously labeled as retreatment episodes.

To examine evidence of potential retreatment, patients were required to be continuously enrolled in both medical and prescription benefits for a minimum of 6 months following the completion of the initial DAA regimen. Completion was defined as no further prescriptions for a DAA within an 84-day window of the last DAA prescription in the initial treatment episode. During the data cleaning process, a small number of patients with potentially missing data were excluded given evidence of zero or negative day supply for their DAA prescription or evidence of a negative payment for DAAs, which was thought to represent claims reversals not otherwise adjudicated. We also excluded a small number of patients (N = 14) using DAA treatment that was not concordant with treatment guidelines (eg, 2 patients with DCV monotherapy and 12 patients with SMV monotherapy) (Figure 1).

STUDY OUTCOMES

The primary outcome of this study was retreatment with a DAA following an initial treatment regimen. Retreatment was defined as evidence of a secondary DAA treatment following a minimum gap of 84 days following completion of the initial treatment regimen. Retreatment was identified regardless of whether a patient started again on the same DAA as the initial treatment or a different DAA treatment option. We chose a minimum 84-day gap to identify retreatment, as this represents clinical guidance for testing an SVR following the completion of initial treatment.¹⁵ Patients were followed until they disenrolled from the MarketScan data resource or the end of available data (December 31, 2019). It is important to note that, given a lack of clinical information in the claims data to show conclusive evidence of SVR, retreatment in this study may represent both initial treatment failure, which is treated with additional DAAs, and reinfection following a previously successful regimen of DAA.

Additionally, average patient OOP payment for the initial treatment, first retreatment, second retreatment, and third retreatment were determined for patients taking each initial treatment regimen. OOP payment was calculated by adding coinsurance, copayment, and deductible amounts associated with all fills in each treatment episode. Average net payment for each treatment episode was also calculated and represents the plan payment minus any patient OOP cost.

To examine potential predictors of retreatment risk, we examined several patient characteristics in the MarketScan data over a 6-month observation window following the initiation of DAA treatment. These characteristics included patient demographics (age, sex, geographic region of the country in which a patient resides, and urban/rural residence), risk factors for HCV infection (eg, hepatitis B coinfection and evidence of liver disease severity), comorbid health conditions (eg, hypertension, renal disease, and diabetes), drug dependence, alcohol abuse, HIV status, mean number of Elixhauser conditions,²³ health plan characteristics (enrollment in a high-deductible health plan and initial DAA copayment rate), and the year of initial treatment initiation. Factors associated with severe liver disease that were examined include cotreatment with ribavirin (RBV), evidence of hepatocellular carcinoma, hepatic fibrosis, and history of a liver transplantation. Copayment levels associated with initial DAA treatment were categorized into varying amounts to capture a broad distribution of copayments found in the data ($0, $1-$25, $26-$50, $51-$100, $101-$250, $251-$500, $501-$1,000, $1,001-$2,500, $2,501-$5,000, and above $5,000). Urban residence was dichotomized to represent patients living in a zip code designated by MarketScan as a metropolitan statistical area or otherwise.

STATISTICAL ANALYSES

We compared patient characteristics between patients identified as receiving retreatment and those with no evidence of retreatment. Continuous variables are reported as means and SD whereas categorical variables are reported as counts and percentages. Group comparisons for continuous variables were analyzed using the Student’s t-test statistic. Statistical testing of differences between groups utilizes the chi-square test for categorical variables. Results were considered statistically significant at an α of 0.05. Descriptive statistics were used to report rates of retreatment as a function of the initial treatment a patient started. Similar descriptive statistics were used to report the number of days until second treatment among patients labeled as retreated, the type of treatment used as a retreatment option, rates of second and third retreatment rates among patients that have any evidence of a second treatment, average net payment by treatment regimens, and average patient OOP payment by treatment regimens. All statistical analyses were carried out using SAS version 9.4 (SAS Institute Inc).

Results

A total of 39,947 patients initiated one of the study DAA medications between January 1, 2013, and December 31, 2019 (Figure 1). After applying exclusion criteria related to claims anomalies (eg, negative/zero-day supply or negative payment information) and requirements for continuous enrollment in both prescription and health benefits for 6 months following each DAA prescription, we obtained a population of 31,553 DAA users. Among the 31,553 patients initiating DAA treatment, we identified 1,017 (3.2%) patients who received a second DAA treatment following a gap of 84 days or longer after their initial treatment (Table 1).

Table 1.

Sociodemographic Characteristics of Patients (N = 31,553)

Characteristic	No retreatment (N = 30,536)	Retreatment (N = 1,017)	P value
Demographics
Age, mean (SD), years	53.3 (10.4)	54.7 (8.1)	< 0.001
Age categories (in years), n (%)
< 18	23 (0.1)	0 (0)	< 0.001
18-34	2,626 (8.6)	42 (4.1)
35-44	2,248 (7.4)	44 (4.3)
45-54	6,689 (21.9)	264 (26.0)
55-64	18,950 (62.1)	667 (65.6)
Sex, n (%)
Male	18,961 (62.1)	730 (71.8)	< 0.001
Female	11,575 (37.9)	287 (28.2)
Geographic region, n (%)
Northeast	5,841 (19.1)	206 (20.3)	0.414
North central	4,680 (15.3)	167 (16.4)
South	14,919 (48.9)	467 (45.9)
West	4,935 (16.2)	173 (17.0)
Unknown	161 (0.5)	4 (0.4)
Urban residence, n (%)	25,914 (84.9)	896 (88.1)	0.005
Specific Elixhauser conditions, n (%)
Hypertension	8,145 (26.7)	325 (32.0)	< 0.001
Diabetes	3,647 (11.9)	172 (16.9)	< 0.001
Uncomplicated diabetes	3,096 (10.1)	151 (14.8)	< 0.001
Complicated diabetes	1,309 (4.3)	54 (5.3)	0.114
Liver disease	12,756 (41.8)	594 (58.4)	< 0.001
Drug dependence	1,065 (3.5)	30 (2.9)	0.357
Alcohol use	713 (2.3)	34 (3.3)	0.038
HIV/AIDS	598 (2.0)	23 (2.3)	0.494
Congestive cardiac failure	350 (1.1)	12 (1.2)	0.921
Renal failure	350 (1.1)	12 (1.2)	0.899
Coagulopathy	1,387 (4.5)	101 (9.9)	< 0.001
Deficiency anemia	2,266 (7.4)	113 (11.1)	< 0.001
No. of Elixhauser conditions, mean (SD)	1.50 (1.70)	2.01 (1.93)	< 0.001
Risk factors, n (%)
Hepatic fibrosis	459 (1.5)	13 (1.3)	0.561
Liver transplant	715 (2.3)	35 (3.4)	0.024
Hepatocellular carcinoma	591 (1.9)	62 (6.1)	< 0.001
Hepatitis B	374 (1.2)	17 (1.7)	0.205
Ribavirin	6,818 (22.3)	481 (47.3)	< 0.001
Insurance plan, n (%)
High-deductible plan	1,646 (5.4)	39 (3.8)	0.030
Not high-deductible plan	28,890 (94.6)	978 (96.2)
Index out-of-pocket payment, n (%)
$0	2,229 (7.3)	81 (8.0)	< 0.001
$1-$25	4,871 (16.0)	180 (17.7)
$26-$50	7,036 (23.0)	257 (25.3)
$51-$100	6,637 (21.7)	256 (25.2)
$101-$250	3,697 (12.1)	95 (9.3)
$251-$500	842 (2.8)	18 (1.8)
$501-$1,000	683 (2.2)	15 (1.5)
$1,001-$2,500	1,674 (5.5)	40 (3.9)
$2,501-$5,000	1,610 (5.3)	32 (3.2)
> $5,000	1,257 (4.1)	43 (4.2)
Year of treatment initiation, n (%)
2013	138 (0.5)	25 (2.5)	< 0.001
2014	6,789 (22.2)	474 (46.6)
2015	8,880 (29.1)	284 (27.9)
2016	5,587 (18.3)	114 (11.2)
2017	4,270 (14.0)	84 (8.3)
2018	3,455 (11.3)	36 (3.5)
2019	1,417 (4.6)	0 (0)

Patients who were re-treated were older (54.7 vs 53.3; P < 0.001) and more often male (72% vs 62%; P < 0.001) than patients not requiring retreatment. Patients requiring retreatment also had higher rates of comorbid health conditions, including hypertension (32.0% vs 26.7%; P < 0.001), diabetes (16.9% vs 11.9%; P < 0.001), liver disease as defined by the Elixhauser Index (58.4% vs 41.8%; P < 0.001), coagulopathy (9.9% vs 4.5%; P < 0.001), deficiency anemia (11.1% vs 7.4%; P < 0.001), and alcohol abuse (3.3% vs 2.3%; P = 0.038), than patients not re-treated. Among the risk factors we examined, patients who were re-treated had higher rates of a prior liver transplant (3.4% vs 2.3%; P = 0.024), hepatocellular carcinoma (6.1% vs 1.9%; P < 0.001), and evidence of RBV use (47.3% vs 22.3%; P < 0.001). One potential indicator of health insurance generosity suggested that patients who were re-treated had a lower rate of enrollment in a high-deductible health plan (3.8% vs 5.4%, P = 0.030). Finally, rates of retreatment appeared to be declining annually. Among the 1,017 patients requiring retreatment, 46.6% initiated treatment in 2014 compared with only 8.3% in 2017 and 3.5% in 2018.

Table 2 describes patterns of DAA treatment and rates of retreatment for patients initiating different DAA regimens. SOF/LDV was the most common DAA initiated (53.0%), followed by SOF (16.3%) and SOF/VEL (10.3%). Very few patients started treatment with EBR/GRZ and SOF + DCV (0.9% and 1.7% of patients, respectively). Patients receiving the SOF + SMV combination as initial treatment had the highest retreatment rate of 8.2%, followed by patients receiving SOF alone (7.6%), and the OBV/PTV/RTV and DSV combination (4.9%). Lower retreatment rates were found for patients initiating GLE/PIB (1.0%), SOF/VEL (1.2%), and SOF/LDV (1.7%). The average gap between initial treatment and subsequent retreatment suggests that initiating on the SOF + DCV combination resulted in the highest mean number of days (512 days) until retreatment was needed, followed by SOF/LDV (420 days) and SOF alone (412 days). In addition to primary retreatment, several patients were identified as needing a second and third retreatment in the data as well. Second retreatment is most common in OBV/PTV/RTV and DSV users (9.48%), followed by SOF (4.33%) and SOF/LDV (4.3%). In addition, 2 of the 17 (11.8%) patients receiving a third DAA treatment (secondary retreatment) after initiating SOF required a fourth DAA treatment (tertiary retreatment).

Table 2.

Retreatment Rate and Mean Number of Days From Initial Treatment to the First Retreatment, by Initial Treatment

Treatment regimen	No. (%) of patients initiating DAA treatment	No. (%) of patients requiring secondary treatment (first retreatment)	Mean no. of days to first retreatment (±SD)	No. (%) of re-treated patients requiring a third treatment (secondary retreatment)	No. (%) of re-treated patients requiring a fourth treatment (tertiary retreatment)
SOF	5,155 (16.3)	393 (7.6)	412 (±275)	17 (4.3)	2 (11.8)
SOF+SMV	1,904 (6.0)	157 (8.2)	305 (±238)	4 (2.5)	0 (0)
SOF/LDV	16,733 (53.0)	279 (1.7)	420 (±288)	12 (4.3)	0 (0)
OBV/PTV/RTV and DSV	2,346 (7.4)	116 (4.9)	385 (±256)	11 (9.5)	0 (0)
SOF/VEL	3,237 (10.3)	39 (1.2)	298 (±182)	0 (0)	0 (0)
SOF+DCV	523 (1.7)	15 (2.9)	512 (±248)	0 (0)	0 (0)
GLE/PIB	1,261 (4.0)	13 (1.0)	206 (±76)	0 (0)	0 (0)
EBR/GRZ	281 (0.9)	5 (1.8)	402 (±302)	0 (0)	0 (0)
SOF/VEL/VOX	111 (0.4)	0 (0.0)	N/A	0 (0)	0 (0)
Total	31,551	1,017 (3.2)	389 (±270)	44 (4.3)	2 (4.5)

Two patients using SOF+GLE/PIB and SOF+EBR/GRZ were excluded from analysis because of insufficient sample size.

DAA = direct-acting antiviral; DCV = daclatasvir; DSV = dasabuvir; EBR = elbasvir; GLE = glecaprevir; GRZ = grazoprevir; LDV = ledipasvir; N/A = not available;

OBV = ombitasvir; PIB = pibrentasvir; PTV = paritaprevir; RTV = ritonavir; SMV = simeprevir; SOF = sofosbuvir; VEL = velpatasvir; VOX = voxilaprevir.

Figure 2 and Supplementary Table 1 ^{(476.4KB, pdf)} describe patterns of retreatment as a function of which treatment a patient originally initiated. Out of 1,017 patients receiving retreatment, SOF/LDV was the most common retreatment regimen received by a total of 454 patients (44.6%), followed by SOF/VEL/VOX and SOF/VEL in 174 (17.1%) and 130 (12.8%) patients, respectively (Supplementary Table 1 ^{(476.4KB, pdf)} ). SOF/LDV as retreatment regimen was received predominantly by patients initiating SOF (185/454), SOF/LDV (94/454), SOF + SMV (120/454), or OBV/PTV/RTV + DSV (54/454) as the initial treatment. SOF/VEL/VOX was received as retreatment predominantly by patients using SOF/LDV (100/174), SOF/VEL (20/174), or OBV/PTV/RTV + DSV (23/174) as the initial treatment. SOF/VEL was received as retreatment predominantly by patients using SOF (61/130), SOF/LDV (34/130), or SOF/VEL (15/130) as the initial treatment.

Figure 2.

Retreatment Options Used Among Patients Requiring Secondary Treatment

A total of 883 patients who received SOF-based regimens as the initial treatment required retreatment. Among these patients, SOF/LDV (399/883), SOF/VEL/VOX (142/883), and SOF/VEL (118/883) were the most common regimens used as retreatment. In patients using NS5A plus NS3/4A inhibitors as the initial treatment, SOF/VEL/VOX (9/18), SOF/VEL (4/18), and GLE/PIB (3/18) were the most common regimens used as retreatment. Among patients using NS5A plus NS5B plus NS3/4A inhibitors as the initial treatment, 116 patients received retreatment: SOF/LDV (54/116), SOF/VEL/VOX (23/116), and OBV/PTV/RTV + DSV (22/116) were the most common regimens used as retreatment.

We calculated OOP cost by adding the coinsurance, copayment and deductible. After looking at the final values, we can say that PAY-NETPAY = OOP. So, PAY is total payment, NETPAY is payment done by the insurance, and OOP is the payment done by the patient. Table 3 describes both the average net payment representing the plan payment amount associated with each treatment regimen as well as the patient OOP cost associated with each treatment episode. In general, the average net payment and OOP payment for the first retreatment period increased from the first treatment period. However, payments decreased later in the second and third retreatment periods. SOF + DCV, SOF + SMV, and SOF had the highest average net payment of $147,239, $131,448, and $95,277, respectively, for the initial treatment period. Although GLE/PIB had the lowest average net payment of $21,683, it had the highest OOP payment of $1,884 for the initial treatment. GLE/PIB was followed by SOF/LDV and SOF + SMV, having an average OOP cost of $1,349 and $1,344, respectively, for the initial treatment. For both the first and the second retreatment periods, SOF + SMV, SOF/LDV, and SOF had the highest average net payment and OOP cost.

Table 3.

Variation in Plan and Out-of-Pocket Payment Amounts Across Initial Treatment Strategy

Initial treatment regimen	No. of patients needing retreatment	Average net payment, USD (SD)				Average out-of-pocket payment, USD (SD)
		Initial treatment	First retreatment	Second retreatment	Third retreatment	Initial retreatment	First retreatment	Second retreatment	Third retreatment
SOF	393	$95,277 (50,655)	$128,789 (69,367)	$97,112 (63,618)	$70,068 (.)	$1,205 (5,211)	$1,304 (3,902)	$616 (1,065)	$75 (.)
SOF/LDV	279	$74,977 (48,692)	$85,119 (58,811)	$94,303 (44,351)	—	$1,349 (4,555)	$1,668 (7,959)	$1,341 (3,228)	—
SOF+SMV	157	$131,448 (46,708)	$139,676 (58,266)	$164,034 (100,836)	—	$1,344 (5,261)	$1,418 (6,537)	$846 (1,137)	—
OBV/PTV/RTV and DSV	116	$54,340 (38,836)	$84,090 (39,772)	$75,309 (19,740)	—	$617 (2,192)	$720 (2,403)	$615 (1,173)	—
SOF/VEL	39	$51,480 (29,818)	$54,442 (19,074)	—	—	$750 (1,588)	$611 (1,979)	—	—
SOF+DCV	15	$147,239 (38,953)	$78,001 (25,963)	—	—	$609 (760)	$262 (269)	—	—
GLE/PIB	13	$21,683 (9,991)	$65,481 (28,875)	—	—	$1,884 (3,980)	$114 (93)	—	—
GRZ/EBR	5	$50,874 (20,403)	$83,762 (59,858)	—	—	$1,056 (2,087)	$896 (1,559)	—	—
Average spending among all re-treated patients	1,017	$88,550 (53,435)	$109,915 (65,211)	$97,106 (58,317)	$70,068 (.)	$1,181 (4,618)	$1,304 (5,531)	$815 (1,819)	$75 (.)

DCV = daclatasvir; DSV = dasabuvir; EBR = elbasvir; GLE = glecaprevir; GRZ = grazoprevir; LDV = ledipasvir; OBV = ombitasvir; PIB = pibrentasvir; PTV = paritaprevir; RTV = ritonavir; SMV = simeprevir; SOF = sofosbuvir; USD = US dollar; VEL = velpatasvir; VOX = voxilaprevir.

Discussion

This is the most comprehensive study among a US population, to our knowledge, to examine patterns of retreatment for HCV following the initiation of DAA treatment. Among a large, broad geographic representation of commercially insured patients, we identified 31,553 patients initiating DAA treatment, of whom 1,017 (3.2%) required retreatment with a second DAA regimen. Of note, 44 patients in this sample required a third treatment regimen and another 2 patients required a fourth treatment regimen. The cost associated with retreatment was significant to both health plans and patients, with an average net payment of $109,683 and an average patient OOP cost of $1,287.

The initial clinical trial evidence that accompanied FDA approval of each DAA treatment demonstrated high, but imperfect, rates of SVR following initial treatment.^2,24 It is therefore expected that some patients may require secondary treatment with a subsequent DAA regimen despite fully completing their initial regimen. The clinical trial evidence supporting drug approval generally presents an ideal clinical circumstance whereby patients have high rates of adherence and tend to be healthier than the general population, such as that represented in this study.

Descriptively, we found higher rates of retreatment for SOF (7.6%) and SOF + SMV (8.2%) and lower rates for SOF/VEL/VOX (0%), GLE/PIB (1%), and SOF/VEL (1.2%). The high retreatment rate with SOF and SOF + SMV is not surprising, given the single mode of action of SOF and the efficacy gap of SOF and SOF + SMV against the most prevalent HCV genotype (genotype 1).^6,25,26 In addition, given that these products were FDA approved early, there may be a longer follow-up time to observe retreatment in these agents. In general, later-introduced SOF in combination with different NS5A inhibitors (SOF/LDV or SOF/VEL) or NS5A inhibitors and NS3A/4A inhibitors (SOF/VEL/VOX) showed low retreatment rates (< 2%). These results are not surprising given the evidence of pangenotypic efficacy, shorter duration of treatment, and lower adverse effects^18,27 of the newest members of the DAA family: GLE/PIB, SOF/VEL, and SOF/VEL/VOX. This is favorable for ensuring low failure rates, increased adherence, and low discontinuation rates due to adverse effects.

Previous research found that success or failure with DAA treatment depends on host factors (cirrhosis or hepatocellular carcinoma), viral factors (viral genotype or resistance-associated substitutions), and treatment-related factors (previous therapy, therapy duration, and adherence).^13,19 With chronic HCV infection cure rates of over 90% using the new-generation DAAs in just 8 to 12 weeks,^4,25 nonadherence is the most important risk factor for HCV treatment failure.^28,29 Younossi et al conducted a study on patients from 13 multinational clinical trials to assess the impact of patient-reported outcomes on adherence to different anti-HCV regimens. In this study, the adherence to interferon- and RBV-containing, interferon-free RBV-containing, and interferon- and RBV-free regimens were 77.6%, 84.3%, and 96.2%, respectively.³⁰ Although newer interferon- and RBV-free DAA regimens show higher adherence compared with RBV-containing regimens, there is still some gap in adherence to fill. Although adherence in clinical trials is mainly influenced by treatment-associated side effects, other factors such as treatment complexity (increased pill burden, self-administration, or duration of treatment)^{30 31} and patient-related factors (age, sex, education, current alcohol and drug use, affordability, short-term memory issues, or preoccupation with family/work responsibilities) can influence adherence in real-world settings.^30,31 Clinicians may consider these factors to ensure adherence to DAA regimens to decrease treatment failure. The rates of retreatment observed in this study suggest that patients initiating on DAA treatment may benefit from frequent monitoring to ensure that adherence to treatments is maintained and the patient has the best opportunity to achieve SVR with their initial treatment.

In addition to the potential need for retreatment following initial treatment failure (not achieving SVR), the rate of retreatment observed in this study also likely includes reinfection with HCV following a successful initial treatment. Our study found the highest rates of retreatment among DAA treatments that had been on the market for the longest time (SOF+SMV and SOF). These agents have a longer available follow-up, which may enable more ability to uncover higher rates of reinfection than newer DAA agents that have not been on the market as long. Risk factors for HCV reinfection that have been reported in the literature include coinfection with HIV,³² substance use (injection drug use and alcohol use disorders),³² sharing injecting equipment,³³ depression,³⁴ mental illness,³⁴ and homelessness.³³ Multicomponent interventions combining harm reduction and behavioral interventions based on specific behaviors and contributing conditions (eg, depression or alcohol use disorders) have been shown to be effective in reducing rates of HCV reinfection.^32,35 Clinicians may educate patients on the risks of reinfection, including the practices likely to increase risk; provide brief evidence-based behavioral interventions; and link patients to harm-reduction services to minimize HCV reinfection.

The need to monitor for potential reinfection risk may be of particular importance to patients with HCV who are injectable drug users. Among people who inject drugs (PWID), linkage to opioid substitution therapy has been shown to reduce injection frequency significantly,³³ and needle and syringe programs aided in the adoption of safer injection practices.³⁵ Past research showed that HCV treatment complemented with opioid substitution therapy and mental health counseling reduced HCV reinfection risk among PWID.³² In persons with depression or untreated mental illness, concurrent screening for and treatment of mental illness should be strongly considered at the time of HCV care engagement. Finally, a postclearance follow-up of PWID for HCV reinfection can facilitate timely retreatment with DAAs and prevent further transmission.

LIMITATIONS

The results of this study should be interpreted in light of the potential limitations of this study. As with any secondary data analysis, the potential for missing data is present. This study examined rates of retreatment that may be indicative of a failure of initial DAA treatment or reinfection risk. Given the lack of ability to confirm SVR through laboratory findings, we cannot say for certain the reason for retreatment, nor can we be assured that patients who were not re-treated obtained SVR and no longer have HCV risk. Although many variables were considered as potential predictors of HCV retreatment, several variables may also be of interest that were not available in these data, including race/ethnicity, sociodemographic variables, HCV genotype, severity of liver disease, and history of alcohol or drug abuse. There may be residual confounding due to a lack of measuring these variables. Finally, the results are generalizable to a large, commercially insured population in the United States. Caution should be used in extrapolating these results to other populations, such as patients in the Medicaid and Medicare programs.

Conclusions

Although uncommon, some patients receiving DAAs still require a second regimen of DAA treatment at substantial cost to both health plans and patients. Differences in several factors, including patient comorbidities and potential markers of hepatic disease severity, were observed between the retreatment and no retreatment groups. Providers should carefully consider these characteristics when screening for retreatment risk and providing clinical guidance and education to patients receiving initial DAA treatment regimens.