FormalPara Key Summary Points

Why carry out this study?

While clinical trial and real-world analyses have consistently demonstrated the efficacy/effectiveness of dolutegravir/lamivudine (DTG/3TC), additional data on the real-world impact of DTG/3TC use on costs would be beneficial to prescribers and payers.

Healthcare costs with DTG/3TC versus other single-tablet regimens were analyzed overall and among treatment-naive and treatment-experienced individuals.

What was learned from the study?

Unadjusted and adjusted healthcare cost differences generally favored DTG/3TC.

DTG/3TC had the lowest pharmacy costs before and after adjusting for confounders.

Results highlight cost benefits of DTG/3TC as an initial or switch option for the treatment of HIV-1.

Introduction

An important component in effectively suppressing HIV infection is adherence to daily antiretroviral therapy [1]. With optimal use of treatment, people living with HIV are anticipated to attain normal life expectancy [2]. However, the lifelong requirement for pharmacotherapy is costly, posing a high economic burden [3]. Moreover, with prolonged life expectancy, prescription medication for age-related comorbidities and potential drug interactions with antiretroviral therapies can exacerbate the burden of HIV. A retrospective analysis of Medicaid administrative claims data from January 2012 to March 2017 observed an increasing trend over time of concomitant medication use, including those contraindicated with antiretroviral therapies [4]. Due to the requirement for lifelong treatment, there is a need for effective therapies that reduce costs as well as the potential risk of drug interactions and long-term toxicity.

Streamlining antiretroviral therapy regimens from 3 or 4 drugs to effective 2-drug regimens could minimize cumulative drug exposure and reduce costs [5]. This strategy is supported by results from the GEMINI-1/-2 clinical trials (NCT02831673/NCT02831764), which demonstrated that the 2-drug regimen dolutegravir (DTG) + lamivudine (3TC) was non-inferior to the 3-drug regimen DTG + tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) in treatment-naive adults with HIV-1 [6]. In GEMINI-1/-2, 82% of the DTG + 3TC group versus 84% of the DTG + TDF/FTC group had virologic suppression (HIV-1 RNA < 50 copies/mL) at 3 years. In the TANGO trial (NCT03446573) of virologically suppressed adults who switched to DTG/3TC or continued 3- or 4-drug tenofovir alafenamide (TAF)-based regimens, 86% who switched to DTG/3TC maintained suppression versus 82% who continued TAF-based regimens at 3 years [7]. Results also showed that DTG/3TC has a high barrier to resistance and good safety and tolerability [6, 7]. Accordingly, the 2-drug regimen DTG/3TC is recommended as an initial antiretroviral therapy regimen for most people with HIV-1, with exceptions for those with HIV-1 RNA > 500,000 copies/mL, hepatitis B virus (HBV) co-infection, or in whom therapy is started before the results of HIV genotypic resistance testing for reverse transcriptase or HBV testing are available, and for virologically suppressed individuals switching therapies [8,9,10].

Real-world economic analyses have demonstrated substantial cost savings when switching to DTG/3TC. In Spain, savings of €2741 and €4164 per patient per year were observed with use of multi-tablet DTG + 3TC relative to DTG/abacavir(ABC)/3TC and elvitegravir (EVG)/cobicistat (COBI)/FTC/TAF, respectively [11]. In Canada, an economic analysis reported savings of CA$1.6 million in 2020 in a cohort of 989 people, only 40% of whom switched to DTG/3TC [12]. These real-world analyses also reported effectiveness results consistent with the efficacy from clinical trials. While these analyses are promising, additional data on the real-world impact of DTG/3TC use on costs would be beneficial to prescribers and payers. Here, we conducted a real-world analysis of costs associated with DTG/3TC use in the United States and compared those with other currently recommended first-line single-tablet regimens, including DTG/ABC/3TC, BIC/FTC/TAF, EVG/COBI/FTC/TAF, and DRV/COBI/FTC/TAF, using a large US claims database.

Methods

Data Source

The IQVIA PharMetrics® Plus database (IQVIA, Durham, NC, USA) contains fully adjudicated medical and pharmacy claims data for approximately 40 million patients with an average enrollment length of about 39 months, and covers all 50 US states. The enrollee population is generally representative of commercially insured individuals aged < 65 years in the US with respect to both age and sex. The data are de-identified and comply with the Health Insurance Portability and Accountability Act; as such, this study was exempt from ethics committee or institutional review board approval and informed consent.

Study Design

This study employed a retrospective cohort design. Adjudicated claims data were identified from the IQVIA PharMetrics Plus database between November 8, 2018, and March 31, 2020 (study period). The index date was defined as the date of the first pharmacy claim for DTG/3TC on or after April 8, 2019 [the date of US Food and Drug Administration (FDA) approval of DTG/3TC for treatment of HIV-1]. If no dispensing of DTG/3TC was observed, then the first use of the other selected antiretroviral single-tablet regimens defined the index date. The 6 months before the index date was used to evaluate patient demographics and clinical characteristics (baseline period). The observation period spanned from the index date until the end of continuous eligibility, unless the end of data availability was earlier, and was used to evaluate healthcare costs.

Study Population

Individuals were included in the study population if they received ≥ 1 dispensing of DTG/3TC, DTG/ABC/3TC, BIC/FTC/TAF, EVG/COBI/FTC/TAF, or DRV/COBI/FTC/TAF on or after April 8, 2019 (first dispensing defined as the index date), had ≥ 6 months of continuous health insurance coverage before the index date (baseline period), had ≥ 1 diagnosis of HIV-1 in the primary or secondary position at any time during the study period [as defined by the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM): B20, R75, Z21], and were aged ≥ 18 years. Individuals were excluded if they received DTG twice daily (BID), DRV BID, enfuvirtide, etravirine/maraviroc, or ibalizumab any time before the index date, as these regimens are typically prescribed for multidrug-resistant HIV-1. Other exclusion criteria included diagnosis of HIV-2 (ICD-10-CM: B97.35) in the primary or secondary position any time during the period of continuous health insurance coverage, or chronic hepatitis B viral infection (ICD-10-CM: B18.0, B18.1, Z22.51) in the primary or secondary position during the baseline period or on the index date. Included individuals were further stratified by treatment experience and classified as treatment-naive or treatment-experienced based on history of HIV antiretroviral therapy use in the baseline period.

Healthcare Cost and Utilization

All-cause and HIV-related total healthcare costs, including medical and pharmacy cost components, were evaluated during the observation period. HIV-related costs were identified as any claim with a primary or secondary diagnosis of HIV (ICD-10-CM: B20, R75, Z21), and HIV-related pharmacy claims were identified as any pharmacy claims for HIV treatment, including single-tablet and multi-tablet regimens. Total healthcare costs consisted of paid medical and pharmacy costs: medical costs included costs of hospitalizations, emergency department visits, and outpatient visits (further stratified by primary care office, specialist, and other visits). Costs were reported from the payer perspective per patient per month (PPPM) to account for varying lengths of follow-up and calculated by dividing costs incurred over the observation period by person-months of observation. The number and percentage of individuals with at least 1 all-cause or HIV-related healthcare utilization by type including hospitalizations, emergency department visits, office visits (primary care and specialist), and laboratory visits are reported in each cohort.

Statistical Analyses

Unadjusted healthcare costs were summarized using descriptive statistics, including mean and standard deviation (SD), and unadjusted cost differences were calculated by subtracting the comparator cost from the DTG/3TC cost. Adjusted cost differences and adjusted cost ratios were estimated using multivariable regression models controlling for differences in baseline covariates [i.e., age, sex, region, insurance plan type, year, and quarter of index date, Quan–Charlson comorbidity index, Elixhauser and HIV-related comorbidities (prevalence ≥ 5%), concomitant medications, prior use of single-tablet and multi-tablet regimens, and baseline healthcare resource utilization and costs] between cohorts. P values and 95% confidence intervals were calculated using non-parametric bootstrap procedures. All analyses were conducted using SAS® Enterprise Guide software version 7.15 (SAS Institute, Cary, NC, USA).

Results

Study Population

Overall, 22,061 people with HIV-1 met the study criteria for this analysis, with most individuals in the BIC/FTC/TAF cohort (n = 9068), followed by EVG/COBI/FTC/TAF (n = 7081), DTG/ABC/3TC (n = 4355), DRV/COBI/FTC/TAF (n = 967), and DTG/3TC (n = 590; Fig. 1). There were more treatment-naive individuals in the DTG/3TC (13%) and BIC/FTC/TAF (16%) cohorts compared with the other cohorts (DTG/ABC/3TC, 6%; EVG/COBI/FTC/TAF, 5%; DRV/COBI/FTC/TAF, 9%), although most people were treatment-experienced (84–95%). Before the index date, 39% (202/514) of treatment-experienced individuals in the DTG/3TC cohort had prior exposure to DTG/ABC/3TC, while 52–98% of those in the comparator cohorts had previously used their respective single-tablet regimens (Table 1).

Fig. 1
figure 1

Selection criteria. ABC abacavir, BIC bictegravir, BID twice daily, COBI cobicistat, DRV darunavir, DTG dolutegravir, EVG elvitegravir, FTC emtricitabine, TAF tenofovir alafenamide, 3TC lamivudine

Full size image
Table 1 Demographics and clinical characteristics of people living with HIV-1 by single-tablet regimen cohort
Full size table

Demographics and clinical characteristics were generally consistent across cohorts. Most claims data were from men with a mean age of approximately 46 years who were from the southern region of the US (63–77%) and had a mean Quan–Charlson comorbidity index score ranging from 3.5 to 3.8 (Table 1). Although the mean observation period was shorter among the DTG/3TC cohort (135 days) relative to the other single-tablet regimen cohorts (234–288 days), the proportion of the observation period covered with index medication was higher among the DTG/3TC cohort (91%) relative to the other single-tablet regimen cohorts (81–88%). The DRV/COBI/FTC/TAF and DTG/3TC cohorts had the highest baseline all-cause and HIV-related total healthcare costs. The most common baseline Elixhauser comorbidities were hypertension (24–29%), obesity (10–12%), and chronic pulmonary disease (6–9%), and the most common HIV-related comorbidities were cardiovascular disease (25–30%) and depression (10–11%). Antihyperlipidemics (23–26%) and antihypertensives (27–33%) were the most common concomitant medications used during baseline across all cohorts.

All-Cause Healthcare Costs

Mean unadjusted all-cause total healthcare costs PPPM were significantly lower among individuals with HIV-1 receiving DTG/3TC compared with those receiving BIC/FTC/TAF (P = 0.040) or DRV/COBI/FTC/TAF (P = 0.004; Fig. 2A), with unadjusted cost differences of $415 and $661, respectively. Unadjusted all-cause total healthcare cost differences were primarily driven by reduced pharmacy costs for the DTG/3TC cohort compared with all other single-tablet regimen cohorts (all P < 0.001). The largest pharmacy cost differences were between DTG/3TC compared with BIC/FTC/TAF ($521 difference) and DRV/COBI/FTC/TAF ($963 difference). Unadjusted medical costs did not significantly differ across cohorts.

Fig. 2
figure 2

a Unadjusted all-cause healthcare costs by single-tablet regimen cohort and b adjusted all-cause healthcare cost differences for DTG/3TC versus other single-tablet regimens. ABC abacavir, BIC bictegravir, COBI cobicistat, DRV darunavir, DTG dolutegravir, EVG elvitegravir, FTC emtricitabine, PPPM per patient per month, TAF tenofovir alafenamide, 3TC lamivudine. *P < 0.05 and **P < 0.001 when compared with DTG/3TC for the corresponding cost domain (i.e., comparisons within all-cause total healthcare, pharmacy, and medical costs)

Full size image

Similar outcomes were observed after adjusting for baseline covariates, with lower mean adjusted all-cause total healthcare costs PPPM among individuals with HIV-1 receiving DTG/3TC compared with those receiving BIC/FTC/TAF, EVG/COBI/FTC/TAF, and DRV/COBI/FTC/TAF, primarily driven by significantly lower pharmacy costs (Fig. 2B). As in the unadjusted analysis, adjusted all-cause pharmacy costs were significantly lower for the DTG/3TC cohort in all comparisons, and no significant differences were seen in medical costs (Fig. S1).

HIV-Related Healthcare Costs

Mean unadjusted HIV-related total healthcare costs PPPM were significantly lower among individuals with HIV-1 receiving DTG/3TC compared with those receiving DRV/COBI/FTC/TAF (P < 0.001; Fig. 3A), with an unadjusted cost difference of $747. After adjusting for baseline covariates, significantly lower HIV-related total healthcare costs were also observed for the DTG/3TC cohort compared with the BIC/FTC/TAF ($851 difference; P < 0.001), EVG/COBI/FTC/TAF ($966 difference; P < 0.001), and DRV/COBI/FTC/TAF cohorts ($995 difference; P < 0.001; Fig. 3B). As observed in the all-cause data, the difference in HIV-related costs resulted from significantly lower pharmacy costs, with the largest cost differences seen between DTG/3TC and BIC/FTC/TAF ($428 difference) and DRV/COBI/FTC/TAF ($997 difference). Mean unadjusted HIV-related medical costs were generally comparable across cohorts, except for significantly lower costs among the EVG/COBI/FTC/TAF cohort compared with the DTG/3TC cohort, with a cost difference of $251 (P = 0.044).

Fig. 3
figure 3

a Unadjusted HIV-related healthcare costs by single-tablet regimen cohort and b adjusted HIV-related healthcare cost differences for DTG/3TC versus other single-tablet regimens. ABC abacavir, BIC bictegravir, COBI cobicistat, DRV darunavir, DTG dolutegravir, EVG elvitegravir, FTC emtricitabine, PPPM per patient per month, TAF tenofovir alafenamide, 3TC lamivudine. *P < 0.05 and **P < 0.001 when compared with DTG/3TC for the corresponding cost domain (i.e., comparisons within HIV-related total healthcare, pharmacy, and medical costs)

Full size image

More broadly, most adjusted cost ratios for HIV-related total healthcare and pharmacy costs significantly favored DTG/3TC, with almost no significant differences in HIV-related medical cost ratios (Figure S2). The lone exception was significantly lower medical costs for the DTG/ABC/3TC cohort ($408; P = 0.032). With adjusted HIV-related pharmacy costs being significantly lower for the DTG/3TC cohort ($479; P < 0.001), the net impact on total HIV-related healthcare costs between DTG/3TC and DTG/ABC/3TC was limited ($71 difference; P = 0.713).

Treatment-Naive and Treatment-Experienced Individuals

Unadjusted HIV-related total healthcare costs PPPM were significantly lower with DTG/3TC compared with BIC/FTC/TAF ($830 difference; P < 0.001) and DRV/COBI/FTC/TAF ($651 difference; P = 0.016) in treatment-naive individuals and remained significantly lower in adjusted analyses (Fig. 4A, B). HIV-related pharmacy costs were significantly lower in the DTG/3TC cohort compared with the BIC/FTC/TAF, EVG/COBI/FTC/TAF, and DRV/COBI/FTC/TAF cohorts in unadjusted analyses; in adjusted analyses, HIV-related pharmacy costs remained significantly lower with DTG/3TC compared with BIC/FTC/TAF and DRV/COBI/FTC/TAF. HIV-related medical costs were generally similar across cohorts in the treatment-naive subgroup.

Fig. 4
figure 4

a Unadjusted HIV-related healthcare costs by single-tablet regimen cohort and b adjusted HIV-related healthcare cost differences for DTG/3TC versus other single-tablet regimens for treatment-naive individuals. ABC abacavir, BIC bictegravir, COBI cobicistat, DRV darunavir, DTG dolutegravir, EVG elvitegravir, FTC emtricitabine, PPPM per patient per month, TAF tenofovir alafenamide, 3TC lamivudine. *P < 0.05 and **P < 0.001 when compared with DTG/3TC for the corresponding cost domain (i.e., comparisons within HIV-related total healthcare, pharmacy, and medical costs)

Full size image

Among treatment-experienced individuals, only DRV/COBI/FTC/TAF was significantly more costly than DTG/3TC in unadjusted analyses of total healthcare costs ($724 difference; P < 0.001), despite unadjusted HIV-related pharmacy costs being significantly higher for all comparator cohorts (Fig. 5A) and only the EVG/COBI/FTC/TAF cohort having significantly lower medical costs ($282 difference; P = 0.048). After adjustment, treatment-experienced subgroups in the BIC/FTC/TAF, EVG/COBI/FTC/TAF, and DRV/COBI/FTC/TAF cohorts all had significantly higher HIV-related pharmacy costs than the DTG/3TC cohort (Fig. 5B). The difference in HIV-related medical costs between the DTG/3TC and EVG/COBI/FTC/TAF cohorts did not remain after adjustment ($58 difference; P = 0.733).

Fig. 5
figure 5

a Unadjusted HIV-related healthcare costs by single-tablet regimen cohort and b adjusted HIV-related healthcare cost differences for DTG/3TC versus other single-tablet regimens for treatment-experienced individuals. ABC abacavir, BIC bictegravir, COBI cobicistat, DRV darunavir, DTG dolutegravir, EVG elvitegravir, FTC emtricitabine, PPPM per patient per month, TAF tenofovir alafenamide, 3TC lamivudine. *P < 0.05 and **P < 0.001 when compared with DTG/3TC for the corresponding cost domain (i.e., comparisons within HIV-related total healthcare, pharmacy, and medical costs)

Full size image

Unlike the other cohorts, there was no marked difference in HIV-related total healthcare costs between DTG/3TC and DTG/ABC/3TC in either treatment-naive or treatment-experienced subgroups in unadjusted or adjusted analyses. However, while the treatment-naive subgroup had numerically similar pharmacy and medical costs, the treatment-experienced subgroup had significant differences in HIV-related pharmacy costs ($547 adjusted difference; P < 0.001) and medical costs ($604 adjusted difference; P = 0.016), the former favoring DTG/3TC and the latter DTG/ABC/3TC.

Healthcare Utilization by Treatment History

All-cause healthcare utilization demonstrated similar rates of hospitalizations and emergency department visits across all cohorts regardless of treatment history (Table 2). All-cause hospitalization rates were between 3 and 6%, while emergency department visits were between 12 and 26%. At least half of individuals had at least 1 office visit with a specialist (eg, infectious disease experts, allergists/immunologists, hematologists, endocrinologists). Laboratory visits were slightly more frequent among treatment-experienced individuals (29–40%) compared with those who were treatment-naive (18%-25%).

Table 2 Healthcare utilization by type and treatment history among people living with HIV-1 with ≥ 1 visit
Full size table

HIV-related healthcare utilization also revealed similar rates of hospitalizations (between 3 and 5%) and emergency department visits (between 7 and 13%) across all cohorts and by treatment experience (Table 2). Slightly fewer treatment-experienced individuals in the DTG/3TC cohort had HIV-related office visits with general practitioners or specialists compared with the other treatment-experienced cohorts. As with all-cause laboratory visits, HIV-related laboratory visits were also slightly more frequent among treatment-experienced individuals than those who were treatment-naive.

Discussion

For people with HIV-1, lifelong pharmacotherapy can be costly to both the individual and the US healthcare system. As such, treatment simplification with single-tablet formulations is favorable, as they are associated with lower costs compared with those composed of ≥ 2 separate tablets and can also facilitate treatment adherence, which can potentially further factor into cost savings. The present study found that single-tablet formulations of the 2-drug regimen DTG/3TC can lead to lower costs compared with other traditional single-tablet regimens. Up to approximately $1000 USD PPPM in all-cause pharmacy cost savings (unadjusted difference relative to DRV/COBI/FTC/TAF) and approximately $660 USD PPPM in total all-cause healthcare costs (unadjusted difference relative to DRV/COBI/FTC/TAF) were observed. Significant cost savings remained even after adjusting for baseline covariates, and patterns observed in the overall analysis were generally consistent in treatment-naive and treatment-experienced individuals. Together, these data demonstrate cost savings associated with DTG/3TC, suggesting that this regimen can be a beneficial option from an economic point of view for people living with HIV.

In all comparisons evaluating the overall cohorts, DTG/3TC provided significantly lower pharmacy costs than comparator regimens. Based on adjusted PPPM HIV-related pharmacy costs, savings could potentially range from $5748 to $14,424 per patient per year, depending on the comparator regimen. Significantly lower costs with DTG/3TC were also evident in most subgroup analyses, with the exceptions of HIV-related pharmacy costs for treatment-naive individuals using DTG/ABC/3TC or EVG/COBI/FTC/TAF. The wholesale acquisition cost of DTG/3TC ($2527) is less than that of DTG/ABC/3TC ($3182) and EVG/COBI/FTC/TAF ($3394), so the underlying reason for the equivalent pharmacy costs is not immediately clear [13].

While pharmacy costs strongly favored DTG/3TC, all-cause and HIV-related medical costs were more variable. For the BIC/FTC/TAF, EVG/COBI/FTC/TAF, and DRV/COBI/FTC/TAF cohorts, this variability resulted in non-significant differences after adjusting for baseline covariates. However, the DTG/ABC/3TC cohort had significantly lower HIV-related medical costs such that the net economic benefit of DTG/3TC in the adjusted analysis amounted to only $71. Despite the variability in medical costs, healthcare utilization rates for hospitalizations and emergency department visits were similar across all cohorts, including the DTG/3TC and DTG/ABC/3TC cohorts.

In general, the results from this analysis are consistent with cost-minimization and cost-effectiveness studies assessing DTG/3TC [14,15,16,17], with the caveat that these studies preferentially considered only treatment-naive populations while this study included primarily treatment-experienced individuals. In a 2021 analysis set in the United States with a lifetime horizon, DTG/3TC resulted in cost savings versus DTG/ABC/3TC, DRV/COBI/FTC/TAF, and BIC/FTC/TAF in all simulations, and also increased quality-adjusted life-years in 87% to 88% of tested scenarios [16]. Similar results were also seen when DTG/3TC was compared with BIC/FTC/TAF and DRV/COBI/FTC/TAF in a test-and-treat model in the United States [17]; with DTG/ABC/3TC, BIC/FTC/TAF, and EVG/COBI/FTC/TAF in Canada [14]; and with DTG/ABC/3TC, BIC/FTC/TAF, and EVG/COBI/FTC/TAF in Australia [15]. In alignment with the data presented here, the primary reason for DTG/3TC cost-effectiveness in these studies was lower drug acquisition costs [14,15,16,17]. Altogether, results from these analyses and the real-world observations presented here support DTG/3TC as a highly cost-saving option with similar efficacy proven in clinical trials to alternative single-tablet 3-drug regimens.

This study has limitations to consider. Although this study compared single-tablet regimens, both single- and multi-tablet regimens are recommended by the 2023 US Department of Health and Human Services guidelines. Yet, the guidelines also indicate that regimen simplification via reduction of pill burden and/or dosing frequency is an acceptable reason for regimen optimization in the setting of maintaining virologic suppression. Most claims data were from men in the southern region of the US who were treatment-experienced, which may limit the generalizability of results beyond this population and region. As expected with retrospective analyses of claims data, there are limitations to interpreting results from this study, including the inability to verify whether individuals consumed medication as prescribed, to correlate economic results with disease status at the time of presentation for care or with clinical outcomes, any channeling bias on why a particular regimen was prescribed, or to determine whether potential rebates impact these results with individual payers. Moreover, although multivariable adjustment on baseline characteristics was used to account for potential differences between patient cohorts, confounding adjustments can only account for factors that are observable and recorded in the claims data. Additionally, the IQVIA database primarily includes data from individuals with commercial insurance, limiting generalizability of results to those with public health insurance coverage (i.e., Medicaid). As individuals with HIV-1 and Medicaid may face disproportionate socioeconomic challenges, which can factor into low adherence rates and more comorbidities, further research to evaluate economic benefits in this population is warranted. Lastly, the observation period of this study spanned the first year since DTG/3TC received FDA approval, and therefore represents the experiences of early adopters of DTG/3TC. It is possible that a longer observation period could alter the extent of cost differences observed between cohorts in either direction. Nonetheless, consistent with the results from this study, a separate 5-year budget impact analysis evaluating DTG/3TC in the US healthcare setting projected considerable healthcare savings ranging from $1.1 to $3.4 billion USD based on the lowest and highest uptake scenarios, respectively [16].

Conclusion

This retrospective analysis of commercially insured individuals from a large US claims database found that all-cause and HIV-related total healthcare costs among adults with HIV-1 receiving DTG/3TC were significantly lower PPPM compared with other current standard 3- or 4-drug single-tablet regimens, including BIC/FTC/TAF, EVG/COBI/FTC/TAF, and DRV/COBI/FTC/TAF, and comparable to those receiving DTG/ABC/3TC. These results collectively highlight the economic benefits of the 2-drug regimen DTG/3TC for the treatment of HIV-1.