Abstract
Background
The use of dual antiretroviral therapy (ART) regimens for treatment of HIV is increasing. The contemporary combination of boosted darunavir with dolutegravir has not been widely studied.
Methods
This was a retrospective cohort study that evaluated treatment-experienced individuals within three large urban clinics prescribed boosted darunavir with dolutegravir (study regimen) dual therapy. Follow-up was defined as the number of days from regimen initiation until the last HIV RNA determination on the study regimen. Virological outcomes, HIV RNA ≤50 copies/ml (undetectable), were assessed overall and by baseline HIV RNA status.
Results
Of 65 individuals included, 83% were at least 3-class antiretroviral-experienced and median time since starting ART was 19 years (IQR 13–22). Median follow-up was 419 days (IQR 286–744). An undetectable HIV RNA was achieved by 62/65 (95%) individuals at any time point on the study regimen. At the end of follow-up 61/65 (94%) individuals remained undetectable, including 48/49 (98%) with an undetectable HIV RNA at baseline (those changing for optimization) and 13/16 (81%) with viraemia at baseline (those changing therapy during virological failure). At the end of follow-up, 55 (85%) individuals were still taking the study regimen. No individuals stopped therapy due to virological failure or intolerance.
Conclusions
In a highly treatment-experienced cohort, boosted darunavir with dolutegravir dual therapy demonstrated high rates of virological success, even in those with detectable HIV RNA prior to initiating the study regimen. Further study of this potent, simple, high-barrier dual-class regimen is warranted.
Introduction
The potency and durability of dual-therapy antiretroviral regimens for both treatment-naive and treatment-experienced individuals living with HIV have been increasingly investigated [1–4]. Dual antiretroviral therapy (ART) has the potential to reduce regimen complexity, lessen costs associated with life-long ART, and decrease long-term toxicity [5]. The US Department of Health and Human Services (DHHS) adult and adolescent ART treatment guidelines include the dual-therapy regimen of a boosted protease inhibitor (PI) plus an integrase strand transfer inhibitor (INSTI) as an option for individuals with virological failure and in individuals with prior ART experience who are optimizing therapy while virally suppressed. Use of this dual-therapy regimen in treatment-naive individuals is currently only recommended in cases in which abacavir or tenofovir cannot be used or is not optimal [6]. There are relatively few published studies of this combination. In particular, boosted darunavir with dolutegravir offers a simple, potent combination with a high genetic barrier to resistance and low side effect/toxicity profile [7–10]. Additionally, these medications do not need to be renally dosed which allows for convenient dosing in chronic kidney disease (CKD), including end-stage renal disease (ESRD) [11,12].
As part of routine quality assurance in our health systems, we sought to verify that viral suppression rates on boosted darunavir with dolutegravir were similar to other optimization and salvage regimens. The objectives of this article were to describe a real-world cohort of treatment-experienced individuals on boosted darunavir with dolutegravir, reasons for being switched to this regimen and to evaluate treatment outcomes.
Methods
Study population
This retrospective cohort study evaluated data from 1 January 2013 through 31 December 2017. All individuals aged ≥18 years who received ritonavir- or cobicistat-boosted darunavir with dolutegravir dual therapy (study regimen) within three large HIV clinics in the Denver, Colorado metropolitan area were identified by electronic medical record (EMR) query. All individuals identified were verified by chart review. Individuals were excluded if they were treatment-naive, had no follow-up HIV RNA while on the study regimen, or the last recorded HIV RNA was within 8 weeks of study regimen initiation. Individuals on once- or twice-daily dosing were included. This study was approved by the Colorado Multiple Institutional Review Board.
Data
Age at regimen initiation, gender and race/ethnicity were collected. Individual chart review was used to determine ART class experience, total prior ART duration, ART regimen immediately prior to the study regimen, pre-existing antiretroviral resistance associated mutations (RAMs) when available, and the reason the individual was switched to the study regimen. The following categories were assigned: optimization with proven or suspected tenofovir (TFV) resistance, TFV toxicity or intolerance with need or desire to avoid abacavir (ABC) and CKD (including ESRD) with need or desire to avoid ABC. Optimization included switches to more contemporary regimens and individuals wishing to reduce pill burden. Changes in pill count, boosting agent used and dosing frequency of the study regimen were also recorded.
Pharmacy refill data was used as a measure of adherence to the study regimen. Refill data was obtained by chart review and was available for those individuals that filled at the clinic-associated pharmacies (the majority of individuals). Refill adherence was measured over a 6-month period prior to the last HIV RNA level on the study regimen or the duration of time the individual was on the study regimen if less than 6 months. Adherence was calculated using the medication possession ratio (the number of days’ supply dispensed divided by the total number of days adherence was measured) [13].
Immunological and virological treatment outcomes on the study regimen were measured. Change in CD4+T-lymphocyte count from the study regimen start to last follow-up on the study regimen was calculated. Virological treatment success was measured in multiple ways. The ability of the individual to achieve or maintain an HIV RNA level of ≤50 copies/ml (undetectable) any time on the study regimen and at the last HIV RNA determination was assessed. Further, this same virological outcome was assessed based on baseline HIV RNA level at the start of the study regimen: baseline HIV undetectable (those changing therapy for optimization with baseline HIV RNA ≤50 copies/ml) versus baseline detectable (those changing therapy during virological failure with baseline HIV RNA >50 copies/ml). Additionally, if the study regimen was stopped prior to the end of the follow-up period, the reason for stopping was recorded.
Lastly, serum creatinine closest to but not after study regimen initiation and last serum creatinine measurement prior to end of follow-up or the study regimen termination was recorded. Individuals were excluded from creatinine measurements if they were on haemodialysis at baseline.
Statistical analysis
Basic descriptive statistics were utilized to describe the included population, prior ART exposure, presence of RAMs, the reason the individual was switched to the study regimen, regimen characteristics, treatment outcomes and laboratory data where applicable. χ2 test was employed to determine associations between virological outcome data. Microsoft Excel 2013 and GraphPad, Prism 8 were used to perform all analyses.
Results
Description of cohort on boosted darunavir with dolutegravir (study regimen)
Of approximately 3,600 individuals receiving care within the three clinics, 73 individuals treated with the dual antiretroviral therapy regimen of boosted darunavir with dolutegravir were identified. Three individuals were excluded due to having no virological outcome data, three individuals for having the last recorded HIV RNA within 8 weeks of study regimen initiation, and two individuals were excluded for being treatment-naive. Of those included (n=65), 14% were female (n=8) or male-to-female transgender (n=1). Median age of the cohort was 55 years (interquartile range [IQR] 51-61). As shown in Table 1, 48% of individuals identified as non-Latino White, 28% as Latino and 23% as non-Latino Black. At baseline, 49 (75%) individuals had an undetectable HIV RNA (≤50 copies/ml), thus 16 (25%) individuals had detectable HIV RNA on their prior regimen (Table 1).
Characteristics of individuals treated with the dual-class regimen of boosted darunavir and dolutegravir (n=65)
Baseline undetectable equals HIV RNA ≤50 copies/ml at study regimen initiation.
Baseline detectable equals HIV RNA >50 copies/ml at the study regimen initiation.
At end of follow-up, 55/65 (85%) individuals remained on the study regimen. ABC, abacavir; bDRV, boosted darunavir; CKD, chronic kidney disease; DTG, dolutegravir; ESRD, end stage renal disease; RAMs, resistance associated mutations; TFV, tenofovir.
Median duration since HIV diagnosis was 22 years (IQR 16-28). Median duration of prior ART use was 19 years (IQR 13-22). Of the 65 individuals, 100% had prior exposure to nucleoside(tide) reverse transcriptase inhibitors (NRTI), 85% to non-nucleoside reverse transcriptase inhibitors (NNRTI), 89% to protease inhibitors (PI) and 57% to integrase strand transfer inhibitors (INSTI). Overall, 83% had prior 3-class antiretroviral experience and 48% had prior 4-class antiretroviral experience. Immediately prior to switching to the study regimen, 9 (14%) were on NNRTI-based therapy, 14 (22%) on PI-based, 6 (9%) on INSTI-based, 9 (14%) on other dual therapy and 28 (43%) on more complex ≥3 class regimens. Among the 65 individuals, 42 (65%) had genotypic or phenotypic protease resistance data available for review at baseline. Pre-existing major RAMs to darunavir (L33F, I47V, I54L, T74P and I84V) were present in five (8%) individuals (Table 1). No one had integrase resistance testing available prior to study regimen initiation.
Reasons for switching to study regimen
Optimization in the setting of previously proven or suspected TFV resistance accounted for 46% of switches, TFV toxicity or intolerance with need or desire to avoid ABC accounted for 34% of switches, and CKD (including ESRD) with need or desire to avoid ABC accounted for 20% of switches (Table 1).
Study regimen characteristics
After switching to the study regimen, average ART regimen pill count decreased from 5 to 3 pills/day. Near equal proportions of the cohort were on cobicistat versus ritonavir as a boosting agent and the majority (59 [91%]) were on daily dosing of both components of the study regimen (Table 1).
Adherence
Adherence was measurable in 50 individuals (Table 1). Median adherence was 99% (IQR 91–100%). When stratified into those with baseline undetectable HIV RNA prior to starting the study regimen for whom adherence data was available (n=36) median adherence was 99% (IQR 92–100%) compared with those with baseline detectable HIV RNA (n=14) where median adherence was 97% (IQR 80–100%).
Virological and immunological outcomes on the study regimen
As detailed in Table 2, median follow-up duration on the study regimen was 419 days (IQR 286–744). Of the 65 individuals, 62 (95%) achieved an undetectable HIV RNA (≤50 copies/ml) at some time point on the study regimen. We also assessed achievement of an undetectable HIV RNA based on whether or not the individuals were undetectable at baseline: 49/49 (100%) of baseline undetectable and 13/16 (81%) of baseline detectable individuals achieved an HIV RNA ≤50 copies/ml at some time point on the study regimen.
Virological outcomes of 65 individuals treated with boosted darunavir with dolutegravir
Refers to achieving an undetectable HIV RNA at any point while on dolutegravir/boosted darunavir (DTG/bDRV).
Last refers to the last recorded HIV RNA value while on DTG/bDRV.
Baseline undetectable defined as HIV RNA ≤50 copies/ml at study regimen initiation.
Baseline detectable defined as HIV RNA >50 copies/ml at study regimen initiation.
Next we assessed HIV RNA at the end of follow-up or termination of the study regimen. At the last recorded HIV RNA while on the study regimen, 61/65 (94%) had an HIV RNA ≤50 copies/ml. Among those individuals who were baseline undetectable, 48/49 (98%) had an HIV RNA ≤50 copies/ml. Among those who had a detectable HIV RNA at baseline, 13/16 (81%) had an HIV RNA ≤50 copies/ml at the last recorded HIV RNA. Those that were baseline undetectable were more likely to have an HIV RNA of ≤50 copies/ml at the last recorded HIV RNA level compared with those with detectable HIV RNA at baseline (98% versus 81%, respectively; P=0.02). Figure 1 details viral suppression rates at different time points on the study regimen.

Percentage of individuals treated with dual therapy boosted darunavir with dolutegravir with HIV RNA <50 copies/ml over time based on baseline HIV RNA
Of the four individuals who had a detectable HIV RNA at the last recorded HIV RNA while on the study regimen, three of those individuals had detectable HIV RNA at baseline. One individual had a recorded adherence of 16%, never had an undetectable HIV RNA on the study regimen and remained on the study regimen at the end of follow-up. The second individual did not have adherence data available but had periods of both viraemia and viral control while on the study regimen and remained on the study regimen at the end of follow-up. The third individual had a calculated adherence of 90% and, similar to the second person, had periods of viraemia and viral control over time but was switched off the study regimen due to a drug–drug interaction (DDI) with ritonavir. All of these individuals were on the study regimen for >68 weeks. None of these individuals had genotypic resistance testing while on the study regimen or after being switched off the study regimen. The one individual who had an undetectable HIV RNA at baseline but not at the last HIV RNA determination had an undetectable HIV RNA throughout the study period but developed an HIV RNA of 56 copies/ml at week 45 and an HIV RNA of 112 copies/ml at week 59. Lamivudine was added at week 59 without any resistance testing at that time.
Median baseline CD4+ T-cell count rose from 527 (IQR 340–767) to 585 (IQR 470–649) cells/mm3 at the end of follow-up. There was no difference in the number of individuals with HIV RNA of ≤50 copies/ml at the last recorded HIV RNA level in those on cobicistat versus ritonavir as the boosting agent (97% versus 91%; P>0.05), with twice daily dosing of meds versus daily (100% versus 93%; P>0.05) and those with pre-existing darunavir RAMs versus none (80% versus 95%; P>0.05). Five individuals had ESRD and were on haemodialysis for the duration of the follow-up period and all five individuals were undetectable at the last HIV RNA determination on the study regimen. Of the five individuals with pre-existing darunavir RAMs, all five were undetectable at baseline and 4/5 (80%) were undetectable at the last recorded HIV RNA. One individual with a pre-existing I54L did develop an HIV-RNA of 112 copies/ml at week 59 and lamivudine was added, as described above. Of the individuals who were baseline undetectable and then developed viraemia (HIV RNA >50 copies/ml) while on the study regimen, 7/7 (100%) re-suppressed without any changes to ART.
Study disposition
At the end of follow-up, 55 (85%) individuals remained on the study regimen. Of the 10 (15%) individuals who were switched off the study regimen, 5 (50%) were due to DDIs with a boosting agent, 2 (20%) due to cardiovascular disease risk and desire to avoid darunavir, 1 (10%) due to low-level viraemia at 112 copies/ml (detailed above), 1 (10%) participant was lost to follow-up and 1 (10%) participant was transitioned to hospice care for non-HIV-related reasons. Of 10 individuals switched off the study regimen, 8 were undetectable when switched off the study regimen. In addition to the individual switched due to low level viraemia, one individual had an HIV RNA of 550 copies/ml at time of switch. This individual was in the baseline detectable group and never achieved an undetectable HIV RNA on the study regimen (detailed under virological outcomes section). This individual was switched due to a DDI.
Study regimen tolerability
No one switched off this regimen due to lack of tolerability. Change in serum creatinine was calculated in 59 individuals. Median change in serum creatinine from study regimen initiation to last creatinine on the study regimen was a gain in serum creatinine of 0.09 mg/dl (IQR −0.12–0.17).
Discussion
In a retrospective review of individuals switching to a regimen of boosted darunavir with dolutegravir there were high rates of viral control, even in individuals with viraemia prior to initiation of the study regimen. Of 65 individuals included in the study, after a median follow-up of 419 days, 95% of individuals were able to achieve an HIV RNA ≤50 copies/ml at some point on the study regimen. Among the 49 individuals with an undetectable HIV RNA prior to switching to the study regimen, 98% were undetectable at the end of follow-up. Of the 16 individuals with detectable viraemia prior to initiation of the study regimen, 81% were undetectable at the end of follow-up. There was no difference in virological outcomes by use of boosting agent, daily versus twice daily dosing of regimen components, presence of ESRD at baseline, or based on the presence of pre-existing darunavir RAMs. No one stopped the regimen due to virological failure or tolerability issues.
Despite recent interest surrounding the use of dual-therapy ART regimens, the combination of a boosted PI plus INSTI has not been widely studied. Prospective studies evaluating a switch to a boosted PI in combination with an INSTI have taken place in resource limited settings and compared ritonavir-boosted lopinavir with raltegravir to a standard 3-drug regimen consisting of 2 NRTIs and boosted lopinavir [14–16]. These studies demonstrated that boosted lopinavir with raltegravir was non-inferior to a standard regimen after first failure.
More recently, an observational study evaluated a switch to ritonavir-boosted darunavir with dolutegravir in 130 individuals receiving care in Italy. This study showed that viral suppression among individuals increased from 60% at baseline to 91% after 48 weeks on the study regimen. Additionally, the regimen was concluded to be safe with minimal metabolic impact [17]. Similarly, a study of 31 treatment-experienced individuals in Korea demonstrated high success rates with a switch to cobicistat-boosted darunavir with dolutegravir. Of 18 individuals who were baseline suppressed, 100% maintained viral suppression and of 13 individuals who were baseline non-suppressed, 11 reached viral suppression [18]. A smaller study evaluated a switch from complex salvage regimens to ritonavir-boosted darunavir with dolutegravir among 13 individuals with transmitted NRTI resistance and found that all 13 individuals tolerated the regimen with continued viral suppression after 12 months [19]. There is an ongoing prospective, randomized trial (NCT02486133) to assess safety, tolerability, and efficacy of a switch to boosted darunavir with dolutegravir in individuals with sustained virological suppression on ‘standard of care’ ART. A sub-analysis of this study has demonstrated adequate plasma concentrations of both study regimen components (darunavir and dolutegravir) following daily dosing of the regimen [20].
Studies evaluating the use of a boosted PI with an INSTI in treatment-naive individuals are even more limited. The ACTG A5262 trial was a single arm study evaluating boosted darunavir with raltegravir. The rate of virological failure at week 48 on therapy was relatively high at 24% but there was no comparison group in this study [21]. The French National Agency for AIDS Research (ANRS) 143 trials compared standard therapy (tenofovir disoproxil fumarate/emtricitabine and boosted darunavir) to a dual therapy PI/INSTI regimen of raltegravir and boosted darunavir. The boosted PI/INSTI regimen was non-inferior to the boosted PI/NRTI regimen though success was limited when baseline viral load was greater than 100,000 copies/ml or the CD4+ T-cell count was <200 cells/mm3 [22].
This study adds to the growing body of evidence that boosted darunavir in combination with dolutegravir offers a simple, potent and efficacious alternative ART regimen in people living with HIV with prior treatment experience, including individuals with ESRD. In theory, boosted darunavir with dolutegravir could be considered after failure with a standard initial regimen of 2NRTIs + NNRTI without the need for resistance testing. This could have implications particularly in resource-limited settings although this would need to be further studied in clinical trials prior to widespread implementation.
Reassuringly, in this study, 98% of individuals who were undetectable at baseline maintained an undetectable HIV RNA at the last recorded HIV RNA while on the study regimen. Success of this regimen in this study was likely tied to the high adherence rates within the cohort (median 6-month adherence of 99%). Virological outcomes were very good but not quite as strong for those with detectable HIV RNA at baseline, 13/16 (81%) individuals who had detectable HIV RNA at baseline were undetectable at the end of follow-up. Viraemia at baseline is likely a surrogate marker of poor adherence and it is not surprising that, as a group, those with baseline viraemia demonstrated lower rates of success. Indeed, a thorough review of the four individuals who had detectable HIV RNA at end of follow-up revealed that, in three of the four individuals, adherence may have been the cause of failure rather than regimen durability. These individuals had periods with and without viral control suggesting adherence rather than failure with treatment emergent resistance was likely responsible for the periods of viraemia.
While the retrospective nature of the paper does not allow for formal tolerability or safety assessments, only 15% of individuals were switched off the study regimen during the follow-up period and none of these were due to virological failure or intolerance of the regimen. There was only a slight increase in serum creatinine (median 0.09 mg/dl) when switching to the study regimen. This is consistent with the impact of dolutegravir on inhibition of tubular creatinine secretion rather than a true reduction in glomerular filtration rate, though many factors could be contributing to creatinine fluctuations in this uncontrolled, retrospective study [23]. Overall, the effect of this regimen on serum creatinine is not likely to be clinically significant.
Strengths of this study include the assessment of real-world outcomes regarding the use of a regimen not widely reported in the literature. The demographics of the cohort mirror the demographics of people living with HIV in Denver, CO and make the results translatable to the population we treat. There are several limitations to this study. A retrospective observational study has inherent limitations and relies on the accuracy and completeness of prior records. Total duration of prior ART, class experience, and extent of underlying cumulative genotypic resistance were likely underestimated given inability to completely review older records either due to individuals being in care elsewhere or incomplete records. Further, due to the low frequency in which this regimen is prescribed, sample sizes were limited.
In conclusion, the dual therapy regimen of boosted darunavir with dolutegravir appeared to be an efficacious and well-tolerated regimen among individuals with prior treatment experience. The regimen was most successful among those with an undetectable HIV RNA at baseline. However, the majority of individuals with baseline viraemia also had success on this regimen. Further study of this potent, well-tolerated, high-barrier regimen is warranted.
Footnotes
Acknowledgements
This work was supported in part by the University of Colorado Health Services Research Development Grant awarded to KLH.
SER receives research support from Gilead Sciences. The remaining authors declare no competing interests.
