Clinical Pharmacology at CROI 2021.
New Drugs & Formulations; PrEP; HIV Treatment
Courtney V. Fletcher, Pharm.D. |
UNMC Center for Drug Discovery
University of Nebraska Medical Center
986000 Nebraska Medical Center
Omaha, NE 68198
The 2021 (28th) Conference on Retroviruses and Opportunistic Infections (CROI) was the 2nd Virtual CROI because of the COVID-19 pandemic. In this report I will highlight abstracts focused on pharmacologic issues that are of broad interest or might benefit from some expert clarification. Abstracts will be discussed in the categories of: (i) the therapy of HIV infection, (ii) PrEP, (iii) pharmacokinetics, pharmacodynamics and pharmacogenomics, and (iv) new drugs and formulations. You can find more information on these abstracts on the CROI website and many are covered in depth elsewhere on the NATAP http://natap.org website.
I. The Pharmacotherapy of HIV Infection
NADIA - DTG or DRV/r both with dual NRTIs for second line failure (94LB).
The NADIA trial was designed to address 2 questions: first, whether DTG was non-inferior to DRV/r in second line therapy and second, whether TDF/FTC was non-inferior to ZDV/3TC. 464 PLWH with treatment failure were enrolled at 7 sites in Uganda, Kenya and Zimbabwe. Participants were randomized to DTG or DRV/r (800/100 mg once daily) and randomized to TDF/3TC or ZDV/3TC. 90.2% of DTG and 91.7% of DRV/r achieved the primary outcome of plasma HIV-RNA <400 copies/mL at week 48. Rates of confirmed virologic rebound (>1000 copies/mL) were 6% and 5.7% for DTG and DRV/r, respectively. Virologic rebound with confirmed ≥ 1 major resistance mutation to DTG and DRV were n=4 and n=0 for DTG and DRV/r, respectively. For the comparison of TDF vs ZDV, 92.3% of TDF recipients and 89.6% of ZDV recipients achieved plasma HIV-RNA < 400 copies/mL at week 48. Thus, DTG was non-inferior to DRV/r and TDF was non-inferior to ZDV.
These data support recommendations for DTG in second line therapy and for programmatic switches from EFV- to DTG-based regimens. The emergence of DTG resistance will need close monitoring. As for TDF vs. ZDV, the outcome was surprising - you would have thought TDF would have demonstrated superiority and not just non-inferiority to ZDV. While it seems preferable to use TDF over ZDV, the final chapter of the history of ZDV hasn't quite been written.
ATLAS-2M: every 8-weeks long-acting CAB and RPV remains non-inferior to every 4-weeks at 96-weeks (abstract 401).
ATLAS-2M is evaluating the safety and efficacy of every 8-weeks (q8w) vs. every 4-weeks (q4w) LA-CAB and RPV. At CROI 2020, 48-week results were presented: 94.3% of q8w and 93.5% of q4w participants had HIV-RNA <50 copies/mL, meeting non-inferiority. Virologic non-response rates (HIV-RNA ≥ 50 copies/mL at week 48) were 1.7% (n=9) vs. 1% (n=5) for q8w and q4w, respectively. Adverse reactions, including injection site reactions (ISR) were similar between the arms, though interestingly, the q8w ISR rate was 30% vs. 20% for q4w.
At CROI 2021, 96-week results were presented. Q8w remained non-inferior to q4w: 91% of q8w and 90.2% of q4w participants had HIV-RNA <50 copies/mL. There were 9 confirmed virologic failures in the q8w arm and 2 in the q4w arm. ISRs remained more commonly reported in q8w recipients: 26.5% vs. 17.4% for q4w.
Important take home points. Virologic suppression with both dosing regimens was durable to 96-weeks and q8w is non-inferior to q4w. No new adverse reactions were found. ISR were common with both dosing regimens though more frequent with q8w but did appear to decrease over the study duration with both dosing regimens. And the not so good news: confirmed virologic failure was more common with q8w dosing: 1.7% vs 0.4% for q8w vs q4w.
My interpretations. There is no doubt that LA-CAB and RPV is an important advance for treatment particularly because if offers another choice and convenience. It is important to remember, however, that q4w dosing even with perfect adherence (as in the ATLAS and FLAIR studies) wasn't superior to oral ART. There remains a lot we need to learn about the effectiveness of LA-CAB and RPV. Two particular needs are:
First, adherence. What are the consequences and optimal strategies for management of late doses or missed doses, life events such as traveling particularly out of the country, or the prolonged PK tail after treatment discontinuation. Two PK modeling abstracts at CROI 2021 provided some useful information (see abstracts #373 and #403). But more work and real-life data are needed.
Second, why does LA-CAB and RPV therapy fail in some persons even with perfect adherence? While rates of confirmed virologic failure are low, they nonetheless happen and are more than twice as likely with q8w compared with q4w dosing. These might be potential factors. Intrinsic: biologic sex, age, weight and body composition, co-infections and inflammation, and pharmacogenomic. Extrinsic: lead-in regimen, maintenance dosing regimen and pharmacokinetics, administration technique, pharmacodynamic, concomitant medications, and adherence. Some information is now emerging, showing that BMI ≥ 30kg/m2 , in addition to at least one other factor, is associated with an increased risk of confirmed virologic failure in ATLAS and FLAIR (see: https://pubmed.ncbi.nlm.nih.gov/33730748/ ). We really need to understand the reasons for these virologic failures - it could have important implications not only for preventing failures with LA-CAB and RPV, but for the development of other ARVs.
My own perspective on q4w or q8w dosing. The primary objective is to sustain virologic suppression. Though both dosing regimens are highly effective, the risk of virologic failure is higher with q8w, and unless lifestyle or other considerations preclude q4w dosing, I'd recommend q4w.
DTG has superior virologic efficacy in pregnant women compared with EFV, and TAF/FTC had fewer adverse pregnancy outcomes in pregnant women who started therapy at 14-28 weeks gestation (DolPHIN-2, abstract 175) and IMPAACT2010, abstract 177).
DolPHIN-2 randomized pregnant women in the third trimester to DTG or EFV (with dual NRTIs). At CROI 2021, investigators reported final 72 week post-partum outcomes of mothers and their infants. DTG remained well tolerated: maternal drug-related adverse events were DTG, 2.2% vs EFV, 3.8%. The primary results, women at the first post-partum visit with viral loads <50 copies/mL were presented at CROI 2019 and published last year: 74% of the DTG group had viral loads <50 copies/mL, compared with 43% in the EFV group (risk ratio 1⋅64, 95% CI 1⋅31-2⋅06). See: DolPHIN-2, Lancet 2020, https://pubmed.ncbi.nlm.nih.gov/32386721/ .
IMPAACT 2010 randomized ART-naïve, pregnant women between 14-28 weeks gestation to one of 3 treatment arms: DTG+TAF/FTC, DTG+TDF/FTC and EFV+TDF/FTC. Just over 200 women were enrolled in each arm. The primary virologic endpoint, reported at CROI 2020, was proportion with HIV-RNA <200 copies/mL at delivery. This study found the combined DTG arms had superior virologic efficacy vs EFV: 97.5% of DTG recipients vs. 91% of EFV had HIV-RNA <200 copies/mL. The DTG arms also had a shorter time to viral suppression. At CROI 2021, investigators reported results through 50 weeks post-partum. 96.3% in combined DTG arms and 96.4% in EFV arm had HIV-RNA <200 copies/mL. Maternal virologic failure rates were: 4.1%, DTG+FTC/TAF; 5.1%, DTG+FTC/TDF; and 10.4%, for EFV/FTC/TDF. Rates of maternal and infant grade ≥3 adverse events were not different across the 3 regimens; however, they were numerically lower in the DTG+FTC/TAF arm. The estimated probability of infant death was higher in the EFV (6.9%) compared to DTG+FTC/TAF (1.0%, p<0.001) and DTG+FTC/TDF (2.0%, p=0.008) arms.
Collectively, these two studies demonstrating superior virologic results of DTG compared with EFV and low rates of maternal adverse events strongly support current guidelines that recommend DTG as a preferred agent for ARV-naïve pregnant women. Additionally, the IMPAACT 2010 study that found DTG+FTC/TAF had the lowest rate of adverse pregnancy outcomes, provides reassuring information on TAF use during pregnancy.
Dose-adjusted emergency contraception for management of EFV-perpetrated drug interaction (abstract 91).
Levonorgestrel (LNG) often referred to as the "morning after pill" is approved for emergency contraception. LNG concentrations are reduced with EFV co-administration. Abstract 91 described the results of A5375 that investigated a doubling of the LNG dose in 52 women taking an EFV-based regimen. Participants were randomized (1:2) to receive either the standard LNG dose of 1.5mg or the double dose of 3mg. EFV, as expected, induced (increased) LNG metabolism, as evidenced by the shorter elimination half-life of ≈12 hours compared with the usual 27 hours. Cmax is the proposed PK marker of LNG efficacy, and was increased 51% in the women who received the double dose: 15.1 ng/mL vs. 24.9 ng/mL for standard vs. double dose LNG, respectively. The Cmax of 24.9 is comparable to that expected in the absence of EFV of 18 ng/mL. These data support doubling the dose of LNG as a management strategy for the EFV-induced drug interaction. This study is important because it provides confirmation of that recommendation in the patient population who use the drugs.
II. Pre-Exposure Prophylaxis (PrEP)
Prevenir study - no difference in prevention with on demand vs. daily TDF/FTC PrEP (Abstract 148).
The Prevenir study is an ongoing open-label evaluation of daily or on demand PrEP with TDF/FTC. Participants were allowed to switch regimen. On demand dosing is one tablet of TDF/FTC <24 hours before sex and a second tablet <24h after sex. 3067 participants have been enrolled across the Paris region. 49.5% of participants chose on demand and 50.5% daily. Median follow-up for the present analysis was 22 months. 3 participants in each regimen acquired HIV infection. Investigators estimated that 361 HIV infections were avoided. Worth noting is the high incidence of sexually transmitted infections that was associated with an increase in condomless sex over the study duration. These data show no difference in HIV incidence between these 2 approaches to PrEP.
Updated analysis of HIV infections in HPTN 083, LA-CAB vs oral TDF/FTC for PrEP (abstract 153).
This is an updated analysis of HIV infections that occurred during HPTN 083, a randomized study of LA-CAB 600mg q8weeks vs daily oral TDF/FTC in cisgender men and transgender women who have sex with men. Among those who received oral TDF/FTC, there were 39 infections; 37 were associated with either sub-optimal or non-adherence based on measured TFV-DP concentrations in DBS. Among those who received LA-CAB, there were 12 infections. Most concerning were infections in four individuals who all received their doses on-time. CAB concentrations well above the protein-adjusted IC90, except between 5-10 weeks after enrollment in 3 of 4, where concentrations declined to 4x or lower than the protein-adjusted IC90.
In HPTN 083, both oral TDF/FTC and LA-CAB were highly effective in preventing HIV-infection; LA-CAB, however, was superior with a HIV incidence of 0.37 vs. 1.22 for oral TDF/FTC, and hazard ratio of 0.32. Whether those early low concentrations are associated with regimen failure is not known, but as I stated in comments above there is an urgent need to understand the reasons for this. Finally, an interesting observation is that in HPTN 084, a study of LA-CAB vs oral TDF/FTC for PrEP in cisgender women, the hazard ratio favoring LA-CAB was 0.11. Could this suggest that LA-CAB works better in cisgender women than cisgender men and transgender women?
III. Pharmacokinetics, Pharmacodynamics and Pharmacogenomics
Some brief bullets:
• Rifampin (RIF) was approved by the FDA in 1971, and the "best" dose is still uncertain. Abstract 90 described an evaluation of high-dose (35 mg/kg) and standard-dose (10 mg/kg) RIF for TB treatment in HIV-infected persons. Sputum culture conversion at week 8 was 86.2% with high-dose and 62.9% with standard dose. There did not appear to be a difference in DTG concentrations with high vs. standard dose RIF. However, EFV concentrations were lower, by about 50%, with high-dose RIF. There were 6 serious adverse events, 3 in each RIF dosing group and notably all in those receiving DTG. Further investigations of high-dose RIF seem appropriate but there are some concerning safety signals with DTG that require particularly close monitoring.
• FTC-triphosphate (FTC-TP) has a half-life of 35 hours in the dried blood spot (DBS) matrix. Abstract 92 investigated FTC-TP in DBS, as a marker of short-term adherence, to predict future viremia in 433 PLWH over 48 weeks. The adjusted odds ratio for future viremia if FTC-TP in DBS was quantifiable vs. not quantifiable (BLQ) was 3.4 (p<0.0001). Interestingly, participants who reported 100% 3-day adherence and who had a FTC-TP value that was BLQ (indicating a mismatch between reported adherence and actual adherence), had even greater odds of future viremia (6.0). This points to a particular patient population who could benefit from adherence counseling - those who report high recent adherence but who don't have measurable FTC-TP concentrations.
• TFV-diphosphate (TFV-DP) has a half-life of ≈400 hours (17 days) in DBS, and is a marker of long-term ART adherence. Abstract 93 reported an evaluation of TFV-DP in DBS to predict future viremia in 81 post-partum women who initiated TDF/FTC/EFV during pregnancy. 17% of women who had a DBS concentration >1850 (consistent with 7 doses/week) had a viral load >20 copies/mL at their next visit compared with 91% of women who had a DBS concentration <350 (consistent with <2 doses/week).
Measuring TFV-DP in DBS has been shown strongly associated with viral suppression and to predict future viremia (see also Castillo-Mancilla J, https://pubmed.ncbi.nlm.nih.gov/30942881/ and https://pubmed.ncbi.nlm.nih.gov/30137238/ ) and to predict virologic failure and resistance (see Abstract #520 from CROI 2020). Combining DBS measurement of FTC-TP and TFV-TP offers the ability to assess short-term and long-term adherence.
• The safety and PK of a dapivirine vaginal ring designed to deliver drug over 3 months vs. 1 month were presented in Abstract 147. The PK look good and safety was acceptable, and support further development.
IV. New Drugs and Formulations
I thought data on new drugs and formulations were a highlight of CROI 2021.
Islatravir (Abstracts 87, 88, 376, 416).
Islatravir (ISL) is a potent NRTI with a long plasma (≈ 60 hrs) and longer intracellular half-life (≈ 190 hrs). The IC50 is approximately 0.2 nM, approximately 8400 times more potent than tenofovir (see Antimicrob Agents Chemother 2015;59:4190. doi: 10.1128/AAC.05036-14).
In Abstract 87, investigators integrated PK data from macaques and phase 1 data from HIV-infected persons to determine a once-monthly oral ISL dose for PrEP. A 60mg oral dose was selected based on simulations that predicted it would maintain ISL-triphosphate concentrations above the target of 0.05 pmol/10^6 cells and allow some forgiveness for a delayed dose.
In Abstract 88, investigators evaluated the PK of an ISL implant for PrEP. In this phase I study, HIV negative men and women had an implant with one of 3 doses, 48mg, 52mg and 56 mg placed subdermal in the upper arm of the nondominant hand. PK data showed the 56mg implant maintained ISL-TP concentrations above the target of 0.05 pmol/10^6 cells for the entire 12-week post-placement period. After the implant was removed, ISL-TP concentrations declined with a half-life of 198 hours. PK modeling projected the 56mg implant could maintain concentrations above target for 52 weeks. The implant was well tolerated. These data support continued development of a once-yearly ISL implant for PrEP.
ISL is also under development for HIV treatment, including in combination with MK-8507. MK-8507 is a NNRTI that has a half-life of 60-70 hrs, and was shown to reduce plasma HIV-RNA for up to 7 days after a single dose (see:
http://www.hivglasgow.org/wp-content/uploads/2020/11/O416_Ankrom.pdf ). In Abstract 376, PK and viral dynamics modeling were used to explore oral ISL doses that could be combined with MK-8507. A 20mg oral dose of ISL combined with MK-8507 oral doses of 100, 200 and 400 mg was identified for further clinical development as a once weekly regimen, which now is recruiting (see: NCT04564547). http://www.croiwebcasts.org/console/player/47401?mediaType=slideVideo&
Finally, the combination of ISL with doravirine (DOR) is now in phase 3 development for HIV treatment. 96 week results of a dose ranging trial in treatment-naïve PLWH of ISL, 0.25, 0.75 and 2.25 mg once daily in combination with 100mg DOR were reported at Glasgow 2020 (see: http://www.hivglasgow.org/wp-content/uploads/2020/11/O415-HIV-DTG-ePoster-Molina.pdf ).
At 96 weeks, proportions of participants with HIV-RNA <50 copies/mL were 86.2%, 90% and 67.7% for ISL 0.25, 0.75 and 2.25mg, respectively. It was noted the lower suppression rate for the 2.25mg dose was driven by discontinuations. Abstract 416 at CROI 2021 investigated viral blips in this 96-week study. Blips were not frequent and less than the comparator regimen of DOR/TDF/3TC and no participant who had a viral blip developed virologic failure. Based on these results, collectively, the 0.75mg once daily oral dose of ISL has been selected for phase 3 evaluation. http://www.croiwebcasts.org/console/player/47448?mediaType=slideVideo&
Lenacapavir (89, 127, 315, 717).
Lenacapavir (LEN, formerly GS-6207) is a first in class, highly potent inhibitor of the HIV-1 capsid protein. HIV capsid is essential for virus replication at multiple stages of the life cycle. Safety and PK data in healthy volunteers who received single doses subcutaneously (SC) were presented at CROI 2019 and dose-response information at CROI 2020. The elimination half-life ranged from 30-40 days.
CROI2020: DOSE-RESPONSE RELATIONSHIP OF SUBCUTANEOUS LONG-ACTING HIV CAPSID INHIBITOR GS-6207 (03/16/20)
CROI2019: GS-6207, A POTENT AND SELECTIVE FIRST-IN-CLASS LONG-ACTING HIV-1 CAPSID INHIBITOR
CROI2019: SAFETY AND PK OF SUBCUTANEOUS GS-6207, A NOVEL HIV-1 CAPSID INHIBITOR (03/08/19)
The drug-drug interaction (DDI) profile of LEN was evaluated in Abstract 89. This is an excellent use (and example) of probe substrates to assess the drug interaction potential. For those of you interested in this area of work, I'd encourage you to listen to the presentation. The take home conclusions are as follows. LEN is a substrate of CYP3A, P-gp and UGT1A1. When given with inhibitors of CYP3A (voriconazole, darunavir, cobicistat) and P-gp (cobicistat), LEN concentrations increased 2.1 to 2.3-fold. These increases are presently considered not clinically significant and indicate, at least going forward, that LEN can be combined with such inhibitors without dose adjustment. However, when LEN was combined with inhibitors of UGT1A1, CYP3A and P-gp (atazanavir plus cobicistat), LEN concentrations increased 3-fold and coadministration is not recommended. When LEN is given with potent inducers of UGT1A1 and CYP3A (rifampin), LEN concentrations decreased 85% and coadministration is contraindicated. LEN was evaluated for its potential to cause DDIs. LEN increased the concentrations of midazolam 3-fold, indicating the LEN is a moderate inhibitor of CYP3A, and that it will have to be used cautiously with substrates of CYP3A. LEN did not appear to be an inhibitor of P-gp, BCRP and OATP. Finally, LEN was not affected by coadministration with famotidine (inhibitor of gastric acid secretion).
LEN in heavily ART-experienced persons (Abstract 127). This phase 2/3 trial investigated LEN or placebo added to ART regimens of persons with virologic failure (HIV-RNA > 400 copies/mL and resistance to ≥ 2 agents). LEN (or placebo) was given orally (600mg) on days 1 and 2, and 300mg on day 8. On day 15, those on oral LEN received SC LEN in the abdomen, 927mg every 6 months. Additionally, an investigator-selected optimized background regimen (OBR) was started on day 15. At day 15, 88% of LEN recipients compared with 17% of placebo had at least a 0.5 log10 decline in plasma VL; median declines were -2.0 vs -0.08 log10. During LEN + OBR, at 4 wks after SC dosing 21/36 (58%) had VL < 50 copies/mL. No serious adverse events were reported. In these highly treatment experienced persons on failing existing regimens, LEN produced substantial declines in viral load with a high proportion achieving an undetectable level. These compelling findings strongly support continued evaluation of LEN for treatment of HIV. http://www.croiwebcasts.org/console/player/48167?mediaType=slideVideo&
Phase II proof of concept (POC) trial of the maturation inhibitor GSK3640254 (Abstract 126).
Maturation inhibitors target the late stage of the HIV life cycle and are designed to prevent the correct processing of the Gag-Pol polyprotein by preventing the cleavage into component proteins. Bevirimat was an early maturation inhibitor discontinued from development because of preexisting resistance mutations. GSK'254, is a next generation maturation inhibitor that does not appear to have this limitation with activity against all HIV-1 subtypes. Phase 1 data from a single and multiple dose ranging study in healthy volunteers have been published (see: https://pubmed.ncbi.nlm.nih.gov/33200887/ ). GSK'254 was slowly absorbed reaching maximum concentrations in ≈ 6 hrs and had a mean elimination half-life of ≈ 24 hours. Abstract 126 presented results from a phase 2 POC trial, which showed HIV-RNA declined 1.5 log10 copies/mL with 140 mg and 2 log10 copies/mL with 200 mg. There were no serious short-term adverse events. These findings have allowed a phase 2b trial of GSK'254 in doses of 100, 150 and 200mg with dual NRTIs, which is underway (see: https://clinicaltrials.gov/ct2/show/NCT04493216 ).
The science of CROI 2021 was outstanding, as usual. The efforts of the organizers to sustain not just the meeting but the quality expected during the COVID-19 pandemic were extraordinary. CROI 2022 is scheduled to be in Denver, Colorado March 13-16, 2022. I look forward to being there - in person.
In Memory of John G. Gerber, M.D.
I have been writing this CROI Pharmacology Update for NATAP since 2011 (though it might be a few years earlier). I took over from John Gerber after his retirement. John established the approach that I continue to follow: to cover pharmacologic issues that are of broad interest or might benefit from some expert clarification, and at times, to say what I think. John passed away February 6, 2021. John was an outstanding physician, scientist and educator. His work in HIV pharmacology made important differences in improving treatment and the lives of persons living with HIV infection. He was a colleague to many, including me, and was a dear friend. https://www.legacy.com/obituaries/nytimes/obituary.aspx?n=john-g-gerber&pid=198270222
%CV, percent coefficient of variation
ACTG, adult AIDS clinical trials group
ARV, antiretroviral drug
ART, antiretroviral drug therapy
AUC, area under the concentration-time curve
BID, twice daily
C12, drug concentration at 12 hours post dose
Cmax, maximum drug concentration
Cmin, minimum drug concentration
CNS, central nervous system
c or COBI, cobicistat
CSF, cerebrospinal fluid
CVF, Cervicovaginal fluid
Ctrough, concentration immediately before the next dose
CYP, cytochrome P450 drug metabolizing enzymes
DBS, dried blood spot
DHHS, Department of Health and Human Services
DSMB, data safety monitoring board
GMR, geometric means ratio
HAND, HIV-associated neurocognitive disorders
HDAc, histone deacetylase
IC50, concentration of drug required to inhibit viral replication in vitro by 50%
IC90, concentration of drug required to inhibit viral replication in vitro by 90%
IQ, inhibitory quotient
IVR, intra-vaginal ring
MPA, medroxyprogesterone acetate
NRTI, nucleoside reverse transcriptase inhibitor
NNRTI, non-nucleoside reverse transcriptase inhibitor
PACTG, pediatric AIDS clinical trials group
PBMCs, peripheral blood mononuclear cells
PEG-IFN, pegylated Interferon
PI, inhibitor of HIV protease
PrEP, pre-exposure prophylaxis
QD, once daily
r or RTV, ritonavir
TAF, tenofovir alafenamide
TDF, tenofovir disoproxil fumarate
TDM, therapeutic drug monitoring