Released: November 06, 2020
Expiration: November 05, 2021
Jean-Michel Molina, MD, PhD:
In this section, we will consider emerging treatment options—some that are still undergoing early-stage investigation and others than have been approved in select geographic locations but have not yet been authorized globally.
FLAIR is a multicenter, randomized, open-label phase III noninferiority trial examining monthly injection of long-acting cabotegravir (CAB) plus rilpivirine (RPV) after a 20-week oral DTG/abacavir (ABC)/3TC induction and a 4-week oral CAB plus RPV lead-in.21 At Week 100, the “extension phase” of FLAIR began during which patients in the comparator arm who had received daily oral DTG/ABC/3TC throughout the entire trial were switched to an injectable CAB plus RPV regimen; in consultation with a physician, they elected either to recapitulate the 4-week oral CAB plus RPV lead-in undergone in the experimental arm or to proceed directly to injections.22
In designing the extension phase this way, investigators were able to discern whether an oral CAB plus RPV lead-in was necessary for patients stably virologically suppressed with oral DTG/ABC/3TC. Because CAB is a structural analogue of DTG,23 it is reasonable to hypothesize that virologic control with oral DTG-based ART might obviate any need for an oral CAB lead-in.
Jean-Michel Molina, MD, PhD:
This Snapshot analysis is chronologically occurring at Week 124 of the FLAIR trial, but in the group being assessed, patients switched from oral DTG/ABC/3TC at Week 100, making this an examination of efficacy 24 weeks post switch. The patient numbers are small because the n = 283 originally randomized to continue oral DTG/ABC/3TC have been split into oral lead-in (OLI, n = 121) and direct-to-inject (DTI, n = 111) CAB plus RPV subgroups. With such short follow-up and reduced patient numbers per treatment group, the data on efficacy are limited. However, virologic suppression rates were 93% for OLI vs 99% for DTI.
Marina Klein, MD, MSc, FRCP(C):
The rationale for an oral lead-in has typically been to test for major AEs during a phase where the option for more immediate drug cessation exists. Once a patient has been injected with the long-acting formulations, the component drugs last for an extended time in the system. Thus, the design was intended more to address concerns for safety rather than efficacy.
What is interesting here is that there seem to be more issues in the OLI subgroup than with a DTI strategy. Eight patients in the OLI group discontinued, including 2 for AEs, compared with 1 patient in the DTI group who discontinued for virologic failure. As Dr. Molina emphasized, all of these patients have already been exposed to DTG, which should be similar to CAB in terms of potential toxicity, and RPV, in general, has a low toxicity profile. These data do suggest that a DTI strategy may be feasible without much worry regarding emergent AEs.
Jean-Michel Molina, MD, PhD:
I agree. The DTI CAB plus RPV approach seems safe for those already suppressed with DTG-based ART, at least within this limited number of patients. What I like to tell my students is to consider the confidence interval. When you do not see an event, there is still a confidence interval around that zero, so we cannot be fully reassured.
Marina Klein, MD, MSc, FRCP(C):
Numerically, there were numerically more rashes and grade 3/4 chemistry toxicities for DTI vs OLI patients. Among DTI patients, this included grade 3 aspartate aminotransferase, creatine phosphokinase, GFR, lipase, and phosphate changes as well as grade 4 creatine phosphokinase and lipase changes.
Jean-Michel Molina, MD, PhD:
For these patients switching from DTG/ABC/3TC to CAB plus RPV, the brand new drug is RPV, and mild rashes as well as liver toxicities have been associated with use of this drug.
Jean-Michel Molina, MD, PhD:
In a pooled analysis of long-acting CAB plus RPV Q4W/Q8W, virologic suppression at Week 48 was 94% (1531/1636) across the phase III FLAIR,21 ATLAS,24 and ATLAS-2M25 trials; there were 17 confirmed virologic failures (ie, 2 consecutive HIV-1 RNA measurements ≥ 200 copies/mL) for a treatment failure rate of approximately 1%. At HIV Glasgow 2020, we saw a multivariable post hoc analysis aiming to identify factors that could predict 13 of these failures in a subset of 1039 patients.26
One of the first issues noted was that the 7 patients with subtype A6/A1 who experienced treatment failure all had the integrase polymorphism L74I, which contrasts with the 4 subtype B patients experiencing failure in whom none had evidence of this mutation.
Marina Klein, MD, MSc, FRCP(C):
My only concern regarding this approach is that there were so few events. The investigators fit a final model based on backwards variable selection, focusing on 10 covariates with known or suspected influence on virologic suppression. This included drug pharmacokinetics, dosing interval, viral subtype, polymorphisms, sex, and body mass index (BMI). With this small number of outcomes, there is a risk for false positivity in terms of associating factors.
Jean-Michel Molina, MD, PhD:
I was surprised to see that they were able to identify so many factors. In total, 4 factors associated with an increased odds ratio of confirmed virologic failure: RPV RAMs at baseline, Week 8 RPV trough concentration, HIV-1 subtype A6/A1, and BMI.
Marina Klein, MD, MSc, FRCP(C):
Yes, these long-acting CAB plus RPV trials included HIV subtypes different from what we mostly see in Europe and North America. However, the CI for HIV-1 subtype A6/A1 extended from 1.82 to 25.26, so I am not sure how to interpret the extent of the effect.
Jean-Michel Molina, MD, PhD:
In their final model, the L74I integrase polymorphism itself did not predict failure. Really, the predictive factors had more to do with the RPV component of the regimen—RPV RAMs and RPV pharmacokinetics.
Marina Klein, MD, MSc, FRCP(C):
Yes, and this makes some biologic sense in that RPV is a drug with a lower barrier to resistance. If there is a break point, it is not surprising that it would be with that portion of the regimen.
Jean-Michel Molina, MD, PhD:
Historically, in clinical trials enrolling treatment-naive patients slated for RPV-containing ART, NNRTI RAMs were used as an exclusion criterion. Consequently, we lack data regarding the efficacy of RPV in the context of such RAMs, particularly with infrequent subtypes such as A6/A1.
The study investigators took their model one step further and asked whether the baseline predictive factors they identified in patients with confirmed virologic failure (ie, subtype A6/A1, RPV RAMs, BMI ≥ 30) commonly occurred together or alone. They found that in most cases (9/13), these individuals had more than 1 factor. Enthusiasm for long-acting injectable ART rightfully exists, but these data provide clinicians with a set of factors that may warrant some level of caution in deploying this strategy, particularly when they are present in combination.
Jean-Michel Molina, MD, PhD:
Islatravir (ISL) is a first-in-class nucleoside reverse transcriptase translocation inhibitor (NRTTI) undergoing clinical evaluation for HIV treatment and prevention. Protocol 11 (P011) assessed this novel drug in an international, randomized, double-blind phase IIb trial.27 ISL doses of 0.25 mg, 0.75 mg, and 2.25 mg were paired with doravirine (DOR) plus 3TC and compared against the single-tablet regimen DOR/3TC/TDF. After 24 weeks of treatment, ISL recipients with HIV-1 RNA < 50 copies/mL discontinued 3TC and continued receiving ISL plus DOR dual therapy to Week 96. In total, 121 ART-naive adults with HIV-1 RNA ≥ 1000 copies/mL, CD4+ cell count ≥ 200 cells/mm3, and no ARV resistance or hepatitis C virus/HBV coinfection at baseline were enrolled. Patients were predominantly recruited from the United States and Europe.
Jean-Michel Molina, MD, PhD:
At HIV Glasgow 2020, I presented P011 results from Week 96,28 and the bottom line is that, in terms of safety and efficacy at this time, the combination of ISL plus DOR looks good. It is not perfect, but it looks good. I would draw your attention to the first set of bars in this graph. When the 3 ISL arms were combined in Snapshot analysis, virologic suppression at Week 96 was 81.1% with ISL plus DOR vs 80.6% with DOR/3TC/TDF.
Marina Klein, MD, MSc, FRCP(C):
Relatively speaking, however, these efficacy rates are not as high as we have come to expect from first-line ART.
Jean-Michel Molina, MD, PhD:
I agree. DOR-based ART is not yet standard of care. For instance, DOR/3TC/TDF is recommended as an alternative first-line regimen in EACS10 and DHHS8 guidance. The trial design is something to consider as well, because patients only moved to the dual regimen if they were suppressed on the triple combination of ISL plus DOR plus 3TC for the first 6 months.
It is a bit counterintuitive to see the highest rate of HIV-1 RNA ≥ 50 copies/mL with the highest dose of ISL, but to gain a real sense of the antiviral potency, it is important to look at the protocol-defined virologic failures. The absolute numbers are small; there were 2 cases in each ISL arm and 1 in the DOR/3TC/TDF arm. It is important to note that among these 7 patients experiencing treatment failure, 5 had a baseline HIV-1 RNA > 100,000 copies/mL, and 5 had suppressed to HIV-1 RNA < 50 copies/mL before switching back to a triple-drug regimen. Thus, these virologic failures involved low-level viremia and did not meet criteria for resistance testing (> 400 copies/mL). That is why no resistance data are available at this point.
Jean-Michel Molina, MD, PhD:
ISL plus DOR was generally well tolerated. Between Week 48 and Week 96, no new drug-related AEs emerged. However, cumulatively, there was a numerically higher rate of drug-related AEs with DOR/3TC/TDF vs ISL plus DOR (22.6% vs 7.8%, respectively). In terms of discontinuations directly caused by these drug-related AEs, 1 DOR/3TC/TDF recipient experienced worsening of congenital long QT syndrome, and 2 ISL plus DOR recipients reported either diarrhea/nausea/vomiting or HBV reactivation.
Based on the totality of P011 results, ISL plus DOR has moved into phase III study with selection of the middle dose of ISL, 0.75 mg.29
Jean-Michel Molina, MD, PhD:
Next, we will consider the novel NNRTI MK-8507. Like DOR, this potent drug demonstrates activity against common NNRTI RAMs.
Marina Klein, MD, MSc, FRCP(C):
What makes MK-8507 particularly compelling is that it is one of the first agents to be tested as a weekly treatment. Because preliminary pharmacokinetic data supported this dosing interval, 18 ART-naive, White, male PWH were enrolled on an open-label phase I trial to receive a single oral dose of MK-8507 (40, 80, or 600 mg).30 At baseline, they had HIV-1 RNA ≥ 10,000 copies/mL, CD4+ cell count > 200 cells/mm3, and no NNRTI resistance or hepatitis C virus/HBV coinfection.
Marina Klein, MD, MSc, FRCP(C):
Overall, decline in viremia was quite rapid, occurring within hours. By 96 hours or 4 days post dose, mean reduction in plasma HIV-1 RNA had already reached the ≥ 1 log10 copies/mL target for the 80-mg and 600-mg doses. It was interesting not to see a substantial dose response between these 2 doses; at 168 hours or 7 days post dose, mean reductions in HIV-1 RNA were -1.53 and -1.50 log10 copies/mL, respectively.
Marina Klein, MD, MSc, FRCP(C):
The safety data looked good for this small group.31 Headache, which occurred in 3 participants, was the sole AE deemed drug related.
Jean-Michel Molina, MD, PhD:
The last thing I would emphasize from this proof-of-concept study is emergence of the RT resistance mutation F227C in 1 patient delaying initiation of standard-of-care ART until > 14 days from MK-8507 administration. Clinicians may not be familiar with this nonpolymorphic mutation, as experience with DOR is still generally limited and relatively few treatment failures have consequently occurred. However, F227C and F227L, particularly in combination with V106A, have emerged in patients receiving DOR.32,33
Marina Klein, MD, MSc, FRCP(C):
MK-8507, like all drugs, can select resistance if used as a monotherapy. It does appear that the pathway of resistance observed here mimics previous reports for DOR. Of note, the pharmacokinetic data from these 18 participants continued to support the approach of once-weekly dosing.
Jean-Michel Molina, MD, PhD:
Moving forward, MK-8507 once weekly will be combined with ISL for comparison against bictegravir (BIC)/FTC/TAF in the switch setting; this phase II trial is not yet recruiting but is targeting an enrollment of 140 participants.34
Jean-Michel Molina, MD, PhD:
The next agent we will discuss is the capsid inhibitor lenacapavir (LEN). This is another first-in-class drug undergoing early-phase investigation. At HIV Glasgow 2020, we saw resistance results from Study 4072, a double-blind, dose-ranging phase Ib study randomizing 39 INSTI-naive and capsid inhibitor–naive patients with HIV-1 RNA levels of 5000-400,000 copies/mL who had not received ART within 12 weeks and with CD4+ cell counts > 200 cells/mm3 to a single subcutaneous dose of either LEN (20, 50, 150, 450, or 750 mg) or placebo.35,36 At Day 10, the maximum on-study reduction in plasma HIV-1 RNA was determined, and all participants transitioned to oral BIC/FTC/TAF.
Jean-Michel Molina, MD, PhD:
Genotypic and phenotypic resistance were analyzed at baseline and Day 10. None of the participants harbored LEN resistance mutations at entry, nor did they exhibit reduced susceptibility to LEN. It is notable that 2 participants developed Q67H by Day 10, 1 receiving the 20-mg dose of LEN and 1 receiving the 50-mg dose. This mutation emerged during a phase of viremic decline in the 20-mg patient and was detected in > 10% of reads by next-generation sequencing. That is somewhat surprising. In the 50-mg recipient, detection was only at > 2%, but HIV-1 RNA did trend marginally upward between Days 7 and 8 before declining again toward Day 10.
Marina Klein, MD, MSc, FRCP(C):
It is worth emphasizing that this Q57H mutation occurred in patients with the lowest dose exposures, but regardless, it suggests a relatively low threshold for resistance during LEN monotherapy. Study investigators pointed out that these exposures are below what is expected for the phase II/III studies.
Marina Klein, MD, MSc, FRCP(C):
In terms of mean plasma HIV-1 RNA reduction, the higher LEN doses exhibited quite potent activity—> 2 log10 copies/mL reductions at Day 10, but even the lowest dose managed to reach a mean decline of 1.3 log10 copies/mL at this time point.37
Jean-Michel Molina, MD, PhD:
The ongoing phase II CALIBRATE trial evaluates both daily oral and biannual subcutaneous dosing of LEN in combination with other ARVs.38 Such an extended dosing interval would make this a great candidate for pre-exposure prophylaxis as well.
Back
