Factors Associated With Virologic Failure With Long-Acting ART: Perspectives From Canada

Key Takeaway

• In expanded multivariable analyses from ATLAS, ATLAS-2M, and FLAIR of people with HIV receiving long-acting cabotegravir plus rilpivirine, factors most predictive of confirmed virologic failure were preexisting rilpivirine resistance-associated mutations and HIV-1 subtype A6/A1.

Uptake of long-acting (LA) injectable cabotegravir (CAB) plus rilpivirine (RPV) is gradually increasing in Canada, as provincial drug plans in many parts of the country are offering public reimbursement. At HIV Glasgow 2022, updated analyses were presented regarding risk factors for virologic failure with this regimen. The presentation was based on pooled data from the pivotal randomized phase III trials that led to the original licensure of LA CAB plus RPV in Canada (ATLAS, ATLAS-2M, and FLAIR), in which 19 of 1651 participants experienced confirmed virologic failure (defined as 2 consecutive HIV-1 RNA levels ≥200 copies/mL). The investigators analyzed data on 1363 participants who received injections either every 4 weeks or every 8 weeks, and they constructed a multivariable logistic regression model to identify baseline demographic and clinical variables associated with confirmed virologic failure.

Predictors of Virologic Failure

Similar to the results of a previously published analysis, 3 variables were predictive of subsequent virologic failure in this “baseline model.”

• Having RPV resistance-associated mutations (RAMs) at baseline (incidence rate ratio [IRR]: 21.7; 95% CI: 5.80-80.8)
• Having HIV-1 subtype A6/A1 (IRR: 12.9; 95% CI: 4.42-37.5)
• Higher baseline BMI (IRR: 1.09 per 1-unit increase in kg/m²; 95% CI: 1.00-1.19)

Of note, receipt of the regimen every 4 weeks vs every 8 weeks was not a significant predictor; neither was sex at birth, cabotegravir RAMs, other nonnucleoside reverse-transcriptase inhibitor RAMs, other integrase strand transfer inhibitor RAMs, or the integrase L74I mutation. It also is important to recognize that only when 2 or more of these factors were present was the risk of virologic failure substantially increased.

In a complementary analysis that incorporated more longitudinal data from the participants, the key predictors of failure were RPV RAMs, HIV subtype A6/A1, and low predicted trough concentrations for CAB and RPV at specific timepoints. BMI was not a significant predictor in that model—perhaps because of collinearity with the drug concentration data—but because such pharmacokinetic parameters are not often available to practicing healthcare professionals (HCPs), most likely will want to focus on the simpler baseline model results instead.

Practical Approaches
Overall, these data probably should increase our confidence in prescribing this exciting new regimen for carefully selected patients who lack these important risk factors of RPV RAMs, subtypes A6/A1, and high BMI—after a thorough discussion regarding the data on virologic outcomes.

Of these factors, BMI is readily calculable. Subtypes A6 and A1 are uncommon in Canada—although it may be important to check whether migration from Ukraine related to the current geopolitical situation could change this—and the presence or potential presence of RPV RAMs often can be identified through careful review of a patient’s treatment history and previous genotype data. A practical application of these findings may be to assist HCPs in performing a systematic, cross-sectional evaluation regarding which of their patients may be good candidates for LA CAB plus RPV, which may assist with program planning.

In addition, although regimens every 4 weeks and every 8 weeks are both approved in Canada, dosing every 4 weeks is used less commonly, likely because of the inconvenience—both for patients and HCPs—associated with more frequent injection visits. Monthly injections may have a niche role for patients who need close follow-up for other health issues or who are anxious about the switch to injections, but the updated data provide important reassurance that the choice of dosing interval is not associated with a clear difference in virologic failure.

However, there was a word of caution from another study presented at the meeting. The CARLOS study is an observational cohort of patients in Germany who were switched from oral therapy to LA CAB plus RPV. Among the 200 patients who had Month 6 virologic outcome data and whose LA CAB plus RPV use was on label, there was a single patient (0.5%) who experienced virologic failure despite receiving on-time injections. In contrast to the analysis described above from the clinical trial–confirmed virologic failure cases, this person did not seem to have RPV RAMs at baseline, had HIV-1 subtype B, and had a normal BMI of 23 kg/m². To me, this suggests that we are still learning when it comes to the performance of this regimen in real-world settings—underscoring the importance of thoughtful, patient-centered counseling of individuals who are contemplating a switch to this innovative treatment option.

Your Thoughts?
What is your experience with long-acting ART regimens? Join the discussion and share your experiences by posting a comment.