Predicting Virologic Failure Risk With LA CAB + RPV: Perspectives From Spain

Key Takeaways

• At HIV Glasgow 2022, investigators presented data showing that the strongest predictor of confirmed virologic failure for patients switching from suppressive daily oral antiretroviral therapy to long-acting injectable cabotegravir plus rilpivirine is the presence of ≥2 of the following baseline factors: rilpivirine resistance-associated mutations, HIV subtype A6/A1, and BMI ≥30 kg/m².

To me, long-acting (LA) therapy with cabotegravir (CAB) plus rilpivirine (RPV) is a crucial addition to our antiretroviral armamentarium. A substantial number of people with HIV place a high value on not having to take daily oral therapy. The main drivers for choosing this type of therapy are reducing internal or external stigma, difficulties with adherence to daily oral therapy, and difficulty swallowing pills.

Somewhat unexpectedly, we have learned that virologic failures of LA therapy with CAB plus RPV are frequently associated with the selection of mutations conferring resistance to integrase strand transfer inhibitors (INSTIs) and non-nucleoside reverse-transcriptase inhibitors (NNRTIs). I say “unexpectedly” because one of my expectations was that people with HIV who adhere to LA medication administered intramuscularly would have the “guarantee” of having no risk of resistance selection.

The ATLAS, FLAIR, and ATLAS-2M studies have shown that most patients receiving CAB plus RPV intramuscularly every 4 or 8 weeks remain virologically suppressed for up to 3 years, with an overall rate of confirmed virologic failure of 1.4%. Although 1.4% is a very low absolute risk, we must put this figure in the context of 2 important factors: (1) There is essentially no risk of resistance with daily oral regimens based on dolutegravir or bictegravir, and (2) some of these virologic failures lead to resistance to second-generation INSTIs, our most recommended antiretroviral drug class. For these reasons, it is essential that healthcare professionals (HCPs) have tools that can help them minimize the risk of virologic failure after switching to LA therapy.

Predicting Virologic Failure: Model of Baseline and Postbaseline Factors
At HIV Glasgow 2022, investigators presented the results of an analysis of predictors of confirmed virologic failure with LA CAB plus RPV that included expanded multivariable models encompassing both baseline and postbaseline factors to assist with patient selection. In this analysis, factors significantly associated with the risk of confirmed virologic failure were:

• Presence of RPV-associated resistance mutations before switching to LA treatment, with an adjusted incident rate ratio (IRR) of 25.7, making this by far the strongest predictor for risk of virologic failure
• HIV-1 subtype A6/A1, with an adjusted IRR of 15.5
• Low predicted CAB trough concentration at Weeks 4 and 44 and predicted RPV trough concentration at Week 4 after initial injections, with adjusted IRRs of 2.20-5.99

Of interest, BMI was not predictive of virologic failure in this model because it is correlated with CAB pharmacokinetics. Other factors that were also not predictive of virologic failure were the dosing interval (every 8 weeks vs every 4 weeks), the L74I integrase polymorphic mutation, sex at birth, other NNRTI resistance-associated mutations, CAB resistance-associated mutations, other INSTI resistance-associated mutations, and predicted RPV trough concentration at Week 4.

Predicting Virologic Failure: Model of Baseline Factors Only
Investigators went a step further to help HCPs by evaluating a combination of predictive factors that can be used at baseline before switching to LA treatment. Few clinics worldwide have access to therapeutic drug monitoring; thus, most cannot measure CAB plus RPV trough levels on treatment. The 3 baseline factors significantly associated with risk of confirmed virological failure were:

• Presence of RPV-associated resistance mutations, with an adjusted IRR of 21.7
• HIV subtype A6/A1, with an adjusted IRR of 12.9
• Baseline BMI ≥30 kg/m², with an adjusted IRR of 3.97 (1.09 per 1 kg/m² unit increase when evaluated on a continuous scale)

Similar to the model that included postbaseline factors, the dosing interval (every 8 weeks vs every 4 weeks) and the integrase L74I mutation were not predictive of confirmed virologic failure risk.

The final guidance point for HCPs is that the presence of 0-1 baseline risk factor is associated with a low risk of virologic failure. Only when ≥2 baseline factors are present is the rate of confirmed virologic failure considerably increased to 19.3%. If none of these factors―or even only 1 factor―is present, the negative predictive value for virologic failure occurrence is 99.1%, which is very reassuring. These findings confirm the results of a previous multivariable analysis that was based on data from 1 year of LA CAB plus RPV treatment, now adding an expanded post hoc multivariable analysis that includes data beyond 1 year (ATLAS through Week 96, ATLAS-2M through Week 152, and FLAIR through Week 124).

I consider this information quite relevant for my clinical practice. LA CAB plus RPV will soon be commercially available in Spain. Before recommending this therapy as a switch option for my patients with virologic suppression, I will carefully assess for a history of virologic failure with NNRTIs. If there is such a history, I would be reluctant to recommend a switch to LA therapy, particularly if the person also has a high BMI. The other factor predictive of virologic failure risk, HIV subtype A6/A1, is rare in our practice.

Your Thoughts?
Have you begun offering a switch to LA HIV treatment for appropriate patients in your clinical practice? If so, how are you considering patient eligibility for the switch based on currently available data regarding predictors of virologic failure? Join the discussion and share your experiences by posting a comment.