Using HIV-ASSIST to Guide ART Selection in a Medically–Complex Patient

Opportunities to intervene in patients’ HIV treatment can be varied and unanticipated, so we should grasp these as they arise. In this commentary, I share an example of opportunistic antiretroviral therapy (ART) evaluation centered around a patient who presented to his HIV provider with concerns about sexually transmitted infections. Of importance, I explain how we used HIV-ASSIST to help sort through his complex medication issues.

Patient Case: 58-Year-old Man With Complex Medical History and Intermittent ART Adherence
This patient is a 58-year-old man who, like many people with HIV (PWH), has significant comorbid conditions including hypertension, opioid use disorder, alcohol use disorder, chronic untreated hepatitis C virus (HCV) infection, and chronic hepatitis B virus (HBV) infection. He was diagnosed with HIV infection approximately 20 years ago and has a history of intermittent adherence to ART with infrequent follow-up, partly because of several episodes of incarceration.

This patient recently presented to his HIV provider with a 10-day history of penile discharge plus a 7-day history of nonpruritic rash, most noticeable over the chest and abdomen. He requested an antibiotic for the discomfort associated with the penile discharge and a lotion for his skin rash.

On examination, he had yellow urethral discharge with a maculopapular rash over the torso and back, with palmar–plantar pustulosis. He was treated empirically for gonorrhea, chlamydia, and secondary syphilis, with ceftriaxone 250 mg intramuscularly (the dosing recommended by CDC at the time), azithromycin 1 g orally, and benzathine penicillin G 2.4 million units intramuscularly. Although he has a long history of intermittent adherence, his HIV-1 RNA was undetectable, and is CD4+ cell count was 125 cells/mm³ (4.4%).

Because of this patient’s complicated social and medical issues, his HIV provider took the opportunity to review the patient’s electronic medical record and compiled the following ART history:

• 2004: He was started on stavudine (d4T), lamivudine (3TC), and efavirenz (EFV); his HIV-1 RNA at the time was 75,000 copies/mL with a CD4+ cell count of 8 cells/mm³.
• Later in 2004: He was switched from EFV to nevirapine (NVP), continuing d4T and 3TC; his HIV-1 RNA at the time of switch was 1000 copies/mL. Subsequently, he had an undetectable HIV-1 RNA and was then lost to follow-up.
• 2010: He returned to care, having been on and off treatment several times in the intervening years. His CD4+ cell count was 15 cells/mm³, with an HIV-1 RNA of 22,000 copies/mL, and genotypic testing revealed the M184V mutation. He was started on lopinavir (LPV)/ritonavir (RTV), zidovudine (ZDV), and tenofovir disoproxil fumarate (TDF). In June, he had an undetectable HIV-1 RNA, but was again lost to follow-up.
• 2016: He returned to care, with a recent history of viral suppression (undetectable) during incarceration in the previous 2 years. He reported chronic diarrhea with LPV/RTV, ZDV, and TDF, and so he was switched to darunavir (DRV)/cobicistat (COBI) plus FTC/tenofovir alafenamide (TAF). His chronic diarrhea persisted, however, so his boosted PI was switched to dolutegravir (DTG). Shortly thereafter, he was again lost to follow-up.
• 2018: He returned to care in September, reporting that he had been on and off ART; testing revealed wild-type HIV with dual tropism. He was re-started on dolutegravir (DTG) plus TAF/FTC. Approximately 4 weeks later, he was seen in the emergency department for severe facial swelling with angioedema. An allergy specialist diagnosed an allergic reaction, likely to DTG, and advised against rechallenge. He was switched back to his previous regimen of DRV/COBI/FTC/TAF, with the addition of loperamide for diarrhea (after opportunistic gastrointestinal infections had been ruled out).
• 2020 to 2021: Between September 2020 and February 2021, his viral load was undetectable on DRV/COBI/FTC/TAF.  In the same interval, however, his serum creatinine had increased (from 1.1 to 1.5 to 2.0 to 2.1 mg/dL) with a concomitant decrease in estimated glomerular filtration rate (eGFR) from 72 to 35 mL/min. Concomitant with this decline in eGFR, he had glycosuria, 1+ proteinuria, and low serum phosphate.

With that background, let’s summarize his current ART-related issues. He is virally-suppressed on his current regimen, but his recent renal dysfunction is very concerning. Although TAF is considered a safe drug with respect to renal toxicity, it is difficult to argue that TAF is completely unimplicated in his current renal dysfunction, and so contingency planning is in order. Discontinuing TAF, in light of his M184 V mutation, would mean effectively treating his HIV infection with DRV/COBI monotherapy. And although his genotype tests did not show NNRTI resistance mutations, his history of poor adherence on both EFV and NVP poses the potential for archived NNRTI mutations such as K103N or Y181C. In thinking about alternative HIV drugs for this situation, an obvious choice would be reintroduction of an integrase inhibitor, but there is the risk that his aforementioned allergic reaction to DTG represents a class effect. Also of concern is his chronic HBV infection, with a high HBV DNA (4.4 x 10⁶ IU/mL) despite recent adherence to FTC/TAF.

Treatment Decisions Supported by HIV-ASSIST
At this point, we turned to the HIV-ASSIST tool, a free, online, decision-support tool available at hivassist.com. HIV-ASSIST allows a clinician to input virus-specific and patient-specific factors (including HIV mutations, HIV-1 RNA, CD4+ cell count, HLA-B*5701 status, tropism, comedications, comorbidities, and patient adherence preferences) that support individualized ART selection.  We entered information about this case into the tool, which yielded a ranked list of potential ART regimens, each with a weighted score indicating the relative strength of evidence underpinning it. Each regimen is conveniently linked to educational summaries, including current HIV guidance from the DHHS and the IAS-USA.

In this case, the highest ranked regimen is the current regimen of DRV/COBI/TAF/FTC. This recommendation is reassuring, and supports a decision to maintain this regimen for as long as possible, with the patient’s renal decline being the main source of concern in relation to HIV treatment.

What might our future options be if we did decide he can no longer receive tenofovir-based therapy? In that scenario, HIV-ASSIST recommends the addition of doravirine (DOR) or rilpivirine (RPV) to DRV/COBI. In support of this potential strategy, HIV-ASSIST provides information about the clinical trials using an NNRTI plus a boosted PI. For example, the PROBE study showed that a dual ART regimen of rilpivirine plus boosted DRV was noninferior to the standard-of-care ART in 60 treatment-experienced patients. Two other studies, NEKA and MULTINEKA, looked at the efficacy of LPV/RTV plus NVP vs LPV/RTV plus 2 NNRTIs. In both of those studies, LPV/RTV plus NVP was shown to be noninferior to the alternative, with the patients who received LPV/RTV plus NVP maintaining virologic suppression.

Moving further down the list of potential regimens, HIV-ASSIST suggested that we could consider adding ibalizumab or fostemsavir to the current regimen. However, this patient is currently suppressed, so we preferred to first consider options such as adding an NNRTI.

Expert Panel Insights
Of particular value in complex cases like ours, HIV-ASSIST also offers an option to engage in dialogue with an HIV expert advisory panel. In this case, we wanted to explore options to treat his HIV and HBV if TAF and integrase inhibitors were off the table.

Discussions with the panel resulted in several strategies to consider. The first was to continue DRV/COBI/FTC/TAF while monitoring the eGFR closely. The panel also suggested confirming the patient’s HLA-B*5701 status, which if negative, would allow for future consideration of abacavir. Another angle was to try to obtain archived genotypes. Negative results would not be helpful, but if genotypes were positive for any NNRTI mutations, the results would inform decisions about the potential addition of DOR or RPV to his boosted PI.

(Although outside the scope of HIV-ASSIST tool, it is important to recognize that this patient’s HBV had likely developed resistance to 3TC. Thus, we have to consider using a drug such as high-dose entecavir, in addition to continuing tenofovir-based ART.)

Take-home Lessons
We ultimately elected to continue DRV/COBI/FTC/TAF with close renal monitoring. Let’s conclude by considering what can be learned from this case. First, sexually transmitted infection screening gave us an opportunity to re-engage this patient with HIV care. Second, the HIV-ASSIST decision support tool can help providers evaluate ART options, and conveniently links to clinical data that provide evidence for potential regimens. (Another helpful feature of HIV-ASSIST is that it allows a clinician to exclude certain drug options. In this case, because of the patient’s prior allergic reaction, I marked drugs in the integrase class as drugs the tool should omit from its recommendations.) Third, in persons with HIV who are treatment-experienced (particularly with a history of nonadherence), archived mutations need to be considered even if absent from the genotypic record. Fourth, HIV-ASSIST can connect clinicians with an expert advisory panel for consultation on complex cases. In this case, the HIV expert panel advised that TAF can be continued in renal dysfunction with close monitoring (and eGFR >30 mL/min). In summary, I highly recommend this tool as an adjunct in clinical care of HIV, both for initiating ART in uncomplicated situations and for evaluation of complex, treatment–experienced patients.

Your Thoughts?
How would you have approached management for this complex patient case? Answer the polling question and join the conversation by posting in the discussion section.