Acute HIV at the Time of Initiation of Pre-Exposure or Post-Exposure Prophylaxis

Impact on Drug Resistance and Clinical Outcomes

Kelly A. Johnson, MD, MPH; Miao-Jung Chen, PhD, MPH; Robert Kohn, MPH; Darpun Sachdev, MD, MPH; Oliver Bacon, MD, MPH; Sulggi Lee, MD, PhD; Stephanie E. Cohen, MD, MPH


J Acquir Immune Defic Syndr. 2021;87(2):818-825. 

In This Article


This retrospective case series analysis describes long-term outcomes in individuals prescribed PrEP or PEP at the time of as-yet-undiagnosed AHI infection. To the best of our knowledge, this is the largest review of undiagnosed AHI prevalence at the time of initiation of PEP or PrEP in a clinical setting and the only case series of outcomes of AHI in this clinical context. AHI was a rare event in this series, occurring in 13 of 4000 (0.3%) of PrEP and PEP starts. This finding is consistent with recent results reported from the largest HIV testing center in Thailand, wherein AHI was subsequently diagnosed in just 7 of 2442 (0.2%) of PrEP starts.[22] When undiagnosed AHI was present at the time of PrEP/PEP initiation, all 13 SFCC patients (100%) were rapidly linked to care with a median time of 1 week. The majority (over 90%) of these individuals received ART prescriptions within a median of 2 weeks, and most (nearly 85%) had documentation of viral suppression within a median of 2 months.

Despite the relative infrequency with which undiagnosed AHI is present at the time of PrEP initiation, this scenario can lead to HIV resistance. The M184 mutation developed in 23% of the 13 patients in our case series (and in 3/11 or 27% of those with available genotypes at the time of AHI diagnosis), while the K65R mutation was not seen. In the overall study population from which our 13 cases were drawn, an M184 mutation correlating with same-day PrEP/PEP start in the face of undiagnosed AHI was even more rare, occurring in just 0.075% of the n = 4000 PrEP/PEP starts. These findings echo 2 previous meta-analyses of PrEP studies in which (1) the FTC-associated M184 mutation emerged more frequently than the TDF-associated K65R, and (2) HIV resistance mutations occurred in 18%–41% of patients with undiagnosed AHI at the time of PrEP enrollment, compared with just 2%–3% of individuals who acquired HIV during PrEP follow-up.[2,12] Although a recent Thai study concluded that HIV resistance mutations were unlikely to develop clinically within the first 15 days of PrEP use among patients with AHI,[22] our case series demonstrates that the M184 mutation can develop more rapidly—within just 1–2 weeks (median 8 days, IQR 7–12) of TDF/FTC exposure. This is consistent with earlier in vitro and in vivo data demonstrating rapid development of M184 in the setting 3TC/FTC.[23–25]

All 3 SFCC patients who developed an M184 mutation were diagnosed with AHI in 2012–2013 and were virally suppressed after a median of 112 days; as of 12/2020, the most recently documented HIV viral loads for each of these 3 individuals (in 2014, 2017, and 2020, respectively) remain <40 copies/mL. It is unknown how clinical outcomes might have changed if AHI had not been diagnosed. It is conceivable that delayed diagnosis in the setting of ongoing ART exposure could have resulted in accumulation of additional resistance mutations.

Given the rapidity with which HIV resistance mutations can emerge when PrEP is prescribed in the setting of undiagnosed AHI, our study highlights the importance of screening for AHI at the time of PrEP initiation—particularly in programs prescribing same-day PrEP. The World Health Organization (WHO), Centers for Disease Control and Prevention (CDC), and International Antiviral Society-USA (IAS-USA) all recommend screening for HIV before starting PrEP, at minimum with an antibody-based test and optimally with a laboratory-based antibody/antigen (Ab/Ag) assay and/or HIV RNA (both of which detect HIV earlier than antibody tests alone)—particularly if there is any suspicion for AHI.[26–28]

In our STD clinic, all patients initiating PrEP are screened with both a rapid point-of-care antibody test and an HIV RNA pool, allowing for rapid identification of AHI cases and linkage to HIV care where indicated. Pooled RNA testing has been shown to be a particularly effective strategy for detecting AHI in high prevalence communities, increasing the number HIV cases identified by 10.5% over antibody-based screening alone.[29] Laboratory-based fourth-generation Ab/Ag testing also increases the diagnostic yield beyond antibody-only tests—but missed 18% of HIV infections detectable by pooled HIV RNA in a multisite prospective US study[30]—and point-of-care rapid Ab/Ag tests are less sensitive than laboratory-based assays.[31]

In resource-limited settings, where universal screening with HIV RNA and/or HIV Ab/Ag assays may not be feasible, PrEP programs could consider prioritizing such testing for patients at higher epidemiologic risk of AHI. Nearly all (12/13, or 92.3%) of the AHI patients in our case series were asymptomatic at initial presentation, suggesting that symptom-driven testing for AHI would miss most AHI cases in this setting. In our study, all AHI patients were men who have sex with men (MSM) with multiple anal sex partners, many of whom reported sex with HIV-positive individuals in the days leading up to their AHI diagnosis. The majority (nearly 70%) of our AHI patients had been diagnosed with bacterial STIs, primarily rectal chlamydia (CT) or gonorrhea (GC) and/or syphilis, in the year preceding their HIV diagnosis. Previous literature has indicated that rectal GC/CT and syphilis are independently associated with HIV acquisition within the upcoming year.[32,33] Patients with these risk factors should be not only screened for AHI when they first present to care but also (1) encouraged to start PrEP and (2) offered tools to support PrEP retention over time.

Strengths of our study include the large sample size of 4000 same-day PrEP and (primarily 2-drug) PEP starts, combined with detailed electronic medical records and chart abstraction linked to San Francisco Department of Public Health HIV outcome registry data dating back to 2011. We were also able to leverage eHARS to ascertain longer-term clinical outcomes for our AHI patients who received HIV care anywhere in California, regardless of whether these individuals ever returned to SFCC after their initial HIV diagnosis. The integration of the SFCC EMR and SF DPH STD registry systems[15] similarly allowed us to achieve complete ascertainment of all reportable STIs diagnosed anywhere in the city of San Francisco. For the 3 AHI patients with an M184 mutation at the time of HIV care initiation, the ability to retroactively perform genotyping on stored serum from the date of TDF/FTC start provided us with a unique opportunity to assess the rapidity with which M184 mutations can emerge in a routine clinical setting.

In terms of limitations, our study was conducted at a single clinic in which all cases of undiagnosed AHI in the setting of PrEP/PEP initiation occurred among MSM, in a jurisdiction with a robust HIV linkage to care and rapid ART programs.[34] As a result, our findings regarding long-term patient outcomes may not be generalizable to other demographic groups, including cis-women and transgender persons, or to other geographic regions—particularly those with less robust infrastructure in place to support HIV RNA testing to detect AHI, linkage to HIV care, and rapid ART start.

We do not have information on how many pills each patient took between the day of PrEP or PEP start and the day of HIV disclosure or resistance testing. It is thus possible that lack of selective drug pressure in any nonadherent PrEP/PEP patients contributed to the lack of drug resistance mutations in some of the patients in our case series. Finally, we were unable to fully confirm that the HIV resistance mutations we observed were the result of PrEP/PEP exposure rather than transmitted drug resistance from sexual partners and we do not know if any of the AHI infections that were negative by HIV STAT-PAK would have been detected by a point of care rapid Ag/Ab assay.

Despite these limitations, our study has important implications for PrEP programs, particularly those considering same-day PrEP prescribing to increase PrEP uptake and reduce barriers to care. Although same-day PrEP programs do not guarantee long-term retention in care, our case series demonstrates that AHI in the setting of same-day PrEP/PEP initiation is an infrequent event in the STD clinic setting and that patients found to be acutely infected with HIV can be rapidly linked to care and achieve excellent clinical outcomes. Overall, our results offer reassurance that the benefits of same-day PrEP prescribing outweigh the risks associated with initiating PrEP in the setting of AHI, provided systems are in place for early diagnosis and connection to HIV care.