Unsupervised PrEP in routine practice: a new challenge?

Introduction

Pre-exposure prophylaxis (PrEP) for the prevention of HIV infection with 300 mg tenofovir daily co-formulated with 200 mg emtricitabine (TDF/FTC) is recommended as one prevention option for people who are at substantial risk of acquiring an HIV infection.^1,2 PrEP was implemented in France on 25 November 2015, first as part of a Temporary Recommendation for Use ³ based on individual-level risk factors for HIV infection, followed by a marketing authorization extension without advance fee for the patient in February 2017. ⁴

According to French guidelines, two negative HIV tests performed four weeks apart without HIV risk behaviour during these four weeks are required ⁵ in order to avoid PrEP initiation during the HIV seroconversion period.

Indeed, such a situation could cause a drug resistance emergency ¹ or could cause atypical patterns of seroconversion. ⁶ Here, we report a case of unsupervised PrEP with an impact on the HIV seroconversion pattern.

Case report

A 28-year-old man who has sex with men (MSM) was referred to our unit in June 2017 for a primary HIV infection diagnosed based on positive HIV-1 antibodies (positive antibodies ratio at 87.51 for a threshold positive value >1.25), negative p24 antigen (<3 pg/mL) and positive p18, p24 and gp160 bands on Western blot analysis (three bands that become positive during the earliest stages of HIV infection). However, the HIV plasma viral load (HIV-VL) (VERIS Beckman Coulter technique) was 56 copies/mL, which was unusually low in the context of a primary infection. The HIV subtype was subtype B. Thus, the low level of HIV viral load could not be explained by an unusual subtype that required specific techniques of viral load quantification. HIV DNA was detectable at a level of 148 copies/10⁶ peripheral blood mononuclear cells, dismissing a possible false HIV-positive serology. The control test for viral quantification performed with a Biocentric assay confirmed the low HIV-VL with 1236 copies/mL. Of note, the patient reported frequent unprotected sex with about four partners per month with the use of recreational drugs, and screening for sexually transmitted infections was positive for Neisseria gonorrhoeae. A previous HIV test performed in January 2017 was negative, and the patient declared to have taken a medicated post exposure prophylaxis for a few days with combination antiretroviral therapy (ART) at this time. The low HIV-VL in the context of primary HIV infection in an MSM with a high number of sex partners and self medicated post exposure prophylaxis practice led us to suspect undeclared PrEP. Further information confirmed this hypothesis as the patient disclosed self-medication with TDF/FTC for four weeks at the time of his positive HIV test without any HIV testing before PrEP initiation.

Despite this unsupervised PrEP initiated during an HIV seroconversion, the HIV genotype was still wild-type. A single-tablet formulation with an integrase strand transfer inhibitor, in order to avoid drug–drug interactions with recreational drugs, was initiated. At baseline, CD4 and CD8 T-cell counts were 771 cells/mm³ and 459 cells/mm³, respectively. After four weeks of treatment, HIV-VL was undetectable.

The patient provided written informed consent for the use of his medical record on NADIS®. This electronic medical record was approved by the French Commission National Informatique et Liberté (Registration number: 2001/762876/nadiscnil.doc).

Discussion

Although HIV misdiagnosis should always be considered in cases of atypical seroconversion pattern with a low viral burden, this case underscores that unsupervised PrEP should now be systematically investigated. ⁷

We could not establish that this case of primary HIV infection was due to a failure of PrEP because PrEP was taken in an uncontrolled and unstructured manner. However, if so, it is important to consider that in our patient, the concomitant infection with N. gonorrhoeae might have enhanced HIV transmission risk. ⁸ The low HIV-VL might be explained by suboptimal ART use in the context of a primary infection but also might reflect HIV-specific T cell responses that were primed by multiple HIV exposures. Unfortunately, we did not evaluate an HIV-specific cytotoxic T-lymphocyte response in our patient. However, the sequence of events observed in this patient seems to be more in favour of PrEP initiation during unrecognized acute HIV infection.

Although rare, emergence of a drug-resistant virus with M184V/I and K65R mutations has been reported in such situations in the Partners PrEP, TDF2 and VOICE trials and in patients who acquired the infection under PrEP in the FEM-PrEP trial. ¹ No resistance was observed in our case.

This case underscores the need to strengthen information regarding the usefulness of laboratory HIV testing before PrEP initiation, PrEP monitoring and the fact that PrEP efficacy depends on consistent use, ⁹ although one case of seroconversion was reported in the AMREP study. ⁶

Moreover, clinicians should be alerted for atypical discordance between HIV seroconversion with negative or low HIV-VL with PrEP and the need to systematically investigate for undeclared PrEP use.