HIV-1 diagnosis with unquantifiable viraemia: don’t be naive, look for antiretroviral drugs Free

Sir,

In drug-naive patients, HIV-1 viral load (VL) is usually measurable, regardless of clinical stage.¹ When this is not the case, the causes must be clarified to ensure appropriate patient care.

We searched the databases of two Parisian hospital laboratories for patients with an unquantifiable VL although they were referred to the hospital for HIV care after a first positive HIV test. Patients were included if they claimed they were unaware of their infection until now, and if the western blot was positive for HIV-1 and negative for HIV-2. No ethics committee approval was required for this retrospective study.

Twenty-seven patients met the criteria over the last 4 years: 89% were originally from Africa. The demographic parameters and test results for these patients are summarized in Table 1. VL was routinely determined with the Roche Cobas Ampliprep/Cobas TaqMan v2.0 assay (CTAM). A control test was possible for 12 patients and confirmed unquantifiable with the Abbott RealTime assay (ABB) (Table 1).

Two patients (nos 22 and 23) who had unquantifiable viraemia with both assays after >6 months of follow-up were considered to be elite controller patients.² Their serotyping tests confirmed HIV-1 group M infection,³ and no antiretroviral drugs (ARV) were detected in their plasma. A woman (patient 21), was suspected to have stopped treatment before the HIV blood test, given the gradual increase in VL to 4.85 log copies/mL over three successive tests in a month. Four other patients (nos 24–27) did not return for subsequent consultations, making their follow-up too short (< 2 months) to determine whether they had recently stopped treatment or had elite controller status.

Finally, ARV intake was detected in the 20 remaining patients (74%): 18 of them had ARV detected in the plasma, and, at the second visit, two others (nos 19 and 20) admitted taking ARV. Some women said they were taking unknown drugs provided by a close friend/relative. The use of a false identity was suspected for three men. One of them (patient 16) retested negative for HIV antibody at a third visit and was believed to have sent an infected and treated person for the previous two blood tests. For the other two (patients 17 and 18), the same drugs were detected in plasma samples and CD4 cell counts were identical. Reverse transcriptase and protease genotyping on proviral DNA showed the sequences from these two patients to be 99.5% identical. By extending our search for similarities (Smartgene System) to the database of the same laboratory, containing data for >4500 patients, we identified four other patients with the same strain, which harboured only one or two nucleotide differences between the samples. These four additional patients were diagnosed with quantifiable viraemia several months before patients 17 and 18. Phylogenetic analysis and bootstrap confirmed the high degree of similarity of the HIV strains harboured by these six patients,⁴ whose other biological parameters were also very similar. These patients each consulted only twice, always postponing blood sampling to another day and requesting a medical certificate, after which they were never seen again. These findings suggest that, rather than a transmission cluster, we are dealing here with a single person who returned for all the blood tests over a period of 2 years, with treatment initiation in the meantime. At the second visit, we asked two of these patients to undergo a fingerstick control test, to remove all possible doubt. They categorically refused, before leaving the consultation and never returning.

This study aimed to determine the reasons for which viraemia was sometimes not quantifiable at the time of HIV-1 diagnosis. No HIV-1 quantification discrepancy was observed between CTAM and ABB assays, confirming the efficiency of the CTAM assay v2.0.^5,6 In these two Parisian hospitals, the presence of ARV in plasma was responsible for viraemia being unquantifiable in most patients (74%). Two situations were observed. In the first, the patient knew he was infected and taking ARV, but he said nothing. The strong taboo concerning HIV in Africans is a real issue and can lead the patient to conceal the actual history of his infection. However, residence permits are facilitated in France for migrants with diseases for which it is difficult to obtain appropriate treatment in the country of origin. Therefore, as related by some patients to our African mediator, ongoing treatment may be concealed during the visit, in order to obtain a residence permit more easily. This may also explain the second situation, in which non-infected individuals sent an infected person (sometimes on treatment) for the blood tests, to obtain a medical certificate. Plasma should therefore be systematically tested for the presence of ARV in this context. Adjustment of the dialogue with patients, to render the discourse more appropriate to the migrant situation, might also help to reduce patient fears, making it easier to obtain correct information about ongoing treatment.

Funding

This work was supported by the Agence Nationale de Recherches sur le SIDA (ANRS)

Transparency declarations

None to declare.

References