Monitoring of HIV-1 Resistance in Tunisia (North Africa) With a Dried Plasma Spot Strategy

To the Editor: The HIV-1 epidemic is particularly severe in the resource-limited countries, where more than 30 million people are living with HIV infection.1 The expansion of access to antiretroviral drugs has made it necessary to develop appropriate tools for the therapeutic follow-up of patients in these regions. Epidemiologic surveillance of the spread and transmission of HIV resistant to antiretroviral drugs is also a major issue for the future use of these drugs in the resource-limited countries. Resistance genotyping has proved useful for such monitoring in the more developed countries.2 This method requires a sequencer, however, and its routine use outside a few reference centers is therefore difficult in resource-limited countries. It also poses logistic problems in terms of the transport and storage of plasma samples. Many studies have demonstrated the utility of dried sample spot technology as an alternative method for serologic and molecular analyses of HIV infection.3-7 We recently demonstrated the feasibility of using a dried plasma spot/dried serum spot (DPS/DSS) method for resistance genotyping, which could be a useful alternative in clinical and epidemiologic settings.8 In this study, we tested a sequencing-based strategy for monitoring from a distance, in which plasma samples were collected far from the center with the sequencer and the expertise to interpret the results. Blood samples were taken from patients positive for HIV-1 in Tunisia. Plasma samples were deposited on filter paper, allowed to dry, and sent by post to our laboratory in France for analysis of resistance to antiretroviral drugs. We obtained 33 plasma samples from 24 HIV-1-infected Tunisian patients with a plasma viral load (VL) ranging from 1300 copies/mL (3.1 log) to more than 750,000 copies/mL (5.9 log) (Table 1). Samples were collected from patients who had never received highly active antiretroviral treatment (HAART) (n = 13) and from treated patients (n = 5). Serial plasma samples were collected during treatment (patient [P] 5, P15, P18, and P22) or before and during treatment (P6 and P8). VL was measured in native samples (stored at −80°C) with an Amplicor HIV-1 Monitor (version 1.5; Roche Diagnostics, Meylan, France) before spotting. A total of 20 μL of each plasma sample was spotted onto the centers of 6 circles drawn on S&S903 filter paper (Schleicher & Schuell, BioScience/Whatman, GmbH, Dassel, Germany). Filters were left to dry for 1 hour at room temperature, placed individually in small plastic bags with a sachet of desiccant, stored, and mailed to France at room temperature. A period of 9 days elapsed between sample collection and the arrival of the sample at the laboratory in Rouen. On arrival, the DPS samples were transferred to a −80°C freezer, in which they were stored until testing.TABLE 1: Results of Amplification, Resistance Genotyping, and Genetic Subtyping of the 33 SamplesRNA was extracted from a DPS, as previously described.8 In the absence of amplification, the extraction was repeated with another spot but with a supplementary 10-fold concentration step, using a Microcon YM-50 device (Millipore, Saint-Quentin en Yvelines, France); briefly, 100 μL of RNA eluate was deposited into the Microcon YM-50 device, centrifuged for 2 minutes at 14,000g to obtain a concentrate of 10 μL, and recovered by centrifugation for 3 minutes at 1000g turning the Microcon YM-50 device, as recommended by the manufacturer. A nested reverse transcriptase (RT) polymerase chain reaction (PCR) assay was used to amplify the protease and RT regions of the pol gene, as previously described, yielding fragments of 507 and 798 base pairs (bp), respectively.8 A 655-bp fragment of the gp41 region encompassing the enfuvirtide (T20) target was also amplified, as follows: RNA (10 μL) was amplified in a 50-μL reaction mixture containing 2 mM of MgSO4, 20 pmol of each outer primer (1S TGGAGGAGGAGATATGAGG and AS1 GTGAGTATCCCTGCCTAACTCTAT), and 1 μL of RT/platinum Taq. Amplification reactions were carried out in a Perkin-Elmer (Courtaboeuf, France) 2400 thermocycler as follows: 1 cycle at 50°C for 30 minutes and 94°C for 2 minutes; followed by 35 cycles of 94°C for 30 seconds, 55°C for 30 seconds, and 68°C for 60 seconds; and a final extension step at 72°C for 7 minutes. A nested PCR assay was performed, as previously described for the RT and protease fragments,8 and using 20 pmol of gp41 inner primers (T20S1 GAGGGACAATTGGAGAAGTGAATT and 2AS CTACCAAGCCTCCTACTATC) with amplification conditions as follows: heating at 95°C for 15 minutes; 30 cycles of 94°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds; and a final extension step at 72°C for 7 minutes Amplification products were visualized by electrophoresis in a 1% agarose gel, with ethidium bromide staining. Resistance genotyping and genetic subtyping were carried out by sequencing and phylogenetic analysis, as previously described.8 The resistance mutations were interpreted with the Agence Nationale de Recherches sur le Sida et les Hépatites Virales (ANRS) algorithm (July 2006 version; available at: http://www.hivfrenchresistance.org/). The results obtained are presented in Table 1. The RT-PCR techniques used to amplify the protease, RT, and gp41 regions were validated on native reference plasma samples as controls. The protease region of the 33 native plasma samples could be amplified with the protocol used, whatever the VL. The RT and gp41 regions of 3 patients (P31 for RT and P21 and P27 for gp41) could not be amplified from the native reference plasma sample, and from the corresponding DPS sample. Difficulties were encountered in amplification of the gp41 region of samples from P8: no amplification was obtained with 2 of the 3 samples available. These difficulties with the amplification of gp41 in 3 patients (P8, P21, and P27) suggest that the PCR was less robust for this fragment. The protease and RT regions were successfully amplified from all DPS samples with a VL ≥5 log (21 of 21 samples), and the gp41 region was successfully amplified in 19 of 20 cases (1 native plasma sample was not amplified with our protocol; see Table 1). The use of a second spot for the amplification of gp41 was necessary for 2 patients. For samples with a VL ≥4 log and <5 log (n = 6), the protease region was successfully amplified in 100% (6 of 6) of cases, the gp41 was successfully amplified in 75% (3 of 4) of cases, and the RT region was successfully amplified in 67% (4 of 6) of cases (see Table 1). For the samples with a VL ≥3 log and <4 log (n = 6), the protease region was amplified in 50% (3 of 6) of cases and the gp41 regions was amplified in 17% (1 of 6) of cases, whereas the RT region was not amplified from any of the samples, even after re-extraction from another spot (see Table 1). A natural polymorphism of the protease gene, with no primary mutations, was identified in patients who had never been treated. In treated patients, sequencing revealed mutation profiles consistent with treatment status, with L100I, K103N, and Y188L mutations identified in patients treated with efavirenz (EFV) and M184V mutations found in patients treated with lamivudine (3TC). In 5 of the treated patients [P5, P8, P13, P15(1), and P19], an absence of resistance mutations in the RT and protease regions associated with a high VL (>5 log) suggesting poor compliance with treatment. In P8, for whom sequences were available before and during treatment, the appearance of resistance mutations, correlated with the change in treatment, was observed in the second sample (see Table 1), and the absence of mutations associated with an increase of VL in the third sample indicates poor compliance by likely stopping treatment. Subtyping by phylogenetic analysis on the available sequences (1 region [protease], n = 2; 2 regions [RT and protease or gp41], n = 3; or 3 regions, n = 18; see Table 1) showed that B-subtype viruses predominated (n = 20 of 23 patients for whom sequences were amplified [87%]; see Table 1), with only 3 patients having non-B-subtype viruses (CRF02_AG in 2 patients and CRF13_cpx in 1 patient). Our results indicate that this strategy for the therapeutic monitoring of resistance based on DPS samples obtained in the resource-limited countries and analyzed in more developed countries is feasible and could be used in field conditions on samples from patients who have never been treated or displayed ARV escape. The 9-day period between sample deposition on the filter and reception at the laboratory did not seem to affect the stability of the virus or its RNA, as shown by comparison with our previous results (storage for 14 days at 4°C).8 The sequencing results revealed a correlation between the resistance status of the patient and the treatments used. No resistance mutations were detected in populations never treated with antiretroviral drugs. In patients for whom we were able to amplify and sequence the gp41 region, the target of T20, we found a high frequency of the N42S mutation (5 [26%] of 19 patients), which was naturally present, with no implications for the algorithms currently in use. Sequential samples taken from treated patients demonstrated the feasibility of therapeutic follow-up. Optimization of our technique to achieve a detection threshold of 1000 to 5000 copies/mL, compatible with individual follow-up, is in progress by increasing the volume deposited on the filter paper (50 μL instead of 20 μL) and using extraction kits and highly sensitive enzymes. Our technique is highly suitable for molecular epidemiologic surveys of diversity and resistance and may be applied to populations and regions that cover large areas or are difficult to reach. We found that molecular typing was possible, for at least 1 gene, in 23 of the 24 patients, making possible molecular epidemiology studies of diversity and of the emergence of mosaic forms. Our results show that B-subtype viruses are circulating in Tunisia, together with 2 circulating recombinant forms (CRFs), as previously observed.9 The absence of complex mosaic forms is consistent with the recent nature of the epidemic. In conclusion, we demonstrate here the feasibility of analysis for resistance mutations for therapeutic follow-up and of molecular epidemiologic analyses of diversity based on the use of DPS samples deposited in the resource-limited countries and sent to laboratories in more developed countries by post for analysis. The samples collected in the resource-limited countries could also be sent to a local reference center, but the use of the DPS method makes it possible to store samples and to send them to the more developed countries in the absence of such a center. This technologic simplification, if combined with the serologic tools for diagnosis and quantification methods currently in place locally, would make it possible for poor countries to improve the therapeutic management of HIV infection in close collaboration with regional reference centers. ACKNOWLEDGMENTS The authors thank the laboratory technicians who participated in this work, the direction of the basic health in the Tunisian Ministry of Health, the ANRS (Paris, France), and Rouen University Hospital for financial support. Raoudha Dachraoui, PharmD*† Agnès Depatureaux, MD* Mohamed Chakroun, MD‡ Imène Fodha, PharmD, PhD† Amel Letaief, MD§ Abdelhalim Trabelsi, PharmD† François Simon, MD, PhD∥ Jean-Christophe Plantier, PharmD, PhD* *Unité de Virologie Laboratoire Associé au Centre National de Référence du VIH EA2656, CHU Charles Nicolle Rouen, France †Laboratoire de Microbiologie Unité de Virologie UR06SP20 CHU Sahloul Sousse, Tunisie ‡Service des Maladies Infectieuses Hôpital Fattouma Bourguiba Monastir, Tunisie §Service des Maladies Infectieuses Hôpital Farhat Hached Sousse, Tunisie ∥Service de Microbiologie Hôpital Saint-Louis, AP-HP Paris, France