Of the 22.5 million individuals infected with the human immunodeficiency virus (HIV) in
sub-Saharan Africa, 62% of patients requiring treatment had access to highly active
antiretroviral therapy (HAART) in 2011. The delivery of HAART and the appropriate
laboratory monitoring of HIV positive individuals in sub-Saharan African countries has
become a public health priority, an intervention which has and will continue to dramatically
reduce HIV-related morbidity and mortality. Routine laboratory monitoring of HIV infected
individuals should ideally include CD4+ T cell testing to assess when to start ART, viral load monitoring to assess virological failure on ART and when indicated, HIVDR genotyping.However, this is often not implemented in resource limited settings due to challenges such as inadequate infrastructure and laboratory capacity, amongst others. Thus the Affordable Resistance Testing for Africa (ART-A) initiative was established to develop an affordable HIV drug resistance testing (HIVDR) algorithm applicable to Africa. The objective of this study was to evaluate the role of in vitro HIVDR phenotyping in the context of HIV-1 subtype C (the most prevalent circulating subtype in sub-Saharan Africa), genotyping and genotypic interpretation tools using existing algorithms, as well as novel virological failure detection tools for clinical patient management.
Current gold standard HIVDR phenotyping technologies use an HIV-1 subtype B backbone to create recombinant viruses with patient-derived polymerase (protease and partial reverse transcriptase). This backbone could impact on the in vitro phenotyping results of non-B subtypes, and therefore it was deemed necessary to establish the applicability of HIVDR phenotypic testing of subtype C polymerase when a commercially available subtype B backbone is used. One hundred and fourteen HIV-1 subtype C samples were HIVDR phenotyped against 17 antiretroviral drugs using both subtype B and C backbones and showed a high level of concordance between the two backbone phenotypic resistance profiles (95.8%; 1590 of 1660 fold change comparisons). Natural assay variability was largely responsible for discordant results. Results confirmed that HIV-1 phenotypic reverse transcriptase inhibitor drug resistance test interpretation is independent of the virus backbone subtype. No conclusions could be made for protease inhibitor resistance since limited samples from 2nd line failure were available. Subsequently, the HIVDR genotypic and phenotypic results of the 114 patient samples were compared to determine whether genotyping is a viable alternative to phenotyping. Results showed a 92.3% concordance between genotyping and phenotyping of individual drug comparisons for a number of HIVDR profiles. Discrepancies were attributed to phenotypic assay variability in addition to the role of mutation mixtures, which impacted genotypic interpretations. Overall, HIVDR genotyping is a reliable tool to detect and interpret antiretroviral drug resistance in HIV-1 subtype C infected patients, and can thus be used for clinical patient management.
Once the accuracy of HIVDR genotyping was established, the development, validation and evaluation of a potential virological failure assay (ARTA-VFA) and a simplified HIVDR
(ARTA-HIVDRultralight) assay was undertaken. A simplified and conceptually novel approach using a qualitative viral load assay with a pre-determined cut-off that gives a threshold above which virological failure (VF) could be confirmed and below which treatment success was likely, was tested. A real-time PCR (ARTA-VFA) assay was developed which involved the amplification of a short sequence of the HIV-1 LTR region from RNA extracted either from plasma and/or dried blood spots (DBS). The ARTA-VFA was tested on 409 patient samples,and successfully amplified samples from all major HIV-1 group M subtypes with equal specificity. The VF was qualitatively classified as a viral load >1000 RNA copies/ml in plasma samples, and >5000 RNA copies/ml in DBS samples. Comparative testing yielded accurate VF determination for therapy-switching in approximately 93% of clinical cases tested, compared to current gold standard quantitative viral load assays.
A simplified HIVDR genotyping assay (ARTA-HIVDRultralight) targeting the region of RT
harboring all major RT inhibitor resistance mutation positions, thus providing all relevant
susceptibility data for first-line regimen failures was developed and assessed. The ARTAHIVDRultralight assay was designed to be practical, faster, and more affordable, show flexibility with respect to equipment (open platform), use DBS or plasma as starting material and amplify and sequence a smaller amplicon (RT). The assay performed well when compared to the in-house assay used in the laboratory at the time for both 212 plasma and 25 DBS samples, yielding identical mutations and subsequent resistant profiles. Furthermore, a theoretical in silico exercise to investigate the consequences of using 125,329 shortened RT genotype (ARTA-HIVDRultralight) as compared to full-length RT sequences showed >95% and >90% concordance when using the Stanford HIVdb algorithm and the virco®TYPE tool, respectively. Differences noted were minor and unlikely to have any impact on clinical decision-making. Overall, this study illustrated that the short RT sequences can be reliably used to generate HIVDR genotypes using the Stanford HIVdb and virco®TYPE algorithms and reduce sequencing costs substantially. A field evaluation using the ARTA-VFA and ARTA-HIVDRultralight on 288 clinical samples was conducted, showing that the accuracy and precision of both assays (using 248 plasma or 40 DBS sampling methods) compared well to the reference methodology, thereby extending access of testing to more remote settings.These assays were designed to either be used as a testing strategy of initially assessing VF,and once confirmed performing an HIVDR assay, or alternatively to be used separately as
stand-alone, or within different laboratory tiers in resource limited settings. It is envisaged
that the ARTA-VFA could be used in the middle laboratory tier, and if confirmatory, patient
samples can be referred to a reference laboratory with the available infrastructure for HIVDR testing using the ARTA-HIVDRultralight. Lastly, an automated sequence analysis and editing software for use in correct base calling of nucleotide/mutation mixtures in HIVDR genotyping was validated on 1624 sequences. Compared to reference software, where interpretation is often operator dependent, this software performed extremely well, with minor discrepancies noted. The automated software can be used to reduce subjectivity, time taken for analysis which is often the rate-limiting step and thus improving the turn-around time and clinical relevance of HIVDR genotyping.
Overall, the results obtained describe the validation of using HIVDR genotyping as an
alternative tool to phenotyping, and the subsequent development and validation of simple,
affordable, "open-platform" alternatives to currently used methods for virological failure
monitoring, and accommodate a centralized approach to HIVDR with DBS testing in
resource limited settings.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/14402 |
| Date | 28 March 2014 |
| Creators | Bronze, Michelle Saltao |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | application/pdf |
Page generated in 0.0028 seconds