Extending HIV drug resistance testing to low levels of plasma viremia: 'Reliability and Clinical Relevance of the HIV-1 Drug-Resistance Test in Patients with Low Viremia Levels'; 'Performance of HIV-1 Drug Resistance Testing at Low Level Viraemia and Its Ability to Predict Future Virologic Outcomes and Viral Evolution in Treatment-Naïve Individuals'
Download the PDF here
Download the PDF here
Download the PDF here
Download the PDF here
1st study: "We evaluated reliability and clinical usefulness of genotypic-resistance- testing (GRT), in patients failing combination-antiretroviral- therapy (cART) with viremia levels 50-1000 copies/mL, for whom GRT is generally not recommended by current guidelines......In conclusion, our study, carried out in standard clinical practice, confirms that drug resistance mutations can be detected even at low viral load, regardless of the antiretroviral target genes, and can remarkably reduce the current therapeutic options for further regimens. Our findings emphasize the importance of using the genotypic test at the first failures even at low viremia, to guide the choice of an effective alternative regimen.......Overall, success rate of amplification/sequencing was 96.4%. Viremia-levels of 50-200 and 201-500 copies/mL afforded success rates of 67.2% and 88.1%, respectively, reaching 93.2% at 501-1000 copies/mL and ≥97.3% above 1000 copies/mL.".....2nd study: "Low-level HIV viraemia (LLV; 50-999 copies/mL) occurs frequently in patients receiving antiretroviral therapy (ART), but there are little or no data available demonstrating that HIV drug resistance testing at viral loads (pVL) <1000 provides potentially clinically useful information. Here, we assess the ability to perform resistance testing by genotyping at LLV and whether it is predictive of future virologic outcomes in patients beginning ART......We have shown that routine HIV genotyping of low-level viraemia samples can be performed with a reasonably high success rate (74% in samples <250 copies/mL and 90% >250 copies/mL). In addition, we have shown that genotyping of low-level viraemia samples is predictive of future virologic outcomes in treatment-naïve patients on their first antiretroviral therapy regimen.....Patients with resistance at LLV (GSS<3) had a 2.1-fold higher risk of virologic failure (95%CI 1.2-3.7) than those without resistance (p=0.007). Progressively lower GSS scores at LLV were associated with a higher increase in pVL over time (p<0.001). .......The results obtained in this study suggest that resistance detected during LLV may be clinically relevant to future outcomes and this information should be provided to healthcare practitioners to better monitor their patients.....Here we show a strong association between risk of virologic failure and GSS scores (p-value <0.007)"
Extending HIV drug resistance testing to low levels of plasma viremia - Editorial Commentary (2 original studies below)
Clinical Infectious Diseases Advance Access published January 14, 2014
Douglas D. Richman
VA San Diego Healthcare System and University of California San Diego, Distinguished
Professor of Pathology and Medicine, Director, Center for AIDS Research
Antiviral treatment that is not sufficient to completely suppress viral replication imposes the selective pressure that results in the emergence of drug resistant virus escape mutants. HIV drug resistance testing has become part of the standard management of patients, both those newly presenting in order to identify transmitted drug resistance and those who fail to suppress virus replication, which results in the emergence of acquired drug resistance[1-3].
We have known for a long time that escape mutations can emerge early after treatment failure, even with low levels of plasma HIV RNA[4, 5]. Guidelines from most clinical laboratories advised HIV drug resistance testing only on specimens with viral loads of at least 1000 copies HIV RNA/mL[1, 3]. This practice was based on the ratioale that successful sequencing diminished in efficiency with diminishing levels of RNA, and that with smaller populations of HIV RNA molecules being tested, the results might not be representative of the variants in the circulation. Moreover, FDA approved drug HIV drug resistance genotyping platforms have specified application to specimens with >1000 copies HIV RNA/ml. Two companion papers in this issue of CID provide data from large clinical programs in British Columbia and central Italy to demonstrate that in fact the success rates for PCR amplification to perform drug resistance testing in specimens from patients with 50-1000 copies HIV RNA/mL plasma are both reasonably efficient and clinically predictive[6, 7]. Success rates were above 90% for specimens above 250 copies HIV RNA/mL in one study and for specimens above 200 copies in the other. For specimens with detectable viral loads below these levels successful sequencing results were still near 75%. In addition the paper by Santoro et al showed that similarly successful rates were obtained with non-B subtype infections.
Each of these studies provides a robust practical experience with data from a large number of subjects and plasma samples (6,617 plasma samples with 50-1000 copies HIV RNA/mL were studied between the two studies) over periods of over a decade, and they provide evidence for beneficial clinical outcomes as a result of their drug resistance testing at low levels of plasma HIV RNA. Gonzalez-Serna et al found that 8% of 212 treatment naïve patients had evidence of transmitted drug resistance. That study also showed that the detection of acquired drug resistance was predictive of subsequent treatment failure. Both studies showed a range of resistance mutations to different classes of antiretroviral drugs and the benefits of using such results to guide the design of subsequent regimens is well established[8, 9].
The results are subject to several limitations, which the authors themselves acknowledge. One concern about either phenotypic or phenotypic drug resistance testing with low viral loads has been whether the results from the amplified RNA reflected the population of genetic variants in a representative manner. Reverse transcription of any RNA population followed by PCR amplification reflects only a minority of that population. When the population tested is small, for example in the hundreds, the amplified products may include a selected subset of the population that is not representative. In fact in the Gonzalez-Serna et al study, the proportion of sequence reads without evidence of nucleotide reads with more than one base supports the likelihood of amplification of a single molecule. The concern thus remains about whether the resulting sequence reflects an unrepresentative "founder effect"; however, Figure 3 in Gonzalez-Serna et al shows a substantial rate of resistance even in those samples from which only a single molecule was likely amplified, and even these results were predictive of treatment failure.
Another issue is that these two highly experienced and specialized laboratories used "in house" methods, rather than the standard FDA-approved platforms that generate the majority of results in most developed countries. The British Columbia approach was to implement a second try on those samples that fail to generate results, initially by using alternative primers and a shorter amplicon. The value of this modification became progressively more important the lower the level of viremia in the specimen. The Italian approach was to concentrate RNA from a larger volume of plasma by centrifugation along with a nested amplification for those specimens negative on the first round. There too, the proportion of positives only after the nested PCR amplification progressively increased with lower levels of viremia. Nevertheless, both studies provide robust data showing the practical feasibility of these approaches for the clinical management of large numbers of patients. The results generated also proved useful to inform and improve clinical management. Moreover, results obtained earlier during treatment failure can help guide modifications of regimens before additional resistance accumulates. These studies thus provide an impetus for investigators designing drug resistance assays and for laboratories providing diagnostic services to extend drug resistance assays to specimens with lower levels of viremia.
Implementing the conclusions from these two studies will expand the proportion of patients who will benefit from HIV drug resistance testing. In addition, by earlier detection of failure and resistance, interventions can be made before the continuing accumulation of resistance would inevitably occur. As we learn how better to manage failure, it is ironic, but not unwelcome, that with the availability of earlier treatment and with more effective and more tolerable drugs, the proportion of patients failing treatment is substantially diminishing in developed countries.
The increasing challenges regarding drug resistance are occurring in resource-limited settings with the global rollout of antiretroviral therapy now reaching approximately 10 million people. Substantial benefits on morbidity, mortality, and likely even transmission are resulting from this effort[10, 11]. Nevertheless, the reality of laboratory support in such settings is compromising some of the benefits of the rollout. Current antiretroviral management in resource limited settings results in the much delayed determination of treatment failure and the progressive accumulation of drug resistance mutations under the selective pressure of continuing drug treatment[12, 13]. This accumulation results in higher and broader antiretroviral resistance, thus diminishing effective treatment options. These options are further restricted by second line regimens that are much more limited than in resource rich countries. There is an urgent need for practical virus load assays and drug resistance assays that will permit more prompt and informed detection and management of treatment failure to improve further the benefits of access to antiretrovirals in the settings where most infections occur.
The author has no reported conflicts of interest.
Reliability and Clinical Relevance of the HIV-1 Drug-Resistance Test in Patients with Low
Clinical Infectious Diseases Advance Access published January 14, 2014
Maria Mercedes Santoro1, Lavinia Fabeni2, Daniele Armenia1, Claudia Alteri1, Domenico Di Pinto1, Federica Forbici2, Ada Bertoli1,3, Domenico Di Carlo1, Caterina Gori2, Stefania Carta2, Valentina Fedele2, Roberta D'Arrigo2, Giulia Berno2, Adriana Ammassari2, Carmela Pinnetti2, Emanuele Nicastri2, Alessandra Latini4, Chiara Tommasi2, Evangelo Boumis2, Nicola Petrosillo2, Gianpiero D'Offizi2, Massimo Andreoni1,3, Francesca Ceccherini-Silberstein1, Andrea Antinori2, and Carlo Federico Perno1,2,3
1University of Rome Tor Vergata, Rome, Italy
2L. Spallanzani Hospital, Rome, Italy
3University Hospital Tor Vergata, Rome, Italy
4San Gallicano Hospital, Rome, Italy
The manuscript was in part presented at the 11th European Meeting on HIV and hepatitis treatment and strategies and antiviral drug resistance, 20 - 22 March 2013, Rome, Italy (Abstract O_23); and at the XX International Workshop on HIV and hepatitis virus drug resistance and curative strategies, June 4 - 8 2013, Toronto, Canada - Antiviral Therapy 2013 (18 Suppl 1:A128).
Background: We evaluated reliability and clinical usefulness of genotypic-resistance- testing (GRT), in patients failing combination-antiretroviral- therapy (cART) with viremia levels 50-1000 copies/mL, for whom GRT is generally not recommended by current guidelines.
Methods: Genotyping-success-rate was evaluated in 12828 HIV-1 plasma-samples with viremia >50 copies/mL, tested using the commercial ViroSeq HIV-1 Genotyping-System or a homemade-system. Samples were stratified in 6 groups according to different viremia-levels (50-200; 201-500; 501-1000; 1001-10000; 10001-100000; >100000 copies/mL). Phylogenetic analysis was performed to test the reliability and reproducibility of the GRT also at low-viremia-levels. Drug-resistance was evaluated in 3895 samples from 2200 treatment-failing-patients (viremia >50copies/mL) by considering the resistance-mutations panelled in the IAS list (2013).
Results: Overall, success rate of amplification/sequencing was 96.4%. Viremia-levels of 50-200 and 201-500 copies/mL afforded success rates of 67.2% and 88.1%, respectively, reaching 93.2% at 501-1000 copies/mL and ≥97.3% above 1000 copies/mL. Phylogenetic analysis revealed a high homology among sequences belonging to the same subject for 96.4% of patients analyzed. The overall resistance prevalence was 74%. Drug-resistance was commonly found also at low-viremia levels. Detection of at least one resistance-mutation was: 50-200 copies/mL=52.8%; 201-500=70%; 501-1000=74%; 1001-10000=86.1%; 10001-100000=76.7%; >100000=63% (P<0.001). Similar bell shaped results were found when the GRT-analysis was restricted to 2008-2012, though at slightly lower prevalence.
Conclusions: In patients failing cART with viremia-levels 50-1000 copies/mL, HIV-1 genotyping provides reliable and reproducible results, that are informative about emerging drug-resistance also at low-viremia-levels. Results may be helpful for the therapy optimization in patients under virologicalfailure, to decrease the risk of virological failures with drug-resistance accumulation.
Over the past 15 years, antiretroviral therapy for the treatment of human immunodeficiency virus type-1 (HIV-1) infection has improved; to date, about 90% of HIV-1 infected patients who start a first-line regimen achieve virological suppression [1-10]. However, therapy failures are still observed in clinical practice; particularly at early time points, many are characterized by low viremia levels (LLV). Standard of care management recommends use of resistance testing to guide further therapy. One area of uncertainty is the evaluation of treatment failure in patients with LLV. Treatment guidelines usually do not recommend genotypic resistance testing (GRT) for plasma HIV RNA <500-1000 copies/mL. This potential limitation of GRT mostly derives from the detection limits of commercial assays, as well as by the technical difficulty of many laboratories in obtaining consistent results with such LLV, yet some studies support the use of GRT, and laboratories increasingly report success in performing genotypes at this level [11-26].
In this study we provide data supporting reliability and usefulness of GRT at viremia levels ≤500-1000 copies/mL by analyzing a large population of HIV-1 patients followed in Central-Italy, who underwent GRT in routine clinical practice. Moreover, we evaluated whether different viremia levels affect the detection of drug-resistance in HIV-1 patients who failed therapy.
MATERIALS AND METHODS
This retrospective study included 13926 HIV-1 plasma samples that were genotyped over the years 1999-2012 in two clinical centers in Rome (Italy) for routine clinical purposes. Sample information (date of sampling, final results of sequencing, nucleotide sequences obtained, mutations found in each sequence), together with the data of patients for whom genotyping was performed (i.e., viroimmunological, clinical, and therapeutical data) were recorded in an anonymous database. For each sample, viremia value at genotyping was known. We focused our analyses on samples with viremia >50 copies/mL (N=12828) that were stratified in six groups according to different viremia ranks (copies/mL): 50-200, 201-500, 501-1000; 1001-10000, 10001-100000, >100000.
HIV-1 RNA viral load
Depending on methodologies available over years 1999-2012, plasma viremia was determined using three different assays: the bDNA v3.0 (until January 2009; Bayer Corporation, Diagnostics Division, Tarrytown, New York), the Abbott RealTime HIV-1 (February 2009-February 2012; Chicago, Illinois) and the Roche Cobas CA/CTM v2.0 (starting from March 2012; Mannheim, Germany). These assays quantify HIV-1 RNA over the range of 50-500000 copies/mL, 40-10000000 copies/mL and 20- 10000000 copies/mL, respectively. Previous studies demonstrated the results obtained by these assays to be well correlated, with a difference >0.5 log10 copies/mL, only for few samples [27-29].
HIV-1 pol sequencing
HIV-1 genotype analysis was performed on plasma samples by using either the ViroSeq HIV-1 genotyping system (Abbot Molecular) and/or a homemade system, designed to improve the performance of the ViroSeq system itself . Indeed, genotyping success by this commercial kit is generally guaranteed for samples with viremia ≥2000 copies/mL [31, 32]. Therefore, some steps of the ViroSeq system were modified, in order to test HIV-1 pol sequences also in subjects with viremia <2000 copies/mL. All the details on the amplification and sequencing procedure can be found in Supplementary Methods and Supplementary Figure 1.
All HIV-1 pol sequences were aligned in Bio-Edit and compared to reference sequences for major HIV-1 subtypes and circular recombinant forms (CRFs), available at Los-Alamos database (http://www.hiv.lanl.gov); a phylogenetic tree was performed. To analyze trends in subtype genetic diversity over time, genetic distances were calculated by using maximum-likelihood method of MEGA (http://www.megasoftware.net/), by using Kimura two-parameter model as the best-fitting evolution model for tree reconstruction . The tree was shown by using the graphical user-interface FigTree. Subtype classification was confirmed also by the REGA subtype tool (http://www.bioafrica.net/regagenotype/html/subtypinghiv.html), the COMET subtype tool (http://comet.retrovirology.lu/) and the
DataMonkey subtype tool (http://www.datamonkey.org/dataupload_scueal.php). To improve the accuracy of recombinant and unique forms, RDP3 software (http://web.cbio.uct.ac.za/~darren/rdp.html) and Splits Tree software (http://en.bio-soft.net/tree/SplitsTree.html) were used.
Evaluation of genotypic success rate and genotyping reliability
Genotyping success rate was determined on the overall population and according to the different viremia ranks (50-200, 201-500, 501-1000; 1001-10000, 10001-100000, >100000 copies/mL), regardless the genotyping platform upgrades (equipment, kits and reagents) that occurred from 1999 to 2012.
To ensure that there was no cross-contamination of samples analysed and in order to test genotyping reliability for samples with viremia ≤500 copies/mL, a phylogenetic analysis was performed on a subgroup of 1613 pol sequences, obtained from 470 patients with at least 1 GRT performed on samples with viremia ≤500 copies/mL and at least 1 GRT with viremia >1000 copies/mL. The phylogenetic analysis of pol sequences was performed by using the Kimura two-parameter model of MEGA version 5.05, with the same parameters as previously described .
Evaluation of resistance in patients who had failed therapy
The prevalence of drug-resistance was evaluated, and stratified according to different viremia levels, in a subset of 3895 samples successfully genotyped from 2200 patients with complete therapeutic history, for whom a GRT was required because of virological failure (defined as viremia >50 copies/mL). Resistance to an antiretroviral drug class was defined by the presence of at least one primary resistance mutation (PRM) included in the mutation list panelled by the International Antiviral Society in 2013 , considering the nucleos(t)ide RT inhibitors (NRTIs), non-NRTIs (NNRTIs) and PR inhibitors (PIs). In particular, we have defined: i) the resistance to any drug-class in the overall samples analysed; ii) the resistance to NRTIs among samples from patients who received regimens that contained NRTIs; iii) the resistance to NNRTIs among samples from patients who received regimens that contained NNRTIs; iv) the resistance to PIs among samples from patients who received regimens that contained ritonavir-boosted PIs (PI/r); v) the resistance to PIs among samples from patients who received regimens that contained ritonavir-unboosted PIs.
To better understand the clinical relevance of GRT in patients failing with LLV at the time of modern anti-HIV therapies, the prevalence of single PRMs was also evaluated on the 1317 samples from patients for whom a GRT was required because of virological failure in the years 2008-2012. This analysis was performed by dividing the samples into two groups according to viremia levels ≤1000 copies/mL (N=436) or >1000 copies/mL (N=881).
Patient outcome analysis
To evaluate the effect of LLV resistance on subsequent virologic outcome, further analyses were restricted to 51 previously drug-naïve patients at first-line regimen for whom a GRT was requested at viremia levels of 50-1000 copies/mL. Patients were included only if they were followed as long as they were receiving constant therapy without any changes or interruptions.
Table 1 shows the characteristics of 12828/13926 plasma-samples with viremia >50 copies/mL, processed for genotyping in routine clinical practice from 1999 to 2012. Among them, 4861 (37.9%) were obtained from 4111 drug-naïve patients, and 7967 (62.1%) from 3841 drug-experienced patients. Among drug-experienced patients, viremia levels of 50-1000 copies/mL accounted for 19.2%, (1535/7967) of total genotypic requests (Figure 1). This prevalence significantly increased over time from 1.5% in 1999-2001 to 28.4% in 2012 (P<0.001). A consistent proportion of samples with LLV was with viremia 50-500 copies/mL (1158/1535, 75.4%, versus 377, 24.6%, with viremia 501-1000 copies/mL).
Phylogentic analysis revealed that B subtype was the most prevalent strain (80.1%). All the other subtypes were present with a prevalence <5%; the most prevalent ones were the recombinant form CRF02_AG (4.7%) and the subtypes C (4.3%) and F (3.3%).
Genotyping success rate
Overall success of genotype amplification and sequencing was 96.4%. The rate of success was 93.2% for samples with viremia levels 501-1000 copies/ml, 88.1% for those with viremia 201-500 copies/mL, and decreased to a still relevant 67.2% for viremia 50-200 copies/mL (Table 2). Genotyping success rate was independent of subtype in all viremia groups (Table 2). By focusing the attention on the three most prevalent non-B subtypes analysed (C, F, CRF02_AG), no differences in the success rate were found (data not shown).
Interestingly, the additional use of a nested PCR (or modified amplification protocol; see Supplementary Methods and Supplementary Figure 1) has significantly improved the overall success rate in samples with LLV (P<0.001). In particular, the nested amplification contributed to 60.4%, 55.3% and the 44.0% of the total genotypic successes with viremia levels 50-200, 201-500 and 501-1000 copies/mL, respectively. In samples with viremia levels >1000 copies/mL, the contribution of nested amplification was less relevant (from 19.2% to 3.6%, data not shown).
In order to test genotyping reliability for samples with VL ≤1000 copies/mL, we performed phylogenetic analysis on 1613 sequences from 470 patients having at least one genotypic sample with viremia 50-1000 copies/mL and at least another with viremia >1000 copies/mL. By evaluating each cluster, we found that sequences belonging to the same subject showed a high homology (bootstrap value >90%) in 96.4% of cases (453/470 patients) (Supplementary Figure 2). Only 25/1613 sequences (1.5%) of the remaining 17 patients did not properly cluster within the same subject.
Evaluation of resistance according to different viremia ranges in patients failing therapy
Prevalence of PRMs was analyzed on 3895 samples from a subgroup of 2200 patients at therapy failure. Patients' characteristics of this subgroup are reported in Supplementary Table 1. Overall, the median (interquartile range, IQR) year of genotyping was 2006 (2003-2009) and the proportion of samples from subtype B infected patients was about 86%.
The overall prevalence of samples with at least one PRM was 74% (Table 3). PI-resistance in patients treated with ritonavir-boosted PI (PI/r) was in general less frequent than NRTI- or NNRTI-resistance (40.5%, vs. 66% and 77.7%, P<0.001) (Table 3).
If we consider PI-resistance only in patients who had failed at first-line regimen containing a PI/r, the rate of resistance dropped dramatically to 3.7%. By contrast, PI-resistance in patients treated with unboosted-PI was more similar to that of NRTI/NNRTI (61.7%) and remained high also among patients tested at first-line failure (46.6%).
The prevalence of resistance varied significantly by viremia strata (P<0.001), and was characterized by a bell-shaped curve in which the highest prevalence was in the 1001-10000 copies/mL stratum, with lower prevalence values at lower and higher viremia strata. Detection of resistance was consistent also at LLV. In particular, for viremia levels of 50-200 copies/mL, NRTI-resistance was 41.3%, NNRTI-resistance was 40.2%, PI/unboosted-resistance was 51.6% and PI/r-resistance was 20.8%. For viremia 201-500 copies/mL, rates of resistance were 62.3%, 69.3%, 30.8%, and 28.0% respectively, that increased, for viremia 501-1000 copies/mL, to 67.1%, 79.5%, 79.2%, 39.0% for each respective drug class (Table 3). Therefore, substantial levels of resistance can be detected also at LLV for all drug classes, with higher rates for NRTI and NNRTIs.
The distribution of drug-resistance stratified for viremia was similar also considering samples only from patients failing their first-line regimen. In particular, a consistent proportion of NRTI and NNRTI resistance was found also at viremia levels 50-1000 copies/mL, while PI-resistance was very low in samples from patients failing their first-line PI/r containing regimen (for viremia 50-200 copies/mL: NRTI-resistance, 19.2%; NNRTI-resistance, 13.6%; PI/r-resistance, 4.9%; for viremia 201-500 copies/mL: NRTI-resistance, 38.3%; NNRTI-resistance, 54.5%; PI/r-resistance, 0%; for viremia 501-1000 copies/mL: NRTI-resistance, 59.5%; NNRTI-resistance, 73.3%; PI/r-resistance, 7.1%).
The resistance to NRTI and NNRTI varied according to viremia strata also by restricting the analysis over the years 2008-2012, with a still considerable prevalence of resistance in samples with viremia levels ≤1000 copies/mL (Figure 2). By contrast, the prevalence of PI-resistance was not influenced by viremia strata because it was very limited among all failures and was almost zero in patients failing their first-line PI/r containing regimen.
Finally, by characterizing the prevalence of each single PRM in samples genotyped over the years 2008-2012, no major differences were found by analyzing samples with viremia ≤1000 vs. >1000 copies/mL (Supplementary Table 2). In particular, only the NNRTI PRM K103N was found with a significantly higher prevalence in patients failing with viremia >1000 copies/mL (43.3%) vs. ≤1000 copies/mL (20.2%, P<0.001, after multiple comparison correction).
By Kaplan-Meier analysis, we found that the probability of reaching viremia >1000 copies/mL after LLV was significantly higher in patients with resistance than in those without resistance, as follows: at 24 weeks, 49.7% versus 4.2%; at 48 weeks, 58.1% versus 8.7%; at 72 weeks, 72.1% versus 15.2% (P<0.001, data not shown).
This study aimed at evaluating the reliability and usefulness of GRT in HIV-1 infected patients, with detectable LLV in a large dataset of samples tested in two clinical centers in Italy. Our results showed that the genotyping success rate was 96% for the overall population. In particular, this success rate was very high also for viremia above 200 copies/mL (about 88%), reaching about 93% at 501-1000 copies/mL and greater than 97% above 1000 copies/mL. Reasonable results in terms of success rate were obtained also for samples with viremia between 50 and 200 copies/mL. The ability to easily detect samples with LLV is mainly due to the improvement of the amplification step performed in our laboratories. The success of sequencing was very similar in B and non-B strains, thus suggesting that the subtype diversity does not represent a limit. Our findings are in agreement with those recently obtained in other studies, showing a high success of amplification and sequencing also at LLV [16, 19, 21, 26]. Our results with LLV may not reflect the true population, but rather reflect founder effects, especially when nested amplification is needed. Nevertheless, phylogenetic analysis confirmed the reliability and reproducibility in our laboratories of genotypic tests at different viremia levels. Indeed, by evaluating 1613 pol sequences obtained from 470 patients with at least two GRTs performed at different times and with different viremia levels (ranging from <50 to >100000 copies/mL), very high similarity among sequences from the same patient was observed.
It should be emphasized that the additional step of the nested PCR does not affect the total cost of genotyping test because the reagents used (see Supplementary Methods) are inexpensive. Indeed, by adding the Nested PCR step, the total amount of HIV-1 genotyping costs is increased only of about 10-15 Euros per sample performed. Therefore, we can conclude that the use GRT for treatment optimization in HIV-infected patients with treatment failure at LLV is in any case cost-effective.
The clinical relevance of our findings is related to the fact that in the last few years there has been an increased demand for GRTs for drug-experienced patients failing with LLV (mainly ≤500 copies/mL, as shown in our analysis; see Figure 1), explained by a greater tendency to closely monitor patients in terms of response to treatment and drug-resistance. In our dataset the proportion of requests from patients failing with LLV has been about 30% since 2009.
Moreover, our results corroborate the already discussed recruitment about drug-resistance presence also at viremia levels ≤1000 copies/mL [26, 35, 36, 39], underlining the importance of GRTs also at LLV for the optimization of therapy in patients under virological failure. In this regard, it should be emphasized that the optimization of the sequencing protocol in the last years has led to a higher accuracy in detecting the PRMs for each viremia level. In our study, a considerable prevalence of resistance was found also at LLV among the samples analyzed from patients failing therapy. This finding proves that the detection of drug-resistance is not a rare event in these low viremia ranges.
A decline in the prevalence of PRMs was observed also at the very high viremia strata among drug experienced individuals. This decline is likely to reflect suboptimal medication adherence, with lower drug resistance selection .
A considerable prevalence of resistance to NRTIs and NNRTIs at LLV was found also when the analysis was restricted to 1317 samples from patients failing therapy in the last few years. This prevalence can be due to the large usage of low genetic barrier drugs such as lamivudine/emtricitabine or efavirenz/nevirapine. By the evaluation of the effect of LLV resistance on subsequent virology outcome, we found that the probability of reaching viremia >1000 copies/mL by 72 weeks after LLV was significantly higher in patients with resistance than in those without resistance (). This strongly suggests that the early detection of resistance (when viremia is still below 1000 copies/mL) may prevent the evolution toward a) a virological failure with higher viremia and b) the accumulation of additional mutations, thus affecting the choice of future therapeutic regimens. A potential limitation of this analysis could be it was performed only on a very small dataset of patients. In line with our data, a recent study, performed in a larger cohort of patients, confirmed that the LLV resistance is predictive of subsequent virological failure . Taken together these results reinforce the concept that GRT may be useful in the management of failure even at LLV.
Data presented in our study, in agreement with previous articles [35, 36] and with data recently presented [26, 38, 39], suggest that newer guidelines may reconsider the importance of GRT in clinical practice even at LLV. Indeed, in spite of the technical improvements achieved in the last few years, treatment guidelines still do not usually recommend GRT in patients with a plasma viral load ranging between >50 and 1000 copies/mL [2, 4].
In conclusion, our study, carried out in standard clinical practice, confirms that drug resistance mutations can be detected even at low viral load, regardless of the antiretroviral target genes, and can remarkably reduce the current therapeutic options for further regimens. Our findings emphasize the importance of using the genotypic test at the first failures even at low viremia, to guide the choice of an effective alternative regimen.
Performance of HIV-1 Drug Resistance Testing at Low Level Viraemia and Its Ability to Predict Future Virologic Outcomes and Viral Evolution in Treatment-Naïve Individuals
Clinical Infectious Diseases Advance Access published January 14, 2014
A Gonzalez-Serna1,*, JE Min1, C Woods1, D Chan1, V Lima1, JSG Montaner1,2, PR
Harrigan1,2, LC Swenson1
1BC Centre for Excellence in HIV/AIDS, Vancouver, Canada
2Division of AIDS, Department of Medicine, University of British Columbia
Summary: This study shows that routine HIV genotyping of low-level viraemia samples can be performed with a reasonably high success rate and the results are predictive of future virologic outcomes and viral evolution in treatment-naïve individuals.
Background: Low-level HIV viraemia (LLV; 50-999 copies/mL) occurs frequently in patients receiving antiretroviral therapy (ART), but there are little or no data available demonstrating that HIV drug resistance testing at viral loads (pVL) <1000 provides potentially clinically useful information. Here, we assess the ability to perform resistance testing by genotyping at LLV and whether it is predictive of future virologic outcomes in patients beginning ART.
Methods: Resistance testing by genotyping at LLV was attempted on 4915 plasma samples from 2492 patients. A subset of previously ART-naïve patients was analyzed who achieved undetectable pVL and subsequently rebounded with LLV (N=212). A genotypic sensitivity score (GSS) was calculated based on therapy and resistance testing results by genotyping, and stratified according to number of active drugs
Results: 88% of LLV resistance assays produced useable sequences, with higher success at higher pVL. Overall, 16/212 (8%) patients had pre-therapy resistance. 38/196 (19%) patients without pre-therapy resistance evolved resistance to 1 or more drug classes – primarily the nRTIs (14%) and/or NNRTIs (9%). Patients with resistance at LLV (GSS<3) had a 2.1-fold higher risk of virologic failure (95%CI 1.2-3.7) than those without resistance (p=0.007). Progressively lower GSS scores at LLV were associated with a higher increase in pVL over time (p<0.001). Acquisition of additional resistance mutations to a new class of antiretrovirals during LLV was not found in a subset of patients.
Conclusions: Routine HIV genotyping of low-level viraemia samples can be performed with a reasonably high success rate and the results appear predictive of future virology outcomes.
An increasing proportion of persons living with HIV are receiving suppressive antiretroviral therapy (ART) [1, 2]. However, despite stable ART, many patients experience episodes of low-level viraemia (LLV), defined as plasma viral load (pVL) measurements between 50 and 1000 HIV-RNA copies/mL which may include repeated pVL episodes, intermittent pVL and blips. An increased risk of virologic failure has been associated with episodes of LLV in several studies [3-6], but not in others [7, 8]. In addition, LLV episodes have been associated with higher immune activation [9, 10] and even possible increased mortality . A main factor in increased risk of virology failure appears to be the accumulation of drug resistance mutations, either released from stable HIV reservoirs  and/or from ongoing cycles of replication .
However, data on resistance during low-level viraemia is limited, in part because FDA-approved genotypic HIV resistance assays require at least 1,000 copies/ml (TRUGENE HIV-1 Genotyping Assay) or 2000 copies/mL (ViroSeq® HIV-1 Genotyping System) . Another concern in the possibility that amplification of a very small number of HIV copies may render inaccurate genotypic results. Thus, studies of antiretroviral resistance during low-level viraemia on patients receiving first-line ART are scarce . The aims of our study were to evaluate emergence of HIV drug resistance mutations during periods of low-level viraemia, assess the ability to successfully sequence them using an in-house assay and evaluate the association of LLV resistance with subsequent virologic outcomes in a cohort of patients beginning their first antiretroviral therapy regimen.
Materials and Methods
We evaluated genotype success rates of all HIV-infected adults who enrolled in the British Columbia (BC) Drug Treatment Program between 1996 and 2012 with any detectable plasma viral load (pVL) <1000 copies/mL by Roche COBAS® Amplicor™ HIV-1 Monitor® Test v1.5 (detection limit of 400 copies/mL from 1996 to 1999, and 50 copies/mL from 1999-2009) or Roche COBAS TaqMan HIV-1 v1.0 or v2.0 (detection limit set to 50 copies/mL, and in use from 2009-2012). A total of 4915 results were obtained from samples with detectable pVL<1000 copies/mL. These samples came from a total of 2492 patients. Many of these results (27%) were obtained retrospectively for research purposes on stored specimens, since the cutoff for ordering a low-level viraemia resistance genotype was initially 500 copies/mL, lowered subsequently to 250 copies/mL, though samples with pVL <250 could be tested by physician request.
HIV RNA extraction and drug resistance analysis
Drug resistance testing was performed on physician-requested samples with pVL as defined above. In addition to changes in the viral load assay over the years, various methods and equipment have also been used for resistance genotyping in BC. From 1998 to 2006, HIV RNA was manually extracted from frozen plasma samples using guanidinium based lysis buffer followed by isopropanol/ethanol washes (Qiagen), or automatically using a BioRobot (Qiagen), and from 2006 to 2012 by automated extraction using a NucliSENS easyMAG (bioMerieux). Amplification of the protease (PR) and reverse transcriptase (RT) regions was performed using nested RT-PCR followed by sequencing in both the 5' and 3' directions on an ABI 3100, 3130, or 3700 sequencer from 1996 to 2006, and an ABI 3730 sequencer from 2006 to 2012. Primers used span all of protease, and to codon 400 of RT (Primary PCR product). Amplification was repeated with different primers (spanning to codon 250 of RT; "Backup PCR product") when the first attempt was unsuccessful. Although a second PCR attempt must be made, this "backup" method simply uses a different primer set which spans a smaller region, so is no more resource intensive than the primary method. A test was reported as failed when a second attempt with re-extraction and back-up primers was unsuccessful. Sequence data were analyzed using Sequencher (Genecodes) from 1996 to 2007 and RECall (BC Centre for Excellence in HIV/AIDS) automated sequencing software from 2007 to 2012 [15-17]. Nucleotide mixtures were identified if the secondary peak height exceeded approximately 17.5% of the dominant peak height.
Genotypic Sensitivity Scores
The genotypic sensitivity score (GSS) was obtained using the Stanford HIVdb genotypic resistance interpretation system . The Stanford algorithm generates five levels of resistance to a drug, ranging from fully susceptible (i.e., wild-type), low to intermediate resistance, and high level resistance. Here, we assigned a GSS value of 1 to each drug categorized as susceptible, potential low-level resistance and low-level resistance; a GSS value of 0.5 to the intermediate resistance category; and a GSS value of 0 to the high-level resistance category. The GSS values for all drugs in a regimen were added together to give a final total GSS. Patients were grouped into 4 categories depending on their GSS scores at LLV, corresponding to the number of active drugs prescribed: 0-0.5; 1-1.5; 2-2.5; and ≥3. In a subsequent analysis, patients were grouped into resistant (GSS<3) and not resistant (GSS≥3) categories. The virtual phenotypic sensitivity score Virco®TYPE HIV-1 (vPSS) was also used to re-evaluate the results .
Patient outcome analysis
To evaluate the effect of LLV resistance on subsequent virologic outcome, further analyses were restricted to previously ART-naïve patients who achieved undetectable viral loads but whose virus rebounded with repeated pVL between 50-999 copies/mL. Patients were included only if they had not had a previous blip (≥1000 copies/mL), and they were followed as long as they were receiving constant therapy without any changes or interruptions. Many of these results (24%) were obtained retrospectively.
Overall, 4312 of 4915 (88%) low-level viraemia assays attempted produced usable sequences (Table 1). When ≥2 different viral strains are amplified within the same sample >1 mixtures will be found in the final sequence. Successful genotypes were obtained more frequently at higher pVL strata approaching 1000 copies/mL. These higher viral load samples tended to have more observed sequence mixtures, suggesting >1 viral input copies amplified from the samples. Successful results were obtained from 74% of samples with viral loads <250 c/mL and from approximately 90% of samples with viral loads above 250 c/mL. Results were similar regardless HIV subtype (data not shown). Unsuccessful genotypes and the use of Backup PCR product progressively increased with decreasing viral load (Figure 1a). In addition, we analyzed whether the age of the sample (time from collection to testing) affected the success in testing finding that the longer age of the samples the success in testing was slightly lower (Figure 1b). However, even in samples stored at -20ľC for more than four years, success in testing remained around 70-80%.
Patient characteristics and resistance testing by genotyping
Patient characteristics are shown in Table 2 for 212 subjects previously drug naïve who rebounded with LLV during their first ART regimen. At time of resistance testing, patients had moderately high CD4 counts (median 415 cells/mm3; 25th-75th percentile: 260-580 cells/mm3) and low pVL (median 374 copies/mL; 25th-75th percentile: 267-559 copies/mL). Resistance testing by genotyping before treatment (baseline) and at time of genotype testing are shown in Table 3. Overall, 16 of the 212 patients (8%) had baseline resistance prior to therapy. Of those without baseline resistance (N=196), 38 patients (19%) evolved resistance to any class of medication at follow up with LLV, a median of 6.9 months (25th-75th percentile: 3.3–18) after their pVL became undetectable. In these patients, resistance was most common to the NRTI (N=28; 14%) and/or NNRTI (N=18; 9%) drug classes. Of note, only 2 cases (1%) of emerging PI mutations arose (D30N), both in patients taking nelfinavir, despite 67% of patients receiving a PI. No patients evolved triple-class drug resistance during the study period. The most common mutations found at LLV were M184V/I (10%V, 4%I), K103N (6%), T215Y/F/C/D/E/S/I (4%), M41L (4%), Y181C (3%), K70R/E (3%) and T69D/N/S (3%).
Before treatment, baseline (i.e., transmitted) resistance was marginally more common in patients who were male (p=0.02) and slightly older (median 46 vs 43 years; p=0.04). Analyses were then further restricted to the 196 patients without baseline resistance. Those patients who evolved resistance at LLV had a non-statistically significant trend towards higher viral load levels at LLV versus those who did not evolve resistance (median 472 vs 369 copies/mL, p-value=0.067). Moreover, we observed that the prevalence of resistance increased at higher viral load strata at LLV. Only 5% of patients (N=2) with 50-249 c/mL at LLV had resistance whereas 24% (N=22), 17% (N=7) and 30% (N=7) had LLV resistance at 250-499, 500-749 and 750-999 c/mL respectively (p-value=0.041). Other patient characteristics (risk group, HCV coinfection, CD4, ethnicity, time from undetectable viraemia to LLV) were not associated with resistance at LLV (p-value ≥0.1).
Kaplan-Meier curves were used to evaluate time to virologic failure (≥1000 copies/mL) while remaining on the same therapy. Figure 2A indicates that patients with resistance (GSS<3) at LLV had significantly increased risk of subsequent virology failure compared to those without resistance (GSS≥3) (p=0.007). Furthermore, linear mixed-effect models showed that progressively lower GSS scores at LLV are were significantly associated with an increased change in median pVL over time (overall p-value <0.001) (Figure 2B).
Bivariate analysis of patient characteristics associated with virologic failure after LLV indicated that higher pVL at LLV (p=0.02), history of injection drug use (p=0.03), HCV coinfection (p=0.04) and female gender (p=0.05) were associated with an increased likelihood of a subsequent pVL ≥1000 copies/mL. To determine predictors associated with the hazard of virologic failure, unadjusted and adjusted Cox proportional hazards models were applied to all the variables related to virologic failure. Among covariates considered, only gender and pVL at LLV were included in the final model. Adjusted hazard ratios were 2.12 (95% CI 1.23-3.66) for patients with resistance at LLV (GSS<3), 2.34 (95% CI 1.11-4.93) for patients with pVLs 500-749 copies/mL vs. 50-249 copies/mL and 1.64 (95% CI 0.97-2.78) for patients reporting female gender. The other pVL strata had p-values>0.1 compared with pLVs 50-249 copies/mL.
Additionally, in a subset of 29 patients maintaining LLV with follow-up resistance test results, there was no evidence of acquisition of additional resistance mutations to a new class of antiretrovirals (data not shown) suggesting that the selection of resistance to an additional family drug during a LLV episode on the same regimen may not be a common event, though the follow-up times were relatively short (8 months). Moreover, when the evolutionary distance between the first and last genotype was analyzed by the TN93 model  a slightly trend towards increased genetic diversity over time was observed. However, this did not reach statistical difference, probably because of the low number of patients (data not shown). This trend may suggest that even with no apparent resistance evolution, HIV-1 may be able to evolve at LLV.
Data were also stratified by whether or not any nucleotide mixtures were observed in the sequence chromatograms. An absence of mixtures suggested that only a single molecule may have been successfully reverse transcribed and amplified during sample processing. As is shown in Figure 3, the time to virologic failure curves are different between patients with resistance and no resistance, whether or not mixtures were observed (p=0.007). Cox proportional hazards model adjusting for an indicator for mixture still showed a significant p-value for resistance (p-value=0.007), implying that even clonal products with only one sequence (no mixtures) gave useful results for predicting virologic failure. Furthermore, all the above analyses were performed using Virco®TYPE virtual phenotype interpretation instead of the Stanford algorithm, and very similar results were obtained (Supplementary data).
We have shown that routine HIV genotyping of low-level viraemia samples can be performed with a reasonably high success rate (74% in samples <250 copies/mL and 90% >250 copies/mL). In addition, we have shown that genotyping of low-level viraemia samples is predictive of future virologic outcomes in treatment-naïve patients on their first antiretroviral therapy regimen.
Diagnosis and management of emerging drug resistance during LLV is a clinical challenge, as FDA-approved genotyping assays require at least 1000 copies/ml  and some standard genotypic tests (including ours) have higher failure rates at amplifying HIV-1 RNA at low-level viraemia . However, the in-house PCR method used in this study shows a high success rate for genotyping LLV samples (about 74% at 50-249 copies/mL and 90% at 250-999 copies/mL) suggesting that genotyping can be performed routinely on clinical low-level viraemia samples. Actually, based on the pVL and the volume used for testing, genotypes sequences were obtained ranging from 2 to 33 actual input copies. In addition to lower assay success in samples with especially low pVL levels, we found that the longer time elapsed between sample collection and testing also negatively affected success of genotyping. However, success was still relatively high, at 70-80%, even for samples older than 4 years old. The relatively high success rate of resistance genotyping at low-level viraemia has also been demonstrated by an independent laboratory, who report strikingly similar success rates to ours .
The results obtained in this study suggest that resistance detected during LLV may be clinically relevant to future outcomes and this information should be provided to healthcare practitioners to better monitor their patients. In fact, use of resistance testing by genotyping testing has been shown to improve virologic outcomes in several prospective studies of patients failing ART [23-25]. To infer resistance in this study, we used the Stanford HIVdb genotypic resistance interpretation system . Other rules based systems as ANRS, Rega and AntiRetroScan are available, and similar results are often obtained regardless of the system used [26, 27]. In addition, all our results were confirmed by using the Virco®TYPE HIV-1 as an alternative classification algorithm. Although some studies have reported that GSS had no impact on viral responses [28-30], several other investigations support genotypic or phenotypic sensitivity scores as predictors of viral response [4, 26, 27,31, 32].
Here we show a strong association between risk of virologic failure and GSS scores (p-value <0.007). This is in agreement with a previous study analyzing LLV of 50-500 copies/mL . We also found a weak association between virologic failure and injection drug use history, HCV co-infection, and female gender. This may be explained because all three of these patient characteristics co-associate in British Columbia and have been previously linked to sub-optimal adherence to ART . However, after adjusting the model for these variables, the only variables independently associated with virologic failure were GSS<3 (p=0.007) and pVL 500-749 copies/mL vs. 50-249 (p=0.026), while female gender was no longer significant (p=0.069).
Only 1% of patients evolved resistance to protease inhibitors (specifically D30N mutations, while taking nelfinavir), despite 67% of patients receiving a PI. This result confirms that emerging PI resistance rarely occurs during LLV, probably due to their high genetic barrier to resistance . Also, our results showed that patients with higher pVL at LLV were more likely to have resistance mutations (mainly M184V/I and K103N) which is in agreement with a previous study analyzing 59 patients with LLV on first-line therapy .
A previously unaddressed concern of genotyping LLV samples has been whether the higher likelihood of amplifying only one viral input copy would still be informative. To clarify that concern, we analyzed our results by whether or not mixtures were found in the sequence chromatograms, and found no difference in the outcome between these two groups. This indicates that even clonal products, where only one viral input copy appeared to have been amplified, are informative of the future outcome of patients.
It has been recently reported that carrying a non-B HIV subtype was a predictor of very low-level viremia (<50 copies/mL) following initiation of HAART in 57 patients . However, in our study with samples between 50 and 999 copies/mL, we found no difference either in genotypic testing success or patient outcomes for patients with B or non-B HIV subtypes (data not shown). It is possible that the low viral loads (<50), low number of patients and differences in patient characteristics between the B and non-B groups in that study are impacting its results.
The strengths of our study include its large sample size (about 5000), the duration of the study (from 1996 to 2012) and the duration of clinical follow-up (up to 8 years). The analysis of only previously ART-naïve patients achieving undetectable viral load before LLV and remaining on the same regimen during the entire follow-up were also major strengths of the study. Moreover, when calculating the GSS we considered only intermediate and high-level resistance mutations, leaving aside low-level and potential low-level resistance mutations, which have a weaker influence on regimen efficacy. Furthermore, different from other groups , we considered only the resistant mutations found at sampling and did not assume that mutations detected in any previous analysis (e.g., before therapy) were still present even if not detected . All these factors provide a more consistent and cleaner picture of LLV resistance.
Nevertheless, our study has several limitations. It was an observational study, and some our results were obtained retrospectively. A randomized controlled trial would be more definitive in proving that resistance testing at low-level viraemia is a useful strategy. We did not exclude patients based on the therapies they were receiving, so two patients with suboptimal regimens (<3 full-dose drugs) were included in our study. Resistance was assessed only through sequencing of the protease and RT regions of HIV, though other regions may also play a role in drug susceptibility . We also did not address adherence in this study. Nevertheless, although adherence levels may play a hypothetical role in LLV, a previous study showed that adherence levels did not modify the associations between LLV resistance and virologic outcome . Furthermore, the low viral loads of these patients likely indicate some level of adherence.
In conclusion, we show that routine HIV genotyping of low-level viraemia samples can be performed with a reasonably high success rate and the results obtained appear predictive of future virologic outcomes.