Drug Resistance to Integrase Strand-Transfer Inhibitors among HIV-1-Infected Adults in Guangdong, China

Background: Integrase strand-transfer inhibitor (INSTI)-containing regimens have gradually been administered in Guangdong Province, China beginning in 2016, and INSTI-related drug resistance (DR) may occur and should be monitored among HIV-1-infected patients. Objective: To investigate the prevalence of INSTI-related resistance among HIV-1-infected individuals in Guangdong and provide evidence for the optimal administration of INSTIs. Methods: This study recruited 1208 HIV-1-infected patients (including 404 ART-naive and 804 ART-experienced patients) between June 2021 and April 2022. The entire integrase gene was amplified from blood plasma. Demographic and epidemiological information were collected. INSTI mutations and susceptibility were interpreted using the Stanford HIV Drug Resistance Database HIVdb program. Results: Of the 1208 enrolled individuals, 2.65% (32/1208) carried at least one INSTI major or accessory drug resistance mutation (DRM), with 1.49% (6/404) being from ART-naive individuals and 3.23% (26/804) from ART-experienced individuals. Among them, seven polymorphic major mutations were detected. Although no INSTI drug resistance was found among treatment-naive patients, seven ART-experienced patients (0.87%, 7/804) carried mutations conferring resistance to INSTIs. Conclusion: The overall prevalence of INSTI DRMs and DR was comparatively low among ART-naive and ART-treated populations in Guangdong; however, INSTI-related polymorphic mutations were observed. Surveillance should be reinforced before transfer to INSTI-containing regimens.


Introduction
The HIV epidemic remains one of the most serious global health threats [1,2]. The World Health Organization (WHO) has estimated that approximately 37.7 million people were living with HIV infection globally in 2020, with 1.5 million new infections with HIV and 0.68 million HIV-related deaths [3]. About 1.05 million people living with HIV and AIDS (PLWHIV), with another 351 thousand deaths, were cumulatively reported in China by the end of 2020 [4]. Heterosexual contact accounted for 74.2% newly diagnosed HIV infections, while men who have sex with men accounted for 23.3% in 2020. In Guangdong, a cumulative 78.2 thousand PLWHIV were reported by the end of October 2021, with 8.8 thousand individuals newly diagnosed in Guangdong during January to October of 2021 [5].
Combined antiretroviral therapy (cART) suppresses HIV replication, allowing immune reconstitution to occur, dramatically decreasing HIV/AIDS-related morbidity and mortality as well as the risk of further HIV transmission [6,7]. By the end of 2020, approximately 73% (27.5/37.7 million) of PLWHIV globally were receiving ART [3]. By the end of October 2021, 85.67% (67.0/78.2 thousand) of PLWHIV in Guangdong were receiving ART [5].
Overall, increased use of reverse transcriptase inhibitor-based regimens has been accompanied by the emergence of drug resistance, which leads to treatment failure and can be transmitted to those with new infections [8][9][10][11]. Indeed, a multicountry HIV drug resistance analysis in five sub-Saharan African countries showed that 53.0% and 8.8% of newly diagnosed infants exhibited resistance to one or more nonnucleoside reverse transcriptase inhibitors (NNRTIs) or nucleoside reverse transcriptase inhibitors (NRTIs), even before treatment was initiated [12].
Integrase strand-transfer inhibitors (INSTIs) are the latest class of drugs available that target the HIV integrase enzyme; they offer novel treatment options for patients with both ART-naive and acquired or transmitted drug resistance (ADR or TDR) to protease inhibitors (PIs) and NNRTIs [13,14]. The first-generation INSTIs raltegravir (RAL) and elvitegravir (EVG) were approved by the Food and Drug Administration for clinical use in 2007 and 2012, respectively; the second-generation INSTIs dolutegravir (DTG), bictegravir (BIC), and cabotegravir (CAB) were approved in 2012, 2018, and 2021, respectively [15]. In US Department of Health and Human Services Adult and Adolescent HIV treatment guidelines, one INSTI plus two NRTIs are recommended regimens for ART-naive patients [13,16]. RAL or DTG plus two NRTI regimens are recommended as first-line treatment in the 2018 Chinese Guidelines for Diagnosis and Treatment of HIV/AIDS [17], and have been widely used in Guangdong since 2016.
Although INSTIs have been proven to be an effective antiretroviral drug against HIV infection [18,19], the occurrence of drug resistance mutations may be inevitable [20]. In the present study, we conducted a province-wide survey to investigate INSTI resistance mutations and drug susceptibility among ART-naive and ART-experienced patients in Guangdong. Because drug resistance to INSTIs is not routinely included in HIV-1 genotypic testing in China, our results provide important evidence for clinicians as well as for the development of preventive HIV/AIDS control strategies.

Study Population and Data Collection
HIV-1-infected individuals were eligible for inclusion in this study if they were 16 years old or older, treatment-naive, or ART-experienced with HIV-1 viral load above 1000 copies/mL. We recruited 1208 individuals from Guangdong between June 2021 and April 2022, comprising 804 ART-experienced and 404 ART-naive individuals. Epidemiological data for the patients (including age, sex, transmission route, geographical region, year at diagnosis, and CD4+ T-cell counts) were downloaded from the National Free Antiretroviral Treatment Database for Disease Control and Prevention.

Sample Collection, Viral Load Determination, and Genotyping
Anticoagulant EDTA peripheral blood samples were collected and plasma was separated after centrifugation. Real-time molecular beacon detection was applied to detect HIV-1 viral load (Daan, China). HIV-1 genotypes were determined using the online tool COMET HIV (https://comet.lih.lu/index.php, accessed on 8 November 2022) and confirmed using a maximum likelihood (ML) phylogenetic tree based on protease (PR) and partial reverse transcriptase (RT) sequences.

RNA Extraction, Amplification, and Sanger Sequencing
Viral RNA was extracted from plasma samples using an automatic magnetic-beadbased Virus RNA Extraction Kit (Daan, China) according to the manufacturer's instructions. Amplification of the entire Integrase (IN) gene (HXB2 4230-5093, covering all 288 amino acids of integrase) was performed with an in-house RT-PCR procedure, as previously described [21]. Positive PCR products were separated by agarose gel electrophoresis and sent to Tianyi Huiyuan Genomics Company for Sanger sequencing.

Genotype Identification and Genotypic Drug Resistance Analysis
The obtained sequences were assembled and cleaned with Sequencher DNA sequence analysis software (Version 5.4.6) and then aligned using BioEdit software (Version 7.2). The Stanford HIV-1 drug resistance database (HIVdb version 9.1, https://hivdb.stanford. edu/hivdb/by-sequences/, accessed on 2 June 2022) was employed to identify INSTI mutations and sensitivity to BIC, CAB, DTG, EVG, and RAL. Sequences associated with lowlevel, intermediate, or high-level categories of resistance were defined as conferring INSTI drug resistance.

Sequence Data and Gene Evolution Analysis
All IN sequences from the 1208 HIV-1 individuals were submitted to the GenBank website (https://www.ncbi.nlm.nih.gov/genbank, accessed on 22 July 2022) under accession numbers OP032752-OP033959.
The ML phylogenetic tree was generated using PhyML version 3.0 (https://www. hiv.lanl.gov/content/sequence/PHYML/interface.html, accessed on 28 September 2017) with the GTR model [22]. Branch support was estimated using the approximate likelihood ratio test (aLRT) SH-like supports, and aLTR values higher than 0.9 were used to identify lineages. The final trees were visualized using Figtree V1.4.2.

Statistical Analysis
All statistical analyses were performed using IBM SPSS V25.0. Quantitative statistics were described using the median (IQR). The chi-square test was applied for comparisons between two groups, and the level of significance for the evaluation of two-sided p values was set at 0.05.
Complete IN sequences were obtained from all 1208 enrolled participants. There were no duplicate samples from the same subject at different time points. According to COMET HIV-1 analysis based on PR/RT sequences, CRF01_AE was the most frequently occurring genotype, with a proportion of 40.81% (493/1208), followed by CRF07_BC (28.64%, 346/1208) and CRF55_01B (28.64%, 346/1208). A total of six subtypes or circulating recombinant forms (CRFs) were confirmed (91.39%, 1104/1208) according to the ML phylogenetic tree based on 1104 PR/RT sequences ( Figure 1A), which was constructed to determine the evolutionary relationship of these sequences. CRF07_BC and CRF08_BC cannot fall into clusters; meanwhile, sequences in CRF55_01B and CRF59_01B clusters cannot match the genotypes confirmed by the ML phylogenetic tree based on 1104 PR/RT sequences, as they lack the necessary breakpoints for subtyping ( Figure 1B).

DRMs Associated with INSTIs in ART-Naive and ART-Experienced Participants
Of the 1208 individuals, 2.65% (32/1208) carried at least one INSTI major or accessory DRM. Among them, 1.49% (6/404) were obtained from ART-naive individuals, whereas 3.23% (26/804) were from ART-experienced individuals (Table 1, Figure 2   Reference sequences were downloaded from the Los Alamos HIV Sequence Dat (https://www.hiv.lanl.gov/content/index, accessed on 7 September 2021); subtype H sequ were selected as the outgroup. The phylogenetic trees were generated using PhyML versio with GTR mode. aLTR values higher than 0.9 were used to identify lineages, as indicated corresponding nodes of the tree.

DRMs Associated with INSTIs in ART-Naive and ART-Experienced Participants
Of the 1208 individuals, 2.65% (32/1208) carried at least one INSTI major or a sory DRM. Among them, 1.49% (6/404) were obtained from ART-naive individ whereas 3.23% (26/804) were from ART-experienced individuals (Table 1, Figure 2).   ; subtype H sequences were selected as the outgroup. The phylogenetic trees were generated using PhyML version 3.0 with GTR mode. aLTR values higher than 0.9 were used to identify lineages, as indicated at the corresponding nodes of the tree.

DR Associated with INSTIs in ART-Naive and ART-Experienced Participants
According to the HIVdb program, 0.58% (7/1208) of the enrolled individuals carried INSTI-related DRMs associated with low-level, intermediate, or high-level resistance ( Figure 4). The characteristics of the patients with INSTI-related mutations and the corresponding drug resistance levels are shown in Table 2 E138K and Q148R (both 0.17%, 2/1208) were the most frequent mutations among INSTI major DRMs, and E157Q (1.41%, 17/1208) was the most frequent mutation among INSTI accessory DRMs. Moreover, INSTI major and accessory DRMs both varied among ART-naive and ART-experienced individuals (Figure 3).

DR Associated with INSTIs in ART-Naive and ART-Experienced Participants
According to the HIVdb program, 0.58% (7/1208) of the enrolled individuals carried INSTI-related DRMs associated with low-level, intermediate, or high-level resistance ( Figure 4). The characteristics of the patients with INSTI-related mutations and the corresponding drug resistance levels are shown in Table 2. Four patients carrying IN-STI-related major mutations all showed at least low-level resistance to INSTIs. Among the 28 patients carrying INSTI-related accessory mutations, we found only three cases carrying the H51Y or G163R mutation, which cause low-level resistance to INSTIs.

DR Associated with INSTIs in ART-Naive and ART-Experienced Participants
According to the HIVdb program, 0.58% (7/1208) of the enrolled individuals carried INSTI-related DRMs associated with low-level, intermediate, or high-level resistance ( Figure 4). The characteristics of the patients with INSTI-related mutations and the corresponding drug resistance levels are shown in Table 2. Four patients carrying IN-STI-related major mutations all showed at least low-level resistance to INSTIs. Among the 28 patients carrying INSTI-related accessory mutations, we found only three cases carrying the H51Y or G163R mutation, which cause low-level resistance to INSTIs.   Conversely, no drug resistance to INSTIs was observed among treatment-naive patients. The prevalence of drug resistance to INSTIs among ART-experienced patients was 0.87% (7/804) ( Figure 2). As shown in Figure 4, the percentages of resistance levels for the five INSTI drugs were all lower than 1.00%. Among the ART-experienced patients, the percentages of resistance were 0.37% (3/804), 0.50% (4/804), 0.37% (3/804), 0.50% (4/804), and 0.50% (4/804) for BIC, CAB, DTG, EVG, and RAL, respectively, and a greater proportion of high-level resistance to EVG was found (0.37%, 3/804) (Figure 4).

Discussion
INSTIs, a novel class of anti-HIV agents, show high activity in inhibiting HIV-1 replication and play a critical role in therapy for infection with this virus [18]. However, as shown in previous research, first-generation INSTIs (RAL and EVG) have lower genetic barriers and possible cross-resistance. While the second-generation INSTIs (DTG, CAB and BIC) have a higher resistance barrier [23], the surveillance of resistance to INSTIs is necessary with more frequent use of the single-tablet regimens (STRs) containing INSTIs.
HIV-1 has a high degree of natural variability due its lack of gene expression proofreading and high frequency of genetic recombination [24]. We observed six cases (1.49%) of the accessory INSTI resistance mutation E157Q among ART-naive individuals (Figure 3), which is a polymorphic mutation that appears to have little effect on INSTI susceptibility [25] and is common in ART-naive individuals with a very low proportion [21,26,27]. The absence of INSTI-associated drug resistance among ART-naive individuals suggests low circulation of INSTI-resistant variants prior to treatment in Guangdong, which is in accordance with reports both from other provinces and worldwide [21,24,[28][29][30]. However, when compared with the prevalence of INSTI-related DRMs among ART-naive HIV-1-infected patients in 2018 (1.45%, 12/827) [21], no statistically significant difference in the prevalence of INSTIrelated DRMs among ART-naive HIV-1 infections was observed in this study (χ 2 = 0.002, p = 1.00 > 0.05). With the increasing number of patients using INSTI-containing regimens in Guangdong (patients who used INSTI-related regimens accounted for approximately 12.8% of all patients under ART in the Pearl River Delta region of Guangdong in 2021, unpublished data), surveillance should be reinforced with respect to the further choice of or transfer to INSTI-containing regimens.
For the ART-experienced individuals, 28.13% (9/32) of the INSTI resistance mutations detected are major resistance mutations. The most frequent major INSTI resistance mutations were E138K and Q148R (both 0.25%, 2/804), followed by T66I, E92Q, G118R, G140A, and S147G (all 0.12%, 1/804). All these major INSTI resistance mutations were detected in four patients (three of them had ever used regimens containing integrase), though the frequency of these major INSTI resistance mutations was very low (from 0.12% to 0.25%). E138K is a non-polymorphic mutation occurring in patients receiving RAL, EVG, and DTG [34]. In this study, two patients carrying the E138K mutation had used RAL and DTG. Q148R has been reported in patients with virological failure during DTG [35], which can reduce RAL and EVG susceptibility 30-100-fold [36]. Most of the major INSTI DRMs detected in this study may have been caused by the use of INSTIs, though the low rates of other INSTI DRMs (mainly accessory mutations) may have been due to natural polymorphisms in the IN region and the long-term use of 2NRTI + NNRTI regimens. Surveillance is necessary for the further use of INSTIs, especially in ART-experienced individuals.
According to previous studies [37,38], drug resistance to NNRTIs and/or NRTIs prior to treatment increases the risk of resistance to INSTIs. In this study, seven patients showed drug resistance to INSTIs, and six individuals were coupled with other class DRMs (Table 2). Whether other classes of DRM represent a risk factor for resistance to INSTIs remains to be investigated in the future.
According to previous studies [39][40][41][42][43], mutations outside of integrase gene, such as those in and near the 3 polypurine tract (3 PPT) and in envelope glycoproteins, are able to confer resistance to INSTI. The continuous monitoring of mutations outside the IN gene warrants special importance in surveillance of development of drug resistance. As INSTIs are now widely used even in first-line therapy, we intend to focus on the continuous monitoring of the mutations outside the IN gene in subsequent research.

Conclusions
The overall prevalence of INSTI DR in Guangdong remains low (0.58%, 7/1208), which suggests that INSTIs currently have good applicability and that the use of INSTIs results in ideal viral suppression. Nevertheless, the proportion of DR in ART-experienced individuals was higher than that in ART-naive individuals (χ 2 = 3.188, p = 0.088), and seven INSTIrelated polymorphic major mutations were detected among HIV-1 patients in Guangdong, emphasizing the importance of monitoring drug resistance prior to administration of INSTI-containing regimens.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board of Guangzhou Eighth People's Hospital (protocol code 202053186 and date of approval 14 February 2020).

Informed Consent Statement:
All participants in this study signed informed consent.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest:
The authors declare no conflict of interest.