Drug Resistance and Its Related Factors in ART-failure Patients From 2018 to 2019 in Liaoning Province, China

Qianqian Wei Author a liations: Liaoning Provincial Center for Disease Prevention and Control, Shenyang, 110001, China. Yan Zhao Author a liations: Liaoning Provincial Center for Disease Prevention and Control, Shenyang, 110001, China. Yani Lv Author a liations: Liaoning Provincial Center for Disease Prevention and Control, Shenyang, 110001, China. Xu Kang Author a liations: Liaoning Provincial Center for Disease Prevention and Control, Shenyang, 110001, China. Shan Pan Author a liations: Liaoning Provincial Center for Disease Prevention and Control, Shenyang, 110001, China. Shujie Yao Author a liations: Liaoning Provincial Center for Disease Prevention and Control, Shenyang, 110001, China. Li Wang (  wangli363@sina.com ) Author a liations: Liaoning Provincial Center for Disease Prevention and Control, Shenyang, 110001, China. https://orcid.org/0000-0002-3600-1254


Introduction
The acquired immunode ciency syndrome (AIDS) is a chronic infectious disease caused by human immunode ciency virus (HIV) infection. Since the rst case was reported in 1981, the number of people living with HIV/AIDS has increased year by year, which has been one of the serious diseases endangering public health. The highly active antiretroviral therapy (HAART) can suppress virus replication, promote body immune reconstruction, and signi cantly reduce HIV/AIDS associated morbidity and mortality [1,2]. Given the potential bene ts of HAART, with the support of the "Four Frees and One Care" policy, China has gradually implemented free antiretroviral therapy (ART) program nationwide since 2002 [3]. This policy has made great achievement in providing the necessary care and treatment to HIV-infected people in China.
According to the o cial report, the coverage of the ART program has been increasing since the introduction of free ART in China in 2002. The number of people receiving antiviral treatment increased from 171,000 in 2012 to 610,000 million in 2017. And the coverage rate of ART was 80.4% and the treatment success rate remained above 90% in 2017 [4].
However, with the increasing coverage of antiretroviral therapy, HIV drug resistance (HIVDR) has become increasingly serious. According to the 2017 HIV drug resistance report released by the WHO, the prevalence of HIV drug resistance has increased from 11% to 29% since the global rollout of ART in 2001 [5]. In China, a prospective cohort study conducted in four urban sentinel sites nds that the proportions of drug resistance to nucleoside reverse transcriptase inhibitors (NNRTIs), non-nucleoside reverse transcriptase inhibitors (NRTIs) and protease inhibitors (PIs) were 100%, 81.8% and 0% after a year of rst-line ART [6] . The enhanced drug resistance is a great challenge to the effectiveness and sustainability of ART, which may directly lead to a decrease availability of the future drug options in resource-limited countries. And drug resistance is considered to be one of the important factors leading to the failure of ART [7,8]. Continuing using of failed treatment regimens may lead to more complex mutation patterns, the development of cross-resistance, and the increasing occurrence of transmitted HIV drug resistance (TDR). To ensure the long-term effectiveness of the ART, the WHO recommends that drug resistance should be monitored and assessed in these receiving HIV ART individuals.
Liaoning lies in the northeast of China, separated from North Korea by a river and across the sea from Japan and South Korea. As the gateway to the opening up of the Northeast region, understanding the prevalence and mutation patterns of HIV-1 drug resistance is conducive to controlling the spread of HIV drug-resistant strains. Nationwide, the prevalence of HIV/AIDS was at a low level in Liaoning Province [9,10]. A ten-year cross-sectional study in Liaoning's capital city of Shenyang showed that the average reported incidence of HIV/AIDS was 7.51 per 100,000, and the main route of infection was sexual intercourse, especially MSM [11]. Zhao Yan's study revealed that the drug resistance rate of 239 patients who failed ART in Liaoning province in 2018 was 61.9% [12].
Therefore, it is particularly necessary to grasp and understand the prevalence in ART-failure patients in Liaoning Province. This study aims to understand the prevalence and mutation patterns of HIV-1 drug resistance among HIV-infected people experiencing ART failure from 2018 to 2019 in Liaoning Province as well as to determine its related factors.

Ethical statement
Each of the participants in the study signed an informed consent prior to the enrollment. This study was approved by the Ethics Committee of Liaoning Provincial Center for Disease Control and Prevention (ethical approval number: 2020-006).

Study design and participants
The National Free Antiretroviral Treatment Handbook (Version 4) recommends that HIV-positive patients receiving ART for more than 12 months should perform a HIV drug resistance genotypic test [13]. We conducted a crosssectional study on HIV-positive individuals who experienced ART failure from April 2018 to September 2019. All the samples were collected through the routine report and then genotyped at Liaoning Provincial Center for Disease Prevention and Control. Samples were obtained from 14 cities in Liaoning Province. The local centers for disease control and prevention (CDC) were responsible for the collection of plasma specimens and transporting them to Liaoning Provincial Center for Disease Prevention and Control. And then the virus load (VL) testing and drug resistance testing were performed in the Liaoning Provincial AIDS Con rmatory Laboratory. All the testing results were uploaded to the HIV/AIDS integrated prevention and control data information management system of China. This system contained the demographic information, the self-reported transmission route, antiretroviral therapy information and HIV testing results and so on. The basic information of participants in this study was derived from this system. Samples with VL≥1000 copies/mL after 12 months of ART were de ned as ART-failure individuals. The ART-failure individuals con rmed between April 2018 and September 2019 were enrolled in the study. The inclusion criteria were: 1) HIV-infected; 2) a minimum of 18 years of age; 3) on ART more than 12 months; 4) a plasma HIV RNA≥1000 copies/ mL ;5) consent and willingness to participate in the study.

Laboratory analysis
The virus load testing and drug resistance testing were performed at the Liaoning Provincial AIDS Con rmatory Laboratory. The plasma HIV-1 virus load was tested by using the COBAS Ampliprep TaqMan 96 (Roche) detection system. And the lowest detection limit value was 20 copies / mL. Samples with VL≥1000 copies/mL were selected for HIV drug resistance testing by using an in-house method developed in core laboratories in China. The HIV-1 RNA was extracted from the plasma using the QIAamp Viral RNA Mini Kit. The 1,300 bp of pol protease region (PR) and transcriptase region (RT) gene regions were ampli ed by nested PCR. And the ampli ed regions included the full length of PR (amino acids1-99) and RT region (amino acids 1-300).
All the primers, PCR reaction system and amplifying condition was described in previous studies [14,15]. The resulting sequences were spliced together by the software (Chromaspro, Conting Express, BioEdit) and analyzed for drug resistance mutations by Stanford HIV Drug Resistance Database (http://hivdb.stanford.edu/). And the Stanford HIV Resistance Database will give the corresponding resistance levels and resistance mutation sites for various antiviral drugs.
In this study, we included mutation results conferred low-level, intermediate-level and high-level resistance [16]. All experimental protocols were carried out in accordance with the manufacturer's instructions, including any relevant details.

Statistical analysis
The database was established by excel, and SPSS 24.0 (IBM Corp., Armonk, NY, USA) software was used for statistical analysis. The demographic variables were analyzed using descriptive statistics. Logistic regression model was used to analyze the risk factors of drug resistance in ART-failure individuals. A stepwise approach was used for variable selection in the multivariate regression model. Factors with p<0.05 in the univariate models were included in the multivariate logistic regression model, and a two-sided p<0.05 was de ned as statistically signi cant.

Basic characteristics of the participants
We collected a total of 468 ART-failure individuals who met the inclusion criteria through routine reporting, and the pol gene sequence in the plasma sample of 256 participants were successfully ampli ed and analyzed. Of these, 166 were considered drug resistant and 90 were drug sensitive (Table 1). Of these,91.41% were men; the median age was 35.61 years (IQR:28.73-48.67); and more than half of the participants (55.86%) were unmarried, 28.13% were married or cohabiting, and 16.02% were divorce or separation; the most common infection routes were mainly homosexual contact (56.25%) and heterosexual contact (30.86%).
The percentage of drug resistance associated mutations in NNRTIs, NRTIs, and PIs among 256 ART-failure patients are listed as follows (

Discussion
In this study, we nd that the most common infection route was sexual transmission, accounting for 87.11%, of which MSM accounted for 56.25% of transmission. A national systematic review showed that the HIV prevalence in MSM increased from 1.77% in 2000 to 5.98% in 2010 [17]. Actually, the transmission route of HIV/AIDS in China has undergone great changes since 2007: the main transmission route has changed from the injection drug use route to the sexual transmission route, and the proportion of MSM transmission increased signi cantly [18] [12]. MSM has a higher risk of HIV infection due to their multiple sexual partners and unprotected anal intercourse behavior [19]. Furthermore, study showed that only 20% of MSM used condoms with their wives [20], which increased the risk of HIV acquisition among their female partners. Therefore, comprehensive intervention should be expanded and taken to reduce the prevalence of HIV among MSM, which would be bene cial to both MSM groups and their female groups. The most dominance subtype in our study was CRF01_AE (77.73%). CRF01_AE has been the absolutely dominant genotype in Northeast China in recent years (≥60%) [21], suggesting that the prevalence of CRF01_AE was relatively stable in this region. Moreover, the distribution of HIV-1 strain was closely related to the route of transmission. Here we found that the most common subtype among the ART failure who were infected by MSM and heterosexual was CRF01_AE. Since the rst CRF01_AE was detected among commercial sex workers of Yunnan province of China in 1994 [22], the prevalence of CRF01_AE had increased from 9.6% in 1998 to 42.5% in 2016 [23]. Study showed that there were four major clusters of HIV-1 CRF01_AE strains caused a real epidemic in China, while cluster 4 was epidemic among homosexuals in Jilin and Liaoning province of China [22]. Importantly, it was reported that the high proportion of CXCR4 usage for CRF01_AE strains may result in the loss of susceptibility to maraviroc (CCR5 antagonists) [24], which would reduce antiretroviral regimen options for ART-failure patients..
The prevalence of HIVDR among ART-failure individuals in this study was much higher than the results of previous studies among other HIV-infected populations [25,26]. It suggested that failure of antiretroviral therapy was closely related to the occurrence of HIVDR, and should be paid more attention on clinical treatment. The prevalence of resistance to NNRTIs was high, which was 62.89% for efavirenz and nevirapine and 50.00% for rilpivirine. The high prevalence of DR in NNRTIs may related to widespread use of efavirenz in the initial regimen in this study. Of concern, 50.39% participants in our study showed dual-class resistance (NNRTIs and NRTIs), which was similar to the patterns of studies reported from other studies from low and middle-income countries [27][28][29]. In addition, we noted that the prevalence of resistance to PIs was signi cantly lower than that of NNRTIs and NRTIs. Several studies have found that mutations in gag protein cleavage sites and other sites may itself contribute to PIs resistance [30][31][32], while this study we only concentrated on the pol gene, the prevalence of PIs resistance may be underestimated.
K65R and M184V were the most prevalent NRTIs mutations identi ed in this study. M184V, the most common discriminatory mutations (DAM), was selected by 3TC and FTC and could reduce susceptibility to these drugs >100-fold [33]. The mutation also caused low-level resistance to ABC and ddI [34,35], and increased susceptibility to AZT, d4T and TDF [36][37][38]. K65R was known to reduce susceptibility to TDF, ABC, d4T, ddI, of note, a retrospective study showed that patients who developed ART failure on a TDF-containing regimen were at a somewhat higher risk of developing K65R if they were infected with a subtype C virus [39]. Although the prevalence of subtype C was lower in our study, the more importance should be attached to the drug resistance associated with K65R in the high prevalence of subtype C region. The major NNRTIs mutations identi ed in our study were G190S, Y181C andK101E.G190S reduced susceptibility to NVP and EFV [40]. Y181C was an uncommon nonpolymorphic mutation selected by NVP and EFV. It caused high-level resistance to NVP and EFV.K101E usually occurred in combination with other NNRTIs-resistance mutations. All the mutations were in accordance with the two NNRTIs drugs selected in our rst-line therapy regimen. The frequencies of major or minor DRM of PIs were very low in our study.M46I were associated with decreased susceptibility to ATV, FPV, IDV, LPV and NFV. In this study, only 3.13% of the participants showed drug resistance to NFV. Fortunately, no one showed drug resistance to LPV/r, which was the core component in our second-line regimens.
Our study revealed that viral load and current ART regimen were associated with drug resistance among participants experiencing treatment failure. First, compared with the higher VL (>80,000 copies/ml), the lower VL has a higher risk of developing HIV-1 DR, which was consistent with the study in Yunnan province [41]. Some studies had proved that compared with the wild-type HIV virus, the HIV resistant strains had poor replication adaptability during a course of viral DNA synthesis [42] [43]. Therefore, some patients may have a high viral load without resistance. Second, the individuals using LPV/r+3TC+TDF in current ART regimen had a lower risk for developing drug resistance compared with those using EFV+3TC+TDF regimen. It was recommended that the medical staff should make timely judgments and change to the second-line treatment regimen to reduce the risk of DR when drug resistance occurred in patients with rst-line regimen. Although the rate of DR to PIs was relatively low in this study, the second-line regimens DR surveillance should attract more attention with the increasing use of the second-line regimen.

Limitations
We also acknowledge there are several limitations in our study. First, of the 468 ART-failure participants, only 256 were successfully sequenced, and 212 individuals didn't have sequence analysis data, meaning sample size was relatively small. The HIVDR rate in this study may not better represent the true drug resistance rate. Second, due to the lack of baseline HIVDR data, we are unable to determine whether the mutant variants pre-existed in the patients. Third, good medication adherence can reduce the occurrence of drug resistance, which is critical for people living with HIV. Regrettably, due to some objective reasons, there is no discussion on adherence in our study.

Conclusion
This study aims to illustrate the prevalence of HIV-1 DR of ART-failure individuals in Liaoning Province, China. The genotype distribution of patients experiencing treatment failure is diverse, with CRF01_AE as the predominant genotype. We observe that these patients show high resistance to NRTIs and NNRTIs, meanwhile, there are multiple DR mutation sites in NRTIs and NNRTIs. In the end, the prevalence of HIV-1 DR in the ART-failures is high. Long-term and ongoing surveillance of HIV DR is necessary, which will help us to better grasp the prevalence of resistance and guide the clinical treatment.   Figure 1 Distribution of HIV-1 genotypes among ART-failure patients in Liaoning Province.

Figure 2
Page 20/20 The proportion of drug resistance in ART-failure individuals according to drug classes.

Figure 3
Percentage of resistance to three classes of antiretroviral drugs among ART-failure patients.