High Rate of Missed HIV Infections in Individuals With Indeterminate or Negative HIV Western Blots Based on Current HIV Testing Algorithm in China

Transcription

1 Journal of Medical Virology 88: (2016) High Rate of Missed HIV Infections in Individuals With Indeterminate or Negative HIV Western Blots Based on Current HIV Testing Algorithm in China Man-Qing Liu, 1 Ze-Rong Zhu, 1 Wen-Hua Kong, 1 Li Tang, 1 Jin-Song Peng, 1 Xia Wang, 1 Jun Xu, 1 Robert F. Schilling, 2 Thomas Cai, 3 and Wang Zhou 1 * 1 Department of Virology, Wuhan Centers for Disease Prevention and Control, Wuhan, China 2 Department of Social Welfare, Luskin School of Public Affairs, University of California, Los Angeles California 3 AIDS Care of China, Nanning, Guangxi, China It remains unclear if China s current HIV antibody testing algorithm misses a substantial number of HIV infected individuals. Of 196 specimens with indeterminate or negative results on HIV western blot (WB) retrospectively examined by HIV-1 nucleic acid test (NAT), 67.57% (75/111) of indeterminate WB samples, and 16.47% (14/85) of negative WB samples were identified as NAT positive. HIV-1 loads in negative WB samples were significantly higher than those in indeterminate WB samples. Notably, 86.67% (13/15) of samples with negative WB and double positive immunoassay results were NAT positive. The rate of HIV-1 infections missed by China s current HIV testing algorithm is unacceptably high. Thus, China should consider using NAT or integrating fourth generation ELISA into current only antibodies-based HIV confirmation. J. Med. Virol. 88: , # 2016 Wiley Periodicals, Inc. KEY WORDS: HIV-1 nucleic acid test (NAT); enzyme-linked immunosorbent assays (ELISA); western blot (WB); China INTRODUCTION Essential to achieving 2014 UNAIDS % targets for 2020 target [Brostrom et al., 2014] are more frequent HIV testing and improved screening algorithms. China s current recommended antibody-based HIV testing algorithm includes an initial screening, repeated screening with a different kit, and confirmatory/supplemental testing (western blot [WB] [Li et al., 2010]). Fourth generation HIV enzyme-linked immunosorbent assays (ELISA) kits that can detect both HIV-1/ 2 antibody and p24 antigen are regularly used in clinical laboratories in China. However, as the established final confirmation method [Guan, 2007], HIV WB only detects HIV-1/2 antibody, potentially missing many individuals with acute HIV infection, or late-stage AIDS. Based on the sequence of the appearance of laboratory markers for HIV infection [CDC, 2014], HIV RNA can be firstly detected in plasma by HIV-1 nucleic acid testing (NAT), followed by HIV p24 antigen and HIV antibody. Recent reports [Tang et al., 2008; Patel et al., 2012; Masciotra et al., 2013; CDC, 2014; Krajden et al., 2014; Li et al., 2014] have recommended NAT in HIV screening, potentially detecting more HIV infections, particularly in individuals with acute HIV infection, and late-stage AIDS, decreasing transmission risk in high risk populations and ensuring the safety of the blood supply [Krajden et al., 2014]. Although the superior sensitivity and specificity of the HIV-1 nucleic acid assays are well established, there are fewer reports, especially in China, comparing the results of such testing relative to the fourth generation ELISA or WB. The intent of the present study was to retrospectively determine, in specimens found to be indeterminate or negative on the WB, the proportion screening positive by the HIV-1 nucleic acid assay. METHODS Setting and Data Collection From January 2012 to July 2014, plasma or serum specimens were collected from local district clinics of Grant sponsor: National Natural Science Foundation of China; Grant number: ; Grant sponsor: Foundation of Young Medical Talents in Wuhan, China; Grant sponsor: Pangaea Global AIDS Foundation; Grant number: SF Man-Qing Liu and Ze-Rong Zhu contributed equally to this work. Correspondence to: Wang Zhou, Wuhan Centers for Disease Prevention and Control, 24 N. Jianghan Road, Wuhan , China. rising-up@hotmail.com Accepted 4 February 2016 DOI /jmv Published online 24 February 2016 in Wiley Online Library (wileyonlinelibrary.com). C 2016 WILEY PERIODICALS, INC.

2 Missed HIV Infection in Clinical Specimens 1463 the Centers for Disease Prevention and Control (CDC), community hospitals, blood centers, or other health screening centers in Wuhan City, and were initially screened HIV suspected positive (Fig. 1) by third or fourth generation ELISA/chemiluminescence immunoassay (CLIA)/rapid test kits. Written informed consent and demographic information were collected when individuals first visited the test site, and the institutional review board of Wuhan Centers for Disease Prevention and Control (Wuhan CDC) approved this study. In the Wuhan CDC laboratory, specimens were further tested with two fourth generation HIV enzyme immunoassays (Diagnostic Kit for Antibody to Human Immunodeficiency Virus Type 1 and/or 2 and HIV-1 Antigen [ELISA] and Genscreen TM ULTRA HIV Ag-Ab, manufactured by BioMerieux [Shanghai, China] and Bio-Rad [Marnes-la-Coquette, France], respectively), both of which can detect HIV p24 antigen and antibodies to HIV-1 and HIV-2 in human plasma/serum. Specimens with at least one positive result were further confirmed with HIV BLOT 2.2 WB kit (MP Diagnostics, Singapore). Interpretation of a positive result was made according to the manufacturer s instructions that suggested two envelope proteins (gp160/gp41 and gp120) with one of the core proteins (p17, p24, and p55) or one of the enzyme proteins (p31, p51, and p66). Specimens, showing HIV bands (except p17) but less than minimum criteria for HIV positive WB, are defined as indeterminate WB [Linley et al., 2013; Moon et al., 2015]. CD4þ T Cells Counts and HIV-1 Loads Because of conversion to a positive result on the HIV WB at follow-up, some patients were tested for absolute CD4þ T cell counts by single-platform flow cytometry (BD FACSCounter, San Jose, CA) [Liu et al., 2013]. Some individuals had been tested on multiple occasions; their HIV results may have been either negative or indeterminate at the first or second test, and they would not have been tested for CD4þ T cell count and HIV-1 loads in these instances. In the subsequent 4 or more weeks, these individuals were re-tested for HIV antibody in our laboratory, and had converted to HIV WB positive. In accordance with China s HIV/AIDS policy, these patients were then followed up immediately, and CD4 and HIV-1 load counts were performed at this Fig. 1. Flowchart of HIV-1 nucleic acid test (NAT) in clinical specimens with indeterminate or negative HIV western blot (WB). Two specimens that were repeatedly tested with Bio-Rad fourth generation ELISA kit and Alere Determine TM HIV-1/2 (Alere Medical, Co., Ltd., Japan) rapid test kit were not included.

3 1464 Liu et al. time. Individuals who remained HIV negative or indeterminate were not followed up with tests for CD4 count and HIV-1 loads. Nucleic acid isolation, amplification, and detection of plasma HIV-1 RNA were performed with COBAS 1 TaqMan 1 HIV Test v2.0 kit on COBAS 1 AmpliPrep Instruments (Roche), and were conducted according to the manufacturer s instructions. COBAS 1 TaqMan 1 HIV Test v2.0 kit has high sensitivity and specificity and can quantitate HIV-1 RNA from 20 to 1,00,00,000 copies/ml. Statistical Analysis Statistical analyses were performed with GraphPad Prism (GraphPad Software, San Diego, CA). Where appropriate, data were expressed as means standard deviations (SD). Categorical variables were tested with Chi-Square test, and continuous variables were tested with Student s t-test. A P-value of <0.05 was considered as statistical significance. RESULTS WB Indeterminate Specimens A total of 3,798 suspected specimens (including 3,051 males [80.33%] and 747 females [19.67%], with an average age of years) were obtained between January 2012 and July A total of 894 were defined as HIV antibody negative by two fourth generation ELISA kits, 2,584, 164, and 156 were confirmed as positive WB, indeterminate WB, and negative WB, respectively (Fig. 1). Within the group of 164 indeterminate specimens, we excluded 53, including those previously tested as indeterminate on WB (n ¼ 12), and those with insufficient volume (at least 100 ml plasma for ; n ¼ 41). The remaining 111 specimens with WB indeterminate were tested with. Because of the expense of the and low rate of HIV positivity among 141 specimens with negative WB and single ELISA positive, we selected every other sample (70) for testing with (Fig. 1). Of 111 specimens defined as indeterminate WB for HIV antibody, 75 (67.57%) were positive for HIV-1 RNA. Specifically, in the 81 specimens with double ELISA reactive results, 73 (90.12%) were positive (two specimens were repeated with Bio-Rad fourth generation ELISA kit and Alere Determine TM HIV-1/ 2 [Alere Medical, Co., Ltd., Japan] rapid test kit) (Fig. 1). Males were disproportionately represented among this latter group, significantly different from the gender distribution among individuals with nonreactive NAT results (P < , Table I). Age was not significantly different across the NAT positive and NAT negative groups (Table I). All NAT positive specimens were reactive to both the BioMerieux and Bio-Rad antigen/antibody combination immunoassays (Table I). The bulk of the NAT positive specimens were collected from hospitals (48.00%) or CDCs (34.67%), whereas most NAT negative specimens were obtained from blood centers (61.11%; P < , Table I). We also analyzed the HIV antibodies in plasma/serum by WB (Table I), and found that most (73.33%) NAT positive specimens had two or more HIV bands, whereas most (77.77%) NAT negative specimens had only one HIV band (p24 or gp160). WB Negative Specimens We also performed NAT on HIV WB negative specimens. Of 85 negative WB specimens, 14 (16.47%) were reactive and 71 (83.53%) were nonreactive including 11 males and 3 females, and 43 males and 28 females, respectively. Notably, all of specimens with reactive results were received from hospitals or local CDCs (Table I). The HIV-1 loads of individuals with negative WB were significantly higher than those of individuals with indeterminate WB ( IU/ml vs IU/ml, P < ). Repeat results with two fourth generation ELISA kits showed a significant difference (P < ) between the specimens with reactive and nonreactive (Table I). Compared to NAT negative specimens, NAT positive specimens, whether indeterminate or negative HIV WB, had significantly higher signal/cutoff (S/CO) values (P < ) for both ELISA kits. No significant difference was found for the S/CO values between indeterminate WB specimens and negative WB specimens. Follow Up Individuals with initial indeterminate or negative WB test results, who then convert to positive status on the WB, are followed-up by CDC personnel. Retrospective inspection of records from the Wuhan CDC laboratory information system and the national HIV epidemiology and treatment database revealed that 56 NAT-positive patients had been followed-up for HIV-related examinations, that is, HIV WB or CD4þ T cell counts. A significant difference between indeterminate WB individuals and negative WB individuals was observed for the time interval before becoming positive HIV WB (P ¼ ), but not for CD4þ T cell counts (P ¼ ) or possible infection routes (P ¼ ). Four patients remained HIV WB indeterminate even after 6 months follow-up. Some (34.25%, 25/73) individuals with indeterminate WB were missed at follow-up, despite being reminded to re-test for HIV after 4 weeks. DISCUSSION The early stage of HIV infection is characterized by high viral loads that render the individual highly infectious, but early-stage HIV infection is not detectable by HIV antibody-based test methods. Newly infected individuals, particularly those in high-risk groups, such as men who have sex with men (MSM), will be unaware of their HIV status when they are most infectious and likely to be engaging in risk

4 Missed HIV Infection in Clinical Specimens 1465 TABLE I. Demographic Descriptors and Test Findings for Individuals with Indeterminate or Negative Results on the HIV Antibody Western Blot (WB) Indeterminate HIV antibody western blot (WB) Negative HIV antibody western blot (WB) reactive nonreactive reactive nonreactive Variables No. % No. % P value a No. % No. % P value a N Sex < Male Female Age (years) First-screen facilities < < CDCs Hospitals Blood centers Others ELISA results < < Single reactive Double reactive 75 b Western blot (WB) results < gp160, gp120, gp gp160, p24, p gp160, gp gp160, p gp120, p gp160, p p24, p gp gp gp p NAT, nucleic acid test; CDCs, District Centers for Disease Prevention and Control in Wuhan city. a By x 2 test. b Two specimens were repeated with Bio-Rad fourth generation ELISA kit and Alere Determine TM HIV-1/2 [Alere Medical, Co., Ltd., Japan] rapid test kit. behavior [Masciotra et al., 2013]. Thus, the lag between the time of infection and production of HIV antibodies measured in accepted HIV test methods increases the transmission risk among high-risk populations and from blood donors. In this study, the HIV-1 nucleic acid test was used to retrospectively identify HIV positive individuals whose WB confirmatory HIV tests had been indeterminate or negative. Our study has found that many specimens identified as indeterminate or negative by HIV WB methods were detected as positive on the NAT, especially those specimens with indeterminate WB or negative WB /ELISA double reactives indicating a high rate of missed HIV infections in these two groups. The rate of missed HIV infections was markedly higher than the report by Bi et al. [2012], which may be due to differences in the study populations patients in the present study were primarily screened as being HIV suspected positive. More importantly, 14 specimens with HIV negative WB showed higher HIV-1 loads than the indeterminate WB specimens that were earlier reported as HIV antibody negatives. Following the National Guideline for Detection of HIV/AIDS in China (2009) ( n273736/n273781/n292709/n /36174.html), these individuals were not flagged for follow-up and retesting, whereas individuals with indeterminate WB resultswereremindedtobere-testedafter4weeks. The substantial cost of precludes its use for everyone. However, the present study has shown that, among individuals with indeterminate WB tests, most of those with NAT positive tests had two or more HIV bands. They were very likely to be male, suggesting a link to a local HIV epidemic in which HIV infection is spreading rapidly among MSM [Han et al., 2011; Liu et al., 2013] (Table I). More interestingly, most NAT positive specimens were double ELISA (fourth generation) reactive and had significantly higher S/CO values than those of NAT negative specimens. Thus, health care providers or laboratory personnel in settings without access to the costly NAT test could plausibly consider S/CO values in determining HIV status. Limitations Several limitations must be acknowledged. First, the criteria for positive HIV WB were made according to the manufacturer s instructions, considered to be tighter than WHO or USA CDC criteria, possibly

5 1466 Liu et al. increasing the proportion of indeterminate HIV WB. Second, a few specimens in this study were serum, perhaps impacting the assay of ; therefore, missed HIV-1 infections in individuals with indeterminate or negative HIV WB may be higher than we reported. Finally, we cannot estimate or extrapolate to the whole study population from which the 3,798 samples were drawn. In conclusion, our data showed a high rate of positive results with the HIV-1 nucleic acid assay in specimens with indeterminate or negative WB results. Because of high HIV-1 load, especially in the HIV-infected individuals with negative WB, some of the individuals who were miss-diagnosed were likely to have spread HIV to other persons. Consistent with other studies, these findings suggest that in order to identify HIV-infected individuals as early as possible, in particular those with acute HIV infection or latestage AIDS, China should consider the HIV testing recommendations recently issued by the USA CDC [CDC, 2014], including use of the fourth generation immunoassay in the first round screening and HIV-1 NAT in supplemental testing. REFERENCES Bi X, Ning H, Wang T, Li D, Liu Y, Yang T, Yu J, Tao C Comparative performance of electrochemiluminescence immunoassay and EIA for HIV screening in a multiethnic region of China. PLoS ONE 7:e Brostrom M, Granich R, Gupta S, Badara S Reimagining HIV testing in an era of ART. AIDS Res Hum Retroviruses 30: A87 A87. Centers for Disease Control and Prevention and Association of Public Health Laboratories Laboratory testing for the diagnosis of HIV infection: Updated recommendations. Guan M Frequency, causes, and new challenges of indeterminate results in Western blot confirmatory testing for antibodies to human immunodeficiency virus. Clin Vaccine Immunol 14: Han X, Xu J, Chu Z, Dai D, Lu C, Wang X, Zhao L, Zhang C, Ji Y, Zhang H, Shang H Screening acute HIV infections among Chinese men who have sex with men from voluntary counseling & testing centers. PLoS ONE 6:e Krajden M, Cook D, Mak A, Chu K, Chahil N, Steinberg M, Rekart M, Gilbert M Pooled nucleic acid testing increases the diagnostic yield of acute HIV infections in a high-risk population compared to 3rd and 4th generation HIV enzyme immunoassays. J Clin Virol 61: Li J, Zhang H, Shen Z, Zhou Y, Fang N, Wang L, Wang B, Wang J, Tang Z Screening for acute HIV infections and estimating HIV incidence among female sex workers from low-grade venues in Guangxi, China. PLoS ONE 9:e Li Y, Zhao JK, Wang M, Han ZG, Cai WP, Zheng BJ, Xu HF Current antibody-based immunoassay algorithm failed to confirm three late-stage AIDS cases in China: Case report. Virol J 7:58. Linley L, Ethridge SF, Oraka E, Owen SM, Wesolowski LG, Wroblewski K, Landgraf KM, Parker MM, Brinson M, Branson BM Evaluation of supplemental testing with the multispot HIV-1/HIV-2 rapid test and APTIMA HIV-1 RNA qualitative assay to resolve specimens with indeterminate or negative HIV-1 western blots. J Clin Virol 58: e108 e112. Liu MQ, Tang L, Kong WH, Zhu ZR, Peng JS, Wang X, Yao ZZ, Schilling R, Zhou W CD4þ T cell count, HIV-1 viral loads and demographic variables of newly identified patients with HIV infection in Wuhan, China. J Med Virol 85: Masciotra S, Smith AJ, Youngpairoj AS, Sprinkle P, Miles I, Sionean C, Paz-Bailey G, Johnson JA, Owen SM Evaluation of the CDC proposed laboratory HIV testing algorithm among men who have sex with men (MSM) from five US metropolitan statistical areas using specimens collected in J Clin Virol 58:e8 e12. Moon HW, Huh HJ, Oh GY, Lee SG, Lee A, Yun YM, Hur M Evaluation of the Bio-Rad geenius HIV 1/2 confirmation assay as an alternative to Western Blot in the Korean population: A multi-center study. PLoS ONE 10:e Patel P, Bennett B, Sullivan T, Parker MM, Heffelfinger JD, Sullivan PS, Group CAS Rapid HIV screening: Missed opportunities for HIV diagnosis and prevention. J Clin Virol 54: Tang JW, Wong BC, Lam E, Tai V, Lee N, Cockram CS, Chan PK Failure to confirm HIV infection in two end-stage HIV/ AIDS patients using a popular commercial line immunoassay. J Med Virol 80: