Article Text

Original article
False-negative type-specific glycoprotein G antibody responses in STI clinic patients with recurrent HSV-1 or HSV-2 DNA positive genital herpes, The Netherlands
  1. Martijn S van Rooijen1,
  2. Wim Roest2,
  3. Gino Hansen1,
  4. David Kwa1,3,
  5. Henry J C de Vries1,4,5
  1. 1Department of Infectious Diseases, Public Health Service of Amsterdam, Amsterdam, The Netherlands
  2. 2Department of Dermatology, Flevoziekenhuis, Almere, The Netherlands
  3. 3Department of Medical Microbiology, Onze Lieve Vrouwe Gasthuis, Amsterdam, The Netherlands
  4. 4Department of Dermatology, Academic Medical Centre, Amsterdam, The Netherlands
  5. 5Centre for Infection and Immunity Amsterdam (CINIMA), Academic Medical Centre, Amsterdam, The Netherlands
  1. Correspondence to Martijn van Rooijen, Public Health Service of Amsterdam, Cluster of Infectious Diseases, STI clinic, P.O. Box 2200, Amsterdam 1000 CE, The Netherlands; mvrooijen{at}ggd.amsterdam.nl

Abstract

Objectives Herpes simplex virus (HSV) type-discriminating antibody tests (glycoprotein G (gG) directed) are used to identify naïve persons and differentiate acute infections from recurrences. We studied test characteristics of three commercially available antibody tests in patients with recurrent (established by viral PCR tests) herpes simplex virus type 1 (HSV-1) or herpes simplex virus type 2 (HSV-2) genital herpes episodes.

Methods Serum samples (at minimum 3 months after t=0) were examined for the presence of gG-1-specific or gG-2-specific antibodies using the HerpeSelect 1 and 2 Immunoblot IgG, the HerpeSelect 1 and 2 enzyme linked immunoassays IgG and the LIAISON HSV-1 and HSV-2 IgG indirect chemiluminescence immunoassays.

Results The immunoblot was HSV-1 positive in 70.6% (95% CI 44.0% to 89.7%), the LIAISON in 88.2% (95% CI 63.5% to 98.5%) and the ELISA in 82.4% (95% CI 56.6% to 96.2%) of the 17 patients with a recurrent HSV-1 episode. From 33 patients with a recurrent HSV-2 episode, the immunoblot was HSV-2 positive in 84.8% (95% CI 68.1% to 94.9%), the LIAISON in 69.7% (95% CI 51.3% to 84.4%) and the ELISA in 84.8% (95% CI 68.1% to 94.9%). Among 15/17 (88.2%; 95% CI 63.5% to 98.5%) patients with HSV-1 and 30/33 (90.1%; 95% CI 75.7% to 98.1%) patients with HSV-2, HSV-1 or HSV-2 antibodies, respectively, were detected in at least one of the three antibody tests.

Conclusions Commercial type-specific gG HSV-1 or HSV-2 antibody assays were false negative in 12–30% of patients with recurrent HSV-1 or HSV-2 DNA positive genital lesions. The clinical and epidemiological use of type-specific HSV serology can be hampered by false-negative results, especially if based on a single test.

  • HERPES SIMPLEX (CLINICAL)
  • GENITAL HERPES
  • SEROLOGY
  • VIROLOGY CLINICAL
  • VIROLOGY GENERAL LABORATORY

Statistics from Altmetric.com

Introduction

Genital herpes is the major cause of genito-ulcerative disease affecting a considerable number of individuals worldwide.1 Genital herpes infections can cause meningitis, life-threatening infections in newborns (neonatal herpes) and contribute to the spread of HIV.1 Genital herpes is caused by herpes simplex virus type 1 (HSV-1) and type 2 (HSV-2). Antibodies against the structural glycoprotein G (gG-1 in HSV-1 and gG-2 in HSV-2) are used for type-discriminating serology.2 At present, there are several US Food and Drug Administration (FDA)-approved commercial kits available with high sensitivity and specificity to establish type-specific immunoglobulin antibodies (IgG) to gG-1 and gG-2.2 These serological tests are indicated for testing sexually active adults or expectant mothers to diagnose past HSV-1 and/or HSV-2 infections and can be used in seroepidemiological studies as a proxy marker to identify high-risk sexual behaviour.3–8

Type-specific herpes simplex virus (HSV) antibodies can take from 2 weeks to 3 months to develop,1 but it has been shown that seroconversion for HSV-gG can take longer.9 Additionally, the loss of antibodies (seroreversion) has been described.10–12 To gain insight into the clinical reliability of type-specific HSV immunoglobulin antibody tests, we aimed to examine antibodies to gG-1 and gG-2 in sequential serum samples from patients with recurrent HSV-1 or HSV-2 DNA positive genital lesions.

Materials and methods

Study population and setting

The sexual transmittable infections (STI) outpatient clinic of the Public Health Service in Amsterdam annually performs approximately 40 000, new, free-of-charge and anonymous STI consultations.13 In patients with anogenital ulcerative disease, ulcer swabs were obtained to investigate with PCR the presence of DNA from HSV-1 and HSV-2 (targets two conserved regions; gB and IE2 genes).14 Serum samples taken at the time of each consultation were stored at –20°C for research purposes. Patient characteristics, clinical findings, diagnoses and subsequent treatment were routinely recorded in an electronic patient file. For this study, we used anonymised routinely collected data; therefore, ethical clearance was not sought.

Patient and sample selection

Between 2000 and 2012, patients with a first episode (t=0), and at least one return visit with recurrent HSV-1 or HSV-2 DNA positive genital lesions more than 3 months after the first episode, were selected. Serum samples were obtained from all recurrent patients with HSV-1, and (due to financial constraints) from the 35 most recent patients with HSV-2.

Serological testing procedure

All sequential serum samples were examined for the presence of gG-1-specific or gG-2-specific antibodies using three gG-specific commercial tests. The HerpeSelect 1 and 2 Immunoblot IgG (Focus Diagnostics, San Juan Capistrano, California, USA) was performed on de Auto-Lipa 48 from Fujirebio (Chuo-ku, Tokyo, Japan). The HerpeSelect 1 and 2 enzyme linked immunoassays IgG (ELISA, Focus Diagnostics, San Juan Capistrano, California, USA) were performed on the Dynex DSX (Dynex Laboratories, Chantilly, Virginia, USA). The LIAISON HSV-1 and HSV-2 type-specific IgG indirect chemiluminescence immunoassays (CLIA) (Diasorin, Stillwater, USA) were performed on the LIAISON analyzer. The above mentioned tests were executed according to the product inserts and were all performed blinded by one lab technician and for the interpretation of the immunoblot by an additional lab technician. In addition, all negative or equivocal sequential serum samples in at least one of the three assays were tested for HSV-1 and HSV-2 antibodies at the University of Washington by western blot (WB).2 ,15 The above tests were executed at different moments; therefore, samples were thawed and re-frozen.

Statistical analysis

Allowing 3 months of antibody maturation, the selected sequential sera were assumed to contain at least type-specific IgG antibodies to the HSV type found with PCR at t=0. If the antibody test was negative or equivocal, this result was denoted false negative. Additionally, we focused on indicators—pregnancy, anti-herpes therapy ((val)acyclovir) and HIV status—for false-negative test results. Groups were compared with the Fisher's exact test for categorical variables. Exact 95% CIs around sensitivities were calculated.

Results

Genital HSV-1 and HSV-2 diagnosis

Between January 2000 and December 2011, 19 (1.5%) patients returned with recurrent genital HSV-1 DNA positive lesions and 66 (2.9%) with recurrent genital HSV-2 DNA positive lesions. From the latter group, the 35 (53.0%) most recent patients were selected. Two patients with HSV-1 and two patients with HSV-2 were excluded because their serum samples were not available.

Patient characteristics

The majority of both patients with HSV-1 (n=10) and patients with HSV-2 (n=18) were men who have sex with men (respectively, 58.9% and 54.5%), and, respectively, 4 (23.5%) and 17 (51.5%) patients were HIV positive (table 1).

Table 1

Main characteristics of patients with recurrent HSV-1 or HSV-2 DNA positive genital lesions, STI clinic, Public Health Service of Amsterdam, The Netherlands, 2000–2011*

Serological test results

HSV-1

At the first recurrent HSV-1 episode, the immunoblot was HSV-1 positive in 12/17 (70.6%; 95% CI 44.0% to 89.7%), the LIAISON in 15/17 (88.2%; 95% CI 63.5% to 98.5%) and the ELISA in 14/17 (82.4%; 95% CI 56.6% to 96.2%) patients (table 2).

Table 2

Type-specific HSV IgG outcomes for HerpeSelect 1/2 immunoblot, HerpeSelect-1 and HerpeSelect-2 ELISA and LIAISON HSV-1 and HSV-2 CLIA in sera from patients with recurrent HSV-1 or HSV-2 DNA positive genital lesions, STI clinic, Public Health Service of Amsterdam, The Netherlands, 2000–2011*

For samples that were negative or equivocal LIAISON – from the 3 used tests the most sensitive assay – no gG-1 antibodies were detected in either the immunoblot or the ELISA. None of the three women with discrepant test results were pregnant. The proportion of patients with seronegative or equivocal results at the recurrent HSV-1 episode who had received anti-herpes therapy at t=0 (3/5, 60.0%) or had an HIV positive status (1/5, 20.0%) did not differ significantly from patients with seropositive results at the recurrent HSV-1 episode (5/12, 41.7%, p=0.62 and 3/12, 25.0%, p=1.00, respectively) (see online supplementary tables S1 and S2a). From the five patients with seronegative or equivocal results at the recurrent HSV-1 episode, three proved HSV-1 positive in the WB analysis, for one patient not enough serum was available for WB, and for one patient the WB was consistently HSV-1 negative (but HSV-2 positive). However, after repeated discrepancy testing this sample was both HSV-1 and HSV-2 positive with HSV-1 on the border of a positive reaction.

HSV-2

At the first recurrent HSV-2 episode, the immunoblot was HSV-2 positive in 28/33 (84.8%; 95% CI 68.1% to 94.9%), the LIAISON in 23/33 (69.7%; 95% CI 51.3% to 84.4%) and the ELISA in 28/33 (84.8%; 95% CI 68.1% to 94.9%) patients (table 2). Combining the results of the ELISA and the immunoblot, the sensitivity raised from 84.8% to 90.9% (95% CI 75.7% to 98.1%).

Among the serum panel with seronegative or equivocal results—all men—the proportion of patients who had received anti-herpes therapy at t=0 (3/11, 27.3%) or with an HIV positive status (5/11 45.5%) was not significantly different from the panel with concordant results (6/22, 27.3%, p=1.0 and 12/22, 54.5%, p=0.62, respectively) (see online supplementary tables S1 and S2b). From the 11 patients with seronegative or equivocal results at the first recurrent HSV-2 episode, two proved HSV-2 positive in the WB analysis, for three patients WB was performed in serum from another recurrent genital HSV-2 episode (all HSV-2 positive) and for four patients not enough serum was available. Samples from the remaining two patients with an inter-recurrent window of 4 and 8 months were consistently WB HSV-2 negative (but HSV-1 positive) at first and after discrepancy testing.

Out of four patients with negative or equivocal HSV-2 serology at the first recurrent genital HSV-2 episode, three patients had one and one patient had three additional recurrent episodes with HSV-2 DNA positive lesions (see online supplementary table S2b, including type-specific HSV antibody test results showing cases with seroconversion and seroreversion).

Discussion

In, respectively, 29.4% (5/17) and in 33.3% (11/33) of the patients with a recurrent episode of genital HSV-1 or HSV-2 DNA positive lesions, type-specific antibodies to the causative HSV-type were not detected in at least one of the three used gG directed serological tests.

A strength of this study is the unique data on recurrent genital HSV-1 episodes that occur less often than for HSV-2.1 Moreover, all recurrent HSV-1 and HSV-2 episodes, including antiviral treatment, and epidemiological data are well documented.

Unfortunately, for some patients the serum volume left was insufficient or the WB analyses were performed on sera from another episode. The patient selection in our study might have influenced the sensitivity, and—due to the relative small numbers—the CIs around the calculated sensitivities are wide. The quality of the used sera might have decreased over time. Although no association of HIV status with false-negative HSV type-specific serology has been shown, no data were available about the actual immune status of the patients.

Earlier, 89% sensitivity was shown for the HerpeSelect-1 ELISA, 96% for the HerpeSelect-2 ELISA, 99–100% for the HerpeSelect-1 immunoblot and 97–100% for the HerpeSelect-2 immunoblot.16 Comparable with our study, a lower sensitivity (82.6%) of the HerpeSelect-2 ELISA was shown.17 In a previous Swedish study, 106 WB positive patients with a recurrent genital culture-proven infection were all HerpeSelect-2 ELISA positive.18 Here, we found two WB HSV-2 positive patients with HSV-2 DNA positive recurrent genital lesions and HerpeSelect-2 ELISA negative results. In a comparison between the HSV-1 and HSV-2 LIAISON CLIA and the HerpeSelect ELISA, the overall agreement was 99.6% for HSV-1 and 100% for HSV-2.19 In our study, the LIAISON was more often positive in patients with recurrent HSV-1 (88.2%) than in patients with recurrent HSV-2 (69.7%).

In two patients with HSV-2, the WB result was HSV-1 positive only. Based on WB results from newly infected patients with HSV, it can take up to 6 months to seroconvert to HSV-gG.9 In our study, out of eight patients with an inter-recurrent HSV-2 window between 3 and 6 months, six patients were discordant in at least one of the three serological tests.

In two patients with more than one documented recurrent HSV-2 episode, gG-2 serology was positive at the first recurrent episode but reconverted to negative at subsequent recurrent episode in all three commercial tests (see online supplementary data). This phenomenon of seroreversion for gG-2 antibodies was previously described.10–12

Next to seroreversion and a seroconversion interval of up to 6 months, a mutated gG-2 gene could explain the false-negative type-specific serological results.20 ,21 Possibly, the ethnic background of our study population has influenced the type-specific serological results as previously geographic variation in the performance of the Focus HerpeSelect ELISA test has been shown.22

False-negative HSV type-specific serological test results can have far-fetching consequences for pregnant women (erroneous indication for caesarean delivery), for HSV serodiscordant sexual couples (unnecessary prophylactic measures) and for sero epidemiological studies (biased proxy for sexual risk behaviour).8 Development of a test algorithm for commercially available type-specific tests should be considered to confirm negative HSV serological results more accurately. Further research is needed to develop more sensitive type-specific antibody tests.

In conclusion, type-specific antibodies to HSV-1 or HSV-2 were frequently (29.4% and 33.3%, respectively) not detected with gG directed commercial assays in sera from patients with recurrent, viral DNA positive genital HSV-1 or HSV-2 lesions. The clinical and epidemiological use of type-specific HSV serology can be hampered by false-negative results, especially if based on one test only.

Key messages

  • Glycoprotein G directed type-specific human herpes simplex virus type 1 (HSV-1) and herpes simplex virus type 2 (HSV-2) antibody tests are used to identify naïve persons for partner and maternity counselling purposes.

  • Here, we show that three serological herpes simplex virus (HSV) tests can be false negative in 12–30% of patients with recurrent genital HSV-1 and HSV-2 lesions.

  • When used for clinical management and counselling purposes in a comparable population, the lower sensitivity of type-specific serological HSV tests should be taken into consideration.

Acknowledgments

The authors would like to thank Sylvia Bruisten for her help to retrieve the original herpes PCR results from three patients who were type-specific HSV serology positive for the opposite HSV type only.

References

View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

  • Abstract in Dutch

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Information from this paper has been orally presented at the STI & AIDS World Congress (17 July 2013, Vienna, Austria; abstract number O18.5).

  • Handling editor Jackie A Cassell

  • Contributors MSvR, WR, DK and HJCdV designed the study protocol. MSvR was responsible for the data collection at the STI clinic. DK and GH were responsible for the type-specific serology at the laboratory. MSvR performed the statistical analyses. MSvR, WR and HJCdV drafted the paper, all authors commented on draft versions and all approved the final version.

  • Funding This work was supported by a grant (no. 2387) of the Research and Development Fund of the Public Health Service of Amsterdam.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.