- Email: [email protected]
Mumps vaccine failure investigation in Novosibirsk, Russia, 2002–2004
ORIGINAL ARTICLE
10.1111/j.1469-0691.2007.01727.x
Mumps vaccine failure investigation in Novosibirsk, Russia, 2002–2004 A. V. Atrasheuskaya1, M. V. Kulak1, S. Rubin2 and G. M. Ignatyev1 State Research Center of Virology and Biotechnology Vector, Koltsovo, Russia and 2Center for Biologics Evaluation and Research, FDA, Bethesda, MD, USA 1
ABSTRACT The aims of this study were to estimate the importance of vaccine failure (VF) in cases of mumps during 2002–2004 in the city of Novosibirsk, Western Siberia, Russia, and to genotype the responsible virus strain. Mumps virus-specific RT-PCR testing of saliva was performed for 18 cases of mumps. Sera were tested for IgM and IgG, IgG avidity, and the ability to neutralise a panel of mumps viruses, including the Leningrad-3 mumps vaccine virus. Of the 12 patients for whom vaccination status was positively determined, 11 showed serological evidence of primary VF. Sequence analysis of virus RNA amplified from saliva revealed a genotype C2 virus in 2002, a genotype H2 virus in 2003, and both genotypes in 2004. Although several vaccinated patients were positive for mumps virus IgG at the time of first sampling, only nominal levels of neutralising antibody were detected, and these were effective in neutralising the vaccine strain, but not genotype C and H mumps virus strains. These results suggest that the majority of cases of mumps in vaccinees are caused by primary VF, defined as either a lack of seroconversion or a lack of IgG maturity, as based on avidity testing. The results also support the hypothesis that sera of low neutralising antibody titre have a limited ability to neutralise heterologous mumps virus strains, suggesting that antigenic differences between circulating and mumps vaccine virus strains may play a role in cases of breakthrough infection. Consistent with previous reports, mumps virus genotypes C and H continue to circulate in Novosibirsk. Keywords
Genotyping, mumps virus, Russia, serology, vaccine failure
Original Submission: 25 July 2006;
Revised Submission: 20 January 2007; Accepted: 1 February 2007
Clin Microbiol Infect 2007; 13: 670–676
INTRODUCTION Despite evidence of effective, long-lasting immunity following natural infection or vaccination [1,2], strains of wild-type mumps virus (MuV) continue to circulate worldwide. Recently, two unusually large mumps epidemics have been reported, one in the UK in 2005, involving over 70 000 cases [3], and one in the USA [4], which began in early 2006 and involved 5783 cases (the background number of mumps cases in the USA has been c. 250 annually for the past decade). The mumps epidemic in the UK has been attributed mostly to a large cohort of unvaccinated
Corresponding author and reprint requests: A. V. Atrasheuskaya, Laboratory of Immunology Safety, State Research Center of Virology and Biotechnology Vector, Koltsovo, Novosibirsk Region, Russia 630559 E-mail: [email protected]
individuals, mostly of college age, who were not eligible for vaccination during childhood. Similarly, the 2006 US epidemic involved mostly college-age students, and has been linked to an insufficient proportion of the population receiving the recommended two-dose schedule of mumps-containing vaccine. In such situations, in which herd-immunity may be lost, outbreaks or epidemics are easily started. The occurrence of sporadic mumps outbreaks in populations with high vaccine coverage is also a well-known phenomenon. These outbreaks are usually attributed to pockets of unvaccinated individuals and ⁄ or vaccine failure (VF), divided into primary (lack of seroconversion) or secondary (waning immunity) failure [5–8]. It has also been suggested that antigenic differences between MuV strains may allow certain strains to escape neutralisation in vaccinees [7,9,10]. While this phenomenon has been demonstrated in the
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases
Atrasheuskaya et al. Mumps vaccine failure 671
laboratory (S. Rubin, personal communication), conclusive evidence of a causal relationship between virus strain-specific antigenic differences and outbreaks or epidemics has not yet been demonstrated. However, for surveillance of vaccine protection against mumps, it is important to follow the distribution of different genotypes of MuV and to measure the genotype-specific immunity in the population. A prospective study to monitor mumps cases was therefore initiated in Novosibirsk, Western Siberia, Russia, in 2002– 2004. Novosibirsk is a large city with a population of c. 1.4 million inhabitants. The Leningrad-3 (L-3) mumps vaccine has been used since 1984 in Novosibirsk as part of the national immunisation programme, and the mumps vaccine coverage rate in Novosibirsk has been calculated at 95%. The most widely circulating MuV strains in Novosibirsk between 1994 and 2003 belonged to genotypes C and H [11–14]. According to official data, 142 cases of mumps were reported in Novosibirsk during 2001, 189 cases in 2002, 24 cases in 2003, and 27 cases in 2004. The large decrease in the number of cases in 2003 and 2004 is probably a reflection of the then-instituted requirement for laboratory confirmation of cases, suggesting that MuV may not actually have caused many of the cases reported before 2003.
Vaccination status was determined for all clinical cases of mumps included in this study, and saliva and acute and convalescent sera were obtained. RT-PCR testing of saliva using primers specific for the MuV SH gene was performed to confirm the presence of the virus and to identify the virus genotype [15]. Sera were tested by ELISA for IgM and IgG antibody titres, and for IgG antibody avidity. In addition, sera were tested for their ability to neutralise the vaccine strain (L-3) and two wild-type viruses isolated previously in Novosibirsk.
PATIENTS AND METHODS Subjects The SRC VB Vector Ethical Committee approved the study (IRB00001360). Informed consent was obtained from the parents of all children and from the adult patients. The patients’ ages and their vaccination status, taken from the official medical records, are presented in Table 1. In total, 18 patients (aged 3–56 years; six females, 12 males) were enrolled in the study. Of the 18 patients, 12 (67%) had been immunised with the L-3 mumps strain vaccine. All 18 patients were diagnosed clinically with mumps according to the WHO case definition (http://www.who.int/vaccines/globalsummary/ timeseries/tsincidenceMUM.htm). Clinical signs of meningitis (severe headache, vomiting, nuchal rigidity) were observed in patients P1 ⁄ 2002, P3 ⁄ 2003, P8 ⁄ 2004, P9 ⁄ 2004, P16 ⁄ 2004 and P18 ⁄ 2004, but spinal taps were not performed; thus, the diagnosis could not be confirmed. Pancreatitis was diagnosed
Table 1. Serum levels of mumps virus (MuV) IgM, IgG, IgG avidity and neutralising antibody in mumps patients
Patient
Age (years)
Statusa (years)
Days sera were takenb
Serum IgM
Serum IgG (titre)
IgG avidity (%)
PRN L-3 (titre)
PRN H (titre)
PRN C (titre)
RT-PCRc
MuV genotype
P1 ⁄ 2002 P2 ⁄ 2002 P3 ⁄ 2003 P4 ⁄ 2003 P5 ⁄ 2003 P6 ⁄ 2003 P7 ⁄ 2004 P8 ⁄ 2004 P9 ⁄ 2004 P10 ⁄ 2004 P11 ⁄ 2004 P12 ⁄ 2004 P13 ⁄ 2004 P14 ⁄ 2004 P15 ⁄ 2004 P16 ⁄ 2004 P17 ⁄ 2004 P18 ⁄ 2004
8 4 18 16 15 25 17 49 20 17 35 9 28 18 22 30 3 56
R (1.5) V (2.5) V (4) R (4) R (4) V (4) Unknown NV V (5) Unknown NV R (2.5) V (5) V (5) V (5) NV V (2) NV
5 ⁄ 26 5 ⁄ 26 5 ⁄ 11 ⁄ 27 4 ⁄ 26 5 ⁄ 26 5 ⁄ 26 5 ⁄ 26 5 ⁄ 26 4 ⁄ 26 5 ⁄ 26 2 ⁄ 25 3 ⁄ 25 5 ⁄ 26 5 ⁄ 26 5 ⁄ 26 3 ⁄ 24 3 ⁄ 25 3 ⁄ 24
+⁄– +⁄– –⁄+⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄–
1:900 ⁄ 1:2000 1:2000 ⁄ 1:8200 neg ⁄ neg ⁄ 1:500 1:1500 ⁄ 1:9000 1:8500 ⁄ 1:17 300 1:5700 ⁄ 1:11 800 1:600 ⁄ 1:1300 neg ⁄ 1:800 neg ⁄ 1:300 1:500 ⁄ 1:900 neg ⁄ 1:500 1:1400 ⁄ 1:3000 1:1400 ⁄ 1:3000 neg ⁄ 1:300 1:300 ⁄ 1:800 neg ⁄ 1:1000 1:1500 ⁄ 1:3000 neg ⁄ 1:500
27 31 ND ⁄ 8 29 26 30 32 ND ⁄ 10 ND ⁄ 6 32 ND ⁄ 11 27 28 ND ⁄ 4 29 ND ⁄ 14 34 ND ⁄ 13
1:4 ⁄ 1:8 1:4 ⁄ 1:4 0 ⁄ 1:4 1:4 ⁄ 1:4 1:4 ⁄ 1:4 1:4 ⁄ 1:8 1:4 ⁄ 1:8 0⁄0 0 ⁄ 1:4 1:4 ⁄ 1:4 0⁄0 1:4 ⁄ 1:8 1:4 ⁄ 1:8 0⁄0 1:4 ⁄ 1:4 0 ⁄ 1:4 1:8 ⁄ 1:8 0 ⁄ 1:4
0 ⁄ 1:4 0⁄0 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:4 0 ⁄ 1:8 0 ⁄ 1:4 0 ⁄ 1:4 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:4 0 ⁄ 1:4 0 ⁄ 1:4 0 ⁄ 1:4 0⁄0
0 ⁄ 1:8 0 ⁄ 1:8 0⁄0 0⁄0 0⁄0 0 ⁄ 1:4 0 ⁄ 1:4 0⁄0 0⁄0 0⁄0 0⁄0 0 ⁄ 1:4 0 ⁄ 1:4 0⁄0 0⁄0 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:8
+⁄– +⁄– +⁄+⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄–
C C H H H H H H H H H H H H H C C C
a Vaccination status: V, vaccinated with one dose; R, re-vaccinated (two doses); NV, not vaccinated. Figures in parentheses indicate the time (years) since the most recent vaccination. b Relative to day of fever onset. The first sample was taken upon admission to the hospital and the second was taken following discharge (except P3 ⁄ 2003). c RT-PCR results are shown for saliva samples. PRN, plaque reduction neutralisation assay; ND, not detectable; neg, negative.
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
672 Clinical Microbiology and Infection, Volume 13 Number 7, July 2007 in patient P12 ⁄ 2004. The duration of the clinical course of mumps infection was 7–14 days. All patients recovered completely. Serological methods Acute and convalescent paired serum and saliva samples were obtained from all patients. Serological detection of antiMuV IgM and IgG was done by ELISA (Enzygnost; Dade Behring, Marburg, Germany). Assessment of IgG avidity was used to differentiate between primary vaccine failure (PVF) and secondary vaccine failure (SVF). IgG avidity was measured using the IgG ELISA and 6 M urea as described previously [16]. According to the manufacturer’s instruction, OD values of 0.1–2.5 allow calculation of IgG titres. Thus, avidity was also determined within this OD range. Samples with OD values <0.1 were regarded as not detectable, and samples with OD values >2.5 were diluted appropriately and retested. The avidity index (AI) was calculated as the ratio of the absorbance with and without urea treatment. Using a system established previously, an AI £31 was considered to represent low avidity, while values ‡32 were considered to represent high avidity [16]. Other aetiologies were excluded by testing serum for antibodies to parainfluenza virus types 1–3 and Epstein–Barr virus (EBV) [17]. IgG to the parainfluenza viruses was measured with a commercial enzyme immunoassay kit able to detect all three types of parainfluenza virus (Parainfluenza 1 ⁄ 2 ⁄ 3 IgG-ELISA; Immuno Biological Laboratories, Hamburg, Germany). IgM antibodies to EBV were measured with a commercial enzyme immunoassay kit (Enzygnost Anti-EBV ⁄ IgM; Dade Behring). Any patient who was IgM-positive for EBV was also tested for IgG to EBV. Neutralising anti-MuV antibody titres were determined using a plaque reduction neutralisation (PRN) assay, with the L-3 MuV vaccine strain (kindly provided by the Tarasevich State Institute of Standardisation and Control of Immunobiological Preparations, Moscow, Russia) and wild-type MuVs of genotypes H (strain PetroNov, accession no. AY681495) and C (strain Dragoon, accession no. AY669145) as the targets. The wild-type MuVs of genotypes C and H were isolated in Novosibirsk, Russia in 1994 [13] and 2003 [14], respectively. In brief, sera were heat-inactivated at 56C for 45 min. Two-fold serial dilutions of heat-inactivated serum (or medium alone as a negative control) were mixed with equal volumes containing 25–35 PFU of the target virus to give a final sera dilution range of 1:4–1:128. Serum– virus mixtures were incubated at 37C in CO2 5% v ⁄ v for 1 h and then placed on Vero cell monolayers in 24-well plates and incubated for 1 h at 37C in CO2 5% v ⁄ v. The virus–serum mixture was removed by aspiration and the cell monolayers were rinsed with MEM (Quality Biologicals, Gaithersburg, MD, USA) immediately before covering with Nobel agar (BD Difco, Franklin Lakes, NJ, USA) 0.75% w ⁄ v in 2 · MEM (Quality Biologicals) supplemented with fetal bovine serum 10% v ⁄ v. Plates were then incubated at 37C in CO2 5% v ⁄ v for 5 days. A second layer of agar containing neutral red (Quality Biologicals) 0.01% w ⁄ v was added and incubated overnight for visualisation of the plaques. For each serum, the neutralising antibody titre was defined as the highest dilution capable of reducing the number of virus plaques by ‡50% compared to control values. The cut-off for seropositivity was a neutralising antibody titre of ‡1:4.
Mumps PRN titres were normalised by logarithmic transformation to calculate mean titres (MT). Quantitative results were expressed as MT ± SD (standard deviation). The statistical significance between MT was tested using Student’s t-test or chi-square tests, with p £0.05 considered to be significant. RT-PCR and sequencing RNA was extracted and purified from throat swabs and sera using an RNeasy Mini Kit (Qiagen, Hilden, Germany), followed by reverse transcription using a Titan One Tube RT-PCR System (Roche, Mannheim, Germany). The primers for RT-PCR and sequencing were specific for the SH gene, as described previously [13,18,19]. Primer 1 (sense), 5¢-TCAACACAATATCAAGTA (positions 2964–2981), and primer 2 (antisense), 5¢-TTCTGTGTTGTATTGTGA (positions 3401–3418), were used for RT-PCR. The PCR products were excised from the gel and purified using a QIAquick Gel Extraction Kit (Qiagen). The purified products were directly sequenced using primer 3 (sense), 5¢-ATGATCTCATCAGGTAC (positions 2988–3004), and primer (antisense), 5¢-TCCTAAGTTTGTTCTGG (positions 3384–3400), for sequencing. Nucleotide sequences were determined using a CEQ 2000 Dye Terminator Cycle Sequencing Kit and a CEQ 2000 XL DNA Analysis System (Beckman Coulter, Fullerton, CA, USA) according to the manufacturer’s instructions. Nucleotide alignments were performed using the Vector NTI v.8.0 (InforMax, Bethesda, MD, USA) software package. Definitions PVF was defined as a previously vaccinated subject with a clinical diagnosis of mumps and evidence of a primary immune response, i.e., IgM-positive ⁄ IgG-negative, or IgGpositive with low avidity [20]. SVF was defined as a previously vaccinated subject with a clinical diagnosis of mumps and evidence of a secondary immune response, i.e., high-avidity IgG antibody in acute-phase sera [16].
RESULTS Serum levels of IgM and IgG to MuV, IgG avidity, neutralising antibody titres and RT-PCR results are summarised in Table 1. All patients were negative for EBV, with the exception of P18 ⁄ 2004, who suffered from an acute severe course of the mixed infection. All patients were negative for parainfluenza viruses 1, 2 and 3, based on enzyme immunoassay tests as described above. Serological results: estimation of immune response MuV-specific admission for tion between between IgG
IgM antibodies were detected at all patients. There was no correlaIgM and vaccination status, or levels in acute-phase sera and
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
Atrasheuskaya et al. Mumps vaccine failure 673
vaccination status. Seven patients were IgG-negative at the acute phase of the disease, although three of these patients had been vaccinated previously. Of the six patients with symptoms of meningitis (P1 ⁄ 2002, P3 ⁄ 2003, P8 ⁄ 2004, P9 ⁄ 2004, P16 ⁄ 2004 and P18 ⁄ 2004), five were initially seronegative. Unexpectedly high initial IgG levels were found in P5 ⁄ 2003 and P6 ⁄ 2003. Both of the patients of unknown vaccination status (P7 ⁄ 2004, P10 ⁄ 2004) appeared to have developed a secondary immune response, based on the results of IgG avidity testing. Four patients, P8 ⁄ 2004, P11 ⁄ 2004, P16 ⁄ 2004 and P18 ⁄ 2004, being unvaccinated, were seronegative at admission, and thereafter were considered to have developed a primary immune response. Taking into account a low AI or IgG negativity, the vaccinated patients P1 ⁄ 2002–P6 ⁄ 2003, P9 ⁄ 2004 and P12 ⁄ 2004–P15 ⁄ 2004 were considered to be cases of PVF. Thus, only one of the 12 vaccinated patients, P17 ⁄ 2004, showed evidence of SVF, characterised by a high AI, and also by the highest PRN titre against the L-3 vaccine virus. The PRN assay results were very informative in characterising VFs. Acute-phase sera from all IgGpositive vaccinated patients with PVFs showed equally low titres of neutralising antibodies (1:4) in the PRN assay with the L-3 vaccine virus as the virus target. The acute-phase sera from all patients in this study lacked neutralising activity against wild-type MuV of both genotypes C and H. However, all 13 patients infected with a genotype H MuV strain had PRN activity against the genotype H virus (MT ± SD = 2.61 ± 0.47 –log2 dilution) at discharge, and only four (31%) of these 13 patients had PRN activity against the genotype C virus (MT ± SD = 0.62 ± 0.89 –log2 dilution) (p <0.005). Similarly, of the five patients infected by the genotype C virus, all five had PRN activity against the genotype C virus (MT ± SD = 3.0 ± 0.0 –log2 dilution), but only three (60%) had activity against the genotype H virus (MT ± SD = 1.2 ± 0.89 –log2 dilution; p <0.002) at discharge. Nine (50%) of the subjects showed increased PRN titres against the L-3 vaccine MuV at discharge (1.28 ± 1.02 –log2 dilution vs. 2.0 ± 0.97 –log2 dilution; p <0.05). It should be noted that the measured PRN titres in the convalescent sera may be artificially low, since these sera were drawn 3–4 weeks following the onset of symptoms (Table 1). Longer post-exposure periods (2–3 months) are probably needed to reach peak titres.
The only difference observed when comparing the immune responses of patients vaccinated with one or two doses of mumps vaccine was that IgG was not detected in acute-phase sera from three of the eight patients who received one dose of vaccine, whereas IgG was detected in acute-phase sera from all four patients who received two doses of vaccine. All other parameters assessed were similar. All patients, regardless of the number of vaccine doses received, were IgMpositive at the time of hospital admission. With the exception of the IgG-negative patients, the mean IgG titre in acute-phase sera of one-dose vaccinees (1:2180) did not differ statistically from that of two-dose vaccinees (1:3075), and both of these means increased, by 3-fold and 2.6-fold, respectively, by the time of the second blood sample approximately 21 days later. The mean IgG AI measured in the one-dose vs. two-dose vaccinees was also similar (30.4 vs. 27.3), as were the PRN titres. All patients in both groups were considered to be cases of PVF, with the exception of P17 (a one-dose vaccinee assessed as a case of SVF). However, these findings were based on only a small number of subjects per study group (eight one-dose and four two-dose vaccinees), and the results may not be representative. RT-PCR results, nucleotide sequence analysis and assignment to genotype The clinical diagnosis of all 18 cases of mumps was confirmed by RT-PCR (Table 1). Five (P1 ⁄ 2002, P2 ⁄ 2002, P16 ⁄ 2004, P17 ⁄ 2004, P18 ⁄ 2004) of the 18 patients had been infected with a genotype C MuV (based on the SH gene sequence). The nucleotide sequences obtained from these patients were each identical to the Dragoon strain used in the PRN assays, which was isolated in Siberia several years previously (accession no. AY669145) [13]. Based on phylogenetic analysis, the genotype C MuV isolated in Novosibirsk in 2002 and in 2004, as well as the Dragoon strain, belonged to subgenotype C2 [15,21]. The other 13 SH gene sequences (P3 ⁄ 2003– P15 ⁄ 2004) belonged to genotype H. The sequences obtained from these patients were identical to each other and to the sequence of PetroNov, sequenced previously in Siberia [14] (accession no. AY681495). Based on phylogenetic analysis, the genotype H MuV isolated in Novosibirsk during 2003–2004 belonged to subgenotype H2 [21].
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
674 Clinical Microbiology and Infection, Volume 13 Number 7, July 2007
DISCUSSION Co-circulation of genotypes H2 and C2 was identified during the study period. This is the first report of the H2 genotype in Novosibirsk. The genotype C MuV has been associated previously with a small (64 cases) outbreak in Novosibirsk during 1994 [13]. The SH gene sequence of the genotype C2 MuV strain reported here is identical to that of the 1994 outbreak. Genotypes H2 and C2 both caused only sporadic cases of mumps infection during 2002–2004. Notably, 44% of all registered mumps cases in Novosibirsk during 2004 were included in this investigation. RT-PCR analysis is indispensable for confirmation of the diagnosis of mumps in highly vaccinated populations [22]. This is highlighted by the serological data in the present study, which showed that, in many cases, mumps patients have substantial levels of virus-specific IgG at the time of first sampling, and do not show the customary four-fold increase in titre in convalescent sera. Despite high ELISA-based MuV-specific IgG titres in acute-phase sera in some patients, IgG avidity was typically low. This absence of an association between IgG titres and IgG avidity has been reported previously for a number of infectious agents, including MuV [22–26], and emphasises the importance and relevance of avidity testing in investigating vaccine failures. It is important to emphasise that serum antibody is but one facet of the immune response. Thus, testing of IgG should be interpreted with caution. The measurement of IgG avidity (functional affinity) is critical in differentiating PVF from SVF, and provides important information for improving the mumps prevention strategy [27], e.g., a second dose of mumps-containing vaccine [28]. Vaccine failures are far less common in recipients of a two-dose regimen [29,30]. In eight of the 12 cases of VF identified in this study, only one dose of vaccine had been administered, and PVF was identified in all eight of these cases. A mandatory second vaccination was not introduced in Novosibirsk until 1999. It is also important to note that the incubation period for mumps is ‡3 weeks, and the possibility that the observed ELISA-based levels of IgG in acute-phase sera are, in part, a response to exposure to the wild-type virus (as opposed to being solely a vaccine response) can therefore not be ignored. Also, the question of whether the
presence of high (borderline) avidity IgG in acutephase sera in vaccinees is conclusive evidence of SVF will require further investigation. However, this does not appear to be a significant issue in this study, since IgG avidity-based SVF was assigned to only one of the subjects investigated. Also critical is the examination of neutralising antibody activity. An important finding in the present study was that all patients initially lacked neutralising antibody activity against the C2 and H2 viruses, but that the majority of these patients were able to neutralise the vaccine strain. Such apparent strain-specific immunity has been reported previously, especially in cases with low-titre sera, and it has been suggested that this phenomenon is responsible for cases of breakthrough infection in vaccinees [31]. In general, the majority of the patients in the present study failed to induce markers suggestive of a mature functional immune response; e.g., they had low IgG avidity and low sera-neutralising activity, and thus, not surprisingly, mounted primary immune responses despite a history of vaccination. These data are consistent with findings in a previous study of horizontal transmission of the L-3 vaccine virus [22]. The present data demonstrate a MuV strain-specific neutralising antibody response, in that all convalescent sera could neutralise an MuV genotype similar to the one detected by RT-PCR in patient saliva, but were often unable to neutralise a different contemporary wild MuV genotype. The goal of vaccination is the generation of a memory response that can rapidly clear infection, thus providing protection to the host. A significant effort has been made towards identifying the attributes of the immune response that are associated with optimal protection. As humoral and cellular immune responses to vaccination play important roles in protection, both must be considered when examining a possible surrogate marker or indicator of protection, as has been described previously for measles virus [32–35]. Earlier findings [22], as well as those presented here, support the hypothesis that a combination of high virus-specific IgG avidity (‡32%) and at least a nominal level of neutralising activity (‡1:8) decreases the risk of mumps infection, even among close contacts. Taken together, the serological and virological findings described here highlight the importance of the specific methods employed in case investigation and in assessing
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
Atrasheuskaya et al. Mumps vaccine failure 675
VF. A VF case investigation protocol should also consider the influence of the specific virus strains used in the serological IgG avidity and PRN assays [36]. The latter is very important in understanding the correlation between the maturation of mumps-specific IgG after infection or immunisation and the neutralising capacity spectrum of immune sera. Notably, antibody avidity testing appears to be a particularly useful assay for examining potential cases of VF. In conclusion, only one dose of vaccine had been administered to the majority of cases of VF identified in this study, and primary VF was implicated in most of these cases, based on a failure to induce markers suggestive of a mature and functional immune response. Whether the absence of sufficient titres of neutralising antibody in response to vaccination, or the absence of neutralising antibody specific for the endemic MuV strain, was responsible for susceptibility of vaccinees to infection cannot be determined conclusively, but both are indicated. What is clear is that causes of VF are complex and multifactorial in nature. A proper assessment of VFs should include IgG avidity testing, as well as a comprehensive assessment of the functional antibody response, i.e., the differential ability of serum antibody to neutralise both the vaccine virus and the endemic challenge virus strains. Other protective responses not captured in this analysis of VFs include cellular and mucosal responses. Such responses deserve more attention and may further clarify the occurrence of VFs. ACKNOWLEDGEMENTS The work was performed in the framework of grant BTEP ID # 16, 2002. No official support or endorsement of this article by the Food and Drug Administration is intended or should be inferred. We thank A. Neverov, T. Malisheva and L. D’achkova for excellent technical assistance, and especially thank A. E. Nesterov for providing data concerning some patients.
REFERENCES 1. Gut J-P, Lablache C, Behr S, Kirn A. Symptomatic mumps reinfections. J Med Virol 1995; 45: 17–23. 2. Crowley B, Afzal MA. Mumps virus reinfection—clinical findings and serological vagaries. Comm Dis Public Health 2002; 5: 311–313. 3. Anonymous. Mumps epidemic—United Kingdom, 2004– 2005. CDC. MMWR 2006; 55: 173–175.
4. Anonymous. USA—mumps activity update; 1 January to 7 October 2006. CDC. MMWR 2006; 55: 1152–1153. 5. Hersh BS, Fine PEM, Kent W et al. Mumps outbreak in a highly vaccinated population. J Pediatr 1991; 119: 187–193. 6. Briss PA, Fehrs LJ, Parker RA et al. Sustained transmission of mumps in a highly vaccinated population: assessment of primary vaccine failure and waning vaccine-induced immunity. J Infect Dis 1994; 169: 77–82. 7. Strohle A, Eggenberger K, Steiner CA, Matter L, Germann D. Mumps epidemic in vaccinated children in West Switzerland. Schweiz Med Wochenschr 1997; 127: 1124–1133. 8. Vandermeulen C, Roelants M, Vermoere M, Roseeuw K, Goubau P, Hoppenbrouwers K. Outbreak of mumps in a vaccinated child population: a question of vaccine failure. Vaccine 2004; 22: 2713–2716. ¨ rvell C, Alsheikhly A-R, Kalantari M, Johansson B. 9. O Characterization of genotype-specific epitopes of the HN protein of mumps virus. J Gen Virol 1997; 78: 3187–3193. ¨ rvell C. Mumps virus 10. No¨jd J, Tecle T, Samuelsson A, O neutralizing antibodies do not protect against reinfection with a heterologous mumps virus genotype. Vaccine 2001; 19: 1727–1731. 11. Agafonov AP, Nicheuchina SN, Patrusheva I et al. Characteristics of parotitis virus, isolated in Siberia. Vopr Virusol 1997; 42: 222–226. 12. Agafonov AP, Streltsova MA, Suslaparov IM, Ignatyev GM. Immune parameters in mumps patients. Vopr Virusol 2001; 1: 30–33. 13. Heider A, Ignatyev G, Gerike E, Schreier E. Genotypic characterization of mumps virus isolated in Russia (Siberia). Res Virol 1997; 148: 433–435. 14. Nesterov AE, Kameneva CN, Maksimov NL, Ignatyev GM. Lymphadenopathy in a previously vaccinated patient with mumps. Infect Dis 2004; 4: 92–94. 15. Utz S, Richard JL, Capaul S, Matter HC, Hrisoho MG, Muhlemann K. Phylogenetic analysis of clinical mumps virus isolates from vaccinated and non-vaccinated patients with mumps during an outbreak, Switzerland 1998–2000. J Med Virol 2004; 73: 91–96. 16. Narita M, Matsuzono Y, Takekoshi Y et al. Analysis of mumps vaccine failure by means of avidity testing for mumps virus-specific immunoglobulin G. Clin Diagn Lab Immunol 1998; 5: 799–803. 17. Davidkin I, Jokinen S, Paananen A, Leinikki P, Peltola H. Etiology of mumps-like illness in children and adolescents vaccinated for measles, mumps, and rubella. J Infect Dis 2005; 191: 719–723. 18. Kunkel U, Driesel G, Henning U, Gerike E, Willers H, Schreier E. Differentiation of vaccine and wild mumps viruses by polymerase chain reaction and nucleotide sequencing of the SH gene: brief report. J Med Virol 1995; 45: 121–126. 19. Elliott GD, Afzal MA, Martin SJ, Rima BK. Nucleotide sequence of the matrix, fusion and putative SH protein genes of MuV and their deduced amino acid sequences. Virus Res 1989; 12: 61–75. 20. Paunio M, Hedman K, Davidkin I, Peltola H. IgG avidity to distinguish secondary from primary measles vaccination failures: prospects for a more effective global measles elimination strategy. Expert Opin Pharmacother 2003; 4: 1215–1225. 21. Muhlemann K. The molecular epidemiology of mumps virus. Infect Genet Evol 2004; 4: 215–219.
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
676 Clinical Microbiology and Infection, Volume 13 Number 7, July 2007
22. Atrasheuskaya AV, Neverov AA, Rubin S, Ignatyev GM. Horizontal transmission of the Leningrad-3 live attenuated mumps vaccine virus. Vaccine 2006; 24: 1530–1536. 23. Rizzo LV, DeKruyff RH, Umetsu DT. Generation of B cell memory and affinity maturation. Induction with Th1 and Th2 T cell clones. J Immunol 1992; 148: 3733–3739. 24. Lazzarotto T, Varani S, Spezzacatena P et al. Delayed acquisition of high-avidity anti-cytomegalovirus antibody is correlated with prolonged antigenemia in solid organ transplant recipients. J Infect Dis 1998; 178: 1145–1149. 25. Weissbrich B. The use of semi-automated EBV IgG avidity determination for the diagnosis of infectious mononucleosis. J Med Virol 1998; 54: 145–153. 26. Atrasheuskaya AV, Blatun EM, Neverov AA, Karpov IA, Ignatyev GM. Measles in Minsk, Belarus, 2001–2003: clinical, virological and serological parameters. J Clin Virol 2005; 34: 179–185. 27. Paunio M, Hedman K, Davidkin I, Peltola H. IgG avidity to distinguish secondary from primary measles vaccination failures: prospects for a more effective global measles elimination strategy. Exp Opinion 2003; 4: 1215–1225. 28. WHO. Global status of mumps immunization and surveillance. Wkly Epidemiol Record 2005; 48: 417–424. 29. Sanz-Moreno JC, Limia-Sanchez A, Garcia-Comas L et al. Detection of secondary mumps vaccine failure by means of avidity testing for specific immunoglobulin G. Vaccine 2005; 23: 4921–4925.
30. Harling R, White JM, Ramsay ME, Macsween KF, van den Bosch C. The effectiveness of the mumps component of the MMR vaccine: a case control study. Vaccine 2005; 23: 4070–4074. 31. Rubin S, Mauldin J, Chumakov K, Vanderzanden J, Iskow R, Carbone K. Serological and phylogenetic evidence of monotypic immune responses to different mumps virus strains. Vaccine 2006; 24: 2662–2668. 32. Muller CP, Huiss S, Schneider F. Secondary immune responses in parents of children with recent measles. Lancet 1996; 348: 1379–1380. 33. Huiss S, Damien B, Schneider F, Muller CP. Characteristics of asymptomatic secondary immune response to measles virus in late convalescent donors. Clin Exp Immunol 1997; 109: 416–420. 34. Bautista-Lopez N, Ward BJ, Mills E, McCormick D, Martel N, Ratnam S. Development and durability of measles antigen-specific lymphoproliferative response after MMR vaccination. Vaccine 2000; 18: 1393–1401. 35. Ovsyannikova IG, Dhiman N, Jacobson RM, Vierkant RA, Poland GA. Frequency of measles virus-specific CD4+ and CD8+ T cells in subjects seronegative or highly seropositive for measles vaccine. Clin Diagn Lab Immunol 2003; 10: 411–416. 36. Pipkin PA, Afzal MA, Heath AB, Minor PD. Assay of humoral immunity to mumps virus. J Virol Methods 1999; 79: 219–215.
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
10.1111/j.1469-0691.2007.01727.x
Mumps vaccine failure investigation in Novosibirsk, Russia, 2002–2004 A. V. Atrasheuskaya1, M. V. Kulak1, S. Rubin2 and G. M. Ignatyev1 State Research Center of Virology and Biotechnology Vector, Koltsovo, Russia and 2Center for Biologics Evaluation and Research, FDA, Bethesda, MD, USA 1
ABSTRACT The aims of this study were to estimate the importance of vaccine failure (VF) in cases of mumps during 2002–2004 in the city of Novosibirsk, Western Siberia, Russia, and to genotype the responsible virus strain. Mumps virus-specific RT-PCR testing of saliva was performed for 18 cases of mumps. Sera were tested for IgM and IgG, IgG avidity, and the ability to neutralise a panel of mumps viruses, including the Leningrad-3 mumps vaccine virus. Of the 12 patients for whom vaccination status was positively determined, 11 showed serological evidence of primary VF. Sequence analysis of virus RNA amplified from saliva revealed a genotype C2 virus in 2002, a genotype H2 virus in 2003, and both genotypes in 2004. Although several vaccinated patients were positive for mumps virus IgG at the time of first sampling, only nominal levels of neutralising antibody were detected, and these were effective in neutralising the vaccine strain, but not genotype C and H mumps virus strains. These results suggest that the majority of cases of mumps in vaccinees are caused by primary VF, defined as either a lack of seroconversion or a lack of IgG maturity, as based on avidity testing. The results also support the hypothesis that sera of low neutralising antibody titre have a limited ability to neutralise heterologous mumps virus strains, suggesting that antigenic differences between circulating and mumps vaccine virus strains may play a role in cases of breakthrough infection. Consistent with previous reports, mumps virus genotypes C and H continue to circulate in Novosibirsk. Keywords
Genotyping, mumps virus, Russia, serology, vaccine failure
Original Submission: 25 July 2006;
Revised Submission: 20 January 2007; Accepted: 1 February 2007
Clin Microbiol Infect 2007; 13: 670–676
INTRODUCTION Despite evidence of effective, long-lasting immunity following natural infection or vaccination [1,2], strains of wild-type mumps virus (MuV) continue to circulate worldwide. Recently, two unusually large mumps epidemics have been reported, one in the UK in 2005, involving over 70 000 cases [3], and one in the USA [4], which began in early 2006 and involved 5783 cases (the background number of mumps cases in the USA has been c. 250 annually for the past decade). The mumps epidemic in the UK has been attributed mostly to a large cohort of unvaccinated
Corresponding author and reprint requests: A. V. Atrasheuskaya, Laboratory of Immunology Safety, State Research Center of Virology and Biotechnology Vector, Koltsovo, Novosibirsk Region, Russia 630559 E-mail: [email protected]
individuals, mostly of college age, who were not eligible for vaccination during childhood. Similarly, the 2006 US epidemic involved mostly college-age students, and has been linked to an insufficient proportion of the population receiving the recommended two-dose schedule of mumps-containing vaccine. In such situations, in which herd-immunity may be lost, outbreaks or epidemics are easily started. The occurrence of sporadic mumps outbreaks in populations with high vaccine coverage is also a well-known phenomenon. These outbreaks are usually attributed to pockets of unvaccinated individuals and ⁄ or vaccine failure (VF), divided into primary (lack of seroconversion) or secondary (waning immunity) failure [5–8]. It has also been suggested that antigenic differences between MuV strains may allow certain strains to escape neutralisation in vaccinees [7,9,10]. While this phenomenon has been demonstrated in the
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases
Atrasheuskaya et al. Mumps vaccine failure 671
laboratory (S. Rubin, personal communication), conclusive evidence of a causal relationship between virus strain-specific antigenic differences and outbreaks or epidemics has not yet been demonstrated. However, for surveillance of vaccine protection against mumps, it is important to follow the distribution of different genotypes of MuV and to measure the genotype-specific immunity in the population. A prospective study to monitor mumps cases was therefore initiated in Novosibirsk, Western Siberia, Russia, in 2002– 2004. Novosibirsk is a large city with a population of c. 1.4 million inhabitants. The Leningrad-3 (L-3) mumps vaccine has been used since 1984 in Novosibirsk as part of the national immunisation programme, and the mumps vaccine coverage rate in Novosibirsk has been calculated at 95%. The most widely circulating MuV strains in Novosibirsk between 1994 and 2003 belonged to genotypes C and H [11–14]. According to official data, 142 cases of mumps were reported in Novosibirsk during 2001, 189 cases in 2002, 24 cases in 2003, and 27 cases in 2004. The large decrease in the number of cases in 2003 and 2004 is probably a reflection of the then-instituted requirement for laboratory confirmation of cases, suggesting that MuV may not actually have caused many of the cases reported before 2003.
Vaccination status was determined for all clinical cases of mumps included in this study, and saliva and acute and convalescent sera were obtained. RT-PCR testing of saliva using primers specific for the MuV SH gene was performed to confirm the presence of the virus and to identify the virus genotype [15]. Sera were tested by ELISA for IgM and IgG antibody titres, and for IgG antibody avidity. In addition, sera were tested for their ability to neutralise the vaccine strain (L-3) and two wild-type viruses isolated previously in Novosibirsk.
PATIENTS AND METHODS Subjects The SRC VB Vector Ethical Committee approved the study (IRB00001360). Informed consent was obtained from the parents of all children and from the adult patients. The patients’ ages and their vaccination status, taken from the official medical records, are presented in Table 1. In total, 18 patients (aged 3–56 years; six females, 12 males) were enrolled in the study. Of the 18 patients, 12 (67%) had been immunised with the L-3 mumps strain vaccine. All 18 patients were diagnosed clinically with mumps according to the WHO case definition (http://www.who.int/vaccines/globalsummary/ timeseries/tsincidenceMUM.htm). Clinical signs of meningitis (severe headache, vomiting, nuchal rigidity) were observed in patients P1 ⁄ 2002, P3 ⁄ 2003, P8 ⁄ 2004, P9 ⁄ 2004, P16 ⁄ 2004 and P18 ⁄ 2004, but spinal taps were not performed; thus, the diagnosis could not be confirmed. Pancreatitis was diagnosed
Table 1. Serum levels of mumps virus (MuV) IgM, IgG, IgG avidity and neutralising antibody in mumps patients
Patient
Age (years)
Statusa (years)
Days sera were takenb
Serum IgM
Serum IgG (titre)
IgG avidity (%)
PRN L-3 (titre)
PRN H (titre)
PRN C (titre)
RT-PCRc
MuV genotype
P1 ⁄ 2002 P2 ⁄ 2002 P3 ⁄ 2003 P4 ⁄ 2003 P5 ⁄ 2003 P6 ⁄ 2003 P7 ⁄ 2004 P8 ⁄ 2004 P9 ⁄ 2004 P10 ⁄ 2004 P11 ⁄ 2004 P12 ⁄ 2004 P13 ⁄ 2004 P14 ⁄ 2004 P15 ⁄ 2004 P16 ⁄ 2004 P17 ⁄ 2004 P18 ⁄ 2004
8 4 18 16 15 25 17 49 20 17 35 9 28 18 22 30 3 56
R (1.5) V (2.5) V (4) R (4) R (4) V (4) Unknown NV V (5) Unknown NV R (2.5) V (5) V (5) V (5) NV V (2) NV
5 ⁄ 26 5 ⁄ 26 5 ⁄ 11 ⁄ 27 4 ⁄ 26 5 ⁄ 26 5 ⁄ 26 5 ⁄ 26 5 ⁄ 26 4 ⁄ 26 5 ⁄ 26 2 ⁄ 25 3 ⁄ 25 5 ⁄ 26 5 ⁄ 26 5 ⁄ 26 3 ⁄ 24 3 ⁄ 25 3 ⁄ 24
+⁄– +⁄– –⁄+⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄–
1:900 ⁄ 1:2000 1:2000 ⁄ 1:8200 neg ⁄ neg ⁄ 1:500 1:1500 ⁄ 1:9000 1:8500 ⁄ 1:17 300 1:5700 ⁄ 1:11 800 1:600 ⁄ 1:1300 neg ⁄ 1:800 neg ⁄ 1:300 1:500 ⁄ 1:900 neg ⁄ 1:500 1:1400 ⁄ 1:3000 1:1400 ⁄ 1:3000 neg ⁄ 1:300 1:300 ⁄ 1:800 neg ⁄ 1:1000 1:1500 ⁄ 1:3000 neg ⁄ 1:500
27 31 ND ⁄ 8 29 26 30 32 ND ⁄ 10 ND ⁄ 6 32 ND ⁄ 11 27 28 ND ⁄ 4 29 ND ⁄ 14 34 ND ⁄ 13
1:4 ⁄ 1:8 1:4 ⁄ 1:4 0 ⁄ 1:4 1:4 ⁄ 1:4 1:4 ⁄ 1:4 1:4 ⁄ 1:8 1:4 ⁄ 1:8 0⁄0 0 ⁄ 1:4 1:4 ⁄ 1:4 0⁄0 1:4 ⁄ 1:8 1:4 ⁄ 1:8 0⁄0 1:4 ⁄ 1:4 0 ⁄ 1:4 1:8 ⁄ 1:8 0 ⁄ 1:4
0 ⁄ 1:4 0⁄0 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:4 0 ⁄ 1:8 0 ⁄ 1:4 0 ⁄ 1:4 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:4 0 ⁄ 1:4 0 ⁄ 1:4 0 ⁄ 1:4 0⁄0
0 ⁄ 1:8 0 ⁄ 1:8 0⁄0 0⁄0 0⁄0 0 ⁄ 1:4 0 ⁄ 1:4 0⁄0 0⁄0 0⁄0 0⁄0 0 ⁄ 1:4 0 ⁄ 1:4 0⁄0 0⁄0 0 ⁄ 1:8 0 ⁄ 1:8 0 ⁄ 1:8
+⁄– +⁄– +⁄+⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄– +⁄–
C C H H H H H H H H H H H H H C C C
a Vaccination status: V, vaccinated with one dose; R, re-vaccinated (two doses); NV, not vaccinated. Figures in parentheses indicate the time (years) since the most recent vaccination. b Relative to day of fever onset. The first sample was taken upon admission to the hospital and the second was taken following discharge (except P3 ⁄ 2003). c RT-PCR results are shown for saliva samples. PRN, plaque reduction neutralisation assay; ND, not detectable; neg, negative.
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
672 Clinical Microbiology and Infection, Volume 13 Number 7, July 2007 in patient P12 ⁄ 2004. The duration of the clinical course of mumps infection was 7–14 days. All patients recovered completely. Serological methods Acute and convalescent paired serum and saliva samples were obtained from all patients. Serological detection of antiMuV IgM and IgG was done by ELISA (Enzygnost; Dade Behring, Marburg, Germany). Assessment of IgG avidity was used to differentiate between primary vaccine failure (PVF) and secondary vaccine failure (SVF). IgG avidity was measured using the IgG ELISA and 6 M urea as described previously [16]. According to the manufacturer’s instruction, OD values of 0.1–2.5 allow calculation of IgG titres. Thus, avidity was also determined within this OD range. Samples with OD values <0.1 were regarded as not detectable, and samples with OD values >2.5 were diluted appropriately and retested. The avidity index (AI) was calculated as the ratio of the absorbance with and without urea treatment. Using a system established previously, an AI £31 was considered to represent low avidity, while values ‡32 were considered to represent high avidity [16]. Other aetiologies were excluded by testing serum for antibodies to parainfluenza virus types 1–3 and Epstein–Barr virus (EBV) [17]. IgG to the parainfluenza viruses was measured with a commercial enzyme immunoassay kit able to detect all three types of parainfluenza virus (Parainfluenza 1 ⁄ 2 ⁄ 3 IgG-ELISA; Immuno Biological Laboratories, Hamburg, Germany). IgM antibodies to EBV were measured with a commercial enzyme immunoassay kit (Enzygnost Anti-EBV ⁄ IgM; Dade Behring). Any patient who was IgM-positive for EBV was also tested for IgG to EBV. Neutralising anti-MuV antibody titres were determined using a plaque reduction neutralisation (PRN) assay, with the L-3 MuV vaccine strain (kindly provided by the Tarasevich State Institute of Standardisation and Control of Immunobiological Preparations, Moscow, Russia) and wild-type MuVs of genotypes H (strain PetroNov, accession no. AY681495) and C (strain Dragoon, accession no. AY669145) as the targets. The wild-type MuVs of genotypes C and H were isolated in Novosibirsk, Russia in 1994 [13] and 2003 [14], respectively. In brief, sera were heat-inactivated at 56C for 45 min. Two-fold serial dilutions of heat-inactivated serum (or medium alone as a negative control) were mixed with equal volumes containing 25–35 PFU of the target virus to give a final sera dilution range of 1:4–1:128. Serum– virus mixtures were incubated at 37C in CO2 5% v ⁄ v for 1 h and then placed on Vero cell monolayers in 24-well plates and incubated for 1 h at 37C in CO2 5% v ⁄ v. The virus–serum mixture was removed by aspiration and the cell monolayers were rinsed with MEM (Quality Biologicals, Gaithersburg, MD, USA) immediately before covering with Nobel agar (BD Difco, Franklin Lakes, NJ, USA) 0.75% w ⁄ v in 2 · MEM (Quality Biologicals) supplemented with fetal bovine serum 10% v ⁄ v. Plates were then incubated at 37C in CO2 5% v ⁄ v for 5 days. A second layer of agar containing neutral red (Quality Biologicals) 0.01% w ⁄ v was added and incubated overnight for visualisation of the plaques. For each serum, the neutralising antibody titre was defined as the highest dilution capable of reducing the number of virus plaques by ‡50% compared to control values. The cut-off for seropositivity was a neutralising antibody titre of ‡1:4.
Mumps PRN titres were normalised by logarithmic transformation to calculate mean titres (MT). Quantitative results were expressed as MT ± SD (standard deviation). The statistical significance between MT was tested using Student’s t-test or chi-square tests, with p £0.05 considered to be significant. RT-PCR and sequencing RNA was extracted and purified from throat swabs and sera using an RNeasy Mini Kit (Qiagen, Hilden, Germany), followed by reverse transcription using a Titan One Tube RT-PCR System (Roche, Mannheim, Germany). The primers for RT-PCR and sequencing were specific for the SH gene, as described previously [13,18,19]. Primer 1 (sense), 5¢-TCAACACAATATCAAGTA (positions 2964–2981), and primer 2 (antisense), 5¢-TTCTGTGTTGTATTGTGA (positions 3401–3418), were used for RT-PCR. The PCR products were excised from the gel and purified using a QIAquick Gel Extraction Kit (Qiagen). The purified products were directly sequenced using primer 3 (sense), 5¢-ATGATCTCATCAGGTAC (positions 2988–3004), and primer (antisense), 5¢-TCCTAAGTTTGTTCTGG (positions 3384–3400), for sequencing. Nucleotide sequences were determined using a CEQ 2000 Dye Terminator Cycle Sequencing Kit and a CEQ 2000 XL DNA Analysis System (Beckman Coulter, Fullerton, CA, USA) according to the manufacturer’s instructions. Nucleotide alignments were performed using the Vector NTI v.8.0 (InforMax, Bethesda, MD, USA) software package. Definitions PVF was defined as a previously vaccinated subject with a clinical diagnosis of mumps and evidence of a primary immune response, i.e., IgM-positive ⁄ IgG-negative, or IgGpositive with low avidity [20]. SVF was defined as a previously vaccinated subject with a clinical diagnosis of mumps and evidence of a secondary immune response, i.e., high-avidity IgG antibody in acute-phase sera [16].
RESULTS Serum levels of IgM and IgG to MuV, IgG avidity, neutralising antibody titres and RT-PCR results are summarised in Table 1. All patients were negative for EBV, with the exception of P18 ⁄ 2004, who suffered from an acute severe course of the mixed infection. All patients were negative for parainfluenza viruses 1, 2 and 3, based on enzyme immunoassay tests as described above. Serological results: estimation of immune response MuV-specific admission for tion between between IgG
IgM antibodies were detected at all patients. There was no correlaIgM and vaccination status, or levels in acute-phase sera and
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
Atrasheuskaya et al. Mumps vaccine failure 673
vaccination status. Seven patients were IgG-negative at the acute phase of the disease, although three of these patients had been vaccinated previously. Of the six patients with symptoms of meningitis (P1 ⁄ 2002, P3 ⁄ 2003, P8 ⁄ 2004, P9 ⁄ 2004, P16 ⁄ 2004 and P18 ⁄ 2004), five were initially seronegative. Unexpectedly high initial IgG levels were found in P5 ⁄ 2003 and P6 ⁄ 2003. Both of the patients of unknown vaccination status (P7 ⁄ 2004, P10 ⁄ 2004) appeared to have developed a secondary immune response, based on the results of IgG avidity testing. Four patients, P8 ⁄ 2004, P11 ⁄ 2004, P16 ⁄ 2004 and P18 ⁄ 2004, being unvaccinated, were seronegative at admission, and thereafter were considered to have developed a primary immune response. Taking into account a low AI or IgG negativity, the vaccinated patients P1 ⁄ 2002–P6 ⁄ 2003, P9 ⁄ 2004 and P12 ⁄ 2004–P15 ⁄ 2004 were considered to be cases of PVF. Thus, only one of the 12 vaccinated patients, P17 ⁄ 2004, showed evidence of SVF, characterised by a high AI, and also by the highest PRN titre against the L-3 vaccine virus. The PRN assay results were very informative in characterising VFs. Acute-phase sera from all IgGpositive vaccinated patients with PVFs showed equally low titres of neutralising antibodies (1:4) in the PRN assay with the L-3 vaccine virus as the virus target. The acute-phase sera from all patients in this study lacked neutralising activity against wild-type MuV of both genotypes C and H. However, all 13 patients infected with a genotype H MuV strain had PRN activity against the genotype H virus (MT ± SD = 2.61 ± 0.47 –log2 dilution) at discharge, and only four (31%) of these 13 patients had PRN activity against the genotype C virus (MT ± SD = 0.62 ± 0.89 –log2 dilution) (p <0.005). Similarly, of the five patients infected by the genotype C virus, all five had PRN activity against the genotype C virus (MT ± SD = 3.0 ± 0.0 –log2 dilution), but only three (60%) had activity against the genotype H virus (MT ± SD = 1.2 ± 0.89 –log2 dilution; p <0.002) at discharge. Nine (50%) of the subjects showed increased PRN titres against the L-3 vaccine MuV at discharge (1.28 ± 1.02 –log2 dilution vs. 2.0 ± 0.97 –log2 dilution; p <0.05). It should be noted that the measured PRN titres in the convalescent sera may be artificially low, since these sera were drawn 3–4 weeks following the onset of symptoms (Table 1). Longer post-exposure periods (2–3 months) are probably needed to reach peak titres.
The only difference observed when comparing the immune responses of patients vaccinated with one or two doses of mumps vaccine was that IgG was not detected in acute-phase sera from three of the eight patients who received one dose of vaccine, whereas IgG was detected in acute-phase sera from all four patients who received two doses of vaccine. All other parameters assessed were similar. All patients, regardless of the number of vaccine doses received, were IgMpositive at the time of hospital admission. With the exception of the IgG-negative patients, the mean IgG titre in acute-phase sera of one-dose vaccinees (1:2180) did not differ statistically from that of two-dose vaccinees (1:3075), and both of these means increased, by 3-fold and 2.6-fold, respectively, by the time of the second blood sample approximately 21 days later. The mean IgG AI measured in the one-dose vs. two-dose vaccinees was also similar (30.4 vs. 27.3), as were the PRN titres. All patients in both groups were considered to be cases of PVF, with the exception of P17 (a one-dose vaccinee assessed as a case of SVF). However, these findings were based on only a small number of subjects per study group (eight one-dose and four two-dose vaccinees), and the results may not be representative. RT-PCR results, nucleotide sequence analysis and assignment to genotype The clinical diagnosis of all 18 cases of mumps was confirmed by RT-PCR (Table 1). Five (P1 ⁄ 2002, P2 ⁄ 2002, P16 ⁄ 2004, P17 ⁄ 2004, P18 ⁄ 2004) of the 18 patients had been infected with a genotype C MuV (based on the SH gene sequence). The nucleotide sequences obtained from these patients were each identical to the Dragoon strain used in the PRN assays, which was isolated in Siberia several years previously (accession no. AY669145) [13]. Based on phylogenetic analysis, the genotype C MuV isolated in Novosibirsk in 2002 and in 2004, as well as the Dragoon strain, belonged to subgenotype C2 [15,21]. The other 13 SH gene sequences (P3 ⁄ 2003– P15 ⁄ 2004) belonged to genotype H. The sequences obtained from these patients were identical to each other and to the sequence of PetroNov, sequenced previously in Siberia [14] (accession no. AY681495). Based on phylogenetic analysis, the genotype H MuV isolated in Novosibirsk during 2003–2004 belonged to subgenotype H2 [21].
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
674 Clinical Microbiology and Infection, Volume 13 Number 7, July 2007
DISCUSSION Co-circulation of genotypes H2 and C2 was identified during the study period. This is the first report of the H2 genotype in Novosibirsk. The genotype C MuV has been associated previously with a small (64 cases) outbreak in Novosibirsk during 1994 [13]. The SH gene sequence of the genotype C2 MuV strain reported here is identical to that of the 1994 outbreak. Genotypes H2 and C2 both caused only sporadic cases of mumps infection during 2002–2004. Notably, 44% of all registered mumps cases in Novosibirsk during 2004 were included in this investigation. RT-PCR analysis is indispensable for confirmation of the diagnosis of mumps in highly vaccinated populations [22]. This is highlighted by the serological data in the present study, which showed that, in many cases, mumps patients have substantial levels of virus-specific IgG at the time of first sampling, and do not show the customary four-fold increase in titre in convalescent sera. Despite high ELISA-based MuV-specific IgG titres in acute-phase sera in some patients, IgG avidity was typically low. This absence of an association between IgG titres and IgG avidity has been reported previously for a number of infectious agents, including MuV [22–26], and emphasises the importance and relevance of avidity testing in investigating vaccine failures. It is important to emphasise that serum antibody is but one facet of the immune response. Thus, testing of IgG should be interpreted with caution. The measurement of IgG avidity (functional affinity) is critical in differentiating PVF from SVF, and provides important information for improving the mumps prevention strategy [27], e.g., a second dose of mumps-containing vaccine [28]. Vaccine failures are far less common in recipients of a two-dose regimen [29,30]. In eight of the 12 cases of VF identified in this study, only one dose of vaccine had been administered, and PVF was identified in all eight of these cases. A mandatory second vaccination was not introduced in Novosibirsk until 1999. It is also important to note that the incubation period for mumps is ‡3 weeks, and the possibility that the observed ELISA-based levels of IgG in acute-phase sera are, in part, a response to exposure to the wild-type virus (as opposed to being solely a vaccine response) can therefore not be ignored. Also, the question of whether the
presence of high (borderline) avidity IgG in acutephase sera in vaccinees is conclusive evidence of SVF will require further investigation. However, this does not appear to be a significant issue in this study, since IgG avidity-based SVF was assigned to only one of the subjects investigated. Also critical is the examination of neutralising antibody activity. An important finding in the present study was that all patients initially lacked neutralising antibody activity against the C2 and H2 viruses, but that the majority of these patients were able to neutralise the vaccine strain. Such apparent strain-specific immunity has been reported previously, especially in cases with low-titre sera, and it has been suggested that this phenomenon is responsible for cases of breakthrough infection in vaccinees [31]. In general, the majority of the patients in the present study failed to induce markers suggestive of a mature functional immune response; e.g., they had low IgG avidity and low sera-neutralising activity, and thus, not surprisingly, mounted primary immune responses despite a history of vaccination. These data are consistent with findings in a previous study of horizontal transmission of the L-3 vaccine virus [22]. The present data demonstrate a MuV strain-specific neutralising antibody response, in that all convalescent sera could neutralise an MuV genotype similar to the one detected by RT-PCR in patient saliva, but were often unable to neutralise a different contemporary wild MuV genotype. The goal of vaccination is the generation of a memory response that can rapidly clear infection, thus providing protection to the host. A significant effort has been made towards identifying the attributes of the immune response that are associated with optimal protection. As humoral and cellular immune responses to vaccination play important roles in protection, both must be considered when examining a possible surrogate marker or indicator of protection, as has been described previously for measles virus [32–35]. Earlier findings [22], as well as those presented here, support the hypothesis that a combination of high virus-specific IgG avidity (‡32%) and at least a nominal level of neutralising activity (‡1:8) decreases the risk of mumps infection, even among close contacts. Taken together, the serological and virological findings described here highlight the importance of the specific methods employed in case investigation and in assessing
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
Atrasheuskaya et al. Mumps vaccine failure 675
VF. A VF case investigation protocol should also consider the influence of the specific virus strains used in the serological IgG avidity and PRN assays [36]. The latter is very important in understanding the correlation between the maturation of mumps-specific IgG after infection or immunisation and the neutralising capacity spectrum of immune sera. Notably, antibody avidity testing appears to be a particularly useful assay for examining potential cases of VF. In conclusion, only one dose of vaccine had been administered to the majority of cases of VF identified in this study, and primary VF was implicated in most of these cases, based on a failure to induce markers suggestive of a mature and functional immune response. Whether the absence of sufficient titres of neutralising antibody in response to vaccination, or the absence of neutralising antibody specific for the endemic MuV strain, was responsible for susceptibility of vaccinees to infection cannot be determined conclusively, but both are indicated. What is clear is that causes of VF are complex and multifactorial in nature. A proper assessment of VFs should include IgG avidity testing, as well as a comprehensive assessment of the functional antibody response, i.e., the differential ability of serum antibody to neutralise both the vaccine virus and the endemic challenge virus strains. Other protective responses not captured in this analysis of VFs include cellular and mucosal responses. Such responses deserve more attention and may further clarify the occurrence of VFs. ACKNOWLEDGEMENTS The work was performed in the framework of grant BTEP ID # 16, 2002. No official support or endorsement of this article by the Food and Drug Administration is intended or should be inferred. We thank A. Neverov, T. Malisheva and L. D’achkova for excellent technical assistance, and especially thank A. E. Nesterov for providing data concerning some patients.
REFERENCES 1. Gut J-P, Lablache C, Behr S, Kirn A. Symptomatic mumps reinfections. J Med Virol 1995; 45: 17–23. 2. Crowley B, Afzal MA. Mumps virus reinfection—clinical findings and serological vagaries. Comm Dis Public Health 2002; 5: 311–313. 3. Anonymous. Mumps epidemic—United Kingdom, 2004– 2005. CDC. MMWR 2006; 55: 173–175.
4. Anonymous. USA—mumps activity update; 1 January to 7 October 2006. CDC. MMWR 2006; 55: 1152–1153. 5. Hersh BS, Fine PEM, Kent W et al. Mumps outbreak in a highly vaccinated population. J Pediatr 1991; 119: 187–193. 6. Briss PA, Fehrs LJ, Parker RA et al. Sustained transmission of mumps in a highly vaccinated population: assessment of primary vaccine failure and waning vaccine-induced immunity. J Infect Dis 1994; 169: 77–82. 7. Strohle A, Eggenberger K, Steiner CA, Matter L, Germann D. Mumps epidemic in vaccinated children in West Switzerland. Schweiz Med Wochenschr 1997; 127: 1124–1133. 8. Vandermeulen C, Roelants M, Vermoere M, Roseeuw K, Goubau P, Hoppenbrouwers K. Outbreak of mumps in a vaccinated child population: a question of vaccine failure. Vaccine 2004; 22: 2713–2716. ¨ rvell C, Alsheikhly A-R, Kalantari M, Johansson B. 9. O Characterization of genotype-specific epitopes of the HN protein of mumps virus. J Gen Virol 1997; 78: 3187–3193. ¨ rvell C. Mumps virus 10. No¨jd J, Tecle T, Samuelsson A, O neutralizing antibodies do not protect against reinfection with a heterologous mumps virus genotype. Vaccine 2001; 19: 1727–1731. 11. Agafonov AP, Nicheuchina SN, Patrusheva I et al. Characteristics of parotitis virus, isolated in Siberia. Vopr Virusol 1997; 42: 222–226. 12. Agafonov AP, Streltsova MA, Suslaparov IM, Ignatyev GM. Immune parameters in mumps patients. Vopr Virusol 2001; 1: 30–33. 13. Heider A, Ignatyev G, Gerike E, Schreier E. Genotypic characterization of mumps virus isolated in Russia (Siberia). Res Virol 1997; 148: 433–435. 14. Nesterov AE, Kameneva CN, Maksimov NL, Ignatyev GM. Lymphadenopathy in a previously vaccinated patient with mumps. Infect Dis 2004; 4: 92–94. 15. Utz S, Richard JL, Capaul S, Matter HC, Hrisoho MG, Muhlemann K. Phylogenetic analysis of clinical mumps virus isolates from vaccinated and non-vaccinated patients with mumps during an outbreak, Switzerland 1998–2000. J Med Virol 2004; 73: 91–96. 16. Narita M, Matsuzono Y, Takekoshi Y et al. Analysis of mumps vaccine failure by means of avidity testing for mumps virus-specific immunoglobulin G. Clin Diagn Lab Immunol 1998; 5: 799–803. 17. Davidkin I, Jokinen S, Paananen A, Leinikki P, Peltola H. Etiology of mumps-like illness in children and adolescents vaccinated for measles, mumps, and rubella. J Infect Dis 2005; 191: 719–723. 18. Kunkel U, Driesel G, Henning U, Gerike E, Willers H, Schreier E. Differentiation of vaccine and wild mumps viruses by polymerase chain reaction and nucleotide sequencing of the SH gene: brief report. J Med Virol 1995; 45: 121–126. 19. Elliott GD, Afzal MA, Martin SJ, Rima BK. Nucleotide sequence of the matrix, fusion and putative SH protein genes of MuV and their deduced amino acid sequences. Virus Res 1989; 12: 61–75. 20. Paunio M, Hedman K, Davidkin I, Peltola H. IgG avidity to distinguish secondary from primary measles vaccination failures: prospects for a more effective global measles elimination strategy. Expert Opin Pharmacother 2003; 4: 1215–1225. 21. Muhlemann K. The molecular epidemiology of mumps virus. Infect Genet Evol 2004; 4: 215–219.
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676
676 Clinical Microbiology and Infection, Volume 13 Number 7, July 2007
22. Atrasheuskaya AV, Neverov AA, Rubin S, Ignatyev GM. Horizontal transmission of the Leningrad-3 live attenuated mumps vaccine virus. Vaccine 2006; 24: 1530–1536. 23. Rizzo LV, DeKruyff RH, Umetsu DT. Generation of B cell memory and affinity maturation. Induction with Th1 and Th2 T cell clones. J Immunol 1992; 148: 3733–3739. 24. Lazzarotto T, Varani S, Spezzacatena P et al. Delayed acquisition of high-avidity anti-cytomegalovirus antibody is correlated with prolonged antigenemia in solid organ transplant recipients. J Infect Dis 1998; 178: 1145–1149. 25. Weissbrich B. The use of semi-automated EBV IgG avidity determination for the diagnosis of infectious mononucleosis. J Med Virol 1998; 54: 145–153. 26. Atrasheuskaya AV, Blatun EM, Neverov AA, Karpov IA, Ignatyev GM. Measles in Minsk, Belarus, 2001–2003: clinical, virological and serological parameters. J Clin Virol 2005; 34: 179–185. 27. Paunio M, Hedman K, Davidkin I, Peltola H. IgG avidity to distinguish secondary from primary measles vaccination failures: prospects for a more effective global measles elimination strategy. Exp Opinion 2003; 4: 1215–1225. 28. WHO. Global status of mumps immunization and surveillance. Wkly Epidemiol Record 2005; 48: 417–424. 29. Sanz-Moreno JC, Limia-Sanchez A, Garcia-Comas L et al. Detection of secondary mumps vaccine failure by means of avidity testing for specific immunoglobulin G. Vaccine 2005; 23: 4921–4925.
30. Harling R, White JM, Ramsay ME, Macsween KF, van den Bosch C. The effectiveness of the mumps component of the MMR vaccine: a case control study. Vaccine 2005; 23: 4070–4074. 31. Rubin S, Mauldin J, Chumakov K, Vanderzanden J, Iskow R, Carbone K. Serological and phylogenetic evidence of monotypic immune responses to different mumps virus strains. Vaccine 2006; 24: 2662–2668. 32. Muller CP, Huiss S, Schneider F. Secondary immune responses in parents of children with recent measles. Lancet 1996; 348: 1379–1380. 33. Huiss S, Damien B, Schneider F, Muller CP. Characteristics of asymptomatic secondary immune response to measles virus in late convalescent donors. Clin Exp Immunol 1997; 109: 416–420. 34. Bautista-Lopez N, Ward BJ, Mills E, McCormick D, Martel N, Ratnam S. Development and durability of measles antigen-specific lymphoproliferative response after MMR vaccination. Vaccine 2000; 18: 1393–1401. 35. Ovsyannikova IG, Dhiman N, Jacobson RM, Vierkant RA, Poland GA. Frequency of measles virus-specific CD4+ and CD8+ T cells in subjects seronegative or highly seropositive for measles vaccine. Clin Diagn Lab Immunol 2003; 10: 411–416. 36. Pipkin PA, Afzal MA, Heath AB, Minor PD. Assay of humoral immunity to mumps virus. J Virol Methods 1999; 79: 219–215.
2007 The Authors Journal Compilation 2007 European Society of Clinical Microbiology and Infectious Diseases, CMI, 13, 670–676