Comparison of Enzyme-Linked Immunosorbent Assay and Respiratory

INFECTION AND IMMUNITY, July 1982, p. 160-165 0019-9567/82/070160-06$02.00/0

Vol. 37, No. 1

Comparison of Enzyme-Linked Immunosorbent Assay and Neutralization Techniques for Measurement of Antibody to Respiratory Syncytial Virus: Implications for Parenteral Immunization with Live Virus Vaccine MUFSON,' EUGENE B. BUYNAK,2 ARLENE A. McLEAN,2 AND MAURICE A. HILLEMAN2 Section of Infectious Diseases, Department of Medicine, Marshall University School of Medicine, Huntington, West Virginia 25701,' and Division of Virus and Cell Biology Research, Merck Institute for Therapeutic Research, West Point, Pennsylvania 194862

ROBERT B. BELSHE,1* LEE P. VAN VORIS,1 MAURICE A.

Received 8 May 1981/Accepted 22 February 1982

The sensitivity of an enzyme-linked immunosorbent assay (ELISA) to detect low levels of antibody to respiratory syncytial (RS) virus was compared with a tube dilution neutralization test (NEUT) on sera obtained from children who received a parenteral live RS virus vaccine. Among the children who developed antibody in response to live RS virus vaccine, ELISA was as sensitive as NEUT at detecting antibody increases. Some children who did not have detectable prevaccine ELISA antibody possessed NEUT antibody; these children were generally less than 12 months pld, suggesting that they had low levels of maternal antibody. Low levels of NEUT or ELISA antibody were associated with the absence of antibody increases after injection of live RS virus vaccine. The quantity of antibody stimulated by this live RS virus vaccine was small compared with that which was stimulated by naturally acquired RS virus infection. We concluded that ELISA is a satisfactory test for determining antibody to RS virus in vaccine field trials, given the understanding that low levels of preexisting antibody are not detected in some instances. No effective vaccine has been developed to prevent infection with respiratory syncytial (RS) virus, the single most important respiratory pathogen of early life (14). RS virus vaccines previously evaluated in infants and children have included an inactivated, alum-precipitated vaccine for parenteral administration (10), a cold-adapted, live attenuated virus vaccine for intranasal administration (9), and two live attenuated temperature-sensitive mutants of RS virus for intranasal administration (8; Belshe, unpublished data). To date, none of these four vaccines has proven satisfactory for use in the immunoprophylaxis of RS virus infection. Recently, we tested a parenterally administered live RS virus vaccine (6). During the initial evaluations of this vaccine, Buynak and coworkers reported that greater than 90% of seronegative infants and children who received this live RS virus vaccine developed postvaccine serum-neutralizing antibody (4, 6). However, during the course of our evaluation of this live RS virus vaccine, we observed that one-third of seronegative infants failed to seroconvert after vaccination as measured by enzyme-linked immunosorbent assay (ELISA) (2a). ELISA, as described by Richardson et al. (15), was superior

to the plaque reduction and complement fixation (CF) tests for determining antibody responses

during naturally acquired RS virus infection in young infants. The tube dilution neutralization test (NEUT) used by Buynak et al. (6) to assay serum antibody levels to RS virus differed from the standard plaque reduction assay (12) generally used to measure RS virus antibody. Subsequently, Buynak et al. (4) reported that the addition of complement increased the sensitivity of their assay, analogous to the findings of Mills et al. (12), who showed that added complement increased the sensitivity of the plaque reduction assay. To determine the relative abilities of the ELISA and NEUT techniques to detect low levels of antibody to RS virus and also to detect antibody increases after parenteral vaccination with live RS virus vaccine, we tested a group of sera by both methods. We also compared the level of antibody induced by this vaccine with that induced by natural infection with RS virus. This communication summarizes the findings of our comparative studies. MATERIALS AND METHODS Vaccine. Live RS virus vaccine (lot 592; Merck & Co., Inc., Rahway, N.J.) was stored lyophilized in 160

VOL. 37, 1982

ANTIBODY RESPONSE TO LIVE RS VIRUS VACCINE

two-dose vials at -20°C (6). The vaccine was reconstituted with sterile water immediately before use. Each 0.5-ml dose contained 103 50o tissue culture infective doses of RS virus. Placebo vaccine consisted of tissue culture fluid and was given in 0.5-ml doses. Volunteers. At the time children visited their pediatricians for routine health maintenance, their parents were asked if they wished to enroll their child in a placebo-controlled efficacy field trial of live RS virus vaccine. Only well children who were 6 months to 3 years 11 months old were enrolled in the study. After written informed consent was obtained from the parents, either live RS virus or placebo vaccine was administered subcutaneously. The vaccines were administered according to a double-blind and randomized protocol. This investigation was approved by the Human Studies Committee of the Marshall University School of Medicine, Huntington, W. Va. Sera. Sources of sera used in this study are shown in Table 1. Serum specimens were collected from each vaccinated child before and 1 month after vaccination with either live RS virus vaccine or placebo vaccine. Sera were also collected from placebo-vaccinated children who subsequently became naturally infected with RS virus; these sera were collected before the onset of illness (as part of ongoing sero-epidemiological surveillance for naturally occurring RS virus infection) and at least 21 days after infection with RS virus. RS virus infection was confirmed in these individuals by isolation of RS virus from respiratory secretions. Serological Tests. ELISA tests were performed as previously described by Richardson and co-workers, using an RS virus-infected HEp-2 cell suspension that had been adsorbed to polyvinyl microtiter plates (Dynatech Laboratories, Inc., Alexandria, Va.) (15). After washing with phosphate-buffered saline-Tween to remove unadsorbed material, test sera in dilutions of 1:100, 1:400, 1:1,600, 1:6,400, and 1:25,600 were added to duplicate wells previously adsorbed with either an RS virus-infected cell suspension (antigen) or an uninfected HEp-2 cell suspension (control). A lower dilution of serum (1:25) frequently reacted with the control cell suspension; therefore, 1:100 was the lowest dilution of serum regularly employed (15). Reac-

TABLE 1. Sources of sera Immune stimulus

No. of sera

(No. of pairs)

112a (56) Live RS virus vaccine 126a (63) Placebo vaccine 74a ( )b Live RS virus or placebo vaccine 40 (20) Naturally acquired RS virus infectionc a Sera included in Fig. 1 (n = 312). b Pre- or postvaccine serum that was unpaired. The corresponding serum from the pair could not be tested by NEUT owing to insufficient quantity of available serum.

c In each instance, RS virus was isolated from respiratory secretions to confirm infection with this agent. All 20 pairs were tested by ELISA, 6 pairs were tested by NEUT, and 6 pairs were tested by CF. Each pair selected for testing was from a placebo vaccine recipient undergoing infection in the first epidemic after enrolling in the vaccine study.

161

ELISA

CR.c

caO

42444'8 2 4 8 16 32 6412Sk25S

Neutralizing Antibody Titer (Reciprocal)

FIG. 1. Comparison of ELISA titers with NEUT titers among 312 sera from children given live RS virus or placebo vaccine. Sera were tested by NEUT at an initial dilution of 1:2 when sufficient volume was available; otherwise, they were tested at 1:4 or 1:8 initial dilution.

tion of serum and antigen was carried out overnight at 4°C, and the unreacted serum was removed by washing. Subsequently alkaline phosphatase-conjugated goat anti-human immunoglobulin G (Miles Laboratories, Inc., Elkhart, Ind.) was added to each well, and after 2 h of incubation, substrate (p-nitrophenyl phosphate [Sigma 104]; Sigma Chemical Co., St. Louis, Mo.) was added. The test plates were incubated further for 30 mi at 37°C, and the absorbance at 400 nm was determined for each well with a Multiskan ELISA reader (Flow Laboratories, Inc., Rockville, Md.). A single reference serum was included in each group of tests as a standard. This reference serum has been used many times, both in Richardson's laboratory and in our own, for this purpose, with the antibody titer set at 1:1,600 as previously described (15). Antibody titer for each test serum was calculated as follows: a calibration line was fitted by the least-squares method to a series of log dilutions and the corresponding logit absorbances of this standard. The fitted logit absorbance (y*) of a preselected dilution (1:1,600) was derived from the slope and intercept of the calibration line. For each experimental serum, a line was fitted by the least-squares method to a series of log dilutions and the corresponding logit absorbances. The titer of the experimental serum was taken to be the antilog of the fitted log dilution on the line in question which corresponded to y*. A similar procedure which omits the use of the logit transformation of the absorbance

162

INFECT. IMMUN.

BELSHE ET AL.

RESULTS Antibody titers to RS virus measured by ELISA and NEUT were compared for the 312 pre- or postvaccine sera (Fig. 1). Of the 211 ELISA antibody-negative sera (ELISA titer,