Immunologic responses to vaccinia vaccines administered by different parenteral routes.

756 Immunologic Responses to Vaccinia Vaccines Administered by Different Parenteral Routes David J. McClain, Shannon Harrison,* Curtis L. Yeager,* Divisions of Virology and Medicine, US Army Medical Research Institute of Infectious Diseases, Fort Detrick, Frederick, Maryland; John Cruz, Francis A. Ennis, Paul Gibbs, Brooke Army Medical Center, Fort Sam Houston, San Antonio, Texas; Michael S. Wright,* Peter L. Summers, James D. Arthur, Division of Infectious Diseases and Immunology, Department of and Jess A. Graham* Medicine, University of Massachusetts Medical School, Worcester, Massachusetts To develop a less reactogenic but equally immunogenic vaccine, this study of 91 human volunteers compared the safety and immunogenic potency of a new, cell culture–derived vaccinia virus vaccine administered intradermally and intramuscularly with the licensed vaccinia vaccine administered by scarification. Cutaneous pox lesions developed in a higher proportion of scarification vaccinees. Scarification and intradermal vaccine recipients who developed cutaneous pox lesions had more local reactions but also achieved significantly higher cell-mediated and neutralizing antibody responses than those who did not develop pox lesions. Although less reactogenic, intradermal or intramuscular administration of vaccinia vaccine without the concomitant development of a cutaneous pox lesion induced lower immune responses. Vaccinia virus is a member of the Orthopox genus of the Practices. Certain US military units may require smallpox vacpoxvirus family and has little virulence for immunocompetent cination during future deployments. Apart from military needs, humans. Apart from its critical role in the eradication of enthe Centers for Disease Control and Prevention (CDC) has demic smallpox, vaccinia virus has several biologic properties recommended vaccination for persons working in the laborathat make it an excellent candidate for introducing foreign tory with vaccinia or recombinant vaccinia viruses [7]. Theregenes that prompted the investigation of recombinant vaccinia fore, it is important that modern production techniques be apvaccines [1–3]. However, cellular immunity and antibody replied to the manufacture of a vaccinia virus vaccine to make sponses to vaccinia virus administered by different routes have it safe and effective as a replacement for the current calf lymph not been compared in humans. These issues bear importance smallpox vaccine. Given the attendant risks associated with the on how such vaccines will be used in humans and perhaps traditional method of cutaneous inoculation of vaccinia, we why immune responses to recombinant vaccinia vaccines have sought to evaluate the clinical and immune responses of alternarequired multiple injections [4–6]. tive routes of inoculation by using a new, cell culture–derived The only vaccinia vaccine currently licensed in the United vaccinia vaccine. States was prepared from calf lymph and is stored as a freezeThe vaccinia vaccines used in the smallpox eradication effort dried product. Supplies of this licensed vaccine cannot be rewere prepared on a large scale by inoculating the shaved abdoplaced, as the production method for calf lymph vaccine is mens of calves, sheep, or water buffalo with seed stocks of archaic and there is no adequate facility in which supplies vaccinia virus, harvesting the infected exudative lymph from can be regenerated in compliance with Good Manufacturing the inoculation sites, and bottling the product with phenol and brilliant green as bacteriostatic agents [8, 9]. Partly because of the nature of this production, in which bacterial contamination was expected, the vaccines were administered percutaneously Received 5 August 1996; revised 19 November 1996. with a bifurcated needle, a process that became known as scariInformed consent was obtained from human volunteers in accordance with human experimental guidelines of the US Department of Health and Human fication because of the permanent scar that resulted. Formation Services and the Office of the Surgeon General of the US Army (ref: AR 70– of a cutaneous pox lesion that healed with a scar formed an 25). The views, opinions, and/orfindings contained herein arethose of the authors important method for verifying vaccination status during the and should not be construed as an official Department of the Army position, era of endemic smallpox. This method proved effective and policy, or decision unless so designated by other documentation. successful when applied by the World Health Organization to Reprints or correspondence: Dr. David J. McClain, Division of Virology, USAMRIID, Ft. Detrick, Frederick, MD 21702-5011. a campaign to globally eradicate smallpox [10]. *Present affiliations: 465 Pine Dr., Jackson Hole, Wyoming (S.H.); MCHJAs a consequence of percutaneous inoculation, infectious CI, Madigan Army Medical Center, Tacoma, Washington (C.L.Y.); Abbott vaccine virus was present in the local lesion after scarification. Diagnostic Division, Abbott Laboratories, Irving, Texas (M.S.W.); Department of Radiology, Brooke Army Medical Center, Fort Sam Houston, San Antonio, Subsequently, there were cases of inadvertent autoinoculation Texas (J.A.G.). and inoculation of susceptible vaccinee contacts. Additional The Journal of Infectious Diseases 1997;175:756–63 complications included severe local spread of vaccine virus in q1997 by The University of Chicago. All rights reserved. 0022–1899/97/7504–0003$01.00 persons with chronic skin diseases [11]. Consequently, if it is / 9d21$$ap43 02-11-97 18:22:00 jinfa UC: J Infect Downloaded from https://academic.oup.com/jid/article/175/4/756/943818 by guest on 02 May 2024 757JID 1997;175 (April) Pox Lesion Required for Vaccinia Immunity screened by medical history and physical examination before being possible to do so without loss of immunogenicity of the vacenrolled, with specific attention to the presence of a vaccinia vaccicine, it is desirable to administer vaccinia by a route that does nation scar or any contraindication to vaccination. Laboratory panot result in cutaneous lesions containing transmissible virus. rameters before enrollment included HIV serologic test, serum Because of these safety concerns, human trials with recombichemistries, a complete blood cell count, and a pregnancy test. nant vaccinia vaccines have used the injectable route. Subjects were accepted if they were in good health, had no vaccinia We tested a new investigational vaccinia vaccine (BB-IND scar or history of vaccination, and had no significant abnormalities 4984) produced in cell culture. It was developed for parenteral that indicated an increased risk for vaccinia immunization (i.e., injection to preclude the potential complication of inadvertent exfoliative skin disease or disorders of cellular immunity). The inoculation of virus attendant to scarification. During preclinicriteria that excluded a person from participating in the protocol cal studies, this cell-cultured vaccinia vaccine candidate proved were the same as those recommended by the CDC [7]. Volunteers who lacked a vaccination scar but were subsequently proved to be comparable to the Bureau of Biologics (New York Board of vaccinia-immune by their baseline serologic test (50% plaqueHealth) reference strain with respect to pock formation on reduction neutralization titer [PRNT] of §1:20) were excluded chorioallantoic membranes of embryonated chicken eggs, forfrom statistical analyses. mation of lesions after adult rabbits were inoculated intraderVaccinations and study design. Volunteers were randomized mally (id), and intracerebral and intraperitoneal virulence in into 3 groups, with one-third receiving licensed vaccinia vaccine adult and suckling outbred mice (unpublished data). A subseby scarification, the cell culture–derived vaccinia vaccine im, or quent phase I dose-escalation trial evaluated this vaccine in the cell-cultured vaccinia vaccine id. In addition, one-half of volunhumans given subcutaneous inoculations. However, cutaneous teers vaccinated id were randomly selected to have the inoculation vesicular (pox) lesions developed in an increasing number of site wiped with alcohol immediately after injection. This procedure volunteers as the subcutaneous dose was increased. Neutralizwas to determine if immune recognition afforded by processing ing antibody and lymphocyte proliferation assays indicated a of antigen id could be achieved without the formation of a cutaneous pox lesion. All volunteers vaccinated im underwent alcohol higher and earlier immune response in vaccinees with cutawiping at the injection site after inoculation to minimize the chance neous lesions than in vaccinees who did not develop pox lesions of accidental dermal inoculation with the vaccine. All inoculation (McClain DJ, unpublished observations). Since the majority of sites were initially covered with a semipermeable dressing until pox lesions occurred in volunteers receiving the highest dose, any pox lesion had scabbed or until day 10 after inoculation (if confounding of pox lesion formation with dose could not be no pox lesion developed). During the month following vaccination, excluded in the statistical analysis. Therefore, we undertook a subjects were seen as outpatients twice a week during the first 2 larger study to examine whether formation of a cutaneous pox weeks after vaccination, and then weekly for the next 2 weeks. lesion was critical for optimal immune responses compared These outpatient examinations consisted of clinical examinations with injection of the cell-cultured vaccinia vaccine. and laboratory tests to assess adverse reactions, potential complications, and immunogenicity. Clinical assessments. Volunteers underwent semiweekly asMethods sessment for local signs or symptoms around the inoculation site Vaccines. The licensed vaccinia (or smallpox) vaccine (Wyand for potential systemic symptoms related to vaccination. Any eth-Ayerst Laboratories, Philadelphia) is a lyophilized vaccinia local reaction at the inoculation site was measured to determine virus derived from the New York Board of Health strain [7] and the diameter of any local erythema and induration and scored for is the only remaining licensed vaccinia vaccine in the United the presence or absence of warmth, tenderness, lymphadenopathy, States. The cell-cultured vaccinia vaccine was manufactured acsubcutaneous nodule, or a vesicle (pox) lesion. Systemic reactions cording to Good Manufacturing Practices after three successive were assessed by questioning and examining the volunteers for the passages in MRC-5 cells. This vaccine lot was derived from the presence and severity of fever, chills, malaise, headache, myalgia, master seed of the smallpox vaccine previously licensed by Conarthralgia, loss of appetite, nausea, vomiting, diarrhea, pruritus, naught Laboratories (Philadelphia). The Connaught vaccine was or rash. These symptoms were quantitatively scored based upon one of several licensed vaccinia products used in the United States severity (0 Åno symptom; 1 Åmild, symptom can be ignored; until the end of smallpox vaccination in the early 1970s. 2Åmoderate, symptom affects activity but is relieved by analgeThe cell culture–derived vaccinia vaccine was administered in sics; and 3 Åsevere, symptom cannot be relieved by analgesics). a dose of 5.1 log 10 pfu, either as 0.1 mL id or 0.5 mL intramuscuData were entered into a database for subsequent statistical analylarly (im). The licensed vaccinia vaccine was administered by sis. The total local or systemic symptom score for a given vaccinee scarification (as directed in the package insert) by dipping a sterile was defined as the sum of all scores for either local or systemic bifurcated needle into the vaccine and then pricking the skin inocusymptoms. lation site three times. The licensed vaccinia vaccine has a virus Clinical laboratories. Blood samples were obtained by weekly titer of Ç10 8 pfu/mL, and the administration of 1 drop via scarifiphlebotomy of volunteers, beginning before vaccination until Ç1 cation with a bifurcated needle is estimated to deliver 2.5 m L. month after vaccination. A complete blood cell count with a fiveTherefore, Ç10 5 pfu is delivered by this method. All vaccinations part differential cell count was done using a cell counter (CellDyn were administered into the deltoid region of the arm. 3000; Abbott Laboratories, Abbott Park, IL). Serum samples were Subjects. The vaccines were administered in an open-label analyzed by an Ektachem 700XR (Eastman Kodak, Rochester, NY) for a panel of chemistry measurements (sodium, potassium,study to healthy vaccinia-naive volunteers. Volunteers were / 9d21$$ap43 02-11-97 18:22:00 jinfa UC: J Infect Downloaded from https://academic.oup.com/jid/article/175/4/756/943818 by guest on 02 May 2024 758 McClain et al. JID 1997;175 (April) chloride, bicarbonate, urea nitrogen, creatinine, glucose, calcium, 2% heat-inactivated FBS. Cultures were incubated at 357Cina humidified atmosphere of 5% CO 2 for 4–6 days. Cultures werephosphorus, lactate dehydrogenase,aspartate transaminase, alanine transaminase, g glutamyl transferase, alkaline phosphatase, total decanted and fixed with 1 mL (200 m L for the 96-well plates) of 10% formalin for 15 min at room temperature. Plates were incu-bilirubin, and creatine phosphokinase). Test value means were calculated for each group on a given day of measurement. The bated with 1 mL (200 m L for the 96-well plates) of HBSS containing 1% bovine serum albumin (BSA) for 30 min. The blockerdata from each hematologic and serum chemistry test were analyzed using Statistical Analysis System procedure GLM (version buffer was removed and 0.5 mL (50 m L for the 96-well plates) of a 1:1000 dilution of vaccinia mouse hyperimmune ascitic fluid6.10; SAS, Cary, NC) repeated measures analysis of variance (ANOVA). The overall differences between groups over the course (ATCC, Rockville, MD) or normal mouse ascitic fluid was added to the appropriate wells for 1 h at 357C. After three washes, 0.2of the study period were compared by using the univariate tests of hypothesis for within-subjects effects. mL (35 m L for the 96-well plates) of a 1:2000 dilution of peroxidase-labeled anti–mouse IgG (Kirkegaard & Perry) was added toSerologic assays. Serum specimens from days 0, 10, 14, 20, and 27 were frozen for subsequent ELISA and PRNT assay. An all wells for 1 h at 357C. Plates were then washed five times and incubated 30 min at 357C with 0.5 mL/well (80 m L/well for theELISA was performed as previously described [12] to assay for antibodies reactive with cell lysate antigens from vaccinia virus– 96-well plates) of ABTS (2,2’-azino-di[3-ethyl-benzthiazoline sulfonate]) substrate (Kirkegaard & Perry). Test samples and positiveinfected cells, with the modification of human sera as the test specimen and goat anti–human IgG as the detector antibody and negative controls were tested in duplicate. A positive control was prepared using 10-fold dilutions of vaccinia virus in negative(Kirkegaard & Perry, Gaithersburg, MD). Given a lower limit of detection of the assay at a 1:100 test serum dilution, negative titers antibody serum. Using these ‘‘spiked’’ serum samples, 2 wells were inoculated with each dilution. Two wells were also inocu-(i.e., optical density comparable to background) were reported as equal to 1:50. lated with normal (noninfectious) human serum as a negative control. Results were read visually or spectrophotometrically atPRNT, an in vitro test of serum’s ability to neutralize the Wyeth strain of vaccinia virus, was determined using a modification of 414 nm. Titers were calculated as TCID 50 values according to the method of Reed and Muench [16] or as the highest specimenthe method of Earley et al. [13]. Briefly, each coded serum sample was incubated at 567C for 30 min, then diluted 1:10 in Eagle MEM dilution with an optical density of 0.2 units over that of the negative controls.(EMEM) containing 10% heat-inactivated fetal bovine serum (FBS). A suspension of vaccinia virus, calculated to yield a dose Statistical methods. The PRNT and ELISA antibody responses were analyzed by variance with repeated measures followed byof Ç40–100 pfu/0.1 mL, was prepared in Hanks’ balanced salt solution (HBSS) with 40 mMHEPES. Two-fold dilutions of serum multiple comparisons for the study groups using the Tukey-Kramer adjustment for multiplicity [17]. The data from each hematologicsamples were then mixed 1:1 with 40–100 pfu of vaccinia virus suspension and incubated at 377C for 1 h. After incubation, test and serum chemistry test and lymphocyte transformation assays were analyzed for each study group by using SAS procedure GLMsamples and controls were inoculated onto monolayers of Vero cells in 12-well cell culture plates. After adsorption for 1 h at ANOVA. The overall differences between groups over the course of the study were compared by using the univariate tests of hypoth-377C, each monolayer was overlaid with 1 mL of 0.5% agarose (FMC Bioproducts, Rockland, ME) containing HEPES-buffered esis for within-subjects effects. All statistical tests were performed at the a Å.05 level unless otherwise indicated.Eagle basal medium with Earle’s salts, and 5% heat-inactivated FBS. After incubation for 2 days at 377C in a humidified atmosphere of 5% (vol/vol) carbon dioxide (CO 2 ), each monolayer was stained with 1 mL of the agarose overlay containing 0.167 mg/ Results mL neutral red dye. The plates were returned to the incubator for 24–36 h and the plaques were counted. The number of plaques Subjects. Ninety-one volunteers participated in this phase for each test sample was entered into a computer program to deterII study. Data from 8 volunteers were excluded from statistical mine the PRNT using probit analysis. The end-point titer was analysis because of preexisting vaccinia immunity as deterthe highest serum dilution demonstrating ú50% reduction in the mined by baseline serologies. Of the remaining 83 subjects, 11 number of plaques for the average dose. were female and 72 were male. During the study, 3 volunteers Lymphocyte proliferation assays. Proliferative responses of developed medical problems that were judged as unrelated to peripheral blood mononuclear cells (PBMC) to live vaccinia virus and heat-inactivated antigen were tested as previously described the protocol: 1 volunteer developed folliculitis 1 week after [14, 15]. Results were expressed as a stimulation index, derived inoculation; another developed lower extremity cellulitis 2 by dividing counts in wells containing antigen by counts in wells weeks after inoculation secondary to an infected leg laceration; without vaccinia antigen or virus. and a third developed nausea, vomiting, and abdominal pain, In situ ELISA viremia assay. Volunteers underwent periodic which led to subsequent laparoscopic appendectomy on day 8 serum sampling within the first 2 weeks after inoculation, with an after inoculation, with normal appendiceal histopathology and aliquot frozen at 0707C until later assay for vaccinia viremia. eventual full recovery. Serum was assayed for viremia by a modification of an in situ Clinical assessments. ANOVA for vaccine effects at each ELISA. Briefly, Vero cells grown in 24or 96-well plates were week yielded no significant differences between the study inoculated with 0.1 mL or 50 m L, respectively, of a serum sample. groups’ temperature profiles (temperature maximums recorded After adsorption for 1 h at 357C, the 24-well plates were re-fed with 1 mL (100 m L for the 96-well plates) of EMEM containing each week). / 9d21$$ap43 02-11-97 18:22:00 jinfa UC: J Infect Downloaded from https://academic.oup.com/jid/article/175/4/756/943818 by guest on 02 May 2024 759JID 1997;175 (April) Pox Lesion Required for Vaccinia Immunity The vaccinees inoculated id who received alcohol wiping areas of contact dermatitis from dressing adhesive were observed.after inoculation and the vaccinees inoculated im did not significantly differ with respect to incidence of pox lesion (25% Clinical laboratories. Laboratory findings were based on the analysis of the weekly mean for each subject for weeksvs. 4.5%, PÅ.141). However, failure to wipe with alcohol increased the incidence in the group inoculated id to 62.5%, 01, 0, 1, 2, 3, and 4. Due to some missing values when volunteers omitted scheduled phlebotomies, no repeated measureswhich was significantly higher then that in the group inoculated im (Põ.001). This was significantly less than the incidence ANOVA could be done. Hence, an ANOVA was done at each week. Only sporadic differences were detected for any post-of pox lesions in the scarification group (96.6%), which was higher than that in all the other groups (Põ.001). treatment week (noted in table 1.) Serologic and virologic assays. PRNTs were significantlyThe scarification group had the highest incidence of local reactions, the group inoculated im experienced the fewest, and greater for the scarification group than for the groups inoculated either im or id beginning at day 13 after inoculation (Põ.001)the subgroups inoculated id were intermediate and indistinguishable from each other. Local symptom analysis indicated using a repeated measures ANOVA. Moreover, volunteers from either scarification or groups inoculated id who developedhighly significant statistical differences between the study groups, using both ANOVA and nonparametric Wilcoxon analcutaneous pox lesions had significantly higher neutralization titers than those without pox lesions in comparisons adjustedyses (Põ.001). There was a higher total symptom score for the scarification group (mean, 169.2) and a significantly lower for prevaccination baseline. There was no significant difference between volunteers inoculated by scarification or id who hadscore for the group inoculated im (mean, 3.0). Meantotal scores for vaccinees inoculated id with and without alcohol wiping developed a cutaneous vesicle. However, there was a statistically insignificant trend toward higher titers in the scarificationwere 61.7 and 90.4, respectively. Additional analysis exclusive of the group inoculated im revealed that the scarification group (Wyeth vaccine) recipients. These PRNT responses are illustrated in figure 1.still differed from the subgroups inoculated id (Põ.001 by ANOVA or nonparametric analysis). When the two subgroups As illustrated in figure 2, vaccinia ELISA antibody responses were significantly higher for the scarification group at day 27inoculated id were compared, no differences were found in local symptom scores (PÅ.215 by ANOVA, PÅ.406 by after vaccination than for either of the groups inoculated id or im (Põ.001). In contrast to PRNT responses, significantlynonparametric analysis). When local symptom scores were compared between volunteers who did and did not develop higher ELISA titers were found in scarification vaccinees than in those inoculated id who formed a cutaneous pox lesion (Pcutaneous pox lesions, overall and weekly score differences were significant (PÅ.0001, nonparametric Wilcoxon test). õ.001). There was greater standard error in this assay than in the PRNT assay. No viremia was detected in any volunteer.Systemic symptom analysis indicated no significant differences in the study groups with respect to total symptom scores Lymphocyte proliferation assays. Lymphocyte proliferation assays examined stimulation indices before and after vacci-either by ANOVA (PÅ.352) or by nonparametric Wilcoxon test (PÅ.412). However, higher mean systemic scores ocnation to both live vaccinia virus and heat-inactivated antigen. For the heat-inactivated antigen, responses to scarification werecurred 2 weeks after inoculation in volunteers who developed a cutaneous pox lesion (PÅ.038). All volunteers initially had their immunization sites covered with a vapor-permeable surgical dressing until pox lesions Table 1. Statistically significant differences in laboratory values of groups of subjects inoculated against vaccinia virus intradermally (id) scabbed. However, 4 volunteers with primary vesicles from or intramuscularly (im) or by scarification (scar). their vaccinations subsequently developed secondary pox lesions adjacent to the inoculation site, underneath the dressing. id We noted that the dressings were occlusive enough to accumulate perspiration underneath and around the inoculation site. Study With Without Test week wipe wipe im Scar P Consequently, in the majority of volunteers with pox lesions, vesicular exudate accumulated under these dressings despite Creatinine (mg%) 1 1.2 1.08 1.12 1.18 .0175 frequent dressing changes. This virus-containing exudate Potassium (mEq/L) 3 4.2 4.49 4.25 4.49 .0043 seeped under the dressing and covered areas much greater than Phosphorus (mEq/L) 4 4.05 4.09 4.57 4.45 .0274 that of the primary pox lesion, with subsequent secondary or % monocytes 2 7.23 8.53 7.6 9.34 .0032 % monocytes 1 6.16 8.1 8.24 7.54 .0392 autoinoculation. Subsequently, a nonocclusive and smaller dressing (i.e, dry gauze) was applied over the inoculation sites NOTE. Data were analyzed for each study group by using SAS Procedure of the remaining volunteers. Dressings were changed if there GLM analysis of variance. Overall differences between groups over study period were compared by using univariate tests of hypothesis for within-subwas a scheduled clinical check, the bandage became wet, or jects effects ( a Å.05). Monocytes are expressed as % of total white blood there were any visible signs of dried exudate on the exterior cells in complete blood cell count. With wipe Åinoculation id with local surface of the bandage. After the institution of the dry gauze alcohol wiping after inoculation; without wipe Åinoculation id without alcohol wiping after inoculation. dressings, no further incident of secondary pox lesions or large / 9d21$$ap43 02-11-97 18:22:00 jinfa UC: J Infect Downloaded from https://academic.oup.com/jid/article/175/4/756/943818 by guest on 02 May 2024 760 McClain et al. JID 1997;175 (April) vaccines is not a new idea. Between 1930 and 1975, at least 8 strains of vaccinia virus were developed for parenteral administration in order to decrease virulence [18]. All were attenuated when compared with standard lymph strains, and some were possibly overattenuated, producing low neutralizing antibody levels after primary and booster inoculations [19]. With the eradication of smallpox following closely upon initial development of most of these vaccines, comparable safety and efficacy data were not obtained on a scale comparable to that for lymph vaccines. In addition, data about cell-mediated responses were not documented for these injectable products. In this study, vaccination reactions were readily apparent from clinical observation alone. Extensive laboratory monitoring of vaccinees identified no specific safety concerns or issues about the routes of vaccine administration. There was no evidence of viremia in any volunteer. Serum chemistry or hematology value differences between study groups were sporadic and relatively small. Given the number of data points and comparisons in this study, these were likely to be type I errors. As Figure 1. Vaccinia plaque-reduction neutralization titer (PRNT) these statistical differences do not represent medically im50% responses as illustrated by geometric mean titers (GMTs). PRNTs were significantly higher for scarification inoculees on reportant differences in a contiguous time frame, we do not attripeated measures analysis of variance (Põ.001) but not significantly bute them to causal events related to respective treatments. higher than intradermal (id) injection recipients who developed a The study indicated that, although not severe, local reactions cutaneous pox lesion (id w/ pox) after vaccination (PÅ.09). id corresponded to formation of a cutaneous pox lesion. As exw/o pox Åintradermal injection recipients without pox lesion after pected, pox lesion incidence was significantly higher in the vaccination; im Åintramuscular injection recipients. scarification group than inthe other groups, as were local symptoms. In addition, higher systemic scores occurred 2 weeks after inoculation in volunteers who developed a cutaneous pox significantly greater than those of either subgroup inoculated im or id with alcohol wiping after inoculation (PÅ.0258) but not significantly different from those in the group inoculated id without alcohol wiping (PÅ0.46). With live vaccinia virus as the antigen, scarification vaccinees had significantly higher stimulation indices (PÅ.0037) than did either group inoculated im or id. Both higher indices and greater standard deviation were seen in the assaywith heat-inactivated antigen in comparison to the live virus assay. For volunteers who developed a pox lesion (whether id or by scarification), the stimulation indices before and after vaccination were significantly greater than for vaccinees who did not develop a cutaneous pox lesion, whether assayed using live virus or heat-inactivated antigen (Põ.001). There was no significant difference in lymphocyte stimulation responses between those who formed pox lesions from the licensed vaccinia vaccine via scarification and those with pox lesions from the cell-cultured vaccinia vaccine. These mean lymphocyte responses for heat-inactivated antigen are illustrated in figure 3. Figure 2. ELISA responses to vaccinia-infected whole cell lysates Discussion as illustrated by geometric mean titers (GMTs). ELISA titers were significantly higher for scarification inoculees on repeated measures This human study compared the safety and immunogenicity analysis of variance (Põ.001), even in comparison to intradermal of administration id and im of a cell culture–derived vaccinia (id) injection recipients who formed cutaneous pox lesions (Põ vaccine with those of the licensed vaccine administered by the .001). id w/o pox Åid recipients without poxlesions after vaccination; im Åintramuscular injection recipients. traditional route of scarification. The use of injectable vaccinia / 9d21$$ap43 02-11-97 18:22:00 jinfa UC: J Infect Downloaded from https://academic.oup.com/jid/article/175/4/756/943818 by guest on 02 May 2024 761JID 1997;175 (April) Pox Lesion Required for Vaccinia Immunity to a dose of 10 7.8 pfu, but with an attendant increased risk of a cutaneous pox lesion. Although the crossing of an additional tissue plane with injection im may have lessened this occurrence, we were concerned that injection im of a similar high dose of vaccinia virus might risk viremia. The study also demonstrated that inoculation id will not reliably prevent the formation of a cutaneous pox lesion. Phase I data had raised similar concerns regarding the subcutaneous route of inoculation. In a study by Connor et al. [21], typical Jennerian vesicles were seen in 4%–9% of children vaccinated subcutaneously with 10 3 ,10 4 ,or10 5 pfu. Furthermore, it is apparent that the presence of such a lesion, although undesirable for safety reasons, is necessary for the most robust immune responses. By cleansing the inoculation site after inoculation, we sought to determine if immune recognition afforded by intradermal processing of antigen could be achieved without the safety risks attendant to a cutaneous pox lesion. This concept stemmed from a previous observation that the cutaneous vesicle was prevented after inoculation id with vaccinia if the needle wound was immediately cleansed with alcohol [22]. Figure 3. Lymphocyte proliferation transformation responses to These results indicate that replication of the virus in the skin, heat-inactivated vaccinia antigen. Stimulation indices were significantly higher for scarification inoculees on analysis of variance (PÅ not merely exposure to antigen-presenting cells (i.e., Langer- .0258) but not significantly higher than those for intradermal (id) hans cells) in that location, is essential to generate optimal injection recipients who developed cutaneous pox lesions after vacciimmune responses to vaccinia antigens. nation (PÅ.46). id w/o pox Åid injection recipients without pox Vaccinees who developed cutaneous pox lesions had siglesions after vaccination; im Åintramuscular injection recipients. nificantly higher ELISA and neutralizing antibody responses. The poor responses to inoculation id (in those without pox lesions) and im confirm and expand on earlier observationslesion. Either vaccine dissimilarities or the route of vaccination could account for the symptomatic differences between scariconcerning injectable vaccinations in children [19]. The investigators of that study concluded that ‘‘... until more is knownfication vaccinees and vaccinees inoculated id who developed a pox lesion: Impurities and animal proteins peculiar to calf about the importance of neutralizing antibody in the immunity to smallpox, it would seem unwise to administer vaccine bylymph exudate could explain in part the reactivity of the licensed vaccinia vaccine, or scarification could permit greater the sub-cutaneous route...’’ [21]. This conclusion was based on the relatively poor rates of seroconversion (67%), as mea-viral replication than would intradermal inoculation. Given comparable results in preclinical tests and the fact that both sured by vaccinia-neutralizing antibodies, in children vaccinated subcutaneously compared with their percutaneously vac-vaccines studied derived from New York Board of Health strains, it is unlikely that the cell culture–derived vaccinia virus cinated counterparts (87%), and the unimpressive neutralizing antibody responses upon successful percutaneous revaccinationhas enough sequence variation or intrinsic biologic disparities from the calf lymph–derived Wyeth vaccine to account for of children whose primary vaccination was subcutaneous. In addition to antibody responses, cell-mediated responsessuch differences. The vaccinees inoculated id who received alcohol wiping after inoculation and those inoculated im did were also superior in volunteers who developed pox lesions (whether inoculated id or by scarification). There was no sig-not significantly differ with respect to incidence of pox lesions, but the latter group had significantly fewer local symptoms nificant difference in lymphocyte stimulation responses between those who formed pox lesions from the licensed vacciniathan did those inoculated id, who in turn had fewer than the scarification vaccinees. vaccine by scarification and those with pox lesions from the cell culture–derived vaccinia vaccine. Inferior cell-mediatedThis study strongly indicated that, although less reactogenic, vaccinia vaccine administered im at a dose of 10 5 pfu fails to responses of vaccination without a pox lesion, as observed in this study, may explain why Connor et al. [21] reported thatinduce an immune response comparable to that elicited by standard scarification. Although a higher dose of vaccinia virus children who had initially receivedan injectable (subcutaneous) vaccine exhibited ú30% incidence of ‘‘primary-type’’ re-might have been attempted im, the WHO Smallpox Eradication Campaign [10] and other clinical trials [18–21] had supported sponses to percutaneous revaccination. This study also confirms the observation by Cherry et al. [20] of a correlation betweenthe safety of a parenteral dose of 10 5 pfu, as used in this trial. In a phase I dose-escalation human trial, the cell culture– neutralizing antibody and lymphocyte stimulation responsiveness to vaccinia after vaccination.derived vaccinia vaccine was used safely subcutaneously up / 9d21$$ap43 02-11-97 18:22:00 jinfa UC: J Infect Downloaded from https://academic.oup.com/jid/article/175/4/756/943818 by guest on 02 May 2024 762 McClain et al. JID 1997;175 (April) An unexpected event in this phase II study was the high rus vectors that are incapable of productive replication in human-derived cell lines [25, 26].incidence of secondary pox lesions adjacent to the inoculation site in the first set of volunteers (4/31). The use of semipermeWhen administration id of the cell-cultured vaccinia resulted in a cutaneous pox lesion, immune responses were not signifi-able dressings appeared to have caused this complication, as no further instances occurred after instituting the use of dry cantly different from those seen with scarification with the licensed vaccinia vaccine. Although there is a statistically insig-gauze dressings. Although vapor-permeable, the semipermeable dressings caused accumulation of perspiration around the nificant trend towards higher neutralization titers with the latter, future studies will examine the cell-cultured vaccine’s immuno-inoculation site despite frequent dressing changes, permitting virus-containing exudate to secondarily inoculate areas proxigenicity by this same scarification route. mate to the primary pox lesion. The large adhesive area of this dressing caused various degrees of contact dermatitis in several Acknowledgment volunteers, but this did not trigger autoinoculation, since the secondary lesions were under the dressing area and not at the We gratefully acknowledge the skillful assistance of Tamara point of adhesive contact. Ironically, the use of these semiperLewis in the completion of ELISAs. meable dressings was originally mandated as a protective measure against contact and environmental spread of the vaccine References virus. Apart from the four instances of autoinoculation precipitated by semipermeable dressings, there were no serious or 1. Piccini A, Paoletti E. Vaccinia: virus, vector, vaccine. Adv Vir Res 1988; 34:43–64. unexpected complications from the study; all inoculation site 2. Moss B. Vaccinia virus: a tool for research and development. Science lesions healed completely. 1991;252:1662–7. If it is necessary to achieve immunity to vaccinia or related 3. Tartaglia J, Pincus S, Paoletti E. Poxvirus-based vectors as vaccine candiOrthopoxviridae such as variola or monkeypox, these results dates. Crit Rev Immunol 1990;10:13–30. warrant vaccination by scarification in a population without 4. Rooney JF, Wohlenberg C, Cremer KJ, Moss B, Notkins AL. Immunization with a vaccinia virus recombinant expressing herpes simplex virus contraindications, despite the attendant risks. Formation of a type 1 glycoprotein D: long-term protection and effect of revaccination. cutaneous pox lesion (i.e., a vaccine ‘‘take’’), although less J Virol 1988;62:1530–4. desirable from a safety standpoint, engenders higher PRNT 5. Kundig TM, Kalberer CP, Hengartner H, Zinkernagel RM. Vaccination titers and cell-mediated responses and has been historically with two different vaccinia recombinant viruses: long-term inhibition validated as indicating protective immunity [10]. This obserof secondary vaccination. Vaccine 1993;11:1154–8. 6. Etlinger HM, Altenburger W. Overcoming of inhibition of antibody revation holds true when comparing administration id or im of sponses to a malaria recombinant vaccinia virus caused by prior expothe cell culture–derived vaccinia vaccine when such does not sure to wild type virus. Vaccine 1991;9:470–2. result in the development of a cutaneous pox lesion. These 7. Centers for Disease Control and Prevention. Vaccinia (smallpox) vaccine. data appear to reflect the epithelial tropism of OrthopoxviriRecommendations of the Immunization Practices Advisory Committee dae, which may explain many of the disappointing immune (ACIP). MMWR Morb Mortal Wkly Rep 1991;40:1–10 (erratum: MMWR Morb Mortal Wkly Rep 1992;41:31). responses in humans to vaccinia-vectored gene inserts [23, 8. Baxby D. Vaccinia Virus. In: Quinnan GV, ed. Vaccinia viruses as vectors 24]. In addition, it remains dubious whether the greater risk for vaccine antigens. New York: Elsevier, 1985:3–8. of high-dose (e.g., 10 7 pfu) inoculation im would be warranted 9. Henderson DA, Arita I. Utilization of vaccine in the global eradication of in an effort to achieve higher PRNT responses without a smallpox. In: Quinnan GV, ed. Vaccinia viruses as vectors for vaccine cutaneous pox lesion, despite evidence from both preclinical antigens. New York: Elsevier, 1985:61–7. 10. Fenner F, Henderson DA, Arita I, Jezek Z, Ladnyi ID. Smallpox and its and clinical studies that the cell culture–derived vaccinia vaceradication. Geneva: World Health Organization, 1988:289–94. cine is biologically comparable to other New York Board of 11. Lane JM, Ruben FL, Neff JM, Millar JD. Complications of smallpox Health vaccines. Although multiple injections of vaccinia via vaccination, 1968. N Engl J Med 1969;281:1201–8. a parenteral route might improve immune responses closer to 12. Chu YK, Rossi C, LeDuc JW, Lee HW, Schmaljohn CS, Dalrymple JM. those seen with scarification, repeated injections would inSerological relationships among viruses in the Hantavirus genus, family Bunyaviridae. Virology 1994;198:196–204. crease costs and logistic burdens and thereby negate the typi13. Earley E, Peralta PH, Johnson KM. A plaque neutralization method for cal advantages of a live vaccine, yet with greater safety conarboviruses. Proc Soc Exp Biol Med 1967;125:741–7. cerns. 14. Demkowicz WE, Ennis FA. Vaccinia virus–specific CD8 / cytotoxic T These results have important implications for the use of lymphocytes in humans. J Virol 1993;67:1538–44. recombinant poxvirus vaccines as vectors for immunogens and 15. Demkowicz W, Littaua R, Wang J, Ennis F. Human cytotoxic T-cell memory: long-lived responses to vaccinia virus. J Virol 1996;70: may explain why immune responses to some of these vaccines 627–31. have required multiple injections [4–6]. Although deliberate 16. Reed LJ, Muench H. A simple method for estimating fifty percent endinoculation by scarification may provide improved immune points. Am J Hyg 1934;27:493–7. responses for vaccinia virus–vectored immunogens, this is with 17. Stoline MR. The status of multiple comparisons: simultaneous estimation a greater risk of local reactions and possible secondary inoculaof all pairwise comparisons in one-way ANOVA designs. Am Stat 1981; 35:134–41. tion. This rationale would not apply to highly attenuated poxvi- / 9d21$$ap43 02-11-97 18:22:00 jinfa UC: J Infect Downloaded from https://academic.oup.com/jid/article/175/4/756/943818 by guest on 02 May 2024 763JID 1997;175 (April) Pox Lesion Required for Vaccinia Immunity 18. Hekker AC. Tissue culture smallpox vaccine. In: Quinnan GV, ed. Vac22. Rivers TM, Ward SM. Jennerian prophylaxis by means of intradermal injections of culture vaccine virus. J Exp Med 1935;62:549–60. cinia viruses as vectors for vaccine antigens. New York: Elsevier, 1985:23. Israel ZR, Edmonson PF, Maul DH, et al. Incomplete protection, but 113–6. suppression of virus burden, elicited by subunit simian immunodefi19. McIntosh K, Cherry JD, Benenson AS, et al. Standard percutaneous revacciency virus vaccines. J Virol 1994;68:1843–53. cination of children who received primary subcutaneous vaccination. J 24. Graham S, Green CP, Mason PD, Borysiewicz LK. Human cytotoxic T Infect Dis 1977;135:155–66. cell responses to vaccinia virus vaccination. J Gen Virol 1991;72: 20. Cherry JD, Rolfe UT, Dudley JP, Garakian AJ, Murphy M. Clinical and 1183–6. immunological study of percutaneous revaccination in children who 25. Tartaglia J, Perkus ME, Taylor J, et al. NYVAC: a highly attenuated strain originally received smallpox vaccine subcutaneously. J Clin Microbiol of vaccinia virus. Virology 1992;188:217–32. 1978;7:158–64. 26. Fleury B, Janvier G, Pialoux G, et al. Memory cytotoxic T lymphocyte 21. Connor JD, McIntosh K, Cherry JD, et al. Clinical and serologic study responses in human immunodeficiency virus type 1 (HIV-1)–negative of four smallpox vaccines comparing variations of dose and route of volunteers immunized with a recombinant gp160 canarypox expressing administration. Primary subcutaneous vaccination. J Infect Dis 1977;gp160 of HIV-1 and boosted with a recombinant gp160. J Infect Dis 135:167–75. 1996;174:734–8. / 9d21$$ap43 02-11-97 18:22:00 jinfa UC: J Infect Downloaded from https://academic.oup.com/jid/article/175/4/756/943818 by guest on 02 May 2024