Western blot analysis of antibody to varicella-zoster virus

Updated efficacy of adjuvant epirubicin plus cyclophosphamide followed by taxanes versus carboplatin plus taxanes in early triple-negative breast cancer in phase 2 trial: 8.1-year median follow-up

BACKGROUND

Paclitaxel/docetaxel after doxorubicin plus cyclophosphamide (ECT) is considered as an adjuvant chemotherapy and improves the survival of early triple-negative breast cancer (TNBC) patients. We aim to assess whether carboplatin plus taxanes (TP) is non-inferior to ECT in prolonging the survival time.

METHODS

TNBC patients were randomized (1:1) to receive ECT (90 mg/m² epirubicin + 600 mg/m² cyclophosphamide followed by 75 mg/m² docetaxel or 175 mg/m² paclitaxel every 3 weeks, n = 154) or TP (75 mg/m² docetaxel or 175 mg/m² paclitaxel + carboplatin AUC 5 every 3 weeks, n = 154). These expression of SPARC, PD-L1, and BRCA were studied. Patients were followed up for disease-free survival (DFS), overall survival (OS), and safety.

RESULTS

We recruited 308 TNBC patients (median follow-up of 97.6 months). The median DFS and OS were not reached; the 8-year DFS rate of ECT and TP arms was 78.4% and 81.7%, respectively, while the 8-year OS rate were 87.2% and 89.1%, respectively. In the SPARC (> 50%) subgroup analysis, the TP arm had longer DFS (P = 0.049) and a tendency with better OS (P = 0.06) than ECT arm. No significant differences were observed in the DFS and OS between the ECT arm and TP arm in TNBC with SPARC (≤ 50%), PD-L1 (-) PD-L1 (+), and BRCA mutation or BRCA wild (all P values > 0.05).

CONCLUSION

TP showed non-inferiority for DFS and OS compared with ECT in early TNBC. TP may be an effective alternative chemotherapy for TNBC patients whom the standard ECT regimen is not being used.

TRAIL REGISTRATION

ClinicalTrials.gov identifier NCT01150513.

Molecular mimicry between varicella, measles virus and Hsp60 in type 1 diabetes associated HLA-DR3/DR4 molecules

BACKGROUND AND AIMS

Type 1 diabetes (T1D) is a multifactorial autoimmune disease that combines genetics and environmental factors. The aim of this study is to determine the environmental risk factors and to investigate how virals infections are risks factors for type 1 diabetics whom have HLA DR3/DR4 predisposition in our population.

METHODS

This study includes 233 subjects, 145 diabetics and 88 controls from regions of the extreme western of Algeria. All the informations related to the disease were collected using predesigned questionnaire. Using in silico approach, we attempt to improve the understanding of this analytical result by molecular mimicry, which is associated with the breakdown of several autoimmune pathologies.

RESULTS

The statistical study showed that history of varicella and measles infection and T1D related inheritance and type 2 diabetes are risk factors for T1D in the population of Tlemcen. We have determined the homologous antigenic regions between the glycoprotein "gE" of the varicella virus, the "hemagglutinin" of measles and the human protein "HSP60" at the level of their sequence and 3D structure. These cross-reactive epitopes bind to MHC class II molecules (HLA DR3/DR4) that predispose to T1D but not to MHC class II molecules (HLA DR2) that protect against T1D. This epitopes induce Th2 cells but only "hemagglutinin" and "Hsp60" can activate Th1 differentiation. This indicates their potential to destroy pancreatic cells β.

CONCLUSION

Our study can allow us to adapt biological markers to genetically predisposed T1D and to establish a preventive strategy for healthy genetic predisposed individuals in Tlemcen population.

Immune Responses to a Recombinant Glycoprotein E Herpes Zoster Vaccine in Adults Aged 50 Years or Older

Background

The herpes zoster subunit vaccine (HZ/su), consisting of varicella-zoster virus glycoprotein E (gE) and AS01_B Adjuvant System, was highly efficacious in preventing herpes zoster in the ZOE-50 and ZOE-70 trials. We present immunogenicity results from those trials.

Methods

Participants (ZOE-50: ≥50; ZOE-70: ≥70 years of age) received 2 doses of HZ/su or placebo, 2 months apart. Serum anti-gE antibodies and CD4 T cells expressing ≥2 of 4 activation markers assessed (CD4²⁺) after stimulation with gE-peptides were measured in subcohorts for humoral (n = 3293) and cell-mediated (n = 466) immunogenicity.

Results

After vaccination, 97.8% of HZ/su and 2.0% of placebo recipients showed a humoral response. Geometric mean anti-gE antibody concentrations increased 39.1-fold and 8.3-fold over baseline in HZ/su recipients at 1 and 36 months post-dose 2, respectively. A gE-specific CD4²⁺ T-cell response was shown in 93.3% of HZ/su and 0% of placebo recipients. Median CD4²⁺ T-cell frequencies increased 24.6-fold (1 month) and 7.9-fold (36 months) over baseline in HZ/su recipients and remained ≥5.6-fold above baseline in all age groups at 36 months. The proportion of CD4 T cells expressing all 4 activation markers increased over time in all age groups.

Conclusions

Most HZ/su recipients developed robust immune responses persisting for 3 years following vaccination.

Clinical Trials Registration

NCT01165177; NCT01165229.

A critical appraisal of 'Shingrix', a novel herpes zoster subunit vaccine (HZ/Su or GSK1437173A) for varicella zoster virus

HZ/Su, branded as 'Shingrix', is one of the newest vaccines to be submitted for multi-national regulatory approval. It is targeted to prevent shingles, a global concern with aging populations. A live attenuated vaccine for shingles has been available for over a decade, however it is contraindicated in specific subgroups of people, and there are added concerns regarding long-term immunogenicity. HZ/Su is the first subunit vaccine developed to protect against shingles. This paper provides a critical appraisal of current evidence regarding HZ/Su.

Microbiology laboratory and the management of mother-child varicella-zoster virus infection

Varicella-zoster virus, which is responsible for varicella (chickenpox) and herpes zoster (shingles), is ubiquitous and causes an acute infection among children, especially those aged less than six years. As 90% of adults have had varicella in childhood, it is unusual to encounter an infected pregnant woman but, if the disease does appear, it can lead to complications for both the mother and fetus or newborn. The major maternal complications include pneumonia, which can lead to death if not treated. If the virus passes to the fetus, congenital varicella syndrome, neonatal varicella (particularly serious if maternal rash appears in the days immediately before or after childbirth) or herpes zoster in the early years of life may occur depending on the time of infection. A Microbiology laboratory can help in the diagnosis and management of mother-child infection at four main times: (1) when a pregnant woman has been exposed to varicella or herpes zoster, a prompt search for specific antibodies can determine whether she is susceptible to, or protected against infection; (2) when a pregnant woman develops clinical symptoms consistent with varicella, the diagnosis is usually clinical, but a laboratory can be crucial if the symptoms are doubtful or otherwise unclear (atypical patterns in immunocompromised subjects, patients with post-vaccination varicella, or subjects who have received immunoglobulins), or if there is a need for a differential diagnosis between varicella and other types of dermatoses with vesicle formation; (3) when a prenatal diagnosis of uterine infection is required in order to detect cases of congenital varicella syndrome after the onset of varicella in the mother; and (4) when the baby is born and it is necessary to confirm a diagnosis of varicella (and its complications), make a differential diagnosis between varicella and other diseases with similar symptoms, or confirm a causal relationship between maternal varicella and malformations in a newborn.

The host immune response to varicella zoster virus

Varicella zoster virus (VZV) is a highly successful human pathogen, which is never completely eliminated from the host. VZV causes two clinically distinct diseases, varicella (chickenpox) during primary infection and herpes zoster (shingles) following virus reactivation from latency. Throughout its lifecycle the virus encounters the innate and adaptive immune response, and in order to prevent eradication it has developed many mechanisms to evade and overcome these responses. This review will provide a comprehensive overview of the host immune response to VZV infection, during the multiple stages of the virus lifecycle and at key sites of VZV infection. We will also briefly describe some of the strategies employed by the virus to overcome the host immune response and the ongoing challenges in further elucidating the interplay between VZV and the host immune response in an attempt to lead to better therapies and a ‘second generation’ vaccine for VZV disease.

Human anti-varicella-zoster virus (VZV) recombinant monoclonal antibody produced after Zostavax immunization recognizes the gH/gL complex and neutralizes VZV infection

Varicella-zoster virus (VZV) is a ubiquitous, highly cell-associated, and exclusively human neurotropic alphaherpesvirus. VZV infection is initiated by membrane fusion, an event dependent in part on VZV glycoproteins gH and gL. Consistent with its location on the virus envelope, the gH/gL complex is a target of neutralizing antibodies produced after virus infection. One week after immunizing a 59-year-old VZV-seropositive man with Zostavax, we sorted his circulating blood plasma blasts and amplified expressed immunoglobulin variable domain sequences by single-cell PCR. Sequence analysis identified two plasma blast clones, one of which was used to construct a recombinant monoclonal antibody (rec-RC IgG). The rec-RC IgG colocalized with VZV gE on the membranes of VZV-infected cells and neutralized VZV infection in tissue culture. Mass spectrometric analysis of proteins immunoprecipitated by rec-RC IgG identified both VZV gH and gL. Transfection experiments showed that rec-RC IgG recognized a VZV gH/gL protein complex but not individual gH or gL proteins. Overall, our recombinant monoclonal anti-VZV antibody effectively neutralizes VZV and recognizes a conformational epitope within the VZV gH/L protein complex. An unlimited supply of this antibody provides the opportunity to analyze membrane fusion events that follow virus attachment and to identify multiple epitopes on VZV-specific proteins.

Immune response to vaccinia virus recombinants expressing glycoproteins gE, gB, gH, and gL of Varicella-zoster virus

Immunogenicity of Varicella-zoster virus glycoproteins gE, gB, gH, and gL expressed by recombinant vaccinia viruses (VV) separately or simultaneously was determined in mice and guinea pigs by ELISA, Western blotting, radioimmunoprecipitation, plaque reduction assay, and skin test. Single VV-gE and VV-gB recombinants and double VV-gH/gL recombinant elicited specific antibodies with VZV neutralizing activity in mice. Co-expression of gE and gB by one recombinant VV resulted in an increased antibody response in comparison with immunization with single recombinants or their mixtures. Unlike anti-gB and anti-gH/gL antibodies, the gE-specific antibodies had no virus neutralizing activity in absence of complement, and when used alone, they even caused considerable increase of VZV infectious units. Moreover, immune sera containing anti-gE antibodies antagonized complement independent virus-neutralizing activity of anti-gB- and anti-gH/gL-positive sera. The ability to induce delayed hypersensitivity reaction to VZV antigens was observed after immunization of guinea pigs with gE- and/or gB-expressing VVs.

Neutralizing antibody responses induced by varicella-zoster virus gE and gB glycoproteins following infection, reactivation or immunization

The purpose of this study was to compare the antibody responses to varicella-zoster virus (VZV) gE and gB after natural VZV infection and after vaccination with live attenuated OKA vaccine in order to determine the relative importance of these proteins as components of a subunit vaccine. Anti-VZV antibody titers determined by IFA were of the same order of magnitude in sera from individuals with a history of varicella and in vaccinated children but higher in individuals given booster vaccination. The titers of anti-gE and anti-gB antibodies were measured by ELISA using recombinant gE or gB as capture antigen. From these experiments, it appears that the ratio of anti-gE to anti-gB antibody is highly variable from one individual to another but relatively stable over a long period of time for a particular individual, even after a zoster episode. Neutralizing antibodies directed against gE or gB were also measured by subtracting the neutralization titers obtained before and after depletion of the specific antibodies on immobilized recombinant gE, gB, or both. This showed that, with respect to neutralization, anti-gE and anti-gB are equally prevalent in vaccinated children and that anti-gE is generally, but not always, predominant over anti-gB in VZV-infected individuals. Finally, antibodies to these two glycoproteins appear to be predominant among the neutralizing antibodies directed to other VZV antigens.

Assays for antibodies to varicella-zoster virus

Anti-varicella-zoster virus serum antibody assays and their use in vaccine development are described. Of particular interest are FAMA and neutralization assays and the gpELISA. These and other assays are compared and summarized in terms of characteristics including biologic relevance, sensitivity, specificity, and suitability for different laboratory and clinical applications.

Immune responses to varicella-zoster virus

The host response to VZV is critical to the outcome of primary VZV infection. The maintenance of immune memory to the virus is required to prevent symptomatic re-infection on exogenous re-exposure to VZV and to prevent symptomatic reactivation of endogenous virus. Immunization with live varicella (Oka) vaccine elicits primary and memory immunity to VZV. Humoral and cell-mediated host responses induced by the wild-type virus and by the vaccine strain are comparable, which is consistent with the clinical observation that varicella vaccine protects against or significantly reduces the clinical symptoms caused by primary VZV infection. Widespread use of the varicella vaccine in healthy children will yield further knowledge about host-virus interactions, such as the role of exogenous re-exposure in maintaining persistent immunity, which will be relevant to vaccine strategies to prevent other human herpesvirus infections.

Purification, characterization and immunogenicity of recombinant varicella-zoster virus glycoprotein gE secreted by Chinese hamster ovary cells

The gene of the varicella-zoster virus (VZV) glycoprotein gE, engineered to code for a truncated molecule lacking the anchor and carboxy-terminal tail domains, was transfected into Chinese hamster ovary (CHO) cells via the pEE14 mammalian expression vector. One recombinant cell line, CHO-gE-2-9, secreted high levels of truncated gE into the culture medium. The product was purified to near homogeneity by a combination of anion exchange, hydrophobic and metal-chelate chromatographies. Purified recombinant gE showed the expected amino-terminal sequence and its glycosylation pattern proved similar to that of the natural product. When injected into mice, using either Freund's or alum as adjuvant, the native truncated gE induced complement-dependent neutralizing antibodies. In contrast, when the molecule was first denatured, it lost immunogenicity with alum. These data show that the recombinant gE, although truncated, could potentially be included in a subunit vaccine against VZV infection.

Combined use of complement and anti-immunoglobulin in an enhanced neutralization assay for antibodies to varicella-zoster virus

An enhanced neutralization assay was developed to permit the sensitive, specific, and reproducible measurement of antibodies to varicella-zoster virus (VZV). Optimal neutralization was achieved using a combination of guinea pig complement (C') and rabbit anti-human IgG. This provided 625-, 160- and 13- to 64-fold increases in dilution endpoints of human post-zoster serum, varicella-zoster immune globulin and representative sera from recipients of live attenuated varicella vaccine, respectively, above those measured in the absence of C' and anti-IgG. The specificity of the assay was shown by the absorption of serum neutralization capacity with VZV-specific antigen and the lack of concordance between antibody titers to VZV with those to either herpes simplex virus type-2 or cytomegalovirus. The antibody status of recipients of live attenuated varicella vaccine was established from the amount of neutralizing activity produced at a single optimal serum dilution.

Varicella-zoster virus antibody avidity and IgG-subclass patterns in children with recurrent chickenpox

The normal immune response after primary varicella-zoster virus (VZV) infection includes IgG subclass evolution to predominantly IgG1-type antibodies, and maturation from low to high avidity antibodies which are maintained for life. Twenty-three healthy and apparently immunocompetent children with a history of 2-5 episodes of chickenpox were studied after repeat disease. Serial sera were tested for VZV-IgG subclass patterns and VZV IgG and G-subclass antibody avidity by urea elution enzyme linked immunoassay (ELISA). Of 11 patients studied within 8 weeks of repeat chickenpox (Early Group), mean antibody avidity was significantly lower (31.3 +/- 26.81) than control (65.1 +/- 12.38) (P < .001). Seven had low avidity antibody (< 30 percent) and an abundance of IgG3 which was a pattern like primary chickenpox, and 2/11 had high avidity antibody characteristic of anamnestic responses. Early Group patients and 12 others studied over 8 weeks after repeat disease (Late Group) showed avidity maturation and attrition of IgG subclass antibodies other than IgG1. At least nine children failed to show VZV-specific secondary (memory) immune responses early in the course of repeat disease. It is possible that failure to maintain or evoke a secondary immune response could explain their susceptibility to repeat chickenpox.

Low antibody avidity in elderly chickenpox patients

A small outbreak of chickenpox confirmed serologically in 3 elderly patients from a geriatric home is described. Disease was probably due to exogenous reinfection, yet nevertheless the avidity of specific antibodies measured by the urea denaturation test was even lower than in primary chickenpox controls, which themselves were, as expected, significantly lower than zoster controls. In elderly individuals susceptibility to reinfection with varicella-zoster virus (VZV) with clinical manifestation such as chickenpox may well be associated with the decay of specific humoral immunity detectable by antibodies of particularly low avidity, in contrast to reactivation of latent VZV presenting clinically as zoster, which is related to deficiencies in specific cellular immunity.

Acyclovir treatment for varicella does not lower gpI and IE-62 (p170) antibody responses to varicella-zoster virus in normal children

The varicella-zoster virus (VZV) membrane glycoprotein gpI elicits a major immunoglobulin G antibody response after naturally acquired VZV infection; antibody to a nonglycosylated immediate-early protein, IE-62 (p170), represents a response to a nonmembrane VZV component. We evaluated antibody response to VZV gpI and IE-62 (p170) at 28 days and 1 year following infection in 34 children (ages 5 to 16 years) enrolled in a randomized placebo-controlled study of oral acyclovir for the treatment of varicella. All children were VZV antibody negative at enrollment, were previously healthy, and had laboratory-documented varicella. Compared with placebo recipients, acyclovir recipients had lower geometric mean titers by the fluorescent antibody to membrane antigen technique at 28 days (620 versus 836) but similar titers at 1 year (122 versus 122). All children had antibodies to gpI and IE-62 detectable by enzyme-linked immunosorbent assay at 28 days and 1 year. No difference in gpI at 28 days compared with 1 year was noted in acyclovir recipients. No difference in antibody to IE-62 (p170) was noted when acyclovir and placebo recipients were compared at either 28 days or 1 year. Antibody responses to gpI and IE were similar when children were stratified by age (5 to 6 years, 7 to 11 years, 12 to 16 years). A short course of oral acyclovir for the treatment of varicella did not affect antibody responses to gpI or IE-62 (p170) in healthy children at 28 days and 1 year following varicella.

The glycoprotein products of varicella-zoster virus gene 14 and their defective accumulation in a vaccine strain (Oka)

Many characteristics of the putative protein encoded by varicella-zoster virus (VZV) open reading fram (ORF) 14 indicate that it is a glycoprotein, which has been designated gpV. To identify the protein products of the gene, the coding sequences were placed under the control of the vaccinia virus p7.5 promoter and recombinant vaccinia viruses were constructed. Heterogeneous polypeptides with molecular weights of 95,000 to 105,000 (95K to 105K polypeptides) were expressed in cells infected by a vaccinia virus recombinant (vKIP5) containing ORF 14 from VZV Scott but were not expressed by control vaccinia viruses. These polypeptides were recognized by antibodies present in human sera that contained high levels of anti-VZV antibodies. Conversely, antisera raised in rabbits inoculated with vKIP5 reacted specifically with heterogeneous 95K to 105K polypeptides present in VZV Scott-infected but not uninfected cells; these polypeptides show a patchy plasma membrane fluorescence pattern in VZV Scott-infected cells. These same antisera neutralized VZV strain Scott infectivity in the absence of complement. Endoglycosidase F treatment of isolated gpV polypeptides and tunicamycin treatment of cells infected with the vKIP5 recombinant indicated that the polypeptides were glycosylated. Three sets of data imply that the VZV strain Oka, which has been used to produce a live attenuated virus vaccine, accumulates low levels of gpV polypeptides relative to wild-type strains: (i) blocking of antibodies in human sera with excess VZV Oka-infected cell antigen yielded residual antibodies which were reactive with the 95K to 105K gpV polypeptides expressed in cells infected by VZV strain Scott and by the vKIP5 vaccinia virus recombinant, but not with Oka-infected cell polypeptides; (ii) antisera raised to vKIP5 detected very low levels of reactive polypeptides made in VZV Oka-infected cells and neutralized VZV Oka virus much less efficiently than VZV Scott; and (iii) comparisons of the reactivity of sera from live attenuated virus vaccine vaccinees with sera derived from patients recovering from wild-type infections indicated greatly reduced levels of gpV-specific antibodies in some vaccinees.

HSV-1 gB and VZV gp-II crossreactive antibodies in human sera

The specificity and prevalence of human IgG antibodies crossreactive between HSV-1 (ANG) and VZV (Ellen) was examined in immunoblots. Using antibody fractions purified on HSV- and VZV-coated affinity chromatography columns and by preadsorption of sera with HSV and/or VZV lysates a crossreactivity between HSV-1 gB and VZV gp-II was demonstrated. Crossreaction of human IgG antibodies among other structural and nonstructural viral proteins, however, was not detected. The frequency of human IgG antibodies crossreactive between HSV-1 gB and VZV gp-II was highest in HSV-seropositive patients experiencing an acute primary VZV infection (4 out of 5 sera tested). In contrast, no crossreactive antibodies were found in sera of HSV-seronegative patients with acute primary VZV infection (0/6) or in sera from individuals with acute recurrent HSV or VZV infection (0/12). Analysis of sera from individuals with previous HSV and/or VZV infection showed the presence of antibodies crossreactive between HSV-1 gB and VZV gp-II in 3 out of 30 sera tested.

Viral vaccines of the future

Varicella vaccine seems close to licensure in the United States. It is likely to be recommended for routine use in healthy children, either administered singly or as a combination of MMR. Healthy children who have been immunized develop excellent antibody and cell-mediated responses to VZV in the absence of significant adverse effects, and they are well protected against subsequent infection with VZV. Although it will not be known for certain for many years, it seems most unlikely that immunization will result in an increased incidence of zoster, a secondary type of infection with VZV that is caused by reactivation of latent VZV. Varicella vaccine may also be given to healthy adults who have never had chickenpox with a great degree of success, although its protection is somewhat less effective than in children. Leukemic children who are at high risk for developing severe or fatal varicella also derive a great deal of protection from varicella vaccine, but the vaccine must be administered to them with great caution. Other viral vaccines that may be licensed in the future but that are not as fully developed as varicella vaccine include vaccines against CMV, hepatitis A, HSV, and AIDS. These are in various degrees of study, and, should any of them be licensed, it is uncertain whether they would be live attenuated, recombinant, or subunit vaccines. It is hoped, however, that they will eventually be licensed for future use because if effective, they could significantly decrease morbidity and mortality of infants, children, and adults.

Antibody responses in recipients of varicella vaccine assayed by immunoblotting

Sera taken from 35 children with cancer who had been vaccinated with live varicella vaccine were assayed using immunoblotting for the presence of IgG class antibodies to proteins present in varicella-zoster virus (VZV)-infected cells. Sera from 23 of these patients were also assayed for IgM class antibodies. The patterns of immunoreactivity observed for these patients following vaccination were substantially weaker and more variable than those detected following natural VZV infection in otherwise healthy individuals. The IgG responses detected following vaccination involved up to 10 protein bands between 28 and 188 kDa. Bands were particularly frequent in the 78-96 kDa region. IgM responses involved up to 10 bands between 28 and 114 kDa, with the bands in the 78-96 kDa region and at 32-36 kDa being detected most frequently. Repeated vaccination generally produced a stronger IgG antibody response than did single vaccination, and subsequent exposure of vaccinees to natural VZV infection resulted in an increased level of reactivity for IgG antibodies, but not for IgM. Similar reaction patterns were obtained with sera from vaccinees when the vaccine virus and wild-type VZV were used as antigens. Immunoblotting showed good correlation with indirect radioimmunoassay for the detection of a vaccine-induced IgG response.

Detection of anti rotavirus coproantibodies by immunoblotting technique

IgA and IgG coproantibodies to individual simian rotavirus (SA 11) structural polypeptides were detected in healthy infants in nursery homes. The number of immunoblottable peptides differed from individual to individual. Coproantibodies were also detected at the convalescent stage of rotavirus infection in two patients but not during the acute stage. This method is useful for confirming the diagnosis of rotavirus infection serologically without the need for paired sera.

Persistence of immunity to varicella in children with leukemia immunized with live attenuated varicella vaccine

To determine the safety and efficacy of varicella vaccine, we studied 437 children in remission from leukemia who were immunized with live attenuated varicella virus. Three hundred one of the patients received two doses of vaccine and 136 received a single dose of vaccine from 1 of 10 lots from two manufacturers. The patients have been followed for an average of three years (range, one to six). Seroconversion occurred in 88 percent of the 437 children after the first dose of vaccine and in 98 percent after one or two doses. The proportions of patients who were seronegative after one, three, and five years were 20, 25, and 30 percent, respectively, with little change over time in the geometric mean titers of specific antibody (6.3, 6.5, and 5.7, respectively). Chickenpox has been documented in 36 vaccinated patients (8 percent) who had 3 to 640 vesicles (mean, 100), mild illness, and no complications. Of the 83 vaccinated patients exposed to varicella within their families, 11 had chickenpox; the attack rate was 14 percent (8 percent among seropositive patients, 29 percent among seronegative patients). There was no relation between the time since vaccination and either the attack rate or the severity of the breakthrough illness. Two doses of vaccine appeared to be no more effective than a single dose. Of the 372 patients receiving maintenance chemotherapy when immunized, 149 (40 percent) had a rash, which was treated with acyclovir in 16 children (4 percent) and became a severe febrile illness in 4. These reactions were not fatal and were all associated with vaccine lots, the use of which has since been discontinued. We conclude that in children in remission from leukemia, varicella vaccine is safe and induces an immunity to chickenpox that persists for more than three years.