Varicella-zoster-virus vaccination in immunosuppressed children with rheumatic diseases using a pre-vaccination check list

Background The goal of this study was to apply the varicella zoster virus (VZV) vaccine to patients with pediatric rheumatic diseases (PRD) at risk for severe chickenpox, without interrupting their current immunosuppression, including biological agents, using an immunological-based pre-vaccination checklist to assure safety. A pre-vaccination checklist was implemented to ensure adequate immune competence prior to immunization. Methods This prospective study included seronegative patients (VZV-IgG ≤200 mIU/ml) and patients who had previously received only a single dose of VZV vaccine. All vaccinees demonstrated clinically inactive PRD. Patients were categorized according to their actual treatment in low-intensity IS (LIIS) and high-intensity IS (HIIS) including biological therapy. The pre-vaccination checklist defined thresholds for the following basic laboratory tests: white blood cell count ≥3000/mm3, lymphocytes ≥1200/mm3, serum IgG ≥500 mg/dl, IgM ≥20 mg/dl, tetanus toxoid antibody ≥0.1 IU/ml. In case of HIIS additional specifications included a CD4+ lymphocyte count ≥200/mm3 and a positive T-cell function (via analyzable positive control of a standard tuberculosis interferon-gamma-release-assay (TB-IGRA) indicating mitogen-induced T cell proliferation). Patients who met the criteria of the pre-vaccination checklist received the first and/or second VZV vaccination. Immunologic response and side effects were monitored. Results Twenty-three patients were recruited of whom nine had already received one VZV immunization before initiating IS. All patients met the pre-vaccination checklist criteria despite ongoing IS. There was no overall difference in VZV-IgG levels when comparing the LIIS (n=9) and HIIS (n=14) groups. In total, 21 patients (91%) showed a positive vaccination response, after the first immunization the median VZV-IgG across all patients was 224 (59-1219) mIU/ml (median (range)), after booster immunization it increased to 882 (30-4685) mIU/ml. Two patients in the HIIS group failed to raise positive VZV-IgG, despite booster immunization. All nine patients receiving only the second immunization on IS reached high titers of VZV-IgG >500 mIU/ml (1117 (513-4685) mIU/ml). There were no cases of rash or other vaccine-induced varicella disease symptoms and no evidence of PRD flare. Conclusions VZV vaccination is safe and largely immunogenic in children with ongoing IS fulfilling an immunological based pre-vaccination checklist. This new approach is based on immunologic function rather than on type of medications. Trial registration number ISRCRTN trial registration number 21654693, date of registration February 12, 2018, retrospectively registered. Electronic supplementary material The online version of this article (10.1186/s12969-018-0231-3) contains supplementary material, which is available to authorized users.


(Continued from previous page)
Conclusions: VZV vaccination is safe and largely immunogenic in children with ongoing IS fulfilling an immunological based pre-vaccination checklist. This new approach is based on immunologic function rather than on type of medications.
Trial registration number: ISRCRTN trial registration number 21654693, date of registration February 12, 2018, retrospectively registered.
Keywords: Biologic agents, DMARDs, Immunologic tests, Juvenile chronic arthritis, Pediatric rheumatic diseases, Varicella zoster virus, Vaccination Background Varicella-zoster virus (VZV) the etiologic agent of chickenpox and herpes zoster [HZ], is highly contagious and still endemic worldwide. Immunosuppressed individuals are at substantial risk for severe disease courses [1]. Therefore, the primary motivation for the development of the live attenuated VZV vaccine in the 1970s was to use it as a targeted vaccine to protect vulnerable children during cancer chemotherapy [2,3]. The safety of administering the VZV vaccine has been studied extensively in immunosuppressed children with malignant and non-malignant disease, e.g. following kidney, liver or intestinal transplantation, nephrotic syndrome or inflammatory bowel disease, demonstrating marked safety and good efficacy [2][3][4][5][6][7][8][9][10]. However, in cases of extreme T cell deficiency with CD4+ T cell count <100/mm 3 in the context of severe combined immunodeficiency and the acquired immunodeficiency syndrome, severe chickenpox induced by the VZV vaccine strain was observed [11,12]. In trials of VZV vaccination in immunosuppressed children, immunological testing was heterogeneous and, for example, included ruling out severe lymphopenia, and the demonstration of preserved immunoreactivity by measurement of total IgG level, antibody titers to inactivated vaccines, CD4+ T cell counts and intracutaneous or in vitro T-cell function tests [2,6,7,9,10,13]. Smaller trials focused on children with pediatric rheumatic diseases (PRD) such as systemic lupus erythematosus and juvenile idiopathic arthritis (JIA) [14][15][16]. Children within these trials were treated with glucocorticoids (prednisone up to 0.7 mg/kg body weight), methotrexate, leflunomide, azathioprine, 6mercaptopurine, cyclosporine, tacrolimus, and, in individual cases, biologics. Again, there were no relevant adverse effects and efficacy, as measured by the prevention of breakthrough VZV-associated disease and an increase in VZV-IgG level, was good. However, the VZV-IgG response appeared slightly diminished when compared to healthy controls in one study [15].
The current recommendations for vaccination of children with rheumatic disease by the European League Against Rheumatism (EULAR) state not to administer live-virus vaccines to patients on therapy with high-dose diseasemodifying antirheumatic drugs (DMARD), high-dose glucocorticoids or biological agents except on a case-to case basis [17]. However, guidance on how to decide on a case-to-case basis is not offered.
The goal of this study was to apply the VZV vaccine to patients with PRD at risk for severe chickenpox, without interrupting their current IS, including biological agents, using an immunological-based pre-vaccination checklist to assure safety. The checklist contained only easy-toobtain clinical and immunologic parameters and could be used for the whole spectrum of currently used/ standard IS. This new approach is based on immunologic function rather than on type of medications. Patients demonstrating immunoreactivity despite IS received live vaccine and safety and immunogenicity data were obtained.

Participants and stratification method
A prospective single-center study was conducted at the German Center for Pediatric and Adolescent Rheumatology, Garmisch-Partenkirchen (ISRCRTN: 21654693, retrospectively registered). The study was approved by the Institutional Board of Ethics in Medical Research of the Bavarian Chamber of Physicians. Written informed consent was obtained from all parents and the patients. Within the inclusion period from April 2012 to Mai 2013, 2802 patients between the ages of 2 to 17 years were screened by reviewing their immunization records, a questionnaire asking about prior chickenpox or herpes zoster and, in case of absence of chickenpox, herpes zoster and vaccination, by VZV-IgG screening. Inclusion criteria for the intervention part of the study were: (1) negative medical history for chickenpox and herpes zoster, (2) ≤ 1 prior dose of the VZV vaccine, (3) in case of first VZV vaccination, laboratory evidence of susceptibility for chickenpox defined as VZV-IgG either classified as negative (<160 mIU/ml) or equivocal (160-200 mIU/ml) (test system: VZV-IgG-ELISA (medac GmbH, Wedel) calibrated based on WHO reference preparation allowing quantitative measurements), (4) diagnosis of an inflammatory PRD, (5) clinically inactive disease as defined by the American College of Rheumatology (ACR) criteria in case of JIA [18] or a physician's global score of <1 in case of other PRDs, and (6) no change of IS for at least 3 months prior to the vaccination. Exclusion criteria were as follows: (1) acute febrile disease, (2) current clinical or laboratory evidence for lack of immunologic reactivity (see pre-vaccination checklist below), (3) known hypersensitivity to constituents of the varicella vaccine, (4) measles, mumps, rubella (MMR) vaccination within 4 weeks prior to VZV vaccination, (5) treatment with IS other than those mentioned in the pre-vaccination checklist, i.v. glucocorticoid pulse therapy or a prednisoneequivalent dose of ≥2mg/kg/day or ≥20mg/day for > 2 weeks within less than 4 weeks prior to vaccination, cyclophosphamide pulse <6 months ago, rituximab without B-cell reconstitution, intravenous immune globulins (IVIG) <6 months ago (high-dose IVIG (2g/kg) <11 months), therapy with aspirin until 6 week post-vaccination or vi) any blood products <3 months prior to vaccination.

Pre-vaccination checklist
A checklist was developed, consisting of items regarding medical history, physical examination, contraindications, grading the level of IS and basic laboratory examinations for all patients (Table 1). Basic laboratory testing included white blood cells (WBC) and lymphocyte count, serum IgG and IgM levels and tetanus toxoid antibody level. Cut-off levels were WBC ≥3000/mm 3 , lymphocytes ≥1200/mm 3 , serum IgG ≥500 mg/dl, IgM ≥20 mg/dl, tetanus toxoid antibody ≥0.1 IU/ml [20]. If tetanus antibody-level was <0.1 IU/ml, tetanus (booster) vaccination was administered and antibody testing repeated 4 weeks later. Additional (extended) laboratory examinations were pursued for patients receiving HIIS or patients on LIIS with abnormal basic laboratory test results and included the following parameters and cut-off levels: normal CD4+ T cell count (cut-off ≥200/ mm 3 if age >5 years or ≥500/mm 3 if age 2-5 years) and a positive T cell function test (CMI). As an easy-to-obtain T cell function test we used a commercial interferon-gamma release assay (IGRA), the TB-EliSpot® test, which contains a mitogen as positive control (phytohemagglutinin). Patients exceeding the respective cut-off levels had adequate immunologic reactivity for the purpose of this study and VZV vaccination was offered.

Vaccination
Patients stratified to fullfill the pre-vaccination criteria, also met the immunological precautions requested by the manufacturer of Varilrix® (Glaxo-Smith-Kline) for the in-label application of the VZV vaccine on IS (absolute lymphocyte count >1200/mm 3 and/or adequate CMI). Varilrix® contains the live attenuated Oka strain at a concentration of at least 10 3.3 plaque-forming units per 0.5ml. Participants received a standard dose of the vaccine 0.5ml subcutaneously. A second dose was given at an interval of at least 6 weeks on LIIS and 3 months on HIIS [19]. Patients, parents and the primary care pediatrician received a written instruction to initiate treatment with acyclovir in case of VZV disease with >50 skin lesions or a rash lasting >7 days and to contact their pediatric rheumatologist to decide on A reduction of the IS.

Determination of safety
Patients/parents and pediatricians completed a separate questionnaire between 4-12 weeks following vaccination asking for vaccination-associated side effects or signs for flare of the PRD. If any sign of side effect was present, further details were obtained by telephone interview.

Determination of immunogenicity and efficacy
VZV-IgG response was measured using a blood sample drawn between 4 to 12 weeks after vaccination. Furthermore, patients were interviewed after a longer interval (median 3 years) to determine if post-vaccination contact to VZV and/or breakthrough chickenpox or HZ had occurred. Table 1 Checklist regarding clinical and immunological requirements prior to varicella zoster virus vaccination Step 1: Rating of immunosuppressive therapy Step 2: Prerequisites Step 3: Testing protocol Step 4: Decision about vaccination Patientsfullfilling these pre-vaccination criteria, also meet the immunological precautions requested by the manufacturer of Varilrix (Glaxo-Smith-Kline) for the in-label application.
c Especially in case of breakthrough or vaccine-induced VZV disease with >50 skin lesions or a rash lasting >7 days: strongly consider treatment with acyclovir and contact pediatric rheumatologist for a possible reduction of immunosuppressive therapy

Statistical analysis
Statistical analysis was performed using GraphPad Prism 6 (GraphPad Software, La Jolla, CA, USA) and Microsoft Excel 2010 (Microsoft, Redmond, WA, USA). Demographic and baseline laboratory characteristics were described using medians and range. Differences between patients receiving LIIS and those receiving HIIS were estimated with the chi-squared test for proportions and with the Mann-Whitney test for interval variables.

Results
Screening for lack of varicella immunity and recruitment indicating prior contact to VZV. Additionally, we observed titers classified as negative (<160 mIU/ml) in 4 out of 9 patients that had received a single VZV vaccination prior to the study (median 230 mIU/ml, range 57-1003 mIU/ml). Twenty-three patients entered the interventional part of this study (Fig. 1).

Participants and baseline laboratory and immunologic data
Demographic data, disease type, current treatments and baseline laboratory test results of the patients vaccinated in the study are shown in Table 2. One patient had a low tetanus toxoid antibody level and received a successful tetanus booster-vaccination. All other parameters were above the predefined thresholds in all patients.

Administration of the varicella vaccine
Nine patients who had previously received one dose of the VZV vaccine received a second dose in the context of this study (2 in the LIIS group, 7 in the HIIS group). Of the 15 patients naïve for VZV, six only received one vaccination (3 in the LIIS group, 3 in the HIIS group) while nine received two (4 in the LIIS group, 5 in the HIIS group). Among the six patients who received only one vaccination, one patient refused a second vaccination, and five patients (and their local pediatricians) deferred a second vaccination due to an increase in VZV-IgG-level after just one vaccination. We suggested to repeat VZV-IgG level measurements in these instances and to offer the booster vaccine in case of a negative or borderline VZV-IgG level.

Safety assessments
The VZV vaccinations were generally well tolerated. There were only minor adverse events, Table 3. There were no cases of rash or other vaccine-induced varicella disease symptoms and no evidence of PRD flare following vaccination.

Efficacy following vaccination
After a median follow-up of three years, no patient had suffered from breakthrough chickenpox or from HZ. A known exposure to VZV was documented in three cases (one to herpes zoster and two to chickenpox).

Discussion
To our knowledge, this is the first study employing a pre-vaccination stratifying algorithm centered on immunologic criteria, rather than on type of IS medication used, to decide whether a IS patient got vaccinated. The implementation of the pre-vaccination checklist ensured immune competence prior to vaccination. On that basis we demonstrate the safety and immunogenicity of VZV vaccination in an insufficiently protected cohort of patients with PRD receiving LIIS or HIIS, including biologic DMARDs.
All patients receiving LIIS and/or only a booster vaccination developed an increase in VZV-IgG above 200 mIU/ml (defined as positive by the manufacturer of the assay). Two out of 14 patients receiving HIIS did not: one on MMF (which has anti-proliferative activity in B-and Tcells) and one on leflunomide and abatacept (an agent blocking T cell co-stimulation). Abatacept and MMF have previously been associated with poor responses to non-live vaccine immunization [21][22][23]. Our study supports previous studies on VZV vaccination in patients with PRD receiving synthetic DMARD and a small cohort on biological agents in which a low rate of flares and rash were demonstrated, even though immunogenicity has been lower in these studies [15,16,24,25].
For the two-step checklist, we selected parameters regarding humoral and cellular immunity that in previous studies indicated the ability to control the standard-dose Oka virus strain of the VZV vaccine even during severe IS [26][27][28][29]. We emphasize that testing T cell function may be simplified using a standard TB-IGRA, such as the widely available QuantiFERON-TB Gold® and the TB-EliSpot® tests which both include a mitogen-induced T cell proliferation (to phytohemagglutinin, PHA) as positive control.
In our opinion, the decision to administer both (first and booster) live vaccination might be better based on the actual immunoreactivity rather than on the type of IS medication.
The alternative is to delay the initiation of IS, use a low-dose glucocorticoid bridging therapy or interrupt IS for several weeks [17,27]. In our experience, the latter approach is often refused both by treating physicians and patients due to an increased risk of disease flares. An intermediate solution would be to administer the first dose of the VZV vaccine prior to IS, and the second dose while on IS to optimize safety and immune response while decreasing the risk of disease flare.
In 2009, Germany initiated a general 2-dose VZV vaccine schedule for healthy children receiving the first dose between the age of 11 to 14 months and the second dose between the age of 15 to 23 months [30,31]. General  Table S1) [32][33][34]. Nonetheless, in juvenile idiopathic arthritis (JIA), the most common PRD, the peak age of onset is the second year of life, i.e. IS is often required before VZV immunization has been completed [35]. We demonstrate a rather high proportion of pre-existing VZV protection in our large cohort of around 2800 patients; 94% of screened patients were considered immune to VZV based on history of appropriate vaccination (92.1%) or chickenpox. This compares favorably to published national data on VZV vaccination coverage in Germany (around 70%) [36]. One reason may be that in our center a strong emphasis is placed on the recommendation to catch-up on vaccinations prior to initiating IS if possible. In countries with targeted immunization strategies the pre-vaccination checklist may be even more helpful protecting susceptible patients with urgent need for IS and high risk of chickenpox exposure. In patients without a 2-dose VZV vaccination and uncertain history of VZV disease, testing for VZV-IgG seems advisable prior to long-term IS, and negative or borderline VZV-IgG levels should lead to vaccination. On the other hand, post-exposure prophylaxis with acyclovir and VZV hyperimmune globulin may fail in preventing varicella disease during IS [1,[37][38][39]. Furthermore, reactivation and HZ often cause substantial morbidity during IS and occur much more often from latent wild-virus infection than after reactivation of the latent VZV vaccine strain [40][41][42][43].
One limitation of our study is that this is a rather small cohort of patients treated with various combinations of IS. We only documented an increase in VZV-IgG but CMI may be just as important. Long-term data for VZV-IgG and cell-mediated immunity (CMI) in patients with leukemia or immunosuppression (IS) who received two doses of the VZV vaccine showed that VZV-IgG peaks and then declines over time, whereas long-lasting VZV-CMI persists and protects between 75-100% of vaccinees from clinically relevant VZV disease after exposure [25,39,40,41,44,45]. Furthermore, post-vaccination titers are less in magnitude when compared to that following wild type VZV infection [25,40,39]. In a cohort of 20 patients receiving HIIS at our center, including rituximab, all patients with a history of chickenpox maintained very high seroconversion levels of VZV-IgG (range 1052 to greater than 2000 mIU/ml) (unpublished data).
It is noteworthy that none of the screened patients "failed" the stratification process. Based on further experience and data the pre-vaccination checklist may be simplified in the future, especially concerning booster-live vaccinations. Moreover, the application of the pre-vaccination checklist for first measles-mumps-rubella (MMR)  vaccination on LIIS and booster MMR vaccination on LIIS and HIIS should be discussed. Two minor modifications have been proposed by experts regarding the application of the MMR vaccine: first, proof of an increase in antibody titer following "diagnostic immunization" with inactivated vaccines and, second, a higher threshold for CD4 + T cells (>500/μl) [46].

Conclusions
In summary, we demonstrated that the monovalent VZV vaccine was well-tolerated and mostly efficacious in a cohort of VZV-susceptible patients with PRD receiving ongoing IS (including biologic DMARDs) after proving immunologic reactivity by means of a simple checklist. We believe that the safety and efficacy of this easy-toobtain approach should be prospectively evaluated in a larger patient cohort.

Funding
This study has been supported by a scientific grant of the "Verein Hilfe für das rheumakranke Kind" a non-profit foundation supporting pediatric rheumatology.

Availability of data and materials
The patient data were entered in Excel spread sheets including their identification as only the investigator team had an access to them. Data are however available from the authors upon reasonable request and with an extended permission of patients who have not yet consented on presenting their full data to any third parties.
Authors' contributions FS developed the pre-vaccination checklist and was involved in data analysis and drafting the manuscript; CH statistically analysed data and drafted the manuscript; SA diagnosis and management of the patients, data analysis; TM was involved in the development of the pre-vaccination checklist and assisted to draft the manuscript; JPH diagnosis and management of the patients and assistance to draft the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate The procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation and with the Declaration of Helsinki. The study was approved by the Institutional Board of