Rationale. Cancer patients are more likely to develop herpes zoster (HZ) and its complications. An acute episode of HZ implies in postponement of current cancer treatments, hospitalization, high risk of development of postherpetic neuralgia and high costs for paying sources. A better understanding of disease incidence, hospitalization, development of postherpetic neuralgia, and costs generated for funding sources may increase awareness of the impact of vaccination on oncology population.

Objectives. Obtain HZ incidence and percentage of hospitalization in study population according to cancer type, topography, and stage. Estimate hospital costs arising from hospitalizations for different paying sources: National Public Health System (SUS - Unified Health System), health operators and private source (patient).

Methods. We propose a retrospective descriptive study of HZ episodes occurred in oncological population attended at AC Camargo Cancer Center between 2019 September and 2022 August.

Rationale

Cancer patients are more likely to develop herpes zoster (HZ) 1, as well as reactivation after first episode, complications such as skin infection, postherpetic neuralgia, vision loss (ophthalmic involvement) and severe forms of disease (disseminated, encephalitis). Other relevant implications to be mentioned: 1) an acute episode of HZ implies in postponement2 of current cancer treatments, which can impact the chances of cure/control of cancer; 2) compared to general population there is greater need for hospitalization3; 3) postherpetic neuralgia can be a big detractor of quality of life and performance status4. Besides that, episodes of HZ in oncologic population generate high costs for paying sources to treat acute episodes and its sequelae5. Treatment of HZ is based on antiviral agents and analgesics for postherpetic neuralgia. For prevention of an episode there are two types of vaccines: 1) live attenuated herpes zoster vaccine is available in Brazil for use in private network since 2008, but oncological population (immunocompromised) or those under 60 years of age are not eligible; 2) In August 2021 a HZ recombinant vaccine6,7 was approved by ANVISA for immunocompromised patients over 18 years old; this vaccine has been available at private vaccination clinics since June 2022. A better understanding of disease incidence, hospitalization, development of postherpetic neuralgia, and costs generated for funding sources may increase awareness of the impact of vaccination on oncology population.

Objectives

Primary goals

To estimate HZ prevalence in study population according to cancer type, topography, and stage.

To estimate percentage of hospitalization in study population.

To estimate hospital costs arising from hospitalizations for different paying sources: National Public Health System (SUS - Unified Health System), health operators and private source (patient);

Secondary goals

Describe temporal correlation between occurrence of HZ episode and oncological therapy in progress (chemotherapy, radiotherapy, immunotherapy, surgery in the last 90 days) to stablish a timeline of occurrence of HZ during oncological journey;

2.2.2. Estimate the occurrence of neuropathic pain in this population;

2.2.3. Estimate the impact of the episode in delaying cancer treatment;

2.2.4. Estimate the prevalence of recurrence within one year of the initial episode.

3. Methods

3.1. Study design

We propose a retrospective descriptive study of HZ episodes occurred in oncological population attended at AC Camargo Cancer Center between 2019 September and 2022 August. AC Camargo Cancer Center is a private hospital, located in the city of São Paulo - Brazil, specialized in cancer treatment, which annually performs around 23,000 attendances. Hospital costs in 2022 were funded approximately 85% by health insurers, 10% by National Health Service (SUS) and 5% on private basis. Inclusion of patients will be carried out in two ways: 1) search in electronic medical records of attendances occurred between 2019 september and 2022 august in which HZ ICD - 10 (Table1) (International Statistical Classification of Diseases and Related Health Problems) has been included among diagnoses; 2) through internal records from hospital infection control sector of patients who remained in contact and/or aerosol precaution, in this period, under the hypothesis of HZ. After initial selection, a detailed search for mentions of HZ diagnosis in medical evolutions will be carried out to confirm inclusion. Diagnosis of HZ is clinical and is based on a painful, polymorphic, acute cutaneous lesions composed of macules, papules, vesicles and crusts at different stages of development, affecting delimited dermatomes or disseminated (when it affects two or more dermatomes).

Selected cases will be submitted to following data collection:

3.1.1. Demographic data;

3.1.2 Type of cancer (hematological or solid tumor - solid tumor topography such as: central nervous system, skin, head and neck, chest/lung, breast, gastrointestinal tract, urinary tract, sarcomas, bone tumors) and stage (TNM Staging System)8;

3.1.3. Ongoing oncological therapy modality - chemotherapy, radiotherapy, immunotherapy and surgery performed up to one year before hospital admission; in case of hematologic malignancies, if patient underwent bone marrow transplantation; individuals not receiving any oncological therapy modality will be classified as treatment naïve or palliative care. Individuals submitted to more than one treatment modality will be grouped according to the combinations found.

3.1.4. Severity of HZ episode (local disease or disseminated, eye or central nervous system involvement);

3.1.5. Medical fees and fixed hospital costs related to attendance in emergency ward and hospitalization; ambulatory costs will not be included;

3.1.6. Medical records related to schedule of oncological therapies (chemotherapy, radiotherapy, immunotherapy and surgery) and records of changes or postponements of treatment after an acute episode of HZ - to avoid bias, will only be included for this analysis cases in which HZ is the only complication reported in the period;

3.1.7. Records of persistent neuropathic pain three months after initial HZ episode;

3.1.8. Records of HZ recurrence after one year.

3.2. Study population

Adult cancer population (solid tumor or hematologic malignancies) from Ac Camargo Cancer Center, São Paulo - Brazil.

3.2.1. Inclusion criteria.

Patients aged 18 years and older

Admission at AC Camargo Cancer Center between 2019 september and 2022 august at emergency department and/or inpatient ward, including intensive care unit

Acute episode of Herpes Zoster or recurrence (see item 5.1. for ICD-10 codes).

Acute episode of HZ is clinically defined as painful, polymorphic, acute cutaneous lesions composed of macules, papules, vesicles and crusts at different stages of development, affecting delimited dermatomes or disseminated (when it affects two or more dermatomes).

Recurrence of HZ is defined as any occurrence of HZ that happens after the initial episode.

Active cancer is defined as: diagnosis of local, regionally advanced or metastatic cancer; treatment naïve, receiving oncological therapies or palliative care in the last year.

Exclusion criteria.

Age under 18 years

Absence of evidence of oncological disease: cases classified as complete remission or without evidence of oncological disease (not receiving any kind of oncological treatment for at least one year)

3.2.3. Sampling

It is estimated that 150 individuals will be included in the study through convenience sampling through all available chart reviews.

Study endpoints

3.3.1. Primary endpoint: profile the occurrence of HZ in oncological population treated in a Brazilian cancer center.

3.3.2. Secondary endpoints: estimate hospital costs of HZ; estimate impact in changing or delaying treatment planning in cancer patients.

3.4. Data collection.

Data will be collected from Tasy Philips electronic medical record. Data analysis will be performed using data exported to IBM SPSS Statistics version 28 software.

3.5. Analysis.

3.5.1. Costs analysis

For patients hospitalized will be calculated average and mean revenue per patient/per day. We choose to use "per day" to decrease the impact of patients that have longer length of stay. For patients not hospitalized (emergency room attendances) we will use average and mean revenue per patient. Costs will be presented in American dollars.

3.5.2. Statistical analysis.

The point prevalence is calculated as follows:

Formula:

Point Prevalence= total number of HZ cases in time period/ total number of all patients in time period (per year). The total number of patients will be the "catchment population" or oncology patients attending the hospital each year

In terms of temporal correlation or incidence, the calculation of time will consider dates related to the oncological therapy scheme (chemotherapy, radiotherapy, immunotherapy, and surgery), dates of records of alterations or postponements of treatment after an acute episode of HZ, dates of records of persistent neuropathic pain three months after the initial episode of HZ, and dates of HZ recurrence records after one year.

To avoid bias of incidence all individuals diagnosed with HZ in the period will be included (according to inclusion criteria) independent of stage (severe or mild disease) of oncological disease.

We will use instrumental variables to control for confounding such as cancer type and stage, tumor topography. An instrumental variable is a factor that is associated with the exposure of interest (often a determinant of the exposure of interest), so that if we categorize the study population by different levels of the instrumental variable, these categories will have different levels of exposure of interest. However, an important condition is that the instrumental variable is not directly associated with the outcome or indirectly associated with the outcome by variables other than the exposure of interest. If these requirements are met and the risk of the study outcome varying between groups with different levels of the instrumental variable, then this variation can only be explained by difference in exposure levels of interest between groups or by chance. As the instrumental variable is unrelated to the outcome, except for exposure, even unknown confounding is removed.

For continuous numerical variables, we will calculate measures of central tendency, that is, means (standard deviation) or medians (range and range interquartile).

We will use the Kolmogorov-Smirnov test and the Shapiro-Wilk test to test the normality of continuous variables. Student's t test or ANOVA will be used to compare means between 2 or more groups. For variables that do not meet the requirements of a normal distribution, we will use the Mann-Whitney test for two groups and the Kuskal-Wallis test for three or more groups. For categorical variables, when comparing each group, the percentage or risk will be evaluated using the chi-square or Fisher's exact test. We will use instrumental variables to control for confusion. An instrumental variable is a factor that is associated with the exposure of interest (often a determinant of the exposure of interest), so that if we categorize the study population by different levels of the instrumental variable, these categories will have different levels of the exposure of interest. . However, an important condition is that the instrumental variable is not directly associated with the outcome or indirectly associated with the outcome through variables other than the exposure of interest. If these requirements are met and the risk of the study outcome varying between groups with different levels of the instrumental variable, then this variation can only be explained by difference in exposure levels of interest between groups or by chance. As the instrumental variable is unrelated to the outcome, except for exposure, even unknown confounding is removed.

Univariate and Multivariate Logistic Regression will also be performed to calculate the association and estimate risk-OR, with its respective confidence interval-CI/95%. P values will be considered significant when <0.05. We will select the variables for the regression models after

• Careful analysis of the categories - 2x2 contingency table - without reset cells.
• Quantitative independent variable, will be categorized, with cutoff point from the literature.
• Chi-square test - p<0.20 - UNIVARIATE - increasing input order of p-value, for Multivariate Regression.

3.5 Study Conduct, Management and Ethics

The diagnosis of Herpes Zoster can be established, in most patients, based on history and classic dermatomal appearance of the rash (painful, polymorphic, acute cutaneous lesions composed of macules, papules, vesicles and crusts at different stages of development, affecting delimited dermatomes or disseminated (when it affects two or more dermatomes). Varicella-zoster DNA PCR tools can be used for confirmation in patients with atypical clinical presentations; of all clinical specimens, the yield of the first vesicular lesions is the greatest. If vesicular fluid cannot be obtained, other acceptable alternatives include lesion scrapings, scabs, tissue biopsy, saliva, cerebrospinal fluid (CSF), and blood. In Brazil, molecular methods are not widely available. At Ac Camargo, this diagnostic tool is available at a high added cost (not covered by health insurers), and is not routinely used, only as an exception, when the clinical aspect is not typical and there is diagnostic doubt9.

In Ac Camargo Cancer Center (research Center) the information is electronic. The PI will lead this study's data extraction and management. Anonymized aggregated HZ patient data will be extracted from the medical chart databases following the study timeframe. The data integrity will be maintained throughout the data extraction process by placing a quality assurance process to avoid missing data. This study will comply with all applicable laws regarding participant privacy and the guiding principles of the Declaration of Helsinki. No direct subject contact or primary collection of individual human subject data will occur. Study results will be in tabular form, and aggregate analyses that omit subject identification; therefore, informed consent are not required. Any publications and reports will not include subject identifiers. The study will be submitted to the ICE/IRB review and approval and will be submitted to Plataforma Brasil, the national and unified base of records of research involving human beings for the entire Research Ethics Committees system, which articulates different primary sources of information on research involving human beings in Brazil.

3.6. Limitations

Cancer patients usually have many clinical intercurrences, which may make it difficult to attribute the delay in oncology schedule to HZ episode. Deficits in register of ICD in medical records may limit inclusion of subjects.

Data Dissemination

We intend to disseminate the results of the study to the scientific community by publishing the study in searchable, peer-reviewed scientific literature within 12 months from the completion of the analysis. We intend to disseminate the results in congresses such as the Brazilian Congress of Infectology, as well as in international congresses, as the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) and IDWEEK Annual World Congress.

4. External Collaborators, Experts and Institutions

Contributing External Authors:

Rosa Nascimento Marcusso - statistics professional; will contribute with statistical analysis.

Contributing Authors from A.C. Camargo Cancer Center

Karen Cristina Migotto - Pharmacist; will contribute to data collection and preparation of manuscript.

Ivan Leonardo Avelino Franca E Silva - infectious diseases physician; will contribute to the preparation of the manuscript.

Natalia Martinez Martos - Costs Specialist; will contribute with costs analysis.

Tables

Table 1. ICD - 10 included codes

B02

Zoster [herpes zoster]

B02.0

Zoster encephalitis

B02.1

Zoster meningitis

B02.2

Zoster, other, with CNS involvement

B02.3

Zoster eye disease

B02.7

Disseminated Zoster

B02.8

Zoster, with other complications

B02.9

Zoster, uncomplicated

G53.0

Postzoster neuralgia

International Statistical Classification of Diseases and Related Health Problems 10th revision

References

Forbes HJ, Bhaskaran K, Thomas SL, et al. Quantification of risk factors for herpes zoster: population cased case-control study. BMJ 2014;348:g2911.

Pedrazzoli P, Lasagna A, Cassaniti I, Ferrari A, Bergami F, Silvestris N, Sapuppo E, Di Maio M, Cinieri S, Baldanti F. Vaccination for herpes zoster in patients with solid tumors: a position paper on the behalf of the Associazione Italiana di Oncologia Medica (AIOM). ESMO Open. 2022 Aug;7(4):100548. doi: 10.1016/j.esmoop.2022.100548. Epub 2022 Jul 16. PMID: 35853350; PMCID: PMC9434335.

Curran D, Hunjan M, El Ghachi A, et al. Herpes zoster related healthcare burden and costs in immunocompromised (IC) and IC-free populations in England: an observational retrospective database analysis. BMJ Open 2019;9:e023502. doi:10.1136/ bmjopen-2018-023502

van Wijck AJM, Aerssens YR. Pain, Itch, Quality of Life, and Costs after Herpes Zoster. Pain Pract. 2017 Jul;17(6):738-746. doi: 10.1111/papr.12518. Epub 2016 Oct 13. PMID: 27611885.

Prosser LA, Harpaz R, Rose AM, Gebremariam A, Guo A, Ortega-Sanchez IR, Zhou F, Dooling K. A Cost-Effectiveness Analysis of Vaccination for Prevention of Herpes Zoster and Related Complications: Input for National Recommendations. Ann Intern Med. 2019 Mar 19;170(6):380-388. doi: 10.7326/M18-2347. Epub 2019 Feb 19. PMID: 30776797.

Anderson TC, Masters NB, Guo A, Shepersky L, Leidner AJ, Lee GM, Kotton CN, Dooling KL. Use of Recombinant Zoster Vaccine in Immunocompromised Adults Aged ≥19 Years: Recommendations of the Advisory Committee on Immunization Practices - United States, 2022. MMWR Morb Mortal Wkly Rep. 2022 Jan 21;71(3):80-84. doi: 10.15585/mmwr.mm7103a2. PMID: 35051134; PMCID: PMC8774159.

BLA Clinical Review Memorandum. Shingrix - Zoster vaccine recombinant, adjuvanted.Clinical Reviewers: Paula Ehrlich Agger, MD, MPH and Rebecca Reindel, MD. Available in: https://www.fda.gov/media/108793/download.

Rosen RD, Sapra A. TNM Classification. [Updated 2023 Feb 13]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2023 Jan-. Available from: https://www.ncbi.nlm.nih.gov/books/NBK553187/

John AR, Canaday DH. Herpes Zoster in the Older Adult. Infect Dis Clin North Am. 2017 Dec;31(4):811-826. doi: 10.1016/j.idc.2017.07.016. PMID: 29079160; PMCID: PMC5724974.

Inclusion Criteria:

-Admission at AC Camargo Cancer Center between 2019 september and 2022 august at emergency department and/or inpatient ward, including intensive care unit

Acute episode of Herpes Zoster or recurrence (see item 5.1. for ICD-10 codes).

• Acute episode of HZ is clinically defined as painful, polymorphic, acute cutaneous lesions composed of macules, papules, vesicles and crusts at different stages of development, affecting delimited dermatomes or disseminated (when it affects two or more dermatomes).
• Recurrence of HZ is defined as any occurrence of HZ that happens after the initial episode.
• Active cancer is defined as: diagnosis of local, regionally advanced or metastatic cancer; treatment naïve, receiving oncological therapies or palliative care in the last year.

Exclusion Criteria:

• Age under 18 years
• Absence of evidence of oncological disease: cases classified as complete remission or without evidence of oncological disease (not receiving any kind of oncological treatment for at least one year)

Adult cancer population (solid tumor or hematologic malignancies) from Ac Camargo Cancer Center, São Paulo - Brazil.