Study on the persistence of Zika virus (ZIKV) in body fluids of patients with ZIKV infection in Brazil

ZIKV is a mosquito-borne arbovirus in the family Flaviviridae, genus Flavivirus whose first isolation occurred in 1947 from a rhesus monkey in the Zika forest of Uganda [1]. The genus includes more than 70 viruses comprising important pathogens such as dengue, yellow fever, Japanese encephalitis and West Nile virus [2]. All flaviviruses include in their cycle a vertebrate host and insect vectors and many flaviviruses are clinically important as they elicit central nervous system diseases in infected individuals [3]. The primary mode of ZIKV transmission is through mosquito bites. The virus is almost exclusively transmitted by the Aedes aegypti mosquito but has been isolated from several other species of the genus Aedes such as Aedes albopictus, Aedes africanus and Aedes luteocephalus [4]. Any country where competent vectors are present bears risk for ZIKV outbreaks. In addition to the ongoing outbreak started in 2015 in South America, ZIKV has previously caused outbreaks in the Yap Islands, French Polynesia, New Caledonia, Cook Island and Easter Island [5]. The mosquito has a predilection for indoor environments which might explain its distribution reflecting patterns of human migration and urbanization [6]. Between 01/01/2007 and 02/06/2016 Zika was documented in 60 countries and territories of which 46 were experiencing their first outbreak since 2015 [7].

The most common symptoms of ZIKV infection are rash, fever, arthralgia, myalgia and conjunctivitis [8]. Recently a study in Rio de Janeiro, Brazil, identified maculopapular rash (97%) and pruritus (79%) as the two most common symptoms among confirmed cases [9]. Most of the patients with ZIKV infection have mild disease, but severe neurologic complications have been described in patients in French Polynesia and during the current outbreak in Brazil [10, 11].

ZIKV infection has become a major public health concern due to suspected causal relationship between ZIKV and microcephaly, amongst other neurological manifestations. The Pan American Health Organization/World Health Organization (PAHO/WHO) issued an alert in November 2015 in response to a large increase in the number of cases of microcephaly notified by the Brazil International Health Regulations National Focal Point. The country, which had its first autochthonous case confirmed in May 2015, and by November 2015 had spread to 15 states, experienced a staggering increase in rates of microcephaly in live births from 5.7/100,000 in 2010 to 99.7/100,000 by November 2015 [12]. This led the World Health Organization to declare the current outbreak as a Public Health Emergency of International Concern on 1 February 2016 [13].

Microcephaly is a rare neurological condition where babies are born with a small head in comparison to other babies of the same sex and age. The symptoms accompanying microcephaly depend on the level of disease severity and vary from mild to life threatening. There is no specific treatment for this condition. In addition to genetic factors, microcephaly can also be associated with other so-called STORCH infections, (i.e. syphilis, toxoplasmosis, rubella, cytomegalovirus and herpes simplex virus), maternal exposure to toxic substances and malnutrition during pregnancy [14].

ZIKV is the first flavivirus to be known to be sexually transmittable; current evidence suggests that sexual transmission has been increasing.

Sexual transmission of ZIKV was first suggested by Foy et al. in 2008 [15]. The authors described two cases that, after returning to their home in Colorado, US from Senegal, developed symptoms of ZIKV infection. Case 1 had sexual intercourse with his wife 1 day after his return and she developed symptoms of ZIKV infection 4 days later. She had no history of traveling outside the US during the previous year.

On 2 February 2016 the United States Centers for Disease Control and Prevention announced the first documented case of sexual transmission of ZIKV between two men through anal sex [16]. Suspicion of transmission through oral sex was raised in a case report published in April 2016 [17]. The case had sexual contact (vaginal intercourse, with no condom and no ejaculation, and oral sex with ejaculation) with her ZIKV infected symptomatic partner. Sexual transmission has been reported to occur up to 41 days after symptom onset in the index case [18].

By June/2016 ten countries have reported cases of sexually transmitted ZIKV (Argentina, Chile, Peru, US, France, Italy, Portugal, New Zealand, Canada and Germany) [7]. All index cases reported were male and presented symptoms consistent with ZIKV infection after travelling to areas of high ZIKV activity. Transmission of the infection to their sexual partners (male or female) could have occurred before, during or after symptoms onset [19]. None of secondary cases had other risk factors apart from having sexual activities with male partners who have been to areas of ongoing high ZIKV transmission activity.

Around 80% of individuals infected with ZIKV remain asymptomatic [4]. A likely man-to-woman sexual transmission of ZIKV between two asymptomatic persons was first reported in June 2016 [20]. Until this date all reported sexual transmissions implicated a male index case with symptoms of ZIKV infection. This finding may play a more important role than expected in the dynamic of ZIKV transmission. This is of concern mainly for pregnant women or those considering pregnancy and needs to be further investigated to inform counselling and recommendations.

Studies on the detection of the ZIKV in sweat and tears have not been published up to June 2016. This is an opportunity to increase generalizable knowledge and reduce uncertainty about ZIKV infection and, according to the results, to contribute to public health recommendations and actions.

The isolation of ZIKV in cell culture from urine was first described in 2014 in Canada [31]. The study reported the case of a Canadian traveler returning from Thailand who was initially diagnosed with dengue fever as result of preliminary laboratory investigations. Atypical dengue serological results, however, led the investigators to a reconsideration of infection by another flavivirus.

In February 2016 ZIKV was also isolated in Brazil where urine and saliva samples from 9 patients in Zika acute phase symptoms were inoculated in Vero cell culture resulting in one isolate returned from urine (and one from saliva) [32].

It was suggested that renal excretion of ZIKV can start around 4 days after symptom onset [33]. ZIKV RNA, however, has been found in urine 20 days after it was no longer detected in serum [34]. The urine has shown to yield positive results through PCR, in a patient with Guillain-Barré Syndrome after 21 days of the neurological symptoms onset [35].

Gingival crevicular fluid (GCF), a serum transudate or inflammatory exudate present in the gingival crevice surrounding teeth, and saliva are identified as distinct body fluids. The constituents of GCF originate from serum, gingival tissues, and from both bacterial and host response cells. Whole saliva and GCF can be distinguished by the fact that the IgG content of the GCF is several times greater than that of saliva.

As this is a non-invasive means of diagnosis, monitoring of viral markers and assessing the host response in infectious disease, and because saliva collection is less technique-sensitive than GCF, collection the latter will also be performed in the participants. No reports on the presence of ZIKV in GCF have been published so far.

This is a prospective observational cohort study. After signing a screening consent, male and female potential study participants presenting rash compatible with ZIKV infection in collaborating clinics will be tested for the presence of ZIKV in blood and urine. The following day, after providing informed consent for enrolment, those found to be RT-PCR positive for ZIKV infection, fulfilling all inclusion criteria will be enrolled and subject of regular collection of body fluids over a period of 12 months according to a pre-established schedule. Household/sexual contacts will be screened at the reference centers where the index case was diagnosed or if preferable at their household, and recruited following the same criteria. The standard care will be provided to all participants.

Participants will comprise men and women aged 18 years and over with diagnosis of ZIKV infection. They will include symptomatic cases diagnosed at the study collaborating clinics (index cases) and their asymptomatic or symptomatic household/sexual contacts.

ZIKV is an emerging pathogen about which little information is known. This is a prospective observational cohort study, whose prediction of the exact sample size required cannot be prospectively determined. However, considering the scarcity of existing information on viral shedding, even a few participants would contribute to the body of evidence on the topic and lead to improved public health recommendations.

A multistage cluster sampling will be used to identify potential participants for inclusion and follow-up in the persistence cohort. All consecutive individuals RT-PCR positive for ZIKV in urine and/or blood identified during the screening process will be included in the cohort. In addition, all adult household/sexual contacts (symptomatic or asymptomatic) of these clinical cases, will be offered participation, and to be tested and included in the cohort if found RT-PCR positive for ZIKV in blood and/or urine.

The following is the rationale for estimating the sample size needed for body fluids other than blood, when we hypothetically estimate the prevalence of persistence will be 50% after 12 months. The below sample size calculations refer to estimating a population proportion at a certain point (or period) in time.

Using a population size (N) of 1,000,000 for finite population correction factor, and hypothesized rate of ZIKV persistence in at least one body fluid that is near 50% by the end of 12 months post-ZIKV infection, with a two-sided 95% confidence interval width of 10% (or margin of error (ME) of ±5%), a sample size of 385 males and 385 females who are either positive for ZIKV in urine or in blood at baseline will be needed.

The potential introduced design effect of the different clinic facilities used in the consecutive recruitment of study participants, will be explored both in stratified analyses by clinic, and in addition clinic will also be run as an independent variable in multivariate models.

To account for the design/clustering effect at household level, we applied design effect DE of 1.5 assuming a cluster (family) size of 3 ZIKV-infected participants and an assumed an intra-cluster correlation (ICC) of 0.25, resulting into a sample size of 576 males and 576 females. We further assume that 10% of participants, index cases and household/sexual contacts, will be lost to follow up, by the end of the first year and hence the final sample size will be 634 males and 634 females. We rounded this number up to 650 males and 650 females. The sample size is tentatively estimated, given that there is little information on ZIKV persistence rates in bodily fluids, available in literature. The final sample size of 650 males and females is estimated for evaluation of the primary outcome which is the overall ZIKV persistence rate by 12 months (taking into consideration a design / clustering effect and loss to follow up). Since the role of different risk factors is not yet well known, the subgroup analyses will be done in an exploratory way and the effect of different factors on the primary outcome will be further explored in multivariable analysis models.

Participants with positive RT-PCR in either blood or urine will be asked to return for a follow up visit according to the study pre-established schedule (Table 1) which will have been discussed when the informed consent was sought. There will be one day flexibility from visit 2 in case for some reason the participant is not able to attend the clinic for specimen collection. Individuals who consent and have yielded a positive result for ZIKV infection will be followed for 12 months in order to evaluate for persistence of virus, reactivation and reinfection.

Menstruating women will be asked to provide a specimen as close to the first day of their period as possible, and a second specimen before the end of the period. Menstrual blood sampling may or may not be done at the same time as other body fluid testing. If the woman’s period does not coincide with other specimens’ collection times, she will be asked to attend for specimen collection on different days as appropriate. This will be explained by the study nurse at the time of the recruitment.

A study card ID check will be produced for each participant taking part in the study and the information with the correspondent identity for each ID will be locked in a safe file. In case of ID loss by the participant, the locked in ID may be accessed to allow verification of the participant’s identity.

Contact information will be reviewed at each visit. Text messages will be sent by the administration assistant prior to the appointments as reminders. If applicable, calls will be made to participants who have missed their appointments, and flexibility will be provided in terms of appointment bookings and any needs for additional information. In addition, community healthcare workers from the areas where the study will take place will be involved in the study to be in contact with the participants.

Appropriately trained study nurses will provide counselling on ZIKV infection to index cases, their household and sexual contacts. This will include the following information:

The prospective data collection from a participant including specimen collection will be discontinued after one year of the first specimen collection.

If male participants are unable to provide a semen specimen at two consecutive visits, they will be referred to specialized urological care if necessary. Additional eligible participants may be recruited to replace these cases.

A participant who cannot be contacted after 3 attempts (phone contact and home visit) for two consecutive visits or who does not present to the study clinic for follow-up visit for two consecutive visits will be considered as lost to follow-up.

After careful evaluation of risks and benefits of the research for pregnant women, and considering the frequency of collection of body fluids, pregnancy was considered as one of the exclusion criteria of the study, aiming at not provoking additional stress to pregnant women due to collections of body fluids and frequent visits to the study centers. According to this criterion of exclusion, and aiming above all the benefit for pregnant woman, women that become pregnant during the course of the study will have their participation discontinued for the same reasons.

They will be, however, promptly referred to existing specialized units for the care of pregnant women with Zika, where they will receive adequate care for their special needs and be followed up to delivery. All information and exams related to this patient and collected during the study period will be shared with the referred unit. In addition there is also the possibility of invitation to participate in ongoing cohort studies of pregnant women infected with ZIKV. This is available in all cities where the study is being conducted.

All laboratory results will be filled into a form marked with the participant’s study ID and kept in confidential storage for entering of follow-up results.

Depending on the tests to be performed, different tubes will be used for blood collection:

Visit 0: blood and urine

Pregnancy test: rapid urine test.

Diabetes screening: blood drop – glucometer tests

Considering that for the following fluids is not possible to predict the final volume that will be available after collecting, it is important to consider that is necessary to have at least 500 μl for RT-PCR and serology (if applicable) and an additional 500 μl to be kept frozen at − 70 °C for future studies.

For the following specimen types, one fresh aliquot should be processed for RT-PCR for ZIKV and 2 aliquots should be stored frozen at − 70 °C for Ag test and culture:

Note: Participants will be asked for consent their specimens to be used for future research. Specimens of consenting individuals will be kept at − 70 °C. Specimens that test positive by RT-PCR will also be processed for culture/viral isolation.

WHO will provide guidance on assay selection. Laboratories from different sites will be using the same assays, reagents and equipment.

Laboratory confirmation of recent ZIKV infection will comprise:

All laboratory results will be filled into a form marked with the participant’s study ID and kept in confidential storage for entering of follow up results.

Additional genomic analysis will be performed on a subset of positive samples by full-length sequencing methods. Whole genome sequencing and cognate viral load measures will be explored. Confirmatory follow-on procedures including real-time quantitative PCR will be performed as necessary, preferably with commercial reagents or alternatively those laboratory developed assays adapted from the literature. Viruses may be obtained by co-culturing with activated peripheral blood mononuclear cells (PBMCs), if other molecular methods for virus sequence determination fail.

Assays for cellular and humoral immune responses will be measured with the specimens stored during the study and will be determined based on the level of technology at the time. CD4+ and CD8+ T cells, B cells and NK cellular responses may be evaluated by a variety of assays including multi-parameter flow cytometry, intracellular cytokine staining and ELISPOT assays. Cytokine levels and inflammatory biomarkers may be measured by sandwich-ELISA method using validated commercial cytokine kits or alternative methods such as Luminex or other multiplexed platforms. Further advances in immunologic assays will occur and may be utilized. Immunophenotyping according to multiple lineage and differentiation markers may be carried out depending on the availability and salience of cutting edge assays. Additional cell types may also be evaluated by flow cytometry. Additionally, single cell immunophenotyping analysis, single cell sorting, antibody Fc-effector profiling and functionality assays and antibody cloning may also be performed.

Investigators will perform targeted analyses of genetic polymorphisms within host genes reported to influence the rates of pathogen acquisition and/or disease severity. These genes may function pathogen host restriction, and innate and adaptive immune system responses. Host restriction genes of interest include but are not limited to entry receptors and post-entry factors. Targeted genes of the innate immune system will include, but not limited to, killer immunoglobulin receptors (KIR) and cognate Class 1 ligands; dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN); and toll-like receptors (TLR). Those involved in humoral and cell-mediated immune responses will include Fcγ receptors, capable of binding diverse immunoglobulin isotypes; and the alleles comprising Class I HLA-A, -B, and -C loci, respectively. HLA typing procedures that will be used in this study have yet to be validated for clinical use, but represent essential research tools for the analyses of adaptive immune responses across multiple types of infections. HLA typing and other genetic testing data will be only collected from those who have provided specific consent in a separate genetic testing consent form. Data will be unlinked from personal identifiers, and reported in aggregate for the populations studied.

Isolated cross sectional identification of genomic expression is fraught with a lack of reproducibility due to patient heterogeneity, latent confounding and non-biological variation. Importantly, the quantitative approach developed for analyzing associations between within-patient gene changes and clinical outcomes conducted longitudinally should provide more robust predictions of outcomes than single measurements of gene expression controlling for a bulk of the intra-patient confounding and heterogeneity. The challenge is how to accurately interrogate and associate early longitudinal gene expression measured at multiple time points with long term clinical trajectories captured by a constellation of clinical variables. We seek to capitalize on the longitudinal structure within an individual, combining bioinformatics and statistical tools to elucidate pathway dynamics from the gene expression data. For host transcriptomics, approximately a small amount of blood will be collected into commercial blood RNA tubes, processed at the hospital, and then frozen for transport to the laboratory. Total RNA will be extracted with commercial kits and subjected to quality control. Specimens meeting quality control criteria will be further processed for use on a human gene expression microarray. Extracted RNA will be reverse transcribed and amplified using kits. Arrays will be scanned raw images will be processed using proprietary software. The gene lists generated will be formatted and subjected to network or biological pathway analysis. Networks will be further analyzed for relevance to the particular biological system and clinical setting.

The following instruments have been produced to provide measure of the variables of interest:

Some of the key variables will be measured through questionnaires administered by the study nurse. Three types of questionnaires will be used to collect information: Screening questionnaire (to be completed during the screening visit – see Additional file 1); Enrolment questionnaire (to be completed after RT-PCR for ZIKV positive results are delivered – see Additional files 2 and 3); and Follow up questionnaires (see Additional files 4 and 5).

Consultation forms (see Additional files 6 and 7).

These are medical forms to be completed by the study consultant in infectious disease during medical consultations. Information collected will include:

Quality checks will be built into the data management system used by the study team and quality control checks of critical data points will be employed to ensure standardization and validity of the data collected. In case of invalid answer an alert will be issued providing the possible answers for that specific question. A manual as well as training of persons who will be involved in the completion of the questionnaire will be provided.

Data queries will be generated and conveyed to each participating site. Ownership of the data entered into the central database will remain with the site that contributed it.

The database will be installed in a dedicated server for the management of the data in Java©. The server will allow authorised members of the study team to enter online data which will provide great flexibility as data entry will be possible at the point where the data is being collected. At the same time, consistency of answers will be verified during the process of questionnaire completion. This will be possible through any internet navigator. In addition to the security measures for accessing the system there will be automatic backups.

Nurses will use ultrabook (tablets) for entering the questionnaire responses collected during the visits to the index case household or at the collaborating clinics. Access to the database will be given to those individuals who will be entering and analyzing the data.

Members of the team with access to the database will be trained before receiving access to the productive server. Likewise only trained interviewers will be used for the interviews and specimen collection. The topics included in the questionnaires include asking for information that may be perceived as sensitive and efforts will be made to ensure a private setting and calm environment to assure internal validity of the data collected.

Handling and treatment of biological sampling and specimen will follow standard operating procedures. Specimens will be handled at appropriate laboratories following relevant protocols.

Study staff will review data collected and entered for quality and consistency. Data quality reports will be developed and generated weekly with feedback to study staff.

An Independent Data Monitoring Committee (IDMC) will be established by the Study management team and sponsors with the aim to ensure that the interests of patients taking part in the study are protected and that the scientific soundness of the study is kept from onset to completion. The IDMC shall be independent of the study and free of significant conflicts of interest.

The IDMC shall review the interim data at four intervals namely: after visit 8 (90 days), after visit 11 (180 days), after visit 14 (270 days) after visit 17 (360 days) for the study progress towards its goals and its integrity (compliance with the protocol, study sites performance and recruitment).

The size of the IDMC will range from 3 to 5 members of which two designated positions should be a chair who will lead the deliberations and sign the official documents and a statistician to minimize subjective judgment.

The IDMC will be supported by the study coordinator, Statistician and/or Data Management Team, as needed. They may be required to present reports at the IDMC meeting, but have no voting rights and may be excluded from parts of the meeting.

The IDMC will report to the study (principal investigator and sponsors) or Trial/study management team and sponsor and notify safety concerns as soon as possible.