Cornelis De Pijper

Introduction In this thesis, two viral diseases are discussed in two separate parts, namely Zika (ZIKV) and rabies (RABV) virus. Both diseases are of military importance to prevent losing strength of manpower. As the title implies: “Quis custodiet ipsos custodes?” (Who guards the guardians?), if we can protect our military personnel, then they will protect us. The general introduction of Chapter 1 starts with a history of military losses, 75% of which are caused by disease and non-battle injuries (DNBI), where preventive measures can be applied to lower their number [1]. In this thesis, for both diseases, measures are available to prevent contracting the disease. Apart from research on preventing infection by both viruses, clinical aspects are also topic of this thesis.

Part 1: Zika virus (ZIKV)

ZIKV In the general discussion, a brief of epidemiology, clinical characteristics and transmission routes are provided. Additionally, international guidelines as well as guidelines of the Dutch Ministry of Defence (MOD), concerning the prevention of sexual transmission of Zika virus infection (ZVI), are discussed in Chapter 1.

The outbreak of ZIKV in the Americas and frequent regular deployments of our armed forces in that area led to the observational study described in Chapter 2. In this study, we included 124 Dutch military personnel, who were deployed in Belize, Curacao or Saint Martin from December 2016 to December 2017. Military personnel was invited to voluntarily participate with a single blood sample to detect ZVI, if they had returned from a ZIKV endemic area two weeks or more before study enrollment. None of them tested positive for ZIKV IgG. This study also included a questionnaire on signs and symptoms that might be consistent with arbo-viral diseases [2]. If an individual had been symptomatic, serologic testing for dengue (DENV) and chikungunya (CHIKV) viruses was performed in addition to ZIKV, since all three viruses were endemic. Ultimately, 20 armed forces had been symptomatic, and one of them was infected by DENV, another one by CHIKV. In conclusion, no ZVI were found among the military personnel in this observational study, despite being deployed in areas with a high transmission rate [3].

In Chapter 3, a rare clinical aspect of ZVI is reviewed. This review was conducted after the first patient with proven ZVI and severe thrombocytopenia had visited our outpatient clinic [4]. This serious complication had never been reported before in the literature. After an extensive search, a further 28 unique cases were found in publications of 2016 until 2018 [4-12]. Unfortunately, five of 28 individuals died. To date, the exact pathophysiology remains unclear. Possible treatment options are platelet transfusion, administration of corticosteroids or intravenous immunoglobulins (IVIG), a watchful waiting strategy, or a combination of these. Because all individual cases were so different and the total number of patients has been limited to date, no firm conclusions can be drawn on the best therapeutic strategy.

Discussion ZIKV

Despite, the fact that ZIKV has first been described around 1947; and probably earlier ZVI epidemics in the past have been mistaken for DENV, it became noticed by the general public since the large outbreak in Brazil and surrounding countries in late 2015. The images from the news of newborns with microcephaly left a strong impression. Shortly after, on 1 February 2016, the World Health Organisation (WHO) announced a public health emergency of international concern (PHEIC) concerning ZIKV. At that time, much was still unknown about ZVI. This is also reflected in the wealth of papers published after 2015. On November 18, 2016, almost 10 months later, WHO already withdrew its ZIKV PHEIC because the epidemic was in decline [13].

In time, it turned out that ZIKV is also sexually transmissible. Several different guidelines were developed with recommendations to prevent sexual transmission. The main difference was the interval, ranging from 2 to 6 months, during which preventive measures were advised [14]. In the introduction, the rationale behind the different recommendations’ time intervals are explained. In the Netherlands, national guidelines advised to practice protected sex for 2 months after exposure. The Dutch MOD guidelines were more conservative, advising a 6-month interval. From the MOD’s point of view, this decision was motivated by the major possible consequences for pregnancies in their relatively young population, and the responsibility of the MOD to provide primary care for deployed military personnel and their families in the Americas. Currently, this MOD directive has been withdrawn after the ZIKV epidemic appeared to be completely halted. Should a next ZIKV outbreak occur, a 2-month interval of protected sex is probably sufficient for military personnel, provided this has not been disproved by new research by then [15]. Although more than 80 percent of ZVI are asymptomatic, we do not recommend standard serological testing after return. In exceptional cases (e.g. pregnancy), a serological test can be considered to rule out a ZVI.

At this moment, the exact pathophysiology with regard to ZVI and thrombocytopenia is unknown. We postulate that ZIKV particles can cause an immune-mediated thrombocytopenic purpura (ITP), because a swift rise in platelet count after IVIG was observed. Thus, a patient with severe thrombocytopenia and bleeding disorders, can recover quickly by a simple and short intervention. This is certainly a feasible treatment strategy in low- and middle-income countries [16, 17].

Future prospective ZIKV

As mentioned briefly in the introduction, immunisation against ZIKV will probably be available in the near future [18]. From a military point of view, this is an excellent way to prevent ZVI. Once effective and safe immunisation are on the market, the only consideration is the cost of the immunisation to prevent ZIKV epidemics worldwide.

A completely different approach is to prevent mosquitoes transmitting the virus. In malaria, ongoing work on genetically modified mosquitoes is performed to prevent transmission [19]. Wolbachia bacteria are introduced in native Aedes aegypti mosquitoes in order to prevent arboviral transmission, including Zika virus [20]. These strategies appear promising, but additional field studies will have to be performed. With a decrease in incidence of Zika virus, this will certainly take some time.

For the time being, only preventive measures against mosquito bites such as impregnated clothing, wearing long sleeves and pants and the use of insect repellent on unprotected skin, remain available to prevent mosquito transmitted ZVI.

Part 2: Rabies virus (RABV)

RABV Most chapters of this thesis cover rabies virus-related topics. The general discussion in Chapter 1, includes the history of rabies and development of immunisations, epidemiology, pathogenesis, clinical stages, diagnostics, and treatment of rabies. In addition, guidelines and the ‘One Health’ approach are discussed in the same chapter. Finally, the immunisation strategy is mentioned with regard to pre- and post-exposure prophylaxis (PrEP, PEP), the most important options to prevent rabies.

Part 2 starts with Chapter 4 ‘intensely’ about clinical rabies, reporting two fatal cases of rabies virus infected patients in our centre. Both patients deceased at the ICU of Amsterdam UMC, despite all treatment efforts, including the intravenous and intrathecal administration of monoclonal antibodies. To date, evidence is lacking for curative treatment of clinical rabies, including the Milwaukee protocol. In literature, the common denominator in rabies ‘survivors’ is a high antibody titre at the time of onset of symptoms [21]. In this chapter, we provided guidance on how to proceed in case of a new rabies presentation: what treatments to avoid and when to choose for supportive / experimental care, versus palliative care.

To prevent this fatal disease, the presence of rabies virus neutralising antibodies (RVNAbs) is important. This is discussed in Chapter 5 with a systematic review and meta-analysis. A total of 5,426 subjects were included, from 36 studies. All, except for one subject retrospectively diagnosed with lymphoma, had adequate RVNAbs after booster immunisation [22]. WHO considers a titre of 0.5 IU/mL or higher as adequate [23]. This systematic review shows that rabies immunisations are effective. According to the meta-analysis, regardless of the route of administration, intradermal (ID) versus intramuscular (IM) immunisation achieved similar results concerning high and fast antibody responses after booster immunisation, which is referred to as ‘boostability’. The interval between the initial immunisation series and booster immunisation (range 1-32 years) does not affect the boostability. Also, divergent immunisation schedules (e.g. at that time a 2-dose PrEP) result in similar boostability.

In 2015, a nation-wide rabies vaccine shortage occurred in the Netherlands and surrounding countries. At that time, a large military cohort had to be immunised for short-notice deployment. In Chapter 6, we describe how the MOD was unable to administer all assigned military personnel with three IM rabies vaccines because of vaccine shortage. Therefore, a single dose ID administration was considered, as due to fractional dosing, only a third of the volume of the vaccine was required to immunise all assigned military personnel. Condition for approval by the MOD was an adequate rabies titre before deployment. In our study design, titres were performed on blood samples taken just before the third ID vaccine administration on day 21 or 28. Although all participants received three ID PrEP immunisations, effectively, immunogenicity data of a 2-dose ID PrEP-schedule was thus obtained. Only three of 430 individuals had an inadequately low titre on day 21. Eventually, all military personnel had adequate titres before deployment.

A follow-up boostability study after an ID PrEP-schedule was conducted, as reported in Chapter 7. Only those military personnel from the cohort described in Chapter 6, were eligible. If individual military personnel were indicated for a booster immunisation due to a new deployment in rabies endemic areas, blood samples for immunogenicity of a booster immunisation were obtained after informed consent. While the study was ongoing, the MOD guidelines changed and no booster immunisation was administrated after a three-dose schedule. Therefore, after 17 inclusions the study was preliminary terminated. The results were as expected, with good boostability after ID PrEP-schedule 1-2.5 years earlier, in line with the results from our systematic review and meta-analysis described in Chapter 5.

Chapter 8, covers long-term rabies vaccine boostability. In the literature, rabies boostability data after more than 10 years is scarce, especially after a PrEP-schedule (n=15) [24]. We performed a pilot study with 28 subjects to test the hypothesis that rabies anamnestic antibody response was reactivated within one week, after only one IM rabies booster immunisation. Study participants were divided in three groups depending on their initial PrEP-schedule, 10-24 years earlier; a 3-dose IM, 3-dose ID, and divergent (DIV) group. Within seven days, all subjects in these different groups showed antibody titres above 3.0 IU/mL, reflecting adequate boostability. In the DIV-group, six individuals with a previous 2-dose PrEP-schedule were included, with similar results to those with a 3-dose PrEP-schedule. This finding is important, as currently the PrEP-schedule consists of only two immunisations.

Discussion RABV

The two clinical rabies cases reported from our centre emphasise the utmost importance to prevent rabies. With adequate pre- and post-exposure measures, this lethal infection is fully vaccine preventable [23]. For WHO, as well as Dutch national authorities and the Dutch MOD, immunisation is considered of eminent importance to prevent clinical rabies. Scientific data worldwide were considered by WHO as part of the ‘Zero to 30’ goals, eventually resulting in simplified PrEP- and PEP-schedules [25].

The immunisation strategy to prevent rabies by PrEP and PEP is different than immunisation strategies for other vaccine preventable diseases. For many diseases, a single or booster immunisation is required to acquire life-long immunity [23]. When immunised against rabies virus (PrEP), additional immunisations for treatment after exposure (PEP) are always required, because high titres of neutralising antibodies at the site of injury are necessary to inactivate the virus upon entry. Failure to initiate PEP after an exposure can be fatal [26].

The rationale for PrEP is best explained in relation to PEP-treatment strategies: After an adequate PrEP-schedule, only two immunisations are required, in case of a potential rabies exposure, because the anamnestic response results in a fast increase of neutralising antibodies at the site of injury; these standard rabies immunisation can be obtained worldwide in most places. On the other hand, without any previous rabies pre-immunisation, PEP consists of four PEP immunisations, which, in absence of an anamnestic response, take time to be effective, PLUS rabies immunoglobulin (RIG) injected at the site of injury, in order to immediately neutralise rabies virus. RIG is scarce, not widely available, and very expensive; sometimes resulting in travel to another country, delaying the treatment and increasing costs [23].

At the start of this thesis, a very strict regimen of intervals between the 3-dose PrEP immunisations had to be adhered to; namely day 0, day 7, and day 21 or 28. Many travellers visit their travel clinic no sooner than two weeks before departure, so they were unable to complete this schedule before their trip. With the new 2-dose PrEP-schedule guideline, it is much easier to complete the immunisation schedule prior travel. Also, the cost of PrEP as well as the number of visits to the clinic are reduced [23].

Studies show that a 2-dose PrEP-schedule is safe and effective with regard to rabies antibody responses [27-29]. Most important is the boostability, reflecting a fast and high antibody response post-exposure to ensure no RIG is needed. Due to memory B-cells, the second immune response is faster and results in higher antibody levels than the primary response [30, 31]. Our systematic review showed adequate antibody responses in all individuals who received 2-dose PrEP.

Shortly after the WHO-recommendations had been issued, some countries including the Netherlands, adopted a 2-dose PrEP-schedule [25]. The data presented in this thesis have partly contributed to this decision. Since then, many countries have followed this strategy. In early 2021, Centres for Disease Control and Prevention (CDC) directly requested information on our long-term 2-dose boostability data, and soon after decided to change their PrEP-schedule to a 2-dose strategy as well [32].

Future prospective RABV

Hopefully, there will be a cure for rabies in the future. At the moment, there is no proven treatment for rabies once clinically manifest. Therefore, the focus lies on preventing the risk of contracting rabies. The worldwide WHO goal ‘zero to 30’ will support this by a ‘One Health Approach’, involving veterinary measures as well. No rabies in animals implies no transmission, and thus no human infections [33]. A recent report on the current worldwide situation on rabies is showing progress, but whether the target will actually be achieved within nine years remains highly uncertain [34].

If the PrEP schedule could be shortened to a ‘single visit’ PrEP-schedule; potentially, more people could be immunised, resulting in easier and cheaper PEP-treatment without RIG. Our Centre for Tropical Medicine and Travel Medicine participated in a Dutch multicentre study on boostability after a single visit PrEP-schedule. Initial results are promising, final results and publication are pending. Recent publications on a single visit schedule support the shortening of PrEP [35-39].

There are only case reports and case series available on immunocompromised patients (ICPs) [40-43]. Clinical research on rabies antibody response in ICPs is absent in