Research Article: Risk assessment for the implementation of controlled human Schistosoma mansoni infection trials in Uganda

Date Published: August 13, 2019

Publisher: F1000 Research Limited

Author(s): Jan Pieter Koopman, Moses Egesa, Anne Wajja, Moses Adriko, Jacent Nassuuna, Gyaviira Nkurunungi, Emmanuella Driciru, Gijsbert van Willigen, Stephen Cose, Maria Yazdanbakhsh, Pontiano Kaleebu, Narcis Kabatereine, Edridah Tukahebwa, Meta Roestenberg, Alison M. Elliott.


Schistosomiasis is a parasitic infection highly prevalent in sub-Saharan Africa, and a significant cause of morbidity; it is a priority for vaccine development. A controlled human infection model for
Schistosoma mansoni (CHI-S) with potential to accelerate vaccine development has been developed among naïve volunteers in the Netherlands. Because responses both to infections and candidate vaccines are likely to differ between endemic and non-endemic settings, we propose to establish a CHI-S in Uganda where
Schistosoma mansoni is endemic. As part of a “road-map” to this goal, we have undertaken a risk assessment. We identified risks related to importing of laboratory vector snails and schistosome strains from the Netherlands to Uganda; exposure to natural infection in endemic settings concurrently with CHI-S studies, and unfamiliarity of the community with the nature, risks and rationale for CHI. Mitigating strategies are proposed. With careful implementation of the latter, we believe that CHI-S can be implemented safely in Uganda. Our reflections are presented here to promote feedback and discussion.

Partial Text

Schistosomiasis is a parasitic infection affecting approximately 230 million people worldwide
1. Infection is caused by trematodes (flukes) of the genus
Schistosoma. Because the infection is responsible for considerable morbidity worldwide, particularly in Africa, schistosomiasis was recently listed among the top 10 infections for which a vaccine should urgently be developed

We identified risks and potential approaches to mitigation based on relevant literature, experience from the Leiden CHI-S model, stakeholder discussions, and discussion with experts. The level of risk and effectiveness of proposed controls was determined by consensus between the authors. The inherent risk was defined as the risk before putting controls in place, calculated as the product of the likelihood and impact scores. The residual risk was similarly calculated, based on likelihood and impact scores after controls have been put in place. Mitigating controls could reduce the residual risk score by reducing the likelihood of an event occurring, or by reducing the impact if it should occur. Likelihood was scored as almost certain/common, 5; likely, 4; possible, 3; unlikely, 2; rare, 1. Impact was scored as critical, 5; major, 4; moderate, 3; minor, 2; insignificant 1. Resulting risk scores of 18–25 were considered high, and unacceptable. Resulting risk scores in the range of 9–17 were considered moderate, with further controls desirable if possible, and caution required if implemented at this risk level. Resulting scores of 0–8 were considered low, and usually acceptable.

According to our first idea, infected snails would be shipped. The WHO report ‘Guidance on regulations for the Transport of Infectious Substances 2017–2018’
11 provides information on how to adequately transport infectious substances. In accordance with these guidelines, shipment of
S. mansoni infected snails falls under ‘CATEGORY B, INFECTIOUS SUBSTANCES’ (UN3373). Shipment of live snails is a time-sensitive undertaking and therefore can only be facilitated by air shipment. Infectious substances cannot be carried on as hand-luggage. Transport of infectious substances are subjected to International Air Transport Association (IATA) requirements. Packaging of Category B substances need to comply with rules set out in the P650 packaging instruction
11. This involves triple packaging and proper marking and documentation. Upon arrival in Uganda, it would be crucial for the package to clear customs as quickly as possible so that snails arrive in good condition. In order to achieve this, the customs office should be notified about the arrival of the shipment. In collaboration with the customs officer, all required documentation should be prepared in advance and approval for import of the products should be sought.

To house the
Biomphalaria glabrata snails in Uganda, they would need to be kept in strict quarantine.
B. glabrata are not a naturally occurring snail host in Uganda, and should therefore not spread to the environment. In order to house snails, an incubator, or room temperature, set and monitored at 28°C is needed. The incubator (if used) door should be fully closed when the laboratory is not in use. Precautionary measures to contain the snails to the facility should be taken and include physical barriers, such as rooms with closed doors and windows. The snail culture basins and water drainage system should be covered with fine mesh to prevent escape (appendix 1 [Extended data
9]). In addition, access to the laboratory should be controlled and restricted to the research team. The incubator (if used) should preferably be positioned away from the door. Additional security measures could be a double door to create a sluice. Appendix 2 (Extended data
9) lists precautionary measures that should to be taken when working with schistosomes. Standard operating procedures (SOPs) will be exchanged with LUMC and reviewed to fit the Ugandan facility. These SOPs deal with culture processes as well as the disposal of infectious material.

This approach only requires transport of
S. mansoni infectious material. This would use the second transport approach described in option 1, within a preserved liver sample from a schistosomiasis-infected hamster. The same regulatory guidelines for transporting infectious material apply. With regard to Ugandan snail species, there is variability between snail species in susceptibility to
S. mansoni infection; however, there is experience of conducting infection of local species at the Vector Control Division
10, so this is expected to be feasible. A major advantage of this approach is that the potential ecological and genetic risks related to introduction of a non-endemic snail species can be avoided.

The alternative to shipping infectious material and snails from The Netherlands is to re-establish the full laboratory life cycle of
S. mansoni using Ugandan snail species and Ugandan isolates of
S. mansoni. The life-cycle has been maintained in the past at the Vector Control Division of the Ministry of Health, but is not currently available. The advantages of using a Ugandan life cycle include reducing the environmental risk associated with non-endemic snail species and schistosome strains. In addition, this model would be most representative of the field infections in Uganda. Similar to option 2, although susceptibility to
S mansoni infection varies between snail species, we do not expect this to be an issue, because the Vector Control Division has experience in infecting local species. There are however several challenges with using Ugandan snails and isolates. With regard to the new schistosome laboratory strain, the characteristics of this would be unknown in terms of virulence and susceptibility to praziquantel treatment. Determining these characteristics would not be simple, since validated tests for schistosome resistance are currently not available. In addition, the new isolate would not be clonal and variability within the newly collected schistosome population might result in variable responses in the host, and to drug treatment. An inbred Ugandan strain could be achieved by crossing clonal males and clonal females to produce a single F1 generation and subsequently cloning the offspring through snails followed by another crossing. This procedure would need to be repeated several times to be able to generate a reasonably monomorphic strain. This process would be laborious and time-consuming and might also result in quite atypical parasites, not necessarily representative of the Ugandan population of schistosomes in general. Ugandan populations have been exposed to regular praziquantel treatment for over a decade, so there is a risk that the initial isolates would include individuals with relative praziquantel resistance
13 and could not be established with certainty in the initial stages of the above process. Starting with a more diverse selection of cercariae would generate a more representative laboratory population of Ugandan schistosomes, but would mean that the characteristics of any particular clone (notably pathogenicity or praziquantel resistance) selected for CHI-S would be unpredictable.

The single-sex
S. mansoni challenge has been designed to prevent the occurrence of egg-associated morbidity. In the current model, volunteers participating in the trial will be infected using only male cercariae which penetrate the skin and result in patent infection. In future, a single-sex female cercariae model may also be used to infect volunteers. The sex of the male cercariae can be determined using a specifically designed multiplex real-time PCR which has been described elsewhere
3. Once infected, individuals should avoid any exposure to contaminated water. If a subject were to be naturally infected over the course of the study, this might lead to mixed, male and female, infections, with mating of the schistosomes resulting in egg production that causes morbidity. If the Puerto Rican strain used in Leiden is imported for use in Uganda, mating and (if adequate sanitation is not used) excretion of eggs into the environment could alter the genetic make-up of Ugandan schistosome populations, with unknown consequences. However, given the fact that the Puerto Rican strain has been kept in rodents for >60 years, it seems likely that fitness in humans will be, if anything, lower than Ugandan human strains. Moreover, given that the Puerto-Rican strain is relatively inbred after prolonged passage in the laboratory, and was shown to be praziquantel-sensitive in the CHI-S, hybridisation with Ugandan schistosome populations is unlikely to result in increased praziquantel resistance or virulence.

Controlled infection with
S. mansoni has been successfully performed in 17 Dutch volunteers. Although the single sex infection does not cause egg-related morbidity in volunteers, it may cause symptoms in response to the infection. These include dermatitis due to the percutaneous penetration of the cercariae and an acute schistosomiasis as a consequence of a systemic hypersensitivity response
16. Severe acute schistosomiasis syndrome (Katayama fever) may present with symptoms such as fever, fatigue, myalgia, malaise, non-productive cough, eosinophilia and patchy infiltrates on chest radiography. In Leiden, several volunteers reported with systemic symptoms which seemed to be an acute schistosomiasis syndrome, with one volunteer presenting with prolonged symptoms of Katayama fever
16. In addition, one volunteer presented with peri-orbital oedema which lasted one day, and may have been related to the infection
16. Such symptoms can be treated symptomatically and all recovered. Both these volunteers had received the highest dose of cercariae (30 cercariae) used in Leiden. The risk of severe symptoms can be minimised by dose escalation in modest increments. The impact can be reduced by careful monitoring, provision of symptomatic relief and abrogation of infection by treatment if necessary. Frequent follow up visits need to be scheduled throughout the trial to discuss adverse events and conduct clinical assessments of the study volunteers. Safety laboratory tests need to be routinely performed. Volunteers can also experience side effects related to the praziquantel treatment. Common side effects include nausea, dizziness, and fatigue. Volunteers can be reassured that these symptoms are well recognised and transient. Their severity can be reduced by taking praziquantel after food. Symptomatic relief can be provided when required.

In this document we have reflected on the potential risks involved in establishing a controlled human infection model for schistosomiasis in Uganda. The opinions expressed and risk scores allocated have been arrived at by discussion between the authors and are therefore subjective. In submitting this document to open peer review through the African Academy of Sciences Open Research Platform we welcome discussion of these issues.




Leave a Reply

Your email address will not be published.