The 1965–1981 St. Lucia study was the first large-scale concurrent comparison trial of strategies for control of S. mansoni

In 1965, the St. Lucia government and the Rockefeller Foundation founded the “Research and Control Department of the Ministry of Health” as a focus for their collaboration on schistosomiasis mansoni control. Their first objective was to get the island’s increasing levels of infection under control. Pre-intervention studies in the high-transmission areas of each targeted valley found a median S. mansoni prevalence of 45% in children [1, p.270], with age-group prevalences reaching up to 91% in individual areas [1, p.53]. Serious illness from schistosomiasis was not uncommon, with a number of children under age 14 having egg counts over 1000, and many more with enlarged livers or spleens [2]. The researchers recognized that while a number of schistosomiasis control methods had been individually tested in trials previously performed in other countries, none had performed head-to-head trials to determine which single method was likely to be most successful in controlling disease or in preventing transmission [1, p.6]. The mountainous terrain of St. Lucia could allow researchers to test individual control methods in relatively isolated valleys (Fig 1), so the newly-constituted Department designed a large-scale, concurrent comparison study of drug-treatment vs. intermediate-host snail control vs. improvement of water supplies, which was the first study of its kind.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Sketch map of St. Lucia Island, indicating valleys selected for intervention assignments in the 1965–1981 comparison trial of S. mansoni control strategies.

Shaded areas indicate the separate valley locations where the different study interventions were tested during the course of the St. Lucia Project. Labels indicate the names of the selected valleys and which interventions were performed at each location. Map created with QGIS v.2.18.10 software (http://www.qgis.org) using a country base map provided by the GADM database of Global Administrative Areas (http://www.gadm.org).

https://doi.org/10.1371/journal.pntd.0006223.g001

The three main control methods tested in St. Lucia were: (i) targeted chemotherapy, (ii) chemical and environmental snail control (both area-wide and then focal), and (iii) installation of household water supplies with eventual provision of community showers, laundry units, and recreational water facilities. Each strategy presented its own specific set of challenges and technical demands, as noted in Table 1. Installation and maintenance of water supplies proved to be generally the most demanding intervention in terms of supervision, and chemotherapy the least. Snail control required the least amount of direct participation from the community.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Comparison of requirements and effects of control strategies tested in the St. Lucia Project (adapted from reference [1]).

https://doi.org/10.1371/journal.pntd.0006223.t001

Comparing targeted chemotherapy to long-term host snail control and to provision of household water supplies

Snail control.

The primary snail control trials took place in Cul de Sac Valley. After an initial period of three years to establish baseline data and to map breeding sites, the Department implemented area-wide mollusciciding with niclosamide from 1970 to 1975 [1, pp.118/131]. This involved initially treating all known snail habitats, and then routinely re-surveying these habitats and treating only those where snails were found. This was performed at 2-week intervals for streams and marshes and at 3 to 4-week intervals for banana drains [1, pp.136/165]. This aggressive, wide-coverage campaign brought human incidence per annum of S. mansoni down from 23% to 6% and prevalence from 45% to 24% among children in high-transmission areas of the Cul de Sac Valley [1, p.270]

Even with this success, the investigators realized that the costs and time demands of such intensive intervention would be prohibitively high for many resource-poor countries where schistosomiasis is endemic. As a result, they chose to test a new, less costly alternative, which involved using only focal snail control at specific transmission sites [6]. For this trial, they chose the Fond St. Jacques region of St. Lucia. In this second test, the researchers identified 12 sections of stream where infected Biomphalaria. glabrata had been found and where humans had frequent contact with the water [1, p.181]. These were selected as the most likely sites of transmission and thus were made the sole locations for monthly niclosamide treatments between 1976 and 1980, regardless of snail presence [6]. In high transmission areas of this valley, yearly incidence dropped from 15% to 8% and prevalence from 43% to 22% among children under ten years old [1, p.270]. Researchers concluded that, although more limited in scope, routine focal snail control was sufficient for decreasing infection levels, and they recommended continuing this intervention at 6-week intervals for the foreseeable future [7–10,] (Fig 2).

In 1978, Project researchers began a trial of supplementary biological control of the intermediate host B. glabrata snails by means of introducing non-native competitor snails such as Helisoma duryi and Melanoides tuberculata to local aquatic habitats, particularly to known transmission sites [11, 12]. Post-Project snail surveys in the late 1980s, the 1990s, and in recent years indicate that M. tuberculata has become the most common freshwater snail in St. Lucia [12]. There is now an absence or very low density of B. glabrata in many former S. mansoni transmission sites, suggesting a major shift in the snail ecology of the island [12]. This environmental change may explain, in part, the very low numbers of human S. mansoni infections now reported by island health service workers.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Comparison of the impact of individual control measures implemented during the St. Lucia study.

The effects of provision of water, of focal and area-wide snail control via mollusciciding, and of targeted chemotherapy in reducing S. mansoni incidence per annum, prevalence, and mean infection intensity (for those with egg-positive stools). Relative reductions over the initial 3–5 year course of each intervention are shown (as percentages) indicated by the horizontal bars. Data from [1, pp. 270–271].

https://doi.org/10.1371/journal.pntd.0006223.g002

Follow-on crossover trials that give us our best estimates for the impact of combining interventions

Anti-schistosomal chemotherapy was arguably the most successful single intervention during the initial phase of the trials. However, after the four annual chemotherapy campaigns were completed in the Marquis Valley (1973–76), the yearly incidence of S. mansoni infection rose again in Fond Assau and Talvern areas of the valley, without a clear explanation [1, p.106]. Three years after the final treatment, their incidence had returned nearly to pre-intervention levels, and infected snails were found in streams around the Marquis Valley [1, p.115; 7]. In addition to offering chemotherapy to residents again in 1979 and 1980, the Department responded by adding focal snail control to this site in March of 1980 [1, pp.95/113]. In the last 12 months of the St. Lucia project, no more infected snails were found in Marquis Valley, suggesting that the combined approach was significantly more successful in achieving interruption of transmission [1, p.115].

In South Richefond, schistosomiasis declined after installation of the new, household level water supplies. As the project progressed, and oral oxamniquine made anti-schistosomal treatment more accessible, chemotherapy was offered to any local Richefond residents who were still infected between 1975 and 1977 [1, p.199]. The household water plus treatment approach was seen by the Department as a way of strengthening the effects of chemotherapy because the household water-related reductions in yearly incidence of S. mansoni infection meant that people were less likely to become immediately re-infected after treatment [1, p.199]. Interestingly, a temporary resurgence in infections was seen when community members, interpreting their reported low incidence as meaning their rivers were safe, returned to the still-infested waters and yearly incidence again increased to pre-control levels [1, p.199]. This added chemotherapy campaign was carried out for 3 years, and afterwards, no further intervention was done other than maintaining the relatively costly water systems [1, pp.199/207]. By the end of the study in 1981, following this combination of interventions, annual incidence among children was at only 1% and prevalence was at 9% [1, p.271]

In Cul de Sac Valley, after the initial 1970–75 area-wide snail control program, the research team launched two further years of focal snail surveillance and control [1, p.152; 8], combined with targeted chemotherapy [1, p.154; 9]. This combined intervention was successful and further reduced prevalence from 24% to 6% [1, p.271]. Subsequently, the Department chose 12 streams and 26 banana drains (out of 145 initial sites) to undergo routine mollusciciding every four weeks regardless of observed snail levels [10]. This regimen continued for the last four years of the St. Lucia project and was sufficient to prevent any resurgence of infection [10].

Importance of reviewing these outcomes in light of upcoming plans for revisions to WHO guidelines for schistosomiasis control

With good reason, mass drug administration is now the mainstay of schistosomiasis morbidity control. But when chemotherapy fails to prevent reinfection, alternative control methods must be considered to interrupt transmission. The Marquis Valley chemotherapy trial was significantly enhanced when supplemented with snail control, and Cul de Sac and South Richefond obtained impressive results by supplementing snail control or WASH-related measures with chemotherapy. The St. Lucia study was not a randomized trial, and very local factors may have biased some of its findings. However, its extensive observational data on the concurrent implementation of very different approaches to schistosomiasis control offers a very strong base of evidence for modern-day decision-making regarding potential elimination strategies. The project’s results are not fully applicable to every country fighting schistosomiasis, particularly given St. Lucia’s island ecology and the fact that the study only dealt with S. mansoni. Nevertheless, the island has significantly minimized Schistosoma infections, and the results of the multiple approaches that it undertook back then deserve a close second look today.