Rethinking detection of pre-existing and intervening Plasmodium infections in malaria clinical trials

Abstract

Pre-existing and intervening low-density Plasmodium infections complicate the conduct of malaria clinical trials. These infections confound infection detection endpoints, and their immunological effects may detract from intended vaccine-induced immune responses. Historically, these infections were often unrecognized since infrequent and often analytically insensitive parasitological testing was performed before and during trials. Molecular diagnostics now permits their detection, but investigators must weigh the cost, complexity, and personnel demands on the study and the laboratory when scheduling such tests. This paper discusses the effect of pre-existing and intervening, low-density Plasmodium infections on malaria vaccine trial endpoints and the current methods employed for their infection detection. We review detection techniques, that until recently, provided a dearth of cost-effective strategies for detecting low density infections. A recently deployed, field-tested, simple, and cost-effective molecular diagnostic strategy for detecting pre-existing and intervening Plasmodium infections from dried blood spots (DBS) in malaria-endemic settings is discussed to inform new clinical trial designs. Strategies that combine sensitive molecular diagnostic techniques with convenient DBS collections and cost-effective pooling strategies may enable more thorough and informative infection monitoring in upcoming malaria clinical trials and epidemiological studies.

Keywords: 18S rRNA; Plasmodium falciparum; at-home DBS; clinical trial; intervening infection; pre-existing infection.

Figure 1

Proposed testing strategy for more frequent DBS collections with pooled qRT-PCR. In this theoretical vaccine clinical trial scenario, enrollment and vaccination take place over the first four months of the study followed by a four-month efficacy phase during the transmission season as shown. In addition to the typical whole venous blood collections (test tube icons), more comprehensive testing can be achieved by adding repeated DBS collections during the intervening time periods (small drop of blood icons). These DBS collections could be at-home or in the clinic as needed. The number of collections could be adjusted to a daily frequency or to less frequent collection as needed. The inset table shows the number of samples collected if only venous blood was specified (‘Traditional Design’) or if venous blood and DBS were collected (‘Added DBS spots’) and then calculates the minimum and maximum number of qRT-PCR tests that would need to be tested to determine each participant’s infection status; this calculation assumes a first qRT-PCR pool size of n = 10. The minimum number of runs would occur if all three pools were negative, whereas the maximum number of runs would occur if all three pools were positive and required deconvolution.