| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Efficacy of pre-erythrocytic vaccines can be determined in Phase IIa trials by sporozoite challenge.2 In trials conducted at the Center for Clinical Vaccinology and Tropical Medicine, Oxford University, UK, malaria-naïve vaccinated and non-vaccinated control volunteers received bites from five Plasmodium falciparum–infected mosquitoes and were monitored for 21 days to detect blood stage parasites.
The trials described here use a heterologous prime-boost strategy designed to induce T-cell responses against the liver stage of the P. falciparum life cycle. Viral (modified vaccinia virus Ankara [MVA] and an attenuated fowlpox strain [FP9]) and plasmid DNA vaccine vectors encoding P. falciparum antigens have been used in various combinations. RTS,S, a protein/adjuvant vaccine3 that induces high levels of antibodies and low T-cell responses and is partially protective against infection,4 was used in prime-boost combinations with MVA expressing the same P. falciparum antigen. A second protein/ adjuvant vaccine, ICC-1132,5 was also tested. Safety, immunogenicity, and efficacy data from vaccine trials included in this paper have been published elsewhere6–8 (Dunachie and others, unpublished data).
For more than 100 years, the gold standard method of malaria diagnosis has been by blood film examination1 with a predicted limit of detection of 5–20 parasites per microliter of blood. Microscopy gives highly variable results and has poor sensitivity when compared with quantitative real-time PCR.9 Previous investigations have established that errors in microscopy do occur and that false blood film results can affect clinical trial outcomes.10 This is especially important when the trial end point is the observation of one parasite on a blood film. PCR analysis can be used to verify blood film readings. Further, from quantitative PCR results generated over the course of a trial, it is possible to estimate the average number of infected hepatocytes for each vaccine regimen and compare regimens based on more precise trial end points.9
A number of PCR methods to detect P. falciparum–infected erythrocytes have already been tested,11–13 and malaria diagnosis PCR kits are available. Here we report the use of a real-time PCR method in clinical trials and compare it to alternative approaches to malaria diagnosis.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sample preparation: Plasmodipur filter method. Leukocytes were removed from the blood by filtration. Samples were centrifuged (5 minutes, 2,500 rpm, Beckman GS-6K centrifuge, Beckman Coulter, Inc., Fullerton, CA) and the sample volume marked on each tube. Plasma was removed and 2.5 mL sterile PBS added. The suspension was mixed by inversion and passed through a Plasmodipur filter (Euro Diagnostica BV, Arnhem, The Netherlands). The vacutainer tube and Plasmodipur filter were washed by adding 2.5 mL PBS to the vacutainer, mixing, and filtering the suspension into the collection tube, then repeating the wash. The sample was centrifuged as before and supernatant removed. The pellet was resuspended in 2.5 mL sterile PBS and the sample returned to its original vacutainer tube. The collection tube was washed with 1.5 mL PBS and this suspension added to the vacutainer tube. After centrifugation as before, the supernatant was removed and PBS added to return the cell pellet to the original sample starting volume, as indicated by the mark on the tube. The blood samples were mixed well by inversion, 0.5 mL was taken for immediate DNA extraction, and the remainder was stored at –80°C.
Sample preparation: Multiwell plate filter method. A second filter method was tested using 24-well plates containing two layers of a glass-fiber based material (Whatman VFE) (Whatman International, Clifton, NJ; product code 7700-9902). Plasma was removed as above. The cells were then resuspended with 1 mL PBS and mixed by inversion. The 24-well filter plate was positioned above a 24-well collection plate in a UniVac3 vacuum manifold (both from Whatman International) and samples added to each well. The vacutainer tubes were washed with 1 mL PBS, and this was poured into the corresponding sample well before drawing the sample through the filter by vacuum. Vacutainer tubes were washed again with 1 mL PBS and blood solutions filtered in the appropriate well. Filter wells were washed with 0.5 mL PBS, and this was filtered through. Samples were transferred back to their original vacutainer tubes, centrifuged (5 minutes, 2,500 rpm, Beckmann GS-6K), and returned to the original blood sample volumes by removing excess supernatant. The blood samples were mixed by inversion, 0.5 mL was taken for immediate DNA extraction, and the remainder stored at –80°C.
DNA extraction. The QIAamp DNA Blood Mini kit (QIAGEN, Crawley, UK) was used for DNA extraction with some modifications to the standard protocol. The sample volume was 0.5 mL filtered blood. Volumes of protease, lysis buffer, and ethanol were 40 µL, 400 µL, and 400 µL, respectively, and wash buffer volumes increased to 750 µL. The second wash buffer allowed to soak on the DNA purification columns for 2 minutes before the vacuum was applied. Columns were transferred to collection tubes and centrifuged at 13,000 rpm for 4 minutes. DNA was eluted with 50 µL sterile 10 mM Tris, pH 8.0, incubating for 1 minute at room temperature before centrifugation (1 minute, 8,000 rpm) to collect the DNA sample.
Quantitative real-time PCR. Parasitemia was detected using the Rotor-Gene 3000 real-time PCR machine (Corbett Research, Sydney, Australia) and Quantitect SYBR Green I chemistry (QIAGEN). The PCR amplified the multicopy 18S (small subunit) ribosomal RNA genes of P. falciparum with the following primers to give a 174-bp product: 18SRank1F (5'-GTTCTGGGGCGAGTAT-3') and 18SRank1R (5'-TGCATCACCATCCAAG-3'). Final concentrations of 1x Quantitect SYBR Green PCR Master Mix, 0.3 µM of each primer, and 4 mM MgCl2 were used in a 25-µL reaction with 5 µL sample or PCR control. Each PCR run included four negative controls (water) and four positive control replicates of a 1 in 8,000 dilution of 6% parasitemia P. falciparum 3D7 DNA. A standard curve was produced using known numbers of ring-stage 3D7 strain P. falciparum parasites diluted in whole blood to give concentrations of approximately 2 x 106, 2 x 105, 6 x 104, 2 x 104, 6,000, 2,000, 200, and 20 p/mL. These were filtered and DNA extracted by the method described. Standard curves were generated using both filter methods. Quantitation was achieved by importing this curve into subsequent PCR runs and adjusting for PCR efficiency using the positive control.
Comparison to blood film. Blood films were prepared from a 10-fold dilution series from a counted stock of 1.25 x 106 ring stage P. falciparum 3D7 parasites and aliquots taken for filtration, DNA extraction, and real-time PCR. The blood films were examined by an experienced microscopist blinded to the parasite numbers. Parasite counts were taken from 400 fields of view and multiplied by three (a whole film being 1,200 fields) or the whole film was counted. As 10 µL blood was used for each film, the total number of parasites per film was multiplied by 100 to produce an estimate of p/mL.
Commercial malaria detection methods. Commercially available rapid diagnostic tests (RDTs) for malaria were carried out on eight volunteers. The OptiMAL (Flow Incorporated, Portland, OR) and Rapimal (Cellabs Pty Ltd, Brookvale, NSW, Australia) immunochromatography dip-stick methods detect Plasmodium-specific lactate dehydrogenase (pLDH) and HRP-2 (histidine-rich protein II), respectively, in blood samples. Tests were carried out using the protocols included in each kit. The RealArt Malaria PCR kit (Artus GmbH, Hamburg, Germany) was tested on the Rotor-Gene 2000 with 5 µL of replicates of a 1 in 8,000 dilution of 6% parasitemia P. falciparum 3D7 DNA. The multiplex PCR uses probes labeled with FAM to detect parasites, with a positive internal control included, and was carried out in accordance with the kit protocol.
Stevor PCR. There are 28 polymorphic genes in the Stevor family, and primers were designed in the most conserved regions with degenerate bases to include polymorphisms.
p17X 5'-AAT CCA CAT TAT CAT AAT GAY CC-3'
p24x 5'-TTC ATA TKT TTY CAA TAA TTC TTT TT-3'
Parasitemia was detected using the Rotor-Gene 3000 real-time PCR machine and Quantitect SYBR Green I chemistry, with final concentrations of 1x Quantitect SYBR Green PCR Master Mix, 0.3 µM of each primer, and 4 mM MgCl2 in a 25-µL reaction with 5 µL sample or PCR control (1 in 8,000 dilution of 6% parasitemia P. falciparum 3D7 DNA).
TaqMan style PCR. Previously published P. falciparum 18S RNA primers with a TaqMan probe12 were tested on the LightCycler real-time PCR machine (Roche, Basel, Switzerland), using 5-µL replicates of a 1 in 8,000 dilution of 6% parasitemia P. falciparum 3D7 DNA in a 20-µL reaction volume. The TaqMan probe was labeled with a FAM reporter and 6-carboxy-tetramethylrhodamine (TAMRA) quencher. The results were compared with the same samples used with the 18S RNA primers designed here on both the LightCycler and Rotor-Gene 3000.
Vaccine regimens. The PCR method was used in clinical trials to test the efficacy of a variety of pre-erythrocytic vaccine constructs and strategies, including prime-boost vaccine regimes. The majority of vaccines used encode the TRAP (thrombospondin related adhesion protein) pre-erythrocytic antigen in different vectors: FP9 (an attenuated fowlpox strain), MVA (modified vaccinia Ankara), or DNA (in plasmid form). These have been described previously.6,8 Other vaccines tested include RTS,S/AS02, a protein vaccine containing the circumsporozoite antigen fused to hepatitis B surface antigen with a lipid adjuvant3 used in combination with MVA-CSO (encoding the circumsporozoite protein, codon optimized for humans), and ICC-1132 vaccine, comprising multiple epitopes of the P. falciparum circumsporozoite protein within a modified hepatitis B virus core particle.5
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). An inverse correlation between threshold cycle and initial DNA quantity was observed, calculated here as parasites per milliliter of blood.
|
Validation of the PCR method was accomplished by generating a number of standard curves over 1 year and demonstrating reproducibility by comparison of curve slope (average y value = –3.55296, range –3.5422 to –3.5588) and correlation coefficient (average R2 = 0.98286, range = 0.9615 to 0.9909). The curves were produced using two blood filtration methods for removal of leukocytes from blood while allowing erythrocytes that may contain parasites to pass through the membranes. This greatly reduced the concentration of human genomic DNA, which may otherwise hinder specific parasite PCR amplification at very low parasitemias. Both the single Plasmodipur and 24-well Whatman filters performed well, giving standard curves of y = –3.2865 and –3.5434, and R2 = 0.9751 and 0.9858, respectively. The 24-well filters were easier, quicker, and cheaper to use due to their 24-sample capacity.
Sensitivity.
PCR analysis on multiple replicates of a range of clinical trial samples was used to determine the level of detection (Figure 2). The replicates were spread over two PCR runs. Figure 2 shows that 20 p/mL can still be detected, but the increase in standard deviation between replicates is dramatic and erratic, making accurate quantitation at very low levels impossible. In addition, DNA was extracted from counted parasites diluted in whole blood at 10, 5, and 2 p/mL and replicates used for real-time PCR. Although the PCR could detect 10, 5, and 2 p/mL, quantitation was not accurate. Therefore, sample results below 20 p/mL were classed as negative. However, in 137 volunteers so far tested, only 73 time points have occurred with positive values below 20 p/mL (4.4% of samples).
|
Clinical trial results.
Figure 3 displays the quantitative PCR results of eight volunteers throughout a vaccine clinical trial. The effect of parasite sequestration can be clearly seen in the pattern of parasitemia, demonstrating the cyclic nature of P. falciparum infection. Flat lines show protected volunteers, with no parasites observed during the 21-day follow-up. Similar results were obtained for every volunteer who takes part in an Oxford malaria vaccine trial, with 137 volunteers having taken part to date in 7 trials in which monitoring was done by both blood film and PCR.
|
gives examples of these data, including both vaccinated and control volunteers. Protected volunteers are excluded from the data set as they did not have positive PCR or blood film results. Symptoms experienced were mainly mild in nature, and once experienced, typically continued until treatment was completed.
|
Comparison to commercially available rapid diagnostic tests.
Optimal and Rapimal RDTs (TCS Biosciences Ltd, Buckingham, UK) were carried out on eight volunteers in one clinical trial. Comparisons were made to the time of blood film and PCR positive results (with a PCR results of > 1,000 p/mL). Table 2 confirms that the PCR method described is more sensitive than these RDTs, identifying parasites in the blood on average 2.43, 2.88, and 2.75 days before Optimal, Rapimal, and blood film methods, respectively, with ranges of 2 to 5.5 days for the Optimal kit, and 0.5 to 4 days for both the Rapimal kit and microscopy. The Optimal kit failed to detect any parasites in one volunteer and the Rapimal kit in three volunteers. In four volunteers, the RDTs did not detect parasites until after treatment had begun. This is likely due to the treatment rupturing parasites and in turn releasing large quantities of the target proteins of the kits into the blood stream.
|
Vaccine efficacy.
Vaccine efficacy is typically determined by number of volunteers completely protected, plus delay to parasitemia compared with controls, as documented by blood film positive reading or PCR result greater than 1,000 p/mL. Figure 4 shows the geometric mean of first positive results for each regimen tested, with confidence intervals. A 22-day delay was used to include protected volunteers in the calculation, and rechallenged volunteers were omitted from this data set. For each method of diagnosis, regimen MRR (MVA-CS, RTS,S, RTS,S) proved to be the most effective vaccination strategy, followed by RRM (RTS,S, RTS,S, MVA-CS).
|
) and that there is little variation in efficacy between the other regimens. Further analysis of vaccine efficacy using more sophisticated statistical techniques to estimate the number of parasites released from the liver9 indicates a substantial reduction of infected hepatocytes in the FFM (92%) and RRM/MRR groups (97%). |
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Two methods of blood filtration were tested to remove leukocytes before DNA extraction, as excessive human DNA can cause inhibition of parasite gene amplification. Filtration greatly reduced quantities of human genomic DNA in each sample but did not completely remove it, as PCR amplification from human genes was still possible. The first filter method implemented, using single Plasmodipur filters, gave satisfactory results. However, due to a need for large number of samples to be processed quickly, a 24-well filter plate was designed and manufactured in conjunction with Whatman International Ltd. These plates have advantages of speed and ease of use, with a 70% reduction in cost per sample compared with single filters. These factors are important when samples are processed in "real-time" during clinical trials and are critical when trials include large numbers of volunteers.
Repeated generation of the standard curve over the course of 1 year, using separate batches of ring-stage 3D7 parasites and whole blood, showed that quantitation was reproducible within and between PCR runs, with no significant difference between the curves produced. Inaccurate initial counts in early parasite batches may explain slight variations in slope crossing points. Additional accuracy was later achieved by thin film parasite counts and RBC counts by hemocytometer on the parasite stock used to generate the standards.
Sensitivity/comparison to blood film. This real-time PCR method has a limit of detection of 20 p/mL. Although parasites are detectable below this level, the deviation in quantitative results is such that a reliable and accurate figure cannot be claimed beneath this point. Trials to date analyzing 1,661 PCR samples have shown that once positive, parasite numbers rarely fall below 20 p/mL, with only 4.4% of samples giving results under this level. There is, therefore, little need for a lower level of detection during monitoring of this type of trial.
The increased sensitivity of PCR over microscopy is shown by the failure of the gold standard method of malaria diagnosis to detect parasites of known concentrations below 1,250 p/mL and by the number of PCR positive time points before detection by blood film. In addition, the level of parasitemia at which parasites were detected by blood film is highly variable during challenge studies (lowest 53 p/mL, highest 36,556,164 p/mL).
These variations may be partially due to sample size. Each PCR reaction amplifies parasite DNA from 50 µL of blood—a volume five times larger than used for blood films—and as only 200 fields of view are examined by microscopy, the actual volume studied is only 1–2 µL blood. We have further evidence that parasites are lost from blood films during preparation, as parasite DNA can be amplified from slide wash buffers (Bejon and others, unpublished data).
Quantitative real-time PCR can provide a specific equivalent trial end point for all volunteers by the choice of a threshold level of parasitemia. In this report, a threshold greater than 1,000 p/mL was chosen for use in analysis.
Trial results. The results shown here are from experimental challenges on malaria-naïve volunteers, and not on travelers returning from malarious areas or for diagnosis of symptomatic malaria in endemic regions, for which other, usually less sensitive methods are available.13 In the studies reported here quantitative real-time PCR was completed within hours of taking blood samples rather than retrospectively on stored samples.13,14
Monitoring of blood stage P. falciparum began on the evening of day 6 postchallenge. Blood samples from Days 4 to 6 were analyzed in early trials but gave consistently negative PCR results. Parasites are thought to emerge from the liver between Days 6 and 7, when each sporozoite-infected hepatocyte can release about 30,000 merozoites into the bloodstream.15
The pattern of parasitemia shown in Figure 3 demonstrates the effect of sequestration, when Plasmodium-infected erythrocytes cytoadhere to vascular channels. This highlights the difficulty of malaria diagnosis by microscopy during these early stages of disease, as parasitemias only reach detectable levels during the cycle peaks.
Trial safety is increased by the use of PCR. If a volunteer shows clear signs of malaria, but has a negative blood film, in current trial protocols this is defined as an end point if more than one positive PCR result has been observed, and the volunteer is treated. The PCR results from the volunteer can still be included in the analysis, whereas in the absence of PCR it would not be possible to ascertain that the symptoms were due to malaria rather than another intercurrent infection. Delays in treatment of symptomatic but blood film negative volunteers can therefore be avoided. PCR results can also be used to estimate whether the next blood sample is likely to be blood film positive by reference to the observed peaks in parasitemia.
An increase in accuracy is also achieved by using these two protocols. Positive blood film results from volunteers who have yet to have a positive PCR can be reexamined both by repeated scrutiny of the film and by the taking of additional blood samples, while treatment can be delayed if the volunteer is asymptomatic. This can help eliminate misdiagnosis caused by false positive microscopy results.
In 137 volunteers tested so far, only 1 volunteer has shown evidence of PCR inhibition, due to the presence of excessive human genomic DNA after sample preparation. Dilution of the samples was effective in removing this inhibition, and parasite DNA could then be amplified. Using the malaria-negative Day-4 blood samples as controls, likelihood of this inhibition may be predicted by spiking these samples with low numbers of parasites and assessing the PCR results.
Not only can quantitative PCR results give more comparable end points between trial volunteers and provide support to the gold standard method of malaria diagnosis, but from these results the multiplication rates of parasites within different vaccine regimens can be calculated, addressing not only affect time to parasitemia but also speed of parasite growth. This has shown that the average rate of multiplication in 48 hours is 14-fold9 in these trials, compared with an 8-fold rate measured with retrospective PCR analysis from neurosyphilis infection studies.16 This difference in rate may indicate that at very low parasite densities, such as those measured here, parasite division rates are above those observed at high parasite densities, perhaps reflecting a lesser effect of immune response at low densities.
The PCR results also provide evidence that the delay in blood film parasitemia observed with some vaccination regimes is not simply due to the vaccines delaying the emergence of parasites from the liver. From the data presented here, there is no evidence that parasite release from the liver is delayed, but that smaller numbers of parasites are released, with no difference in parasite replication in the blood between vaccinees and controls.9
Symptoms.
One potential consequence of using this PCR in subsequent malaria vaccine clinical trials could be to allow earlier treatment of volunteers when a threshold of 1,000 p/mL is reached. This would normally provide at least three PCR positive results, reducing the risk of false positives or sample mix up. Volunteers treated at this PCR threshold would be treated 2.12 ± 0.2265 days earlier than treatment by microscopy, with a range of 0 to 5.5 days, which would be expected to reduce the number and severity of symptoms experienced by volunteers. Evidence for this is provided in Table 1, showing that few symptoms were seen before parasitemia reaches a density of > 1,000 p/mL.
RDTs. It has been claimed that rapid malaria diagnostic tests have a sensitivity of approximately 100 p/µL. Both the OptiMAL and Rapimal dipstick methods were tested during one clinical trial, where they proved to be less sensitive than both microscopy and real-time PCR. They were therefore not suitable for use in determining vaccine efficacy during experimental challenge.
Other PCR methods. The quantitative real-time PCR method described here is a closed PCR system, carried out in nonbreakable plastic tubes, giving a lower probability of contamination than two-step reverse-transcriptase RT-PCR assays, nested PCR, or PCR in glass tubes. The lowest malaria detection limit by PCR methods that do not use real-time technology is 5,000 p/mL13 compared with the 20 p/mL sensitivity of this assay. The commercially available RealArt Malaria PCR kit has a limit of detection of 100 p/mL when using the Rotor-Gene 3000. The RealArt kit also amplifies 18S ribosomal RNA genes in P. falciparum and detects amplification using a FAM-labeled probe. Previously published 18S rRNA primers with a TaqMan probe were tested.12 However, major problems were found with batch to batch variation of the TaqMan probe, and the method gave no increase in sensitivity but an increased cost, compared with our 18S rRNA primers alone.
We have carried out real-time PCR targeting the P. falciparum Stevor genes, using both primers alone and hybridization probes. There are 28 polymorphic genes in the Stevor family, and primers were designed in conserved regions. However, the Stevor PCR methods gave lower sensitivity when compared with primers targeting the five gene copies of 18s rRNA.
Vaccine efficacy.
PCR provides more information about vaccine efficacy than microscopy alone, providing quantitative data. Comparison of vaccine efficacy by PCR and blood film gives slightly different results (Figure 4), due to the variability in blood film positive end points. However, both methods conclude that the MMR and RRM regimens have been the most effective so far, followed by FFM. From the PCR data, the original number of infected hepatocytes and parasite multiplication rates have been calculated for a variety of vaccine regimens, also showing that RRM/MRR and FFM are the most effective to date.9 The method described here has been established for use in clinical trials, and although it has other possible applications, it is not designed as an all-purpose malaria diagnosis method, but to determine the efficacy of vaccination strategies. The cost of real-time PCR makes it unsuitable for general diagnosis, especially as almost all malaria occurs in the poorest regions of the world.
The method can provide information on vaccine efficacy in Phase IIb field studies and has been used to study 220 volunteers in The Gambia thus far.17 P. falciparum is the dominant malaria parasite in The Gambia.18 Resistance to P. vivax infection in Gambians due to their predominant Duffy blood group phenotype (FyFy) rules out the possibility of false PCR positives due to vivax infection.19,20 P. ovale and P. malariae are rare in West Africa,19 and studies in Equatorial Guinea show when these two species do appear, they are usually as mixed infections with P. falciparum.21 Blood films from the Gambian volunteers showed only the presence of P. falciparum during this field trial.
The PCR has also been used to study parasite growth rates after blood stage experimental challenge (Sanderson and others, unpublished data) and within the laboratory to determine P. falciparum and P. berghei infections in mosquitoes.
The addition of this real-time PCR method to Phase IIa malaria vaccine trials should significantly aid comparative evaluation of new vaccines by providing more precise estimates of vaccine efficacy with a safer, more accurate monitoring protocol.
Received September 28, 2004. Accepted for publication January 22, 2005.
Acknowledgments: We are grateful to Ian Poulton and Simon Correa for assistance with the clinical trials and to TCS Biosciences Ltd (Buckingham, UK) for the gift of the Optimal and Rapimal kits. This work was supported by the Wellcome Trust and Malaria Vaccine Initiative at PATH.
* Address correspondence to Sarah C. Gilbert, Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford, OX3 7BN, United Kingdom. E-mail: sarah.gilbert{at}well.ox.ac.uk 
Authors’ addresses: Laura Andrews, Rikke F. Andersen, Adrian V. S. Hill, and Sarah C. Gilbert, Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford, OX3 7BN, UK, Telehone: +44 1865 287500, Fax: +44 1865 287686. Daniel Webster, Susanna Dunachie, R. Michael Walther, Philip Bejon, Angela Hunt-Cooke, Gillian Bergson, and Frances Sanderson, Centre for Clinical Vaccinology and Tropical Medicine, Churchill Hospital, University of Oxford, Oxford, OX3 7LJ, UK, Telehone: +44 1865 857401, Fax: +44 1865 857471.
Reprint requests: Sarah C. Gilbert, Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, UK.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Sanderson, L. Andrews, A. D. Douglas, A. Hunt-Cooke, P. Bejon, and A. V. S. Hill Blood-stage Challenge for Malaria Vaccine Efficacy Trials: A Pilot Study with Discussion of Safety and Potential Value Am J Trop Med Hyg, June 1, 2008; 78(6): 878 - 883. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. B. IMOUKHUEDE, L. ANDREWS, P. MILLIGAN, T. BERTHOUD, K. BOJANG, D. NWAKANMA, J. ISMAILI, C. BUCKEE, F. NJIE, S. KEITA, et al. LOW-LEVEL MALARIA INFECTIONS DETECTED BY A SENSITIVE POLYMERASE CHAIN REACTION ASSAY AND USE OF THIS TECHNIQUE IN THE EVALUATION OF MALARIA VACCINES IN AN ENDEMIC AREA Am J Trop Med Hyg, March 1, 2007; 76(3): 486 - 493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Conway Molecular Epidemiology of Malaria Clin. Microbiol. Rev., January 1, 2007; 20(1): 188 - 204. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. ADEGNIKA, J. J. VERWEIJ, S. T. AGNANDJI, S. K. CHAI, L. PH. BREITLING, M. RAMHARTER, M. FROLICH, S. ISSIFOU, P. G. KREMSNER, and M. YAZDANBAKHSH MICROSCOPIC AND SUB-MICROSCOPIC PLASMODIUM FALCIPARUM INFECTION, BUT NOT INFLAMMATION CAUSED BY INFECTION, IS ASSOCIATED WITH LOW BIRTH WEIGHT Am J Trop Med Hyg, November 1, 2006; 75(5): 798 - 803. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Walther, J. Woodruff, F. Edele, D. Jeffries, J. E. Tongren, E. King, L. Andrews, P. Bejon, S. C. Gilbert, J. B. De Souza, et al. Innate Immune Responses to Human Malaria: Heterogeneous Cytokine Responses to Blood-Stage Plasmodium falciparum Correlate with Parasitological and Clinical Outcomes J. Immunol., October 15, 2006; 177(8): 5736 - 5745. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Dunachie, M. Walther, J. E. Epstein, S. Keating, T. Berthoud, L. Andrews, R. F. Andersen, P. Bejon, N. Goonetilleke, I. Poulton, et al. A DNA Prime-Modified Vaccinia Virus Ankara Boost Vaccine Encoding Thrombospondin-Related Adhesion Protein but Not Circumsporozoite Protein Partially Protects Healthy Malaria-Naive Adults against Plasmodium falciparum Sporozoite Challenge. Infect. Immun., October 1, 2006; 74(10): 5933 - 5942. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Walther, F. M. Thompson, S. Dunachie, S. Keating, S. Todryk, T. Berthoud, L. Andrews, R. F. Andersen, A. Moore, S. C. Gilbert, et al. Safety, Immunogenicity, and Efficacy of Prime-Boost Immunization with Recombinant Poxvirus FP9 and Modified Vaccinia Virus Ankara Encoding the Full-Length Plasmodium falciparum Circumsporozoite Protein. Infect. Immun., May 1, 2006; 74(5): 2706 - 2716. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
