xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ka 4500
controlfield tag 001 002029031
007 cr mnu|||uuuuu
008 090915s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002858
Schilke, Jessica L.
Artemisinin-based combination therapy (ACTs) drug resistance trends in Plasmodium falciparum isolates in Southeast Asia
h [electronic resource] /
by Jessica L. Schilke.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 57 pages.
Thesis (M.S.P.H.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Plasmodium falciparum, one of the parasites that cause clinical malaria, is a continuous public health concern, especially in Asia and Africa. Unfortunately, the parasite has developed resistance to many drugs created to treat and prevent the disease. Artemisinin and its derivatives are the new gold standard for treatment of malaria, yet treatment failures in clinical studies are starting to be reported. Clearly, artemisinin resistance needs to be characterized and dealt with accordingly. In support of the Gates Foundation Artemisinin Consortium, we conducted a blinded study to elucidate the phenotypic response of artemisinin derivatives of parasites derived from patient blood samples from Cambodia and Thailand. Blood samples containing Plasmodium falciparum were cultured and then assayed using SYBR green as an indicator to obtain drug IC50s. The data suggested that many isolates are not demonstrating resistance to artemisinin. However, a select few are showing some resistant characteristics in the form of elevated IC 50s, especially to some of the drugs already identified in previous studies as drugs having resistant characteristics. Compared to studies conducted within the past ten years, no significant changes in parasite susceptibility to the artemisinin drugs have been observed. Additional analysis of clinical outcomes, therapeutic drug levels, and molecular markers needs to be completed before it can be assumed that artemisinin resistance has emerged.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Dennis Kyle, Ph.D.
x Global Health
t USF Electronic Theses and Dissertations.
Artemisinin-Based Combination Therapy (ACTs) Drug R esistance Trends in Plasmodium falciparum Isolates in Southeast Asia by Jessica L. Schilke A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Public Health Department of Global Health College of Public Health University of South Florida Major Professor: Dennis Kyle, Ph.D. Wilbur Milhous, Ph.D. John Adams, Ph.D. Date of Approval: April 10, 2009 Keywords: SYBR green, Plasmodium falciparum resistance, ACTs, Artemisinin Copyright 2009, Jessica L. Schilke
Acknowledgments Thank you to Dr. Kyle, Dr. Adams, and Dr. Milhous f or their guidance and support during my thesis writing process. Also, a huge thank you to everyone in the Â“Kyle labÂ”, who went out of their way to help me le arn the methods of malaria culture. Lastly, thank you to my family and friends for thei r continuous support, encouragement, dog sitting, and apartment cleaning. I could never have successfully accomplished this without them.
Table of Contents List of Tables iii List of Figures iv List of Symbols and Abbreviations v Abstract vi Introduction 1 Malaria 1 Life cycle 1 Clinical features 2 Artemisinin 3 Artemether 5 Artesunate 5 Dihydroartemisinin 6 Artemisinin Combination Drugs (ACTs) 6 Lumefantrine 6 Atovaquone 6 Mefloquine 7 Piperaquine 8 Anti-Malarial Drug Resistance Patterns 8 ACT Resistance Studies 11 Purpose 13 Methods 16 Obtaining Samples 16 Thawing of Isolates 16 Culturing 17 Freezing Protocol 18 Plasma, Media, and Blood Preparation 19 Slide Preparation 20 Drug Assay Preparations 20 Isolate Suspension Assay 20 SYBR Green Assay 21 Isobologram Assay 22 Data Analysis 23 Results 32 Cell Culture 32 Single Drug Assay Data 32 Isobologram Assay Data 33 Discussion 38 Findings 38
Strengths and Weaknesses 39 Future Study Plans 40 List of References 41 Bibliography 43 Appendices 44 Appendix A 45 Appendix B 50
List of Tables Table 1 IC 50s (ng/ml) of Thai Isolates in the 72 Hour SYBR Green Drug Assay 35 Table 2 IC 50s (ng/ml) of Cambodia isolates in the 72 Hour SYBR Green Drug Assay 36 Table 3 IC 50s (ng/ml) of Reference Strains W2, D6 and C2B in the 72 Hour SYBR Green Drug Assay 37
List of Figures Figure 1 Life Cycle of Plasmodium falciparum 14 Figure 2 Chemical Structures of the 8 Compounds Tes ted 15 Figure 3 Map Depicting Isolate Collection Sites 24 Figure 4 6 Well plates Containing Isolates 25 Figure 5 Isolates Growing Within Tissue Culture Fla sks 26 Figure 6 Geimsa Stain of Isolates 27 Figure 7 96 Well Assay Plate Setup (Drug Name and C oncentration in ng/ml) 28 Figure 8 Biomek 3000 (Robot) Used to Complete Drug Assays 29 Figure 9 Plate Reader for SYBR Green Assay 30 Figure 10 Incubator Modulators Containing 6 Well Pl ates (Right) and 96 well Plates (Left) 31
List of Symbols and Abbreviations Symbol and Abbreviation Description QHS Artemisinin DHA Dihydroartemisinin AM Artemether AS Artesunate ATO Atovaquone FIC Fractional Inhibitory Concentration FIC Fractional Inhibitory Concentration Sum MFQ Mefloquine LF Lumefantrine PIP Piperaquine C Degrees Centigrade IC50 50% Inhibitory Concentration nM Nanomolars ng/ml Nanograms Per Milliliter ul Microliter RBC Red blood cells ACT Artemisinin Combination Therapy
Artemisinin-Based Combination Therapy (ACTs) Drug R esistance Trends in Plasmodium falciparum Isolates in Southeast Asia Jessica L. Schilke ABSTRACT Plasmodium falciparum one of the parasites that cause clinical malaria, is a continuous public health concern, especially in Asi a and Africa. Unfortunately, the parasite has developed resistance to many drugs cre ated to treat and prevent the disease. Artemisinin and its derivatives are the new gold st andard for treatment of malaria, yet treatment failures in clinical studies are star ting to be reported. Clearly, artemisinin resistance needs to be characterized and dealt with accordingly. In support of the Gates Foundation Artemisinin Consortium, we conducted a b linded study to elucidate the phenotypic response of artemisinin derivatives of p arasites derived from patient blood samples from Cambodia and Thailand. Blood samples containing Plasmodium falciparum were cultured and then assayed using SYBR green as an indicator to obtain drug IC50s. The data suggested that many isolates are not demonstrating resistance to artemisinin. However, a select few are showing som e resistant characteristics in the form of elevated IC 50s, especially to some of the drugs already identified in previous studies as drugs having resistant characteristics. Compared to studies conducted within the past ten years, no significant changes in parasite susce ptibility to the artemisinin drugs have
been observed. Additional analysis of clinical out comes, therapeutic drug levels, and molecular markers needs to be completed before it c an be assumed that artemisinin resistance has emerged.
Introduction Malaria Malaria is one of the most problematic tropical dis eases worldwide. Endemic to Parts of South and Central America, Africa, and Sou theast Asia, half of the world is at risk for contracting malaria. ItÂ’s responsible for over 300 million illnesses per year and an estimated three million mortalities (WHO, 2008). One malaria related death occurs approximately every 30 seconds, mainly in children and pregnant women. Malaria is caused by a parasite in the genus Plasmodium and five different species affecting humans have been discovered thus far. Of those fiv e species, Plasmodium falciparum by far causes the most morbidity and mortality. Ther e is no vaccine for malaria, but there are treatment options available. Unfortunately, dr ug resistance has rapidly become a major issue in controlling malaria, and has been id entified in both Plasmodium faciparum and Plasmodium vivax (CDC, 2008). This thesis attempts to classify som e of the resistance trends to some currently common treatmen t options in Southeast Asia, specifically in Thailand and Cambodia. Life Cycle The definitive host of the malaria parasite is the female anopheline mosquito (John, 2006). The female mosquito takes a blood me al, and injects sporozoites located in her salivary glands into her host (see Figure 1). These sporozoites then travel from the bloodstream to hepatocytes. Asexual reproduction ( schizogony) then occurs in the liver.
Schizogony occurs immediately in P. falciparum and results in the formation of merozoites, which move into the blood stream due to the rupturing of hepatocytes. The merozoites invade RBCs, and erythrocytic schizogony follow, producing anywhere from 2 to 36 merozoites per infected red blood cell. Th e erythrocyte eventually ruptures, releasing more merozoites into the blood stream. Us ually after the patient becomes ill, some of the merozoites form male and female gametoc ytes, which can then be taken up by the mosquito during the blood meal process. Onc e in the mosquito stomach, these gametocytes mature into micro and macrogametes. Th e male microgamete is then capable of fertilizing the female macrogamete, whic h forms into an ookinate. The ookinate then becomes an oocyst just outside of the mosquitoÂ’s stomach. The oocyst produces sporozoites, which can detach and find the ir way into the salivary glands of the mosquito. The sporozoites are then injected into t he mosquitoÂ’s host when the female mosquito takes a blood meal. This thesis examines the asexual erythrocytic stages of Plasmodium falciparum in vitro under drug pressure. Clinical features Plasmodium falciparum is the malaria species that causes the most morbid ity and mortality in humans. The incubation period in the body is the shortest, only about 8-11 days, and if left untreated, could last anywhere fr om 6-17 months. Fever in P. falciparum can be Â‘continuous, remittent or quotidianÂ” (John, 2006). Prodromal symptoms include aches and pains, photophobia, nausea and even vomit ing. The paroxysm stage begins with the symptoms of feeling severely cold, even th ough fever is present. Ten to fifteen minutes later, the patient begins to feels extremel y warm and becomes flushed and agitated. Restlessness and delirium can also accom pany this stage. This can last hours
and is followed by a sweating phase. The paroxysm stage in P. falciparum patients can occur mere hours after the previous episode. Other symptoms include splenomegaly, anemia, capillary obstruction, decreased blood flow hypoxia, renal failure and even death in cases of cerebral malaria. Blackwater fever is yet another complication of P. falciparum which occurs from severe hemolysis within the patientÂ’s veins and is usually associated with the use of quinine (Rieckmann, 2006). Other less common compl ications include Dysenteric Malaria, Algid Malaria, Pulmonary Edema, Tropical S plenomegaly Syndrome and Hypoglycemia. Pathogenesis in humans is caused by a variety of factors, including hemolysis of red blood cells, red blood cells stick ing to the endothelial cells of the vascular system and decreasing blood flow, the pati entÂ’s own immune response, parasite metabolites released into the body, as well as the malaria pigment (Rieckman, 2006). Artemisinin Artemisinin is a lactone derived from the leaves of Artemisia annua, or sweet wormwood. It is also known to the Chinese as qingh ao, and has been used to treat ailments such as hemorrhoids and fever for over a t housand years. ChinaÂ’s government later decided to try to join traditional Chinese me dicine with western medicine, and a program to screen traditional Chinese remedies for activity began in 1967. It was first proposed that the mechanism of action of artemisinin was due to the endoperoxide bridge, which generates free radicals. It has been thought that the malaria parasite has been susceptible to free radicals. To test the theory that free radicals were involved in the mechanism of action, tests were con ducted in the 1980s and it is was found that free radical generators promoted the act ivity of artemisinin against malaria,
and free radical scavengers were found to inhibit t he activity. Free radicals were also found to inhibit a calcium adenosine triphoshatase PfATPase 6 (Sherman, 1998). Artemisinin is active against the blood schizont st age of the malaria parasite, and reduces parasite load in all malaria species. It is uniq ue in that it is effective in killing all asexual blood stages of the malaria parasite as wel l as gametocytes in P. falciparum Before artemisinin was discovered, the only drug th at killed gametocytes was primaquine. The highest drug concentrations in the plasma occur around 3 hours when administered orally, and around 11 hours when given in a suppository form. It has a molecular weight of 282.3; the elimination half lif e is around 1 hour. Finally, it is converted into inactive metabolites via enzymes inc luding cytochrome P450 and CYP2B6. Side effects include dizziness, gastrointestinal p roblems, reticulocytopenia, neutropenia, and infrequently electrocardiographic abnormalities. However, the artemisinin drugs do not cause nearly as many side effects in humans as some of the other antimalarials, and there are currently no known dru g interactions, making it an almost ideal treatment option. One in 3000 patients repor t hypersensitivity reactions, which is the only serious side effect reported. The recommended use of artemisinin, as well its de rivatives, artemether, artemotil, artelinic acid, dihydroartemisinin, arte mether and artesunate, is in combination with other antimalarials in order to reduce the pot ential emergence of resistance (WHO, 2006).
Artemether The methyl ether form of dihydroartemisinin is arte mether. With a molecular weight of 298.4, peak concentrations of drug in the plasma occur around two to three hours after it is taken orally. It can also be int roduced into the body by being injected directly into the muscle of the patient, but this m ethod causes extreme variability in absorption amount, with peak times ranging from 6 t o over 18 hours. However, when injected, artemether itself is the predominate form of the drug, whereas when administered orally, dihydroartemisinin is the arte misinin derivative found most readily throughout the body. The elimination half life is about 1 hour when given orally but can be longer when injected. Biotransformation occurs via CYP3A4, a cytochrome p40 enzyme. Side effects are very similar to artemisinin, howe ver, it was found in animals to cause damage to the neurons when given intramuscula rly. This is due to the fact that high concentrations of drug are within the blood wh en given in this manner. There are also no known drug interactions with artemether. Artesunate The sodium salt of the hemisuccinate ester of artem isinin is artesunate. It is water soluble, and can be administered orally, intravenou sly, as well as rectally. Peak plasma levels happen at around 1.5 hours after oral admini stration, 2 hours after given rectally, and one half hour after given intramuscularly. The form found most frequently in the body is dihydroartemisinin. Toxicity is again simi lar to artemisinin, and there are no currently known drug interactions.
Dihydroartemisinin An active metabolite of artemisinin is known as dih ydroartemisinin. It can be given both in oral and rectal forms, and its effect iveness can be compared to oral administration of artesunate. Peak plasma levels o ccur at around 2.5 hours when given orally, and slightly slower (around four hours) whe n given rectally. Toxicity is similar to that of artemisinin and there are no currently know n drug interactions. Artemisinin Combination Drugs (ACTs) Lumefantrine Lumefantrine is a drug that is co administered oral ly with artemether. It needs to be consumed with fatty foods which increases absorp tion by 108%. Highest levels in the plasma occur at around 10 hours after administering There is no toxicity reported with lumefantrine, and the only side effects reported in clude headache, dizziness, nausea, and discomfort of the abdomen. It has a molecular weigh t of 528.9. Atovaquone Atovaquone is a drug that is known to interfere wit h the electron transport chain of the parasite. It is effective against all Plasmodium species, and also inhibits oocyst development within the mosquito, as well as develop ment in the liver before it invades red blood cells. It is has been available since th e early 1990s in combination with proguanil in tablet form (Malarone). Similar to lu mefantrine, absorption is increased when taken with food. The half life is around 66-7 0 hours due to enter hepatic recycling, however pregnancy can reduce its half life. Mild side effects reported include diarrhea, vomiti ng, insomnia, skin rash, fever, increased liver enzymes, and sometimes anemia and n eutropenia. When administered to
n patients taking drugs such as tetracycline, metoclo pramide, as well as benzodiazepines, cephalosporins, laxatives, opiods and paracetamol, reduced concentrations in the plasma may occur. It also decreases metabolism of the dru gs zidovudine and cotrimoxazole. Atovaquone/proguanil (Malarone) is usually successf ul after a three day course, however it is not widely used due it its high costs. (Rieck man, 2008). It can also be used as a prophylactic for travelers and military personnel. Malarone is also used to treat travelers not endemic to malarias areas, and is approved for use in North America and Europe (CDC, 2008). Studies examining patient isolates alo ng the Thai-Myanmar and ThaiCambodia border have shown that resistance to atova quone is not yet an issue. It has, however, been reported that the point mutation on c odon 268 on the cytochrome B gene is responsible for treatment failures that have thu s far been reported (Khositnithikul, 2008). This point mutation is key in identifying a tovaquone resistant blood isolates. Mefloquine Mefloquine is effective against all Plasmodium species and came about as a result of screening a massive amount of drugs for antimala rial activity during the 1960s (Reichmann, 2008). It is a 4-methanolquinoline rel ated to quinine, and is given orally as the hydrochloride salt. It is given both as a prop hylactic to travelers, as well as a disease treatment. It is only slightly water soluble, but is completely soluble in alcohol. Since mefloquine has varying time to achieve peak levels in the plasma, it is administered in two separate oral doses 6-24 hours apart in order t o improve absorption. Enterohepatic recycling occurs with mefloquine. It can be co adm inistered with artesunate, which increases concentrations in the blood. Unfortunate ly, side effects are common, including nausea vomiting diarrhea anorexia, abdominal pain, headache, dizziness, dysphoria and
various sleep disturbances such as insomnia and str ange dreams. Less commonly, seizures, encephalopathy and psychosis can occur, a s well as the occasional case of hair loss, skin rash, purities, muscle weakness and prob lems with liver function. Some cardiovascular effects have also occurred when mefl oquine was given. There are also possible risks of drug interactions. These include arrhythmias when taken with beta blockers, calcium channel blockers, amiodarone, pim ozide, digoxin or antidepressants. When taken with chloroquine or quinine, convulsion risk is possible as well. In order to minimize recrudescence and due it its low half life and ability to clear parasites, mefloquine is given in combination with artesunate. Unfortunately, some cross resistance between mefloquine and artesunate has be en reported in Thailand, as well as mefloquine resistance (Wongsrichanalai, 1999). Res earchers have identified an amplification of the gene pfmdr1 that has been link ed to mefloquine resistance (Price, 2004). Piperiquine Piperiquine, otherwise known as bischloroquine, is given in combination with dihydroartemisinin to treat uncomplicated malaria. Piperiquine has a long half life, so it not only able to treat uncomplicated malaria, but a lso to provide protection for reoccurrence. The side effects are minimal, and th e combination drug is inexpensive to dispense. (Tran, 2004) Anti-malarial Drug Resistance Patterns Most of the drug resistance towards anti-malarials was first documented and reported along the south-eastern border of Thailand and Cambodia. For example, resistance began here in the 1950s to chloroquine, followed by the anti-malarials
sulfadoxine-pyrimethamine and mefloquine. This is why the isolates collected were chosen from this location. Drugs that were once Â“g old standardsÂ” to effectively treat patients infected with malaria are now almost compl etely ineffective, the origins of resistance being in Southeast Asia. Therefore, ne wer drug classes are being developed in order to stay one step ahead of the parasite. The World Health Organization has made ACTs the mos t widely used and effective drugs worldwide (CDC.gov). A drug combin ation of mefloquine and artesunate has replaced all previous drugs and been the main t reatment prescribed ever since 1995. However recently, failure rates with artemisinin dr ugs in the clinic have been reported, specifically along the Thai-Cambodian border. If r esistance truly is occurring, control measures must be started in order to characterize a nd contain the resistance so that it does not become widespread, as in the case of past gold standard malarial treatments. First, it is important to discern between drug resistance and Â“treatment failureÂ”. Drug resistance is defined as Â“the ability of a par asite strain to survive, and/or multiply despite the proper administration and absorption of an antimalarial drug in the dose normally recommendedÂ” (Sherman, 1998). Treatment f ailure, on the other hand, is Â“failure to clear malarial parasitemia and/or resol ve clinical symptoms despite the administration of antimalarials (Sherman, 1998). I t also needs to be understood that drug resistance can indeed lead to treatment failure. H owever, treatment failures can be caused by phenomenons other than drug resistance, i ncluding administering the incorrect dose and/or drug, compliance, drug quality not bein g up to standards, drug-drug interactions, issues with absorption of the drug, a nd incorrect diagnoses. In order to combat some of these issues with malaria treatment that could lead to failure and possibly
r resistance, the Â“Roll Back MalariaÂ” strategy has be en created in order to enforce proper compliance with treatment (Ikeoluwapo, 2008). At l east 18 countries in Africa have adopted this strategy, and 44 countries in Africa a re now using artemisinin combination therapies (ACTs) as a first choice to treat uncompl icated malaria. Enforcing proper use of ACTs isnÂ’t as much of an issue in healthcare fac ilities as it is on a community level, which facilities drug failure and resistance. Coun terfeit treatments which do not contain the correct dose also tend to leak into communities facilitating resistance. Drug resistant malaria strains can arise in a vari ety of ways. The first event leading to resistance involves a change in the para sites genetic code, following by a process of selection that allows the new resistance mutants to survive and take over as the dominant parasite (Sherman, 2008). Factors that as sist in resistance development include the amount of genetic mutations that are occurring in a certain region and the number of parasites that are exposed to the drug. In additio n, the amount of immunity in an individual with the disease under drug pressure and the pharmacokinetics are also involved in resistance (Sherman, 2008). Therefore, it is imperative to administer the correct amount of drug needed to kill the parasite, so that the strongest parasites arenÂ’t the only ones that survive and multiply. Again, antima larial drug distribution and dosing isnÂ’t always properly regimented in some areas of t he world. Thus, these issues commonly arise. Since lumefantrine, atovaquone and mefloquine are absorbed at varying rates due to their lipophilic properties, they are more prone to resistance, since the peak drug concentrations rely heavily on absorption. Unfortunately, for reasons still unknown, one of the main issues with the artemisinin drugs is the tendency for patients to r ecrudesce (Giao, 2001). One to four
weeks after being treated with an artemisinin drug, 10%-40% of patients experience recrudescence. Parasites can become undetectable a fter a single day of artemisinin therapy, but patients can recrudesce if not repeate d treated. One attempt to decrease the risk for resistance includes giving antimalarials i n combination forms. The hope here is that the drugs have unique mechanisms of actions an d therefore the way that resistance would arise would vary. This issue of recrudescenc e is also cause for alarm when looking at ACT resistance rates, and discerning bet ween the recrudescent infections of P. falciparum and actual resistant strains. In addition, the sh ort half life of artemisinin makes it more difficult to monitor proper administr ation of the drug to patients, which could also lead to more unclassified resistance. A ll of these issues combined could possible lead to widespread artemisinin resistance. ACT resistance studies Due to the unusually high incidence of drug resista nce in Thailand and Cambodia, and some high clearance and treatment failure rates that were being clinically reported with some ACT use, studies were conducted to examin e patient susceptibility to current ACTs. The Army component of the U.S. Armed Forces Research Institute of Medical Sciences conducted two studies in 2005 and 2006 to determine if drug resistance was the culprit to these clinical failures. The first studi es led to a conclusion that 16.6% of patients failed therapy when given a drug combinati on of mefloquine and artesunate (Penh, 2007). A second study was then conducted ad ministering artesunate alone. In vitro results showed artemisinin IC50s were higher along the Thai-Cambodian border, versus other areas where the same treatment was adm inistered. A conclusion was made that preliminary evidence had been gathered that ar temisinin resistance had emerged
along the Thai-Cambodian border. Recently, another study was published in the New England Journal of Medicine that revealed out of 60 people receiving artesunate treatment, two were classified as having an artemis inin resistant infection (Noedl, 2008). Although this is a small amount of people, it is st ill concerning, as resistance can spread rapidly if not contained. The most concerning data from this study was the prolonged clearance times for parasitemia, which significantl y increased for all previous studies of ACTs.
Purpose The purpose of this thesis was to culture and then conduct blinded phenotypic analyses to identify possible ACT susceptibility tr ends in isolates infected with Plasmodium faciparum from regions in both Thailand and Cambodia. These trends were observed both for single drugs, as well as in combi nation form. Plasmodium falciparum isolates also were cultured and frozen for use in f uture ACT resistance experiments. Characterization of drug resistance is key in order to control the spread of the disease, containment of resistance, as well as successful tr eatment of the malaria parasite.
Figure 1: Life Cycle of Plasmodium falciparum
Figure 2: Chemical Structures of the 8 Compounds Te sted Atovaquone Artemisinin Artesunate Lumefantrine Artemether Piperaquine Dihydroar temisinin
Methods Obtaining samples Isolate samples were collected from both Maesod in Thailand and Palan on the Thai-Cambodian border, cryopreserved and shipped to the University of South Florida in Tampa. Reference strains C2B, W2 and D6 were used in this experiment in order to compare the blind isolate data. C2B is a P. falciparum strain that is resistant to atovaquone, due to a point mutation in cytochrome B W2, a strain isolated from IndoChina, is historically chloroquine resistant an d mefloquine sensitive. Lastly D6, a strain isolated from Sierra Leonne, is historically chloroquine sensitive and naturally mefloquine resistant. Thawing of Field Isolate Samples Each blood sample vial was wiped down with 70% Etha nol to sterilize, and then thawed in a 37C water bath for 2 minutes. The sam ple was then transferred into a 50 ml sterile centrifuge tube and 0.1% of 12% sterile NaC l of the volume of the isolate sample was added drop wise to the sample, swirling the sam ple between each drop addition. The sample containing the 12% NaCl was then allowed to sit untouched for five minutes, followed by the addition of ten times of 1.6% steri le NaCl of the new isolate volume (isolate plus volume of 12% NaCl added). The NaCl was added drop wise for the first ml with swirling between each drop, followed by swirli ng after each 0.5 ml addition of 1.6% NaCl. The sample was then centrifuged at 4,000 rpm at 4C for 2 minutes. Next, the supernatant was aspirated off of the isolate, leavi ng a small amount above the hematocrit,
n with taken not to remove any parasites. The pellet was then resuspended in 10 ml of RMPI with HEPES, and centrifuged again at 4,000 rpm at 4 C for 2 minutes. The supernatant was aspirated, again leaving a small am ount above the pellet, then resuspended in 10ml of complete media containing 15% heat-inactivated AB plasma. In the third round of isolate thawing, the complete me dia was supplemented with 10ul/1ml of penicillin/neomycin/streptomycin solution in ord er to reduce bacterial contamination. The suspension was then centrifuged at 4,000 rpm at 4 C for 2 minutes, and the supernatant removed once more as noted previously. The pellet was then resuspended in 10 ml RPMI with HEPES, and again centrifuged at 4,0 00 rpm at 4C for 2 minutes. The supernatant was aspirated again, and the pellet res uspended in 2ml of complete media containing 15% AB plasma. 50ul of the suspension w as added drop wise to filter paper and left to air dry for future studies of molecular markers. If red blood cell lyses was observed, 15-30 ul of 50% hematocrit O (+) blood wa s added to the suspension prior to incubating the culture. The suspension was then tr ansferred into a six well plate, placed in a modulator incubator, air displaced with mixed gas (5%O2/5%CO2/90%N), and incubated at 37 C. Culturing Incubator modulators containing the six-well plates were gassed every twenty four hours with mixed gas. Twenty four hours after initially thawing the isolates, the sixwell plate was removed from the incubator and tippe d at a 45 angle to allow the hematocrit to settle in order to make slides. 5ul of hematocrit was removed to make both a thin and thick smear. Slides were made initially every day to check for growth of both the parasite and contamination. O+ blood was conti nuously added to each well to keep
the isolate up to 4% hematocrit while in the six-we ll plates. Contamination was contained by spinning the isolate in a 15 ml falcon tube at 4,000 rpm at 4 C for 2 minutes, followed by removing the supernatant. The remaining pellet was then resuspended in 2 ml fresh complete media containing 15% AB plasma, as well as penicillin/streptomycin/neomycin [5,000U/5mg/10mg] at a concentration of 2ul/ml. Each isolate was treated every day until contamination w as no longer seen in the thick smear. Slides were made less frequently (every 2-4 days) when no initial parasites were observed. When 2% parasitemia was observed in an i solate thin smear, the isolate was centrifuged for 2 minutes at 37 C at 4000 rpm, and the old media removed. The parasite was then resuspended in 5 ml 10% CM at 4% hematocri t (O+ blood) and transferred into a tissue culture flask. The flask was then incuba ted at 37 C with the carbon dioxide, nitrogen and oxygen mixture used previously. Media was replaced with fresh 10% AB plasma every day, and both thin and thick smears we re made, until the parasitemia reached 3%, and at least 40% ring stage was observe d. The flask was then split into two, and the remaining blood pellet was frozen down for stock. These flasks were then checked every day until 3% parasitemia was observed At this point one flask was split again to continue growth, and the other flask was u sed for the drug assay. At least four cryopreserved vials of each isolate were frozen in total for stock and future use. Freezing Protocol When the isolate was at least 40% rings and 3% para sitemia, they were frozen for future ACT resistance characterization studies. Th e culture was centrifuged at 4,000 rpm for 15 minutes at 4C, and the supernatant removed. Glycerolyte 57 (57g glycerin, 1.6 g sodium lactate, 30 mg potassium chloride, buffered with 51.7 mg monobasic sodium
phosphate and 124,2 dibasic sodium phosphate, pH 6. 8.) was added slowly drop wise to the pellet at 0.33 times that volume of the isolate pellet. The sample was then allowed to stand for five minutes, and Glycerolyte 57 was adde d again drop wise at 1.33 times the volume of the original isolate pellet and the first Glycerolyte 57 addition. The sample was then transferred into a cryovial and stored at -80 C for future resistance experiments. Plasma, Media, and Blood Preparation O+ RBCs were used when culturing and assaying each isolate. Blood was suspended in RPMI 1640 plus HEPES containing sodium bicarbonate, and centrifuged at 4000 g, 37 C for 15 minutes. The supernatant and buffy coat were removed, and the remaining blood was re-suspended in the (RPMI/HEPES /sodium bicarbonate) wash and centrifuged three more times. The supernatant was removed in between each wash and the blood was resuspended in RPMI wash media, at fi ve, five, and fifteen minute increments. After the supernatant was removed the final time, the remaining hematocrit was diluted to 50% in the RBC wash media, and store d at -40 C for up to one week before new 50% hematocrit was prepared. Plasma for cell culture media was provided by Inte rstate Blood Bank. Plasma was allowed to warm to 60 C for one hour with gent le agitation. The plasma was then cooled at room temperature for 15 minutes, then ref rigerated at -40C for an additional 15 minutes. Plasma was then transferred into 50 ml fa lcon tubes, and centrifuged for 15 minutes, at 4000 rpm for 15 minutes. The plasma lo ts were then pooled and frozen at 80 C. Before use, plasma was thawed in a 37 C wa ter bath.
r Slide Preparation A 5ul aliquot of the isolate sample was taken from culture, and 1ul was dispensed by smearing in a circler fashion on one part of the slide for the Â“thick smear,Â” and the remaining 4ul was dispensed on the other side of th e slide for the thin smear. A separate clean slide was used to spread a thin layer of cell s across the second half of the slide. The sample was allowed to dry, and the Â“thin smearÂ” was dipped in methanol and again allowed to dry. 15% or 20% Giemsa stain was then d ropped onto the slide, and the slides were then rinsed with water after a period of 10-20 minutes. The slides were then dried and ready to be examined. Drug Assay Preparations Drug Â“motherÂ” plates were prepared using the Biomek 3000 robot. Drugs in the mother plates were all prepared at 10 times the fin al concentration to be used in the final assay Â“daughterÂ” plate. Initially, drug stock solu tion was prepared at a concentration of 1 mg/ml and dissolved in DMSO. Next, drugs were dilu ted down to ten times the desired concentration in 10% AB plasma complete media, and 300 ul of each suspension was added to the first row of a 96 well mother plate. The drugs then underwent a 1:2 dilution scheme to complete the Â“mother plate.Â” This plate was used up to 5 times to make the Â‘daughter platesÂ”, which contained both the drug as well as the parasites. This dilution scheme reduced the final DMSO concentration to less than 1%. Isolate Suspension for Assay: Once each isolate was grown to a least 3% parasitem ia, and greater than 40% rings, each isolate culture was prepared for the dr ug assay. First, the cells were centrifuged at 4000 g at 37 C for 5 minutes. Complete media cont aining 10% AB
plasma was prepared, and 1.5% total hematocrit was added to the media, with a parasitemia of 0.5%. The suspension was gently mix ed before added to the daughter assay plate. 15 ul of the Â“mother plateÂ” drug susp ensions was transferred to each corresponding Â“daughterÂ” well by the Biomek robot, and 150 ul of cell suspension was then added to each well. 25 mg/ml of each artemisin in and dihydroartemisinin were added directly in duplicate form to serve as negat ive controls. The parasite suspension without the addition of any drug served as the posi tive controls in the plate. Finally, the plates were placed in a modulator incubator and in cubated for 72 hours at 37 C. The incubator was gassed every 24 hours (5%O2/5%CO2/90% N). After 72 hours, plates were wrapped in foil to protect from light, and pla ced in a -80C freezer until the SYBR green assay was ready to be performed. SYBR Green Assay Drug plates containing lysed isolates were assayed via a fluorescent based SYBR Green assay, a quick and easier method of finding t he IC50 values than alternate methods. (Bacon, 2006) Plates were thawed for two hours after being frozen in a -80 C freezer overnight. Then,100 ul of parasite suspens ion was transferred into a black 96 well plate compatible with the plate reader. The 1 00 ul of SYBR Green I in lyses buffer containing Tris 20 nM (Ph 7.5), EDTA (5mM), saponin (0.008% wt/vol and Triton X100 (0.08% wt/vol) was also added to the wells cont aining the thawed parasites in the drug assay plate. The plates were then immediately allowed to incubate light free for one hour before read. The intensity of the florescence was measured using the SpectraMaxM2 plate reader, with an excitation wavelength of 4 85 nm and an emission wavelength of 535 nm.
Isobologram Assay Three isolates that showed the highest resistance t o the 8 compounds overall were chosen to perform a drug combination assay. Starti ng plate concentrations for each drug were calculated by using the IC50 value calculated in the single compound assay. The starting drug IC50s were chosen in order for the I C50s in the multi drug plate to hypothetically be the original starting concentrati on in the plate if the assay was redone. The plate design consisted of a row of each of the drug pairs alone, followed by the pairs in ratios on 1:3, 1:1 and 3:1. The drug pairs teste d consisted of either QHS or DHA, depending on which one the isolate was showing the least susceptibility to. These drugs were paired with MFQ and PIP, both commonly adminis tered drug combinations in the clinic. Thai isolate 18, 6, and Cambodia isolate 6 were us ed in the isobologram assay plates. W2 was used as a control. The starting co ncentrations in the assay plate of Thai 18 were 8.7 ng/ml for DHA, 234.2 ng/ml for MFQ and 179.9 ng/ml for PIP. Thai 6 had starting drug concentrations of 80 ng/ml for QHS, 175.1 ng/ml for MFQ, and 163.9 ng/ml for PIP. Cambodia isolate 6 had starting con centrations of 6.1 ng/ml for QHS, 56 ng/ml for MFQ and 12.36 ng/ml for PIP. Reference c lone W2 had a starting concentration of 11.88 ng/ml for QHS, 30.22 ng/ml f or MFQ, 225.25 ng/ml for PIP and 4.9 ng/ml for DHA.
Data Analysis Linear and non-linear regression curves for IC50s, IC90s and FIC isobologram curves were calculating using DatAspects Plate Mana ger (DPM) software. Data was imported into the software from the plate reader us ing the Biotec format. Relative Fluorescent Units (RFUs) were converted into percen t inhibition in the program.
Figure 3: Map Depicting Isolate Collection Sites
Figure 4: 6 Well Plates Containing Isolates
Figure 5: Isolates Growing Within Tissue Culture F lasks
n Figure 6 : Geimsa Staining of Isolates
Figure ( D Figure 7: 96 Well Drug Assay Plate Set Up D rug Name and Concentration in ng/ml)
Figure 8: Biomek 3000 Â“RobotÂ” Used to Complete Drug Assays
r Figure 9: Plate Reader for SYBR Green Assay
Figure 10: Incubator Modulators Containing 6 Well Plates (Right) and 96 Well Assay Plates (Left)
Results Cell Culture Success A total of 66 isolates, 42 Cambodia, 24 Thai, were thawed. Out of the 42 Cambodia isolates, 19 grew enough to be assayed (45 %). Out of the 24 Thai isolates thawed, 15 grow well enough to be assayed (63%). O verall, a growth success rate of 52% was obtained. The addition of antibiotic to th e complete media during the thawing process noticeably increased the percentage of isol ates that successfully grew. For instance, a 50% growth success rate was obtained be fore the antibiotic was added during the thawing process, and a 56% growth rate was obta ined after the antibiotic was added. Contamination was also greatly reduced overall when the antibiotic was added during the thawing process. Before antibiotic was added, at le ast 50% of samples showed some degree of bacterial contamination, both rods and co cci, versus around 5% after antibiotic was added. Single Drug Assay Data The IC50s (in nmol/l) varied between isolates, depe nding on type of drug pressure (refer to tables 1 and 2, as well as Appendix B). IC50s of isolates exposed to artemether ranged from 0.7 to 10.3 nmol/l. IC50s of isolates exposed to artesunate ranged from 0.2 to 6.4 nmol/l. IC50s of isolates exposed to atovaqu one ranged from 0.2 to 13.9 nmol/l. IC50s of isolates exposed to dihydroartemisinin ran ged from 0.4 to 7.1 nmol/l. IC50s of isolates exposed to lumefantrine ranged from 6.5 to 395.3nmol/l. IC50s of isolates exposed to mefloquine ranged from 2 to 135.1 nmol/l IC50s of exposed to piperaquine
ranged from 4.7 to 124.1 nmol/l. IC50s of isolates exposed to Artemisinin ranged from 1.4 to 35.4nmol/l. Certain isolates, both Thai and Cambodian consistently produced high IC50s in the standard 72 hours drug assay. For instance, Ca mbodia isolate Â“6Â” produced some of the highest overall IC50s to artemisinin, artemethe r, lumefantrine, dihydroartemisinin, and mefloquine. Thai isolate Â“6Â” was also consiste ntly less susceptible to a majority of the drugs tested. Cambodia isolate Â“5Â” was among t he most susceptible parasites and produced the lowest IC50s to the drugs tested. In addition, when observing IC50 trends in reference to MFQ, none of the 7 drugs showed a s ignificant positive correlation in IC50 trends. Isobologram Assay Data Thai isolate 6, Cambodia isolate 6, and Thai isolat e 18 were chosen for the drug combination studies based on the fact that they sho wed the most evidence of resistance to the drugs when tested individually. Isobolograms d epicting the fractional inhibitory concentrations (FIC) curves were produced using the DPM software (refer to Appx. A). In order for a drug pair to be considered Â“synergis ticÂ”, the sum of the FICs must be under 0.5. For the drugs tested, only Thai 18 under DHA and MFQ pressure had FIC meeting this criteria when DHA and MFQ were in a 1:1 ratio of one another. In order for a drug combination to be considered antagonistic, the FIC sum must be higher than 2 ng/ml Thai 6, when exposed to QHS and MFQ at a ratio of 1 :3 respectively produced an FIC of 2.320, which indicates a significant antagonisti c effect. In addition, Thai 6 treated with QHS and PIP in a ratio of 1:3 respectively als o produced a significant antagonistic FIC of 2.048. The W2 reference clone under similar conditions did not conclusively
lead to synergism or antagonism, which was expected from previous studies in the Kyle lab.
Table 1: IC 50s (ng/ml) of Thai Isolates in the 72 hour SYBR Green Drug Assay nnr nr n rrn rrnrrrnn rrrrrr r rr rrrrr rrr rrrrr n rrrnn r rrrr rrnrrr n rrrnnrnrrrrrr rr rrrr
Table 2: IC 50s (ng/ml) of Cambodia isolates in th e 72 hour SYBR Green Drug Assay n n nr n r r r r n r rn r r r r n n r r r r r r r n r r r r rr r r r n r r nn r r n r r n r r r r r r n n r r r r n r r r n r n r r r r r r r r r r r rn r n r r r r r r r r
n Table 3: IC 50s (ng/ml) of Reference Strains W2, D6 and C2B in the 72 Hour SYBR Green Drug Assay control AM AS ATO DHA LF MFQ PIP QHS W2 .58 .56 .19 .61 6.11 3.78 28.12 1.48 C2B 1.1 .01 62.5 .32 53.78 4.91 52.24 1.69 D6 1.7 .19 .06 1.74 84.66 2.44 32.02 2.66
Discussion Findings This thesis describes the general resistance trends to 8 commonly used antimalarial drugs in recent P. falciparum isolates from Thailand and Cambodia. It is important to determine if resistance is occurring e arly on, in order to devise an efficient containment plan, which can include, but is not lim ited to alteration of drug regiments, education, proper dispersal and formulation of new and improved drugs. As hypothesized, results from this blinded phenotypic analysis show some trends that suggest isolates in Southeast Asia as less suscepti ble to some common ACT drugs. This is consistent with previous clinical and in vitro s tudies performed in the past (Noedl, 2008; Khositnithikul, 2008). In particular, Cambodia isolate 6 and Thai isolate 6 were consistently less susceptible to the majority of drugs tested, in ref erence to the other isolates tested, as well as in comparison to the positive controls, C2b, D6 and W2 (refer to appendix B for details). Atovaquone was the only drug to which al l of the isolates (with the exception of Thai isolate 13) were consistently susceptible. Aga in, molecular markers testing should be conducted on Thai 13 in order to determine if a point mutation in Cytochrome B is indeed occurring, a key sign of atovaquone resistan ce ( Khositnithikul,2008). Drug combination assay studies revealed some unexpe cted trends towards antagonism for several ACT combinations. An exampl e of this was when Thai 6 was exposed to the 1 part QHS and 3 part MFQ drug combi nation (FIC=2.048), as well as
when exposed to the 3 QHS part and 1 part PIP drug combination (FIC=2.32). Trends for some synergism in the drug combination response s were also observed with this assay, the most significant one being Thai 18 when incubated with 1 part DHA to 1 part MFQ (FIC=.362). These blinded results should be validated via other methods. The isolate data on drug susceptibility still needs to be tied back to original parasitemia and specific location to examine the trends more closely. The results co uld be biased in that the majority of the isolates assayed may have originated from isola tes with higher original parasite load. The isolates that did not grow successfully in the lab must be attempted to grow again so that one is not selecting and testing for a severe infection only. Also, it is possible that these isolates already contained other types of dru gs and although a thorough wash protocol was used, previous drug pressure could hav e also negatively affected the accuracy of our results when selecting for a partic ular drug. Strengths and W eaknesses The use of the SYBR Green assay has many advantages opposed to other conventional ways to evaluate drug effectiveness an d ultimately drug resistance trends. For instance, lab personnel safety and clean up is a lot easier than a radioisotope assay. It is faster to carry out than an ELISA, which require s multiple washing steps, and the materials needed to complete the assay are easier t o obtain. Finally, the SYBR Green assay is more consistent and easier than the micro test (morphology based), which requires slides of each of the 96 well plates befor e a conclusion could be made (Bacon, 2006). Unfortunately, there are many variables in the assay that could negatively affect the accuracy of the results obtained. For instance slight alterations in parasite
r parasitemia, hematocrit, and plasma lots could sign ificantly alter IC50s obtained from all of the assay formats, including SYBR green. Testing drug combinations that have major structura l and functional differences can also affect the accuracy of the data obtained. For instance, if the improper ratio of drugs is used in a drug assay and one drug has a mu ch different dose response profile, the FICs could become skewed. Also, FICs can vary base d on parasite strain and parasitemia, and since this is a blinded study, it is difficult to come to a solid conclusion regarding drug-drug interactions. Further informat ion and testing will need to be done in order to validate the assumptions made regarding th e drug combinations. The data in the drug assays provide a phenotype tha t characterizes the drug response. These phenotypes will be useful to ident ify putative genetic markers of resistance, such as SNPs or amplifications. If the mechanism of resistance is known for artemisinin and its derivatives, one can then test whether or not the same mechanism is occurring in a particular isolate of interest. Future study plans This study is the initial step in characterizing re sistance trends in these particular isolates. Data for these isolates will be compared to fresh, ex vivo in vitro drug susceptibility data conducted at the clinical field sites. Isolates that did not successfully grow during this experiment will be attempted to be grown once again. Also, isolates showing decreased susceptibility to ACTs will also have further tests performed. SNP studies, gene amplifications, as well as sequencing entire genomes to figure out resistance mechanisms will occur in the future at the Universi ty of Maryland. Only then, will the real story of current ACT resistance be fully under stood.
List of References Bacon, D.J., Latour, C., Lucas, C, Colina, O., Ring wald, P., & Picot, S. (2006). Comparison of a SYBR Green I-Based Assay with a His tidine-Rich Protein II Enzyme-Linked Immunosorbent Assay for In Vitro Anti malarial Drug Efficacy Testing and Application to Clinical Isolates Antimicrobial Agents and Chemotherapy 51(4), 1172-1178. Centers for Disease Control and Prevention [CDC] (2 001). Malaria Drug Resistance. Available from UR http://www.cdc.gov/Malaria/drug_ resistance.htm Giao, P.T., Bing, T.Q., Kager, P.A., Long H.P., Van Thang, N., Van Nam, N & de Vries, P.J. (2001) Artemisinin for treatment of uncomplica ted falciparum malaria: is there a place for monotherapy? American Journal of Tropical Medicine and Hygiene, 65(6), 690-695. Ikeoluwapo, OA, Browne EN, Bateganya F, et al. (200 8). Effectiveness of artemisininbased combination therapy used in context of home m anagement of malaria: A report from three study sites in sub-Sahara Africa. Malaria Journal 7(190). Ikeoluwapo, OA, Browne EN, Garshong B et al. (2008) Feasibility and acceptability of artemisinin-based combination therapy for the home management of malaria in African Sites. Malaria Journal. 7(6). John, David T and William A. Petri. (2006). Medical Parasitology. St Louis, Missouri: Elsevier, Inc. Khositnithikul, P, Tan-ariya, P and Mathirut Mungth in. (2008). In vitro atovaquone/proguanil susceptibility and characteri zation of the cytochrome b gene of Plasmodium falciparum from different endemic regions of Thailand. Malaria Journal. 7(23). Noedl, Harold. (2008). Evidence of Artemisinin-Res istant Malaria in Western Cambodia. The New England Journal of Medicine. 359 : 2619-2620. Pehn, P. (2007) Containment of malaria multi-drug r esistance on the ThailandCambodian border-report of an informal consultation World Health Organization.
Price, R.N., Uhlemann, A.C., Brockman, A, McGready, R., Ashley, E., Phaipun, L., Patel, R., Laing, K., Looareesuwan, S., White, N.J., Noste rn, F., Krishna, S (2004). Mefloquine in Plasmodium falciparum and increased p fmdr1 gene copy number. Lancet. 364: 428-447. Riechmann, K.H. (2006). The chequered history of m alaria control: are new and better tools the ultimate answer? Annals of Tropical Medic ine and Parasitology. 100(8): 647-662. Sherman, Irwin. (1998). Malaria. Washington D.C., U SA, American Society for Microbiology. Tran, T.H.I., Dolecek, C, Pham Pv et al. (2004) Dih ydroartemisinin-piperaquine against multidrug-resistant Plasmodium falciparum malaria in Vietnam: randomized trial. Lancet. 336, 18-22. Wongsrichanalai, C., Wimonwattrawatee, T., Sookto, P., Laoboonchai, A., Heppner, D.G., Kyle, D.E., Wernsdorfer, W.H. (1999). In vitr o sensitivity of Plasmodium falciparum to artesunate in Thailand. Bulletin of the World Health Organization. 77(5). World Health Organization. Guidelines for the Treat ment of Malaria. 2006.
Bibliography Kyle, D.E., Oduola, A,M., Martin, S,K & Milhous, W. K. (1990). Plasmodium falciparum : modulation by calcium antagonists of resistance t o chloroquine, desethylchloroquine, quinine, and quinidine in vitr o. Trans T Society of Tropical Medicine and Hygiene 84(4), 474-478. Medicines for Malaria Venture (2008). Artemisinin. Available from URL http://www.mmv.org/rubrique.php3?id_rub rique=128. Rogers, W.O., Sem, R., Tero, T., Chim, P., Lim, P., Muth, S., Socheat, D., Ariey, F., Wongsreichanalai, C. (2009) Failure of artesunatemefloquine combination therapy for uncomplicated Plasmodium falciparum mal aria in southern Cambodia. Malaria Journal, 8(10). Shoklo Malaria Research Unit (2008). FAQs in Artemi sinin based Combination therapy. Available from URL http://www.shoklo-unit.com/faq.shokl o-unit.com/index2.html. Wongsrichanalai, C.H., Webster, H.K., Wimonwattrawa tee, T., Sookto, P., Chuanak, N., Thimararn, K., & Wersndorfer, W.H. (1992 ) Emergence of Multidrug-Resistant Plasmodium falciparum in Thailand: In V itro Tracking. American Journal of Tropical Medicine and Hygiene 47(1), 112-116. Yeung, S., Damme., W.V., Socheat, D., White, N.J & Mills, A. (2008) Access to artemisinin combination therapy for malaria in remo te areas of Cambodia. Malaria Journal, 7(96).
nn !"#$!""$ !"#$!%%%$
nn!$ r& !"#$!""$ r& !"#$!%%%$
n nn!$ r&'(n!%'%$!""$ r&'(!%%%$n!%'%$
Appendix A (continued) W2 control QHS(97) PIP(444) W2 QHS (97) and MFQ (99)
Appendix A (continued) W2 DHA (414) PIP (444)
r Appendix B
Appendix B continued
Appendix B continued
Appendix B continued
Appendix B continued
Appendix B continued
Appendix B continued
n Appendix B continued