| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES |
|---|
|
INTRODUCTION |
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES |
|---|
Antimalarial drugs remain at the forefront of malaria prevention and treatment, and with increasing resistance to existing drugs there is an urgent need for new effective and safe medications. 3,4 Tafenoquine, an 8-aminoquinoline analogue of primaquine, is a promising candidate for causal prophylaxis for non-immune travelers, for the radical cure of P. vivax, and for transmission reduction through gametocytocidal action.5 The 8-aminoquinolines are the only approved class of drugs with activity against P. vivax hypnozoites and stage 5 P. falciparum gametocytes. The World Health Organization (WHO) recommends primaquine in combination with chloroquine for the radical cure of P. vivax malaria, although limited compliance with the 14-day dosing regimen is known to impact effectiveness.6 Given the long half-life of approximately 14 days, tafenoquine has the potential to address problems with compliance, as Phase II studies show that 1–3 days of treatment is effective to prevent P. vivax relapse when used alone or in combination with other antimalarials. 7,8 Consequently, tafenoquine is being developed for the radical cure of P. vivax malaria. In addition to effectiveness against hypnozoites, tafenoquine’s antigametocyte activity and long-acting blood stage activity against multidrug resistant strains make it a potentially important agent for P. falciparum eradication under the new initiative by the Bill and Melinda Gates Foundation.9
Detailed ophthalmic assessments in a sub-group of persons in a large Phase III prophylaxis study in East Timor showed that, although tafenoquine 200 mg weekly for 6 months was well tolerated, 69/74 persons in the tafenoquine group developed vortex keratopathy (phospholipid deposits in the cornea). 10 No keratopathy was seen in the mefloquine-treated persons. The corneal deposits were benign and reversible with resolution in more than 90% of persons at 6 months and complete resolution in all persons by 1 year post-prophylaxis. An expert ophthalmology advisory panel concluded that vision had not been affected in any persons, but that the relevance of minor retinal findings, in the absence of baseline assessments, could not be ascertained. Therefore, it was considered that potential ophthalmic effects of tafenoquine required further study.
In clinical studies of tafenoquine, small and transient increases in serum creatinine concentration were identified in tafenoquine persons compared with those receiving placebo or comparator drugs. 10 Although these increases were rarely outside the normal range, were asymptomatic, and did not have any long-term consequences, we believed the potential changes in renal function also warranted further study. To address the long-term safety of tafenoquine, the objectives of this study were to assess the ophthalmic and renal effects of tafenoquine after a 600 mg loading dose and 200 mg weekly dosing for 6 months.
|
METHODS |
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES |
|---|
The person’s health status was determined by a medical history, vital signs, physical examination, and laboratory screening. Volunteers were excluded if they had a history of eye surgery, corneal or retinal abnormalities, current use of eye drops, or were participating in activities that could affect vision (e.g., scuba diving, exposure to high altitude, or excessive sunlight). Volunteers that were allergic to 8-aminoquinolines, had a history of drug or alcohol abuse, used prescription medications or another investigational agent within 30 days, or non-prescription drugs within 14 days of enrolment were excluded from study participation. As assessed by laboratory screening, persons were excluded if they had clinically significant abnormalities, including evidence of renal or hepatic dysfunction, G6PD deficiency, hepatitis B, C, or human immunodeficiency virus (HIV) infection.
Design. Eligible persons were randomized in a 2:1 ratio to tafenoquine or placebo. The oral dosing regimen was 200 mg once daily for 3 consecutive days (i.e., loading dose, 600 mg), followed by 200 mg once weekly for 23 weeks. Study medications were supplied by GlaxoSmithKline (Brentford, Middlesex, UK).
The primary ophthalmic endpoint was the proportion of persons with impaired night vision as measured by the forward light scatter test (FLST), a test that is sensitive to the presence of scatter secondary to corneal deposits. Secondary ophthalmic endpoints included further assessment of night vision, assessment of macular function, visual acuity, color vision, corneal deposits, and changes in retinal morphology.
The primary renal endpoint was tafenoquine’s effect on mean change in glomerular filtration rate (GFR), compared with placebo, after 24 weeks of drug administration. Secondary renal endpoints included the number of persons with significant changes in GFR, serum creatinine, or urinalysis findings anytime after drug administration.
Schedule. After obtaining informed consent, screening assessments included review of past medical history, physical examination, ophthalmic and laboratory testing. Within 21 days of successful screening and eligibility determination, a Day 0 visit was scheduled where baseline laboratory assessments, including GFR, were measured. On Days 0, 1, and 2, persons received a loading dose of tafenoquine 200 mg or placebo once daily. After loading, persons had once-weekly dosing of tafenoquine 200 mg or placebo for a total of 23 weeks (24 weeks effective drug administration).
Ophthalmic tests were performed at screening, and then at Weeks 3, 6, 12, 18, 24, as well as 12 and 24 weeks post dosing. Glomerular filtration rate was assessed on Day 0 before the first loading dose, on Week 12 and on Week 24 of drug administration, with additional testing as indicated by urinalysis or creatinine results (confirmed proteinuria or hematuria > trace, serum creatinine concentration increased from baseline by > 0.3 mg/dL or 26.6 µmol/L). Hematology and biochemistry parameters were assessed at regular intervals from screening to follow-up visits.
Adverse event reporting, pregnancy testing, and concomitant medications review were conducted at screening, weekly throughout the dosing period, and at follow-up. Persons kept a diary card to record compliance. Drug administration was supervised for all loading doses, Weeks 3, 6, 12, and 18; for all other doses persons were required to telephone the study site to confirm they had taken their medication.
Assessments.
Ophthalmic testing included an assessment of night vision using the FLST, low-contrast visual acuity (LCVA), mesopic contrast threshold (MCT), and scotopic contrast threshold (SCT). Failure in FLST (primary endpoint) was defined as an increase from baseline of scattered light k’ 
3 resulting in k’ 22.84 in either eye at any assessment. Best-corrected visual acuity was also assessed using the Early Treatment Diabetic Retinopathy Study (ETDRS) high-contrast visual acuity chart. Color vision was assessed using standard pseudoisochromatic plates (PIP) and a specialized color sorting test (the L’Anthony 40 Hue test) as well as the City University Color Assessment and Diagnosis (CAD) test. 11 Macular function was assessed using the Amsler grid, Humphrey 10-2 visual field, and a macular stress test. Masked reviewers (optometrists) assessed for potential corneal deposits and retinal morphology changes using digital corneal and retinal photographs during and after the dosing period of the entire study. In addition, on study completion, retinal photographs from study persons underwent a masked review and grading assessment by a specialized, independent retinal reading center.
Renal function was primarily assessed by changes in GFR measured using the iothalamate-clearance technique. 12–14 Iothalamate assays for GFR were conducted by the Uniformed Services University of Health Sciences (Bethesda, MD). Briefly, persons received an intravenous bolus injection of 2-mL Conray 30 (600 mg iothalamate meglumine) and a 5-mL sterile normal saline flush. Blood and urine samples were collected approximately at 60, 90, 120, and 150 minutes. Serum and urine iothalamate (IOT) concentrations were measured in duplicate for each specimen using the high-performance liquid chromatography (HPLC) method developed in this study. 12 The GFR was calculated as an average of IOT clearance. IOT = amount of IOT excreted in urine/AUC, where AUC is the area under the serum IOT concentration-time curve, determined using the trapezoidal rule.
Hematology tests included hemoglobin, haptoglobin, hematocrit, red blood cell count, and reticulocyte levels. Laboratory values indicative of hemolysis were defined as a 15% decrease from baseline in hemoglobin or hematocrit, together with a 50% decrease from baseline in haptoglobin.
Adverse events (AEs) were reported using the MedDRA (Medical Dictionaries for Regulatory Activities) data coding system. Serious adverse events (SAEs) were defined as events that resulted in death, were life-threatening, required hospitalization, prolongation of an existing hospitalization, or resulted in incapacity or disability. The protocol defined clinically significant renal and ophthalmic events as SAEs to facilitate expedited reporting.
Monitoring and withdrawal criteria.
Routine blinded assessments of ophthalmic and safety were performed by two specialist independent data monitoring committees (IDMCs). Withdrawal criteria for ophthalmic safety included decreased vision, bull’s eye retinopathy, distortions observed on the Amsler Grid test, abnormal color vision, and development of scotoma on visual field testing. Digital corneal and retinal photographs were assessed by independent, masked expert reviewers, as the presence of corneal deposits had the potential to unblind the reviewer to study treatment. Persons were withdrawn for renal safety if a persistent decrease in GFR of 20% from baseline was detected. Further withdrawal criteria included evidence of renal injury based on urinalysis including persistent and significant proteinuria, hematuria, or glucosuria.
Statistics. It was estimated that a sample size of 120 persons randomized 2:1 for tafenoquine:placebo would yield 70 evaluable tafenoquine-treated persons, providing at least 90% statistical power for the primary analyses. The study was not powered to evaluate the secondary endpoints. For the primary endpoint of the proportion of persons with unimpaired night vision, there was no formal between-group comparison. Clopper-Pearson exact one-sided 95% confidence intervals (CIs) for the proportion of persons with unimpaired night vision were calculated separately by treatment. 15 If the lower limit of the CI for tafenoquine was < 90%, the result was considered clinically important.
The primary renal objective was to show non-inferiority of tafenoquine to placebo on the mean change in GFR from baseline to Week 24. The pre-defined non-inferiority limit was –15% of mean GFR values observed at baseline as calculated while blinded to study group. An analysis of covariance was used to compare mean changes for tafenoquine versus placebo using a two-sided 95% CI and adjusting for baseline GFR, age, sex, race, and study site.
The safety population included all randomized persons who received at least one dose of study medication (Figure 1
). Based on the long half-life of tafenoquine, we defined compliance over the entire dosing period as having completed three loading doses in Week 1, having missed no more than five doses and no more than three consecutive doses from Weeks 2 to 23.
|
).
Renal endpoints were assessed in the renally evaluable population, including all persons in the safety population who had no major protocol violations. The primary renal analysis included persons for whom GFR values were available at baseline and Week 24 or withdrew early from the study for clinically important renal findings, with their last GFR measurement carried forward. Sensitivity analyses were performed separately on the renally evaluable population using worst case where all missing Week 24 GFR measurements were imputed by carrying last observations forward, and on the safety population using observed data (Figure 1).
|
RESULTS |
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES |
|---|
). In the tafenoquine and placebo groups, 72/81 (88.9%) and 35/39 (89.7%) persons, respectively, were compliant with study medication as defined previously (Figure 1).
|
), indicating tafenoquine did not impair night vision in this study. The conservative worst case analysis, with imputed failure for all persons with post-baseline assessments and undetermined success at Week 24, showed a successful FLST in 56/69 (81.2%) with tafenoquine and 26/32 (81.3%) with placebo, respectively. The lower bound of the one-sided 95% CI was < 90% for both tafenoquine and placebo groups at 71.7% and 66.3%, respectively. The sensitivity analysis on the safety population using observed data confirmed the primary result that tafenoquine does not impair night vision as measured by FLST.
|
).
|
There were no meaningful differences between the study groups in changes to high contrast visual acuity measured using ETDRS chart, and there were no meaningful changes in color vision, as assessed by PIP plates, CAD, and L’Anthony 40 hue tests. The majority of persons had normal test results throughout the study at > 98% and > 96% in the tafenoquine and placebo groups, respectively.
Corneal deposits were reported as present at screening in 10/70 (14.3%) and 7/32 (21.9%) in tafenoquine and placebo persons, respectively. Treatment-emergent corneal deposits were observed in 15/60 (25%) and 4/25 (16%) persons in the tafenoquine and placebo groups, respectively, with no observed trend for time to onset. In 14 tafenoquine persons, new-onset corneal deposits resolved within 12 weeks of onset, many while still on study drug, and in the remaining person the corneal deposits resolved by 24 weeks post-dosing. In the 4 persons in the placebo group, the corneal deposits resolved within 6 weeks of onset.
Retinal abnormalities identified by digital photography were reported in one person in each study arm. In the person that received tafenoquine, a single area of retinal hyperpigmentation was detected at follow-up. This was not associated with a decrement in visual acuity, foveal sensitivity, or visual field and did not change 11 months after cessation of therapy. One placebo person had a retinal abnormality noted at the 12-week follow-up visit that resolved within 2 months.
Independent retinal photograph grading was performed for 60 persons who had complete records for Week 3 and Week 24 visits for both eyes (N = 39 in the tafenoquine group and N = 21 in the placebo group). There were no persons who showed evidence of retinal morphology changes over the course of the study, and there were no objective signs of toxicity.
Renal safety.
From baseline to Week 24, the mean treatment difference in GFR was –0.061 mL/s/1.73 m2 (95% CI, –0.168, 0.045), adjusted for baseline GFR, age, study site, race, and gender using analysis of covariance. The lower confidence limit was larger than the non-inferiority margin of –0.247, indicating tafenoquine was non-inferior to placebo for the mean change in GFR (Table 4). The sensitivity analyses supported the non-inferiority of tafenoquine to placebo.
|
Clinically important changes in serum creatinine concentration ( 26.6 µmol/L or 0.3 mg/dL) from baseline were reported in 3/70 (4.3%) and 1/32 (3.1%) persons per visit in the tafenoquine and placebo groups, respectively. Clinically significant increases in creatinine concentration were in the 1.1 to 1.3mg/dL (97.0 to 115.0 µmol/L) range, although in all cases values were inside the extended normal range (< 150% baseline), and were not associated with clinically significant decreases in GFR. In all cases, by the following study visit, changes in serum creatinine concentration were no longer considered to be clinically significant.
Of those with urinalysis results at Week 24, clinically important urinalysis findings were found in 3.6% and 11.5% of persons in the tafenoquine and placebo groups, respectively. Two persons in the tafenoquine group and one in the placebo group had hematuria > trace, and two persons in the placebo group had proteinuria > trace. None of these cases were associated with any significant change in GFR or serum creatinine concentration and all resolved without treatment.
Laboratory tests and urinalysis. No person in this study had laboratory values indicative of hemolysis, as defined by protocol as a 15% decrease from baseline in hemoglobin or hematocrit, together with a 50% decrease from baseline in haptoglobin. However, there was one person who had an apparent hemolytic anemia in the tafenoquine group. At baseline, the person’s hematocrit, haptoglobin, and hemoglobin levels were at or just below the lower limit of normal range. At Week 3, the person showed a 17% decrease in hemoglobin and a 23% decrease in haptoglobin, consistent with mild hemolysis. The person had normal quantitative enzyme and genetic G6PD testing results. Study drug was withdrawn and all hematology values returned to normal 12 weeks after cessation of therapy.
In general, there was a higher incidence of mild reductions in haptoglobin (< 85% baseline) in the tafenoquine group compared with the placebo group (47% versus 31%). From Week 3 to Week 12, there was a higher incidence of an increase in reticulocytes ( 150% baseline) in the tafenoquine group, although no difference was seen from Week 12 to Week 24. There were no clinically significant changes for other hematology parameters.
Adverse events.
The incidence of AEs was similar in the tafenoquine and placebo groups at 74.1% and 79.5%, respectively, and the majority of events were mild or moderate in severity (Table 5). The most common AEs during study drug administration were headache, nausea, and vomiting. The majority of AEs were not considered to be related to the study drug, with only 24.7% and 20.5% of AEs in the tafenoquine and placebo groups, respectively, considered to be related. The incidence of nausea was higher in the tafenoquine than the placebo group, at 14.8% and 5.1%, respectively. Overall, the incidence and nature of all other AEs was similar between the groups, with the exception of headaches that were less common with tafenoquine than placebo at 36% and 53%, respectively. There were five cases in four persons of treatment emergent myalgia/intercostal myalgia in tafenoquine persons, compared with none in the placebo group. Two of the five tafenoquine cases had creatinine phosphokinase (CPK) values outside the normal range during the treatment phase, with an additional case during follow-up. There were 7/81 (8.6%) and 3/39 (7.7%) persons with SAEs in the tafenoquine and placebo groups, respectively (Table 6). The case of mild decrease in macular sensitivity on the Humphrey perimetry test was considered to be possibly related to study drug, and all other SAEs were considered by the investigator to be unlikely to be related to study drug.
|
|
|
DISCUSSION |
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES |
|---|
As with other cationic amphiphilic drugs, tafenoquine has the potential to inhibit lysosomal metabolism and cause phospholipidosis, leading to phospholipid accumulation in the cornea and reversible keratopathy. 16–18 In the previous Phase III 6-month prophylaxis study, pigmented corneal deposits were observed in 69/74 tafenoquine persons compared with 0/21 persons in the comparator group. 10 These corneal deposits did not affect vision, resolved completely within 5 to 12 months post-prophylaxis, and were similar to those associated with other antimalarial drugs, such as chloroquine. 17 A lack of baseline retinal data made it impossible to assess the significance of minor retinal findings seen in a small number of persons on fundoscopy and fundus fluoroscein angiography at follow-up.
The current study is the first to directly measure the impact of tafenoquine on night vision as corneal deposits could affect this, and to collect detailed retinal assessments throughout a 6-month drug administration period. The primary endpoint was the proportion of persons with unimpaired night vision using FLST, with 100% of persons in the tafenoquine and placebo groups considered to be a FLST success. Detailed retinal assessments showed no clinically significant changes in retinal function or morphology during the study, and tafenoquine had no apparent effect in any of the other ophthalmic tests, including visual acuity and color vision.
The incidence of treatment emergent corneal deposits was 25% in the tafenoquine group, which is substantially lower than in the previous Phase III study where the rate was 93%. 10 In addition, whereas in the previous study the majority of persons had corneal deposits at the end of 6 months of drug administration, the duration of the corneal deposits in the current study seemed to be much shorter, resolving within 12 weeks of onset in the majority of cases. One exception was a case that did not resolve until 30 weeks after onset. Moreover, in the current study, corneal deposits were detected in 16% of placebo persons, suggesting some of these differences may be related to the corneal assessment technique.
In the previous study, the effect of debris in the tear film was removed by using direct visualization with slit lamp examination to assess corneal pathology. Debris in the tear film moves with the tear film during this dynamic assessment, whereas corneal deposits are fixed. In the current study, off-site review of digital corneal photographs was used to maintain unbiased assessments of both corneal and retinal pathology. The presence of tear film debris in still images is at times difficult to differentiate from deposits within the corneal epithelial tissues and may have led to false positive results in some cases. Additionally, environmental factors may have contributed. For example, persons with excessive ultraviolet light exposure were excluded from the study, whereas those in the previous study were intensely exposed during their deployment to East Timor. Compliance would have had to be very low to account for the large differences seen between the studies. This is very unlikely because directly observed therapy was conducted for 28% (7/25 doses) and supervised dosing by telephone contact for all other doses.
In the same previous study, elevations in serum creatinine concentration of unknown etiology were observed. 10 The changes were small, were rarely outside the reference range, and there was no evidence of chronic renal injury. However, they were more frequent in tafenoquine persons compared with placebo persons.
This study shows that tafenoquine does not have a deleterious effect on renal function compared with placebo when direct measurement of GFR is assessed. Because serum creatinine concentration represents a surrogate marker of renal function, as an increase in serum creatinine concentration can be caused either by a decrease in overall renal function or because another drug has blocked its secretion, the current study was designed to assess potential effects of tafenoquine on renal function using a direct measurement of GFR using non-radioactive iothalamate as a filtration marker. 19,20 The small changes in serum creatinine concentration (increases to > 125% baseline value, but < 26 µmol/L or 0.3 mg/dL) found using this method were not associated with significant changes in GFR, and the number of reported cases of decreased GFR was similar between the study groups. There were no clinically significant changes in urinalysis findings suggestive of renal dysfunction. Overall, these results suggest that tafenoquine 200 mg weekly over a 6-month period does not have a clinically important effect on renal function.
The incidence of AEs was high at 74.1% and 79.5% in the tafenoquine and placebo groups, respectively. However, a high AE incidence was not unexpected in such a long study with weekly active reporting of AEs, and the majority of events were not considered to be study drug related. The incidence of nausea was higher in the tafenoquine group at 14.8% than the placebo group at 5.1%, and this is consistent with previous studies. Overall, there were no major differences in the incidence or nature of AEs between the tafenoquine and placebo groups. The incidence of SAEs was similar in the tafenoquine and placebo groups at 8.6% and 7.7%, respectively, and the majority of events were not considered to be related to study drug. Overall, there were no clinically significant laboratory findings, and the tolerability profile of tafenoquine was similar to that seen in previous studies. 7,8,21–24
Tafenoquine trials have revealed mild reductions in hemoglobin and cases of hemolysis in G6PD deficient individuals, and the drug is contraindicated in people with G6PD deficiency because of the potential for serious hemolysis. 21,23,24 Persons with G6PD deficiency were excluded from the current study and a clinical dose-escalation study in G6PD-deficient persons is planned to better quantify and characterize this important risk. In this study, one person withdrew for an apparent study drug-related hemolytic anemia that, given the normal G6PD status of this person, appears to have been an idiosyncratic event. Abnormal hematology values (reticulocyte count and haptoglobin levels) were more common in the tafenoquine than the placebo group, although there were no additional unexpected or clinically important hematologic events.
To conclude, the results of this study show that tafenoquine, 600 mg loading dose followed by 200 mg weekly for 6 months has an acceptable ophthalmic and renal safety profile and support the continued clinical development of this important new antimalarial agent.
Received November 26, 2008. Accepted for publication March 29, 2009.
Financial support: This study was supported by GlaxoSmithKline (Brentford, Middlesex, UK) and the U.S. Army Medical Research and Materiel Command.
Disclaimer: This manuscript was reviewed by the Walter Reed Army Institute for Research and the United States Army Medical Research and Material Command. There is no objection to its publication or dissemination. The opinions expressed herein are those of the authors and do not necessarily reflect the official policy, position, or opinions of the Department of the Army, the Department of Defense, or the U.S. Government.
Disclosure: Some of the authors are employed by GlaxoSmithKline, which funded the study. This statement is made in the interest of full disclosure and not because the authors consider this a conflict of interest.
* Address correspondence to Kevin Leary, United States Army Medical Materiel Development Activity, 1430 Veteran’s Drive, Fort Detrick, MD 21702-5009. E-mail: kevin.leary1{at}us.army.mil 
Authors’ addresses: Kevin Leary, United States Army Medical Materiel Development Activity, 1430 Veteran’s Drive, Fort Detrick, MD 21702-5009, Tel: 301-619-1106, Fax: 301-619-2304, and Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814, Tel: 301-295-3239, Fax: 301-295-3976, E-mail: kevin.leary1{at}us.army.mil. Michael A. Riel, Michael J. Roy, Louis R. Cantilena, and Daoqin Bi, Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814, Tel: 301-295-3239, Fax: 301-295-3976. D. Craig Brater, Indiana University School of Medicine, Fairbanks Hall 6200, 340 West 10 Street, Indianapolis, IN 46202, Tel: 317-274-8416, Fax: 317-274-8439. Corina van de Pol, Acufocus, Inc., 32 Discovery, Suite 200 Irvine, CA 92618, Tel: 949-585-9511, Fax: 949-585-9545. Khadeeja Pruett and Caron Kerr, GlaxoSmithKline Research and Development, Greenford Road, Greenford, Middlesex UB6 0HE, UK, Tel: +44-20-8422-3434, Fax: +44-20-8966-5285. James M. Veazey Jr, United States Army Medical Materiel Development Activity, 1430 Veteran’s Drive, Fort Detrick, MD 21702-5009, Tel: 301-619-1106, Fax: 301-619-2304. Ronnie Beboso, Chiltern (Early Phase) Limited Ninewells Hospital and Medical School Dundee DD1 9SY Scotland, U.K., Tel: +44 (0) 1382 646317, Fax: +44 (0) 1382 645606. Colin Ohrt, Walter Reed Army Institute of Research, 503 Robert Grant Avenue, Silver Spring, MD 20910, Tel: 301-319-9280, Fax: 301-319-9449.
|
REFERENCES |
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
