Doi:10.1016/s1473-3099(11)70065-2

Articles
Chloroquine for infl uenza prevention: a randomised,
double-blind, placebo controlled trial
Nicholas I Paton, Lawrence Lee, Ying Xu, Eng Eong Ooi, Yin Bun Cheung, Sophia Archuleta, Gerard Wong, Annelies Wilder Smith Summary
Background Chloroquine has in-vitro activity against infl uenza and could be an ideal candidate for worldwide Published Online
prevention of infl uenza in the period between onset of a pandemic with a virulent infl uenza strain and the development DOI:10.1016/S1473-
and widespread dissemination of an eff ective vaccine. We aimed to assess the effi
cacy of such an intervention.
See Online/Comment
Methods In this randomised, double-blind, placebo-controlled trial done at a single centre in Singapore, we randomly DOI:10.1016/S1473-
assigned (1:1) healthy adults to receive chloroquine phosphate (500 mg/day for 1 week, then once a week to complete 3099(11)70092-5
12 weeks) or matching placebo by use of a computer-generated randomisation list. Participants fi lled an online Yong Loo Lin School of
Medicine, National University
symptom diary every week, supplemented by daily diaries and self-administered nasal swabs when unwell.
of Singapore, Singapore
Haemagglutination-inhibition assays for infl uenza A (H1N1, H3N2) and B were done on blood samples taken at (N I Paton MD, L Lee MBBS,
baseline and after 12 weeks. The primary outcome was laboratory-confi rmed clinical infl uenza defi ned by specifi c S Archuleta MD, G Wong MBBS,
symptoms accompanied by infl uenza RNA on nasal swabs or a four-fold increase in haemagglutination-inhibition A Wilder Smith MD); Singapore
Clinical Research Institute,
titres over the 12-week study period. Analysis was by intention to treat. This trial was registered with ClinicalTrials.gov,
Singapore (Y Xu PhD,
number NCT01078779.
Y B Cheung PhD); Duke-NUS
Graduate Medical School,
Findings From November, 2009, to February, 2010, we recruited 1516 eligible participants. 1496 (96%) returned at Singapore (Y Xu, Y B Cheung,
E E Ooi PhD); DSO National
week 12 and were included in the effi
cacy analysis. Adherence to study intervention was 97%, and 94% of the Laboratories, Singapore
scheduled weekly diaries were completed. Eight (1%) of 738 participants had laboratory-confi rmed clinical infl uenza (E E Ooi); and Investigational
in the placebo group and 12 (2%) of 724 in the chloroquine group (relative risk 1·53, 95% CI 0·63–3·72; p=0·376). Medicine Unit, National
29 (4%) of 738 had laboratory-confi rmed infl uenza infection (symptomatic or asymptomatic) in the placebo group University Health System,
Singapore (G Wong)
and 38 (5%) of 724 in the chloroquine group (1·34, 0·83–2·14; p=0·261). 249 (33%) of 759 participants reported
adverse events (mostly mild) in the placebo group and 341 (45%) of 757 in chloroquine group (p<0·0001). Headache,
dizziness, nausea, diarrhoea, and blurred vision were more common in the chloroquine group, but rarely resulted in Trials Unit, 222 Euston Road,
treatment discontinuation. One serious adverse event (hepatitis) was possibly related to chloroquine.
London NW1 2DA, UK
nick.paton@ctu.mrc.ac.uk
Interpretation Although generally well tolerated by a healthy community population, chloroquine does not prevent
infection with infl uenza. Alternative drugs are needed for large-scale prevention of infl uenza.
Funding National Medical Research Council, Singapore.
Introduction
for the prevention of infl uenza could avert millions of The rapid spread and magnitude of the 2009 H1N1 deaths during the period between onset of a virulent infl uenza pandemic shows clearly that the current infl uenza pandemic and the development and worldwide armamentarium available for prevention of global dissemination of an eff ective vaccine.
infl Chloroquine, an antimalarial drug that has been in production often takes more than 6 months from widespread clinical use for more than 50 years, has broad-identifi cation of a new strain to widespread dissemination spectrum antiviral activity because it increases endosomal to the public, by which time the initial wave of a pH, which disrupts pH-dependent structural changes in pandemic is likely to have passed.1 Neuraminidase viral-synthesised proteins.4 Chloroquine has in-vitro inhibitors are eff ective in pre-exposure prophylaxis of activity against both H1N1 and H3N2 infl uenza strains at infl uenza,2 but such use of these drugs on a large-scale, concentrations that are achievable in vivo at the doses of long-term basis is not feasible for many reasons, chloroquine used for malaria prophylaxis and treatment including insuffi of connective-tissue diseases.5,6 Although chloroquine did cost, and risk of promoting resistance that would impair not prevent infl uenza infection or diminish the weight cacy in the treatment of infl uenza. Even with loss in mice infected with a mouse-adapted H1N1 strain, adequate supply of vaccines and drugs, many additional it did show some eff ectiveness in limiting viral replication challenges to prevention of pandemic infl uenza in ferrets infected with an adapted H3N2 strain.7 It might transmission in developing countries exist—especially be an ideal drug for worldwide pre-exposure prevention limitations of health-care infrastructure and human of infl uenza. Chloroquine can be taken once a week, is resources.3 A simple, well tolerated, internationally well tolerated and safe (including use in pregnant women accessible, and economical pharma cological intervention www.thelancet.com/infection Published online May 6, 2011 DOI:10.1016/S1473-3099(11)70065-2
Articles
worldwide for an immediate roll-out programme if randomised (ie, any household association with other eff ectiveness were to be established, is cost eff ective for participants was ignored). All trial investigators, clinical use in resource-poor countries, and has a putative and laboratory staff , and participants were masked to the mechanism of action that is unlikely to be compromised treatment allocation. A procedure for emergency by antiviral resistance. Several expert reviews have called unblinding of individual participants was established, for an assessment of generic drugs, such as chloroquine, although no unblinding was necessary during the trial.
for prevention of infl uenza.8,9 No trials of chloroquine for the prevention of infl uenza Procedures
have been done in man, but in view of its antiviral Demographic characteristics and a clinical history were properties and overall suitability as a prophylactic drug, recorded at baseline, and height, weight, and vital signs we postulated that it might be of value. Furthermore, the were measured. Blood was drawn for storage of serum. drug also has anti-infl ammatory and immunomodulatory Both study groups received identical capsules containing actions that might be benefi cial for prevention of either chloroquine phosphate 250 mg (equivalent to progression to severe disease, since the pathogenesis of 150 mg base) or lactose placebo (Beacons Pharmaceuticals, severe infl uenza is driven by an infl ammatory immune Singapore). Participants were instructed to take two response to the infection.8,10 We aimed to assess whether capsules once a day with food for the fi rst 7 days of the chloroquine, taken once a week by healthy individuals in trial, starting on the day after randomisation. After this the community during a pandemic, would protect against induction period, they took two capsules once a week, on infection with infl uenza or decrease the severity of a designated day of the week to complete a total course of infl uenza symptoms.
12 weeks. Participants were telephoned at the end of the fi rst week to check adherence to the study intervention, Study design and participants
Medical management of participants with infl uenza or The chloroquine for infl uenza prevention (CHIP) trial other illnesses was provided through the usual was a randomised, double-blind, placebo-controlled community channels, and health-care providers were clinical trial done at a single site (the Investigational asked to complete forms with details of diagnosis and Medicine Unit, National University Health System, any diagnostic tests done to identify infl uenza at each Singapore). Participants were recruited between clinic visit. If the participant decided to receive an November, 2009, and February, 2010, through a study infl uenza vaccination during follow-up they were asked website that provided information about the trial and an to return to the trial site to have an additional blood draw opportunity to register for a screening visit. The website before having the vaccination.
address was advertised around the university campus During follow-up, participants completed weekly and and hospital and widely in the national media. Follow-up daily diaries on a secure, password-protected trial website. Weekly diaries were completed on the day that was Eligible participants were male or female healthy designated for taking study treatment, and included volunteers aged 18–65 years. Exclusion criteria were a information about symptoms, visits to health-care history of psoriasis, porphyria cutanea tarda, epilepsy, workers, and adherence to the study intervention. myasthenia gravis, myopathy of any cause, cardiac Participants received reminder messages each week via arrhythmias, retinal disease, serious hepatic or renal email and mobile phone text message. They completed a disease, known glucose-6-phosphate dehydrogenase daily diary if they developed any infl uenza-like symptoms, defi ciency, current use of medication with known serious and continued this diary until the symptoms resolved. hepatotoxic eff ects or known interaction with chloroquine, This diary consisted of a checklist of symptoms, which severe depression, infl uenza vaccination within the past participants graded as none, mild, or moderate to severe, 3 months, acute clinical infl uenza at screening, and and a report of other symptoms, body temperature (a pregnancy or breastfeeding.
standard digital thermometer was provided), and absence For the CHIP trial protocol see
The trial protocol was approved by the ethics committee from work or school caused by sickness.
of the National Healthcare Group and all participants Participants were given a pack containing two nasal swabs and an RNA-stabilising compound (PrimeStore, Longhorn Vaccines and Diagnostics, San Antonio, USA) Randomisation and masking
and were instructed to take a swab from each nostril on The trial statistician prepared a computer-generated the second day of any illness associated with fever, runny randomisation list with a 1:1 ratio and a prepared block nose, or cough. After use, swabs were sealed and kept in size of ten. A pharmacist, independent of the trial team, the refrigerator until they were returned at the week 12 placed labels with four-digit study identifi ers on bottles of visit. Participants called the trial hotline after they had placebo and chloroquine according to the randomisation used the swabs, and a replacement pack was sent by post list. Randomisation was done by computer after baseline within 1 week for use during any subsequent symptomatic data had been gathered. Participants were individually episode. Pretrial testing showed that infl uenza RNA did www.thelancet.com/infection Published online May 6, 2011 DOI:10.1016/S1473-3099(11)70065-2
Articles
not degrade in PrimeStore media after 30 days, even when maintained at 37°C.
The second trial visit was at 12 weeks (window 3 days before to 21 days after) from the baseline visit. Adherence to study intervention was assessed by self-report of 20 hepatotoxic or other contraindicated drugs missed doses and by pill count. We obtained a participant history of infl uenza vaccination, infl uenza-like illness, rash or itch, and any other clinical events occurring since the baseline visit. Adverse events were graded with standard toxicity criteria,11 and a blood sample was drawn.
haemagglutination-inhibition assays to detect infection with infl uenza H1N1, H3N2, or infl uenza B with standard laboratory methods.12 We did these assays with infl uenza viruses isolated from local patients presenting with acute febrile illness to primary health-care clinics in 2009. Infection was defi ned as at least a four-fold increase in antibody titre from baseline to week 12. We retested 759 included in intention-to-treat analysis 757 included in intention-to-treat analysis specimens with a borderline increase, which did not meet diagnostic criteria, by use of a microneutralisation assay with a previously described method.13 Nasal swabs were tested by RT-PCR for infl uenza virus with a method that
targets the M gene of the virus;14 additional testing was Figure: Trial profi le
done with primers specifi c to the pandemic H1N1 strain.15
The primary endpoint was laboratory-confi rmed outcomes associated with severity of illness: symptomatic ned as the combination of clinical infl uenza; severe symptomatic clinical infl uenza symptoms of clinical infl uenza and laboratory-confi rmed (but with a temperature ≥37·8°C and at least fi ve infl nition used in moderate to severe clinical infl uenza symptoms uenza prevention trials.16,17 We defi ned clinical occurring on the same day); maximum number of infl uenza as a reported temperature of at least 37·2°C, symptoms of clinical infl uenza that were graded as with at least one respiratory symptom (sneezing, runny moderate to severe at the peak-illness day; presence of nose, blocked nose, sore throat, dry cough, coughing up individual constitutional symptoms (feeling feverish, phlegm, wheezing, shortness of breath) and at least one muscle aches, fatigue, headache, diarrhoea) that were constitutional symptom (feeling feverish, muscle aches, graded as moderate to severe; and the number of days fatigue, headache, diarrhoea) occurring on the same day. off school or work due to clinical infl uenza.
Laboratory-confi rmed infl uenza infection was defi ned as
one of the following test results: (1) PCR confi rmation of Statistical analysis
infl uenza on a nasal swab taken by the participant; We estimated that the rate of laboratory-confi rmed clinical
(2) PCR confi rmation or culture for infl uenza obtained infl uenza would be at least 10% over the 12-week study
from a nasopharyngeal swab taken by a health-care period, consisting of 5% incidence of seasonal infl uenza
practitioner in the community; (3) serological (based on Singapore national data18 and data from an
confi rmation by at least a four-fold increase in antibody infl uenza-prevention trial in the USA),16 and a 5%
titre on haemagglutination-inhibition or micro-
incidence of H1N1 pandemic infl uenza infection, which neutralisation assay for H1N1, H3N2, or infl uenza B was considered to be a conservative estimate in view of infection from baseline to week 12. For participants who the much higher transmissibility of the H1N1 strain than had H1N1 infl uenza vaccination during follow-up, the that of seasonal infl uenza.19 We reasoned that a 50% change in H1N1 titre at week 12 was disregarded, but the reduction in clinical cases would justify widespread change in H1N1 titre at a prevaccination follow-up blood clinical use of chloroquine and would be a realistic eff ect sampling was included in the analysis if available. We size in view of the 74% reduction seen with pre-exposure used a similar approach for participants who received prophylaxis with oseltamivir.16 We calculated that a sample seasonal infl uenza vaccination during the study.
size of 1500 participants (750 per group) would give a The main secondary endpoint was laboratory- power of 90% to detect a 50% reduction in incidence of confi rmed infl uenza infection, whether symptomatic or the primary endpoint at the 5% signifi cance level (adjusted asymptomatic (ie, without the requirement for clinical for 10% loss to follow-up and for household correlation, infl uenza symptoms). In those with laboratory- assuming at most two participants per household and an uenza infection, we also assessed intraclass correlation coeffi www.thelancet.com/infection Published online May 6, 2011 DOI:10.1016/S1473-3099(11)70065-2
Articles
cacy analyses were done on an intention-to- who delayed starting the study intervention for more treat basis, including all eligible participants who were than 7 days after randomisation, or who took less than randomised and who returned for a week-12 visit within 50% of the loading dose or less than 25% of the the specifi ed window period. A supplementary analysis maintenance dose. Safety analysis included all eligible was done on a per-protocol dataset that was based on the randomised participants. Analysis of indicators of intention-to-treat population, but excluded participants severity of illness was done on the subset of those with laboratory-confi rmed infl uenza infection.
Placebo (n=759)
Chloroquine (n=757)
We calculated relative risks with 95% CIs and two-sided p values using the exact method.20 Analyses were done on the basis of individual participants (as randomised), without taking into account the eff ects of clustering. The number of symptoms graded as moderate to severe and the number of days off school or work due to clinical infl uenza were compared between the two groups by Wilcoxon rank-sum tests. This trial was registered with ClinicalTrials.gov, number NCT01078779.
Role of the funding source
The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data and had fi nal responsibility for the Of 1516 eligible participants enrolled and randomly assigned to study treatment, 1462 (96%) returned for the week-12 visit (fi gure). Table 1 shows demographic and clinical characteristics for the two study groups. Adherence to study intervention was high (97% doses Data are number of participants (%) or median (IQR).
taken) and only 31 participants reported discontinuation Table 1: Baseline characteristics
(ten in the placebo group, 21 in the chloroquine group; p=0·048). 31 (2%) participants received infl uenza vaccination during the trial follow-up (16 in the placebo Chloroquine Relative risk
group and 15 in the chloroquine group; 23 received H1N1 vaccinations, fi ve seasonal, and three both).
Primary endpoint
A total of 17 521 weekly diaries (94% of those scheduled) Laboratory-confi rmed clinical infl uenza and 4037 elective daily symptom diaries were completed. Main secondary endpoint
120 (8%) participants reported one or more episodes of Laboratory-confi rmed infl uenza infection symptoms that met the defi nition of clinical infl uenza. Other secondary endpoint
There were 60 laboratory-confi rmed infl uenza infections detected by haemagglutination-inhibition testing. Nasal Data are number of participants (%) or relative risk (95% CI).
swabs were taken in 43% of symptomatic episodes. A total of 210 nasal swabs were obtained, from which Table 2: Effi
cacy analysis.
13 cases of infl uenza were detected (six were confi rmed Chloroquine
Relative risk
Number of days off school, college, or work due to clinical infl uenza Number of moderate to severe symptoms at the peak day of clinical-infl uenza symptoms Data are number of participants (%), relative risk (95% CI), or median (range). *n=23.†n=33. ‡n=26. §n=37.
Table 3: Analysis of clinical disease severity in participants with laboratory-confi rmed infl uenza infection
www.thelancet.com/infection Published online May 6, 2011 DOI:10.1016/S1473-3099(11)70065-2
Articles
hepatitis, endometrial carcinoma). The only event in the Chloroquine p value
chloroquine group judged to be possibly related to the study drug was symptomatic hepatitis starting about 5 weeks after initiation of chloroquine and 10 days after Adverse events with frequency ≥5% in any group the participant returned from a holiday on a nearby tropical island. Concentrations of hepatic transaminases were raised to more than ten times the upper limit of normal at presentation, and returned to normal 1 month after discontinuation of the study intervention. The hospital discharge diagnosis was viral hepatitis, but standard viral serology tests were negative. Discussion
We did not fi nd evidence that chloroquine, when given in a malaria prophylaxis dose, had protective effi laboratory-confi rmed clinical infl uenza or laboratory- asymptomatic) in this adult population (panel). We designed this trial to detect a level of protective cacy of chloroquine of at least 50%, which is the Table 4: Adverse events
minimum level of protection that we thought would be necessary to show for chloroquine to be accepted as a by serology, seven by PCR alone; 11 were H1N1). No worthwhile global intervention for prevention of additional laboratory-confi rmed infections were infl uenza transmission during a pandemic. Lower levels identifi ed by microneutralisation tests or by tests done of protection might be of academic interest, but would be in the community. Of 67 laboratory-confi rmed infections, unlikely to lead to widespread acceptance as a public health 51 were identifi ed as H1N1 infection alone, six as H3N2 intervention in a pandemic situation because of the infection alone, eight as infl uenza B infection alone, and probable diminished eff ectiveness of the intervention in a two were mixed. By combination of clinical and real-life setting (as opposed to a clinical trial), the possibility laboratory diagnostic criteria, 20 (1%) participants met of individual risk compensation (ie, decreased compliance the primary-endpoint defi nition of laboratory-confi rmed with other prevention measures because of perceived high-clinical infl uenza.
level protection from pharmacological prophylaxis), and The primary endpoint of laboratory-confi rmed clinical the possibility of competition for scarce health-care- uenza and secondary endpoint of laboratory- infrastructure resources, which might impede the roll-out confi rmed infl uenza infection, whether symptomatic or of other infl uenza risk-reducing strategies.
asymptomatic, did not diff er between groups (table 2). Although continuing local transmission of pandemic We did not identify a signifi cant diff erence between the H1N1 strain infl uenza in Singapore (along with increasing two treatment groups for the number of clinic visits or H3N2 and B strain transmission) during follow-up was admissions to hospital (data not shown). Results were substantial, it did not yield the number of clinical infl uenza similar when the analysis was repeated, taking into infections that we had anticipated—partly because the account the eff ects of clustering, and when repeated transmission had declined after the peak of the pandemic with the 1431 participants in the per-protocol dataset and because many of the infections that occurred were (data not shown).
asymptomatic. As a result, estimates of protective effi We recorded no signifi cant diff erence between the groups in clinical severity of infl uenza (various Panel: Research in context
defi nitions) in the 67 participants who had laboratory-confi rmed infl uenza infection (table 3).
Systematic review
More participants in the chloroquine group reported We searched PubMed from inception to March 7, 2011, with the terms “chloroquine” OR adverse events than in the placebo group (table 4), “hydroxychloroquine” AND “infl uenza” AND “prevention” OR “treatment”; no langauge although these events were judged as mild (or grade 1) in restrictions were used. We identifi ed no previous clinical trials of chloroquine in the almost all cases. Headache, dizziness, nausea, and prevention or treatment of infl uenza.
diarrhoea were more common in the chloroquine group, Interpretation
but rarely resulted in treatment discontinuation (table 4).
We showed that chloroquine taken once weekly does not have protective effi Eight serious adverse events were reported, fi ve in the infl uenza infection, and does not appear to diminish the symptoms of established placebo group (haematemesis, renal stone, acute infection. An eff ective drug is needed that can be used internationally on a large scale for appendicitis, ankle fracture, prolapsed intervertebral disc) and three in the chloroquine group (maxillary cyst, www.thelancet.com/infection Published online May 6, 2011 DOI:10.1016/S1473-3099(11)70065-2
Articles
of chloroquine that we obtained were imprecise. dose given each week should maintain blood Nevertheless, even with the most optimistic estimates concentrations achieved in the once-a-day induction from the 95% CI, we showed that chloroquine has no more phase28 and should be roughly the half maximal inhibitory cacy against laboratory-confi rmed concentration (IC50) of the H1N1 and H3N2 strains used clinical infl uenza and no more than 17% protective effi cacy in the in-vitro studies. Furthermore, the drug is highly against the more precise measure of laboratory-confi rmed concentrated in many tissues with sustained use infection (symptomatic or asymptomatic). Thus, according including the lungs, and in white cells and macrophages, and such intracellular concentrations might be more rule out a useful role of chloroquine as a general relevant for activity of the drug against infl uenza viral intervention to prevent pandemic infl uenza transmission. entry.5,6,28,29 At the dose given, headache, dizziness, nausea, Furthermore, the point estimate of effi cacy that we diarrhoea, and blurred vision were signifi cantly more obtained suggests that chloroquine might actually increase common in participants taking chloroquine than in those the risk of infl uenza infection, although we believe that taking the placebo, although these were mainly transient this increased risk is probably a chance fi nding. However, during the fi rst week of treatment, and rarely led to in view of the diversity of immunological actions of treatment discontinuation. Although a higher dose of chloroquine, it might increase the risk of infl uenza chloroquine could have been sustained for a longer infection through some as yet unknown pathway.
period, despite these side-eff ects, this dose would Several possible reasons exist as to why the effi cacy of probably have restricted the acceptability of the chloroquine against infl uenza shown in vitro did not intervention for many people. The case of symptomatic translate to effi cacy in vivo.5,6 First, infl uenza strains vary hepatitis, which was possibly related to chloroquine in their pH requirements for viral entry and hence (although symptomatic hepatitis is rare with this drug), susceptibility to chloroquine, and effi cacy of chloroquine reinforces the belief that sustained higher doses would against the H1N1 pandemic strain might have been low need systematic laboratory monitoring, which would (the predominant strain in circulation during the trial). seriously limit the generalisability of the intervention. However, in-vitro studies have shown effi cacy of this Thus, although we cannot rule out effi drug against H1N1 infection,5 and our trial provided no dose of chloroquine than the one we tested, we believe evidence of an eff ect of chloroquine against non-H1N1 that the tolerability and safety fi ndings in this study infections either, so this explanation seems unlikely. Our confi rm that the dose selected was appropriate for an results are consistent with fi ndings from studies of intervention to prevent infl dengue infection in which the in-vitro eff ects of generalisable worldwide.
chloroquine did not translate into in-vivo effi Chloroquine has anti-infl ammatory properties that although they have done so in other viral infections, might help to stop the virus-induced infl ammation that especially HIV infection.22–27 drives disease pathogenesis after the fi rst few days.8,10 Second, the possibility of bias should be considered. Treatments directed against the host response to Although chloroquine is easily recognised by its bitter infl uenza are notable, because they might have the taste, this taste was masked eff ectively by formulating the potential to decrease the mortality associated with severe drug in capsules. The use of blinding and the high rates infl uenza at a stage when antiviral drugs are ineff ective. of completion of symptom diaries and the high proportion However, we saw no evidence of a diff erence in severity of participants who returned for follow-up visits make of disease in people who acquired infl uenza on signifi cant bias unlikely to account for the result. chloroquine or placebo that might support this notion.
Furthermore, chloroquine was not eff ective against For short-term treatment of infl uenza, high doses of laboratory-confi rmed infl uenza infection, an entirely chloroquine given every day could possibly be used, and objective outcome variable.
Third, despite very high levels of adherence to study this approach might still be worth testing. Furthermore, intervention, tissue levels of chloroquine might have absence of evidence of benefi been inadequate to protect against infl uenza viral entry. monotherapy does not preclude the possibility of activity The choice of chloroquine dose for this trial was when used in combination with standard antiviral drugs, pragmatic. We took into account the high degree of and this might merit further investigation.
convenience and tolerability that would be needed for We studied healthy volunteers, most of whom were chloroquine, if shown to be eff ective, to gain widespread Chinese. Although our fi nding that chloroquine does not acceptance as an infl uenza pre-exposure prophylaxis have protective effi regimen taken for many months by healthy members of probably generalisable to other populations, its eff ects on the community, and the safety factors that make it the host immune response to infl uenza, and hence acceptable to give a dose of chloroquine each week (but disease severity, might be greater in an older population not a sustained high dose given every day) without with more comorbidity, or in diff erent ethnic groups.
undertaking regular laboratory monitoring. In view of An eff ective, well-tolerated, cheap, and widely available the long half-life of chloroquine, we calculated that the prophylactic drug is needed that can be used as a part of www.thelancet.com/infection Published online May 6, 2011 DOI:10.1016/S1473-3099(11)70065-2
Articles
an international public health response to infl uenza. 11 Division of AIDS N. Division of AIDS table for grading severity Identifi cation of this compound needs to be done with of adult adverse experiences. Rockville: National Institute of Allergy and Infectious Diseases, 1992.
some urgency before a pandemic occurs (or recurs) with 12 WHO. WHO manual on animal infl uenza diagnosis and a virulent strain of infl uenza virus.
surveillance. Geneva: World Health Organization, 2002.
Contributors
13 Rowe T, Abernathy RA, Hu-Primmer J, et al. Detection of antibody to avian infl uenza A (H5N1) virus in human serum by using a NIP designed the trial, led the trial as the chief investigator, and wrote combination of serologic assays. J Clin Microbiol 1999; 37: 937–43.
the fi nal report. LL was the principal investigator of the trial, with overall 14 Fouchier RA, Bestebroer TM, Herfst S, Van Der Kemp L, responsibility for the conduct of the trial at the site, and for medical Rimmelzwaan GF, Osterhaus AD. Detection of infl uenza A viruses oversight of trial implementation. GW was responsible for the from diff erent species by PCR amplifi cation of conserved sequences operational aspects of trial implementation. SA contributed to the in the matrix gene. J Clin Microbiol 2000; 38: 4096–101.
clinical and operational implementation of the trial. YX and YBC were 15 Wang R, Sheng ZM, Taubenberger JK. Detection of novel (swine responsible for analysing the data. EEO was responsible for the origin) H1N1 infl uenza A virus by quantitative real-time reverse laboratory analyses. AWS was responsible for recruitment of transcription-PCR. J Clin Microbiol 2009; 47: 2675–77.
participants. All authors contributed to trial design and interpretation 16 Hayden FG, Atmar RL, Schilling M, et al. Use of the selective oral of data and reviewed the fi nal report.
neuraminidase inhibitor oseltamivir to prevent infl uenza.
N Engl J Med 1999; 341: 1336–43.
Confl icts of interest
We declare that we have no confl icts of interest.
17 Welliver R, Monto AS, Carewicz O. Eff ectiveness of oseltamivir in preventing infl uenza in household contacts: a randomized Acknowledgments
controlled trial. JAMA 2001; 285: 748–54.
We thank Jin Wang, Vivien Fu, Tek Seng Tan, and all the staff of the 18 Ng TP, Pwee KH, Niti M, Goh LG. Infl uenza in Singapore: Investigational Medicine Unit, National University Health System assessing the burden of illness in the community. (NUHS), Singapore, who ran the trial; Damien Hong, Xuan Hui Ng, Ann Acad Med Singapore 2002; 31: 182–88.
Wen Li, and all the staff at Singapore Clinical Research Institute, 19 Fraser C, Donnelly CA, Cauchemez S. Pandemic potential Singapore, who did the data entry and management of the trial; of a strain of infl uenza A (H1N1): early fi ndings. Science 2009; Teck Sin Lim and his staff at KooPrime for developing the online website 324: 1557–61.
and database for participant data entry; Shiqin Howe, Wen Yan Lim, and 20 Greenwood B. Interpreting vaccine effi all staff at Duke-NUS who prepared the blood samples and undertook 40: 1519–20.
the laboratory assays, and Gary Lau and Lynn Tang for their assistance 21 Tricou V, Minh NN, Van TP, et al. A randomized controlled trial in virus propagation; Richard Chiam for his assistance with organising of chloroquine for the treatment of dengue in Vietnamese adults. participant recruitment, and the Corporate Communications team at PLoS Negl Trop Dis; 4: e785.
NUHS for their assistance in advertising the trial. This trial was funded 22 Chiang G, Sassaroli M, Louie M, Chen H, Stecher VJ, Sperber K. by National Medical Research Council, Singapore.
Inhibition of HIV-1 replication by hydroxychloroquine: mechanism of action and comparison with zidovudine. Clin Ther 1996; References
18: 1080–92.
Ferguson NM, Cummings DA, Fraser C, Cajka JC, Cooley PC, 23 Sperber K, Kalb TH, Stecher VJ, Banerjee R, Mayer L. Inhibition of Burke DS. Strategies for mitigating an infl uenza pandemic. human immunodefi ciency virus type 1 replication by Nature 2006; 442: 448–52.
hydroxychloroquine in T cells and monocytes. 2 Jeff erson T, Jones M, Doshi P, Del Mar C, Dooley L, Foxlee R. AIDS Res Hum Retroviruses 1993; 9: 91–98.
Neuraminidase inhibitors for preventing and treating infl uenza 24 Savarino A, Gennero L, Chen HC, et al. Anti-HIV eff ects of in healthy adults. Cochrane Database Syst Rev 2010; 2: CD001265.
chloroquine: mechanisms of inhibition and spectrum of activity. Oshitani H, Kamigaki T, Suzuki A. Major issues and challenges AIDS 2001; 15: 2221–29.
of infl uenza pandemic preparedness in developing countries. 25 Vincent MJ, Bergeron E, Benjannet S, et al. Chloroquine is a potent Emerg Infect Dis 2008; 14: 875–80.
inhibitor of SARS coronavirus infection and spread. Virol J 2005; Savarino A, Boelaert JR, Cassone A, Majori G, Cauda R. Eff ects of chloroquine on viral infections: an old drug against today’s 26 Sperber K, Chiang G, Chen H, et al. Comparison of diseases? Lancet Infect Dis 2003; 3: 722–27.
hydroxychloroquine with zidovudine in asymptomatic patients Ooi EE, Chew JS, Loh JP, Chua RC. In vitro inhibition of human infected with human immunodefi ciency virus type 1. infl uenza A virus replication by chloroquine. Virol J 2006; 3: 39.
Clin Ther 1997; 19: 913–23.
Di Trani L, Savarino A, Campitelli L, et al. Diff erent pH requirements 27 Sperber K, Louie M, Kraus T, et al. Hydroxychloroquine treatment are associated with divergent inhibitory eff ects of chloroquine on of patients with human immunodefi ciency virus type 1. human and avian infl uenza A viruses. Virol J 2007; 4: 39.
Clin Ther 1995; 17: 622–36.
Vigerust DJ, McCullers JA. Chloroquine is eff ective against 28 Wetsteyn JC, De Vries PJ, Oosterhuis B, Van Boxtel CJ. The infl uenza A virus in vitro but not in vivo. pharmacokinetics of three multiple dose regimens of chloroquine: Infl uenza Other Respi Viruses 2007; 1: 189–92.
implications for malaria chemoprophylaxis. Fedson DS. Confronting an infl uenza pandemic with inexpensive Br J Clin Pharmacol 1995; 39: 696–99.
generic agents: can it be done? Lancet Infect Dis 2008; 8: 571–76.
29 Ducharme J, Farinotti R. Clinical pharmacokinetics and Fedson DS. Meeting the challenge of infl uenza pandemic metabolism of chloroquine: focus on recent advancements. preparedness in developing countries. Emerg Infect Dis 2009; Clin Pharmacokinet 1996; 31: 257–74.
15: 365–71.
10 de Jong MD, Simmons CP, Thanh TT, et al. Fatal outcome of human infl uenza A (H5N1) is associated with high viral load
and hypercytokinemia. Nat Med 2006; 12: 1203–07.
www.thelancet.com/infection Published online May 6, 2011 DOI:10.1016/S1473-3099(11)70065-2

Source: http://www.kooprime.com/dl/CHIPLancet.pdf