Microsoft word - unim study descriptions for website _1_

A Randomized Controlled Trial of Male Circumcision to Reduce HIV Incidence in Kisumu,
Robert C. Bailey,1 Stephen Moses,2 Corette B. Parker,3 Kawango Agot,4 Ian Maclean,5 John N.
Krieger,6 Carolyn F. M. Williams,7 Richard T. Campbell,1Jeckoniah O. Ndinya-Achola8

PI Contact Information: rcbailey@uic.edu
Summary
Background
Male circumcision may provide significant protection against HIV-1 acquisition. A randomized controlled trial
was conducted in Kenya to test the protective effect of male circumcision, and to assess safety and changes in
sexual behaviour.
Methods
2784 men aged 18-24 years were randomized to circumcision or control, and assessed through HIV testing,
medical examinations and behavioural interviews during follow-ups at 1, 3, 6, 12, 18 and 24 months. HIV sero-
incidence was estimated using the Kaplan-Meier method, in an intent-to-treat analysis. An as-treated analysis was
also performed, using Cox regression with a time-dependent covariate for circumcision status at follow-up.
Behavioural change was assessed using generalized estimating equations.
Findings
The trial was stopped on December 12, 2006, after a third interim analysis reviewed by the Data and Safety
Monitoring Board. The median length of follow-up was 24 months. 22 and 47 participants tested HIV positive in
the circumcision and control groups, respectively. The relative risk of circumcision was 0.47 (95% CI: 0.28,
0.78), corresponding to a reduction in HIV incidence of 53%. Adjusting for non-adherence to treatment and
excluding four men determined to be seropositive at enrollment, the protective effect was 60% (95% CI: 0.32,
0.77). Adverse events were few (1.5%) and resolved quickly. Behavioural risk compensation after circumcision
was present but small in magnitude.
Interpretation
Male circumcision is confirmed as significantly reducing risk for HIV acquisition among young men in Africa.
Voluntary, safe and affordable circumcision services should be integrated with other HIV preventive interventions
and where appropriate provided as expeditiously as possible.

Summaries of Sub-studies and Principal Investigators are listed below:
1. Resistance and Susceptibility to Chlamydia trachomatis (Robert Brunham, MD)

Purpose:
Little is known about worldwide variation in Chlamydia trachomatis, a sexually transmitted disease that
affects 5-25% of adults. Genetic variation may be associated with different levels of transmissability and
virulence. The purpose of this study is to obtain urethral swabs from men in Kisumu, Kenya who have a
Chlamydia trachomatis infection. The genetic compostition of the specimens will be studied in
Winnipeg, Canada and add to a database of chlamydia collected from various sites around the world.
Progress:
Page 2 of 16
However, the specimens collected for this study were discovered to contain non-viable cells, and as a result, the study team could not perform the necessary analyses in order to determine the immunoepidemiological correlates of incidence C. trachomatis cases. UIC terminated the research in January 2008. 2. The Effect of Male Circumcision on Penile HPV infection in Kisumu, Kenya
(Jennifer S. Smith, PhD, MPH)

Co-investigators / Co-authors:
Stephen Moses, MD, MPH
Ian W. Maclean, PhD
J.O. Ndinya-Achola, MBChB
KawangoAgot, PhD, MPH
Robert C. Bailey, PhD, MPH
Michael Hudgens, PhD
Meijer, Chris J.L.M., MD, PhD
Nicolas F. Schlecht, PhD, MSc
Peter J.F. Snijders, PhD
MaaikeBleeker, PhD
Walter Jaoko, MBchB, PhD

PI Contact information: JenniferS@unc.edu

Purpose:
Human papillomavirus (HPV) is the central cause of invasive cervical cancer (ICC) in women, and other
anogenital cancers, including anal, penile, vaginal and vulvar cancers. Public health intervention
strategies are needed to reduce the genital acquisition of high-risk HPV types to prevent these cancers that
affect women and men. Given that men are important in transmitting HPV to their female sexual partners,
interventions that reduce HPV acquisition and persistence in men could be an important HPV prevention
strategy benefiting both men and women. Limited cross-sectional data suggest that male circumcision
may reduce the carriage of penile HPV infection. However, proof that male circumcision reduces HPV
penile infection acquisition and/or persistence can only be obtained through the conduct of a prospective,
randomized clinical trial (RCT).
This study is investigating the effect of male circumcision on the risk and natural history of penile HPV
infection in young and sexually active men in Kisumu, Kenya. This HPV project proposes to determine
the type-specific HPV DNA prevalence in penile exfoliated cell samples collected as part of the RCT of
male circumcision. An estimated 2,442 men are expected to contribute penile exfoliated cell specimens
from two anatomical sites (the shaft and glans/coronal sulcus). These samples are being collected at
enrollment and at the 6-, 12-, 18- and 24- month follow-up visits. Specimens will be tested using a
sensitive PCR assay to detect a wide-range of HPV DNA types. HPV viral load of samples positive for
HPV types 16, 18, 31, 33 and 45 will be determined using a real-time PCR. A new clinical examination (a
visual inspection with 3% acetic acid and aided magnification) will be added to this ongoing RCT at the
final 24-month visit to determine the presence of penile lesions (i.e., flat lesions, papular lesions, pearly
penile papules or condolymata acuminate).
Page 3 of 16

Study results will provide the most definitive evidence to date regarding the effect of male circumcision
on HPV infection and penile lesions in men. If male circumcision is found to be effective and safe in
reducing HPV infection or penile lesions in men, it could in turn reduce HPV prevalence and associated
cervical neoplasia in women.
Progress:
The baseline HPV results from 2,705 men were published in the International Journal of Cancer (2010).
The HPV DNA testing results from both the glans and the coronal sulcus indicated that HPV prevalence
rates were 47% in the glans and coronal sulcus, and 19% in shaft specimens. Of the glans specimens, 57%
were beta-globin positive, and of the shaft specimens, 36% were beta-globin positive.
In addition, risk factors for penile HPV DNA positivity among men with data on covariates of interest
were examined. Age was not significantly associated with a higher HPV positivity. The strongest risk
factors for penile HPV infection included laboratory-diagnosed C. trachomatis or N. gonorrhea infections,
less frequent bathing, and a self-reported history of STIs. Risk factors were similar when analyses were
limited to men who were beta-globin positive.
An analysis of whether HPV infection at baseline influences acquisition of HIV was performed and the
results were published in the Journal of Infectious Diseases (2010). Estimated cumulative incidence of
HIV infection by 42-months was 5.8% [95% CI 3.6, 7.9] in men with HPV-positive glans specimens
versus 3.7% [1.8, 5.6] in men with HPV-negative glans specimens (p=0.01). In multivariate models, the
risk of HIV infection over 42 months was 1.8 times higher in men with HPV at baseline compared to
those without. HIV rates were similar in men with high-risk HPV compared to low-risk HPV. These
results suggest an independent, increased risk of HIV seroconversion among men who are HPV positive.
If this finding is confirmed in other studies, HPV prevention, particularly through immunization, could be
another tool for HIV prevention.
Results also indicate that male circumcision is highly protective against flat penile lesions in this cohort.
The prevalence of flat lesions in circumcised men was 0.7%, while the prevalence in non-circumcised
men was significantly higher at 26.0%. This translates to a protective odds ratio of 0.02 for circumcised
men against penile flat lesions. This manuscript was published in the International Journal of Cancer
(2011).
A manuscript entitled, “Multiple human papillomavirus infections and type competition in men," is in
press at the Journal of Infectious Diseases. Half of all men were HPV positive, of whom 57% had
multiple HPV types. Findings indicate that men with four or more HPV types more often than expected if
infections were independent. No negative associations between individual HPV types were observed.
Thus, HPV types that are candidates for potential HPV type competition in men were not identified.
A manuscript on the incidence and risk factors of penile human papillomavirus infection among
uncircumcised men is currently under review for publication in the International Journal of Cancer.
Results show that HPV incidence was higher in this study compared to incidence rates previously
reported in the literature. Reducing the number of sexual partners and promoting penile hygiene could
reduce HPV acquisition among uncircumcised men.
Page 4 of 16
A manuscript on baseline HPV infections and the future acquisition of other HPV types is pending submission to the Journal of Infectious Diseases. Among 1,097 HIV-negative, uncircumcised men, 2,303 incident HPV infections were detected over 2,534 person-years of follow-up. Compared to men without the type-specific infection at baseline, men with an infection at baseline had a similar or slightly shorter time to acquisition of any other HPV and HR-HPV types. In conclusion, acquisition of individual HR-HPV types varied by baseline HPV type; however, there was no clear evidence of shorter time to acquisition among men without vaccine-relevant HPV infections that would suggest the potential for HPV type replacement. 3. A Study of Early HIV Infection in the Kisumu Male Circumcision Cohort.(Keith Fowke, PhD)

Co-investigators / Co-authors:
Alan Landay, PhD
Robert C. Bailey, PhD, MPH
Stephen Moses, MD, MPH
Corette Parker, DrPH
Walter Jaoko, MBchB, PhD

PI Contact information: fowkekr@cc.umanitoba.ca

Purpose:
This study focuses on the role of immune activation and CD4 polymorphisms in HIV infection and
disease progression. The goal of this Early HIV Infection protocol is to further study those individuals
within the Circumcision Trial who have seroconverted to HIV during the course of the
study.Seroconversion is an end point for the circumcision trial; however, this protocol would follow
individuals who seroconvert to HIV for an additional 24 months from the date seroconversion is detected.
The first twelve months of an HIV infection is called the “early HIV infection” period and this study will
focus on that period as well as the next 12 months.
There is a paucity of data concerning the role of immune activation in early HIV infection and none that
take into account the unique feature of elevated immune activation observed in Africans. Genetic studies
of the CD4 polymorphism have never been performed in men from East Africa, nor in Luo, the
predominant ethnic group in Kisumu. Through the Kisumu Circumcision Trial, this study’s goals are to
determine the role of immune activation in early HIV infection and determine if the CD4 polymorphism
is a risk factor for HIV seroconversion or disease progression.
Progress:
This study was completed and closed onNovember 28, 2011. The researchers investigated whether certain
polymorphisms with genes around the CD4 locus were associated with HIV seroconversion. This study
focused on the polymorphisms in the CD4, GNB3, and MLF2 genes. Findings show that the CD4 and
GNB3 polymorphisms were not found at increased frequency in the study subjects who seroconverted
versus those who did not seroconvert. However, the MLF2 gene polymorphism CT/TT was found at a
higher prevalence among the seroconverters (P=0.0012) and was independent of circumcision status
(circumcised p=0.0121, uncircumcised p=0.0256).
Page 5 of 16
The SNPs included in the analysis were CD4 C868T (rs28919570), GNB3 C825T (rs5443) and a previously unpublished SNP in MLF2 designated C89T. All polymorphisms were found to be in Hardy-Weinberg Equilibrium (p>0.3 for all). The CD4 868T allele was found at a frequency of 18.9%. There was no significant differences in allele frequency between seroconverters and non seroconverters (p=0.58). Kaplan-Meier survival analysis showed no differences in time to seroconversion between genotypes (p=0.55). Circumcision status did not affect the results. The GNB3 825T allele frequency was 82% overall and did not differ by circumcision status. There were no significant differences in allele frequency between seroconverters and non seroconverters.There was no significant difference in time to seroconversion but the 825TT genotypes trended towards faster seroconversion (p=0.09). The MLF2 89T allele frequency was 5.8%. The 89 CT genotype was significantly enriched among seroconverters (p=0.0049) regardless of circumcision status. 89CT/TT subjects seroconverted significantly faster than people who do not carry the T allele (p=0.0012) regardless of circumcision status. In conclusion, despite the previously observed association with susceptibility to HIV infection, no
association between the CD4 C868T and GNB3 C825T SNPs and HIV acquisition was observed in this
population. The fact that previous study was performed in female commercial sex workers may partially
explain that difference. Indeed, the route of acquisition, number of exposures/different partners and rates
of sexual transmitted infection might all be factors that can explain the observed differences.
Interestingly, the novel MLF2 C89T SNP was significantly associated with seroconversion. Little is
known about the function of ubiquitously expressed MLF2 protein, although it has high sequence
homology to MLF1 (an oncogenic protein in humans). An animal study recently showed that MLF2 is
indeed involved in genetic susceptibility to infectious disease. Further studies need to be performed in
other to understand the function of MLF2 in the human immune system as well as in role in the
acquisition to HIV.
The final manuscript reporting these results is being submitted for publication.
4. A Study of Urethral Swabs for Chlamydia Trachomatis Culture and Genomic Analysis from
Men Enrolled in the Kisumu Male Circumcision Trial. (Grant McClarty, PhD)

Co-investigators / Co-authors:
Ian Maclean, PhD
Robert Bailey, PhD
J.O. Ndinya-Achola, MBChB
Stephen Moses, MD, MPH
PI Contact information:
Dr. Grant McClarty
Dept of Medical Microbiology
University of Manitoba
504-730 William Avenue
Page 6 of 16
Winnipeg, Manitoba, Canada
R3E0W3

Purpose:
The purpose of this study was to obtain urethral swabs from men in Kisumu, Kenya who have a
Chlamydia trachomatis infection. The genetic compostition of the specimens were studied in Winnipeg,
Canada and added to a database of chlamydia collected from various sites around the world. We cultured
chlamydia from the urethral swabs sent to Winnipeg from Kisumu. Positive cultures were then PCRed
and the OMP 1 and plastizy zone loci were sequenced.
Progress:
This study was completed and closed on September 30, 2008. In agreement with the study’s original
hypothesis, findings show significant genomic variation in the plasticity zone of identical omp 1 typed
clinical isolates. Further studies are needd to see if these genetic variations give rise to changes in in vivo
pathogenicity, as assessed in a mouse genital tract model of chlamydial disease. If they do, sequencing the
entire genomes to verify precisely where and how many genetic changes there are may have to be
performed.
5. Investigation of the Role of FREM1 (FRAS Related Extracellular Matriz 1) and HLA Class I and II
Antigens in HIV-1 Infection in Men Enrolled in the Kisumu Male Circumcision Cohort
(Frank Plummer and Ma Luo)
Co-investigators / Co-authors:
Ian Maclean, PhD
Robert Bailey, PhD
J.O. Ndinya-Achola, MBChB
KawangoAgot, PhD

Contact Information:
Ian Maclean (main contact):
imaclea@ms.umanitoba.ca

Purpose:
The purpose of the study is to investigate the role of Frem1 in HIV-1 seroconversion in men enrolled in a clinical trial studying male circumcision and HIV infection in Kisumu, Kenya. The archived blood samples will be shipped to Winnipeg, Canada where the DNAs will be isolated. Frem1 gene segments will be amplified, sequenced and genotypes for SNPs (Single nucleotide polymorphism) within the Frem1 region. Statistical analysis will be carried out to examine the association of specific Frem1 genotype with HIV-1 seroconversion. HLA class I and class II genes will also be typed to control for the existence of HLA alleles that were identified as associated with resistance or susceptibility to HIV-1 infection in a Nairobi sex worker cohort. Even though very little is known about Frem1, the facts that it is a component of extracellular matrix and membrane, expressed in dermis and involved in calcium, protein and sugar binding, make it a very promising candidate gene that is involved in HIV-1 resistance. These results suggest that Frem1 is in Page 7 of 16
some way involved in resistance/susceptibility to HIV-1 infection. In addition to conducting experiments
investigating the involvement of Frem1, we would like to see whether the association of specific genotype
of Frem1 with resistance to HIV-1 infection could be confirmed in other populations, such as other
African and Caucasian populations and in males. The male circumcision cohort is a good population to
examine the association of Frem1 with HIV-1 infection in male because it consists of young males and
there are about 70 seroconverters in the cohort so far. We would like to genotype male patients for the
two SNPs of Frem1 using archived frozen blood samples.
HLA class I and class II genes present antigens to CD8+ and CD4+ T cells and are important in immune
recognition. Genotyping HLA class I and class II genes in the circumcision cohort, while studying the
involvement of Frem1 in HIV-1 seroconversion not only can help to clarify the role of Frem1, but also
that of HLA alleles identified in PumwaniSexworker cohort.
Progress:
This study uses frozen archived blood samples which will be analyzed to examine the association
between a particular FREM1 SNP genotype and HIV susceptibility and resistance. To date, 2000 whole
blood specimens from 2000 clients have been shipped to Winnipeg, Canada. 1532 whole blood specimens
have been processed with 1500 yielding quality DNA analysis. Genotyping of HLA DPA and HLA DPB
have been conducted on 1385 specimens and HLA DRB on 1500. Staff has begun the genotyping of the
FREM SNPs”1889” and “1552”. The quality of gDNA from the 1500 whole blood specimens was
adequate for the analysis of HLA DPA and HLA DPB however re-extraction to produce better quality
gDNA has been done on 606 whole blood specimens resulting in 265 specimens with enough DNA to
produce quality PCR products for the analysis of the FREM1 SNP rs 1552896 and the FREM1 SNP rs
1889050. At this time no sequence analysis has been done on these PCR amplicons. No manuscripts have
yet been written on the data from this study.
6. Behavioral Disinhibition Associated with Male Circumcision in Kenya.
(Christine Mattson, Ph.D.)
Co-investigators / Co-authors:
Robert C. Bailey, PhD, MPH
Stephen Moses, MD, MPH
KawangoAgot, PhD, MPH
JeckoniahNdinya-Achola, MBchB
Daniel Bennett, PhD
Nicholas Wilson, Ph.D, MPA/ID
PI Contact Information:
christine.mattson@gmail.com

Purpose:
Based on evidence from other HIV interventions, there is concern that male circumcision (MC) might
illicit increases in high-risk behavior due to false perceptions of complete protection, despite education
and rigorous risk-reduction counseling. This increase in risky behaviors after the implementation of an
intervention has been termed “behavioral disinhibition.” We propose a nested cohort study of behavioral
Page 8 of 16
disinhibition among the men participating in the RCT of MC in Kisumu, Kenya. We will prospectively
recruit a sub-sample of men participating in the RCT to undergo in-depth interviews lasting up to 90
minutes before circumcision and then at the 6, 12, and 24 month follow-up visits. The information
obtained in these interviews will allow a more effective evaluation of sexual risk behavior before and
after MC, as well as changes in risk behavior by circumcision status.
Progress:
The authors developed an 18 item sexual risk propensity scale to evaluate risk compensation among
participants. This statistical approach can be used to represent each respondent's level of sexual risk
behavior as the sum of his responses on multiple dichotomous and rating scale (i.e. ordinal) items. This
summary "score" can be used to summarize information on many sexual behaviors or to evaluate changes
in sexual behavior with respect to an intervention. Further, we established criterion validity by
demonstrating a statistically significant association between the risk scale and the acquisition of incident
sexually transmitted infections (STIs) at the 6 month follow-up and HIV at the 12 month follow-up visits.
This method has broad applicability to evaluate sexual risk behavior in the context of other HIV and STI
prevention interventions (e.g. microbicide or vaccine trials), or in response to treatment provision
Additionally, longitudinal analyses indicated no statistically significant differences between sexual risk
propensity scores or in incident infections of gonorrhea, chlamydia, and trichomoniasis between
circumcised and uncircumcised men. Lastly, analyses of risk compensation are continuing to investigate
whether men’s perceptions of change in risk are associated with their behaviors.At this time additional
analyses are investigating the incidence, prevalence, and factors associated with concurrent sexual
partnerships and are looking at the belief in the protective effect of male circumcision as it relates to
actual risk behaviors. Three additional manuscripts based on the data from this study are in progress.


7. A Study of GIS Applied to the UNIM Trial in Kisumu, Kenya
(Nelli Westercamp, Ph.D. Candidate)
Co-investigators / Co-authors:
Robert C. Bailey, PhD, MPH
Stephen Moses, MD, MPH
KawangoAgot, PhD, MPH
Ian MacLean, PhD
JeckoniahNdinya-Achola, MBchB

PI Contact information: nweste1@uic.edu
Purpose:
The purpose of this substudy is to observe the geographic distribution of cases with sexually transmitted
infections (STI) and high risk practices among trial participants, to investigate the spatial relationship
between health care facilities and STI cases, and to assess more in depth the relationship between the
socioeconomic status (SES) and risk of having an STI. This substudy is designed to improve the
understanding of the demographic characteristics of the parent-study participants. Analysis will focus on
the geographic distribution of STI cases and risk behaviors.
Page 9 of 16
Progress:
This study was completed and closed on November 24, 2008. Results show that the study population was not uniformly distributed throughout Kisumu. While the majority of participants came from the Kisumu municipality and its surroundings, the two standard distance deviations for this study population covered a larger area including the three neighboring districts. The SES composite measure appears to not only measure SES but also account for urban/rural status. Validity of the SES measure is observationally confirmed by cluster analysis. No patterns were observed in the distribution of STI cases and behavioral risk factors confirming findings from a recent study based in Nyanza Province (Voeten et al., 2004) In conclusion, GIS and spatial statistics methodology can be a useful tool in epidemiologic studies. The fact that no pockets of STIs or high risk behaviors were identified in this sample is informative in that behavioral interventions may need to target a broader geographic area rather than focusing on specific geographic locations in Kisumu (Westercamp et al., International Journal of Health Geographics 2010). 8. Determination of Cellular Markers and Innate Molecules in the Foreskin of Circumcised Men from
Kisumu Kenya. (Kristine Broliden and TahaHirbod, Ph.D, Sweden)
Co-investigators / Co-authors:
Walter Agingu
Robert C. Bailey, PhD
Stephen Moses, MD, MPH
KawangoAgot, PhD, MPH
JeckoniahNdinya-Achola, MBChB, MSc
Walter Jaoko, MBchB, PhD
PI contact Information: kristina.broliden@karolinska.se or taha.hirbod@ki.se

Purpose:
This study aims to evaluate the expression of HIV receptors and define the phenotype of viral target cells as well as characterize the innate immune response against HIV in human foreskin tissues collected from participants in the randomized controlled trial of male circumcision to reduce HIV incidence in Kisumu, Kenya. By studying the biology of HIV transmission at mucosal sites we will gain knowledge that can contribute to the development of microbicide products and vaccine candidates against HIV. Development of prophylactic vaccines and microbicides against HIV-infection has been hindered by our limited knowledge of the mechanisms whereby HIV gains entry at mucosal sites. Thus, it is not yet obvious which critical host cellular and molecular structures should be targeted. The large number of different cellular receptors involved in establishing HIV infection and the rapid kinetics of viral dissemination suggest that no single approach for prevention of primary infection may be effective. Several large clinicaltrials show that circumcision can reduce the rate of HIV infection in heterosexual men by 50%, studies performed to further elucidate the biology of HIV transmission in the male genital tract are of great interest. Mucosal innate immune responses in individuals who are at risk of acquiring HIV infection through sexual intercourse is also of importance to characterize in genital tissue including foreskin. Several innate immune molecules are in fact ligands to HIV cellular receptors (i.e. RANTES being the ligand for CCR5) Page 10 of 16
and can thus potentially block HIV transmission. Other innate immune molecules have a more general anti-viral activity such as interferon-alpha. Studies of these molecules in human foreskin will thus give a more complete picture of the previously described characterization of HIV receptors and target cells. Progress:
Thirty-three foreskin tissue samples, stratified by Herpes simplex virus type 2 (HSV-2) status, have now
been analyzed by confocal in situ imaging microscopy and by RT-qPCR. The distribution of T cells
(CD3+CD4+), Langerhans cells (LCs) (CD1a+Langerin/CD207+), macrophages (CD68+ or CD14+) and
subepithelial dendritic cells (CD123+ or CD11c+) was defined. CLR+ cells were detected in both the
epithelium and submucosa and distinct lymphoid aggregates containing HLA-DR+, CD3+ and CD4+
cells were identified in the submucosa. When comparing asymptomatic HSV-2 seropositive versus
negative men for the presence and size of lymphoid aggregates and for mRNA expression of HLA-DR,
CD3, CD4 and the CLRs, only Mannose receptor/CD206 differed between the groups(HSV-2
seropositive >negative, P=0.038). The detection of abundant and superficially present potential HIV-1
target cells and submucosal lymphoid aggregates in foreskin mucosa demonstrate an anatomical
explanation for the protective effect of male circumcision on HIV-1 transmission (Hirbod et al., Am J
Pathol, 2010).
Ongoing studies of the foreskin are further investigating the structure and function of the lymphoid
aggregates and further studies are designed to measure keratinized tissue at different sites on the inner and
outer foreskin.On September 22, 2011, the KNH/UON - Ethics and Research Committee approved the
shipment of 93 foreskin biopsies to Karolinska Institute in Sweden for analysis. The samples have been
received at the Karolinska Institute and analyses of specimens are on-going.
9. Qualitative Study of sexual Disinhibition related to male Circumcision in Kisumu, Kenya
(Thomas Reiss, Ph.D. Candidate)
Co-investigators / Co-authors:
Robert C. Bailey, PhD, MPH
Walter Jaoko, MBchB, PhD
PI Contact Information: treiss@uic.edu

Purpose:
This qualitative research study will examine men’s and women’s perceptions of sexual risk and sexual
risk behaviors for human immunodeficiency virus (HIV) transmission related to male circumcision (MC)
in Kisumu, Kenya, and provide recommendations to HIV prevention programs. The goals of this study
are: 1) to explore men’s and women’s perceptions of male circumcision with regard to its protective effect
against HIV, 2) to explore how contextual factors related to sex and MC influence sexual risk taking, and
3) to provide recommendations to HIV prevention programs about how to better address risk behaviors
related to MC among men and women.
Male circumcision (MC) has been shown to reduce the transmission of HIV from women to men via
vaginal sex by 60%. As MC becomes a more common practice it is important to understand men’s
attitudes, expectations, and behavioral intentions surrounding circumcision. There is a concern that men
who are circumcised, and their sexual partners, may falsely assume that circumcision offers complete
protection against contracting HIV and become sexually disinhibited, reducing condom use and
Page 11 of 16
increasing sexual practices that place them a increased risk for HIV. Disinhibiting sexual behavior could
nullify the protective effects of circumcision and undermine prevention efforts.
Progress:
Focus group and individual interview data collection was completed in November 2008 as theoretical
saturation has been reached. Four focus groups discussions (FGD) were carried out. The focus groups
were composed of the following: FGD #1) 8 men, FGD #2) 6 women, FGD #3) 5 men, FGD #4) 4
women and 5 men. The total number of FGD participants is 28. Seventy individual interviews were
conducted with residents of Kisumu District: 30 with circumcised men, 10 with uncircumcised men, and
30 with women.
Data analysis is currently underway. All 70 qualitative interviews and four focus group discussions
conducted among men and women have been translated, transcribed, and coded. Several analytical
memos based on major themes emerging from the data have been written and will serve as the basis for
future publications and presentations. Results illustrate that information about MC’s protection against
HIV has disseminated into the larger community and MC accompanied by counseling and HIV testing
can foster positive behavior change and maintain sexual behavior. (Riess, T.H. et al, PLoS ONE,2010)
10. Gonococcal antibiotic resistance among STD Clinic patients in Kisumu, Kenya
(Supriya Mehta, PhD, MHS)
Co-investigators / Co-authors:
Robert C. Bailey, PhD, MPH
Ian Maclean, PhD
Stephen Moses, MD, MPH
Elijah Odoyo-June, MBChB, MSC
KawangoAgot, PhD, MPH
JeckoniahNdinya-Achola, MBChB, MSc
Walter Jaoko, MBchB, PhD
PI Contact information:smehta@uic.edu
Purpose:
The objective of the current proposed study is to determine the prevalence and types of antimicrobial
resistance to gonorrhea, whether this is associated with this previous clinic visits, clinical presentation and
course of infection, and treatment outcome. This has immediate implications for clinical care outcomes
and transmission.
While there is insufficient data collected to quantify to what extent multiple visits for gonorrhea at the
UNIM STD Clinic are due to new infections, re-infection from an untreated partner, or antibiotic drug
resistance, the trial data suggest a potential for gonorrhea resistance. This information is vital for
appropriate and curative treatment and for interrupting transmission. As the results of diagnostic testing
are not immediately available, men are treated syndromically for gonorrhea and chlamydia; meaning,
treatment is administered on the basis of symptoms and signs of infection. Generally, the antimicrobial
agents given for empiric treatment of urethral discharge syndrome that is believed to be related to
Neisseria gonorrhoeae (NG) and Chlamydia trachomatis (CT) are ciprofloxacin 500mg (NG) and
Page 12 of 16
doxycycline 100mg BID (CT). However, if there is quinolone resistance, then clinicians should not be using ciprofloxacin to treat presumptively for gonorrhea, as this would not be effective treatment for some men. Rather, a combination of doxycline and a non-quinolone antibiotic, such as cefixime or ceftriaxone, would be appropriate. Doxycycline and cefixime is standard in most places where there is quinolone resistance to gonorrhea. However, the choice of alternative drug therapies has implications for cost, and should be undertaken with certainty. The proposed study aims to systematically determine the prevalence and types of antibiotic resistance for urogenital gonorrhea infection among STD Clinic attendees in Kisumu, Kenya. Additionally, we will examine whether antibiotic resistant Neisseria gonorrhoeae is associated with features of past and current clinical presentation(s). This has immediate implications for clinical care outcomes and prevention of transmission. Results of this analysis will provide the basis for recommendations for alternative antibiotic therapy that will have greater curative efficacy, if needed. A high prevalence of quinolone resistance will indicate a new regimen is needed. Progress:
This study was completed and closed on September 30, 2011. Records of clinic visits were examined from March 2009 through December 2010. We abstracted data from patient charts, and men and women with symptoms suggestive of urogenital gonococcal infection had specimens submitted for detection of N. gonorrhea (NG) by culture. Culture positive isolates were assessed for antimicrobial resistance to several drug classes by disk diffusion. Statistical analyses examined whether patient factors were associated with quinolone resistance or repeat clinic visit, as a potential marker of treatment failure. Among 412 male client visits, 103 (25%) urethral swabs were obtained from men who had a current complait of urethral discharge or who had urethral discharge on examination. Specimens from these 103 men were tested for N. gonorrhoeae (NG) by culture. Among the 103 men tested for NG, 90 (87%) were positive. Quinolone resistance was assessed by disc diffusion to norfloxacin. Overall, the prevalence of norfloxacin resistance was 15.6% [14/90; 95% confidence interval 8.8-24.7%]. These results confirm the existence of quinolone resistance in western Kenya, and suggest that prevalence may be greater than an acceptable level (5%, as per CDC) for continued treatment with ciprofloxacin or norfloxacin. Of 377 female patients seen at the clinic March 2010 through September 2010, 53 (14%) had specimens obtained for culture for N. gonorrhoeae, 11 (20.8%) of which tested positive. Four (36.4%; 95% CI: 10.9-69.2%) showed resistance to norfloxacin, as demonstrated by zone size <18mm on disc diffusion. The differential diagnosis for vaginal discharge includes vaginal infections as well as cervical infections, which explains why only 21% were culture positive for NG. Because of the large sample size that would be needed to observe a sufficient number of women with gonococcal infection to generate precise estimates of resistance, we discontinued resistance testing to meet our research aim on December 31, 2010. Nevertheless, quinolone resistance in women demonstrated a high rate, similar to men. Our data is included in a multisite analysis of antimicrobial resistance in NG. The manuscript is under review at the journal, Sexually Transmitted Diseases. [Lagace et al.; senior author SD Mehta]. Among 1,473 clients (1,296 single-visit individuals vs. 177 individuals with repeat visits), the median age was 24 years, 67% were male, and 8.6% self-reported being HIV positive. In adjusted analyses, men with repeat visits were more likely to report > 2 recent sex partners (adjusted odds ratio (AOR) =1.60) and being HIV-positive (AOR=2.35). They were less likely to have been referred from other health facilities (AOR=0.14) and more likely to have urethral discharge at initial visit (AOR=2.46). Among women, Page 13 of 16
repeat visits were associated with vaginal discharge (AOR=2.22), but attending the clinic with a partner
was protective (AOR=0.38).

In conclusion, the association between sexual risk, HIV positivity and repeat visits among male clients
highlights the need to focus intervention efforts on this group. For women, attending with a partner may
reflect a decreased risk of re-infection if both partners are treated and counseled together.
11. Study of Gonococcal resistance and Mycoplasma genitalium, Trichomonasvaginalis in men from
Kisumu, Kenya.(Supriya Mehta, PhD, MNS)

Co-investigators / Co-authors:
Robert C. Bailey, PhD, MPH
Walter Jaoko, MBchB, PhD
PI Contact information:smehta@uic.edu
Purpose:
As reported in the Journal of Infectious Diseases, in our cohort of sexually active men aged 18-24 years-
old, the incidence of urogenital Chlamydia trachomatis (4.55 per 100 person-years) and Neisseria
gonorrhoeae (3.48 per 100 person-years) and/or Trichomonasvaginalis (1.32 per 100 person-years) at
follow-up was 8.34 per 100 person-years [95% CI: 7.50 – 9.28]. Of the 148 men with gonorrhea detected
at follow-up, 18 (12.2%) accounted for 38 infections. Conversely, very few men infected with chlamydia
experienced re-infection (3.6%). Anecdotally, the clinicians report similar observations of multiple visits
for gonorrhea infection at the STI clinic in Kisumu, but not for chlamydia infection. We hypothesize that
the differential rate between gonorrhea and chlamydia for repeat infection is due in part to antimicrobial
resistant organisms.
PCR detection of Mycoplasma genitalium and Trichomonasvaginalis as causes of urethral discharge in our population may have implications for syndromic management. We will examine the proportion of symptomatic disease that is further explained by these two PCR-detected infections, and whether or not this will impact recommendations for syndromic management. This has immediate implications for clinical care outcomes and interruption of transmission. The specific aims of this study are to: 1) Analyze results of antimicrobial resistance testing for Neisseria gonorrhoeae and 2) Conduct polymerase chain reaction assays to detect Mycoplasma genitaliumand Trichomonasvaginalisin urine specimens that have been collected from the cohort participants. Analysis of specimens for the prevalence and types of antimicrobial resistance will enable us to determine whether antibiotic resistance potentially affects treatment and outcome of urogenital gonorrhea infection in our population. Results of this analysis can provide the basis for recommendations for alternative antibiotic therapy that will have greater curative efficacy, if needed. The need for this information extends beyond current, local concerns; currently there is no regular gonorrhea antimicrobial susceptibility testing in Kenya. Progresssub-study #1:
Page 14 of 16
Specimen processing for the detection of antimicrobial resistance to Neisseria gonorrhoeae(NG) isolated from men enrolled in a randomized trial of male circumcision to prevent HIV was completed October 2010. Urethral specimens from men with discharge were cultured for NG. Minimum inhibitory concentrations (MICs) were determined by agar dilution. Clinical Laboratory Standards Institute criteria defined resistance: MIC≥2.0 µg/ml for penicillin, tetracycline, and azithromycin; ciprofloxacin MIC≥1.0 µg/ml; spectinomycin MIC≥128.0 µg/ml. Susceptibility to ceftriaxone and cefixime was MIC<0.25 µg/ml. Additionally, PCR amplification identified mutations in parCand gyrAgenes in selected isolates. From 2002–2009, 168 NG isolates were obtained from 142 men. Plasmid mediated penicillin resistance was found in 65%, plasmid mediated tetracycline resistance in 97%, and 11% were ciprofloxacin resistant (QRNG). QRNG appeared November 2007, increasing from 9.5% in 2007 to 50% in 2009. Resistance was not detected for spectinomycin, cefixime, ceftriaxone, and azithromycin, but MICs of cefixime (p=0.018), ceftriaxone (p<0.001), and azithromycin (p=0.097) increased over time. In a random sample of 51 men gentamicin MIC was: 4 µg/ml (n=1), 8 µg/ml (n=49), 16 µg/ml (n=1). Findings from this study conclude that the level of quinolone resistance we detected (26% in 2007-2009) in Kisumu, Kenya, is greater than the level (5% (37)) at which an alternative antibiotic regimen is recommended, and indicates that clinicians should not be using ciprofloxacin or norfloxacin to treat gonorrhea. A non-quinolone antibiotic such as cefixime or ceftriaxone would be appropriate to treat diagnosed or suspected gonorrhea in this area, as indicated by the low MIC90 values we observed. The coexistence of multiple other types of resistance and increased MICs for cephalosporins contribute to growing concern for multidrug resistant NG. Ongoing surveillance for antimicrobial resistance in Kenya is lacking and is an essential public health capacity to ensure timely response to emerging cephalosporin and multi-drug resistant NG, and selection of pharmacologic regimens with effective cure. Prevention of antimicrobial resistance emergence will require a population-level approach, rather than a disease specific approach. Progress sub-study #2:
Specimens processing for detection of urethral Mycoplasma genitalium(MG) infection among men
enrolled in the randomized trial of medical male circumcision to prevent HIV acquisition in Kisumu,
Kenya was completed February 2, 2011. M. genitalium and Trichomonasvaginalis (TV) were detected in
first void urine (FVU) by APTIMA transcription mediated amplification assay. FVU and urethral swabs
were assessed for N. gonorrhoeae(NG) and C. trachomatis (CT) by polymerase chain reaction assay.
HSV-2 antibodies were detected by IgG ELISA.
Specimens were collected between July and September 2010, and 52 [9.9%; 95% CI: 7.3-12.4%] MG infections were detected among 526 men. NG and TV were not associated with MG. CT co-infection was 5.8% in MG-infected men, and 0.8% among MG-uninfected men (p=0.02). Urethral discharge (1.5%) and dysuria (1.7%) were not associated with MG status. The prevalence of MG was 13.4% in uncircumcised men vs. 8.2% in circumcised men (p=0.06). In multivariable logistic regression, being circumcised nearly halved the odds of MG [adjusted OR=0.54; 95% CI: 0.29-0.99], adjusted for other variables significant at the p<0.05 level: HSV-2 infection [aOR=2.05; 95% CI: 1.05-4.00], CT infection [aOR=2.69; 95% CI: 1.44-5.02], and washing the penis <=1 hour after sex [aOR=0.47; 95% CI: 0.24-0.95]. Study findings conclude that prospective study is required to verify the potential protective association of circumcision against urethral MG infection. The role of MG in STI morbidity for men and women in this Page 15 of 16
region should be further evaluated, and syndromic management algorithms and antibiotic regimens reviewed. 12. Microbial Diversity of Genital Ulcer Disease in Men Enrolled in a Randomized Trial of Male
Circumcision in Kisumu, Kenya. (Supriya Mehta, PhD, MNS).

Co-investigators / Co-authors:
Robert C. Bailey, PhD, MPH
Ian Maclean, PhD
Stephen Moses, MD, MPH
Patrick Gillevet
Greg T. Spear, PhD

PI Contact information:smehta@uic.edu
Purpose:
Medical male circumcision (MMC) reduces the risk of HIV acquisition by ~60%, in part by reducing
genital ulcer disease (GUD) and HSV-2. In the Ugandan and Kenyan trials, PCR did not find a sexually
transmissible agent in 50-60% of GUD specimens, and MMC did not protect against HSV-2 in the
Kenyan trial. We sought to better define the etiology of GUD in the Kenyan trial and examine how MMC
affects GUD etiology.
From March 2002 through February 2007, specimens were obtained by clinicians from men with genital
ulcers detected on clinical examination. Specimens were obtained by swabbing the genital ulcers with a
Dacron or polyester swab. The stored specimens were examined using multiplex polymerase chain
reaction (PCR) assay to detect the causal etiology. Three pathogens were assayed: herpes simplex virus
(HSV), Treponemapallidum(TP; syphilis), and Haemophilusducreyi(HD; chancroid). Results of mPCR
found that 29% of specimens were HSV positive, 4% TP positive, and 0% with HD. Overall, 62% of
ulcers did not have a specific etiology identified. Because of the contribution of genital ulcer disease to
the HIV epidemic, determining the etiology of these ulcers is crucial for effective prevention and
treatment.
Therefore, we seek to re-analyze these stored isolates using length-heterogeneity (LH-) PCR followed by
pyrosequencing to identify a broad range of potential bacterial etiology. Pyrosequencing is a method of
DNA sequencing that has high coverage and >99% accuracy at the phylum level compared to traditional
methods of DNA sequencing. It has been used to determine microbial diversity in environments ranging
from the soil to human gut, throat, and hands to recovery of human bacterial colonies from inanimate
objects. We will employ a multiplexed pyrosequencing method to study the microbial diversity of these
clinically identified genital ulcers and will compare microbial profile by a number of clinical and
behavioral factors to suggest mechanism of causation.
Progress:
Processing of specimens is complete and the final analyses and manuscript are pending. Overall, 83
distinct genera were detected. Prevotella spp. was most abundant, accounting for 18% of microbiota on
average, and present in 75% of specimens. Bacterial diversity was greater in GUD of unknown etiology
than STI associated GUD (median number of genera 13 [range 7-20] vs. 11 [range 3-20], p=0.06).
Page 16 of 16
Fusobacterium spp., Sneathia spp. and Anaerococcus spp. were more abundant in GUD of unknown
etiology (7.7%, 7.6%, and 7.3%, respectively) than in GUD of STI etiology (4.6%, 5.0%, and 5.4%).
Fusobacteriales (Fusobacterium spp. or Sneathia spp.) [OR=4.7; 95% CI: 1.3-19.9] and
Anaerococcusspp. [OR=4.6; 95% CI: 1.2-22.5] were more likely to be recovered in GUD of unknown
etiology than STI associated GUD. Fusobacteriales were more often recovered from uncircumcised men
than circumcised men (62% vs. 22%, p=0.04), and Anaerococcus spp. was present in 22% of circumcised
vs. 70% of uncircumcised men (p=0.010). Reported penile coital injuries were more common among men
with Anaerococcusspp. (85% vs. 57%, p=0.01), and condom use was less common (50% vs. 71%,
p=0.11). There was no difference in these bacteria by ulcer location.

Findings concluded that fusobacteriales and Anaerococcus spp. may colonize genital “ulcers” that
develop from a mechanism related to circumcision status. Many such "ulcers" may be epithelial
disruptions that are traumatic in origin. These bacteria have cytotoxic properties that may ulcerate or
exacerbate pre-existing minor epithelial disruptions. MMC may reduce GUD through a reduction in these
anaerobic bacteria.


Page 17 of 16

Source: http://bluewebsafrica.co.ke/nrhs/UNIM%20Study%20Descriptions.pdf