Characteristics and outcomes of individuals enrolled for HIV care in a rural clinic in Coastal Kenya - Thesis

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Characteristics and outcomes of individuals enrolled for HIV care in a rural clinic

in Coastal Kenya

Hassan, A.S.

Publication date

2014

Document Version

Final published version

Link to publication

Citation for published version (APA):

Hassan, A. S. (2014). Characteristics and outcomes of individuals enrolled for HIV care in a

rural clinic in Coastal Kenya.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)

and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open

content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please

let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material

inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter

to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You

will be contacted as soon as possible.

Characteristics

and Outcomes of

Individuals Enrolled

for HIV Care in

a Rural Clinic

in Coastal Kenya

Amin S. HASSAn

Three decades after the first reported case, HIV/AIDS

re-mains a major global public health challenge. Despite the

scale up of HIV/AIDS care and treatment services in

sub-Saharan Africa, characteristics of individuals enrolling for

care and the continuum of care remains less well described.

The changing dynamics of the HIV-1 epidemic warrants an

up-to-date description of the clinical and molecular

char-acteristics of individuals enrolling for care, with an aim of

understanding the epidemic and suggesting potential

in-terventions to improve outcomes. This thesis presents the

results of studies nested in a longitudinal HIV surveillance

system aimed at describing the characteristics and

out-comes of individuals enrolling for HIV/AIDS care and

treat-ment in a rural HIV clinic in Coastal Kenya.

acteristics and Outcomes of Individuals Enr

olled for HIV Car

e in a Rur

al Clinic in Coastal K

enya

Individuals Enrolled for HIV Care

in a Rural Clinic in Coastal Kenya

The studies in this thesis were as a result of a surveillance system set up in the HIV clinic at Kilifi County (formerly District) Hospital in Kenya. The surveillance system was set up by the KEMRI/Wellcome Trust Research Programme and was financially supported by a Wellcome Trust Masters Masters fellowship grant (WT089351MA) and a Wellcome Trust Intermediate fellowship grant (WT083579MA).

The printing of this thesis was financially supported by the KEMRI/Wellcome Trust Research Programme in Kilifi, Kenya.

Individuals Enrolled for HIV Care

in a Rural Clinic in Coastal Kenya

ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam op gezag van de Rector Magnificus

prof. dr. D.C. van den Boom

ten overstaan van een door het college voor promoties ingestelde commissie,

in het openbaar te verdedigen in de Agnietenkapel op dinsdag 9 september 2014, te 10:00 uur

door

Amin Shaban Hassan geboren te Mombasa, Kenia

PROMOTIECOMMISSIE

Promotor: Prof. dr. T.F. Rinke de Wit Co-promotores: Dr. J.A. Berkley

Dr. E.J. Sanders Overige leden: Prof. dr. J.M.A. Lange

Prof. dr. P.R. Klatser Prof. dr. M.D. de Jong Prof. dr. C.A.B. Boucher Dr. A.D. Smith

Abbreviations 9

Chapter 1 Background 11

Chapter 2 Study Setting 23

Part 1 Characteristics of Individuals Enrolling for HIV Care 37

Chapter 3 HIV-1 in a rural coastal town in Kenya shows multiple introductions with many subtypes and much recombination.

39

Chapter 4 Low prevalence of transmitted HIV type 1 drug resistance among antiretroviral-naïve adults in a rural HIV clinic in Kenya.

51

Part 2 Outcomes of Individuals Enrolled for HIV care (but prior to ART Initiation)

67 Chapter 5 Early drop out of recently diagnosed HIV-infected adults

from routine pre-ART care in a rural district hospital in Kenya: A cohort study.

69

Chapter 6 Dynamics and constraints of early infant diagnosis of HIV infection in rural Kenya.

93 Part 3 Outcomes of Individuals Enrolled for HIV Care

(and after ART Initiation)

109 Chapter 7 Incidence and predictors of attrition from antiretroviral

care among adults in a rural HIV clinic in Coastal Kenya: A retrospective cohort study.

111

Chapter 8 HIV-1 virological failure and acquired drug resistance among first-line antiretroviral experienced adults at a rural HIV clinic in coastal Kenya.

131

Chapter 9 General Discussion 155

Addendum Summary

Samenvatting (Dutch Summary) Authors and Affiliations List of Publications Acknowledgements About the Author

171 175 179 181 183 185

AIDS Acquired Immunodeficiency Syndrome ART Antiretroviral Therapy

ARV Antiretrovirals

BMI Body Mass Index

CCRC Comprehensive Care and Research Clinic CRF Circulating Recombinant Forms

DBS Dried Blood Spots DNA Deoxyribonucleic Acid

DTC Diagnostic Testing and Counseling EID Early Infant Diagnosis

HAART Highly Active Antiretroviral Therapy HIV Human Immunodeficiency Virus IAVI International AIDS Vaccine Initiative KDH Kilifi District Hospital

KEMRI Kenya Medical research Institute LTFU Loss to Follow up

MCH Mother and Child Health

NASCOP National AIDS and STI Control Programme NNRTI Non-Nucleoside Reverse Transcriptase Inhibitors NRTI Nucleoside Reverse Transcriptase Inhibitors PCR Polymerase Chain Reaction

PEPFAR President Emergency Plan for AIDS Relief PI Protease Inhibitors

PITC Provider Initiated testing and Counseling PMTCT Prevention of Mother to Child Transmission RNA Ribonucleic Acid

SIV Simian Immunodeficiency Virus sSA sub-Saharan Africa

STI Sexually Transmitted Infections TDR Transmitted Drug Resistance WHO World Health Organization

1

INTRODUCTION

Three decades after the first reported cases in the US, HIV/AIDS remains a major public health burden worldwide [1]. The past decade has experienced an extensive roll out of HIV/AIDS services, particularly in sub Saharan Africa (sSA) where an estimated 22.4 million people – around two-thirds of the people living with HIV globally – are infected [2]. The scale up of HIV/AIDS services has resulted in substantial improvement in the prognosis of HIV-infected people seeking care [3-6]. In 2011 for example, an estimated 1.2 million people died from HIV/AIDS related causes in sSA, compared to 1.8 million deaths in 2005 [7]. Similarly, a decline in new HIV infections has also been reported, from 2.4 million in 2001 to 1.8 million in 2011 [7].

Consequently, the characteristics and outcomes of people enrolling for HIV care have also changed over time. Most studies describing the characteristics and outcomes of HIV infected individuals in sSA are from urban or tertiary settings. This thesis describes the characteristics and outcomes of individuals enrolled and followed up for HIV care in a rural HIV clinic in Coastal Kenya.

The Human Immunodeficiency Virus

Types and subtypes

HIV is an enveloped single stranded reverse transcribing RNA of the Retroviridae family. Based on genetic similarities, HIV strains may be classified into two types: HIV-1 (predominant glob-ally) and HIV-2 (mostly concentrated in West Africa)[8]. The HIV-1 strains are further classified into four groups: the “major” group M, and the other groups N (“non-M”), O (“outlier”) and P (“pending identification of further human cases”). Globally, more than 90% of HIV-1 infections belong to HIV-1 group M.

Within group M, there are at least nine genetically distinct subtypes; A, B, C, D, F, G, H, J and K [9]. Subtypes A and C are most widespread in sSA, while subtype B is most common in Europe, America, Japan and Australia. Subtype C is predominant in Southern and East Africa, India and Nepal. Subtype D is generally limited to East and Central Africa. Subtype G has been observed in West Africa, East Africa and Central Europe. Subtype H has only been found in Central Africa; J only in Central America; and K only in the Democratic Republic of Congo and Cameroon.

Occasionally, two viruses of different subtypes can meet inside cells of an infected person and mix their genetic material to create a new hybrid [10]. Many of these new strains do

14 Chapter 1

not survive for long, but those that infect more than one person are known as circulating recombinant forms (CRFs). For example, the CRF A/D is a recombinant of subtypes A and D. HIV structure

Outside of a human cell, HIV exists as spherical particles called virions, surrounded with viral membranes. Projecting from the membranes are the env proteins, consisting of caps (glycoprotein (gp) 120) and stems (gp41). Below the membrane is the matrix, made from the protein (p) 17. Within the matrix is the viral core, made from p24 (in HIV-1) or p26 (in HIV-2). Inside the core are three enzymes required for HIV replication (reverse transcriptase, integrase and protease) and two single strands of HIV RNA. At either end of each strand of RNA is a sequence called the long terminal repeat, which helps to control HIV replication.

HIV has nine distinct genes playing different roles in the replication cycle [11]. In brief, three of the HIV genes called gag, pol and env, contain information needed to make structural proteins for new virus particles. The other six genes known as tat, rev, nef, vif, vpr and vpu are regulatory proteins responsible for viral replication.

Replication of HIV

HIV targets white blood cells such as the CD-4 T-lymphocytes, macrophages and dendritic cells. The virus binds to the CD4 receptor, and either the CCR5 or the chemokine co-receptors protein CxCR4 on the surface of the CD4 T-lymphocyte via the gp120 protein. The virus membrane fuses via a structural change in the gp41 protein and releases its RNA genetic materials into the host cell.

Once inside the host cell, the HIV reverse transcriptase enzyme converts the single stranded viral RNA into double stranded HIV DNA through a process of reverse transcription. The newly formed DNA material is transported to the host’s cell nucleus, where it is spliced into the host’s cell DNA by the HIV integrase through a process of integration. Once integrated, the HIV DNA is known as provirus. If not activated, the proviral DNA may lie dormant for a long time.

When the host cell becomes activated, it converts HIV DNA into messenger RNA using the enzyme RNA polymerase through a process of transcription. The messenger RNA is trans-ported outside the nucleus and used as a blueprint for producing new HIV proteins and enzymes through a process of translation. Among the strands of messenger RNA produced by the cell are complete copies of HIV genetic material. These gather together with newly made HIV proteins and enzymes to form new viral particles through an assembly process, which are then released from the cell. The protease enzyme chops up long strands of protein into smaller pieces, which are used to construct mature viral cores.

1

membrane through a process of budding. The matured HIV virions are ready to infect other cells and begin the replication process all over again.

Antiretroviral therapy

Because of the complex nature of the virus, the search for an effective vaccine and/or a cure for HIV has so far been elusive. However, treatment for HIV has been available since 1987, initially in the form of a single drug, zidovudine. Two and a half decades later, more than thirty different types of antiretroviral drugs in 5 distinct classes are now available. The main aim of these drugs is to suppress viral replication, improve immunity and thus prevent opportunistic infections. Combinations of at least three antiretroviral drugs targeting differ-ent pathways in the virus replication cycle, known as highly active antiretroviral treatmdiffer-ent (HAART), are currently recommended in order to achieve maximum viral suppression and long term efficacy [11].

HIV drug targets and mechanism of action

Currently, five different classes of antiretroviral drugs exist: Nucleoside/Nucleotide Reverse Transcriptase Inhibitors (NRTIs), Non-Nucleoside Reverse Transcriptase Inhibitors (NNRTIs), Protease inhibitors (PIs), Integrase Inhibitors and Viral Entry inhibitors.

NRTIs are analogues of DNA building blocks. When building a new viral DNA chain, the reverse transcriptase enzyme binds to NRTIs instead of binding to the naturally occurring DNA building blocks. Because the structure of the NRTIs does not allow attachment of the next DNA building block, DNA chain growth is terminated. Hence, NRTIs interrupt the HIV replication cycle via competitive inhibition of HIV reverse transcriptase and termination of the DNA chain [12]. Examples of NRTIs include Zidovudine, Stavudine, Lamivudine, Emtricitabine, Didanosine, Tenofovir and Abacavir.

NNRTIs bind tightly to the enzyme reverse transcriptase, Instead of competing with naturally occurring DNA building blocks, as do the NRTIs,. This noncompetitive binding induces a conformational change in the enzyme that alters the active site and limits its activity, thereby preventing viral RNA from being converted to DNA [13] . Examples of NNRTIs include Nevi-rapine, Efavirenz, Delavirdine and Etravirine.

Protease Inhibitors (PIs) slow down HIV replication after the virus is integrated into the host cell’s DNA in infected cells. During the maturation process of new virions, the HIV protease enzyme systematically cleaves individual proteins from the gag and gag-pol polypeptide precursors into functional subunits for viral capsid formation during or shortly after viral bud-ding from an infected cell. If the polyproteins are not cut, the new virus fails to mature and

16 Chapter 1

is incapable of infecting a new cell. PIs function as competitive inhibitors that directly bind to HIV protease enzyme and prevent subsequent cleavage of polypeptides, thereby rendering the new virion immature and non-infectious [14]. Examples of PIs include Lopinavir, Ritonavir, Darunavir, Indinavir, Atazanavir and Nelfinavir.

HIV integrase is responsible for the transport and attachment of proviral DNA to host-cell chromosomes, allowing transcription of viral proteins and subsequent assembly of virus particles. Proviral integration involves two catalytic reactions: 3’-processing in the host-cell cytoplasm to prepare proviral strands for attachment, and strand transfer whereby proviral DNA is covalently linked to cellular DNA. Integrase inhibitors competitively inhibit the strand transfer reaction by binding metallic ions in the active site [15]. Examples of Integrase Inhibi-tors include Raltegravir and Elvitegravir.

Viral entry inhibitors can be further sub-divided into three mechanistically distinct classes: Attachment inhibitors, Co-receptor inhibitors and Fusion Inhibitors. Attachment inhibitors e.g. BMS-378806/663068, prevent the attachment of HIV to the outer membrane of the CD4 host cell by binding to gp120 [16]. Co-receptor inhibitors, e.g. Maraviroc, prevent the interaction of HIV with co-receptors CCR5 and CXCR4 on the host cell surface [16]. Fusion inhibitors, e.g. Enfurvitide, prevent the fusion of the CD4 T-lymphocyte binding to the HIV surface glycoprotein 41 (gp41) [16].

Antiretroviral drug resistance

HIV is a highly variable virus, characterized by rapid error prone replication and viral re-combination. This extensive genetic variability is primarily attributed to the high error rate of the reverse transcriptase, which results to approximately 10 genomic base changes per replication cycle [17]. The majority of mutations from copying errors confer negative or no survival advantage. However, some may give rise to resistance to antiretroviral drugs. When treatment with HAART is successful, the plasma viral load drops to undetectable levels. However, incomplete suppression of viral replication while on HAART fosters emergence of resistance. This occurs because the wild type virus is inhibited from growing while the mutant strains become dominant if they are not fully inhibited by the drugs because of selective pressure. In the presence of drug pressure, and given the need for lifelong treatment in HIV infected individuals, emergence of some degree of HIV drug resistance is inevitable, even when optimal adherence to therapy is achieved. Because drug-resistant mutants can replicate despite antiretroviral therapy, they may become the dominant circulating strains. This can eventually lead to an increase in viral load, a decline in the number of CD4 cells, and subsequent treatment failure. Antiretroviral treatment failure is to a large extent because of drug resistance.

1

affin-ity of the enzyme reverse transcriptase to NRTIs which impairs incorporation into the proviral DNA, and increased removal of the NRTI from the elongating proviral DNA chain [18]. A good example is the common M184V mutation, which confers high-level resistance to lamivudine and emtricitabine in a single step. Other mutations that selectively impair incorporation into the proviral DNA chain include M184V, Q151M, and K65R. In addition, mutations associated with zidovudine resistance are termed as thymidine analogue mutations (TAMs). These muta-tions work by removing the chain-terminating residue and reinstate an extendable primer for the proviral DNA to resume DNA synthesis. Examples of TAMs include M41L, D67N, K70R, L210W, T215Y, T215F, K219Q, K219E.

Resistance to NNRTIs is normally associated with mutations that are proximal to the drug-binding site on reverse transcriptase, which is in contrast to mutations conferring resistance to NRTI. Mutations within the reverse transcriptase gene domain alter the ability of NNRTIs to bind the enzyme [19]. NNRTIs have a low genetic barrier to resistance; a single mutation in the binding site can decrease the ability of the drug to bind, significantly diminishing activ-ity. Resistance to NNRTI has been associated with mutations at multiple codons. However, the presence of either a K103N or Y181C mutation is sufficient to cause clinical failure of delavirdine, efavirenz, and nevirapine. Other mutations conferring resistance to NNRTIs usually occur at codon positions L100, K101, V106, V179, Y188, G190, P225 and M230. Resistance to HIV protease inhibitors results from mutations both inside and outside the active protease domain [20]. Resistance typically occurs through the development of one or more major mutations, which produce conformational changes in the protease binding site. Examples of primary mutations conferring resistance to PIs are G48V, L90M, M46I, V82A/L/F, I184V, D30N, L90M, I50L, I84V and N88S. Multiple mutations are necessary to cause high-level resistance to Ritonavir-boosted protease inhibitors, which exhibit a higher genetic threshold for resistance than un-boosted protease inhibitors.

Mutations in the integrase gene are associated with resistance to integrase inhibitors. The most common integrase mutational sequences are Q148H, N155H and Y143RC. Additional resistance to Integrase inhibitors can be conferred by the L74M/R, E92Q, T97A, E138A/K, G140S/A, V151I, G163R, H183P, Y226D/F/H, S230R and D232N mutations.

The HIV epidemic

Origin, spread and current status of the epidemic in Africa

It is postulated that the HIV originated in Africa. A study of chimpanzees from Cameroon identified a strain of the Simian Immunodeficiency Virus (SIV) that is believed to be the viral

18 Chapter 1

ancestor of the HIV-1 in humans [21]. Evolution of the HIV-1 has since been simulated using computer models, which suggest that the transfer of SIV to humans occurred as early as the 1930’s [22]. On the other hand, it is believed that HIV-2 was introduced into the human population through the SIV-sooty mangabey lineage around 1940’s in Guinea-Bissau [23]. The first HIV/AIDS epidemic is believed to have occurred in Kinshasa, Congo in the 1970’s where a sudden increase in patients with opportunistic infections was observed [24, 25]. It is thought that HIV was brought to Kinshasa by an infected individual travelling from the neighboring Cameroon [21]. From there, it is speculated that the HIV was carried into Eastern Africa, other parts of Africa and subsequently the rest of the world where it reached epidemic levels in the early 1980’s. By the end of 2009, the global estimate of people living with HIV was 33.3 million, with approximately two-thirds of these residing in sSA [26]. HIV diagnosis and linkage to care

In sSA, HIV infected individuals are identified through various entry points within communities and in health facilities. These entry points include voluntary counselling and testing (VCT) sites established in different strategic locations within communities; in health facilities where diag-nostic testing and counselling (DTC) is done from the inpatients wards and outpatient clinics, including STI clinics, family planning clinics and antenatal clinics. More recently, testing in a healthcare setting has now transitioned into provider initiated counselling and testing (PITC). Testing is done using rapid antibody testing kits. A screening test is first done, followed by a confirmatory test for those screening positive. Discrepant results undergo a third (tie-breaker) test. Individuals diagnosed with HIV infection are subsequently referred to HIV clinic for registration and follow up care.

The WHO recommends a public health approach for HIV disease monitoring and antiretroviral treatment (ART) in developing settings [27]. Disease progression and ART eligibility is currently assessed clinically (WHO clinical staging III/IV) and immunologically (CD4 T-cell lymphocyte count <350 cells/μl in adults) [28]. Virological monitoring and drug resistance testing are not routinely available for disease monitoring and ART initiation in most healthcare settings in sSA. Individuals not eligible for ART initiation continue to be followed up closely so that subsequent eligibility can be identified for timely initiation of treatment. All HIV infected individuals are prescribed daily cotrimoxazole prophylaxis to prevent opportunistic infections. Those eligible are started on lifelong ART. In developing settings, first line ART is mostly comprised of 2 NRTIs and one NNRTI. Individuals failing first line ART are switched to second line regimen, comprised of 2 NRTIs and a boosted PI.

1

major challenges which threaten to reverse the gains made in the fight against the epidemic. These include attrition from care, treatment failure and emergence of HIV drug resistance. a) Attrition from HIV care

The main causes of attrition in sSA have been identified as lost to follow up (LTFU) and death. Most studies assessing attrition in HIV programmes have been amongst patients initiated on ART. Only a handful of studies on pre-ART attrition exist from sSA [29-31]. Of these, only one study from South Africa has assessed for correlates of retention in patients who are not eligible for ART [30]. A better understanding of pre-ART attrition is critical in designing interventions aimed at improving timely initiation of ART.

Among patients in ART programmes in sSA, LTFU and death are estimated to account for up to 56% and 40% respectively [32]. A large proportion of HIV-infected patients initiating ART, up to 60% in some settings, drop out immediately after starting ART [33-35]. Most observa-tional studies assessing attrition followed up individuals over durations ranging 6-24 months. A review of ART programs in sSA found rates of LTFU ranging from 20% at 6 months to nearly 40% at 2 years after ART initiation [36]. Observational studies also show that mMost deaths among patients on ART occur in the early months after treatment initiation and that mortality declines substantially thereafter [37-39].

The main independent risk factors for attrition, as determined by these studies, were lower baseline BMI, lower CD4 count, lower haemoglobin, WHO stage III/IV, younger patients and being male [34, 40-42]. Severe immune suppression at the time of ART initiation as indicated by low CD4 and high viral load counts at start of ART may therefore largely explain the high early mortality observed [42-44].

Attrition amongst children born to HIV-infected mothers is equally important from a public health perspective. Many HIV infected infants die from HIV related causes without their HIV status being known, or receiving HIV care [45]. Without access to cotrimoxazole prophylaxis, ART and supportive care, about a third of infants die by one year of age and a half will be dead by two years [46].

Loss to follow up has been described as one of the major early warning indicators for HIV treatment failure and drug resistance [47].

b) HIV treatment failure

The WHO defines treatment failure clinically, immunologically and virologically [48]. Clini-cally, treatment failure is defined as patients with new or recurrent WHO stage 4 conditions,

20 Chapter 1

despite being on treatment for more than six months. Immunological treatment failure is defined as patients with CD4 counts that have either fallen to baseline values (or below), have fallen from on-treatment peak values or those that have persistently low values of less than 100 cells/μl. Virological treatment failure is defined as two consecutive viral load values of more than 5,000 copies/ml.

In resource-rich settings, viral load testing is used to confirm suspected treatment failure based on immunological and/or clinical criteria. Virological treatment failure and the development of drug resistance are not routinely assessed in patients on ART in resource poor settings. The Ugandan DART trial suggested that viral load and resistance testing are not essential for monitoring patients on ART [49]. However, other data have shown that use of immunologic criteria frequently fails to appropriately identify virologically defined antiretroviral treatment failure [50]. Without virological testing, therefore, HIV infected patients on antiretroviral treatment are at risk of developing undetected virological failure [51]. Therefore, as more people access ART, there is need for routine laboratory monitoring, especially for virological treatment failure.

A number of studies have investigated virological treatment failure in Africa using different viral load thresholds, with most ranging from viral loads of 400 to 10,000 copies/ml [52-57]. Using these outcomes, the percentage of treatment failure observed in these studies, over follow up periods ranging from 6 months to two years, varied between 10% and 32%. Risk factors for virologic treatment failure were poor adherence to antiretroviral medication, tuberculosis co-infection, lower CD4 counts at the initiation of ART, the use of nevirapine-based regimens and younger age at ART initiation [53, 54, 58]. The majority of those failing treatment were found to harbour HIV drug resistance mutations.

c) Antiretroviral drug resistance

HIV drug resistance mutations may be primary (transmitted when a treatment naive indi-vidual is infected by a strain of HIV-1 already resistant to one or more antiretroviral drugs) or acquired (developing in host after being on treatment for a given period of time).

i) Transmitted drug resistance

Transmitted drug resistance (TDR) has been shown to compromise the effectiveness of first line ART regimens (refs). Data from developed settings have shown increasing trends in the prevalence of TDR amongst new HIV infections. Specifically for example, TDR to NNRTIs in newly infected European individuals increased from 2.3% in 1996 – 1998 to 9.2% in 2001 – 2002 [59] while TDR in the US increased from 0% in 1996 – 1997 to 13.2% in 2000 – 2001 [60]. This has been mainly attributed to more widespread and longer use of ART in these settings.

1

some recent data suggest an increase in TDR in some parts of Africa. The IAVI early infection cohorts in East and Southern Africa have reported an increase of TDR amongst high risk adults in Zambia from 0% in 2005 to 15.8% in 2009, whilst there was no evidence of an increasing prevalence of TDR in their other regions [62]. More recently, data from Uganda has shown an increase in the prevalence of TDR amongst adults with new HIV infections from 0% in 2006 – 2007 to 8.6% in 2009 – 2010 [63].

In Kenya, a handful of studies have been done to assess the prevalence of TDR. A cross sectional study carried out among HIV-1 infected individuals prior to ART initiation in Nai-robi in 2005 found 4/53 (7.5%) new clients had TDR [64]. The IAVI early infection cohort reported an overall TDR prevalence of 3.1% from three sites in Kenya [62]. The multi-site cross sectional study by the PASER group, conducted between 2007 and 2009, reports TDR frequencies of 9/200 (4.5%) in Mombasa and 10/204 (4.9%) in Nairobi [65]. Data from the Kisumu Incidence Cohort study carried out between 2007 and 2009 reported TDR frequen-cies of 5/133 (3.8%) among 16-34 year old sexually active participants [66]. Importantly, and more recently, a cross sectional survey amongst newly diagnosed ARV naïve adults attending four VCT centers from Mombasa in 2009-10 reported a TDR prevalence of 13.2% [67] ii) Acquired drug resistance

The prevalence of acquired drug resistance amongst individuals failing first line regimens has been reported to be high in some African settings, ranging from 70% to 84% over follow up periods of 12 to 30 months on ART [54, 68, 69]. Reported risk factors for development of HIV drug resistance amongst individuals on treatment in these studies include lower baseline CD4 counts and higher viral loads, concurrent infections while on treatment and non-adherence to ARTs.

One study has evaluated acquired drug resistance to antiretroviral drugs in Kenyan adults [70]. Steegen and colleagues assessed 132 participants who had been on ART for >6 months at the Coast Province General Hospital in Mombasa. Viral suppression (<50 copies/ml) was observed in 113 (86%) participants. Sixteen of the 19 samples with detectable viral loads were successfully amplified and sequenced. Mutations associated with drug resistance in RT were detected in 14 of the 16 patients (88%). High-level resistance against at least 2 drugs of the ART regimen was observed in 9/14 (64%). The 3TC mutation M184V and the NNRTI mutation K103N were most frequent.

Rationale for thesis

The changing dynamics of the HIV-1 epidemic warrants an up-to-date description of the characteristics and outcomes of individuals enrolling for care, with an aim of understanding

22 Chapter 1

the epidemic and suggesting potential interventions to improve on outcomes. A paucity of data on characteristics and outcomes of patients enrolled for HIV care from rural parts of Kenya exists.

General objective

To describe the characteristics and outcomes of individuals enrolled for care in a rural HIV clinic in Coastal Kenya.

Specific objectives

a) Characteristics of Individuals enrolling for HIV Care:

i. To describe the distribution of HIV-1 subtypes among individuals enrolled for care in a rural HIV clinic in Coastal Kenya

ii. To establish the prevalence of HIV-1 Transmitted Drug Resistance among recently diagnosed individuals enrolling for care in a rural HIV clinic in Coastal Kenya

b) Outcomes of Individuals Enrolled for HIV Care (but prior to ART Initiation):

iii. To determine the incidence and predictors of pre- ART lost to follow-up among recently diagnosed individuals enrolling for care in a rural HIV clinic in Coastal Kenya iv. To describe the effect of maternal attrition from HIV care on retention of HIV-exposed

and HIV-infected infants enrolled for HIV care in a rural HIV clinic in Coastal Kenya c) Outcomes of Individuals Enrolled for HIV Care (and after ART Initiation):

v. To determine the incidence and predictors of retention in care among individuals starting first line ART regimens in a rural HIV clinic in Coastal Kenya

vi. To describe the prevalence and correlates of HIV-1 virologic failure and acquired drug resistance among first-line antiretroviral experienced adults at a rural HIV clinic in Coastal Kenya.

2

STUDY SITE

Kenya is one of the seven sub-Saharan countries with more than 1 million people living with HIV/AIDS by the end of 2011 [71], and has an average adult HIV prevalence of about 5.6% in 2012 [72]. Whilst the country reports a generalized epidemic, pockets of concentrated epidemics, especially among the high risk populations, are increasingly becoming evident [73]. Wide geographic variations in prevalence have also been observed, with Nyanza region having the highest prevalence of HIV infection among adults at 15.1%, whilst the North Eastern region reporting the lowest prevalence at 2.1%. An estimated prevalence of 4.3% was reported from the Coast province [72].

Kenya, like many other developing countries, has adopted a standardized public health ap-proach in the provision of ART. Free ART services became available in public health facilities in 2006. Since then, a rapid scale up has been experienced, with ART coverage among those eligible rising from 38% in 2007 to 72% by the end of 2011 [74]. As at the end of 2012, more than 500,000 individuals had been started ART in Kenya.

The current National ART eligibility guidelines [75], largely adopted from the 2010 WHO guidelines [76] recommend ART initiation in individuals with WHO clinical stage III and IV regardless of the CD4 T-cell count, and/or CD4 T-cell count of <350 cells/μL regardless of the WHO clinical staging [28]. First line therapy is mainly comprised of 2 NRTI’s and one NNRTI. Patients failing first line therapy are switched to a second line regimen, mainly comprising of 2 NRTI’s and a boosted PI.

The Comprehensive Care and Research Clinic

The studies described in this thesis were carried out at the Comprehensive Care and Research Clinic (CCRC) within Kilifi district hospital (KDH), a rural hospital in Coastal Kenya (figure 1). The District is one of the poorest in the country. Subsistence farming is the mainstay socioeconomic activity. The hospital serves a catchment population of more than 260,000 people from both within and outside the district. Even though a few public peripheral sites have been upgraded to offer comprehensive HIV care within the District, majority of the HIV infected individuals seek care from the CCRC.

The CCRC started registering individuals for HIV care in 2004. Standards of care are pro-vided according to the National AIDS and STI Control Programme (NASCOP) guidelines. Participants routinely undergo rapid HIV testing, voluntarily or provider initiated, at different settings within and outside the hospital setting. Whilst some of the confirmed HIV infected individuals are referred to peripheral health centers, majority are referred to the CCRC for enrolment into HIV care.

26 Chapter 2

In the CCRC, care is provided by counselors, nurses and clinical officers. Routine laboratory in-vestigations are undertaken at registration into care and every six months thereafter, or when clinically indicated. Newly registered patients are immediately prescribed daily cotrimoxazole and given a two-week appointment to assess for side effects and discuss laboratory results. Individuals are deemed ART-eligible if they meet the WHO criteria i.e. clinical stage III/IV or CD4 cell count of <350 cells/μl. Individuals not eligible for ART continue to receive a package of pre-ART care, including nutritional assesments and cotrimoxazole prophylaxis, and are monitored for ART eligibility. Individuals meeting the eligibility criteria are initiated ART and followed up monthly or two-monthly thereafter. Over time, the clinic had a consistent supply of the United States PEPFAR funded antiretrovirals offered to eligible patients at no cost. Recently, there is a transition towards government funded antiretroviral drugs in Kenya. Clients requesting to transfer their HIV care to other health facilities at any one point are issued with a standard referral note and their status updated on an electronic database. Data on deaths is also passively captured and dependant on reporting by health staff from in-patient wards, relatives, friends and/or acquaintances.

Electronic surveillance systems

When the clinic was started in 2004, an electronic data system (Filemaker, version 5.5) was implemented to capture sociodemographic characteristics of individuals registered for HIV

Figure 1: A map showing the location of the Comprehensive Care & Research Clinic (CCRC) within

the Kilifi District Hospital (KDH) and the main road linking the hospital to other parts of the district and neighboring districts to the north and the south

2

status, highest level of education and religion were routinely collected at the time of registra-tion into HIV care on standardized forms by trained counselors and fieldworkers. A trained data clerk entered these into the data system.

For long term clinical surveillance, and in addition to sociodemographic data, sub-systems to routinely collect clinical and laboratory data were established and interlinked in 2007. Clinical data including anthropometry, WHO clinical staging, opportunistic infections, ART status, ART start date, ART regimen and appointment dates were routinely captured at every clinic visit on real time in standardized forms by trained clinicians. Laboratory data including full blood count, CD4 T-cell counts and biochemistry were routinely captured after reception of results from the lab. Lab requests were made by clinicians when clinically indicated and according to the national guidelines.

Because of the longitudinal nature of these sub-systems, and to enhance efficiency in the data collection, the electronic systems were migrated to the MySQL platform in 2007. By the end of 2013, the clinic had registered 8911 HIV-infected individuals and HIV exposed infants for care. Of these, 3794 individuals were initiated on ART (figure 2).

Figure 2: Graph showing cumulative number of individuals enrolled for HIV care (N= 8911) and initiated

antiretroviral therapy (N=3794) at the Comprehensive Care and Research Clinic in Kilifi District Hospital

Figure 2: Graph showing cumulative number of individuals enrolled for HIV care (N= 8911) and initiated antiretroviral therapy (N=3794) at the Comprehensive Care and Research Clinic in Kilifi District Hospital

0 2000 4000 6000 8000 Fr eq ue nc y 2005 2003 2004 2006 2007 2008 2009 2010 2011 2012 2013 Year

28 Chapter 2

Figure 3: Distribution of individuals enrolled for HIV Care in the Comprehensive Care and Research

Clinic between 2004 and 2013 by age and gender (N=8911).

Figure 4: Distribution of individuals registering for HIV care at the Comprehensive Care and Research

2

[73.2%]) were adults (≥15 years). Of these, more than two thirds (n=4583 [70.3%]) were female (figure 3).

Whilst most of the females registering for HIV care were aged 25 – 29 years, most of the males were aged 35 – 39 years old. Overall, the mean age at registration into care was 34.9 (95% CI: 34.7 – 35.2) years. Male clients were older at registration into care, compared to their female counterparts (38.4 [38.0 – 38.8] vs. 33.5 [33.2 – 33.8] years). Indeed, this was evident and consistent over the years (figure 4).

Only about a quarter (n=2393 [26.8%]) of registered individuals were children (<15 years). In fact, almost two-thirds of the children enrolled (n=1503 [62.8%]) were infants (<18 months) born to HIV infected mothers.

According to the Kenyan national guidelines, all infants born to HIV-infected mothers should be enrolled for care, started on cotrimoxazole prophylaxis, receive prevention of mother to child transmission (PMTCT) interventions and followed up until they are 18 months old. During follow up, and according to the national HIV early Infant diagnosis (EID) algorithm, a dried blood spot (DBS) sample should be taken for polymerase chain reaction (PCR) at six weeks. Nationally, PCR for EID are done centrally from a handful of government designated laboratories located in different parts of the country. An antibody test is also recommended at 9 months, with another confirmatory antibody test at 18 months. If an infant tests PCR or antibody positive at any point, the current algorithm recommends immediate initiation of antiretroviral therapy. In 2012, national guidelines were revised to recommend follow-up of mother-infant pairs in Maternal and Child Health clinics.

In the CCRC, EID was started in 2006. By 2012, 1503 exposed infants registered for EID and HIV care. Of these, 917 (61.0%) had a DBS taken for PCR. The remaining infants did not have a PCR done because they were either registered at an older age (>6 months, n=346 [23.0%]) or were lost to follow up before the test could be done. The median age at PCR was 1.4 months (range, 0.4 – 6.0) months. Overall, our data suggest that PCR positivity has halved over time, from an estimated 19.3% in 2006 to 10.3% in 2012 (figure 5). Our data therefore suggests that despite the roll out of PMTCT interventions, overall mother to child transmission remains more than 10% in this population.

I have provided herewith a brief description of the basic demographic characteristics of indi-viduals registered for HIV care in the CCRC to put studies in this thesis in context. Detailed clinical and molecular characteristics and outcomes are provided forthwith in the main body of the thesis.

30 Chapter 2

OUTLINE OF THE THESIS

As an introduction to the thesis, I start by giving a brief background on the biology of the human immunodeficiency virus (HIV), the status of the HIV epidemic and a description of the available antiretrovirals used in the management of HIV and AIDS (Chapter 1). The chapter ends with an outline of the objectives of this thesis.

To put the findings of the studies in this thesis in context, we describes the setting under which all the studies were carried out (Chapter 2). This chapter also briefly describes the surveillance systems and the distribution of individuals registered for HIV care in the clinic, where studies for this thesis were done. The chapter ends with a brief outline of the three core parts of the thesis: Characteristics, Pre-ART and On-ART outcomes of Individuals enroll-ing for HIV care.

Part 1 of this thesis describes the characteristics of individuals enrolling for care in a rural HIV clinic at a district hospital in Coastal Kenya. I carried out a study to describe the patient population, distribution and subtype diversity of HIV-1 infections among individuals enrolling for HIV care in setting (Chapter 3). In addition, I also analyzed samples from ART naïve HIV infected adults enrolling for HIV care to determine the prevalence of HIV-1 transmitted drug resistance in this population (Chapter 4).

Figure 5: Distribution of PCR positivity among infants enrolled for Early Infant Diagnosis and HIV Care

2

Initiation. I followed up individuals enrolled for HIV care in a cohort study to assess for incidence and predictors of early drop out from routine pre-ART care in our clinic (Chapter 5). I also followed up infants born to HIV-infected mothers in a mixed methods design to determine uptake and constraints of the early infant diagnosis (EID) process for detection of HIV infection and to describe loss to follow up of these infants from HIV care (Chapter 6). Part 3 of the thesis describes outcomes of individuals enrolled for HIV care after ART initia-tion. I followed up individuals initiated on ART in a cohort design to describe the incidence and predictors of attrition (loss to follow up and death) from care (Chapter 7). I also cross sectionally sampled individuals on ART to deteremine the prevalence and correlates of viro-logic treatment failure and acquired drug resistance (Chapter 8).

In the last chapter of the thesis, I summarize and discuss the main findings (Chapter 9) and their implications on future perspectives. The chapter ends with concluding remarks on the characteristics and outcomes of HIV-infected Individuals enrolled for HIV care in our rural HIV Clinic in Coastal Kenya.

32 Chapter 2

REFERENCES

1. Pneumocystis pneumonia--Los Angeles. MMWR Morb Mortal Wkly Rep, 1981. 30(21): p. 250-2. 2. UNAIDS, Report on the Global AIDS epidemic. 2009.

3. Coetzee, D., et al., Outcomes after two years of providing antiretroviral treatment in Khayelitsha,

South Africa. AIDS, 2004. 18(6): p. 887-95.

4. Laurent, C., et al., Long-term benefits of highly active antiretroviral therapy in Senegalese

HIV-1-infected adults. J Acquir Immune Defic Syndr, 2005. 38(1): p. 14-7.

5. Lohse, N., et al., Survival of persons with and without HIV infection in Denmark, 1995-2005. Ann Intern Med, 2007. 146(2): p. 87-95.

6. Morgan, D., et al., HIV-1 infection in rural Africa: is there a difference in median time to AIDS and

survival compared with that in industrialized countries? AIDS, 2002. 16(4): p. 597-603.

7. UNAIDS, 2012. Sub-Saharan Africa - AIDS epidemic facts and figures. [accessed, 19th June 2013]; Available from: http://www.unaids.org/en/media/unaids/contentassets/documents/epide-miology/2012/gr2012/2012_FS_regional_ssa_en.pdf

8. De Cock, K.M. and F. Brun-Vezinet, Epidemiology of HIV-2 infection. AIDS, 1989. 3 Suppl 1: p. S89-95.

9. Taylor, B.S. and S.M. Hammer, The challenge of HIV-1 subtype diversity. N Engl J Med, 2008.

359(18): p. 1965-6.

10. Buonaguro, L., M.L. Tornesello, and F.M. Buonaguro, Human immunodeficiency virus type 1

subtype distribution in the worldwide epidemic: pathogenetic and therapeutic implications. J

Virol, 2007. 81(19): p. 10209-19.

11. Freed, E.O., HIV-1 replication. Somat Cell Mol Genet, 2001. 26(1-6): p. 13-33.

12. Goldschmidt, V. and R. Marquet, Primer unblocking by HIV-1 reverse transcriptase and resistance

to nucleoside RT inhibitors (NRTIs). Int J Biochem Cell Biol, 2004. 36(9): p. 1687-705.

13. Sluis-Cremer, N. and G. Tachedjian, Mechanisms of inhibition of HIV replication by nonnucleoside

reverse transcriptase inhibitors. Virus Res, 2008. 134(1-2): p. 147-56.

14. Pillay D., B.M., Getman D. and Richard D. D., HIV-1 Protease inhibitors: Their development,

mechanism of action and clinical potential. Rev. Med. Virol., 1995. 5: p. 23-33.

15. Pendri, A., et al., New first and second generation inhibitors of human immunodeficiency virus-1

integrase. Expert Opin Ther Pat, 2011. 21(8): p. 1173-89.

16. Briz, V., E. Poveda, and V. Soriano, HIV entry inhibitors: mechanisms of action and resistance

pathways. J Antimicrob Chemother, 2006. 57(4): p. 619-27.

17. Osmanov, S., W.L. Heyward, and J. Esparza, HIV-1 genetic variability: implications for the

devel-opment of HIV vaccines. Antibiot Chemother (1971), 1996. 48: p. 30-8.

18. Marcelin, A.G., Resistance to nucleoside reverse transcriptase inhibitors, in Antiretroviral

Resis-tance in Clinical Practice, A.M. Geretti, Editor 2006, Mediscript: London.

19. Boyer, P.L., et al., Analysis of nonnucleoside drug-resistant variants of human immunodeficiency

virus type 1 reverse transcriptase. J Virol, 1993. 67(4): p. 2412-20.

20. Ridky, T. and J. Leis, Development of drug resistance to HIV-1 protease inhibitors. J Biol Chem, 1995. 270(50): p. 29621-3.

21. Keele, B.F., et al., Chimpanzee reservoirs of pandemic and nonpandemic HIV-1. Science, 2006.

313(5786): p. 523-6.

22. Korber, B., et al., Timing the ancestor of the HIV-1 pandemic strains. Science, 2000. 288(5472): p. 1789-96.

2

23. Lemey, P., et al., Tracing the origin and history of the HIV-2 epidemic. Proc Natl Acad Sci U S A, 2003. 100(11): p. 6588-92.

24. Worobey, M., et al., Direct evidence of extensive diversity of HIV-1 in Kinshasa by 1960. Nature, 2008. 455(7213): p. 661-4.

25. Zhu, T., et al., An African HIV-1 sequence from 1959 and implications for the origin of the

epidemic. Nature, 1998. 391(6667): p. 594-7.

26. UNAIDS, UNAIDS OUTLOOK REPORT. 2010.

27. Gilks, C.F., et al., The WHO public-health approach to antiretroviral treatment against HIV in

resource-limited settings. Lancet, 2006. 368(9534): p. 505-10.

28. WHO, 2009. Rapid advice: antiretroviral therapy for HIV infection in adults and adolescents. [accessed, 18th June 2013; Available from: http://www.who.int/hiv/pub/arv/advice/en/index.html 29. Larson, B.A., et al., Early loss to follow up after enrolment in pre-ART care at a large public clinic

in Johannesburg, South Africa. Trop Med Int Health, 2010. 15 Suppl 1: p. 43-7.

30. Lessells, R.J., et al., Retention in HIV care for individuals not yet eligible for antiretroviral therapy:

rural KwaZulu-Natal, South Africa. J Acquir Immune Defic Syndr, 2011. 56(3): p. e79-86.

31. Losina, E., et al., The “ART” of linkage: pre-treatment loss to care after HIV diagnosis at two

PEPFAR sites in Durban, South Africa. PLoS One, 2010. 5(3): p. e9538.

32. Fox, M.P. and S. Rosen, Patient retention in antiretroviral therapy programs up to three years on

treatment in sub-Saharan Africa, 2007-2009: systematic review. Trop Med Int Health, 2010. 15

Suppl 1: p. 1-15.

33. Lawn, S.D., et al., Determinants of mortality and nondeath losses from an antiretroviral

treat-ment service in South Africa: implications for program evaluation. Clin Infect Dis, 2006. 43(6): p.

770-6.

34. Toure, S., et al., Rapid scaling-up of antiretroviral therapy in 10,000 adults in Cote d’Ivoire: 2-year

outcomes and determinants. AIDS, 2008. 22(7): p. 873-82.

35. Wools-Kaloustian, K., et al., Viability and effectiveness of large-scale HIV treatment initiatives in

sub-Saharan Africa: experience from western Kenya. AIDS, 2006. 20(1): p. 41-8.

36. Rosen, S., M.P. Fox, and C.J. Gill, Patient retention in antiretroviral therapy programs in

sub-Saharan Africa: a systematic review. PLoS Med, 2007. 4(10): p. e298.

37. Boulle, A., et al., Antiretroviral therapy and early mortality in South Africa. Bull World Health Organ, 2008. 86(9): p. 678-87.

38. Fenner, L., et al., Early Mortality and Loss to Follow-up in HIV-Infected Children Starting

Antiret-roviral Therapy in Southern Africa. J Acquir Immune Defic Syndr, 2010.

39. Russell, E.C., et al., Low haemoglobin predicts early mortality among adults starting antiretroviral

therapy in an HIV care programme in South Africa: a cohort study. BMC Public Health, 2010. 10:

p. 433.

40. Amuron, B., et al., Mortality and loss-to-follow-up during the pre-treatment period in an

anti-retroviral therapy programme under normal health service conditions in Uganda. BMC Public

Health, 2009. 9: p. 290.

41. Bassett, I.V., et al., Loss to care and death before antiretroviral therapy in Durban, South Africa. J Acquir Immune Defic Syndr, 2009. 51(2): p. 135-9.

42. Brinkhof, M.W., et al., Early loss of HIV-infected patients on potent antiretroviral therapy

pro-grammes in lower-income countries. Bull World Health Organ, 2008. 86(7): p. 559-67.

43. Bassett, I.V., et al., Who starts antiretroviral therapy in Durban, South Africa?... not everyone who

34 Chapter 2

44. Lawn, S.D., et al., Early mortality among adults accessing antiretroviral treatment programmes in

sub-Saharan Africa. AIDS, 2008. 22(15): p. 1897-908.

45. Towards universal access: scaling up priority HIV/AIDS interventions in the health sector. Progress Report 2008.

46. Newell, M.L., et al., Mortality of infected and uninfected infants born to HIV-infected mothers in

Africa: a pooled analysis. Lancet, 2004. 364(9441): p. 1236-43.

47. HIV DRUG RESISTANCE EARLY WARNING INDICATORS: World Health Organization indicators to

monitor HIV drug resistance prevention at antiretroviral treatment sites. 2010.

48. WHO (2010) Antiretroviral Therapy for HIV Infection in Adults and Adolescents Recommendation

for a Public Health Approach 2010 Revision. 2011.

49. Mugyenyi, P., et al., Routine versus clinically driven laboratory monitoring of HIV antiretroviral

therapy in Africa (DART): a randomised non-inferiority trial. Lancet, 2010. 375(9709): p. 123-31.

50. Reynolds, S.J., et al., Failure of immunologic criteria to appropriately identify antiretroviral

treat-ment failure in Uganda. AIDS, 2009. 23(6): p. 697-700.

51. Ruel, T.D., et al., Early virologic failure and the development of antiretroviral drug resistance

mu-tations in HIV-infected Ugandan children. J Acquir Immune Defic Syndr, 2011. 56(1): p. 44-50.

52. Barth, R.E., et al., Virological follow-up of adult patients in antiretroviral treatment programmes

in sub-Saharan Africa: a systematic review. Lancet Infect Dis. 10(3): p. 155-66.

53. Datay, M.I., et al., Associations with virologic treatment failure in adults on antiretroviral therapy

in South Africa. J Acquir Immune Defic Syndr. 54(5): p. 489-95.

54. El-Khatib, Z., et al., Viremia and drug resistance among HIV-1 patients on antiretroviral

treat-ment: a cross-sectional study in Soweto, South Africa. Aids. 24(11): p. 1679-87.

55. Emmett, S.D., et al., Predicting virologic failure among HIV-1-infected children receiving

anti-retroviral therapy in Tanzania: a cross-sectional study. J Acquir Immune Defic Syndr. 54(4): p.

368-75.

56. Ford, N., et al., Early adherence to antiretroviral medication as a predictor of long-term HIV

viro-logical suppression: five-year follow up of an observational cohort. PLoS One. 5(5): p. e10460.

57. Fox, M.P., et al., Initiating patients on antiretroviral therapy at CD4 cell counts above 200 cells/

microl is associated with improved treatment outcomes in South Africa. Aids. 24(13): p. 2041-50.

58. Ahoua, L., et al., Evaluation of a 5-year programme to prevent mother-to-child transmission of

HIV infection in Northern Uganda. J Trop Pediatr, 2010. 56(1): p. 43-52.

59. Wensing, A.M., et al., Prevalence of drug-resistant HIV-1 variants in untreated individuals in

Europe: implications for clinical management. J Infect Dis, 2005. 192(6): p. 958-66.

60. Grant, R.M., et al., Time trends in primary HIV-1 drug resistance among recently infected persons. JAMA, 2002. 288(2): p. 181-8.

61. Bennett, D.E., et al., Recommendations for surveillance of transmitted HIV drug resistance in

countries scaling up antiretroviral treatment. Antivir Ther, 2008. 13 Suppl 2: p. 25-36.

62. Price, M.A., et al., Transmitted HIV type 1 drug resistance among individuals with recent HIV

infection in East and Southern Africa. AIDS Res Hum Retroviruses, 2011. 27(1): p. 5-12.

63. Ndembi, N., et al., Transmitted antiretroviral drug resistance among newly HIV-1 diagnosed

young individuals in Kampala. AIDS, 2011. 25(7): p. 905-10.

64. Lihana, R.W., et al., HIV type 1 subtype diversity and drug resistance among HIV type 1-infected

Kenyan patients initiating antiretroviral therapy. AIDS Res Hum Retroviruses, 2009. 25(12): p.

2

65. Hamers, R.L., et al., HIV-1 drug resistance in antiretroviral-naive individuals in sub-Saharan Africa

after rollout of antiretroviral therapy: a multicentre observational study. Lancet Infect Dis, 2011.

11(10): p. 750-9.

66. Otecko, N., et al., 2012. Viral and Host Characteristics of Recent versus Established HIV-1 Infec-tions in Kisumu: Secondary analysis of the Kisumu Incidence Cohort Study (KICoS). 6th INTEREST workshop [accessed; Abstract O_06: [Available from: http://regist2.virology-education.com/ abstractbook/2012_4.pdf

67. Sigaloff, K.C., et al., High Prevalence of Transmitted Antiretroviral Drug Resistance Among Newly

HIV Type 1 Diagnosed Adults in Mombasa, Kenya. AIDS Res Hum Retroviruses, 2012.

68. Ahoua, L., et al., Risk factors for virological failure and subtherapeutic antiretroviral drug

concen-trations in HIV-positive adults treated in rural northwestern Uganda. BMC Infect Dis, 2009. 9: p.

81.

69. Barth, R.E., et al., Rapid accumulation of nonnucleoside reverse transcriptase inhibitor-associated

resistance: evidence of transmitted resistance in rural South Africa. Aids, 2008. 22(16): p. 2210-2.

70. Steegen, K., et al., Effectiveness of antiretroviral therapy and development of drug resistance in

HIV-1 infected patients in Mombasa, Kenya. AIDS Res Ther, 2009. 6: p. 12.

71. UNAIDS, 2012. World AIDS Day Report. [accessed, 23rd October 2013]; Available from: http:// www.unaids.org/en/media/unaids/contentassets/documents/epidemiology/2012/gr2012/ JC2434_WorldAIDSday_results_en.pdf

72. National AIDS and STI Control Programme, September 2013. Kenya AIDS Indicator Survey 2012: Preliminary Report. [accessed, 03rd October 2013]; Available from: http://nascop.or.ke/library/3d/ Preliminary%20Report%20for%20Kenya%20AIDS%20indicator%20survey%202012.pdf 73. Bezemer, D., et al., HIV Type 1 Transmission Networks Among Men Having Sex with Men and

Heterosexuals in Kenya. AIDS Res Hum Retroviruses, 2013.

74. National AIDS Control Council (NACC) and National AIDS and STI Control Program (NASCOP), 2012. Kenya AIDS Epidemic Update 2012. [accessed, 10 September 2013; Available from: http:// www.nascop.or.ke/library/3d/FINAL%20Kenya%20Updat%202012,%2030%20May.pdf 75. Guidelines for antiretroviral therapy in Kenya. [accessed, 23 October 2013]; 4th Edition 2011:

[Available from: http://healthservices.uonbi.ac.ke/sites/default/files/centraladmin/healthservices/ Kenya%20Treatment%20Guidelines%202011.pdf

76. World Health Organization, Antiretroviral therapy for HIV infection in adults and adolescents: recommendations for a public health approach - 2010 rev. [accessed, 20 Nov 2012]; Available from: http://whqlibdoc.who.int/publications/2010/9789241599764_eng.pdf

Characteristics of Individuals

Enrolling for HIV Care

HIV-1 in a rural coastal town in Kenya

shows multiple introductions with many

subtypes and much recombination.

Stéphane Hué S, Amin S. Hassan, Helen Nabwera, Eduard J. Sanders,

Deenan Pillay, James A. Berkley and Patricia A. Cane.

40 Chapter 3

ABSTRACT

The extent of HIV-1 diversity was examined among patients attending a rural district hospital in a coastal area of Kenya. The pol gene was sequenced in samples from 153 patients. Subtypes were designated using REGA, SCUEAL and jpHMM programmes. The most com-mon subtype was A1, followed by C and D; A2 and G were also detected. However, a large proportion of the samples were found to be recombinants, which clustered within the pure subtype branches. Phylogeographic analysis of Kilifi sequences compared with those from other regions of Africa showed that while many sequences were closely related to se-quences from Kenya others were most closely related to known sese-quences from other parts of Africa, including West Africa. Overall, these data indicate that there have been multiple introductions of HIV-1 into this small rural town and surrounds with ongoing diversity being generated by recombination.

3

infection. HIV-1 M group is subdivided into 9 pure subtypes (A-D, F, G, H, J, and K), some of which may be further subdivided into sub-subtypes (eg A1, A2 etc)1_{. There is considerable} geographic influence on circulating subtypes with some countries having a very high propor-tion of infecpropor-tions of a single subtype, for example, HIV-1 in North America is predominantly subtype B while subtype C predominates in Southern Africa. However, multiple subtypes co-exist in populations and this can lead to the generation of inter-subtype recombinant forms. It has been postulated that HIV-1 subtypes A and D were introduced into East Africa after 1950 and spread exponentially during the 1970s, with the rapid spread in part being due to the strong interconnectivity between major population centres in the area2_{. Studies in} Kenya, mainly based in Nairobi, have confirmed the predominance of subtype A with sub-type D being much less common, with occasional other subsub-types and recombinants being detected3,4_{. In a full genome characterisation of 41 strains from blood donors in 1999-2000 it} was found that 25 (61%) were pure subtype (23 A1, 1 C and 1 D) and the rest inter-subtype recombinants of which A1-D was predominant (15%) then A2-D and A1-C (both 7%) with A1-A2-D, A1-C-D, A1-G, and C-D also found5_.

It has previously been reported that, using a fragment of env gene of 86 samples from the Kenyan coastal strip, including 27 samples from Kilifi, 86% of samples were subtype A1, 5% were subtype C, 8% were subtype D and 1% was subtype G6_{. Full-length genome} sequenc-ing of samples from 21 individuals from Mombasa found a 74% of 23 isolates had pure subtype A strains while the rest were recombinants, including A-D, D-G, A-C, A-A2-C-D7_. Here we report on the subtype diversity of HIV-1 infections among patients attending the comprehensive care and research clinic (CCRC, HIV clinic) at Kilifi District Hospital (KDH), Kenya, looking at pol gene sequences from 153 individuals attending the clinic. We confirm the predominance of subtype A1 but also report the presence of multiple other subtypes (A2, C, D, G) together with many novel recombinants. We further analyse the phylogeography of the sequences and show that there have been multiple introduction of HIV-1 into the area. Kilifi is a small town serving a mainly rural population of about 250,000 in coastal Kenya about 50 km north of Mombasa. It lies on the main coastal tarmac road between Mom-basa and Somalia, at a crossing point of the estuary of the Kilifi River, with most of the population being rural subsistence farmers. The prevalence of HIV-1 in the coastal province of Kenya in 15-49 year olds is estimated at 8.1% (male 6.7%, female 8.9%)8_{. KDH is a} government hospital that has been providing comprehensive HIV services including free anti-retroviral therapy (ART) and prevention of mother to child transmission since 2004. At

42 Chapter 3

the end January 2010, 2,618 patients were on active follow up in the HIV clinic, 47% of whom were on ART.

All samples were collected from attendees at the HIV clinic between July 2008 and June 2009, and were either new diagnoses (n=121) or patients undergoing treatment with antiretroviral therapy (n=32). Overall, 72% of patients were female and 23% were children.

Population sequencing of 1245 nucleotides was carried out on PCR amplicons to give sequence for codons 5-99 of protease and 1-320 of reverse transcripase using in-house methods. The sequences have accession numbers HQ441597-HQ441749. The sequences were manually aligned using the sequence editor Se-Al v2.0 (http://tree.bio.ed.ac.uk/software/seal/). Subtypes were assigned to the sequences using 3 methods, SCUEAL9_{, using the default reference} sequences, REGA version 2.0 (http://dbpartners.stanford.edu/RegaSubtyping/), and jpHMM10_. With all three methods, the default settings, including window size, were used. SCUEAL sub-type designation of A or ancestral A were called as A1, based on the phylogenetic clustering. Considerable complexity was observed in the subtypes of the samples, with 42/153 (27%) of the samples not giving concordant results using all 3 methods. However, by all methods sub-type A1 was the most common, comprising 54% of samples by SCUEAL, 59% by jpHMM, and 61% by REGA. The next most common subtypes were C, 8% by SCUEAL and jpHMM and 9% by REGA, and D, 9% SCUEAL and REGA and 10% by jpHMM.

The jpHMM analysis did not detect any pure A2 sequences whereas 1% by SCUEAL and 9% by REGA of samples were found to be A2 although there were no samples found to be A2 by both these latter methods. One sample was designated subtype G by all 3 methods. The rest of the samples were inter-subtype recombinants (27% by SCUEAL, 23% by jpHMM and 12% by REGA). The subtype designations by all three methods are summarised in Table 1. Inter-subtype recombinants detected by the various methods included A1-AE, A1-A2, A1-A2-B, A1-A2-D, A1-C, A1-D, A2-A1-A2-B, A2-D, AE-C, B-C, C-D, CRF-16-like. Concordance between all three methods as to the constituents of the recombinant sequences was present for only 8 samples. The detection of subtype B sequences in 5 sequences by jpHMM and in 1 by SCUEAL was surprising given the rarity of subtype B in Africa, though the lack of agreement between the methods in detection of this subtype possibly indicates that this was an artefact. In addition, some recombinants were designated as “complex” or had regions that were not classified and where there was a lack of high confidence in the subtype designations and breakpoints. These sequences included multiple subtypes and may be the consequence of further recombination between recombinants.

3

The phylogeny of the 153 Kilifi sequences together with HIV-1 subtype A1, A2, B, C, D, G, CRF01_AE and CRF02_AG reference sequences from the Los Alamos HIV sequence database (http://www.hiv.lanl.gov/) was reconstructed using the program MrBayes v.3.1.211_{, under the} general time reversible (GTR) model of nucleotide substitution with gamma-distributed rate heterogeneity. The GenBank accession number of the Los Alamos reference sequences is as follows: A (AB004885, AB253429, DQ253421, DQ676872), A2 (AF286237, AF286238), B (AY173951, AY253311, AY331295, AY423387, K03454, K03455), C (AF067155, AY772699, U46016, U52953), D (AY371157, U88824), G (AF061641, AF084936, AY612637, U88826), AE (AB220944, U54771) and AG (AY271690, L39106). The Bayesian Markov chain Monte Carlo (MCMC) search was set to 8,000,000 iterations, with trees sampled every 100th generations. Convergence of the estimates was determined with the software Tracer v1.5 (http://tree.bio.ed.ac.uk/software/tracer/), as indicated by an effective sampling size > 200. A maximum clade credibility tree (MCCT) was selected from the sampled posterior distribution

Table 1: Subtype designations by SCUEAL, REGA and jpHMM

Subtype Frequency by SCUEAL % (n) Frequency by REGA % (n) Frequency by jpHMM % (n) A1 54% (82) 61% (93) 59% (90) A1-AE recombinant 1% (2) A1-C recombinant 1% (2) 2% (3) 2% (3) A1-D recombinant 7% (10) 5% (8) 7% (11) A1-Unknown <1% (1) A1-A2 recombinant 1% (2) 1% (2) A1-A2-Unknown 1% (2) A1-A2-B 1% (2) A1-A2-D recombinant 2% (3) A2 1% (2) 9% (13) A2-B 2% (3) A2-D recombinant 3% (4) 3% (5) AE-C recombinant <1% (1) B-C recombinant <1% (1) C 8% (12) 9% (14) 8% (13) C-D recombinant <1% (1) 2% (3) <1% (1) C-Unknown <1% (1) CRF16-like 3% (5) Complex 10% (16) <1% (1) D 9% (14) 9% (14) 10% (16) D-Unknown G <1% (1) <1% (1) <1% (1) Unknown <1% (1)

44 Chapter 3

Figure 1: Phylogeny of the 153 HIV-1 isolates from Kilifi. A maximum clade credibility tree was retrieved

from a posterior distribution of Bayesian MCMC 10000 trees, under the GTR+G model of nucleotide substitution. Branches are coloured according to the HIV-1 subtype assigned by the program SCUEAL. HIV subtype reference sequences are indicated in black, with the corresponding GenBank accession number. Differences with REGA subtype assignations are indicated in brackets. Bayesian posterior prob-abilities of 1.00 and above 0.90 are shown on the branches by two and one asterisks respectively. Discordant sequences are indicated as #1 to #14. Branch lengths indicate the number of substitutions per nucleotide sites.