Centre activities will initially comprise five research programs, each with a number of closely related projects, plus supporting programs in research training and post-graduate education. Research projects will focus on a number of areas that fall under the broad umbrella of clinical and translational research. In operational terms, all research areas follow a logical flow of hypothesis generation and hypotheses testing from genetic sequence to 'functional' cell and protein levels.

1. HIV-host interaction
   Dr Mina John
2. Understanding interactions between host human leukocyte antigens and hepatitis C virus adaptive responses
   Dr Silvana Gaudieri
3. Genetic susceptibility to drug hypersensitivity reactions
   Dr David Nolan
4. Antiretroviral drug toxicity: lipoatrophy
   Dr David Nolan
5. Biostatistical methods
   Professor Ian James

Research Program 1: HIV-host interaction

The Centre has innovated the field of interacting population genetics from its work on the interaction of HIV with variable (polymorphic) human proteins such as HLA molecules.

The Centre has developed a highly sophisticated and novel analytical approach to analysing host and viral factors that generate diversity of HIV genetic sequences both at the individual and population levels. This research has shown that HIV diversity is in large part driven and shaped by HLA diversity at population level and this interaction is critical in determining viral load. This observation, published in Science in 2002 [1], is considered a landmark in HIV biology and has effectively opened up a new field of host-pathogen interactions. Work to date has resulted in comprehensive host genetic assessment and full-length HIV genetic sequencing in approximately 240 individuals from the Western Australian HIV cohort study - thus creating the single largest repository of genetic data relating to HIV and host viral diversity in the world.

The Centre will extend these studies to multiple world populations with a view to designing vaccine immunogens optimised to combat HLA-drive adaptation and HIV diversity in these populations. A global collaboration between groups in Europe, USA, Africa and South America has been negotiated. These extensive studies will allow construction of predictive models of viral load set-point according to HLA genotype and HIV-1 sequence, allowing for dissection of the consequences of viral adaptation on disease progression. This program aims to demonstrate that host HLA alleles influence evolution of the population-adapted consensus sequence and HIV-1 clades. Putative CTL and CD4 epitopes will be identified based on this population-based approach, and the correlation to epitopes identified by ex-vivo CTL assays will be confirmed.

1. Expand host and viral population genetic data;
2. Optimise high-throughput HIV genetic sequencing techniques;
3. Innovate methods for epitope prediction and vaccine optimisation based on population viral sequence data; and
4. Confirm putative HLA Class I and II-restricted epitopes using cellular assays to analyse patient specific T cell responses.

Research Program 2: Understanding interactions between host HLA and hepatitis C virus adaptive responses

The work on HIV has been generalised to Hepatitis C and the principles can be applied to any interacting genomes. Therefore the work can be applied to many other areas in medicine including cancer research and any of the life sciences including agriculture, environmental science and Veterinary Medicine.

This research builds on the experience with assessing host-HIV interactions, using a population-based approach to characterise HCV adaptation to HLA-restricted immune responses and to identify new CTL and CD4⁺ T helper epitopes that will represent true in-vivo targets of immune responses. A multi-centre Australian cohort (~25% HCV/HIV co-infected, n=400) and HCV/HIV co-infected patients from the Swiss HIV cohort (n=190) will be utilised. The influence of HLA-specific CTL viral escape mutations on HCV infection outcomes in acute as well as chronic HCV infection will be examined, along with the influence of HIV disease on the progression of HCV in co-infected HIV/HCV patients. It is anticipated that clinically relevant predictive models of HCV infection outcome and disease progression, using HLA genotype and HCV sequence data, will be obtained.

1. Use a population-based approach to characterise HCV adaptation to the HLA-restricted immune pressure in chronic HCV infection;
2. Examine the influence of HLA and HCV adaptation to HLA-restricted immune pressure in acute and newly acquired HCV infection;
3. Compare polymorphic sites within the consensus sequences shaped by the different populations (US and Australia); and
4. Examine the effect of HIV co-infection on HCV adaptation to HLA-restricted immune responses.

Research Program 3: Genetic susceptibility to drug hypersensitivity reactions

The Centre has delineated genetic risk factors for hypersensitivity reactions to antiretroviral drugs previously thought to be essentially unpredictable. Success in this endeavour has been most notable for the NRTI drug abacavir. CCIBS has identified a highly predictive association between HLA-B*5701 and related haplotypic MHC genetic markers and abacavir hypersensitivity reaction. This also serves as an ideal model system to explore multiple genetic determinants of Class I MHC-restricted immune responses. This program aims to map susceptibility MHC gene(s) associated with abacavir hypersensitivity, utilising recombinant haplotype mapping in abacavir-exposed cohorts from Western Australia (n=270), Canada (n=600) and Switzerland (n=1200), with precise diagnostic classification using clinical data and adjunctive tests (eg. epicutaneous patch testing). Genetic associations will be reinforced with functional studies investigating the cellular events involved in priming and initiation of abacavir-specific immune responses in antigen presenting cells, and cytokine production in effector cells. Studies will be undertaken to map biochemical pathways (eg. generation of immunogenic aldehyde derivatives) and immunological pathways (eg. role of chaperones such as heat shock protein) involved in directing hypersensitivity responses to abacavir. Drug-protein interactions will be investigated to identify antigenic peptides derived from the association of abacavir metabolites with endogenous proteins. More recently, this paradigm has been applied to nevirapine hypersensitivity with encouraging results suggesting that nevirapine hypersensitivity may also be largely predictable and therefore preventable. The program aims to extend the work to other drugs that cause potentially life-threatening reactions as well as elucidate the immunological mechanisms underlying these reactions

1. Characterise HLA allele or haplotype association with abacavir and nevirapine hypersensitivity in different ethnic groups;
2. Develop predictive models of drug hypersensitivity incorporating genetic, demographic and HIV disease-related variables;
3. Characterise the pathways in HLA class I (abacavir) and classII (nevirapine) presentation and pathological responses;
4. Elucidate drug bioactivation pathways relevant to HSR; and
5. Elucidate the role of dendritic cells (DC) in priming and development of HSR

Research Program 4: Antiretroviral drug toxicity: lipoatrophy

CCIBS has a large body of work relating to the contribution of non-nucleoside reverse transcriptase inhibitor (NRTI) drug therapy to the pathogenesis of 'mitochondrial toxicity' syndromes, with an emphasis on a highly prevalent syndrome of pathological subcutaneous fat loss referred to as lipoatrophy. The Centre has sought to clarify mechanisms underlying lipodystrophy complications among patients receiving long-term antiretroviral therapy (ART), testing the hypothesis that adipose-specific mitochondrial DNA depletion induced by nucleoside analogue reverse transcriptase inhibitor (NRTI) drugs has a role in lipodystrophy pathogenesis. For ongoing studies the program aims to investigate the effects of antiretroviral drugs on adipocyte mitochondrial DNA content and tissue mitochondrial toxicity (mitochondrial DNA-encoded protein expression, mitochondrial function, markers of mitochondrial biogenesis, tissue morphology at the light and ultrastructural levels, and apoptosis markers), and to assess effects of specific NRTI and HIV protease inhibitor drugs on adipocyte lipid/fatty acid metabolic pathways. Analyses will be conducted on subcutaneous fat samples (n>200) and clinical data (eg serial DEXA scans and lipid assessments) from participants in the Western Australian HIV Cohort, utilising appropriate (including novel) analyses of longitudinal response data. This repository of adipose tissue samples under a wide range of HIV drug therapy represents the largest of its kind internationally.

1. Expand the subcutaneous fat biopsy resource, with an emphasis on longitudinal data and contemporary HIV therapy;
2. Assess mtDNA content, expression of mtDNA-encoded protein and mitochondrial respiratory chain function according to current NRTI and PI therapy;
3. Analyse cellular morphology, utilising confocal and histopathological assessment;
4. Profile adipose tissue cytokine regulation and gene expression; and
5. Evaluate objective measures of fat loss over time.

Research Program 5: Biostatistical methods

The Biostatistical Methods research program aims to facilitate use of the most appropriate and powerful statistical analyses in all aspects of the Centre's work. It is driven by the implementation and modification of existing methods where these exist, and by development of new approaches to answer the complex questions which arise where existing techniques are inadequate or sub-optimal. Non-standard and sophisticated statistical methods have underpinned the work of the Centre and have enabled the accommodation of large numbers of interacting effects, incompleteness of data, longitudinal repeated measures and methods for assessing genetic associations. Novel approaches to the issue of adjustment of significance levels to incorporate huge numbers of related comparisons via permutation tests have been incorporated into Epipop and the Centre has developed new approaches to the flexible modelling of trends in measures such as percentage leg fat, immunological and virological profiles via linear and non-linear mixed models. This work and its application to assessment of the impact of host-viral and other prognostic factors on post-HAART viral control form the basis of an ongoing NHMRC funded project. The group has experience and expertise in all aspects of medical statistics and epidemiology, particularly in censored data survival analyses, longitudinal mixed-effects methods, case-control matching methods and frailty models and will continue its fundamental methodological developments in these and other areas of application. Existing collaborations with local and international biostatistics and bioinformatics researchers will continue and be strengthened.

1. Validate the properties of estimates and inferences associated with the use of piecewise-linear mixed models for the analysis of virologic response profiles;
2. Compare the efficacy and appropriateness of the piecewise-linear mixed model with models specifically designed for correlated binary response data
3. Investigate the use of data-adaptive choice of change-points in the piecewise linear model rather than pre-specified times;
4. Compare the efficacy and applicability of survival analyses for viral response data based on mixed models with those based on the Kaplan-Meier or Cox methods; and
5. Utilise the developed methods to comprehensively evaluate prognostic factors associated with virologic control following initiation of HAART in patients from the Western Australian HIV Cohort.