Treatment/cure

Question 1

disease burden of HIV

Answer 1

1.7 million new infections, 700,000 HIV-related deaths in 2019

Question 2

HAART

Answer 2

entry inhibitors, RTIs, protease inhibitors, integrate inhibitors

Question 3

nucleoside inhibitor

Answer 3

zidovudine

Question 4

non-nucleotide inhibitor

Answer 4

nevirapine

Question 5

protease inhibitors

Answer 5

saquinavir

Question 6

integrase inhibitors

Answer 6

raltegravir

Question 7

CCR5 antagonists

Answer 7

maraviroc

Question 8

fusion inhibitor

Answer 8

enfuvirtide

Question 9

increasing efficiency of fusion inhibitors

Answer 9

Kaminskas (2019) = used multi-enfuvirtide PEGylated complex increased in thoracic lymph duct-cannulated rats
enhances plasma and lymph exposure to enfuvirtide –> 2 and 6 fold increase in anti-viral activity respectively

Question 10

PreP efficacy

Answer 10

Fonner et al (2016) = meta-analysis, 51% reduction in HIV in prep cohorts

Liu et al (2015) = no HIV infections in individuals taking oral prep (527 individuals (MSM and transgender women) for 48 weeks taking oral prep 4+ times a week)

Question 11

what is the viral reservoir?

Answer 11

• Latent reservoir consists of 105 to 106 cells harbouring integrated HIV genome = most are CD4+ memory T cells
• Reservoir cells do not express viral antigens when cell is quiescent = transcriptionally silent = invisible to cytotoxic T lymphocytes
• Replication cycle of viruses resumes when T cells are activated (e.g. by antigens, cytokines…)  upregulate transcription factors (e.g. NF-kB) that facilitate viral gene transcription = produce replication-competent virus
• Cells are source of rebound viremia and drop in CD4+ T cell counts when ART is stopped
• Major barrier to HIV cure

Question 12

evidence that ART does not eliminate viral reservoir

Answer 12

Davey et al (1999)
discontinued in ART in 18 patients who had responded well (CD4 > 350 w undetectable viral load)
all patients experienced rapid relapse = plasma RNA exceeded 50 copies/ml within 2-3 weeks, CD4+ T cell counts also fell

Question 13

sterilising cure vs inducing remission

Answer 13

SC = aviremia + no detectable HIV-infectd cells
Remission = clinically undetectable viremia, no disease progression, stable CD4+ T cell count
but how long do individuals have to stay in remission before considered cured?

Question 14

Berlin patient

Answer 14

Berlin patient (2008) = Hutter et al (2009)
• Patient with HIV-1 infection and acute myeloid leukaemia received haematopoietic stem cell transplant from a donor who was homozygous for the CCR5-delta32 mutation  conveys immunity to HIV
• Achieved prolonged post-treatment control of HIV viremia even after cessation of antiretroviral therapy (ART)
• Three months after transplant, HIV viral load fell to undetectable levels, CD4+ T cell count rose, no detection of HIV in blood and tissue samples
• Died of leukaemia in 2020
• Proof-of-concept for HIV cure = likely that CCR5 mutation, pre-treatment chemotherapy and graft vs reservoir effect all contributed to treatment efficacy

Question 15

London patient

Answer 15

Similar efficacy achieved in London patient (Gupta et al, 2019) = still undetectable viremia at 30 months

Question 16

limitations with SCT cure

Answer 16

Limitations with SCT cure
• Not scalable = very difficult to find tissue matches
• High risk with increased mortality
- E.g. GVHD = donor T cells carried over into the recipient attack host tissues, mainly skin, GI tract, liver  affects 30-50% of patients even with HLA-matched donor
- E.g. risk of infection due to immunosuppression induced by chemotherapy, radiotherapy, and immunosuppressant drugs
• Will fail if patient’s reservoir contains CXCR4-tropic viruses = e.g. Essen patient = stopped ART a week before transplant, viruses evolved to be CXCR4-tropic, then experienced rapid rebound following transplant
• Last resort for HIV+ patients with haematological malignant disease

Question 17

Boston patients

Answer 17

Boston patients (2014)
• Softer HSCT approach to investigate where GVRE made significant contribution to cure
• 2 individuals with haematological malignancies  allogeneic HSCT from WT CCR5 donors
• Patients were subsequently assessed for evidence of on-going HIV infection 🡪 biopsies were taken from the rectum, gut, and lymph nodes and leukapheresis blood samples were taken but no vDNA could be detected,
• For both, residual post-HSCT host cells were <0.00096%
• No HIV-1 specific immune responses could be detected by ELISPOT (suggesting a lack of antigen present)

🡪 Decision was taken to stop therapy …
• Viral load was tested every week,
• DNA was tested for bi-weekly,
o Other tests also carried out 🡪 in short, intensive testing of the patients to ensure nothing detrimental is allowed to happen,
• Both patients initially seemed to do well off therapy – no viral rebound was seen,
• HOWEVER…
o Viral rebound later occurred in both patients 🡪 massive increase in viral load 🡪 both patients symptomatic,

🡪 Although the treatment was ultimately a failure, what it did tell us is that even when we think we are measuring the viral reservoir, we may not be doing so accurately…
• The assays carried out suggested that there was no vDNA in these patients, indicating their reservoirs were empty, yet the virus was ultimately able to rebound with vigour…

Question 18

using CD32a as reservoir marker

Answer 18

Descours et al (2017) = could CD32a act as marker for selective elimination of VR?
• Infected quiescent CD4+ T cells from HIV-negative controls with GFP-expressing HIV
- FACS revealed CD32a expression was selectively induced in T cells following HIV infection
- 90% of CD32a+ CD4+ T cells stained positively for GFP
• FACS also revealed that productive infection of stimulated CD4 T cells was not associated with significant induction of CD32a = CD32a is specific marker of HIV-infected resting CD4 T cells in vitro

• Then isolated CD4 T lymphocytes from ART-treated HIV-infected individuals
• Sorted into subsets with negative, intermediate, and high CD32a expression using FACS
• Found that CD32high cells were greatly enriched for HIV DNA (1,024-fold compared to CD32a- cells) and replication-competent proviruses (up to 3000-fold enrichment compared to total CD4 T cells), as measured using quantitative viral outgrowth assay

Question 19

problems with CD32a

Answer 19

• Reliability of marker has since been questioned as other groups have produced less compelling results = shown no enrichment for DNA/proviruses in CD32a cells
• Possible explanations for Descours results:
- Presence of CD32 on T cells may be due to presence of B cell-T cell doublets

Thornhill et al (2019)
• stained PBMCs and GALT from healthy donors, and tonsil from HIV-infected individual on ART, for CD3, CD4, CD19 and CD32 and analysed using imaging flow cytometry
• revealed that almost all CD3+ CD4+ CD32high cells were B cell-T cell doublets = saw co-localisation of CD3 and CD4 on same cell forming doublet with cell expressing CD19 and CD32
• CD32 only seen on B cells, no evidence of expression on CD4+ T cells
• T cells may still express CD32a RNA intracellularly?

Question 20

measuring viral reservoir

Answer 20

Essentially nobody is sure …
• Eriksson et al. 2013:
o 8 different assays proposed to measure the viral reservoir…
o Examples:
 Viral outgrowth assay 🡪 Leukapheresis of patient 🡪 take blood sample 🡪 culture cells 🡪 measure proportion of cells able to produce replicating virus (V. EXPENSIVE 🡪 5/6K dollars per assay)
 Integrated HIV-1 DNA assay 🡪 simple QPCR assay 🡪 give an idea of all DNA present within infected CD4+ T cells (including junk)
• Problem = massive inconsistency:
 Taking 1 million CD4+ T cells, the viral outgrowth assay suggests the size of the reservoir is on average 1/1million cells,
 In contrast, the QPCR assays suggest the number is 1000/1million
 3log difference in sensitivity of assays currently being used to measure the size of the viral reservoir…

🡪Robert Silicano’s lab 2013 – suggested that 95% of what most people measure as being part of the reservoir is in fact junk,
• Cells were pulled out that were HIV DNA+ but were not producing virus,
• Most had nonsense mutation, large internal deletions or hypermutation,
• However, ~11% of these cells (that wouldn’t produce virus even on exposure to highly activating factors such as CD3 & CD28) possessed HIV with normal, intact genomes,
o These viruses were reconstituted and found to be replication competent, perfectly fit viruses, yet for some reason they were not ‘waking up’ in the latently infected cells (unless repeatedly pulsed with activators such as CD3 and CD28),
• Significance: if latently infected cells are to be activated as a therapeutic strategy, this is a concern as many may not respond to activating stimuli, but yet retain the capacity to subsequently re-activate unwanted.

• True size is likely somewhere in between the estimates provided by contrasting assays 🡪 somewhere around ~60/million CD4+ T cells containing replication-competent virus,
• Problems:
o No assay currently provides a good estimate of the reservoir size
o Proportions of HIV-1 DNA and intact non-induced and viral outgrowth assay (VOA) results are patient-specific; there is no unifying formula 🡪 i.e. the size of reservoir may vary hugely between individuals.

Question 21

what is shock and kill?

Answer 21

One promising mechanism by which viral reservoirs in T cells could be eliminated is through the ‘shock and kill’ strategy = reactivate and destroy latent cells
• Latency reversal agents (LRAs) are used to induce viral replication in latently infected cells
• Triggers expression of viral gene products + presentation of antigen on MHC class I
• Infected cells are then rendered susceptible to cytotoxic killing by CTLs
• New cells are not infected by activated proviruses due to control by ART

Question 22

LRAs

Answer 22

1. Epigenetic modifiers
   - DNA and histone methyltransferase inhibitors, and histone deacetylase inhibitors (HDACi = e.g. valproic acid, vorinostat)
   - Reverse repressive epigenetic marks in HIV promoters  induce chromatin remodelling  activate gene expression
   - CLEAR trial = investigated efficacy of cyclic Panobinostat (1000 times more potent than vorinostat in the lab)  produced robust increases in RNA, so strong evidence that latent reservoir could be activated BUT still no change in markers for viral reservoir size (no change in integrated HIV-1 DNA)
2. Protein kinase C agonists
   - activate downstream NFkB signalling + HIV reactivation
3. TLR agonists (e.g. TLR2, TLR7)

Using checkpoint inhibitors to reverse latency
Uldrick et al (2022)
• Inhibitory checkpoint protein PD-1 upregulated on CD4+ T cells in HIV = limits T cell activation, facilitating reservoir persistence
• Study showed that blocking PD-1 could reverse latency
- 32 patients with well-controlled HIV treated with pembrolizumab w advanced cancer
- Drug induces increases in unspliced HIV RNA and HIV RNA:DNA ratio 1 week after administration, increase in frequency of cells with inducible virus = indicates that drug induced latent transcription
- BUT no change in spliced HIV RNA
- Patient population had cancer = immune system is abnormal, results may not be generalisable
- Increases seemed to be transient = decreased by day 8

Checkpoint inhibitors associated with immune-related adverse effects
- Low tolerance for severe side effects in individuals on ART = high quality of life, good life expectancy

Question 23

problems with LRAs

Answer 23

• LRAs so far ineffective in humans  promote increases in cell-associated and plasma HIV RNA but don’t reduce size of viral reservoir (as measured by levels of integrated HIV DNA in CD4+ T cells)
• Problems with LRAs
- Low potency = reactivate only a small fraction of reservoir cells
- Associated with significant off-target effects (e.g. epigenetic modifiers induce global genomic changes, NF-kB inducers stimulate systemic induction of pro-inflammatory cytokines and global immune activation)
- Many LRAs impair CTL responses (e.g. HDACi reduce MHC-1 expression, impair CD8+ T cell proliferation and killing)
- Difficulties targeting CNS reservoir = restricted LRA penetration and altered immune surveillance may limit shock and kill respectively
- Patients on HAART are usually well = is it ethical to trial effects of drugs?

Question 24

evidence that shock is not enough

Answer 24

Kill mechanisms
Shan et al (2012) = early evidence that ‘shock’ alone is insufficient to reduce reservoir size
• Treated infected resting CD4+ T cells (either isolated from patients on HAART, or generated in vitro by infecting primary CD4+ T cells with replication-deficient GFP+ HIV reporter virus) with the LRA HDACi SAHA
• SAHA treatment alone did not reduce frequency of latently infected cells as measured by standard limiting dilution virus culture assay – does not reliably lead to CD8 T cell induced killing
• Then treated latently infected cells with SAHA and cocultured with autologous CD8+ T cells from elite controllers
- SAHA reactivated latent HIV and induced expression of HIV antigens
- CD8+ T cells only able to kill latently infected cells after virus reactivation at high effector: target ratios after long periods of coculture
- Prestimulation of CD8+ T cells with Gag and IL-2 drastically improved cytotoxic activity
• Study indicates that boosting CTL responses is required for efficient eradication of HIV using LRAs

Question 25

how to kill?

Answer 25

Enhancing immune responses 
•	CD8+ HIV vaccines = e.g. RIVER trial 
•	Broadly neutralising antibodies (see below) 
•	Chimeric antigen (CAR) T cells 
•	Immune stimulators
-	E.g. TRL agonists 
-	Immune checkpoint inhibitors 
-	Vaccine adjuvants 

Tateishi et al (2017) = ‘lock in and apoptosis’
• Investigated efficacy of Gag inhibitor called L-HIPPO = binds with high affinity to the viral protein Pr55Gag, required for viral budding
• L-HIPPO was shown to greatly reduce translocation of Pr55Gag to cell surface membrane, and suppress viral protein release, in HIV-infected HeLa cells
• Also found that L-HIPPO was potent inducer of apoptosis
- Transfected HeLa cells with HIV strain  incubated for 10 hours  treated with L-HIPPO  incubated for 12 hours  stained with annexin V (positive staining indicative of apoptosis)  analysed by FACS
- Observed very strong staining of cells for annexin V when treated with L-HIPPO (up to 81%, compared to 5-6% for controls)

Other pro-apoptotic drugs = BCL-2 antagonists, Akt inhibitors

Question 26

RIVER trial

Answer 26

RIVER study (2020)
• Phase 2 clinical trial comparing ART, HDACi vorinostat (shock) and T cell-inducing vaccine encoding conserved HIV sequences (kill), with ART alone
• No statistically significant difference in mean total HIV DNA, mean integrated HIV DNA, viral outgrowth, HIV RNA between 2 groups = ‘shock and kill’ regime did not alter size of viral reservoir
• Despite evidence that vaccine induced HIV-specific CD4+ and CD8+ T cells and vorinostat inhibited histone deacetylation (BUT did not induce viral transcription)
• Highlights that shock and kill therapy is still just a theory = need to refine agents
- Vorinostat is a weaker LRA = trial may have proved more successful with a different drug

Question 27

block and lock

Answer 27

Block and lock
• Opposite rationale to shock and kill
• Aim to induce deep latency in reservoir cells, through permanent epigenetic silencing = resist reactivation after ART discontinued

Méndez et al (2018) = explored efficacy of 2 si/shRNAs, named PromA and 143
• Induce transcriptional gene silencing (TSG) through epigenetic repression of TF binding sites in HIV-1 LTR  suppress HIV replication
• Effective at inhibiting HIV replication when stably expressed in Jurkat T cell line (robustly suppressed RT activity (measured using RT-assays) and levels of gag mRNA (measured by qRT-PCR) and HIV-1 integrated DNA (measured by qPCR))
• Provided protection from HIV reactivation when T cells were challenged with latency reversal agents = TNF, SAHA (HDACi), byrostatin (PKC agonist)
• Impaired expression of proviral genes following challenge was due to repressive epigenetic marks induced by shRNAs (e.g. H3K27me3) = observed using CHIP
• Problems
- other data show that latency is only maintained for 3 weeks after treatment cessation  therapy needs to be developed for a more sustained response
- difficulties in delivery = how do we ensure shRNAs are targeted to wide range of cell types and anatomical locations that form reservoir?  lack robust markers to identify cells harbouring proviruses

Question 28

bnabs

Answer 28

• Might be useful in suppressing viral rebound from latent reservoir following ART cessation
• Advantages over T cell products = will work in all individuals regardless of HLA-matching, no need for genetic engineering
• Disadvantages = short half-life in vivo so frequent dosing required

Question 29

human trial of bnabs

Answer 29

• Scheid et al (2016) = bnAbs exert selective pressure on latent HIV

• Phase II clinical trial enrolling 13 patients with chronic HIV infection suppressed for at least 12 months  treated with infusions of bnAb against CD4 binding site of Env (3BNC117) following ART interruption
• Delayed rebound in viremia = up to 19 weeks compared to 2.6 weeks for historical controls
• 30% of participants remained suppressed until antibody concentrations dropped below 20ug/ml
• In many individuals, rebound viruses arose from single provirus, often showing resistance to 3BNC117 = indicative of viral escape

Question 30

macaque bnab trial

Answer 30

Study by Nishimura et al.2017 found the administration of 2 bNAbs to macaques during acute SHIV infection was able to reduce plasma viral loads to undetectable in 6 of the 13 macaques whilst another 4 retained normal CD4 count and very low viral load for over 2 years. Contrast this to those treated with ART for 15 weeks sustained rebound plasma viraemia. Shown to be CD8 T cell immunity in bNAb group by administration of anti-CD8α mAb causing rebound of plasma viraemia.
SHIV differs from HIV

Question 31

Sterilising cure without BMSCT?

Answer 31

Turk et al (2021)
• Published report of 30-year-old ‘Esperanza patient’ infected with HIV, who had achieved viral suppression without ART
- No genome-intact HIV-1 proviruses were detected in analysis using near-full-length individual proviral sequencing, of 1.188 billion PBMCs and 503 million mononuclear cells from placental tissue from patient
- 7 defected HIV-1 proviral sequences were detected  1 with APOBEC3G/3F hypermutations  evidence of past infection and active cycles of viral replication
- Also detected HIV-1 specific memory CD4+ and CD8+ T cells using immunologic assays
- No retrieval of replication-competent HIV-1 from 150 million resting CD4+ T cells in viral outgrowth assay
- No HIV-1 RNA detected in 4.5mL of plasma

• This patient is more than an ‘elite controller’
- these patients display undetectable plasma viremia
- but genome-intact DNA and replication-competent viruses can be isolated by in vitro lab assays
- still harbour persistent viral reservoirs
• Resembles phenotype of Berlin patient
• Limitations  impossible to empirically prove sterilising cure
- did isolate a huge number of cells, performed multiple virological assays
• Hope of finding a cure!

Question 32

eliminating integrated HIV DNA

Answer 32

Eliminate integrated proviral DNA using CRISPR/Cas9
• Both ends of viral genome contain long terminal repeats (LTRs)  could use gRNAs to target both LTRs and excise full integrated provirus
• Hu et al (2014) = expression of Cas9 and gRNAs facilitated efficient removal of integrated HIV genome from latent microglial and promonocytic cells
• Problems = need to optimise delivery of Cas9 constructs to all reservoir cells = currently efficiency is insufficient to remove all integrated provirus from all reservoirs
• Potential immunogenicity of Cas9 when stably expressed = bacterial in origin

Question 33

targeting CCR5 and CXCR4

Answer 33

Target CCR5 and CXCR4 = render patient’s T cells resistant to viral entry
• Need to edit both coreceptors, not just R5 (used by most transmitted founder viruses)
• Viruses can evolve to utilise alternate coreceptor in response to selection pressures = occurs in 50% of individuals with chronic infection

Didigu et al (2013)
• Used zinc-finger nucleases to genetically modify CCR5 and CXCR4 in primary human CD4+ T cells  induced dose-dependent reduction in surface expression of both coreceptors
• Cells proliferated normally and were highly resistant to infection by both R5 and X4-using HIV in vitro
• ZFN disruption also provided protection against HIV in humanised mouse model
- NSG immunodeficient mice  infused with WT CD4 T cells, or CD4 T cells treated with R5-ZFN or both ZFNs  then infused with CD4 T cells infected with R5 and X4 viruses
- ZFNs facilitated better control of CD4+ T cell counts = CD4 counts in mice from R5/X4-ZFN group were 200-fold higher than in other groups 55 days post infection
• Translation to humans?
• Problems with ZFNs = prone to off-target cleavage so extensive screening required to ensure optimal editing of target genes without toxicity = expensive