PIs and Adipogenesis

1
IS OLDER AGE A RISK FACTOR FOR ANTIRETROVIRAL-TOXI
CITIES?
CA SABIN, CJ SMITH, T HILL, AN PHILLIPS FOR THE
UK COLLABORATIVE HIV COHORT (CHIC) STUDY, ROYAL
FREE CENTRE FOR HIV MEDICINE, LONDON UK
Correspondence Professor Caroline
Sabin, Department of Primary Care and Population
Sciences, Royal Free UC Medical School, Rowland
Hill Street, London, UK, NW3 2PF. Tel 44
207830 2239 ext 34752. Fax [phone].
E-mail c.sabin_at_pcps.ucl.ac.uk
BACKGROUND
RESULTS (cont.)

• Unadjusted for other factors, older individuals
  had an increased hazard of TRTD (HR 1.23, 95 CI
  1.05-1.45, p0.01). Whilst the hazard of TRTD
  was similar in older and younger individuals in
  the first year after starting HAART, hazards
  diverged thereafter (first year 1.09
  0.83-1.42, p0.55 year two onwards 1.33
  1.09-1.62, p0.005)
• After adjustment for potential confounders, both
  relationships became non-significant, although
  the effect remained stronger from the second year
  onwards (Table 2)

• Whilst most antiretroviral drugs are associated
  with the development of toxicities, the
  relationship between older age and the risk of
  toxicities has not been described
• Such a relationship would be biologically
  plausible for at least two reasons
• 1) As an individual ages, s/he may develop other
  conditions typical of aging these conditions
  and/or the medications used to treat them may
  interact with antiretroviral drugs, thus placing
  these individuals at higher risk of toxicity
• 2) As older individuals are thought to have a
  greater adherence to therapy, the incidence of
  toxicities may be higher because exposure to drug
  is greater
• We aimed to assess the relationship between older
  age and a) the time to toxicity-related treatment
  discontinuation (TRTD) and b) the incidence of
  laboratory-defined toxicities, among a large
  cohort of antiretroviral-naïve patients starting
  HAART

TABLE 2 RESULTS FROM MULTIVARIABLE COX
PROPORTIONAL HAZARDS REGRESSION MODEL OF FACTORS
ASSOCIATED WITH TRTD
METHODS

• All antiretroviral-naïve patients starting HAART
  in the UK CHIC Study were included for the
  purposes of these analyses, HAART was defined as
  any treatment combination that included at least
  one PI, NNRTI, abacavir, tenofovir, or
  enfuvirtide
• Older age was defined as gt50 years although age
  was also considered as a continuous covariate in
  some analyses

Statistical methods i) Analyses of TRTD
Time to first discontinuation of the third drug
in the regimen was calculated. Discontinuations
were defined as toxicity-related if they occurred
whilst the viral load was lt50 copies/ml.
Patients included in this analysis therefore had
to have viral loads measured both pre-HAART and,
if they discontinued treatment, post-HAART but
pre-discontinuation. Time to TRTD was described
using Kaplan-Meier and Cox proportional hazards
regression methods. Treatment discontinuation
with viral failure was treated as a competing
risk, thus follow-up on all patients who did not
experience a TRTD was censored on the date of the
patients last available viral load measurement.
Other potential confounders included sex, risk
group, ethnicity, calendar year, type of HAART
regimen and the CD4 count and HIV viral load at
the time of starting HAART

• Older individuals had significantly higher
  pre-HAART TG (p0.0001), TC (p0.0001) and ALT
  (p0.001) and lower Hb (p0.0002) than younger
  individuals
• In univariable analyses (Figure 1), abnormal TG,
  TC and Hb were all more common in those aged gt50
  years than in younger individuals. After
  adjusting for patient demographics, the
  relationship between age and abnormal TG levels
  became non-significant (Table 3). Further
  adjustment for pre-HAART TG levels removed the
  relationship between older age and abnormal TG
  levels. Only abnormal Hb levels remained more
  common in older individuals than younger
  individuals after adjusting for patient
  demographics and pre-HAART levels

ii) Analyses of laboratory-defined toxicities
The incidence of laboratory-defined toxicities in
the first year after starting HAART was
calculated, ignoring treatment discontinuations/sw
itches. Abnormal laboratory markers were defined
as follows Triglycerides (TG) gt2.3 mmol/L total
cholesterol (TC) gt6.2 mmol/L haemoglobin (Hb)
lt11.5 g/l bilirubin gt17 mmol/L ALT gt40 IU/L.
The incidence of abnormal laboratory markers in
older and younger individuals was compared using
univariable and multivariable logistic regression
analyses, after controlling for the potential
confounders above. Eligible patients had to have
at least one post-HAART measurement of each
marker although those with missing pre-HAART
measurements were not excluded, patients were
excluded if they were known to have an abnormal
pre-HAART value
FIGURE 1 INCIDENCE OF ABNORMAL LABORATORY
MEASUREMENTS IN OLDER AND YOUNGER INDIVIDUALS IN
THE FIRST YEAR OF HAART
P-values TG 0.02 TC 0.0001 Hb 0.007 Bili
1.00 ALT 0.55 Any 0.03
RESULTS

• Overall, 7932 antiretroviral-naïve patients
  started HAART in the UK CHIC Study of these, 693
  (8.7) were aged gt50 years
• Older individuals were more likely to be male,
  homosexual, of white ethnicity, and had lower CD4
  counts and higher HIV RNA levels at the time of
  starting HAART (Table 1), but there were no
  differences in the type of initial HAART regimen,
  the specific drugs received nor the year of
  starting HAART between the two age groups
• Overall, 4794 patients with follow-up viral loads
  discontinued the third drug in their regimen
  of these, 1703 were classified as TRTD.
  Kaplan-Meier estimates of the proportion of
  patients with TRTD were 4.4, 9.7, 14.4 and
  17.9 by months 6, 12, 18 and 24 respectively in
  younger individuals, and 4.4, 10.7, 17.7 and
  22.0 among older individuals (p0.01, log-rank
  test).

TABLE 3 ODDS RATIOS FOR ABNORMAL LABORATORY
MEASUREMENTS ASSOCIATED WITH OLDER AGE,
UNADJUSTED AND ADJUSTED FOR BASELINE DEMOGRAPHICS
AND PRE-TREATMENT VALUES
TABLE 1 CHARACTERISTICS OF OLDER AND YOUNGER
INDIVIDUALS STARTING HAART IN THE UK CHIC STUDY
CONCLUSIONS

• Whilst older individuals do appear to have a
  higher rate of TRTD and a higher incidence of
  laboratory-defined toxicities in the first year
  of HAART, these findings are largely explained by
  the demographic profiles and pre-existing
  age-related changes in these individuals prior to
  starting HAART
• Our definition of TRTD will over-estimate the
  incidence of this phenomenon, as it does not
  exclude the possibility that patients may have
  discontinued their drugs for other reasons
  although UK CHIC does collect information on
  reasons for treatment discontinuation, these data
  may be incomplete, particularly in the early
  HAART era

United Kingdom Collaborative HIV Cohort (UK
CHIC) Medical Research Council Clinical Trials
Unit (MRC CTU), London Abdel Babiker, David
Dunn, Esther Fearnhill, Kholoud Porter HIV
Epidemiology and Biostatistics Unit, Department
of Primary Care Population Sciences, Royal Free
and University College Medical School (RFUCMS)
Caroline Sabin, Teresa Hill, Loveleen Bansi,
Andrew Phillips Kings College Hospital, London
Philippa Easterbrook, Stephen Duffell, Eghosa
Bazuaye, Emma Macfarlane Brighton and Sussex
University Hospitals NHS Trust Martin Fisher,
Duncan Churchill, Wendy Harris, Nicky Perry,
Anthony Pullin Chelsea and Westminster NHS
Trust, London Brian Gazzard, Steve Bulbeck,
Jemima Clarke, Sundhiya Mandalia Mortimer Market
Centre, RFUCMS Richard Gilson, Julie Dodds, Andy
Rider, Ian Williams Health Protection Agency
-Communicable Disease Surveillance Centre
(HPA-CDSC), London Valerie Delpech Royal Free
NHS Trust Margaret Johnson, Clinton Chaloner,
Helen Gumley, Fiona Lampe, Dewi Ismajani
Puradiredja, Colette Smith, Mike Youle St.
Marys Hospital, London John Walsh, Christian
Kemble, Jonathan Weber Barts and the London NHS
Trust, London Chloe Orkin, Kevin Jones, Rachel
Thomas, James Hand Homerton Hospital, London
Jane Anderson, Selina Gann, Kevin Jones
Edinburgh Clifford Leen, Alan Wilson North
Middlesex Hospital Achim Schwenk, Jonathan
Ainsworth.