Balancing HAART and hepatic safety

1
Balancing HAART and hepatic safety
Paul Sax, MD
Brigham and Womens Hospital Harvard Medical
School Boston, MA
2
Liver toxicity is the most common severe adverse
event in the HAART era
Incidence/ 100 persons
Coinfection with hepatitis B or C associated with
4.15 (95 CI 2.267.60) hazard ratio of Grade 4
liver toxicity
Reisler RB, et al. J Acquir Immune Defic Syndr
20033437986
3
NRTI-related hepatotoxicity

• Syndrome of mitochondrial toxicity
• Mitochondria have their own DNA (mtDNA) that
  encodes 13 of mitochondrial proteins
• mtDNA is replicated by DNA polymerase ?
• Differs from nuclear DNA
• NRTIs inhibit DNA polymerase ?
• Mitochondrial dysfunction
• Cellular toxicity
• High-risk drugs
• Stavudine, didanosine, zalcitabine
• Low-risk drugs
• Abacavir, zidovudine, lamivudine, tenofovir

4
NRTIs Inhibition of DNA polymerase ?
100
13
11
9
3.6
10
1
Relative Potency of Inhibition
0.1
0.04
0.02
0.01
0.01
0.001
FIAU
Abacavir
Stavidine
Zalcitabine
Zidovudine
Lamivudine
Didanosine
Martin JL, et al. Antimicrob Agents Chemother
1994382743-9
5
Hepatic steatosis associatedwith D4T exposure
Sulkowski MS, et al. 11th CROI 2004 Abstract 72
6
ddI increases risk of hepatic decompensation
during IFN/RBV treatment
Mauss S, et al. AIDS 200418F215
7
Risk of mitochondrial toxicity NRTI RBV in
HIV/HCV-coinfected patients
Mitochondrial Toxicity Risk

• US FDA Reporting System of HIV/HCV patients
  treated with IFN/RBV
• 31 cases (58 adverse events) suggestive of
  mitochondrial toxicity
• Pancreatitis and/or increased lipase (n21)
• Lactic acidosis (n20)
• Elevated LFTs (n8)
• Hepatic steatosis (n6)
• Elevated creatinine, neuropathy, multiorgan
  failure (n1 each)

RBV
12.4
Didanosine Didanosine stavudine Stavudine Aba
cavir Lamivudine Zidovudine
8.0
3.3
1.1
0.2
0.06
0.01 0.1 1.0 10
100
Odds Ratio (95 CI)
Fleischer R, et al. Clin Infect Dis
200438e79-e80
8
NRTI-based liver toxicityClinical presentation

• Non-specific symptoms
• Abdominal pain, vomiting, anorexia, pain (right
  upper quadrant), weight loss
• Hepatomegaly
• Mixed cholestatic/hepatocellular pattern of liver
  enzymes
• Evidence of extrahepatic mitochondrial toxicity
• Amylase/lipase, CPK, lactate, metabolic acidosis,
  loss of bicarbonate

Falco et al. Clin Infect Dis 2002
9
NNRTI-associated drug-induced liver injury
NVP n256
EFV n312
Severe hepatotoxicity, based on ?ALT/AST Gr 3
(gt5 x ULN) or Gr 4 (gt10 x ULN)
Patients on NVP were more likely to have chronic
viral hepatitis (HCV or HBV) than patients on
EFV (P 0.05). Additionally, NVP was
co-administered with PIs more frequently than EFV
(Plt0.0001).
Sulkowski MS, et al. Hepatology 200235182-9
10
NNRTI-associated hepatotoxicity
Incidence of Grade 3 or 4 change in ALT or AST
levels
Nevirapine
Efavirenz
Incidence/ 100 persons
No PI
No PI
PI
PI
patients with Gr 3/4 ALT or AST NVP 15.6 EFV
8RR 1.9 (CI1.23.1)
Sulkowski MS, et al. Hepatology 200235182-9
11
HCV and NNRTI-associated hepatotoxicity
Sulkowski MS, et al. Hepatology 200235182-9
12
NVP hepatotoxicity Time to onset ofALT or AST
gt5 x ULN
NVP (n1731), placebo (PBO, n1912)
Probability ()
Stern JO, et al. J Acquir Immun Defic Syndr
200334S21-S33
13
Risk factors for NVP-associated symptomatic
hepatic events

• Rash-associated hepatic events
• NVP (RR 11.2 P lt 0.01)
• Female gender (RR3.2 P lt 0.01)
• Female with baseline CD4 ?250 cells/mm3 (RR
  9.8P lt 0.01)
• Male with baseline CD4?400 cells/mm3 (RR 6.4P
  lt 0.01)
• Too few events in HCV/HBV coinfected to assess

• Other symptomatic hepatic events
• Baseline ALT or AST gt2.5 times ULN (RR3.2 P lt
  0.01)
• HBV coinfection (RR3.9 P lt 0.01)
• Baseline CD4 count found to be an inconsistent
  factor

Stern JO, et al. J Acquir Immun Defic Syndr
200334S21-S33
14
Risk factors for severe, life-threatening
hepatotoxicity with nevirapine

• Women with CD4 counts gt250 cells/mm3 HR 12 men
  with CD4 counts gt400 cells/mm3 HR 4
• Greatest risk of severe and potentially fatal
  hepatic events often associated with rash
• Occurs in first 6 weeks of nevirapine therapy
• Close monitoring recommended in some cases,
  hepatic injury progresses despite discontinuation
  of treatment
• Nevirapine should not be used in women withCD4
  gt250 cells/mm3 or men with CD4 counts gt400
  cells/mm3 when other options exist

Viramune (nevirapine) prescribing information.
Boehringer Ingelheim, 2004
15
Nevirapine hypersensitivity
26 cases of NVP reactions in 241 patients
treated (Australia 85 Caucasian)

• All NVP reactions associated with HLA-DRB101
  (P0.014)
• Strong interaction between HLA-DRB101 and ?
  CD4 cells and NVP reactions (p0.00006) in
  logistic regression, but neither significant
  alone
• Rash alone not associated with HLA-DRB101 or
  CD4 cells
• Conclusion Interaction between genetic
  disposition and a immune function that
  predisposes to liver toxicity with rash and/or
  fever in patients treated with NVP

Martin A, et al. XV IAC, 2004, LbOrB13
16
Severe hepatotoxicity in patients receiving
NNRTI-based regimens

• Treatment-naïve patients
• 1216 enrolled
• HBV coinfected 5.2
• HCV coinfected 9.5
• NNRTI d4T 3TC
• Nevirapine qd (400 mg)
• Nevirapine bid (200 mg)
• Efavirenz qd (600 mg)
• Nevirapine efavirenz qd (400 600 mg)

Grade 3/4 Hepatotoxicity
Patients ()
Nevirapine
van Leth F, et al. Lancet 20043631253-63
17
Study 006 Patients withHepatitis B and/or C
coinfection
ALT and AST gt5 x ULN

• Seropositive for hepatitis B and C at study entry
• 137 patients treated with efavirenz-containing
  regimens
• 84 patients treated with indinavir 2 NRTIs
• Discontinuations due to liver or biliary system
  disorders were similar (2 3)

Patients ()
Sustiva (efavirenz) prescribing information
(US). August 2004.
18
PI-related hepatotoxicity

• Incidence of ALT and/or AST elevations in
  registration trials
• Grade 3 (gt5 x ULN) and 4 (gt10 x ULN) 19
• Full-dose ritonavir
• Only PI identified as an independent risk factor
  for severe hepatic injury
• Boosting doses of ritonavir (200 mg/day)
• Not associated with significantly higher
  incidence of severe hepatotoxicity vs other PIs
• Limited data on newer PIs
• Atazanavir or fosamprenavir

Sulkowski MS, et al. Clin Infect Dis
200438S90-S97
19
PI-associated hepatotoxicityJohns Hopkins Cohort

• Prospective cohort 298 patients with new ARV
  drug regimens and serial liver enzymes
• 71 PI-containing regimen
• RTV SQV (n50)
• IDV (n117 )
• NFV (n51)
• Chronic HCV 52 HBV 2.7
• HCV higher baseline ALT and AST levels

• Overall incidence of severe hepatotoxicity 10.4
  (95 CI 7.214.4)
• Hepatotoxicity associated with ritonavir use 30
  (95 CI 17.944.6)

Sulkowski M, et al. JAMA. 200028374-80
20
Incidence of hepatotoxicityby ART regimen
Johns Hopkins Cohort
Ritonavir Saquinavir
Sulkowski M, et al. JAMA. 200028374-80
21
Incidence of severe hepatotoxicity of unboosted
PIs in HIV/HCV-coinfection

• Retrospective case review (n1094)
• On PI-based ART for a mean of 39 months
• Baseline
• CD4 240 cells/mm3
• HIV RNA 4.1 log10
• Similar HIV outcomes among unboosted PIs
• CD4 increase117 cells/mm3
• HIV RNA reduction-1.1 log10
• 58 of patients achieved HIV RNA lt400 copies/mL

Grade 3/4 Hepatotoxicity
Patients ()
AST and/or ALT gt5 x ULN
Dieterich D, et al. 43rd ICAAC, 2003 Abstract
H-831
22
Incidence and risk factors of HAART-associated
hepatotoxicity

• Retrospective cohort analysis (N560)
• Risk factors for Gr 4 liver enzyme elevations(HR
  95 CI)
• Female sex 2.8 1.35.8
• BL ALT levels (per 10 U/L increase) 1.05
  1.011.11
• Chronic HCV 5.0 2.310.7
• Chronic HBV 9.2 4.120.6
• Recent discont, of 3TC 6.8 2.122.7
• Recent start of NVP 9.6 3.2-28.3
• Recent start of RTV 4.9 2.012.1
• No increased risk if RTV dose 200 mg/day

12-week period after start of drug.
Wit FWNM, et al. J Infect Dis 200218623-31
23
Hepatotoxicity of PI-based ART

• Prospective evaluation of Gr 3/4 AST/ALT
  elevations
• Median follow-up for individual PIs 223-365 days

Baseline characteristics
1st PI-containing regimens
SQV/RTV
NFV
IDV/RTV
LPV/r
median
Sulkowski MS, et al. AIDS. 2004. In press
24
Hepatotoxicity of PI-based ART
Sulkowski MS, et al. AIDS. 2004. In press
25
Proportion without revere hepatotoxicity among
PI-based regimens ritonavir
Proportion without severe hepatotoxicity

• Prospective study of 1061 HIV-infected patients
• Hazard ratios for severe hepatotoxicity
  (multivariate analysis)
• HCV positive 1.82
• CD4 gt200 cells/mm3 0.52
• HIV RNA gt10k copies/mL 4.77
• Indinavir ritonavir use 2.97
• Saquinavir ritonavir use 2.41
• No increased risk associated with lopinavir/r or
  nelfinavir

Indinavir ritonavir (n94) Lopinavir/r
(n89) Saquinavir ritonavir (n273) Nelfinavir
(n605)
AST and/or ALT gt5 x ULN
Sulkowski MS, et al. AIDS. 2004. In press
26
60-week safety of lopinavir/r and nelfinavir by
hepatitis status Study 863

• Hepatitis B/C-positive patients
• Patients with ALT/AST gt3 x ULN were excluded at
  screening regardless of hepatitis status
• Lopinavir/r patients (n57) tended to have a
  lower incidence of Grade 3/4 AST and ALT
  elevations compared with nelfinavir-treated
  patients (n68)
• AST lopinavir/r 4 vs nelfinavir 13
• ALT lopinavir/r 12 vs nelfinavir 17
• Similar incidence of discontinuations due to
  hepatic adverse events in both groups (4)
• No discontinuations due to elevated liver enzymes
  in either group
• No hepatic-related events resulted in death

Bernstein B, et al. 1st IAC, 2001 Abstract 525
27
LPV/r vs NFV Mean ALT in HIV/HCV patients



ALT (IU/L)
plt0.05 plt0.01
Week
Sample Size LPV/r 29
24

21 NFV 41
37
35
Study 863
Sherman KE, et al. 11th CROI, 2004 Abstract 811
28
PK of LPV/r in HCV coinfected subjects

• Phase I, multiple-dose, open-label, two-center
  study
• Control and hepatically impaired (HI) groups to
  be similar in age, ideal body weight and gender
• Subjects were not allowed to take NNRTIs, other
  PIs or enzyme inducers

Trough plasma samples were collected on Study
Day 7 and on one day between Study Days 10 and
12, inclusive, and Study Day 14 Plasma samples
were collected predose and at 2, 4, 6, 8, 10 and
12 hours post dose for LPV and RTV and at predose
and 6 h for LPV protein binding breakfast did
not include grapefruit juice
Arribas JR, et al. 9th EACS, 2003 Abstract F2/6
29
Total lopinavir concentration-time profiles (mean
SD) at Study Day 14
For LPV trough values collected on Study Days
7-14 Study Day effect was not
statistically significant (p0.82)
Arribas JR, et al. 9th EACS, 2003 Abstract F2/6
30
BMS AI424 045 Patients with hepatitis B and/or C
coinfection
ALT and AST gt5 x ULN

• Limited data available on the use of atazanavir
  in patients coinfected with hepatitis
• Seropositive for hepatitis B and/or C at study
  entry
• 20 patients treated with atazanavir/ritonavir300
  mg/100 mg od
• 18 patients treated with lopinavir/ritonavir400
  mg/100 mg bid

Patients ()
Reyataz (atazanavir) full prescribing
information. July 2004.
31
Interaction between HIV and HCV
32
Does HCV coinfection influenceHIV disease
progression?

• Negative impact
• Swiss HIV Cohort Study1
• Italian Cohort Naïve Antiretrovirals Study2
• Little or no impact
• Klein and colleagues3
• French Aquitaine cohort4
• CHORUS and VACS 3 cohorts5
• Johns Hopkins cohort6

1Grueb G, et al. Lancet 20003561800-5 2De Luca
A, et al. Arch Intern Med 20021622125-32 3Klein
MB, et al. JAIDS 200333365-72 4Rancinan C, et
al. AIDS 2002161357-625Justice A, et al. 14th
IAC, 2002. Abstract MoOrB1058 6Sulkowski MS, et
al. JAMA 2002288199-206
33
Outcomes After ARTAccording to HCV Status
Swiss HIV Cohort1 Proportion Alive and No
Clinical Progression
Italian Cohort Naïve ARTs2 Cumulative
Proportion With No Clinical Progression
P0.2
Plt0.001
0 6 12 18 24 30
36 42
0 12 25 37
50
Time Since Start of ART (mo)
Time Since Start of ART (mo)
1Grueb G, et al. Lancet 20003561800-5 2De Luca
A, et al. Arch Intern Med 20021622125-32
34
Impact of HCV on Survival and Clinical Outcomes
in HIV Disease

• Johns Hopkins HIV cohort
• Urban setting
• 873 HCV (53 IDUs)
• 1076 HCV- (13 IDUs)
• HCV infection did not negatively impact
• Risk of death
• Risk of developing anAIDS-defining illness
• CD4 increase following effective ART

RH 1.05 (95 CI 0.85, 1.30)
RH 1.03 (95 CI 0.86, 1.23)
Months
Sulkowski MS, et al. JAMA 2002288199-2060
35
LPV/r vs NFV Mean CD4 change from baseline in
HIV/HCV subjects
234 184
CD4 (cells/mm3)
p0.247
Week
Sample Size LPV/r 29
23

20 NFV 41
34
35
Sherman KE, et al. 11th CROI, San Francisco, CA,
February 2004, Poster 811
Study 863
36
HIV/HCV-coinfected patientsTime to HIV RNA lt400
and lt50 copies/mL
All patients received lamivudine stavudine
Sherman KE, et al. 11th CROI, 2004 Abstract 811
37
Does treatment of HIV slow progression of liver
disease?
38
ART and hepatitis C progression

• Retrospective cohort, 19952000
• 182 pts (63 PIs and 119 no-PI)
• Median PI use 14 months (95 CI 1016)
• Liver biospy
• Estimate duration of HCV
• Progression to cirrhosis
• No PI RH 4.74 (1.3416.67)
• EtOH RH 4.71 (1.9211.57)
• CD4 lt200 cells/mm3 RH 2.74 (1.176.41)
• Age at HCV infection RH 2.37 (1.04 5.38)

Benhamou Y. Hepatology 200134283-7
39
Cirrhosis rate in PI-treated and untreated
HCV/HIV infected patients
Benhamou Y. Hepatology 200134283-7
40
Impact of ART on liver fibrosisin
HIV/HCV-Coinfected Patients

• Cross-sectional study
• 152 HIV patients
• HCV coinfected 90
• Alcohol gt5 drinks/day
• 2731
• Fibrosis progression rate
• Median 0.143(range 0.0590.250)
• Nevirapine gt1 year
• More likely to show advanced liver fibrosis
• Faster liver fibrosis progression
• PIs seem to be protective for liver fibrosis

Sample size too small to determine effect
Macias J, et al. AIDS 200418767-74
41
Impact of ART on overall liver mortalityin
HIV/HCV-coinfected patients

• Bonn cohort (19902002)
• 285 HIV/HCV coinfected patients
• Liver-related mortality rates per 100
  person-years
• HAART 0.45
• ART 0.69
• No therapy 1.70
• Predictors for liver-related mortality
• No HAART
• Low CD4 count
• Increasing age

Overall Mortality
Plt0.001
HAART
Cumulative Survival
ART
No therapy
0 1000 2000 3000 4000 5000 6000
Days
Liver-related Mortality
P0.018
HAART
ART
No therapy
Cumulative Survival
0 1000 2000 3000 4000 5000 6000
Days
Qurishi N, et al. Lancet 20033621708-13
42
Impact of NRTIs on HBVin HIV-coinfected patients
1Dore GJ, et al. J Infect Dis 1999180607-13
2Pillay D, et al. AIDS 2000141111-63Harris J,
et al. 11th CROI, 2004 Abstract 836 4Dore GJ,
et al. J Infect Dis 20041891185-92
43
Prevention and management ofART-related
hepatotoxicity

• Assess baseline characteristics of the patient
• Obtain pre-treatment aminotransferase levels
• Determine if any pre-existing liver disease
• In particular hepatitis coinfection
• Consider
• Concomitant medications and OTC products
• Alcohol consumption
• Any potential for drug abuse
• Vaccination if indicated against hepatitis A and
  B
• Treatment for hepatitis C or B

44
Selection of drugs

• Less hepatotoxic drugs should be preferred in
  high-risk patients
• NRTIs Lamivudine, abacavir, emtricitabine,
  tenofovir DF, and zidovudine
• PI Lopinavir/r, nelfinavir limited data on
  newer PIs
• NNRTI Efavirenz
• Risks include
• Alcohol
• Coinfection with hepatitis B and/or C
• Previous episode of hepatotoxicity
• Cirrhosis
• Pregnancy
• Obesity
• Female sex

45
Monitoring on ART

• Without pre-existing liver disease
• Check aminotransferases at 2, 4, and 12 weeks
  after initiating a new regimen
• Every 3 months thereafter if stable
• With pre-existing liver disease
• Check aminotransferases frequently after
  initiating a new regimen
• If stable, then every 3 months thereafter
• Check aminotransferases levels in any case of
  skin rash

46
Management ofelevated liver function tests

• Asymptomatic, minor elevations (lt5 x ULN)
• Generally safe to tolerate if hypersensitivity
  reaction has been excluded
• Discontinue ART if
• Clinical symptoms occur
• Serum lactate levels start to rise
• Evidence of severe liver dysfunction
• ALT gt5 x ULN, coagulopathy, encephalopathy
• Elevated LFTs should always prompt a search for
  other potential causes of hepatotoxicity

47
ART and the liver Conclusions

• Liver-related morbidity and mortality are
  assuming an increasing importance in HIV care
• Each class of antiretroviral agents may be
  associated with hepatic toxicity
• Viral hepatitis co-infection increases the risk
  of drug-induced liver injury all co-infected
  patients must be monitored carefully
• Data suggest that treatment of HIV slows
  progression of fibrosis and reduces liver-related
  mortality