Drug Drug Interactions Review (WHO)

1

SYSTEMATIC REVIEW:

Drug-drug Interactions between Antiretrovirals and medications used to treat
TB, Malaria, Hepatitis B&C and opioid dependence






SH Khoo
1
, S Gibbons
1
, K Seden
2
, DJ Back
1






1 Department of Pharmacology, University of Liverpool, 70 Pembroke Place, Liverpool L69
3GF, UK

2 NIHR Biomedical Research Centre in Microbial Diseases, Royal Liverpool University
Hospital, Prescott St, Liverpool L7 8XP, UK


+44 151 794 5560
+44 151 794 5656 (fax)


Correspondence to [email protected]


2

Table of Contents:

Page

1

Introduction 3
2

Methods 5
2.1 Search Strategy
2.2 Study Selection
2.3 Study Quality Assessment
2.3.1 Strength of recommendation
2.3.2 Grading of evidence
2.4 Data Synthesis

3

Results 8
3.1 Antimalarials
3.2 TB Drugs
3.3 Hepatitis B & C Treatment
3.4 Opioid dependence

4 Discussion

5 Acknowledgements and Declarations

6 References

7 Appendix A – Antimalarial Drugs


Appendix B – Antituberculous Drugs


Appendix 3 – Hepatitis B and C Treatment


Appendix 4 – Drugs used in Opioid Dependence









3


1 INTRODUCTION
Drug-drug interactions (DDIs) are an important and widely under-recognised source of medication
errors, which represent significant risk of harm to patients and opportunity cost for healthcare
systems. The co-administration of contraindicated drugs has been found to account for 5.2% of
209 hospital admissions in the USA in patients receiving antiretrovirals (ARVs) [Rastegar et al,
2006]. Although studies are limited, clinically significant DDIs involving ARVs are common,
affecting at least 14-41% of patients in the US, the Netherlands and UK [Shah et al, 2007; de Maat
et al, 2004; Cottle et al, 2009]. A substantial proportion of these have the potential for an adverse
impact on ARV exposure. Conversely, DDIs may result in increased exposure to ARVs or co-
administered drugs, precipitating drug toxicity or greater severity and incidence of adverse
reactions. Data from developing countries are sparse, though it is likely that clinically significant
DDIs are prevalent [Kigen et al, 2008].

DDIs may be pharmacokinetic or pharmacodynamic in nature. ARVs are among the most
therapeutically risky drugs for DDIs, due to potent inhibition or induction of liver enzymes such as
the cytochrome P450 isoenzymes (CYP450), which metabolise a broad array of other medications.
DDIs involving protease inhibitors (PI) and non-nucleoside reverse transcriptase inhibitors (NNRTI)
are more likely to be attributable to hepatic metabolic pathways than DDIs involving nucleoside or
nucleotide reverse transcriptase inhibitors (NRTI), which in some cases can be due to competition
for renal tubular secretion. DDIs are more prevalent in regimens containing PIs than NNRTIs [Miller
et al, 2007; Cottle et al, 2009].

Although all patients receiving ARVs are potentially at risk of DDIs, this risk is increased in certain
patient groups and clinical scenarios:

1.1 Use of New HIV Drugs
Assessment of the potential for DDIs during the clinical phase of drug development, although
comprehensively undertaken, is at best incomplete. Screening of a new molecular entity for
potential as a substrate, inducer or inhibitor of phase I and II metabolic enzymes and influx/efflux
drug transporters is limited by a lack of validated expression systems and standardised protocols,
particularly for drug transporters.

It cannot be assumed that drugs from the same class have broadly similar potential for interaction.
For example, the integrase inhibitors raltegravir and elvitegravir differ in metabolic pathways and
interaction potential with regard to cytochrome P450 enzymes. Further, raltegravir has significant
interactions with proton pump inhibitors as a result of physiochemical characteristics which lead to
pH-dependent solubility. In addition, there will always be surprises in the form of unanticipated
DDIs which emerge after licensing, and may lead to diminished therapeutic effect of ARVs, such as
lopinavir and rosuvastatin. This highlights the need for standard protocols for interaction screening
of new drugs, as well as clinical vigilance as experience in their use develops.

1.2 Co-infections, particularly in the developing world
Throughout the world, HIV overlaps with other epidemics such as tuberculosis (TB), malaria and
chronic viral hepatitis. TB is the leading cause of death among people living with HIV in Africa, and
globally 456,000 people died of HIV-associated TB in 2007. Difficulties in treating TB in HIV
patients may arise due to interactions with rifampicin which is a potent inducer of liver enzymes.
Several ARVs contraindicate the use of rifampicin and others may require dose modification. HIV
also has a considerable impact on malaria, affecting parasitaemia, disease severity (in areas of
unstable transmission) and mortality during pregnancy. Drug interactions are understudied, but
important interactions have already been identified between antiretrovirals and quinine,
amodiaquine and lumefantrine.

Worldwide, an estimated two billion people have been infected with the hepatitis B virus (HBV),
and more than 350 million have chronic (long-term) liver infections. An estimated 170 million
persons are chronically infected with Hepatitis C (HCV) with 3 to 4 million persons are newly
4
infected each year. Therapy for chronic HCV infection is set to dramatically change with the
proliferation of new drugs directed against HCV polymerase, protease and other targets, and
although data are sparse, interactions between HIV nucleoside reverse transcriptase inhibitors and
PIs, and new or existing HCV drugs have been identified.
The inflexibility of dosing when using fixed-dose combinations of ARVs makes many DDIs harder
to manage. ARV coverage in middle and low income countries has increased 45% between 2006

and December 2007 [UNAIDS, 2008], and with increasing coverage, it is likely that access to other
medication will also improve, for example various integrated programmes for ‘neglected’ tropical
diseases, which aim to combine mass drug administration for conditions such as helminth
infection. This inevitably increases the scope for DDIs.

1.3 Polypharmacy in an ageing population
There are an increasing number of patients over 50 years living with HIV [Nguyen et al, 2008], in
whom chronic conditions associated with ageing may co-exist. These include cardiovascular drugs,
lipid lowering agents, antihypertensives and analgesics. The use of non-prescribed medications in
patients taking ARV in Canada and the UK is widespread [Dhalla et al, 2006, Ladenheim et al,
2008]. These include recreational or illicit drug use.

1.4 Decentralised models of care
In many healthcare settings, the provision of antiretroviral therapy is progressively devolving from
tertiary care- in developing countries, this means the decentralisation of care to district level. Even
with an intensive programme of training and education, practitioners with less expertise in
prescribing of ARVs may be less likely to identify DDIs or recognise their adverse consequences.

1.5 Lack of monitoring in resource-poor settings
Lack of pharmacovigilance structures, and laboratory monitoring coupled with the high background
of febrile and other illness may mask clinically significant DDIs in resource-poor settings. Moreover
the syndromic management of illness, the high rates of self-treatment (especially for malaria) and
widespread use of traditional medicines (which may contain ingredients such as St John’s Wort
and steroids) make a complete list of patient medications difficult to compile.

Minimising Harm From DDIs
While DDIs involving HIV drugs are often unavoidable, many can be better managed. Lack of
awareness and recognition of clinically significant DDIs is a major obstacle to safe ARV
prescribing. This review undertakes a systematic evaluation of potential DDIs between ARVs and
drugs used to treat TB, malaria, chronic Hepatitis B&C infections and opioid dependence.



3
2 METHODS

2.1 Search Strategy
The following searches were used on PubMed (1987-July 31
st
2009). If the standard searched
returned numerous results which were not relevant, then the refined search was used.

Standard Search
DrugName AND CoMed AND english[Language] NOT review[Publication Type] NOT child[MeSH
Terms]

Refined Search
DrugName AND CoMed AND english[Language] NOT review[Publication Type] NOT child[MeSH
Terms] AND (drug interactions[MeSH Terms] OR anti-hiv agents/pharmacokinetics[MeSH Terms]
OR reverse transcriptase inhibitors/pharmacokinetics[MeSH Terms])



For all antiretrovirals, we searchedthe manufacturer’s Summary of Product Characteristics
(Europe) (http://emc.medicines.org.uk/) and Product Information (USA) (from each anti-retroviral
manufacturer’s website). Websites were accessed (to August 21
st
2009) .

We searched the following conference reports for (peer-reviewed) DDI abstracts:
• Conference on Retroviruses and Opportunistic Infections (2004 - February 2009)
• International AIDS Society Conference (2005 – July 2009)
• World AIDS Conference (2004- July 2008)
• Interscience Conference on Antimicrobial Agents and Chemotherapy (2004 - Sept
2008)
• International Workshop on Clinical Pharmacology of HIV Therapy (2004 - April 2009)
• International Congress on Drug Therapy in HIV Infection (2004 - December 2008)
• European AIDS Clinical Society (2005, 2007)



2.2 Study Selection
We included all studies that evaluated pharmacokinetic data when antiretrovirals were combined
with: TB drugs, antimalarials, hepatitis B treatment, hepatitis C treatment, and drug used to treat
opioid dependence. Studies which reported clinical interactions only, or overlapping toxicity were
not included. Drugs in development which were not yet licensed were excluded. Studies involving
children were excluded.


2.3 Study Quality Assessment
In order to develop a system which is robust, easy to apply in a consistent manner and allows the
user to assess the applicability of existing data to clinical practice, we will apply the GRADE
system of classification to the strength of recommendation (Table 1), and the quality of evidence
(Table 2) [Atkins et al, 2004]. The strength of evidence is framed in the following question: Is it safe
to administer both drugs? We will utilise our existing ‘traffic lights’ system, which maps onto
GRADE equivalents outlined in Table 1.


6
2.3.1 Strength of Recommendation

Table 1 ‘Traffic lights’ summary of Drug-drug interactions

L|verpoo| Webs|te Def|n|t|on:

GkADL Lqu|va|ent
Ooestloo. ls lt sofe to oJmlolstet botb Jtoqs ?


no cllnlcally slgnlflcanL lnLeracLlon, or lnLeracLlon
unllkely based on knowledge of drug meLabollsm

?LS

ÞoLenLlal lnLeracLlon LhaL may requlre close
monlLorlng, alLeraLlon of drug dosage or Llmlng of
admlnlsLraLlon

Þrobably ?LS lf
• beneflL ouLwelghs rlsk, or
• Lhe lnLeracLlon can be safely managed

Þrobably nC lf
• rlsk ouLwelghs beneflL, or
• Lhe lnLeracLlon cannoL be safely managed


lnLeracLlon llkely, do noL use or use wlLh cauLlon

nC

∇
1here are no clear daLa, acLual or LheoreLlcal, Lo
lndlcaLe wheLher an lnLeracLlon wlll occur

uCn1 knCW


noLe: Where daLa are lacklng, recommendaLlons wlll be lnfluenced by Lhe ManufacLurer's Summary of ÞroducL
CharacLerlsLlcs (approved by Lhe LMLA), or Þrescrlblng lnformaLlon (approved by Lhe luA). A 8Lu recommendaLlon
wlll be asslgned when 'avold' or 'conLralndlcaLed' ls advlsed. An AM8L8 recommendaLlon wlll be asslgned when
'cauLlon' ls advlsed. Where Lhe comblnaLlon of Lwo drugs ls 'noL recommended', a classlflcaLlon of AM8L8 or 8Lu wlll
be made on Lhe basls of llkely poLenLlal adverse lmpacL of Lhls lnLeracLlon, and Lhe LherapeuLlc lndex of affecLed
drug(s). We wlll ensure LhaL recommendaLlons are conslsLenL wlLh Lhe llcenslng guldance wherever posslble
(assessmenL of Lhe quallLy of evldence supporLlng LhaL guldance ls a separaLe process). Where dlfferences exlsL
beLween luA and LMLA oplnlons around a uul, our recommendaLlons wlll reflecL Lhe more conservaLlve approach.



2.3.2 Grading of Evidence
Grading of quality of evidence will be achieved using a methodology based upon the GRADE
system of classification [Atkins et al, 2005,] (Table 2). Four categories are proposed, which reflect
a hierarchy of methodological design and execution of a study. Ability to up- or down-grade the
assessment of quality is also set out in Table 2 and closely mirrors GRADE.


Table 2 Assessment of Quality of DDI Evidence
C8AuL equlvalenL

uowngrade*

upgrade*
Plgh
Lvldence obLalned from aL leasL one
properly deslgned and execuLed
randomlzed conLrolled Lrlal.

ModeraLe

Low
Lvldence obLalned from observaLlonal
sLudles.

very Low

SLudy CuallLy
• serlous llmlLaLlons (-1)
• very serlous llmlLaLlons (-2)
• lmporLanL lnconslsLency (-1
or -2)

ulrecLness
• some uncerLalnLy (-1)
• ma[or uncerLalnLy (-2)

sparse or lmpreclse daLa (-1)

hlgh probablllLy of publlcaLlon
SLrong assoclaLlon
• sLrong, no confounders,
conslsLenL & dlrecL evldence
(+1)**
• very sLrong, no ma[or LhreaLs
Lo valldlLy, dlrecL evldence
(+2)***
• evldence of dose response
gradlenL (+1)
• all plauslble confounders
would have reduced effecL
(+1)
Formatted Table
Formatted: Bullets and
Numbering
Formatted: Bullets and
Numbering
Formatted: Bullets and
Numbering
Formatted: Bullets and
Numbering
Deleted: reporLlng
7
blas (-1)

* 1= move up or down one grade (e.g. from high to moderate); 2= move up or down two grades (e.g. from high to low)

** A statistically significant relative risk of >2 (<0.5), based on consistent evidence from two or more observational
studies, with no plausible confounders

*** A statistically significant relative risk >5 (<0.2) based on direct evidence with no major threats to validity


Evidence based on population pharmacokinetic modelling was graded according to the quality of
the primary data upon which that model was based. Examples of issues impacting study quality or
directness, which led to downgrading were: trough, or ‘random’ pharmacokinetic sampling or
sparse sampling not supported by a validated population PK model, use of single dose studies
(more acceptable for known enzyme inhibitors, less acceptable for known enzyme inducers),
healthy volunteer data where existing literature suggests different plasma drug exposure in
disease. Studies available as abstracts only, or evidence from the manufacturer’s in-house clinical
studies of drug interactions which were submitted to the regulatory authorities but not published in
peer-reviewed journals were graded as ‘very low’. Evidence based on case reports, descriptive
reports, experience of experts or knowledge of mechanisms of drug disposition which predict
presence or absence of interaction, were also graded ‘very low’. These included recommendations
in Manufacturer’s SPC or Prescribing Information, where published clinical data were lacking.

All grading was carried out by 4 assessors (SK, DJB, SG, KS). Where there was discordance, the
studies involved were discussed to arrive at an agreed grading.


2.4 Data Synthesis
We utilised a standard data extraction template to systematically assess and summarise the
evidence, and extracted relevant data into evidence templates (presented in Appendices)


Formatted: Font: 9 pt,
Complex Script Font: 9 pt
Formatted: Font: (Default)
Arial, 11 pt, Complex Script
Font: Arial, 11 pt
Formatted: Font: (Default)
Arial, 11 pt, Complex Script
Font: Arial, 11 pt
Formatted: Font: (Default)
Arial, 11 pt, Complex Script
Font: Arial, 11 pt
Deleted: in-house
Deleted: ¶
¶
8
3 RESULTS

3.1 Antimalarials
In general there are potential interactions between HIV protease inhibitors and NNRTIs and
lumefantrine, quinine and amodiaquine. However, only few drug interaction studies have
been performed. These studies have varied in design and quality, utilising both healthy
volunteers and HIV-positive subjects. They may not reflect what happens in real life,
particularly as the pharmacokinetics of many antimalarials alter with disease. For example,
the protein binding and plasma half-life of quinine increases with severity of malaria,
lumefantrine absorption is decreased during acute malaria and the pharmacokinetics of
mefloquine also alters with disease. Concentrations of quinine and lumefantrine also
accumulate with multiple dosing, and single dose studies only yield limited data. In
addition, pharmacogenetic effects are not usually explored in small drug interaction studies,
for example pyrimethamine is predominantly metabolised by CYP2C19, and the frequency
of poor metabolisers differs between Africans (3%), South East Asians (20%) and
Caucasians. Specific points to note are:

3.1.1 Quinine – Quinine is extensively metabolised by CYP3A4, and its AUC is increased
over fourfold by ritonavir (200mg) in healthy volunteers. However, the impact of the more
usual dose of ritonavir (100mg) is uncertain, and this study needs repeating. It seems
likely, given the limited data, that HIV positive patients on boosted PIs may be at increased
risk of cinchonism. An important issue is whether or not a loading dose of quinine is
required in patients with severe malaria who are receiving a boosted protease inhibitor.
The AUC of quinine is reduced by approximately a third with nevirapine, although the
clinical significance of this is uncertain.

3.1.2 Amodiaquine – Excessive risk of hepatotoxicity has been reported in healthy
volunteers who were also given efavirenz. There are no data for nevirapine and other
boosted protease inhibitors although such studies should be undertaken very cautiously
(using very low doses initially) in healthy volunteers. Prolonged neutropenia has been
reported in Ugandan children treated with amodiaquine, who were also receiving
antiretrovirals.

3.1.3 Lumefantrine – Lumefantrine is extensively metabolised by cytochrome P450 CYP
3A4. Lumefantrine does not seem to prolong the QT interval, but its pharmacokinetics are
complex and variable and a marked food effect is observed. Interactions with PIs and
NNRTIs are likely, and the manufacturer’s SPC advises that co-administration of CYP3A4
inhibitors such as PIs are contraindicated. An approximately twofold rise in AUC was
reported in healthy volunteers who were given lumefantrine with lopinavir/ritonavir. This
interaction may be beneficial if it could be shown to reduce the marked pharmacokinetic
variability of lumefantrine, or to abolish the food restrictions required with this antimalarial.

3.1.4 Artemether is metabolised via CYP3A4 to dihydro artemesinin (although both
compounds have anti-malarial activity, dihydro artemesinin has greater potency). Inhibition
of 3A4 would reduce dihydro artemesinin, but increase artemether and potentially increase
the short half life of artemether (1-2 h). The effects of PIs and NNRTIs are unclear.

3.1.5 Mefloquine had variable effect on ritonavir metabolism - no interaction was noted
after a single dose but ritonavir plasma AUC was reduced by 31% and Cmax by 36% after
multiple dosing. PK of mefloquine was not significantly influenced by RTV.

3.1.6 Since proguanil is a pro-drug and is partially activated (CYP2C19) to cycloguanil
there is concern that inhibition of metabolism by ritonavir or ritonavir-containing boosted PI
regimens will reduce pharmacological effect. However, synergy with atovaquone is related
to proguanil, not cycloguanil. When both drugs are co-administered, CYP2C19 inhibition
could potentially enhance this synergistic effect, which may off-set decreased cycloguanil
formation.
9

3.1.7 Atovaquone decreases zidovudine oral clearance leading to a 35 % ± 23 %
increase in plasma zidovudine AUC. The clinical significance of this is unknown, and no
dose modification is recommended.

Lopinavir may decrease plasma concentrations of atovaquone, the clinical significance of
which is unknown, however, increases in atovaquone doses may be needed. Atovaquone
lowers indinavir exposure, reducing Cmin by ~23%. Another healthy volunteer study
observed indinavir AUC decrease of 5%, but increase in atovaquone AUC (13%) and Cmax
(16%) when both drugs were co-administered. No dosage adjustments are necessary for
atovaquone when given with indinavir. The clinical significance of lowered indinavir
concentrations is uncertain since these were healthy volunteer studies carried out without
ritonavir boosting (which is no longer the preferred means of giving indinavir ). Moreover,
clinical studies have shown higher plasma indinavir in Thai patients (who have lower body
weight), and given the toxicity of indinavir at higher doses, dosage adjustments are not
indicated for indinavir (boosted with ritonavir) when dosed with atovaquone or malarone.

3.1.8 Previous formulations of ddI (buffered tablets) decrease dapsone concentrations, in
some cases leading to failure of Pneumocystis prophylaxis. No interaction was observed
with newer formulations.

Interactions between co-trimoxazole use and malaria, or antiprotozoal effects of protease inhibitors
are not within the scope of this review.



1ab|e 3 Ant|ma|ar|a|s
Ant|ma|ar|a|s & ÞIs
Þrotease Inh|b|tors
A1V DkV IÞV IDV LÞV NIV k1V SÇV 1ÞV
Amodlaqulne Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
ArLemeLher/
ArLesunaLe
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Creen
(2)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
ALovaquone Amber
(4)
Amber
(4)
Creen
(4)
Amber
(4)
Amber
(4)
Creen
(4)
Amber
(4)
Creen
(4)
Amber
(4)
Chloroqulne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
Creen
(4)
Creen
(4)
Cllndamycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
Creen
(4)
Creen
(4)
uoxycycllne/
1eLracycllne
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LumefanLrlne Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(2)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Mefloqulne Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(3)
Amber
(4)
Amber
(4)
Þrlmaqulne no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
Þroguanll Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
Creen
(4)
Creen
(4)
Culnlne Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Sulfadoxlne/
ÞyrlmeLhamlne
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
Creen
(4)
Amber
(4)

Deleted: 8ed
10
Ant|ma|ar|a|s &NNk1Is, Cthers

NNk1Is Cthers
LIV L1V NVÞ MVC kAL
Amodlaqulne 8ed
(2)
Amber
(4)
Amber
(4)
Amber
(4)
Creen
(4)
ArLemeLher/
ArLesunaLe
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Creen
(4)
ALovaquone Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
Chloroqulne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Cllndamycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
uoxycycllne/
1eLracycllne
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LumefanLrlne Amber
(4)
Amber
(4)
Amber
(4)
Creen
(4)
Creen
(4)
Mefloqulne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Þrlmaqulne no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
Þroguanll Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Culnlne Amber
(4)
Amber
(4)
Amber
(3)
Creen
(4)
Creen
(4)
Sulfadoxlne/
ÞyrlmeLhamlne
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)

11
Ant|ma|ar|a|s & Nk1Is

Nk1Is
A8C ddI I1C 31C d41 1DI 2DV
Amodlaqulne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
ArLemeLher/
ArLesunaLe
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
ALovaquone Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(3)
Chloroqulne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Cllndamycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
uoxycycllne/
1eLracycllne
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LumefanLrlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Mefloqulne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Þrlmaqulne no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
Þroguanll Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Culnlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Sulfadoxlne/
ÞyrlmeLhamlne
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(2)

12
3.2 TB Drugs
Drug-drug interactions should be viewed as only one part of a complex and multi-faceted
clinical problem when treating both diseases. Factors such as timing the introduction of
antiretroviral therapy, its effect in preventing early mortality, use of alternative rifamycins,
overlapping toxicity and immune reconstitution are important issues which are currently
being assessed within clinical trials. Suffice to say, despite recent advances in drug
development, there remains no real alternative to rifampicin use in developing countries in
the foreseeable future.

3.2.1 In general, interactions between rifampicin and protease inhibitors
(boosted/unboosted) result in substantially lowered PI exposure, which renders these
interactions difficult or impossible to manage in clinical practice. Strategies to overcome
this interaction have yielded only limited success. Ritonavir (at full dose) is poorly tolerated.
Doubling the dose of lopinavir/ritonavir, or ‘super-boosting’ with higher ritonavir doses
increased lopinavir exposure in one PK study, but double dose lopinavir/ritonavir failed to
prevent low drug exposure in a significant number of children in South Africa with HIV/TB
co-infection [McIlleron et al 2009]. Moreover, healthy volunteers given rifamycin with
boosted protease inhibitors (saquinavir, lopinavir, atazanavir) appeared to be at excessive
risk of hepatotoxicity, making this pharmacokinetic interaction difficult to study safely.

3.2.2 NNRTIs – Pharmacokinetic studies suggest that rifampicin has a greater impact in
lowering drug exposure of nevirapine (AUC ↓ 40-58%) compared with efavirenz (AUC ↓
26%). One large cohort study reported that when antiretroviral therapy is commenced in
patients receiving rifampicin-containing TB treatment, treatment outcomes with standard
dose efavirenz are superior to nevirapine, and comparable with patients on efavirenz who
were not receiving TB therapy. No difference in efficacy was observed in patients receiving
either efavirenz or nevirapine who subsequently required TB therapy. These differences
could have resulted from the lead-in phase of dosing of nevirapine undertaken during
rifampicin therapy.

3.2.3 Current international treatment guidelines prefer efavirenz to nevirapine in patients
requiring rifampicin. However, there is no universal consensus about how to manage the
efavirenz-rifampicin interaction. Lopez-Cortes et al [2006] conducted a two period
sequential study which supported weight-based dose increment of efavirenz during
rifampicin therapy. However, other studies in Africans and South-East Asians have shown
that while pharmacokinetic variability of efavirenz is markedly increased in the presence of
rifampicin, median exposures are adequate, and outcome is good. Numerous factors may
account for these differences, not least sampling strategy (trough versus ‘random’ versus
AUC sampling), body weight and pharmacogenetic influences (cytochrome P450 CYP2B6
poor metabolisers are more common in black Africans and South East Asians compared
with Caucasians).

3.2.4 NRTIs – Although use of triple NRTI regimens as first line agents has resulted in
inferior outcomes, combination treatment with three (zidovudine, lamivudine, abacavir) or
four (zidovudine, lamivudine, abacavir plus tenofovir) drugs during TB therapy has yet to be
properly assessed.

3.2.5 Newer drugs – Of the newer agents, raltegravir has shown promise as an effective
antiretroviral in patients receiving rifampicin, since plasma exposure is only modestly
reduced. Furthermore, dose ranging studies have shown that the marked antiviral effect of
raltegravir is not blunted even when doses as low as 100mg twelve hourly (a quarter of the
adult daily dose) are administered. This lack of a clear pharmacokinetic-pharmacodynamic
relationship has caused regulatory authorities to differ in recommendations when raltegravir
is co-prescribed with rifampicin. The FDA recommends no dose increment, while the
EMEA suggests that a dose increment of raltegravir could be considered. Clinical trial data
are awaited. Enfuvirtide is also an option, but the high cost and need for twice daily
injections makes this a second line option.
13

1ab|e 4 18 Drugs

18 Drugs & ÞIs

Þrotease Inh|b|tors
A1V DkV IÞV IDV LÞV NIV k1V SÇV 1ÞV
Amlkacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Capreomycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Cycloserlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LLhambuLol Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LLhlonamlde no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
CaLlfloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber*
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
lsonlazld Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
kanamycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Levofloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber*
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Moxlfloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber*
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Cfloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber*
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Þ-Amlnosallcyllc
Acld
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
Þyrazlnamlde Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
8lfabuLln Amber
(4)
Amber
(4)
Amber
(2)
Amber
(2)
Amber
(4)
Amber
(3)
Amber
(2)
Amber
(2)
Amber
(3)
8lfamplcln 8ed
(2)
8ed
(4)
8ed
(3)
8ed
(3)
8ed
(2)
8ed
(4)
Amber
(3)
8ed
(2)
8ed
(4)
8lfapenLlne Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
Amber
(4)
SLrepLomycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)

Deleted: Amber
14
18 Drugs & NNk1Is, Cthers

NNk1Is Cthers
LIV L1V NVÞ MVC kAL
Amlkacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Capreomycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Cycloserlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LLhambuLol Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LLhlonamlde no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
CaLlfloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
lsonlazld Creen
(3)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
kanamycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Levofloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Moxlfloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Cfloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Þ-Amlnosallcyllc
Acld
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
Þyrazlnamlde Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
8lfabuLln Amber
(3)
Amber
(4)
Amber
(4)
Amber
(4)
Creen
(4)
8lfamplcln Amber
(2)
8ed
(4)
8ed
(2)
Amber
(2)
Amber
(2)
8lfapenLlne Amber
(4)
8ed
(4)
Amber
(4)
Amber
(4)
Amber
(4)
SLrepLomycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)

13
18 Drugs & Nk1Is

Nk1Is
A8C ddI I1C 31C d41 1DI 2DV
Amlkacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
Creen
(4)
Capreomycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
Creen
(4)
Cycloserlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LLhambuLol Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LLhlonamlde no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
CaLlfloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
lsonlazld Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber*
(4)
Creen
(4)
Creen
(4)
kanamycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
Creen
(4)
Levofloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(3)
Moxlfloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Cfloxacln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Þ-Amlnosallcyllc
Acld
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
no
uaLa
Þyrazlnamlde Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
8lfabuLln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(3)
8lfamplcln Amber
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(2)
Amber
(3)
8lfapenLlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
SLrepLomycln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Amber
(4)
Creen
(4)

16
3.3 Hepatitis B and C treatment

3.3.1 Hepatitis B
Hepatitis B infection is widely prevalent in Asia and Africa. Infection with HIV is common.
Management of both diseases is complex since in addition to increased risk of liver
dysfunction with antiretrovirals, HIV drugs such as lamivudine, emtricitabine and tenofovir
have activity against hepatitis B virus. Entecavir (a hepatitis B drug) may also exhibit anti-
HIV activity. There is concern that lack of routine hepatitis B testing within national
antiretroviral programmes in developing countries, coupled with the use of first line
regimens based on stavudine/lamivudine or zidovudine/lamivudine plus an NNRTI will
effectively deliver 3TC monotherapy, and may result in widespread drug resistance to
hepatitis B in resource-limited settings.

Clinically significant drug interactions mainly involve tenofovir and are listed below.

3.3.1.1 Tenofovir has known interactions with HIV protease inhibitors, increasing the
exposure of darunavir and saquinavir modestly. In contrast, plasma exposure to atazanavir
is decreased (AUC ↓ 25%) by tenofovir. This interaction may to some degree be offset by
the use of boosted atazanavir (at doses of either 300mg or 400mg combined with 100mg of
ritonavir).

3.3.1.2 Tenofovir exposure is also modestly increased by certain boosted protease inhibitor
combinations such as lopinavir/ritonavir, saquinavir/ritonavir and darunavir/ritonavir.

3.3.1.3 Tenofovir significantly increases didanosine exposure (through inhibition of purine
nucleoside phosphorylation) and the combination is contraindicated.



3.3.2 Hepatitis C
Recent advances in the development of agents that act specifically to inhibit hepatitis C
virus (HCV) look set to fundamentally change the way that patients will be treated. New
directly acting anti-HCV agents such as protease and polymerase inhibitors will initially be
added to standard care with pegylated interferon alfa and ribavirin. However, future therapy
is likely to constitute combinations of agents which act at distinct stages of viral replication
and have differing resistance profiles. While directly acting anti-HCV agents will
undoubtedly improve treatment outcomes, the introduction of combination therapy may not
be without complication in some patient groups. HIV positive patients who are receiving
antiretrovirals are relatively highly represented among those with HCV infection, and are at
high risk of drug-drug interactions.

3.3.2.1 Concomitant administration of abacavir with PEG-IFN and ribavirin has been
associated with an increased risk of non-response to anti-HCV therapy [Bani-Sadr et al,
2007]

and an interaction between abacavir and ribavirin has been suggested. As both drugs
are guanosine analogues and have some metabolic pathways in common, an inhibitory
competition for phosphorylation may be possible between ribavirin and abacavir [Mira et al,
2008].

3.3.2.2 Combinations of zidovudine with ribavirin and PEG-IFN can lead to increased risk of
severe haematological toxicity, including anaemia. The use of zidovudine has been
identified as an independent factor contributing to haematological adverse events in
patients undergoing ribavirin and PEG-IFN treatment; the combination is not recommended
[Mira et al, 2007].

3.3.2.3 The use of didanosine alongside ribavirin is associated with increased risk of
mitochondrial toxicity, which may be attributed to increased exposure to the active
metabolite of didanosine, dideoxyadenosine 5’-triphosphate when didanosine is
17
coadministered with ribavirin [Bani-Sadr et al, 2005; Montes Ramirez et al, 2002; Videx US
Prescribing Information]. Toxicity may be severe and coadministration is not
recommended.

3.3.2.4 Mitochondrial toxicity has also been observed with combinations of stavudine and
ribavirin. In vitro data has shown that ribavirin can inhibit phosphorylation of zidovudine and
stavudine. The clinical significance is not clear, however manufacturers of ribavirin advise
close monitoring of HIV RNA with this combination.

3.3.2.5 Although clinical significance is not thought to be high, the use of atazanavir with
ribavirin and IFN has been associated with hyperbilirubinaemia [Rodriguez-Novoa et al,
2008].

3.3.2.6 In the case of patients receiving efavirenz alongside PEG-IFN, monitoring of central
nervous system effects is important, as incidence of depressive symptoms in patients with
HIV/HCV co-infection treated with IFN is reportedly high [Laguno et al, 2004].

Currently, ARV treatment may be adjusted, as far as is practicable, to enable optimal
administration of anti-HCV therapy, without compromising ARV efficacy. This will become
increasingly complex to manage with the addition of new Hepatitis C agents.








1ab|e S nepat|t|s Drugs

nepat|t|s Drugs & ÞIs

Þrotease Inh|b|tors
A1V DkV IÞV IDV LÞV NIV k1V SÇV 1ÞV
Adefovlr Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LnLecavlr Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
lnLerferon o/
Þeg-lnLerferon o
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Lamlvudlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(2)*
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
8lbavlrln Amber
(2)*
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
1elblvudlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
1enofovlr Amber
(2)
Amber
(4)
Amber
(2)
Amber
(4)
Amber
(2)
Creen
(2)
Amber
(4)
Amber
(2)
Creen
(4)



18

nepat|t|s Drugs & NNk1Is, Cthers

NNk1Is Cthers
LIV L1V NVÞ MVC kAL
Adefovlr Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
LnLecavlr Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
lnLerferon o/
Þeg-lnLerferon o
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Lamlvudlne Creen
(4)
Creen
(4)
Creen
(3)
Creen
(3)
Creen
(4)
8lbavlrln Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
1elblvudlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
1enofovlr Creen
(2)
Creen
(4)
Creen
(3)
Creen
(3)
Creen
(2)



nepat|t|s Drugs & Nk1Is

Nk1Is
A8C ddI I1C 31C d41 1DI 2DV
Adefovlr Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
8ed
(4)*
Creen
(4)
LnLecavlr Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
lnLerferon o/
Þeg-lnLerferon o
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
Lamlvudlne Creen
(3)
Creen
(4)
8ed
(4)
- Creen
(4)
Creen
(2)
Creen
(3)
8lbavlrln Amber
(2)
8ed
(3)
Creen
(3)
Creen
(2)
Amber
(3)
Creen
(2)
8ed
(3)*
1elblvudlne Creen
(4)
Creen
(4)
Creen
(4)
Creen
(4)
no
daLa
Creen
(4)
no
daLa
1enofovlr Amber
(3)*
8ed
(2)
Creen
(4)
Creen
(4)
Creen
(4)
- Creen
(4)

19
3.4 Opioid Dependence
Methadone and buprenorphine are the two most commonly used drugs as replacement
therapy for opioid dependency. Methadone is usually prescribed as a racemic mixture,
containing equal proportions of R-methadone (the active form) and S-methadone (with less
activity, but may be responsible for some toxicity). Buprenorphine undergoes extensive
first pass metabolism, and is consequently administered sublingually. It is metabolised
principally by hepatic cytochrome P450 CYP3A4 (to norbuprenorphine), and by
glucuronidation. As a result there are potential significant interactions with protease
inhibitors.

3.4.1 Methadone – Boosted protease inhibitors and NNRTIs increase the clearance of
R-methadone by enzyme induction. Clinical symptoms of opioid withdrawal are well
documented. If co-administered, consider increasing the dose of methadone.

3.4.2 Buprenorphine – Some studies have reported that exposure of buprenorphine and
its metabolites may be increased by concomitant protease inhibitor use, while others have
failed to observe this. The clinical significance is uncertain. It seems prudent to commence
replacement with a low dose of buprenorphine in a patient receiving boosted protease
inhibitors.



Table 6 Opioid Replacement Therapy

Opioid Replacement Therapy & PIs

Protease Inhibitors
ATV DRV FPV IDV LPV NFV RTV SQV TPV
Buprenorphine Amber
(3)
Amber
(4)
Amber
(4)
Amber
(4)
Green
(3)
Amber
(3)
Amber
(3)
Amber
(4)
Amber
(4)
Methadone Green
(2)
Amber
(4)
Amber
(2)
Amber
(2)
Amber
(2)
Amber
(2)
Amber
(2)
Amber
(2)
Amber
(4)


Opioid Replacement Therapy & NNRTIs, Others

NNRTIs Others
EFV ETV NVP MVC RAL
Buprenorphine Amber
(2)
Amber
(4)
Amber
(4)
Green
(4)
Green
(4)
Methadone Amber
(2)
Amber
(3)
Amber
(2)
Green
(4)
Green
(4)


Opioid Replacement Therapy & NRTIs

NRTIs
ABC ddI FTC 3TC d4T TDF ZDV
Buprenorphine Amber
(4)
Green
(4)
Green
(4)
Green
(4)
Green
(4)
Green
(4)
Green
(3)
Methadone Amber
(4)
Amber
(3)
Green
(4)
Green
(3)
Amber
(3)
Green
(2)
Amber
(2)

20
4 DISCUSSION

Drug-drug interactions are one of the commonest causes of medication error in developed
countries, and antiretrovirals among the most therapeutically risky drugs for clinically significant
drug interactions. Studies in the Netherlands and New York involving 115 and 550 patients
suggest a prevalence of 20-25% CSDIs [de Maat at al, 2004; Shah et al, 2007]. A second study in
New York involving 153 patients reported a prevalence of 41.2% [Miller et al, 2007]. Two recent
studies conducted in Liverpool (159 patients) and Switzerland (771 patients) reported prevalence
of 26.3% and 61% respectively [Cottle at al, 2009; Marzolini at al, 2008]. Although definitions
differed, four out of five of these studies utilised the Liverpool Drug Interactions website to screen
for interactions. There have been no such studies in resource-limited settings where risk is
arguably increased as a result of less laboratory monitoring, high rates of background illness
(which may result in adverse effects being missed), lack of affordable alternative treatments, use of
fixed dose combinations (that offer less flexibility for managing interactions) and lack of
pharmacovigilance data. In addition, there is a higher cost of treatment failure in these settings,
since options are limited compared with developed countries.

Use of therapeutic drug monitoring is not feasible as a strategy for managing CSDIs in resource-
poor settings. Practical steps that can be instituted to reduce the risk of adverse outcomes from
CSDIs include integrating national treatment programmes for HIV and other diseases (with
protocols that minimise drug interactions), establishing regional networks for pharmacovigilance,
and improving the quality of prescribing through training and education of health care workers.
Knowledge of common interactions involving antiretrovirals on a country-specific basis will allow
targeted training, monitoring and protocol development. Finally, we believe that large antiretroviral
programmes should consider undertaking an audit of clinically significant drug interactions as a
proxy for the quality of prescribing within that scheme.



21
5 Acknowledgements & Declarations

We thank Holger Schunemann and Paul Garner for advice on applying GRADE criteria to the
assessment and classification of drug interactions. Our classification system was developed in
Liverpool and it’s use does not imply endorsement from the GRADE Working Group.

The Liverpool HIV Drug interactions Website (www.hiv-drug-interactions.org) receives educational
grants from Abbott, Gilead, Merck, Bristol-Myers-Squibb, Pfizer, Tibotec, GlaxoSmithKline and
Boehringer Ingelheim. Support has also been received from research grant funding from the UK
National Institute for Health Research, and the EU. Editorial content remains entirely independent.
SK and DJB have received research grant support, PhD studentships, travel bursaries and
consultancy fees from Boehringer, GlaxoSmithKline, Tibotec, Merck, Bristol-Myers-Squibb and
Pfizer.

22
6 REFERENCES


Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, et al. Grading quality of evidence and
strength of recommendations. BMJ. 2004 Jun 19;328(7454):1490.

Atkins D, Briss PA, Eccles M, Flottorp S, Guyatt GH, Harbour RT, Hill S, Jaeschke R, Liberati A, Magrini N,
Mason J, O'Connell D, Oxman AD, Phillips B, Schünemann H, Edejer TT, Vist GE, Williams JW Jr; GRADE
Working Group. Systems for grading the quality of evidence and the strength of recommendations II: pilot
study of a new system. BMC Health Serv Res. 2005 Mar 23;5(1):25.

Bani-Sadr F, Denoeud L, Morand P et al. Early Virologic Failure in HIV Coinfected Hepatitis C Patients
Treated With the Peginterferon-Ribavirin Combination Does Abacavir Play a Role? J Acquir Immune Defic
Syndr 2007; 45:123-5

Bani-Sadr F, Carrat F, Pol S et al. Risk factors for symptomatic mitochondrial toxicity in HIV/hepatitis C virus-
coinfected patients during interferon plus ribavirin-based therapy. J Acquir Immune Defic Syndr 2005; 40:
47–52

Cottle LE, Evans-Jones JG, SH Khoo SH. Physician awareness of antiretroviral drug interactions. 15th
Annual Conference of the British HIV Association, Liverpool, April 2009, abstract P138. HIV Medicine, 10
(Suppl. 1), 11–56

De Maat M, De Boer A et al. Evaluation of clinical pharmacist interventions on drug interactions in outpatient
pharmaceutical HIV-care. Journal of Clinical Pharmacy and Therapeutics 2004, 29:121-130

Dhalla S, Chan K et al, Complementary and alternative medicine use in British Columbia - a survey of HIV
positive people on antiretroviral therapy. Complementary Therapies in Clinical Practice 2006, 12: 242-248

Kigen G, Kimaiyo S, Owen A, Edwards G, Back D, Gibbons S, Sang E, Khoo S. Prevalence of drug
interactions between antiretroviral and co-administered drugs at the Moi Teaching and Referral Hospital
(AMPATH), Eldoret-Kenya. 9th International Congress on Drug Therapy in HIV Infection, Glasgow,
November 2008, abstract O122.

Ladenheim D, Horn O et al. Potential health risks of complementary alternative medicines in HIV patients.
HIV medicine 2008, 9:653-659

Laguno M, Blanch J, Murillas J et al. Depressive symptoms after initiation of interferon therapy in human
immunodeficiency virus-infected patients with chronic hepatitis C Antivir Ther 2004; 9:905-9

McIlleron H, Ren Y, Nuttall J, Riddick A, Kleynhans L, Rabie H, Cotton M, Eley B, Merry C, Maartens G.
Double-dose lopinavir/ritonavir provides insufficient lopinavir exposure in children receiving rifampicin-based
anti-TB treatment. 16th CROI. February 2009, Montreal. Oral abstract 98.

Miller C, El-Kholi R et al. Prevalence and risk factors for clinically significant drug interactions with
antiretroviral therapy. Pharmacotherapy 2007, 27(10):1379-1386

Mira J, Lopéz-Corté L, Barreiro P et al. Efficacy of pegylated interferon plus ribavirin treatment in
HIV/hepatitis C virus co-infected patients receiving abacavir plus lamivudine or tenofovir plus either
lamivudine or emtricitabine as nucleoside analogue backbone. J Antimicrob Chemother 2008; 62:1365-73

Mira J, López-Cortéz L, Merino D et al. Predictors of severe haematological toxicity secondary to pegylated
interferon plus ribavirin treatment in HIV-HCV coinfected patients. Antivir Ther 2007; 12:1225-35

Montes Ramírez ML, Rodríguez Zapata M. Report of three cases of hyperlactacidemiae/lactic acidosis after
treatment of hepatitis C with pegylated interferon and ribavirin in HIV coinfected patients Rev Clin Esp 2002;
202:543-5

Nguyen N, Holodniy M. HIV infection in the elderly. Clin Interv Aging 2008, 3(3):453-472

Rastegar D, Knight A, Monolakis J. Antiretroviral errors among hospitalised patients with HIV infection.
Clinical Infectious Diseases 2006, 43 (Oct 1):933-938

Formatted: French France
23
Rodríguez-Nóvoa S, Morello J, González M et al. Increase in serum bilirubin in HIV/hepatitis-C virus-
coinfected patients on atazanavir therapy following initiation of pegylated-interferon and ribavirin AIDS 2008
Nov 30;22(18):2535-7

Shah S, McGowan J et al. Interaction of drug interactions involving ART in New York City HIV specialty
clinics. CROI 2007, Abstract 573

UNAIDS Report on the Global AIDS Epidemic, 2008

Videx US Prescribing Information, Bristol-Myers Squibb, http://packageinserts.bms.com/pi/pi_videx_ec.pdf
01/06/09 date last accessed.