583www.expert-reviews.com ISSN 1747-6348© 2008 Expert Reviews Ltd10.1586/17476348.2.5.583

Perspective

Growing threat of drug-resistant tuberculosisTuberculosis (TB) continues to be one of the most important infectious disease threats to human health [1]. Multidrug-resistant TB (MDR-TB) is TB that is resistant to at least two of the most important first-line antimyco-bacterial drugs – isoniazid (INH) and rifampicin (RIF). Extensively drug-resistant TB (XDR-TB) is a relatively rare but potentially lethal type of MDR-TB. XDR-TB is defined as TB that is resistant to INH and RIF, plus resist-ant to any fluoroquinolone and at least one of three injectable second-line drugs (amikacin, kanamycin or capreomycin) [2].

It has been estimated that over 400,000 new cases of MDR-TB occur each year and, although their exact incidence rates are cur-rently unknown, XDR-TB cases have been

detected in every country where there is the capacity to detect them [2,3]. Growing concerns regarding the spread of MDR-TB and alarm over the emergence of XDR-TB have sparked a great deal of interest in the development and application of rapid diagnostic tests for the detection of drug-resistant Mycobacterium tuberculosis disease, especially in settings with high HIV prevalence [4–7]. Early detection of MDR-TB and XDR-TB is critical to initiate appropriate treatment, reduce morbidity and mortality, and prevent further transmission of drug-resistant strains of TB.

Line probe assays: what is the evidence?Molecular assays to detect gene mutations that signal drug resistance are widely recognized as being most suited for rapid diagnosis, espe-cially since these assays can be directly used on

Daphne I Ling, Alice A Zwerling and Madhukar Pai††Author for correspondenceDepartment of Epidemiology, Biostatistics and Occupational Health, McGill University, Respiratory Epidemiology and Clinical Research Unit, Montréal Chest Institute, 1020 Pine Avenue West, Montreal H3A 1A2, Canada Tel.: +1 514 398 5422Fax: +1 514 398 4503madhukar.pai@mcgill.ca

Growing concerns about the spread of multidrug-resistant tuberculosis (MDR-TB) and the emergence of extensively drug-resistant TB have triggered substantial interest in the development and application of rapid tests for the detection of drug-resistant TB. Molecular assays to detect gene mutations that signal drug resistance are widely recognized as being most suited for rapid diagnosis. Among molecular assays, line probe assays have shown great promise. Currently, two line probe assays are commercially available: the INNO-LiPA® Rif. TB kit (Innogenetics NV, Gent, Belgium) and the GenoType® MTBDRplus assay (Hain Lifescience GmbH, Nehren, Germany). Evidence from a systematic review suggests that INNO-LiPA is a highly sensitive and specific test for the detection of rifampicin resistance in culture isolates. The test, however, appeared to have relatively lower sensitivity when used directly on clinical specimens. Another meta-analysis showed that the GenoType MTBDR assays had excellent accuracy for rifampicin resistance, even when used directly on clinical specimens. While specificity was excellent for isoniazid, sensitivity estimates were modest and variable. Based on evidence and expert opinion, the WHO recently endorsed the use of molecular line probe assays for rapid screening of patients at risk of MDR-TB. Special initiatives have been announced to make these assays accessible and affordable for countries with high MDR-TB prevalence. With strong evidence and new policy directives, the stage is now set for the use of rapid tests for MDR-TB diagnosis. Whether molecular tools, such as line probe assays, will actually make a clinical and public-health impact remains to be determined.

Keywords: diagnostic accuracy • extensively drug-resistant TB • line probe assays • MDR-TB • multidrug-resistant TB • TB • tuberculosis • XDR-TB

Rapid diagnosis of drug-resistant TB using line probe assays: from evidence to policyExpert Rev. Resp. Med. 2(5), 583–588 (2008)

Expert Rev. Resp. Med. 2(5), (2008)584

Perspective

INNO-LiPA® (line-probe assay)

dTTP, dATP, dGTP, dCTPbiotinylated primers

Taq polymerase

DNA target

Thermal cycler

PCR amplification

INNO-LiPA strip

Trough Trough holder

Interpretation

HybridizationStringent washConjugate incubationColor development

Hybridization Conjugate binding EvaluationColor formation

DNA•STRIP®

with specific probes

Alkalinephosphatase

Streptavidine

DNA•STRIPwith developed bands

Genotype

Ling, Zwerling & Pai

clinical specimens, such as sputum [4,5]. Among the molecu-lar assays, line probe assays have shown great promise [4]. Line probe assays are a family of novel DNA strip-based tests that use nucleic acid amplification techniques (e.g., PCR) and reverse hybridization methods for the rapid detection of mutations asso-ciated with drug resistance. A major advantage of line probe assays is that they can be directly used on clinical specimens, such as sputum.

Currently, two line probe assays are commercially available: the INNO-LiPA® Rif. TB kit (Innogenetics NV, Gent, Belgium) (Figure 1) and the GenoType® MTBDRplus assay (Hain Lifescience GmbH, Nehren, Germany) (Figure 2). Since the introduction of these assays, several studies have evaluated the diagnostic accuracy of these tests in various settings. Table 1 provides a brief summary of available evidence, based on two systematic reviews [8,9].

In a systematic review and meta-ana-lysis, Morgan and colleagues summa-rized the evidence on the INNO-LiPA assay for detection of RIF resistance [8]. In 14 studies that applied INNO-LiPA to culture isolates, 12 reported sensitiv-ity estimates greater than 95%, and 12 out of 14 studies reported specificity of 100%. Thus, INNO-LiPA was highly accurate in detecting RIF resistance in culture isolates. By contrast, the four studies that applied INNO-LiPA directly to clinical specimens (i.e., sputum) had 100% specificity and sensitivity estimates ranged between 80 and 100%. Thus, the evidence from this systematic review sug-gested that INNO-LiPA is a highly sen-sitive and specific test for the detection of RIF resistance in culture isolates. The test, however, appears to have relatively lower sensitivity when used directly on clinical specimens.

More recently, Ling and colleagues performed a systematic review and used meta-analysis methods to summarize the evidence on the GenoType MTBDR and MTBDRplus assays for rapid detection

of RIF and INH resistance [9]. They identified 14 comparisons for RIF and 15 comparisons for INH in ten articles that used GenoType MTBDR assays. After meta-ana lysis, the pooled sen-sitivity (98.1%; 95% confidence interval [CI]: 95.9–99.1) and specificity (98.7%; 95% CI: 97.3–99.4) estimates for RIF resist-ance were very high, and consistent across all groups, assay ver-sions (GenoType MTBDR and GenoType MTBDRplus assays) and specimen types (i.e., clinical specimen vs culture isolate). The accuracy for INH was variable, with sensitivity lower (84.3%; 95% CI: 76.6–89.8) and more inconsistent than specificity (99.5%; 95% CI: 97.5–99.9). Thus, the evidence suggested that GenoType MDTBR assays had excellent accuracy for RIF resist-ance, even when used directly on clinical specimens. While spe-cificity was excellent for INH, sensitivity estimates were modest and variable.

Figure 1. Overview of the INNO-LiPA® Rif. TB kit. Image courtesy of Innogenetics NV, Gent, Belgium. Modified with permission, © Innogenetics.

Figure 2. Overview of the GenoType® MTBDRplus assay. Image courtesy of Hain Lifescience GmbH, Nehren, Germany. Modified with permission © Hain Lifescience.

www.expert-reviews.com 585

PerspectiveRapid diagnosis of drug-resistant TB using line probe assaysTa

ble

1. F

ind

ing

s fr

om

sys

tem

atic

rev

iew

s o

f co

mm

erci

al li

ne

pro

be

assa

ys f

or

the

dia

gn

osi

s o

f d

rug

-res

ista

nt

tub

ercu

losi

s.

Focu

s o

f th

e sy

stem

atic

re

view

Test

s ev

alu

ated

St

ud

ies

in t

he

met

a-a

nal

ysis

(n

)Sp

ecim

ens

in t

he

met

a-a

nal

ysis

(n

)St

ud

ies

that

use

d t

he

assa

y d

irec

tly

on

clin

ical

sp

ecim

ens

(n)

Prin

cip

al fi

nd

ing

s an

d c

on

clu

sio

ns

Ref

.

Rap

id d

etec

tion

of

rifa

mpi

cin

resi

stan

ce

INN

O-L

iPA

®

Rif.

TB

assa

y15

3349

412

out

of

14 s

tudi

es a

ppl

ying

LiP

A t

o is

ola

tes

had

a se

nsit

ivit

y gr

eate

r th

an 9

5%, a

nd 1

2 ou

t of

14

had

a sp

ecifi

city

of

100

%; t

he f

our

stud

ies

that

use

d cl

inic

al

spec

imen

s ha

d 10

0%

sp

ecifi

city

, and

sen

siti

vity

ran

ged

fr

om 8

0 to

10

0%

; ove

rall,

the

LiP

A a

ssay

was

hig

hly

accu

rate

whe

n us

ed o

n is

ola

tes;

mor

e ev

iden

ce is

nee

ded

on

its

accu

racy

whe

n us

ed d

irect

ly o

n cl

inic

al s

pec

imen

s

[8]

Rap

id d

etec

tion

of

rifa

mpi

cin

and

ison

iazi

d re

sist

ance

Gen

oTyp

e®

MTB

DR

assa

y an

d G

enoT

ype

MTB

DR

plus

as

say

14 f

or r

ifam

pici

n 15

for

ison

iazi

d17

38

9 fo

r ri

fam

pici

n 10

for

ison

iazi

dTh

e p

oo

led

sens

itiv

ity

(98

.1%

; 95%

CI:

95.9

– 9

9.1)

and

sp

ecifi

city

(9

8.7

%; 9

5% C

I 97.

3–9

9.4

) est

imat

es f

or

rifa

mpi

cin

resi

stan

ce w

ere

very

hig

h an

d co

nsis

tent

acr

oss

al

l gro

ups,

ass

ay v

ersi

ons

and

spec

imen

typ

es; t

he

accu

racy

for

ison

iazi

d w

as v

aria

ble,

wit

h se

nsit

ivit

y lo

wer

(8

4.3

%; 9

5% C

I: 76

.6–8

9.8

) and

mor

e in

cons

iste

nt t

han

spec

ifici

ty (

99.

5%; 9

5% C

I: 97

.5–9

9.9

); o

vera

ll, G

enoT

ype

MD

TBR

assa

ys h

ad e

xcel

lent

acc

urac

y fo

r ri

fam

pici

n re

sist

ance

, eve

n w

hen

used

on

clin

ical

sp

ecim

ens;

whi

le

spec

ifici

ty w

as e

xcel

lent

for

ison

iazi

d, s

ensi

tivi

ty e

stim

ates

w

ere

mo

des

t an

d va

riabl

e

[9]

CI:

Co

nfi

den

ce in

terv

al.

Expert Rev. Resp. Med. 2(5), (2008)586

Perspective Ling, Zwerling & Pai

With overall performance characteristics that are superior to conventional culture and drug susceptibility testing, and the possibility for high throughput with potential cost savings, line probe assays have the potential to greatly enhance MDR-TB diagnosis and treatment. However, evidence on accuracy alone is insufficient to make clinical and public health policies. Available evidence from systematic reviews must be considered together with data from field demonstration projects in real-world set-tings [10]. In fact, field projects are already providing useful data, such as feasibility, affordability, cost and clinical impact [10]. In addition to collecting field data, research is also necessary to develop rapid molecular assays for the diagnosis of XDR-TB. Clearly, the diagnosis of XDR-TB is a tougher challenge, espe-cially since it requires additional data on resistance to several second-line drugs. Unfortunately, second-line drug-resistance

testing is not well standardized and gene mutations for second-line drugs are also not well characterized [11].

From evidence to policyBased on currently available evidence and expert opinion, the WHO recently issued a policy statement on molecular line probe assays for the rapid screening of patients at risk of MDR-TB [101]. In this statement, the WHO endorses the use of line probe assays and provides several guiding principles that need to be consid-ered while adopting the use of these tests in TB programs at the country level (box 1). The policy statement also outlines the agenda for future research.

On 30 June 2008, a major initiative was announced by the WHO, the Stop TB Partnership, UNITAID and the Foundation for Innovative New Diagnostics (FIND) to improve the diagnosis

Box 1. Key elements of the 2008 WHO policy statement on line probe assays.

The use of line probe assays is recommended by the WHO, with the following guiding principles:

Adoption of line probe assays for rapid detection of multidrug-resistant tuberculosis (MDR-TB) should be decided by Ministries of • Health within the context of country plans for appropriate management of MDR-TB patients, including the development of country-specific screening algorithms and timely access to quality-assured second-line anti-TB drugs.

Line probe assay performance characteristics have been adequately validated in direct testing of sputum smear-positive specimens and • on isolates of Mycobacterium tuberculosis complex grown from smear-negative and smear-positive specimens. Direct use of line probe assays on smear-negative clinical specimens is not recommended.

The use of commercial line probe assays, rather than in-house assays, is recommended to ensure reliability and reproducibility of results, • as in-house assays have not been adequately validated or used outside limited research settings. Any new or generic line probe assays should be subject to adequate validation, that is, published laboratory validation studies, adequate data to allow systematic review and meta-ana lysis (including assessment of data quality), and results from field demonstration projects documenting feasibility and consistent performance equal to conventional methods and commercial line probe assays.

Adoption of line probe assays does not eliminate the need for conventional culture and drug-susceptibility testing (DST) capability; • culture remains necessary for definitive diagnosis of TB in smear-negative patients, while conventional DST is required to diagnose extensively drug-resistant TB (XDR-TB). However, the demand for conventional culture and DST capacity may change, based on the local epidemiological situation and the use of line probe assays in country-specific screening algorithms.

As current line probe assays only detect resistance to rifampicin and/or isoniazid, countries with documented or suspected cases of • XDR-TB should establish or expand conventional culture and DST capacity for quality-assured susceptibility testing of second-line drugs, based on current WHO policy guidance.

Adoption of line probe assays for the rapid detection of MDR-TB should be phased in, starting at national/central reference laboratories • or those with a proven capability to conduct molecular testing. Once this has been accomplished, expansion could be considered within the context of country laboratory strengthening plans, and considering availability of suitable personnel in peripheral centers, quality of specimen transport systems and country capability to provide appropriate treatment and management of MDR-TB patients once diagnosed.

Adequate and appropriate laboratory infrastructure and equipment should be provided, ensuring that the required precautions for • biosafety and prevention of contamination are met.

Appropriate laboratory staff should be trained to conduct line probe assay procedures, especially those relating to amplification and • interpretation of results. Supervision of staff by a senior individual with adequate training and experience in molecular assays is strongly recommended.

A detailed commercial sales contract and customer support plan should be negotiated with manufacturers, guaranteeing an ample and • continuous supply of materials, appropriate shipment conditions, customs clearance, equipment installation, maintenance, repair and replacement, and provision of training and ongoing technical support.

Stringent laboratory protocols, standard operating procedures for molecular line probe assays and internal quality control mechanisms • must be implemented and enforced. A program for external quality assessment of laboratories involved in line probe assays should be developed as a matter of priority.

Mechanisms for rapid reporting of line probe assays results to clinicians must be established to provide patients with the benefit of an • early diagnosis.

The WHO and partners should assist countries with operational plans to introduce line probe assays within the appropriate • epidemiological and resource availability context.

Source: WHO 2008 [101].

www.expert-reviews.com 587

PerspectiveRapid diagnosis of drug-resistant TB using line probe assays

Key issues

An estimated 400,000 new cases of multidrug-resistant tuberculosis (MDR-TB) occur each year and, although their exact incidence • rates are currently unknown, extensively drug-resistant TB (XDR-TB) cases have been detected in every country where there is the capacity to detect them.

Molecular assays to detect gene mutations that signal drug resistance are widely recognized as being most suited for rapid diagnosis • of MDR-TB.

Among the molecular assays, line probe assays have shown great promise. Currently, two line probe assays are commercially available: • the INNO-LiPA® Rif. TB kit (Innogenetics NV, Gent, Belgium) and the GenoType® MTBDRplus assay (Hain Lifescience GmbH, Nehren, Germany).

Evidence from a systematic review showed that INNO-LiPA is a highly sensitive and specific test for the detection of rifampicin • resistance in culture isolates. The test, however, appeared to have relatively lower sensitivity when used directly on clinical specimens.

A recent meta-ana lysis showed that MDTBR assays had excellent accuracy for rifampicin resistance, even when used directly on clinical • specimens. While specificity was excellent for isoniazid, sensitivity estimates were modest and variable.

Based on currently available evidence and expert opinion, the WHO recently endorsed the use of molecular line probe assays for rapid • screening of patients at risk of MDR-TB.

Special initiatives have been announced to make line probe assays accessible and affordable for resource-limited countries.•

and treatment of MDR-TB in resource-limited countries [102]. As part of this initiative, over the next few years, 16 countries will begin using rapid methods to diagnose MDR-TB, including line probe assays. These countries will receive the tests at specially negotiated prices through the Stop TB Partnership’s Global Drug Facility, which provides countries with both drugs and diagnostic supplies [102].

Expert commentary & five-year viewWith strong evidence, new policy directives and involvement of funding agencies, the stage appears to be set for the use of rapid tests for MDR-TB diagnosis. In the next 5 years, the use of rapid molecular assays is expected to increase, especially in countries that are dealing with a dual burden of HIV and MDR-TB. Usually, cost considerations would have greatly limited the use of molecular assays in poor countries. However, with the negotiated pricing and special funding initiatives from agencies such as the WHO, Stop TB Partnership, FIND, UNITAID and industry

partners, the situation looks hopeful. In the next 5 years, it will become evident if molecular tools, such as line probe assays, can actually make a clinical and public-health impact in the real world. In the meantime, programmatic field studies and cost–effectiveness data are needed to better understand the real world implications of the new policy [12] and to enable evidence-based MDR-TB diagnosis.

Financial & competing interests disclosureM Pai is an external consultant for the FIND, Geneva, Switzerland, a nonprofit agency that partners with several industries (including Hain LifeScience, Germany) for the development and evaluation of new diag-nostic tools for neglected infectious diseases. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

ReferencesPapers of special note have been highlighted as:• of interest

World Health Organization. Global 1

tuberculosis control. Surveillance, planning, financing. WHO Report 2007. WHO/HTM/TB/2007.376. World Health Organization, Geneva, Switzerland, 1–242 (2007).

World Health Organization. The Global 2

MDR-TB and XDR-TB Response Plan, 2007–2008. WHO/HTM/TB/2007.387. World Health Organization, Geneva, Switzerland (2007).

Zignol M, Hosseini MS, Wright A3 et al. Global incidence of multidrug-resistant tuberculosis. J. Infect. Dis. 194(4), 479–485 (2006).

Migliori GB, Matteelli A, Cirillo D, Pai M. 4

Diagnosis of multidrug-resistant

tuberculosis and extensively drug-resistant tuberculosis: current standards and challenges. Can. J. Infect. Dis. Med. Microbiol. 19(2), 169–172 (2008).

Pai M, Kalantri S, Dheda K. New tools 5

and emerging technologies for the diagnosis of tuberculosis: part 2. Active tuberculosis and drug resistance. Expert Rev. Mol. Diagn. 6(3), 423–432 (2006).

Migliori GB, Loddenkemper R, Blasi F, 6

Raviglione MC. 125 years after Robert Koch’s discovery of the tubercle bacillus: the new XDR-TB threat. Is “science” enough to tackle the epidemic? Eur. Respir. J. 29(3), 423–427 (2007).

Perkins MD, Cunningham J. Facing the 7

crisis: improving the diagnosis of tuberculosis in the HIV era. J. Infect. Dis. 196(Suppl. 1), S15–S27 (2007).

Morgan M, Kalantri S, Flores L, Pai M. A 8

commercial line probe assay for the rapid detection of rifampicin resistance in Mycobacterium tuberculosis: a systematic review and meta-ana lysis. BMC Infect. Dis. 5, 62 (2005).

Synthesized published evidence on the •INNO-LiPA Rif. TB assay.

Ling DI, Zwerling A, Pai M. GenoType 9

MTBDR assays for the diagnosis of multidrug-resistant tuberculosis: a meta-ana lysis. Eur. Respir. J. DOI 10.1183/09031936.00061808 (2008) (Epub ahead of print).

Synthesized published evidence on the •GenoType® MTBDR assay.

Barnard M, Albert H, Coetzee G, 10

O’Brien R, Bosman ME. Rapid molecular screening for MDR TB in a high volume public health laboratory in South Africa.

Expert Rev. Resp. Med. 2(5), (2008)588

Perspective Ling, Zwerling & Pai

Am. J. Respir. Crit. Care Med. (2008).

Large-scale field evaluation study of the •GenoType MTBDR assay in South Africa.

Kim SJ. Drug-susceptibility testing in 11

tuberculosis: methods and reliability of results. Eur. Respir. J. 25(3), 564–569 (2005).

Pai M, Ramsay A, O’Brien R. Evidence-12

based tuberculosis diagnosis. PLoS Med. 5(7), e156 (2008).

Recent ana lysis on how evidence on •various tuberculosis diagnostics is influencing policies, clinical practice and research agendas.

Websites

World Health Organization. Policy 101

statement. Molecular line probe assays for rapid screening of patients at risk of multidrug-resistant tuberculosis (MDR-TB). World Health Organization, Geneva, Switzerland, 2008 www.who.int/tb/features_archive/policy_statement.pdf

New policy from the WHO, endorsing •the use of line probe assays.

World Health Organization. Rapid tests for 102

drug-resistant TB to be available in developing countries. World Health Organization, Geneva, Switzerland, 2008 www.who.int/tb/features_archive/mdrtb_rapid_tests/en/index.html

AffiliationsDaphne I Ling, MPH • Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Respiratory Epidemiology and Clinical Research Unit, Montréal Chest Institute, 1020 Pine Avenue West, Montreal H3A 1A2, Canada Tel.: +1 514 934 1934 ext. 32156 Fax: +1 514 398 4503 daphne.ling@mail.mcgill.ca

Alice A Zwerling, MSc • Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Respiratory Epidemiology and Clinical Research Unit, Montréal Chest Institute, 1020 Pine Avenue West, Montreal H3A 1A2, Canada Tel.: +1 514 934 1934 ext. 32466 Fax: +1 514 398 4503 alice.zwerling@mail.mcgill.ca

Madhukar Pai, MD, PhD • Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Respiratory Epidemiology and Clinical Research Unit, Montréal Chest Institute, 1020 Pine Avenue West, Montreal H3A 1A2, Canada Tel.: +1 514 398 5422 Fax: +1 514 398 4503 madhukar.pai@mcgill.ca