INT J TUBERC LUNG DIS 19(3):339–341

Q 2015 The Union

http://dx.doi.org/10.5588/ijtld.14.0814

Concordance of Mycobacterium tuberculosis fluoroquinolone

resistance testing: implications for treatment

M. R. Farhat,*† C. D. Mitnick,†‡ M. F. Franke,† D. Kaur,§ A. Sloutsky,§ M. Murray,†¶ K. R. Jacobson#**

*Division of Pulmonary and Critical Care Medicine, Massachusetts General Hospital, Boston, †Department of
Global Health and Social Medicine, Harvard Medical School, Boston, ‡Department of Global Health Equity,
Brigham and Women’s Hospital, Boston, §University of Massachusetts Medical School, Massachusetts
Supranational TB Reference Laboratory, Boston, ¶Department of Epidemiology, Harvard School of Public Health,
Boston, #Section of Infectious Diseases, Boston University School of Medicine, Boston, Massachusetts, USA;
**Department of Science & Technology, National Research Foundation Centre of Excellence for Biomedical TB
Research/Medical Research Council Centre for Molecular and Cellular Biology, Division of Molecular Biology and
Human Genetics, Faculty of Medicine and Health Sciences, Stellenbosch University, Tygerberg, South Africa

SUMMARY

Fluoroquinolone (FQ) drug susceptibility testing (DST)
is an important step in the design of effective treatment
regimens for multidrug-resistant tuberculosis. Here we
compare ciprofloxacin, ofloxacin and moxifloxacin
(MFX) resistance results from 226 multidrug-resistant
samples. The low level of concordance observed suggests
that DST should be performed for the specific FQ
planned for clinical use. The results also support the new
World Health Organization recommendation for testing
MFX at a critical concentration of 2.0 lg/ml.
K E Y W O R D S : drug susceptibility testing; multidrug-
resistant tuberculosis; critical concentration

FLUOROQUINOLONES (FQs) are among the most
effective drugs available for the treatment of multi-
drug-resistant tuberculosis (MDR-TB).1 There is,
however, uncertainty about how to interpret results
from drug susceptibility testing (DST) against FQs.
Although the third- and fourth-generation agents
(e.g., levofloxacin and moxifloxacin [MFX]) are
considered to have the most activity against Myco-
bacterium tuberculosis, in vitro FQ DST can still be
performed against second-generation FQs (e.g.,
ofloxacin [OFX] and ciprofloxacin [CFX]). Resis-
tance to OFX or CFX is inferred to mean resistance to
any FQ.2 This extrapolation of resistance across the
class of FQs is implicit in how extensively drug-
resistant TB (XDR-TB) is defined, i.e., MDR-TB that
is resistant to an injectable agent and any FQ.
However, the World Health Organization (WHO)
now recommends against extrapolating resistance
results from second- to third-generation FQs.3
The interpretation of DST results is further
complicated by the need to choose a critical
concentration (CC) for testing. If bacterial growth
occurs in the presence of a drug at this CC, the
bacteria are considered to be resistant. Currently used
CCs are often lower than the serum concentrations
achievable for the drug. In some cases, this means that
isolates are classified as drug-resistant when they may
still respond to treatment.4 In the case of MFX, data
on the optimal CC are limited.5 A CC of 0.125–0.5
lg/ml for MFX is widely used both to guide patient
care and in clinical trials.2,6 In 2012, based on the
experience of several supranational reference labora-
tories, the WHO proposed an interim recommenda-
tion to increase the CC of MFX from 0.5 to 2.0 lg/ml;
however, the WHO did not proceed with formal
policy guidance recommending the revised concen-
trations due to the limited data available.3 A more
refined resistance measure is the minimum inhibitory
concentration (MIC) of the FQ, which involves
testing at a series of increasing drug concentrations.
MICs are seldom measured clinically due to the cost
and labor involved.
To determine whether early-generation DST results
can predict resistance to later-generation FQs, we
compared the DST results for three FQs: CFX, OFX
and MFX. We also compared the distribution of
MICs for MFX to achievable serum drug concentra-
tions to assess the validity of the current standard CC.
METHODS
Using an archive of M. tuberculosis samples from
patients referred for individualized MDR-/XDR-TB
treatment in Lima, Peru, between 1 February 1997

Correspondence to: Maha R Farhat, Department of Global Health and Social Medicine, Harvard Medical School. 641
Huntington Avenue Suite 4A, Boston, MA 02115, USA. Tel: (þ1) 617 432 6783. Fax: (þ1) 617 432 2565. e-mail: mrfarhat@
partners.org
Article submitted 22 October 2014. Final version accepted 26 November 2014.
340 The International Journal of Tuberculosis and Lung Disease

Table Resistance classification and concentrations tested for three fluoroquinolones using the agar indirect proportion method on
7H10 media
Drug tested lg/ml Concentrations tested Peak drug serum concentration Susceptible Intermediate resistance Resistant
4
OFX MIC 1.0, 2.0, 4.0, 6.0, 8.0, 10.0 2–10 ,2.0 2.0 74.0
MFX MIC 0.125, 0.25, 0.5, 1.0, 4.0, 8.0 4–97 ,0.5 0.5,1.0 74.0
CFX* 1.0 or 2.0 — 61 or 62 NA .1 or .2
* Isolates were tested at a critical concentration of 2.0 lg/ml before 2001; from 2001 onwards the critical concentration was reduced to 1.0 lg/ml.
OFX ¼ ofloxacin; MIC ¼ minimum inhibitory concentration; MFX ¼ moxifloxacin; CFX ¼ ciprofloxacin; NA ¼ not available.

and 31 July 2003,1 we selected isolates that had agreement in 69% (156/226) of the samples. Of the
undergone DST for CFX using the indirect agar remaining 70 (31%) samples, 56% of the discrepan-
proportion method on 7H10 media (Table). In 2001, cies (39/70) were among isolates with high-level
the testing laboratory altered its standard CC for resistance to OFX but intermediate resistance to
CFX, reducing it from 2 lg/ml to 1 lg/ml. We MFX. Of the 226 isolates, 30% had MFX MICs in
randomly selected 175 CFX-resistant and 100 CFX- the intermediate range (Figure, B), in contrast to only
susceptible isolates from this archive for OFX and 5% of the isolates with intermediate OFX MIC. One
MFX MIC testing (Table) using the indirect propor- third (22/67) of the MFX-intermediate isolates had
tion method on 7H10 agar. Cut-offs for susceptibility an OFX MIC , 2 lg/ml, which explains most of the
and resistance were chosen based on peak serum other discrepancies (22/70, 31%) between the OFX
concentration data (Table), and the prevalent CCs of and MFX MIC results (Figure, B).
2.0 lg/ml for OFX and 0.25–0.50lg/ml for MFX.8,9 We determined the sensitivity and specificity of the
We excluded 49 isolates from the analysis that were
MIC of 0.5 lg/ml for MFX as a proxy for the OFX
found to be previous samples from the same patient.
MIC of 2.0 lg/ml, in the light of the recent WHO
Of the 45 total isolates that were CFX-resistant but
recommendation to use these interchangeably.3 We
MFX- and OFX-susceptible, 12 (25%) were random-
determined the sensitivity to be 89.1% (66/74 strains
ly selected for repeat MIC testing and 10 were
with OFX MIC . 2.0 have a MFX MIC . 0.5) and
reconfirmed as MFX- and OFX-susceptible. For two
isolates, the measured MIC for MFX increased from the specificity to be 92.8% (141/152 strains with
0.25 lg/ml to 0.5 lg/ml. In eight randomly selected OFX MIC 6 2 have an MFX MIC 6 0.5).
concordant isolates, repeat testing confirmed the
exact MIC, except in two isolates: for the first, the
MIC for MFX changed from 60.125 to 0.25 and for
the second the MIC for MFX decreased from 8 to 4
and the MIC for OFX from 8 to 6. The repeat MIC
results were retained for all data analysis.
R software, version 3.1.0 (R Computing, Vienna,
Austria) and Fisher’s exact test were used for
statistical testing. This study was deemed not to
constitute human subjects research by the Partners
Human Research Committee (Boston, MA, USA).

RESULTS
Of the 226 MDR-TB patient isolates analyzed, 153
were CFX-resistant and 73 were susceptible. Of
these, 56% (86/153) of CFX-resistant isolates had
an MIC for OFX in the intermediate or resistant
range, and 67% of the 153 CFX-resistant isolates had
an MIC in the intermediate or resistant range for
MFX (Figure, A). The proportion of strains that were Figure DST results. A) Venn diagram of a subset of the DST
intermediate or resistant was not different before and results showing overlap between CFX, MFX and OFX resistance
after 2001, when the CFX CC was reduced from 2.0 (number outside the circle represents the number of tested
to 1.0 lg/ml (Fisher P ¼ 0.7 for CFX/MFX and 0.8 for isolates susceptible to all three drugs). B) Cross-tabulation of
CFX/OFX). None of the 73 CFX-susceptible strains MFX and OFX MIC results, all drug concentrations are in lg/ml.
Numbers in the cells represent strain counts (n ¼ 226 strains).
was fully resistant to MFX or OFX, while three were CFX ¼ ciprofloxacin; R ¼ resistant; OFX ¼ ofloxacin; I/R ¼
of intermediate MIC to one of these two drugs. intermediate resistance; MFX ¼ moxifloxacin; MIC ¼ minimum
The MFX and OFX MIC results were in full inhibitory concentration; DST ¼ drug susceptibility testing.

DISCUSSION
Concordance in resistance among the FQ agents
tested was lower than expected, with one third to half
of the strains showing no agreement among the three
agents. The discrepancy between the in vitro resis-
tance results of CFX, OFX and MFX, and in
particular the high rates of CFX/OFX-resistant
isolates that were intermediate or susceptible to
MFX, confirm findings from previous smaller stud-
ies,10 and could account for the continued clinical
efficacy of later generation FQs in XDR-TB treat-
ment.11 This discordance among agents within the
FQ class calls into question the value of defining
XDR-TB as TB resistant to one or more FQ rather
than TB that is resistant to all—or at least the most
active—class members. The latter definition will
arguably carry better prognostic information.
We also found that a substantial proportion (30%)
of strains had MFX MICs in the intermediate range.
Peak serum concentrations for MFX are reported to
be upwards of 4 lg/ml,7 and although serum
concentration may not directly correlate with the
intracellular concentrations needed for TB bacteri-
cidal activity, the high serum concentrations may
indicate that intermediate isolates (MIC 0.5–1.0 lg/
ml) can be clinically treatable with standard doses of
MFX (400 mg/day).2 This finding should be support-
ed with the direct measurement of outcomes in
patients infected with MFX-intermediate isolates.
This finding is consistent with previous reports,2 and
supports the recent interim change in the WHO
guidelines, suggesting the use of a CC of MFX of 2.0
lg/ml. In sum, our results indicate that CFX or OFX
DST results should not be used clinically to guide the
use of MFX and support the WHO recommendation
to test MFX at a critical concentration of 2.0 lg/ml.
Acknowledgements
The authors would like to thank the Peruvian team Socios En Salud
for their patient care and for providing the clinical isolates that
made this study possible.
This work was supported by the Harvard Center for AIDS
Research (CFAR), a National Institutes of Health (NIH) funded
program (P30 AI060354), which is supported by the following NIH
Co-Funding and Participating Institutes and Centers: National
Institute of Allergy and Infectious Diseases (NIAID), National
Cancer Institute, National Institute of Child Health and Human
Development, National Institute on Drug Abuse, National Institute
of Mental Health, National Institute on Aging, National Institute
of Diabetes and Digestive and Kidney Diseases, National Institute
TB FQ cross-resistance 341
of General Medical Sciences, Fogarty International Center (FIC),
and Office of AIDS Research (Bethesda, MD, USA). This work was
also supported by the American Lung Association (Washington
DC, USA) Biomedical Research Grant (RG-270912-N) and the
Parker B. Francis Fellowship (MRF), the NIH/FIC (K01
TW009213 KRJ) and NIH NIAID (U19-AI109755, MM). The
content is solely the responsibility of the authors and does not
necessarily represent the official views of the NIH.
Conflicts of interest: none declared.
References
1 Mitnick C D, Franke M F, Rich M L, et al. Aggressive regimens
for multidrug-resistant tuberculosis decrease all-cause
mortality. PLOS ONE 2013; 8: e58664.
2 Sirgel F A, Warren R M, Streicher E M, Victor T C, van Helden
P D, Böttger E C. gyrA mutations and phenotypic susceptibility
levels to ofloxacin and moxifloxacin in clinical isolates of
Mycobacterium tuberculosis. J Antimicrob Chemother 2012;
67: 1088–1093.
3 World Health Organization. Companion handbook: to the
WHO guidelines for the programmatic management of drug-
resistant tuberculosis. WHO/HTM/TB/2014.11. Geneva,
Switzerland: WHO, 2014. http://apps.who.int/iris/bitstream/
10665/130918/1/9789241548809_eng.pdf?ua¼1&ua¼1
Accessed December 2014.
4 Böttger E C. The ins and outs of Mycobacterium tuberculosis
drug susceptibility testing. Clin Microbiol Infect 2011; 17:
1128–1134.
5 Horne D J, Pinto L M, Arentz M, et al. Diagnostic accuracy and
reproducibility of WHO-endorsed phenotypic drug
susceptibility testing methods for first-line and second-line
anti-tuberculosis drugs. J Clin Microbiol 2013; 51: 393–401.
6 Gillespie S H, Crook A M, McHugh T D, et al. Four-month
moxifloxacin-based regimens for drug-sensitive tuberculosis. N
Engl J Med 2014; 371: 1577–1587.
7 Peloquin C A, Hadad D J, Molino L P D, et al. Population
pharmacokinetics of levofloxacin, gatifloxacin, and
moxifloxacin in adults with pulmonary tuberculosis.
Antimicrob Agents Chemother 2008; 52: 852–857.
8 Isaeva Y, Bukatina A, Krylova L, Nosova E, Makarova M,
Moroz A. Determination of critical concentrations of
moxifloxacin and gatifloxacin for drug susceptibility testing
of Mycobacterium tuberculosis in the BACTEC MGIT 960
system. J Antimicrob Chemother 2013; 68: 2274–2281.
9 World Health Organization Stop TB Partnership. Global
Laboratory Initiative Mycobacteriology Laboratory Manual.
Geneva, Switzerland: WHO, 2014. gli_mycobacteriology_lab_
manual_web.pdf Accessed December 2014.
10 Sulochana S, Rahman F, Paramasivan C N. In vitro activity of
fluoroquinolones against Mycobacterium tuberculosis. J
Chemother 2005; 17: 169–173.
11 Jacobson K R, Tierney D B, Jeon C Y, Mitnick C D, Murray M
B. Treatment outcomes among patients with extensively drug-
resistant tuberculosis: systematic review and meta-analysis.
Clin Infect Dis 2010; 51: 6–14.

Le test de sensibilité à la fluoroquinolone (FQ) est une
étape importante dans la conception de protocoles de
traitement efficace de la tuberculose multirésistante (TB-
MDR). Nous comparons ici les résultats des tests de
résistance à la ciprofloxacine, à l’ofloxacine et à la
moxifloxacine (MFX) de 226 échantillons TB-MDR. Le
faible niveau de concordance observé confirme que les
Las pruebas de sensibilidad (DST) a las fluoroquinolonas
(FQ) constituyen una etapa importante en la elaboración
de pautas terapéuticas eficaces contra la tuberculosis
multidrogorresistente. En el presente estudio se
compararon los resultados de resistencia a
ciprofloxacino, ofloxacino y moxifloxacino (MFX) de
226 muestras de casos multidrogorresistentes. El bajo
TB FQ cross-resistance i
RESUME
tests de sensibilité devraient être réalisés pour la FQ
spécifique de l’utilisation clinique prévue. Les résultats
sont également en accord avec les nouvelles
recommandations de l’Organisation Mondiale de la
Santé de tester la MFX à la concentration critique de 2,0
lg/ml.
RESUMEN
grado de concordancia observado respalda la noción de
que se deben practicar DST para cada FQ, cuyo uso
clı́nico se prevé en la práctica clı́nica. Estos resultados
respaldan también la nueva recomendación de la
Organización Mundial de la Salud de usar una
concentración discriminatoria de 2,0 lg/ml en las DST
al MFX.