INT J TUBERC LUNG DIS 18(3):357–362 © 2014 The Union http://dx.doi.org/10.5588/ijtld.13.0459

Diagnostic dilemma: treatment outcomes of tuberculosis patients with inconsistent rifampicin susceptibility Y. Pang,* Y-Z. Ruan,* J. Zhao,* C. Chen,* C-H. Xu,* W. Su,* S-T. Huan,† R-Z. Li,* Y-L. Zhao,* D. P. Chin,† L-X. Wang* * National Center for Tuberculosis Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing, † Bill & Melinda Gates Foundation, China Office, Beijing, China SUMMARY O B J E C T I V E : A retrospective clinical trial to evaluate treatment outcomes in adults with smear-positive tuberculosis (TB) and discordant rifampicin (RMP) resistance results. D E S I G N : A total of 2156 smear-positive TB patients underwent both conventional and Genechip® drug susceptibility testing (DST) for RMP resistance. All 49 patients with discordant results treated with either a firstline or second-line regimen were analysed. R E S U LT S : Of 30 Type I cases (Genechip-resistant, conventional DST-susceptible) receiving the first-line regimen, 4 had a favourable outcome and 5 failed treatment. The 21 remaining Type I cases were treated with the second-line regimen, of whom 18 had a favourable out-

come. Second-line regimen thus resulted in significantly more favourable outcomes than first-line treatment (P = 0.032). Among Type II cases (Genechip-susceptible, conventional DST-resistant), 13/19 received the first-line regimen, and 7 had a favourable outcome. The six Type II cases treated with the second-line regimen all had favourable outcomes. C O N C L U S I O N : Patients with discordant RMP DST results who receive second-line regimens may have a better clinical response than those treated with the first-line regimen. Patients infected with fluoroquinolone-resistant Mycobacterium tuberculosis strains were observed to have a significantly higher treatment failure rate. K E Y W O R D S : tuberculosis; rifampicin; clinical outcome

TUBERCULOSIS (TB) is one of the most widespread infectious diseases in the world.1,2 According to the World Health Organization (WHO), there are approximately 9 million new TB cases every year, with more than 95% occurring in developing countries.3–5 Although the introduction of anti-tuberculosis drugs has improved outcomes for TB patients over the past decade, the emergence of drug-resistant TB, particularly multidrug-resistant TB (MDR-TB, defined as resistance to at least isoniazid [INH] and rifampicin [RMP]), has become a major obstacle to global TB control.4,6 China is one of the 27 high MDR-TB burden countries, with an MDR-TB prevalence of 5.7% and 25.6% in new and previously treated cases, respectively, according to a recent national survey.7 Laboratory diagnosis of drug-resistant TB by culture-based conventional drug susceptibility testing (DST) is considered the gold standard.8,9 However, due to the slow growth of the tubercle bacillus, this method usually takes >2 months to yield results.8 The cost of biosafety level 2 and 3 laboratories and the need for highly skilled staff are other disadvan-

tages of conventional DST.10 The high incidence of drug-resistant TB and the shortcomings of conventional DST highlight the need for rapid and accurate DST methods to help clinicians prescribe optimal therapeutic regimens that will prevent further drug resistance.5,11,12 Molecular DST techniques provide an excellent solution for overcoming the challenges of conventional culture-based methods.5 Several studies have reported that Mycobacterium tuberculosis drug resistance arises via the selection of subpopulations with spontaneously occurring mutations in particular genes.13,14 The mutations that cause drug resistance most often affect the susceptibility of the bacterium to the most important first-line anti-tuberculosis agents, RMP and INH.6,15 A number of commercial molecular kits have been developed to rapidly diagnose M. tuberculosis susceptibility to RMP and INH, the sensitivity and specificity of which are approximately 90% and 95% for RMP resistance, and 80% and 95% for INH resistance, respectively.5,16–19 Although these tools have demonstrated performance in detecting RMP and INH resistance, inconsistencies are sometimes found when comparing molecular and conventional DST

YP, YZR, JZ and CC contributed equally to this study.

Correspondence to: Renzhong Li, National Center for Tuberculosis Control and Prevention, Chinese Center for Disease Control and Prevention, No 155 Changbai Road, Changping District, Beijing. Fax: (+86) 10 5890 0533. Tel: (+86) 10 5890 0533. e-mail: [email protected] Article submitted 26 June 2013. Final version accepted 11 November 2013.

358

The International Journal of Tuberculosis and Lung Disease

methods. As treatment regimens are based on DST, it is important to determine the best regimen for patients with discordant tests. To resolve this problem, we conducted a retrospective clinical trial to evaluate treatment outcomes in adults with smear-positive TB who had conflicting results for RMP resistance using a molecular tool and conventional DST. Our aim was to provide evidence for the best treatment regimen for these patients.

MATERIALS AND METHODS Patients This study reviewed the results of adult TB patients treated at a local TB dispensary or specialised hospital. Sputum samples from all patients were analysed using both conventional DST and Genechip® (Capital Bio, Beijing, China), a molecular diagnostic tool.18 Conventional DST against four first-line antituberculosis drugs (INH, RMP, ethambutol [EMB] and streptomycin [SM]) and two second-line antituberculosis drugs (kanamycin [KM] and ofloxacin [OFX]) was performed as previously reported.6 The concentrations of the drugs in medium were INH 0.2 μg/ml, RMP 40 μg/ml, EMB 2 μg/ml, SM 4 μg/ml, KM 30 μg/ml and OFX 2 μg/ml. A strain was declared resistant to the specific drug when the growth rate exceeded 1% compared to the control. Genechip was performed according to the manufacturer’s instructions.18 Based on these two methods, patients with discordant laboratory-documented TB susceptibility to RMP were included in the study. The fragment of the rpoB gene containing the RMP resistance-determining region from M. tuberculosis strains harbouring discordant RMP results was amplified and sequenced to confirm the Genechip results. The primer pair and reaction programme were used as previously reported.6 Per the study design, all of the RMP-resistant patients diagnosed by either conventional DST or Genechip should have received the second-line regimen. However, some patients opted for the firstline regimen provided free of charge rather than the second-line regimen due to personal reasons. All patients completed and signed an informed consent form in their local language. Both treatment regimens closely followed WHO recommendations.3 Treatment regimens and monitoring Patients were classified as Type I if they were Genechip-resistant and conventional DST-susceptible, while Type II patients were Genechip-susceptible and conventional DST-resistant. Both Type I and Type II patients received either first-line or second-line treatment regimens; there were thus four treatment groups. As shown in Table 1, patients were treated with one of two anti-tuberculosis regimens. The first-line regimen consisted of INH, RMP, pyrazinamide (PZA)

Table 1 Treatment regimens for Type I and Type II patients* with inconsistent RMP susceptibility Regimen First-line Second-line

Intensive phase

Continuation phase

2HRZE† 2HRZES 6ZAmk(Cpm) Lvx(Mfx)PAS(E)Pth

4HR 6HRE 18ZLvx(Mfx) PAS(E)Pth

* Type I cases: Genechip-resistant, conventional DST-susceptible; Type II cases: Genechip-susceptible, conventional DST-resistant. † Numbers indicate the duration of the phase of treatment in months. RMP, R = rifampicin; H = isoniazid; Z = pyrazinamide; E = ethambutol; S = streptomycin; Amk = amikacin; Cpm = capreomycin; Lvx = levofloxacin; Mfx = moxifloxacin; PAS = para-aminosalicylic acid; Pth = prothionamide; DST = drug susceptibility testing.

and ethambutol, supplemented by SM if the patient was a retreatment case. The second-line regimen consisted of a fluoroquinolone (levofloxacin or moxifloxacin), amikacin (or capreomycin), PZA, paraaminosalicylic acid (or EMB) and prothionamide as the core drugs, modified according to medication history and adverse reactions. Each regimen had an intensive phase and a continuation phase of treatment. Sputum smears were taken and solid cultures were carried out periodically during treatment. A favourable treatment outcome was defined as completion of treatment without relapse over the 2-month followup period for the first-line regimen, and sputum conversion at 12 months for the second-line regimen. Treatment failure was defined as sputum smear or culture positivity at month 5 or later during treatment, or death during treatment. Relapses were defined as patients who had previously been treated for TB, had been declared cured at the end of their most recent course of treatment, and now had been diagnosed with a recurrent episode of TB. Statistical analysis Demographic and clinical factors that might be associated with treatment outcome were obtained from the patient records. Fisher’s exact test and the χ2 test were used to evaluate associations among multiple categorical variables, and results were expressed as odds ratios with 95% confidence intervals. Statistical analysis was performed using SPSS 13.0 (Statistical Product and Service Solutions, Chicago, IL, USA). Differences with P ⩽ 0.05 were considered statistically significant.

RESULTS Study population Patients were enrolled from January to December 2011. A total of 2156 smear-positive TB cases were analysed using both Genechip and conventional DST, 52 of whom were discordant, i.e., the RMP susceptibility results determined using the two methods were different. These patients were included in the

TB patients with inconsistent RMP susceptibility

Table 2

359

Statistics analysis of clinical outcomes of cases Treatment outcomes Patient Type I Patient Type II Molecular tool: resistant Molecular tool: Conventional tool: susceptible susceptible Conventional tool: resistant

Regimen First-line Second-line Total P value

Figure 1 Flow diagram for the diagnosis of cases with discordant rifampicin susceptibility results. TB = tuberculosis.

retrospective clinical trial. Three patients were excluded during data analysis due to default (Figure 1). Treatment outcomes Of the 49 TB strains isolated from patients available for analysis, 30 were Type I, as defined in the Methods. Of these 30 cases, 9 received the first-line regimen and 21 received the second-line regimen. Four of the nine cases on the first-line regimen had fa-

Favourable Adverse Total Favourable Adverse Total 4 18 22

5 3 8

9 21 30

0.032

7 6 13

6 0 6

13 6 19

0.109

vourable outcomes, while five failed treatment. Of the 21 cases treated with the second-line regimen, 18 had a favourable outcome, while three were treatment failures (Figure 2). As shown in Table 2, the second-line regimen resulted in significantly more favourable outcomes among Type I patients than the first-line regimen (P = 0.032). Among the 19 Type II patients, 13 received the firstline and six the second-line regimen (Figure 2). Seven of the 13 cases on the first-line regimen had favourable outcomes, while six failed treatment. All six cases on the second-line regimen had a favourable outcome. However, as shown in Table 2, there was no statistical difference in the frequency of favourable treatment outcomes between the two regimens (P = 0.109). We attribute this to the small number of cases. We also compared the clinical outcome of 51 patients with concordant results (both resistant) with those of Type I and Type II patients given the secondline regimen. Among these 51 patients, 35 cases on the second-line regimen had a favourable outcome,

Figure 2 Treatment outcomes of cases with discordant rifampicin susceptibility results.

360

The International Journal of Tuberculosis and Lung Disease

while 16 failed treatment. Statistical analysis showed no difference in the frequency of favourable treatment outcomes between the concordant group (both resistant) and the Type I (P = 0.135) and Type II (P = 0.170) patients.

Table 4 Relationship between second-line drug susceptibility and clinical outcome

Risk factors influencing treatment outcomes As shown in Table 3, the percentage of favourable outcomes for Type I and Type II patients combined differed by age group. Compared with the percentage of favourable outcomes in the 45–64 year age group, patients aged