591. Multi-center study to assess the non-inferiority of Nipro NTM+MDRTB and Hain GenoType MTBDRplus V2 line probe assays compared to Hain GenoType MTBDRplus V1
Session: Poster Abstract Session: TB: Mycobacteria Diagnostic Testing
Thursday, October 8, 2015
Room: Poster Hall
Posters
  • LPA poster final.jpg (166.9 kB)
  • Background:

    The emergence of multi- and extensively-drug resistant TB (MDR and XDR TB) has led to a pressing need for improved diagnostic tests for drug resistance. In 2008, the WHO endorsed line probe assay (LPA) technology for rapid MDR-TB detection. Newer versions of these assays require performance evaluation for endorsement.

    Methods:

    We performed a two-phase, multi-center, blinded study to evaluate and compare the performance of the Nipro NTM+MDRTB and Hain MTBDRplus V2 assays against the WHO-endorsed Hain Genotype MTBDR plus V1 assay. In phase 1 ~200 test strains and in phase 2 ~300 sputum samples were tested by 3 reference laboratories in Germany, South Africa and Uganda, which were blinded to phenotypic and genotypic results. The new tests were considered non-inferior if the lower limits of their 95% confidence intervals were not lower than those of the study Hain V1 point estimate minus the non-inferiority margin (based on meta-analysis).

    Results:

    Although phase 1 non-inferiority analysis for the Hain V2 assay was inconclusive for rifampin (RIF) sensitivity, non-inferiority was demonstrated by the Nipro assay for RIF sensitivity and by both assays for RIF specificity, isoniazid (INH) sensitivity and INH specificity (Table 1, Figure 1). Phase 2 non-inferiority analysis of RIF sensitivity was inconclusive for Hain V2 and Nipro although non-inferiority was demonstrated by both assays for RIF specificity as well as INH sensitivity and INH specificity (Table 1, Figure 2).

    Conclusion:

    Overall the Hain V2 and Nipro assays performed similarly for RIF and INH resistance detection at the three sites. Non-inferiority (using Hain V1 as a reference) was demonstrated for most but not all instances.

    Table 1. Non-inferiority analysis

    Hain V1

    Hain V2

    Nipro

    Sensitivity

    Specificity

    Sensitivity

    Specificity

    Sensitivity

    Specificity

    %, (95% C.I.)

    %, (95% C.I.)

    %, (95% C.I.)

    PHASE 1 RIF

    94.6

    98.3

    94.2

    98.9

    96.7

    99.0

    (91.6-97.6)

    (96.3-100)

    (90.2-96.9)

    (97.0-99.8)

    (93.6-98.6)

    (97.1-99.8)

    PHASE 1 INH

    96.6

    98.8

    96.7

    99.6

    97.4

    100.0

    (86.6-100)

    (93.8-100)

    (93.6-98.6)

    (97.8-100.0)

    (94.6-98.9)

    (98.6-100.0)

    PHASE 2 RIF

    95.6

    93.8

    96

    96.7

    94.4

    96.4

    (92.6-98.6)

    (91.8-95.8)

    (92.2-98.2)

    (94.4-98.3)

    (90-97.3)

    (93.7-98.1)

    PHASE 2 INH

    90.7

    96

    89.3

    97.5

    91.1

    95.6

    (80.7-100)

    (91.0-100)

    (84.8-92.9)

    (95.2-98.9)

    (86.5-94.5)

    (92.6-97.6)

    Figure 1. Forest plots demonstrating Phase 1 non-inferiority analysis

    Figure 2. Forest plots demonstrating Phase 2 non-inferiority analysis

    Ruvandhi Nathavitharana, MD MPH1, Doris Hillemann, PhD2, Joshua Havumaki, MPH3, Eloise Valli, MPH3, Pamela Nabeta, MD3, Catharina Boehme, MD3 and Claudia Denkinger, MD PhD1,3, (1)Division of Infectious Diseases, Beth Israel Deaconess Medical Center, Boston, MA, (2)National Reference Center for Mycobacteria, National Reference Center for Mycobacteria, Borstel, Germany, (3)Foundation for Innovative New Diagnostics, Geneva, Switzerland

    Disclosures:

    R. Nathavitharana, None

    D. Hillemann, None

    J. Havumaki, Hain: Funder/Industry Partner , Hain supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.
    Nipro: Funder/Industry Partner , Nipro supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.

    E. Valli, Hain: Funder/Industry Partner , Hain supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.
    Nipro: Funder/Industry Partner , Nipro supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.

    P. Nabeta, Hain: Funder/Industry Partner , Hain supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.
    Nipro: Funder/Industry Partner , Nipro supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.

    C. Boehme, Hain: Funder/Industry Partner , Hain supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.
    Nipro: Funder/Industry Partner , Nipro supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.

    C. Denkinger, Hain: Funder/Industry Partner , Hain supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.
    Nipro: Funder/Industry Partner , Nipro supplied funds to FIND that were not ear-marked to the study and checks and balances were in place to not have companies have any influence on the design, monitoring, analysis or reporting of the study.

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.