Safety, tolerability, and concordance with interferon-γ release assays of a recombinant ESAT6–MPT64 skin test: a phase 1 randomized clinical trial

Background Accurate and cost-effective screening for Mycobacterium tuberculosis (Mtb) infection remains a global challenge. While Interferon-γ Release Assays (IGRAs) offer high specificity, their widespread deployment is hindered by high costs and technical complexity. Conversely, the traditional Tuberculin Skin Test (TST) lacks specificity due to BCG cross-reactivity. Emerging recombinant skin tests predominantly rely on the ESAT6-CFP10 antigen combination. To potentially broaden the antigenic repertoire and enhance diagnostic sensitivity, we developed a novel recombinant fusion protein incorporating MPT64 (from Region of Difference 2, RD2) alongside ESAT-6. This study represents the first-in-human evaluation of the ESAT6-MPT64 (EM) skin test. Methods This single-center, randomized, open-label, dose-escalation Phase 1 clinical trial (Registration: ChiCTR2500112887) enrolled 60 participants, comprising 30 healthy controls and 30 patients with active pulmonary tuberculosis (TB). Participants were stratified and randomized into low-, medium-, and high-dose cohorts (n = 10 per cohort/group). The primary endpoint was safety and tolerability. Secondary endpoints included diagnostic performance (sensitivity and specificity) and concordance with the T-SPOT.TB assay (IGRA). Results The EM skin test demonstrated an excellent safety profile. Adverse events were predominantly mild (Grade 1–2), transient, and self-limiting, with no serious adverse events (SAEs) related to the investigational product reported. In terms of diagnostic performance, the test exhibited robust immunogenicity in active TB patients, achieving a peak sensitivity of 87.0% at 48 hours and maintaining 82.6% at 72 hours post-injection. In healthy controls, the test showed high specificity with a low rate of non-specific reactions (13.3%). Receiver Operating Characteristic (ROC) analysis indicated high diagnostic accuracy, with Area Under the Curve (AUC) values exceeding 0.80 across all dose groups at both 48- and 72-hour reading windows. Furthermore, the EM skin test demonstrated substantial concordance (κ > 0.60) with IGRA results, confirming that the specific DTH response is distinct from BCG vaccination background. Conclusion The recombinant EM skin test is safe, well-tolerated, and demonstrates preliminary diagnostic accuracy comparable to IGRAs. By successfully validating the translational utility of the MPT64 (RD2) antigen in a human cohort, this study provides a strong proof-of-concept for the EM skin test as a scalable, specific, and cost-effective immunodiagnostic tool.