S88 Negative Interferon-gamma release assays reliably rule out progression to active TB in patients who have inflammatory conditions and are starting biologic therapy

<h3>Introduction</h3> Interferon-gamma release assays (IGRAs) are the major tool used to screen patients for latent tuberculosis infection (LTBI) before commencing biologic therapy. Anergy due to immunosuppression from drugs may result in false negative IGRAs. There is therefore concern that clinicians may be falsely reassured by negative IGRAs, exposing patients to risk of TB reactivation by commencing biologics. This study assesses the negative predictive value of IGRAs in patients on, or due to start, immunosuppressive therapy with biologics. <h3>Methods</h3> We conducted a retrospective cohort analysis of IGRA test results of patients on biologics between 2013 and 2019, in ICHT. Patients were identified using trust pharmacy data and matched with North-West London Pathology Laboratory records. The London TB Registry was searched to identify cases of active TB. Patients frequently were tested multiple times, and analysis was performed on test rather than patient level. Notification of active TB with a positive IGRA was considered a ‘true positive’. Absence of active TB notification and a negative IGRA was a ‘true negative’. Exclusions were indeterminate or unreadable IGRA results. No distinction has been made between those tested whilst on biologics or immunosuppressants and those tested before commencing them. <h3>Results</h3> 4414 IGRAs in 2705 patients were performed. 3399 were negative with 31 tests belonging to patients who developed active TB during the study period. 254 IGRAs were positive although only 33 of these progressed to active TB, potentially reflecting either LTBI treatment or a poor specificity which is outside the scope of this analysis. Of 99 borderline results, only 2 results (2 patients) developed active TB. <h3>Conclusions</h3> Within this heterogeneous immunosuppressed population, IGRAs have a high NPV. This remains true even if all borderline results are regarded as negative for TB, suggesting suppression of the interferon-gamma response, may have a negligible effect on clinical practice in our study population. IGRAs have a low PPV which may reflect their poor specificity for progression to disease or efficacy of LTBI treatment given. Our results suggest clinicians can generally be confident in negative IGRA results for assessing the risk of developing TB in patients about to start biologic therapy.