Impact of lymphocyte transformation test (LTT) on the diagnostic accuracy of the culprit drug in drug-induced cytopenias: A case-control study

Susana Martín-López¹Olga Rogozina¹Daniela Aguilar-Concepción²Miguel Álvarez- Montero¹Ramón Pardo-Puras¹Ibtissam Akatbach-Bousaid²María Jiménez-González³Ana Martinez Feito²Miguel Gonzalez-Muñoz^2*Elena Ramírez^1*

• ¹Clinical Pharmacology Department, La Paz University Hospital-IdiPAZ, Faculty of Medicine, Universidad Autónoma de Madrid,, Madrid, Spain
• ²Immunology Department, La Paz University Hospital-IdiPAZ, Madrid, Asturias, Spain
• ³Clinical Trials Unit, La Paz University Hospital-IdiPAZ, Madrid, Asturias, Spain

ABSTRACT Background: Drug-induced cytopenias are serious adverse drug reactions (ADRs), often challenging to diagnose. While causality algorithms offer high sensitivity and positive predictive values, they exhibit low specificity and negative predictive values for identifying the causative drug(s). Therefore, complementary diagnostic tools are required. Objective: This study aimed to evaluate the utility of the lymphocyte transformation test(LTT) in supporting the causality assessment of the Spanish Pharmacovigilance System(SPS) causality algorithm in the diagnosis of the implicated drug(s) in drug-induced cytopenias, using a sample of 40 cases and 85 controls. Methods: Suspected cytopenia cases were identified through the Proactive Pharmacovigilance Program for Laboratory Signals in Hospital or via pharmacovigilance consultation. Control patients completed their treatment without experiencing any ADR. A receiver-operating characteristic (ROC) curve analysis was performed to determine the optimal stimulation index (SI) cut-off for the LTT, maximizing the sum of specificity and sensitivity values to accurate identify cytopenias cases. Results: The case group included 29 cases (72.5%) of agranulocytosis, 6 (15.0%) of neutropenia, 2 (5.0%) of haemolytic anaemia, 2 (5.0%) of bicytopenia and 1 (2.5%) of bone marrow aplasia in 39 patients. Most had ≥3 comorbidities (66.7%) and no previous allergies (71.8%). Eighty-four drugs were suspected as causative agents (SPS-score≥+4), with metamizole being the most frequent (17.2%), followed by acetaminophen (9.1%) and amoxicillin-clavulanate (8.1%). Eight cases (20.0%) involved a single suspected drug, while two cases (5.0%) involved polypharmacy (≥5 drugs). LTT was positive in 75% of cases and in 1.2% of controls. Forty one (41.4%) of the 99 2 suspected drugs yielded positive LTT result. With an optimal SI cut-off of 1.95, the LTT achieved a sensitivity of 72% and a specificity of 99% (area under the curve, 0.86; 95% CI 0.77–0.96; p < 0,001). With monitoring, drug re-exposure was fully tolerated in patients with negative LTT results (100%), but poorly tolerated in one-third of those with positive LTT results. A causality score below 6 and a negative LTT yielded a 100% negative predictive value for drug tolerance (95% CI: 94.4-100%). Conclusion: This study demonstrates that the LTT can be a valuable tool for strengthening causality assessment in suspected drug-induced cytopenias.