Incidence of Adverse Drug Reactions Related to Immune Thrombocytopenia Drugs. A Prospective Cohort Study

Background: The incidence ofadverse drug reactions (ADRs) related to immune thrombocytopenia (ITP) drugs is not well known in the real-life practice.

Aim: The principal aim of this study was to assess the incidence of ADRs related to ITP drugs. The secondary aims were to compare the incidence of ADRs depending on the drugs, and to assess the factors associated to corticosteroids-related ADR occurrence.

Methods: Study population was the patients included between June 2013 and December 2014 in the CARMEN (Cytopénies Auto-immunes: Registre Midi-PyréneEN) registry. This multicenter registry is carried out on behalf of the French national center for autoimmune cytopenia and the French national center for rare diseases in immunohematology. It is aimed at the prospective follow-up of all newly diagnosed ITP adults in the French Midi-Pyrénées region (3 million inhabitants). Each investigator prospectively follows every patient newly diagnosed for ITP in routine visit or hospital stay, providing informed consent was received. ITP is defined in accordance with French guidelines: platelet count <150 x 10⁹/L and exclusion of other causes of thrombocytopenia. CARMEN is dedicated to pharmacoepidemiological studies, with detailed recording of treatment exposure and prospective reporting of every ADR that the investigator judge significant. All ADRs are analyzed by two independent investigators at the Midi-Pyrénées Center for Pharmacovigilance. Only ADRs with a World Health Organization causality score at least "possible" were included in this study. For incidence calculations, the denominator was the period of exposure to ITP drugs. To assess the factors associated with corticosteroids-related ADRs, we performed univariate and multivariate (backward procedure) Cox models. The variables included were: age, gender, secondary vs.primary ITP, bleeding score and platelet count at diagnosis, diabetes mellitus, Charlson's comorbidity score, and number of concomitant drugs.

Results: Out of 116 patients, 81 were exposed to at least one ITP drug and had at least one follow-up visit. At diagnosis, median age was 64 years (range: 18-95), 48.1% were female, 74.1% had primary ITP, 69.1% had bleeding symptoms and median platelet count was 7 x10⁹/L. Eighty patients (98.8%) of the patients were exposed to corticosteroids, 46 (56.8%) to intravenous immunoglobulin (IVIg), 11 (13.6%) to thrombopoietin receptor agonists (TPO-RA), 7 (8.6%) to danazol and 7 to dapsone, 6 (7.4%) to rituximab. Thirty-nine patients (48.1%) experienced at least one significant ADR (58 ADRs in total). The most frequent ADRs were infections (25.0%) and endocrinal diseases (18.5%). Twenty-two (37.9%) ADRs were serious (leading to hospitalization, necessitating a new drug) and 2 were lethal. The overall incidence of ADRs related to ITP drugs was 4.2/100 persons-weeks (95% confidence interval - 95% CI: 3.2-5.2). Corticosteroids showed the highest incidence of ADR (5.2/100 persons-weeks; 95% CI: 3.6-6.8). In multivariate analysis, the 2 remaining variables associated with corticosteroids-related ADR occurrence were an age >60 years (hazard ratio - HR: 1.79, 95% CI: 0.77-4.15) and diabetes mellitus (HR: 2.73, 95% CI: 1.08-6.87).

Conclusion: This first study assessing the incidence of ADRs related to ITP drugs in a real-life setting showed that they are frequent. Corticosteroids were the first drug responsible for ADRs, stressing the need of second-line treatments, particularly in older and diabetic patients.