TYPE Original Research PUBLISHED 30 January 2023 DOI 10.3389/fphar.2023.1090707 Characteristics of drug safety alerts OPEN ACCESS issued by the Spanish Medicines EDITED BY Fátima Roque, Instituto Politécnico da Guarda, Portugal Agency REVIEWED BY Xavier Kurz, Eva Montané 1,2* and Javier Santesmases 3,4 EMA, Netherlands 1 Florence Van Hunsel, Department of Clinical Pharmacology, Hospital Universitari Germans Trias i Pujol, Barcelona, Spain, 2 Netherlands Pharmacovigilance Centre Department of Pharmacology, Therapeutics and Toxicology, Universitat Autònoma de Barcelona, Lareb, Netherlands Barcelona, Spain, 3Department of Internal Medicine, Hospital Universitari Germans Trias i Pujol, Barcelona, Claudia Garcia Serpa Osorio-de-Castro, Spain, 4Department of Medicine, Universitat Autònoma de Barcelona, Barcelona, Spain Fundação Oswaldo Cruz (Fiocruz), Brazil *CORRESPONDENCE Eva Montané, emontane.germanstrias@gencat.cat Objectives: To describe the characteristics of safety alerts issued by the Spanish SPECIALTY SECTION Medicines Agency (AEMPS) and the Spanish Pharmacovigilance System over a 7-year This article was submitted to period and the regulatory actions they generated. Pharmacoepidemiology, a section of the journal Methods: A retrospective analysis was carried out of drug safety alerts published on Frontiers in Pharmacology the AEMPS website from 1 January 2013 to 31 December 2019. Alerts that were not RECEIVED 05 November 2022 drug-related or were addressed to patients rather than healthcare professionals were ACCEPTED 11 January 2023 PUBLISHED 30 January 2023 excluded. CITATION Results: During the study period, 126 safety alerts were issued, 12 of which were Montané E and Santesmases J (2023), excluded because they were not related to drugs or were addressed to patients and Characteristics of drug safety alerts issued by the Spanish Medicines Agency. 22 others were excluded as they were duplications of previous alerts. The remaining Front. Pharmacol. 14:1090707. 92 alerts reported 147 adverse drug reactions (ADRs) involving 84 drugs. The most doi: 10.3389/fphar.2023.1090707 frequent source of information triggering a safety alert was spontaneous reporting COPYRIGHT (32.6%). Four alerts (4.3%) specifically addressed health issues related to children. © 2023 Montané and Santesmases. This is ADRs were considered serious in 85.9% of the alerts. The most frequent ADRs were an open-access article distributed under the terms of the Creative Commons hepatitis (seven alerts) and congenital malformations (five alerts), and the most Attribution License (CC BY). The use, frequent drug classes were antineoplastic and immunomodulating agents (23%). distribution or reproduction in other Regarding the drugs involved, 22 (26.2%) were “under additional monitoring.” forums is permitted, provided the original author(s) and the copyright owner(s) are Regulatory actions induced changes in the Summary of Product Characteristics in credited and that the original publication in 44.6% of alerts, and in eight cases (8.7%), the alert led to withdrawal from the market this journal is cited, in accordance with of medicines with an unfavorable benefit/risk ratio. accepted academic practice. No use, distribution or reproduction is permitted Conclusion: This study provides an overview of drug safety alerts issued by the which does not comply with these terms. Spanish Medicines Agency over a 7-year period and highlights the contribution of spontaneous reporting of ADRs and the need to assess safety throughout the lifecycle of medicines. KEYWORDS pharmacovigilance, safety-related regulatory actions, adverse drug reaction, spontaneous reporting systems, post-marketing surveillance 1 Introduction Drug safety is an important health concern that requires continuous evaluation throughout a drug’s lifecycle, not only during its development phase but also after it has entered the market. When a drug is approved, safety information is still limited. A major reason for this is the poor external validity of clinical trials, particularly for some conducted by the pharmaceutical industry (Rothwell, 2005). In these studies, patients exposed to the drug have been selected following strict inclusion and exclusion criteria, leading to significant differences between the population treated and the individuals likely to be seen in real clinical practice who tend to be Frontiers in Pharmacology 01 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 older, have many comorbidities, and use polypharmacy (Martin et al., communication type, and focus (Perry et al., 2020). However, few 2004). Some adverse drug reactions (ADRs) may only be detected after previous studies have analyzed the characteristics of safety alerts marketing authorization has been granted because of the mechanism issued by regulatory authorities, though data are available for a few of production, a long latency period, or very low incidence. Therefore, other countries, including Portugal, the Netherlands, Brazil, and the the safety of medicines should be continuously assessed even after Ivory Coast (Ferreira et al., 2020; Soares et al., 2015; Mol et al., 2010; their authorization and appearance on the market, through N’Guessan-Irié et al., 2012). To our knowledge, no research has been pharmacovigilance systems (European Medicines Agency, 2022a) in published regarding alerts specifically issued by the Spanish Medicines which governmental regulatory agencies typically play a key role. In Authority. the last 50 years, pharmacovigilance systems have been developed in With the goal of filling the gap, this study attempted to describe the many countries around the world (Uppsala Monitoring Centre, 2022). characteristics of safety warnings issued by the Spanish Medicines In the European Union (EU), pharmacovigilance is jointly managed Agency and Pharmacovigilance System during a 7-year period after by EU member states, the European Medicines Agency (EMA), and PRAC began to function and examine the regulatory actions taken as a the European Commission (European Medicines Agency, 2022a). result of each alert. Secondary goals included comparing the alerts When health professionals or patients suspect an ADR, a issued by different institutions, assessing specific safety alerts related to spontaneous report is sent to the local regulatory agency. In Spain, drugs under ‘additional monitoring,’ and analyzing alerts according to reports are first sent to the national pharmacovigilance system the the drug’s target population and medical area of focus. Sistema Español de Farmacovigilancia de Medicamentos de uso Humano (SEFV-H). The SEFV-H consists of 17 regional pharmacovigilance centers and the Spanish Medicine Agency (Agencia Española de 2 Methods Medicamentos y Productos Sanitarios [AEMPS]). Then, reports are sent from the SEFV-H to EudraVigilance, the EMA’s 2.1 Design pharmacovigilance network. Finally, spontaneous reports are included in Vigibase, the database of the World Health Organization This study is a retrospective analysis of drug safety alerts issued by Uppsala Monitoring Centre (Baldo et al., 2018). the Spanish Medicines Agency and published on its website (Notas To ensure that drugs continue to have an optimal benefit–risk informativas and medicamentos de uso humano. Agencia Española de balance after they are put on the market, they are typically subject to Medicamentos y Productos Sanitarios, 2022) during the 7-year period risk management plans involving post-authorization safety studies. from 1 January 2013, after the launch of the PRAC to 31 December However, spontaneous reports of ADRs are also essential in triggering 2019, and before the COVID-19 pandemic started. This cut-off date alarm signals. Then, these signals need to be detected, validated, and was chosen to include homogenous safety alerts. confirmed following a management process (European Medicines All of the safety alerts issued were selected and those involving Agency, 2022b). For this purpose, the EMA published new medicines were included. Alerts that were not drug-related, warnings legislation concerning pharmacovigilance and risk management in addressed to patients, and duplicated alerts were excluded. We 2010 and in the process created the Pharmacovigilance Risk considered an alert was duplicated when the ADR alert was similar Assessment Committee (PRAC), which is the EU-level committee and related to the same drug. responsible for assessing and monitoring the safety of medicinal products for human use across Europe. The PRAC came into existence at the end of 2012 and has since demonstrated its 2.2 Variables effectiveness in detecting and evaluating new drug safety signals (Potts et al., 2020). In Spain, the published law was enacted in For each alert, the following descriptive variables were recorded. 2013 that brought Spain in line with European legislation (Agencia estatal boletín oficial del estado, 2022) by promoting the creation of - Year of publication health data bases, conducting pharmacoepidemiological studies and - Issuing agency: AEMPS or EMA evaluating ADR reports received by the SEFV-H. - Population targeted by the alert: children <18 years old, adults In general, drug safety alerts are motivated by signals obtained from 18–65 years old, and elderly >65 years old databases of pharmacovigilance systems and signals warning of ADRs - Drugs involved: classification was based on the first level of the detected in post-marketing studies and/or clinical trials. A signal Anatomical Therapeutic Chemical (ATC) classification (WHO “suggests a new potentially causal association or a new aspect of a Collaborating Centre for Drug Statistics Methodology, 2018) known association between an intervention and an event or set of - Limitations on the prescription or dispensing of the drug: related events that is judged to be of sufficient likelihood to justify specifically whether it was limited to the hospital setting or not verificatory action” (European Medicines Agency, 2022b). The - Prescribing specialists: the healthcare professionals most likely to generated alerts provide information to health-care professionals or prescribe the drug in question in the consensus view of the patients about the needed regulatory actions based on the signals such as authors; for widely used drugs, prescribing professionals were new restrictions on the use of medicine in particular populations, the assumed to be general practitioners mandatory monitoring of patients using that medicine, or the - Drugs under “additional monitoring:” this term is used by the withdrawal of the drug from the market (Farcaş et al., 2018). EMA to denote medicines that are more intensively monitored In a recent study comparing drug safety alerts issued by national than others, generally because there is less safety information medicine regulators in Australia, Canada, the United Kingdom, and the available; they have an inverted black triangle [▼] displayed on United States from 2007 to 2016, major differences were found in the their packaging and Summary of Product Characteristics use of safety advisories by regulators, including their frequency, content, (SmPC) (European Medicines Agency, 2022) Frontiers in Pharmacology 02 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 - Drug indication: A medical condition for which that medicine is corroborate drug risks reported by the PRAC when issued by the prescribed AEMPS, or include changes to the regulatory measures applied. Drugs - Type of ADR: classified in accordance with the System Organ involved in duplicated alerts were alemtuzumab, canagliflozin, Classes (SOC) of the Medical Dictionary for Regulatory cilostazol, codeine, cyproterone acetate, denosumab, diacerein, Activities (MedDRA) (MEDDRA, 2021) fentanyl, fingolimod, fusafungine, gliflozins, hydroxyethylstarch, - Seriousness of the ADR: a serious ADR was defined according to idelalisib, methotrexate, mycophenolic acid, radium dichloride, the International Conference on Harmonization (ICH) guideline strontium ranelate, tetrazepam, tofacitinib, ulipristal, valproic acid, E2D as one that is fatal, life-threatening, requires hospital and VPH vaccine. All duplicates (n = 22) were eliminated from the admission or prolongation of hospital stay and causes analyzed dataset, which resulted in a total of 92 remaining alerts persistent or significant disability/incapacity, congenital involving 84 drugs. anomaly/congenital defect or medically important (European Forty alerts (43.5%) were issued by the AEMPS, while the Medicines Agency, 2004) remainder were issued by the EMA (47 of which were specifically - Sources of evidence: spontaneous reports, clinical trials, and/or issued by the PRAC). Sources used to support safety alerts were observational studies spontaneous reporting of ADRs (30 alerts, 32.6%), clinical trials - Reiteration of the alert: whether it was a first-time alert or the (16 alerts, 17.4%), observational studies (eight alerts, 8.7%), or reiteration of an alert issued before the study period multiple sources (21, 22.8%). A total of 17 alert (18.5%) sources - Regulatory actions to be taken as noted in the alert such as new were not specified. restrictions on use, risk minimization measures, changes in A total of 147 ADRs were reported in the 92 safety alerts, ranging SmPC, or withdrawal from the market. Regulatory actions are from one to four ADRs per alert; two or more ADRs were reported in based on the scientific decisions of an expert group responsible 42 (45.6%) alerts. The most frequently reported ADRs were hepatitis for assessing safety problems related to medicines such as the (nine alerts), death (seven alerts), congenital malformation (six alerts), Committee for Medicinal Products for Human Use [CHMP]. and drug inefficacy, arrhythmia, and arterial thrombosis (five alerts When the group decides that the degree of risk is acceptable each). The remaining types of ARDs are detailed in Table1. The most under the currently authorized conditions of use, the regulatory frequent ADRs in terms of the MedDRA’s SOC classification were action is to provide information regarding the ADR and cardiac, vascular, and hepatobiliary disorders (19, 15, and 14 alerts, recommending risk minimization measures such as patient respectively) (Table 2). monitoring and follow-up. When the group decides that the The most frequently alerted drugs (by ATC category) were risk is acceptable but only under certain conditions, the L-Antineoplastic and immunomodulating agents (21 alerts, 23%), regulatory agency issues a restriction of use such as restricting N-Nervous system (12, 13%), and J-Anti-infectives for systemic use the indications for which the medicine may be prescribed or the (11, 12%) (Table 3). The most alerts (three) were issued for fingolimod patient population to which it may be given. Any of these (Table 4). The prescription of 37 drugs (45%) was restricted to the measures can lead to the modification and/or updating of the hospital setting. SmPC. When the group decides that the risk is unacceptable and With regard to drugs under ‘additional monitoring,’ a total of the benefit/risk ratio is unfavorable, the drug is withdrawn from 26 alerts (28.3%) were related to 22 of these drugs (Table 5), and the the market (Rodriguez Pascual, 2006). most frequently involved drugs were fingolimod (three alerts) and tofacitinib (two alerts). Hepatitis or hepatobiliary disorders were the The selection of alerts for inclusion in the study and the recording most frequent ADR alerts for this group of drugs (in connection with of study variables were conducted by the two researchers working five drugs, 22.7%). The SmPC was modified for 12 of these drugs independently (SJ and ME). Consensus was reached whenever (54.5%), and one of them (strontium ranelate) was withdrawn from discrepancies arose. the market. All ADRs were considered serious, with the six exceptions being drug inefficacy, cerebral deposits of contrast, and 2.3 Statistical analysis hypertrichosis. Thirty-nine (42.4%) alerts were addressed to adult populations, 37 (40.2%) were addressed to the elderly Descriptive statistics (counts and percentages) were used to population (whether exclusively or not), four (4.3%) alerts were characterize the variables assessed in this study. To compare the specifically addressed to children or to children and adults, and characteristics of alerts issued by AEMPS with those issued by the eight (8.7%) alerts were targeted at populations of all ages. Alerts EMA (mainly by the PRAC), the Pearson chi-square or Fisher’s exact addressed to children are detailed in Table 6. Seven (7.6%) alerts test for categorical variables was used. A bilateral p-value <0.05 was were addressed to pregnant women and warned of consequences to used to determine statistical significance. the fetus (mainly congenital malformations). The features of these All statistics were performed using the SPSS software package for alerts are detailed in Table 7. Windows version 19.0 (SPSS Inc., Chicago, Ill). The prescribing specialists were most often rheumatologists (10 alerts), neurologists, and hematologists (eight alerts each), although 11 alerts involved drugs to treat unspecific symptoms 3 Results such as pain or fever, emesis, cough, rhinosinusitis, and urinary tract infection and therefore were likely of interest to general An initial search of the EMA website yielded a total of 126 safety practitioners. The diseases most likely linked to safety alerts were alerts. Thirty-four alerts were excluded (22 were duplicated alerts) any kind of pain (seven alerts), multiple sclerosis (six alerts), and (Figure 1). These duplicated alerts provide new safety data, osteoporosis (five alerts) (Table 8). Frontiers in Pharmacology 03 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 FIGURE 1 Flow chart of included safety alerts. The most frequent regulatory actions derived from alerts were vs. 34.6%; p < 0.015), while alerts issued by the EMA were more often “restrictions in use” in 30 alerts (32.6%) and “reporting of ADRs and addressed to child populations (25% vs. 7.5%; p < 0.028). recommendation of patient monitoring and follow-up” in 14 alerts (15.2%). ‘Restrictions in use’ included restrictions related to the age of the population receiving the medicine, the specific disease being 4 Discussion treated, and the duration of treatment and/or dosage. Combinations of these regulatory actions were present in 22 alerts The results of this study show that the Spanish Medicines Agency (23.9%). The regulatory actions of 41 alerts (44.6%) involved changes issued, on average, one or two drug safety alerts every month from in the SmPC, and eight (8.7%) required mandated withdrawal from 2013 to 2019, one out of five of these were alert duplications. The rate the market of medicines with an unfavorable benefit/risk ratio. Only of alert issuance did not increase over time in this period, contrary to one of these withdrawn medicines was put on the market in the 5 years other studies conducted in Portugal and the Netherlands (Soares et al., prior to the alert, while five had been on the Spanish market for more 2015; Ferreira et al., 2020). Almost half of the alerts were issued by the than 20 years (Table 9). Two batches of one drug were withdrawn national medicines agency, and some were subsequently issued by the from the market because they contained minoxidil instead of EMA, indicating that the SEFV-H is a proactive local safety system and omeprazole due to an error in the manufacturing process. an effective national signal detection system that anticipates alerts issued by the PRAC (Farcas et al., 2020). Therefore, our findings are in line with those of other similar European studies. However, differences 3.1 Comparison of alerts issued by the AEMPS between European and other regulatory agencies (such as those in the with those issued by the EMA US, Canada, or Australia) may lead to different risk assessments of a drug and hence to different regulatory actions (Farcaş et al., 2018; When safety alerts were compared by the issuing institution, some Bjerre et al., 2018; Bhasale et al., 2021). significant differences were found. Alerts issued by the AEMPS In this study, we found that the main trigger leading to the involved more drugs ‘under additional monitoring’ (42.5% vs. generation of these alerts was spontaneous case reports, as 17.3%; p < 0.008) and drugs restricted to a hospital setting (60% indicated in other studies (Alves et al., 2013; Lester et al., 2013; Frontiers in Pharmacology 04 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 TABLE 1 Type of adverse drug reaction (ADR) related to the alert. ADR Number of alerts % ADR Hepatitis 9 6.1 Death 7 4.8 Congenital malformation 6 4.1 Inefficacy 5 3.4 Arrhythmia 5 3.4 Arterial thrombosis 5 3.4 Allergic reaction 4 2.7 Cardiovascular risk 4 2.7 Progressive multifocal leukoencephalopathy 4 2.7 Venous thrombosis 4 2.7 Acute myocardial infarction 3 2.1 Cardiac failure 3 2.1 Cerebral deposit 3 2.1 Drug toxicity 3 2.1 Infection 3 2.1 Bleeding 2 1.4 Bone fracture 2 1.4 Hypotension 2 1.4 Kidney failure 2 1.4 Lymphoma 2 1.4 Liver failure 2 1.4 Mandibular osteonecrosis 2 1.4 Poisoning 2 1.4 Psychiatric disorder 2 1.4 Skin cancer 2 1.4 Skin reaction 2 1.4 Abuse and dependence, acute pulmonary edema, agranulocytosis, alveolar bleeding, atrial fibrillation, atrioventricular block, 1 each (56) 0.68% each bradycardia, bronchiectasis, cirrhosis, coma, complex regional pain syndrome (CRPS), connective tissue fibrosis, convulsions, cutaneous reaction, diabetes mellitus, diarrhea, difficulty driving, ergotism, extrapyramidalism, hepatocarcinoma, hemophilia, hypercalcemia, hyperkalemia, hypersensitivity, hypertrichosis, hypocalcemia, hypogammaglobulinemia, hyponatremia, ketoacidosis, liver decompensation, liver toxicity, lower extremity amputation, lung disease, lymphohistiocytosis, miscarriage, muscle weakness, nephrotic syndrome, neural tube defects, neurodevelopmental disorder, ocular pigmentation, opportunist infection, pancreatitis, Parkinson’s disease, pneumonia, pneumonitis, postural orthostatic tachycardia syndrome (PoTS), pulmonary fibrosis, pulmonary infection, pulmonary thromboembolism, respiratory depression, restless legs, skin pigmentation, somnambulism, stroke, tendinopathy, thrombocytopenia, thrombotic microangiopathy, tumor, VHB reactivation Total 147 100 Ishiguro et al., 2017). This underlines the importance and usefulness of particularly closely by regulatory authorities either because the spontaneous reporting and national pharmacovigilance systems in medicine has been recently approved or because there are limited detecting issues of concern related to medicines and contributing to data on its long-term use. These findings show that recently marketed the generation of safety alerts during the marketing period (European drugs, with less safety information available, should be a priority for Medicines Agency, 2022b). Other sources, such as clinical trials and/or the reporting of suspected ADRs. Therefore, information about any post-marketing studies, were also necessary to identify or confirm possible ADRs must be collected as early as possible to further inform safety issues in some alerts. More than one out of every four safety healthcare practitioners about the safety of these medicines. alerts concerned drugs under “additional monitoring,” that is, Although most alerts were addressed the adult population, almost medicines authorized for use in the EU but are being monitored half were referred to drugs typically prescribed to elderly patients Frontiers in Pharmacology 05 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 TABLE 2 ADRs classified according to MedDRA’s SOC terminology. Number of alerts % Infections and infestations 10 7 Infections (3), progressive multifocal leukoencephalopathy (4), opportunistic infection (1), pulmonary infection (1), and VHB reactivation (1) Neoplasms benign, malignant, and unspecified 7 4.9 Lymphoma (2), skin cancer (2), hepatocarcinoma (1), lymphohistiocytosis (1), and tumor (1) Blood and lymphatic system disorders 6 4.2 Hemorrhage (2), agranulocytosis (1), hemophilia (1), thrombocytopenia (1), and thrombotic microangiopathy (1) Immune system disorders 6 4.2 Allergic reaction (4), hypersensitivity (1), and hypogammaglobulinemia (1) Endocrine disorders 4 2.8 Hypercalcemia (1), hypocalcemia (1), hyponatremia (1), and hyperkaliemia (1) Metabolism and nutrition disorders 2 1.5 Diabetes mellitus (1) and ketoacidosis (1) Psychiatric disorders 4 2.8 Abuse and dependence (1), psychiatric disorder (2), and somnambulism (1) Nervous system disorders 9 6.3 Cerebral deposit (3), coma (1), convulsions (1), extrapyramidal symptoms (1), neurodevelopmental disorder (1), Parkinson’s (1), and restless legs (1) Eye disorders 1 0.7 Ocular pigmentation (1) Ear and labyrinth disorders 0 0 Cardiac disorders 19 13.4 Arrhythmia (5), cardiovascular risk (4), acute myocardial infarction (3), cardiac failure (3), acute pulmonary edema (1), atrial fibrillation (1), atrioventricular block (1), and bradycardia (1) Vascular disorders 15 10.6 Arterial thrombosis (5), venous thrombosis (4), hypotension (2), ergotism (1), lower extremity amputation (1), postural tachycardia syndrome (PoTS) (1), and pulmonary thromboembolism (1) Respiratory, thoracic, and mediastinal disorders 7 4.9 Alveolar bleeding (1), bronchiectasis (1), pneumonia (1), pneumonitis (1), lung disease (1), pulmonary fibrosis (1), and respiratory depression (1) Gastrointestinal disorders 2 1.5 Diarrhea (1) and pancreatitis (1) Hepatobiliary disorders 14 9.9 Hepatitis (9), liver failure (2), hepatic cirrhosis (1), hepatic decompensation (1), and liver toxicity (1) Skin and subcutaneous tissue disorders 4 2.8 Skin reaction (2), hypertrichosis (1), and skin pigmentation (1) Musculoskeletal and connective tissue disorders 8 5.6 Bone fracture (2), mandibular osteonecrosis (2), complex regional pain syndrome (CRPS) (1), connective tissue fibrosis (1), muscle weakness (1), and tendinopathy (1) Renal and urinary disorders 3 2.2 Kidney failure (2) and nephrotic syndrome (1) Pregnancy, puerperium, and perinatal conditions 1 0.7 Miscarriage (1) Reproductive system and breast disorders 0 0 Congenital, familial, and genetic disorders 7 4.9 Congenital malformation (6) and neural tube defect (1) (Continued on following page) Frontiers in Pharmacology 06 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 TABLE 2 (Continued) ADRs classified according to MedDRA’s SOC terminology. Number of alerts % General disorders and administration site conditions 7 4.9 Death (7) Investigations 0 0 Injury, poisoning, and procedural complications 5 3.5 Drug toxicity (3) and poisoning (2) Surgical and medical procedures 0 0 Social circumstances 1 0.7 Difficulty driving (1) Product issues 0 0 a 142 100 a 5 alerts were not classifiable (alerts of drug inefficacy). TABLE 3 Anatomical Therapeutic Chemical (ATC) drug classification system. ATC category Therapeutic area Number of alerts % A Alimentary tract and metabolism 7 7.6 B Blood and blood forming organs 9 9.8 C Cardiovascular system 6 6.5 D Dermatological 1 1.1 G Genito-urinary system and sex hormones 7 7.6 H Systemic hormonal preparations, excluding sex hormones and insulin 3 3.3 J Anti-infectives for systemic use 11 12 L Antineoplastic and immunomodulating agents 21 22.8 M Musculoskeletal system 9 9.8 N Nervous system 12 13 P Antiparasitic products, insecticides, and repellents 0 0 R Respiratory system 2 2.2 S Sensory organs 0 0 V Various 4 4.3 Total 92 100 (whether exclusively or not), which is a population that frequently women, these mainly concerned the risk of congenital receives multiple drugs, has multiple morbidities, and is rarely malformations, and the source of information was epidemiological included in clinical trials (Davies and O’Mahony, 2015). Pregnant studies (when specified). As pregnant women are traditionally women and children, who accounted for 12% of the analyzed alerts, excluded from clinical trials of therapeutics other than those are also “vulnerable” populations in terms of drug efficacy/ related to diseases of childbirth, these data usually come from effectiveness and safety. These alerts were specifically addressed to observational or pharmacoepidemiological studies (Irl and Hasford, children involved with three drugs (codeine, parenteral nutrition, and 2020). minoxidil), which contrasts with another study where antidepressants Rheumatologists, neurologists (mainly from multiple sclerosis were the only drug alerts related to children (Clavenna and Bonati, units), and hematologists were most likely interested in the safety 2009). Spontaneous reporting in the pediatric population is crucial due alerts. Moreover, almost half of the safety alerts involved drugs to the limited number of clinical trials and the huge “off-label” use of restricted to use in the hospital setting. Therefore, hospital drugs in situations where drug safety is not yet well-established pharmacovigilance programs must play an important role in (Conroy et al., 2000). Regarding alerts addressed to pregnant disseminating safety alerts to their healthcare professionals. These Frontiers in Pharmacology 07 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 TABLE 4 Drugs involved in alerts. Drug ATC code Number of alerts % Fingolimod L04AA27 3 3.2 Cilostazol B01AC23 2 2.1 Codeine R05DA04 2 2.1 Denosumab M05BX04 2 2.1 Fentanyl N02AB03 2 2.1 Mycophenolic acid L04AA06 2 2.1 Tofacitinib L04AA29 2 2.1 Parenteral iron B03AC 2 2.1 Progestogens and estrogens, fixed combinations G03AA 2 2.1 Aceclofenac, aliskiren/ACE inhibitors/sartans, aflibercept, agomelatine, alemtuzumab, atosiban, B01AC06 1 each (73) 1.1% each brivudine, bromocriptine, calcitonin, canagliflozin, carbimazole/thiamazole, cladribine, cholecalciferol/ calcidiol, coagulation factor VIII, corticosteroids, cyproterone acetate and estrogen, daratumumab, diacerein, diclofenac, dimethyl fumarate, direct acting antivirals, direct factor Xa inhibitors, dolutegravir, domperidone, elvitegravir/cobicistat, ergot alkaloids, febuxostat, flutamide, fusafungine, gadobenic acid, gadodiamide, gadolinium, gliflozins, hydrochlorothiazide, hydroxyzine, hydroxyethylstarch, ibuprofen/dexibuprofen, idelalisib, interferon beta, immunosuppressants, ® Inzitan , ivabradine, ketoconazole, leuprorelin, metamizole, methylprednisolone, metoclopramide, methotrexate, natalizumab, nicotinic acid-laropiprant, nitrofurantoin, ombitasvir/paritaprevir/ ritonavir, omeprazole, ondansetron, papillomavirus vaccine, parenteral nutrition, pomalidomide, posaconazole, quinolone, radium dichloride, retigabine, retinoids, riociguat, ritodrine, selexipag, sofosbuvir/ledipasvir/daclatasvir–amiodarone, strontium ranelate, tetrazepam, trimetazidine, tyrosine kinase inhibitors, ulipristal, valproic acid, and zolpidem Total - 92 100 ® Inzitan : dexamethasone, thiamine, cyanocobalamin, and lidocaine results cannot be compared with those of other studies because studies or to be classified with an unfavorable benefit/risk ratio, leading to with similar designs are lacking. Findings of this sort are also their withdrawal from the market. Changes or updates in the SmPC contingent on the study period assessed and are probably not were frequent regulatory actions for almost half of the medicines in informative about drug consumption or the number of drugs this and other studies (Lester et al., 2013; Ferreira et al., 2020; Perry available for particular medical specialties. ADRs for which alerts et al., 2020). This was specifically true for nationally authorized are issued are unlikely to be identified in clinical trials because they are drugs, which are more likely to be updated than EU-authorized rare and unpredictable. In addition, the most frequent types of ADRs products (van Hunsel et al., 2021). In contrast, drug withdrawal in this analysis were infections and neoplasms, which are type A effects tended to be infrequent and occurred in approximately 5%–7% of related to the mechanism of action of antineoplastic agents, the most alerted drugs, with several prior informational alerts preceding the frequent drug class alerted, and consequently, these ADRs are frequent suspension of the medicine (Soares et al., 2015; Ferreira et al., 2020). and predictable. In contrast to other studies, the drugs most frequently In our study, most of these drugs had been available in the Spanish addressed in the alerts reported in this study were antidepressants, market for decades. The causes of withdrawal of these drugs were antidiabetics, anti-infectives, and anti-inflammatory drugs (Farcaş heterogeneous, and for a few drugs the suspension of marketing et al., 2018; Ferreira et al., 2020; Perry et al., 2020; Meddra, 2021). authorization was due to multiple reasons. The most frequent Hepatitis, death, and congenital malformations were the most reported causes for withdrawal from the market were liver, frequent ADR alerts with the first two being the most frequently cardiac, and nervous system toxicity, and the median interval reported ADRs in other studies (Farcaş et al., 2018; Meddra, 2021). between the first report of an ADR pointing to withdrawal and Nevertheless, information about birth defects is often lacking when a actual withdrawal itself was approximately 1 year (Onakpoya et al., drug is marketed because pregnant women are excluded from clinical 2016a). However, there are discrepancies in the patterns of trials for ethical reasons. Both reported drugs and ADRs are in line withdrawal of medicinal products from the market and with safety signals assessed by PRAC between 2014 and 2017 (Farcaş withdrawals are inconsistent across countries (Onakpoya et al., et al., 2018). 2016b). These findings underline the need for continuous Regarding the regulatory actions resulting from these alerts, monitoring of risks throughout the lifecycle of a medicine. one-third led to a restriction in the use of the medicine, which is a finding that was similar to that reported in another study in which the most frequent restriction involved adding new 4.1 Limitations and strengths contraindications for use (Perry et al., 2020). Some of the reported ADRs in this study were sufficiently serious and This study has several limitations. First, only the alerts issued in clinically important enough to require changes to their labeling Spain were evaluated; however, evaluations and decisions taken by the Frontiers in Pharmacology 08 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 TABLE 5 Characteristics of alerts for drugs under ‘additional monitoring.’ ADR Drug ATC first Modification of the level SmPC Osteonecrosis mandibular Aflibercept L Yes Hepatitis and hepatocarcinoma Direct agent antiviral J No Ketoacidosis Canagliflozin and other gliflozins A Yes Limb amputation No Cardiovascular risk and hemorrhage Cilostazol C Yes Hepatitis Daratumumab L Yes Osteonecrosis and hypocalcemia Denosumab M No Neural tube defect Dolutegravir J No Arrhythmia, lymphoma, congenital malformation, skin cancer, and Fingolimod L Yes opportunistic infection Infection Idelalisib L No VHB reactivation Immunosuppressants L No Hypersensitivity Iron isomalthose B No Progressive multifocal leukoencephalopathy Natalizumab L Yes Liver failure and liver decompensation Paritaprevir/ombitasvir/ritonavir J Yes Hepatitis Pomalidomide L Yes Bone fracture and death Radium dichloride V No Ocular and skin pigmentation Retigabine N No Death and respiratory infection Riociguat C Yes Toxicity and inefficacy Selexipag L No Bradycardia and cardiac block Sofosbuvir/ledipasvir/ J Yes daclatasvir– amiodarone Cardiovascular risk and skin reaction Strontium ranelate M Market withdrawal Pulmonary thromboembolism and death Tofacitinib L Yes Venous thrombosis and infection No Hepatitis Tyrosine kinase inhibitors L Yes SmPC, summary of product characteristics TABLE 6 Characteristics of alerts with drugs addressed to children. Year of Drug ADR Mechanism of ADR Issuing Source Regulatory issue institution action 2013 Codeine Respiratory Overdose due to genetic deficiency of PRAC Spontaneous reporting Restriction by age and (analgesic) depression death CYP2D6 and others diseases 2015 Codeine Intoxication Overdose due to genetic deficiency of PRAC Spontaneous reporting Restrictions in use for (antitussive) CYP2D6 ages <12 2019 Parenteral Death Peroxide formation due to exposure to light AEMPS Laboratory and clinical Changes in the SmPC nutrition studies 2019 Omeprazole Hypertrichosis There was minoxidil instead of omeprazole AEMPS Spontaneous reporting Withdrawal of due to a manufacturing error batches AEMPS, Agencia Española de Medicamentos y Productos Sanitarios; PRAC, pharmacovigilance risk assessment committee; SmPC, summary of product characteristics EMA were involved in more than half of them. A second limitation issued alerts. The reason why we did not include alerts issued during was the short length of the study period of only 7 years; nonetheless, 2020 and 2021 should be self-evident because the outbreak of the we consider this period sufficiently representative to characterize the COVID-19 pandemic quickly changed the profile of alerts and their Frontiers in Pharmacology 09 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 TABLE 7 Characteristics of alerts with drugs addressed to pregnant women. Year of Drug ADR Issuing Source of information Regulatory action issue institution 2014 and 2018 Valproic acid Congenital malformation PRAC Observational studies Use restrictions, risk minimization measures Neurodevelopmental disorder 2015 Mycophenolate Congenital malformation AEMPS Not specified Use restrictions, risk mofetil minimization measures Miscarriage 2018 Retinoids Congenital malformation PRAC Not specified Use restrictions, risk minimization measures 2019 Elvitegravir + Inefficacy AEMPS Clinical trial Use restrictions cobicistat Risk of mother-to-child transmission of HIV 2019 Fingolimod Congenital malformation PRAC Not specified Use restrictions 2019 Ondansetron Congenital malformation PRAC Observational studies Use restrictions 2019 Carbimazol/tiamazol Congenital malformation AEMPS Observational studies and Use restrictions, risk spontaneous reporting minimization measures AEMPS, Agencia Española de Medicamentos y Productos Sanitarios; HIV, human immunodeficiency virus; PRAC, pharmacovigilance risk assessment committee TABLE 8 Indication of drugs involved in alerts and potential specialist prescriber. Specialist prescriber Indication (number of alerts) Number of alerts % General practitioner Pain or fever (5), emesis (2), cough (1), dyslipidemia (1), rhinosinusitis (1), and urinary tract infection (1) 11 11.9 Rheumatology Osteoporosis (5), arthritis (3), hyperuricemia (1), and osteoarthritis (1) 10 10.8 Hematology Leukemia (3), iron deficiency (2), multiple myeloma (2), and hemophilia (1) 8 8.6 Neurology Multiple sclerosis (6)and epilepsy (2) 8 8.6 Obstetrics and gynecology Contraception (2), preterm labor (2), lactation (1), uterine myoma (1), and prophylaxis HPV infection (1) 7 7.6 Infectious diseases HIV infection (2), mycosis (2), and bacterial infection (1) 5 5.4 Cardiology Angina (2), anticoagulation (1), cardiac failure (1), and diuretic (1) 5 5.4 Oncology Immunosuppression (2), chemotherapy-induced emesis (1), colorectal cancer (1), and disruptive pain (1) 5 5.4 Dermatology Acne and hirsutism (1), acne and psoriasis (1), and herpes zoster infection (1) 3 3.3 Endocrinology Diabetes mellitus (2) and hyperthyroidism (1) 3 3.3 Pneumology Pulmonary hypertension (2) and asthma/COPD (1) 3 3.3 Hepatology VHC infection (3) 3 3.3 Pediatrics Gastroesophageal reflux (1), malnutrition (1), and pain (1) 3 3.3 Psychiatry Bipolar disorder (1), depression (1), and insomnia (1) 3 3.3 Radiology NMR contrast (3) 3 3.3 Urology Prostate cancer (3) 3 3.3 Allergology Allergy (2) 2 2.2 Gastroenterology Ulcerative colitis (2) 2 2.2 Vascular surgery Intermittent claudication (2) 2 2.2 Intensive care medicine Acute hemorrhage (1) 1 1.1 Nephrology Kidney transplantation (1) 1 1.1 Traumatology Musculoarticular pain (1) 1 1.1 Total 92 100 COPD, chronic obstructive pulmonary disease; HIV, human immunodeficiency virus; HPV, human papillomavirus; NMR, nuclear magnetic resonance; VHC, Hepatitis C virus. Frontiers in Pharmacology 10 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 TABLE 9 Characteristics of medicines withdrawn from the market. Year medicine came Year alert Issuing Drug (medicine) Indication ADRs to the market was issued institution 2008 2013 PRAC Nicotinic acid-laropiprant Hyperlipidemia Highest incidence of bleeding, muscle ® (Tredaptive )▼ weakness, infections, and diabetes mellitus 1978 2013 PRAC Tetrazepam (Myolastan ) ® Muscle pain Cutaneous reaction <1991 2013 PRAC Ketoconazole systemic use Mycosis Hepatitis 1964 2016 PRAC Fusafungine (Fusaloyos )® Rhinopharyngitis Allergic reaction 2004 2014 PRAC ® Strontium ranelate (Protelos ) Osteoporosis Cardiovascular risk and cutaneous ® (Osseor ) reaction 1968 2017 AEMPS Inzitan® (dexamethasone, thiamine, Pain Allergic reaction lidocaine, and cyanocobalamin) 1994 2018 PRAC Gadodiamide (Omniscan ) ® Magnetic resonance Cerebral deposit 2019 2019 AEMPS Batches with minoxidil instead of Gastroesophageal Hypertrichosis omeprazole (Farma-Química Sur S.L) reflux AEMPS, Agencia Española de Medicamentos y Productos Sanitarios; PRAC, pharmacovigilance risk assessment committee focus shifted to drugs used to treat the coronavirus and vaccines. ‘vulnerable’ patient populations, such as children and pregnant Another limitation is that we did not include Direct Healthcare women, are involved. Professional Communications (formerly called ‘Dear Doctor Letter’), which are issued by pharmaceutical companies to inform healthcare providers of important drug-related safety issues and follow Data availability statement a different safety assessment process. Finally, we did not assess the impact of alerts on the prescription of the medicines concerned and The raw data supporting the conclusions of this article will be trends in their consumption. Nevertheless, the impact of risk made available by the authors without undue reservation. minimization has already been assessed in various other studies (The European Network of Centres for Pharmacoepidemiology and Pharmacovigilance). Author contributions Our study also has certain strengths. This is the first study to analyze drug alerts issued by the Spanish Medicines Agency. EM: conceived the idea, contributed to the study design, Additionally, it is the first study to devote special attention to alerts conducted the literature search, reviewed the cases selected, involving drugs under “additional monitoring” and also the first to contributed to data acquisition of included cases, analysis and break down the analysis according to the population targeted and the interpretation of the data, and writing of the first draft of the medical specialty involved. manuscript. JS: contributed to the study design, reviewed the cases selected, contributed to data acquisition of included cases, analysis and interpretation of the data, and writing of the manuscript. All authors 5 Conclusions agree to be accountable for the content of this work. Pharmacovigilance systems should ensure monitoring of all authorized medicines throughout their lifecycle in clinical use, and Conflict of interest regulatory authorities should continuously review their benefit/risk ratio and issue an alert whenever a new signal or risk is detected. Over The authors declare that the research was conducted in the a 7-year period, almost one hundred alerts were issued by the absence of any commercial or financial relationships that could be Spanish’s Medicines Agency. Despite the limitations, spontaneous construed as a potential conflict of interest. reporting was the most frequent source of information for the basis of alerts and reinforced the idea that such reporting is an essential tool for drug and patient safety. Drugs “under additional monitoring” were Publisher’s note frequently implicated in alerts due to their unknown safety profile. The most frequent regulatory action resulting from an alert was changes in All claims expressed in this article are solely those of the authors the SmPC, and one in ten alerts resulted in the withdrawal of the drug and do not necessarily represent those of their affiliated organizations, from the market. Our findings underline the important contribution or those of the publisher, the editors, and the reviewers. Any product of the spontaneous reporting of ADRs and the need for a safety that may be evaluated in this article, or claim that may be made by its assessment throughout the lifecycle of medicines, particularly when manufacturer, is not guaranteed or endorsed by the publisher. Frontiers in Pharmacology 11 frontiersin.org Montané and Santesmases 10.3389/fphar.2023.1090707 References Agencia estatal boletín oficial del estado (2022). RD 577/2013 de farmacovigilancia de Ishiguro, C., Misu, T., Iwasa, E., and Izawa, T. (2017). Analysis of safety-related medicamentos de uso humano. Available at: https://www.boe.es/eli/es/rd/2013/07/26/577 regulatory actions by Japan’s pharmaceutical regulatory agency. Pharmacoepidemiol. (accessed Oct 23, 2022). Drug Saf. 26 (11), 1314–1320. doi:10.1002/pds.4252 Alves, C., Macedo, A. F., and Marques, F. B. (2013). Sources of information used by Lester, J., Neyarapally, G. A., Lipowski, E., Graham, C. F., Hall, M., and Dal Pan, G. regulatory agencies on the generation of drug safety alerts. Eur. J. Clin. Pharmacol. 69 (12), (2013). Evaluation of FDA safety-related drug label changes in 2010. Pharmacoepidemiol. 2083–2094. doi:10.1007/s00228-013-1564-y Drug Saf. 22 (3), 302–305. doi:10.1002/pds.3395 Baldo, P., Francescon, S., and Fornasier, G. (2018). Pharmacovigilance workflow in Martin, K., Begaud, B., Latry, P., Miremont-Salame, G., Fourrier, A., and Moore, N. Europe and Italy and pharmacovigilance terminology. Int. J. Clin. Pharm. 40 (4), 748–753. (2004). Differences between clinical trials and postmarketing use. Br. J. Clin. Pharmacol. 57 doi:10.1007/s11096-018-0711-z (1), 86–92. doi:10.1046/j.1365-2125.2003.01953.x Bhasale, A. L., Sarpatwari, A., De Bruin, M. L., Lexchin, J., Lopert, R., Bahri, P., et al. MEDDRA (2021). Introductory guide MedDRA version 24.0. Available at: https://www. (2021). Postmarket safety communication for protection of public health: A comparison of meddra.org/how-to-use/support-documentation/english (accessed Oct 23, 2022). regulatory policy in Australia, Canada, the European union, and the United States. Clin. Mol, P. G., Straus, S. M. J. M., Piening, S., de Vries, J. T. N., de Graeff, P. A., and Haaijer- Pharmacol. Ther. 109 (6), 1424–1442. doi:10.1002/cpt.2010 Ruskamp, F. M. (2010). A decade of safety-related regulatory action in the Netherlands: a Bjerre, L. M., Parlow, S., de Launay, D., Hogel, M., Black, C. D., Mattison, D. R., et al. retrospective analysis of direct healthcare professional communications from 1999 to (2018). Comparative, cross-sectional study of the format, content and timing of 2009. Drug Saf. 33 (6), 463–474. doi:10.2165/11532840-000000000-00000 medication safety letters issued in Canada, the USA and the UK. BMJ Open 8 (10), N’Guessan-Irié, A. G., Yavo, J. C., Guillaume Amari, A. S., and Yapi, A. D. (2012). e020150. doi:10.1136/bmjopen-2017-020150 Summarizing of medicinal alerts in ivory Coast from 2001 till 2010. Therapie 67 (3), Clavenna, A., and Bonati, M. (2009). Adverse drug reactions in childhood: a review of 251–256. doi:10.2515/therapie/2012029 prospective studies and safety alerts. Arch. Dis. Child. 94 (9), 724–728. doi:10.1136/adc. Notas informativas, medicamentos de uso humano. Agencia Española de Medicamentos 2008.154377 y Productos Sanitarios (2022). Available at: https://www.aemps.gob.es/acciones- Conroy, S., Choonara, I., ImPicciatore, P., Mohn, A., Arnell, H., Rane, A., et al. (2000). informativas/notas-informativas-medicamentos-de-uso-humano/?cat=266&tag= Survey of unlicensed and off label drug use in paediatric wards in European countries. seguridad-8 (accessed Oct 23, 2022). European Network for Drug Investigation in Children. BMJ 320 (7227), 79–82. doi:10. Onakpoya, I. J., Heneghan, C. J., and Aronson, J. K. (2016a). Worldwide withdrawal of 1136/bmj.320.7227.79 medicinal products because of adverse drug reactions: a systematic review and analysis. Davies, E. A., and O’Mahony, M. S. (2015). Adverse drug reactions in special Crit. Rev. Toxicol. 46 (6), 477–489. doi:10.3109/10408444.2016.1149452 populations - the elderly. Br. J. Clin. Pharmacol. 80 (4), 796–807. doi:10.1111/bcp.12596 Onakpoya, I. J., Heneghan, C. J., and Aronson, J. K. (2016b). Post-marketing withdrawal European Medicines Agency (2004). ICH topic E 2 D. Post approval safety data of 462 medicinal products because of adverse drug reactions: a systematic review of the management. CPMP/ICH/3945/03. Available at: https://www.ema.europa.eu/en/ world literature. BMC Med. 1417 (1), 1056. doi:10.1186/s12916-016-0553-2 documents/scientific-guideline/international-conference-harmonisation-technical- Perry, L. T., Bhasale, A., Fabbri, A., Lexchin, J., Puil, L., Joarder, M., et al. (2020). A requirements-registration-pharmaceuticals-human-use_en-12.pdf (accessed Oct 23, descriptive analysis of medicines safety advisories issued by national medicines regulators 2022). in Australia, Canada, the United Kingdom and the United States - 2007 to 2016. European Medicines Agency (2022). Medicines under additional monitoring. European Pharmacoepidemiol. Drug Saf. 29 (9), 1054–1063. doi:10.1002/pds.5072 Medicines Agency. Available at: https://www.ema.europa.eu/en/human-regulatory/post- Potts, J., Genov, G., Segec, A., Raine, J., Straus, S., and Arlett, P. (2020). Improving the authorisation/pharmacovigilance/medicines-under-additional-monitoring (accessed Oct safety of medicines in the European union: From signals to action. Clin. Pharmacol. Ther. 23, 2022). 107 (3), 521–529. doi:10.1002/cpt.1678 European Medicines Agency (2022a). Pharmacovigilance: Overview. European Rodriguez Pascual, A. (2006). “Alertas y reacciones adversas de medicamentos,” in Medicines Agency. Available at: https://www.ema.europa.eu/en/human-regulatory/ Curso de Actualización Pediatria (Madrid: Exlibris Ediciones), 69–76. En: AEPap ed. overview/pharmacovigilance-overview (accessed Oct 23, 2022). Rothwell, P. M. (2005). External validity of randomised controlled trials: "to whom do the European Medicines Agency (2022b). Guideline on good pharmacovigilance practices results of this trial apply? Lancet 365 (9453), 82–93. doi:10.1016/S0140-6736(04)17670-8 (GVP) Module IX—signal management. Available at: https://www.ema.europa.eu/en/ documents/scientific-guideline/guideline-good-pharmacovigilance-practices-gvp- Soares, S., Roque, F., Teixeira Rodrigues, A., Figueiras, A., and Herdeiro, M. T. (2015). module-ix-signal-management-rev-1_en.pdf (accessed Oct 23, 2022). Safety alerts: An observational study in Portugal. Clin. Ther. 37 (9), 2122–2128. doi:10. 1016/j.clinthera.2015.07.015 Farcaş, A., Mahalean, A., Bulik, N. B., Leucuta, D., and Mogoșan, C. (2018). New safety signals assessed by the pharmacovigilance risk assessment committee at EU level in 2014- The European Network of Centres for Pharmacoepidemiology and Pharmacovigilance 2017. Expert Rev. Clin. Pharmacol. 11 (10), 1045–1051. doi:10.1080/17512433.2018. (ENCePP). “ENCePP guide on methodological standards in Pharmacoepidemiology,” in 1526676 Chapter 15.4. Methods for pharmacovigilance impact research. Available at: https://www. encepp.eu/standards_and_guidances/methodologicalGuide15_4.shtml (Accessed Nov 7, Farcas, A., Balcescu, T., Anghel, L., Bucsa, C., and Mogoșan, C. (2020). A description 2022). of medicines-related safety issues evaluated through a referral procedure at the EU level after 2012. Expert Opin. Drug Saf. 19 (6), 755–762. doi:10.1080/14740338.2020. Uppsala Monitoring Centre (2022). The WHO programme for international drug 1744561 monitoring. Available at: https://www.who-umc.org/global-pharmacovigilance/who- programme-for-international-drug-monitoring/(accessed Oct 23, 2022). Ferreira, V. C., Ferreira, G. C., and Baldoni, A. O. (2020). Thirteen years of medication alerts issued by the Brazilian Health Regulatory Agency (ANVISA): What is the profile? van Hunsel, F., de Jong, E., Gross-Martirosyan, L., and Hoekman, J. (2021). Signals from J. Eval. Clin. Pract. 26 (3), 957–961. doi:10.1111/jep.13233 the Dutch national spontaneous reporting system: Characteristics and regulatory actions. Pharmacoepidemiol. Drug Saf. 30 (8), 1115–1122. doi:10.1002/pds.5246 Irl, C., and Hasford, J. (2020). Assessing the safety of drugs in pregnancy: the role of prospective cohort studies. Drug Saf. 22 (3), 169–177. doi:10.2165/00002018-200022030- WHO Collaborating Centre for Drug Statistics Methodology (2018). Guidelines for ATC 00001 classification and DDD assignment 2019. Oslo, Norway. Available at: http://www.whocc.no (accessed Oct 23, 2022). Frontiers in Pharmacology 12 frontiersin.org