ORIGINAL RESEARCH published: 05 May 2021 doi: 10.3389/fphar.2021.670945 Adverse Reactions to Drugs of Special Interest in a Pediatric Oncohematology Service Kristopher Amaro-Hosey 1,2†, Immaculada Danés 1,2,3*†, Lourdes Vendrell 1,2,3, Laura Alonso 3,4, Berta Renedo 5, Luis Gros 3,4, Xavier Vidal 1,2,3, Gloria Cereza 2,3,6‡ and Antònia Agustí 1,2,3‡ 1 Clinical Pharmacology Service, Vall d’Hebron University Hospital, Barcelona, Spain, 2Department of Pharmacology, Therapeutics and Toxicology, Universitat Autònoma de Barcelona, Barcelona, Spain, 3Vall d’Hebron Research Institute, Barcelona, Spain, 4Department of Pediatric Hematology and Oncology, Vall d’Hebron University Hospital, Barcelona, Spain, 5 Pharmacy Service, Vall d’Hebron University Hospital, Barcelona, Spain, 6Catalan Institute of Pharmacology Foundation, Vall Hebron University Hospital, Barcelona, Spain Edited by: Introduction: Drugs used in oncological diseases are frequently related to adverse drug Alfredo Vannacci, reactions (ADR). Few studies have analyzed the toxicity of cancer treatments in children in University of Florence, Italy real practice. Reviewed by: Georg Hempel, Methods: An observational, longitudinal and prospective study has been carried out in an University of Münster, Germany Oncohematology Service of a tertiary hospital. During 2017, patients exposed to one or Ashwin Karanam, Pfizer, United States more drugs of a previously agreed list were identified and followed-up for at least 6 months *Correspondence: each. Characteristics of ADR, incidence, causality and possible preventability, have been Immaculada Danés evaluated. id@icf.uab.cat † These authors have contributed Results: 72 patients have been treated with at least one study drug, and 159 ADR equally to this work and share first episodes involving at least one of these drugs have been identified, with a total of 293 ADR. authorship ‡ Most episodes required hospital admission (35.2%) or happened during the hospital stay These authors have contributed equally to this work and share last (33%), and 91.2% were severe. Blood disorders were the most frequent ADR (96; 32.8%), authorship related to thioguanine (42) and pegaspargase (39) mainly, followed by infections (86; 29.4%) related to thioguanine (32), pegaspargase (27), Erwinia asparaginase (14) and Specialty section: This article was submitted to rituximab (13). Two ADR were unknown. Most ADR were dose-dependent or expectable Obstetric and Pediatric Pharmacology, (>90%). The global incidence of ADR was 3.1/100 days at risk (95% CI 2.7–3.5), with 3.5 a section of the journal ADR/100 days at risk with pegaspargase (95% CI 2.9–4.2), 1.2/100 days at risk with Frontiers in Pharmacology rituximab (95% CI 0.8–1.8) and 11.6/100 days at risk with thioguanine (95% CI 9.4–14.2). Received: 22 February 2021 Accepted: 22 April 2021 Controversial additional measures of prevention, other than those already used, were Published: 05 May 2021 identified. Citation: Amaro-Hosey K, Danés I, Vendrell L, Conclusion: ADR are frequent in pediatric oncohematological patients, mainly blood Alonso L, Renedo B, Gros L, Vidal X, disorders and infectious diseases. Findings regarding incidence and preventability may be Cereza G and Agustí A (2021) Adverse Reactions to Drugs of Special Interest useful to compare data between different centers and to evaluate new possibilities for in a Pediatric action or prevention. Oncohematology Service. Front. Pharmacol. 12:670945. Keywords: pharmacovigilance, adverse drug reactions, pediatrics, hematology, neoplasms, pegaspargase, doi: 10.3389/fphar.2021.670945 thioguanine, incidence Frontiers in Pharmacology | www.frontiersin.org 1 May 2021 | Volume 12 | Article 670945 Amaro-Hosey et al. Drug Toxicity in Pediatric Oncohematology INTRODUCTION knowing better its safety profile in pediatric population in clinical practice. As secondary objectives, we wanted to Adverse drug reactions (ADR) are a major cause of morbidity and identify possible preventive measures, to analyze the mortality in adults and children. Between 2 and 5% of hospital proportion of patients with ADR in whom the use of the admissions in children are due to ADR (Impicciatore et al., 2001), drugs was off-label and to improve the identification of and its incidence among hospitalized pediatric population has suspected ADR in the Pediatric Oncohematology service. been described as 0.6–17.7% (Gallagher et al., 2011; Stausberg and Hasford., 2011; Gallagher et al., 2012; Posthumus et al., 2012; Thiesen et al., 2013). The highest percentage was documented in a ARTICLE TYPES study carried out in a United Kingdom pediatric hospital, in which the oncological treatment was identified as a risk factor for Original Research. ADR (HR 1.90; 95% CI 1.40–2.60) (Thiesen et al., 2013). In spite of that, there are few studies in which the impact of oncological treatments in children has been quantified or analyzed, and their METHODS methodology has varied. A retrospective study including a random sample of adult and children hospitalized at a An observational, longitudinal and prospective study was university oncology centre showed that 22% of ADR happened carried out in the Pediatric Oncohematology Service of Vall in patients younger than 20 (Vaseghi et al., 2016). Cisplatin d’Hebron University Hospital. After reviewing drug (44%), doxorubicin (24%) and 5-fluouracil (20%) were the consumption during 2016 in the Pediatric Oncohematology chemotherapeutic drugs most commonly involved and the Service, a list of drugs with an incomplete knowledge about principal symptoms were nausea, vomiting, neutropenia and their safety profile in the general population or specifically in constipation, both in adults and in pediatric population. With children was identified and agreed between all researchers, and these results the authors could identify prevention strategies to selected to be monitored. Drugs selected were those used during reduce their incidence. Other studies have been focused on the previous year with one or more of the following validating indirect tools or systems for detecting possible characteristics: marketed in exceptional circumstances, newly preventable unwanted effects in oncohematology services by authorized with conditional approval, subject to special follow identifying drug use (protamine, vitamin K, naloxone, up by the European Medicines Agency (EMA), not marketed in flumazenil or hyaluronidase) commonly prescribed when ADR Spain, and drugs frequently used off-label or with little with certain medications occur (Call et al., 2014; Hébert et al., experience in children. Based on these criteria, the following 2015). Other factors that may affect toxicity are the use of recently drugs were selected: clofarabine, nelarabine, thioguanine, marketed drugs with incomplete knowledge of the toxicity profile crizotinib, pegaspargase, asparaginase, defibrotide, infliximab, and off-label drug use (Bellis et al., 2013; Saiyed et al., 2015). rituximab, bevacizumab, imatinib, dasatinib, nilotinib, At the end of 2013, the European Union initiated a new eltrombopag and romiplostim. It is important to note that procedure to identify drugs undergoing particularly rigorous some drugs such as thioguanine, defibrotide and asparaginase monitoring by health authorities (Agencia Española de are not marketed in Spain, and neither was pegaspargase when Medicamentos y Productos Sanitarios, 2013; European the study began. Medicines Agency, 2013). These drugs are subject to During 2017, all patients exposed to at least one drug of the “additional monitoring”, and a black inverted triangle is list were identified at the Pharmacy Service based on the displayed in their package leaflet and in the product summary, prescription requests received daily, and were followed for a together with a short sentence explaining the triangle meaning. minimum of 6 months in order to detect possible ADR. This is generally because there is less information available on it Patients with at least one ADR suspicion were identified than on other medicines, for example because it is new to the using different sources (notification by clinicians attending market or there is limited data on its long-term use. This includes, the patients and/or patients monitoring throughout the for example, new active ingredients, including biological clinical records). ADR were grouped as an episode when medicines, authorized for the first time in the EU after more than one ADR coincided in time in the same patient, January 1, 2011, medicines to which a conditional marketing and the suspected drugs (even in polytherapy) and involved authorization has been granted, have been approved in mechanisms were also considered coincident when assessing the exceptional circumstances, or have been authorized with causal relationship according to the pharmacovigilance specific obligations on the recording or suspected ADR, methods used (detailed in the next paragraph). This allows to sometimes based on advice from the Agency’s have detailed information according to ADR and also about Pharmacovigilance Risk Assessment Committee (PRAC). events or clinical situations with ADR. For example, if a patient For this study the knowledge of the incidence and had thrombocytopenia and hemorrhage, or febrile neutropenia characteristics of ADR related to medications of special and a certain infection, two ADR were considered but only one interest in safety matters in a third level hospital was ADR episode in each case. The number and characteristics of prioritized. The main objective of the study was to evaluate ADR episodes were registered. Cytopenias were only recorded if the incidence and characteristics of suspected ADR to they caused some additional medical act such as tests or medications of special interest, with the final purpose of treatments such as antibiotics, transfusions or a delay in Frontiers in Pharmacology | www.frontiersin.org 2 May 2021 | Volume 12 | Article 670945 Amaro-Hosey et al. Drug Toxicity in Pediatric Oncohematology chemotherapy administration. Demographic data, clinical TABLE 1 | Baseline characteristics of the cohort of patients exposed to study drugs. diagnosis, preventive measures used and other variables necessary to evaluate ADR causality were also obtained from Patients (n = 72) the review of the medical records. Information on prior Age (median [IQR]) years 6 (3.1–10.2) knowledge of ADR suspicions was obtained from drug <2 years, n (%) 9 (12.5%) datasheets and the review of the published bibliography. Use 2–12 years, n (%) 50 (69.4%) was considered off-label if indication, population or dosage were >12 years, n (%) 13 (18.1) different from the medication data sheet approved by the Gender regulatory authority. Female, n (%) 29 (40.3%) Male, n (%) 43 (59.7%) ADR were classified according to the Medical Dictionary for Exposition to study drugsa ® Regulatory Activities MedDRA (Medical Dictionary for Regulatory Activities, 2020). The plausibility of the 1 drug, n (%) 2 drugs, n (%) 27 (37.5) 37 (51.4) association between the ADR and drugs has been analyzed 3 drugs, n (%) 8 (11.1) Indication of study drugs (n) and discussed, according to the Spanish Pharmacovigilance Acute lymphoblastic leukemia 43 System algorithm, which takes into account the temporal Hepatic veno-occlusive disease 12 sequence, prior knowledge of the adverse reaction to the Epstein-Barr virus infection 7 drug, drug withdrawal effect, drug re-exposure effect (if Idiopathic thrombocytopenic purpura 5 available), and possible alternative causes of symptoms Lymphoblastic lymphoma 3 Autoimmune hemolytic anemia 3 (Aguirre and García, 2016). It has also been used to classify Chronic graft-versus-host disease 2 previous knowledge of the ADR in well-known ADR, known Chronic myeloid leukemia 1 from anecdotal reports and unknown. The seriousness of the Dermatofibrosarcoma protuberans 1 ADR was classified in accordance with to the European Union’s Aplastic anemia 1 criteria: ADR were considered serious (when they result in a Fourteen out of 37 patients exposed to 2 study drugs (pegaspargase and thioguanine in death, were life-threatening, required hospitalization or eight patients) had some period of coincidence of risk time. prolonged an existing hospitalization, resulted in persisting disability or were important medical events) or non-serious (the remaining cases) (European Medicines Agency, 2004). risk. 95% confidence intervals were estimated from Poisson Incidence of ADR was calculated taking into account the distribution. number of days of exposure to each drug. Time at risk with The study was conducted according to international ethical each drug was considered taking into account the interval for recommendations and was approved by the local Research Ethics drug administration (related to pharmacokinetic and Committee following the national directives related to post- pharmacodynamic parameters) of drug as the main guide. authorization studies. Exposure was considered continued during all treatment as long as the normal interval of drug administration was maintained, and when a longer interruption happened or RESULTS the treatment ended, the same number of days after the last dose were considered as a time at risk. Normal interval for A total of 72 patients treated with at least one study drug were administration is 14 days for pegaspargase, 2 days for Erwinia identified. Demographic characteristics of these patients and asparaginase, seven for romiplostim, twenty-one for medical conditions for which study drugs were used are nelarabine (with a short half-live but high intracellular shown in Table 1. Patients were followed for a median of accumulation of ara-GTP), 8 weeks (60 days) for 388.5 days (IQR 252–483.5). Table 2 shows the distribution of infliximab and 6 months for rituximab (the interval of drugs and data of exposure. Forty-five patients (62.5%) were administration in some indications, when the CD20 cells exposed to more than one study drug. There were no cases recovery starts) (Yun et al., 2015; Aaltonen et al., 2015; exposed to nilotinib, crizotinib, clofarabine or bevacizumab. European Medicines Agency, 2021). For the rest of drugs, The drugs with a longer total time at risk were pegaspargase administered daily, the treatment period was considered. (3,539 days), rituximab (2,051), thioguanine (794), eltrombopag When one patient was exposed to more than one study (768) and imatinib (597). Study drugs were used to treat 10 drug and time at risk coincided at some point, the same different diseases; the most frequent were acute lymphoblastic time at risk was considered for each drug to calculate leukemia (43), hepatic veno-occlusive disease (12) and Epstein- incidence data. An electronic database was set up and Barr virus infection (7). Thirty-three percent of patients with analyzed with the statistical package SAS 9.4 (SAS® Institute Inc., Cary, NC, United States). Continuous and leukemia had a high-risk disease. Fifty-two out of 72 patients (72.2%) presented some ADR in count variables were summarized using medians with 25th which a study drug was considered suspicious. Median age of and 75th percentiles, and categorical factors were reported these patients was 6.8 years (IQR 3.4–11.5) and 58% were male. using frequencies and percentages. Incidence rates were The total number of ADR episodes was 159, with a median of two calculated by dividing the number of ADR by the episodes per patient (IQR 0–3), and 3 (IQR 2–6) when analysis corresponding time at risk, and expressed for 100 days at was performed only in patients with high-risk leukemia. Most Frontiers in Pharmacology | www.frontiersin.org 3 May 2021 | Volume 12 | Article 670945 Amaro-Hosey et al. Drug Toxicity in Pediatric Oncohematology TABLE 2 | Exposure to study drugs. Drugs Time Number Total number of Median number of at risk considereda of exposed patients days of exposure days of exposure (IQR) Pegaspargase During treatment at normal interval +14 days 38 3,539 63 (42–240) Thioguanine During treatment at normal interval 37 794 15 (14–25) Defibrotide During treatment at normal interval 12 320 25.5 (16.5–31.5) Erwinia asparaginase During treatment at normal interval +2 days 11 384 24 (24–36) Rituximab During treatment at normal interval +180 days 10 2,051 194 (188–201) Eltrombopag During treatment at normal interval 5 768 113 (67–178) Romiplostim During treatment at normal interval +7 days 3 346 62 (56–228) Nelarabine During treatment at normal interval +21 days 2 102 51 (50–52) E. coli asparaginase During treatment at normal interval 2 16 8 (4–12) Imatinib During treatment at normal interval 2 597 298.5 (113–484) Infliximab During treatment at normal interval +60 days 2 168 84 (60–108) Dasatinib During treatment at normal interval 1 402 402 (402–402) a Additional days were considered when treatment ended or an interruption longer than the interval of administration happened. is available as Supplementary Material). Specifically, the most TABLE 3 | Number of ADR with study drugs as suspects. reported ADR (accounting for almost 40% of ADR) were febrile Drugs Number of ADR neutropenia, pancytopenia, thrombocytopenia, fever, hypertriglyceridemia and agranulocytosis. Suspicious drugs Pegaspargase 124 (39.2) Thioguanine 92 (29.2) considered possibly involved in these ADR are detailed in Erwinia asparaginase 47 (14.9) Table 5. Rituximab 25 (7.9) Hypersensitivity/allergic reactions accounted for 9.5% of the Defibrotide 9 (2.9) episodes, while the rest of ADRs (90.5%) were considered dose- E. coli asparaginase 6 (1.9) Eltrombopag 4 (1.3) related or due to mechanisms other than hypersensitivity. For Dasatinib 3 (1.0) cytopenias and infections, the mechanism is well known; Nelarabine 2 (0.6) however, for other common complications such as Imatinib 2 (0.6) hypertriglyceridemia or pancreatitis with pegaspargase, the Infliximab 1 (0.3) mechanism is not fully established (genetic predisposition has Romiplostim 0 been suggested) but hypersensitivity is highly unlikely (Zhang et al., 2020). The incidence of ADR was 3.1/100 days at risk (95% CI 2.7–3.5), and for episodes of ADR incidence was 1.7/100 days episodes (90.5%) were identified by monitoring clinical records. at risk (95% CI 1.4–2). The incidence of ADR with the drugs with Fifty-six episodes required hospital admission (35.2%), 33.3% a longer total time at risk in this study was: 3.5 ADR/100 days at happened during the hospital stay, and 31.5% were managed on risk with pegaspargase (95% CI 2.9–4.2), 1.2/100 days at risk an outpatient basis. The majority of episodes of ADR were serious with rituximab (95% CI 0.8–1.8) and 11.6/100 days at risk with (145; 91.2%): the principal reasons for severity were the thioguanine (95% CI 9.4–14.2). More specifically, the most requirement of hospitalization or prolongation of an existing frequent ADR with pegaspargase were blood disorders (1.1/ hospitalization (60; 37.7% of episodes) and the consideration of 100 days at risk; 95% CI 0.8–1.5) and infections (0.8/100 days the episode medically important (60; 37.7% of episodes). Eight at risk; 95% CI 0.5–1.1); with rituximab ADR were mainly episodes were life-threatening, but no deaths due to adverse infections (0.6/100 days at risk; 95% CI 0.4–1.1), and with effects were registered. thioguanine, blood disorders (5.3/100 days at risk; 95% CI Two-hundred ninety-three ADR were described in the 159 3.9–7.2) and infections (4/100 days at risk; 95% CI 2.9–5.7). episodes. Patients were receiving a median of seven drugs per There were five episodes of ADR despite using specific episode (IQR 6–9) and were considered suspicious a median of 3 prophylaxis to prevent them: three patients with herpes (IQR 2–4). At least one drug of the study list was involved in these infection were receiving acyclovir and two patients with reactions, being pegaspargase and thioguanine the most frequent rituximab infusion-related disorders had received (Table 3). Fifteen percent of suspicious study drugs had been premedication with antihistamines and analgesics. No other given off-label, mainly rituximab for Epstein-Barr virus infection prophylactic methods were applied in the rest of patients with (on 12 different occasions) or for autoimmune hemolytic anemia ADR. The prevention of ADR according to individual (4). As shown in Table 4, the principal ADR with study drugs susceptibility could have been carried out in 30 episodes of were blood disorders related to thioguanine, pegaspargase and hematological adverse reactions to thioguanine by phenotyping Erwinia asparaginase, infections with thioguanine, pegaspargase, of thiopurine methyltransferase. In eight episodes of ADR Erwinia asparaginase and rituximab, and gastrointestinal or consisting of neutropenia/agranulocytosis in patients with high metabolic disorders with pegaspargase (additional information risk leukemia during high-risk blocks, prophylaxis with Frontiers in Pharmacology | www.frontiersin.org 4 May 2021 | Volume 12 | Article 670945 Amaro-Hosey et al. Drug Toxicity in Pediatric Oncohematology TABLE 4 | Principal ADR according to the affected system and suspicious study drugsa. ADR n (%) Suspicious study drugs Other suspicious drugs in ≥5 ADR Blood and lymphatic system 96 (32.8) Thioguanine (42), pegaspargase (39), Erwinia asparaginase Cytarabine (53), cyclophosphamide (37), vincristine (38), disorders (19), defibrotide (3), rituximab (3), dasatinib (2), E. coli dexamethasone (36), methotrexate (20), mercaptopurine (15), asparaginase (2) doxorubicin (16), daunorubicin (8), etoposide (8), methylprednisolone (6) Infections and infestations 86 (29.4) Thioguanine (32), pegaspargase (27), Erwinia asparaginase Cytarabine (38), vincristine (29), cyclophosphamide (24), (14), rituximab (13), E. coli asparaginase (2), defibrotide (1), dexamethasone (23), methotrexate (18), mercaptopurine (13), eltrombopag (1), infliximab (1) daunorubicin (10), methylprednisolone (10), doxorubicin (9), mycophenolic acid (6) Gastrointestinal disorders 23 (7.8) Pegaspargase (11), thioguanine (6), Erwinia asparaginase Cytarabine (11), dexamethasone (11), vincristine (11), (5), E. coli asparaginase (1) methotrexate (10), cyclophosphamide (7), mercaptopurine (6) Metabolism and nutrition 22 (7.5) Pegaspargase (21), Erwinia asparaginase (1) Dexamethasone (11) disorders General disorders and 16 (5.5) Thioguanine (7), pegaspargase (6), rituximab (2), Erwinia Cyclophosphamide (8), cytarabine (8), vincristine (6), administration site conditions asparaginase (1), E. coli asparaginase (1), imatinib (1) dexamethasone (5) Skin and subcutaneous tissue 13 (4.4) Pegaspargase (9), Erwinia asparaginase (2), defibrotide (1), disorders thioguanine (1) Respiratory, thoracic and 10 (3.4) Rituximab (4), thioguanine (3), pegaspargase (2), Erwinia mediastinal disorders asparaginase (1), nelarabine (1) Other 27 (9.2) a More than one drug could be involved in an ADR. TABLE 5 | ADR and suspicious drugsa. ADR according to n (%) Suspicious study drugs Other suspicious drugs the affected system in ≥5 ADR Febrile neutropenia 27 (9.2) Thioguanine (11), pegaspargase (9), Erwinia asparaginase Vincristine (16), cytarabine (15), dexamethasone (13), (8), E. coli asparaginase (1) cyclophosphamide (8), daunorubicin (6), doxorubicin (6), methylprednisolone (5) Pancytopenia 25 (8.6) Thioguanine (13), pegaspargase (12), Erwinia Cytarabine (19), cyclophosphamide (14), dexamethasone (13), asparaginase (4) vincristine (10), methotrexate (6), doxorubicin (5) Thrombocytopenia 22 (7.5) Pegaspargase (10), thioguanine (9), Erwinia asparaginase Cytarabine (12), cyclophosphamide (8), methotrexate (6), (4), defibrotide (2), dasatinib (1), rituximab (1) mercaptopurine (5) Fever 13 (4.4) Thioguanine (7), pegaspargase (4), rituximab (2), E. coli Cyclophosphamide (8), cytarabine (8), vincristine (6), asparaginase (1), Erwinia asparaginase (1) dexamethasone (5) Hypertriglyceridemia 13 (4.4) Pegaspargase (13) Agranulocytosis 12 (4.1) Thioguanine (8), Erwinia asparaginase (3), Cyclophosphamide (7), cytarabine (6) pegaspargase (3) Respiratory tract infection 7 (2.4) Pegaspargase (5), thioguanine (2), rituximab (1) Anemia 5 (1.7) Pegaspargase (3), defibrotide (1), rituximab (1), thioguanine (1) Stomatitis 5 (1.7) Erwinia asparaginase (2), pegaspargase (2), thioguanine (1) Respiratory tract viral 5 (1.7) Pegaspargase (2), Erwinia asparaginase (1), rituximab (1), infection thioguanine (1) Rhinovirus infection 5 (1.7) Thioguanine (3), pegaspargase (2), rituximab (1) Mucosal inflammation 5 (1.7) Pegaspargase (2), thioguanine (2), Erwinia asparaginase (1) Other 149 (50.9) a More than one drug could be involved in an ADR. granulocyte colony-stimulating factor could have been study drugs were detected, with a median number of two episodes considered. per patient (IQR 1–4). One-hundred forty-three (80%) were ADR with drugs of the list were previously unknown or poorly considered serious, mainly because hospitalization was required known: papular acne related to Erwinia asparaginase and or were considered medically important (110; 61.5%). Two- pneumatosis intestinalis related to pegaspargase. Additionally, hundred and ninety-five ADR were described in these episodes, during the follow-up of the global cohort of 72 patients exposed all of them previously known. The principal ADR according to the to some study drug, 179 episodes involving drugs different from affected system were infections and infestations (80; 27.1%), blood Frontiers in Pharmacology | www.frontiersin.org 5 May 2021 | Volume 12 | Article 670945 Amaro-Hosey et al. Drug Toxicity in Pediatric Oncohematology and lymphatic system disorders (73; 24.7%), gastrointestinal Individual Case Safety Reports concerning antineoplastic drugs in disorders (34; 11.5%), general disorders and administration site children and found that only 18 (7.6%) out of 236 were classified conditions (18; 6.1%), hepatobiliary disorders (17; 5.8%), and skin as off-label cases (Mascolo et al., 2020). These authors considered and subcutaneous tissue disorders (14; 4.7%). The most reported the low number of cases were due to underreporting. ADR were pancytopenia (56), anemia (46), stomatitis (38), Rituximab was the most frequently drug used off-label. In thrombocytopenia (36), and hepatitis (35). Median number of children, rituximab has been used in a wide range of autoimmune suspicious drugs per episode was 2 (IQR 1–3). The drugs more diseases that require lymphocyte B depletion, and observational commonly involved as suspicious were mercaptopurine (in 115 studies have been published in patients with immune ADR), cytarabine (85), methotrexate (76), cyclophosphamide (51), thrombocytopenia, hemolytic anemia, multiple sclerosis or dexamethasone (49), and cyclosporine (41). Additional nephrotic syndrome (Zecca et al., 2003; O’Connor et al., 2013; information on the most frequent ADR and suspicious drugs is Maratea et al., 2016; Makis et al., 2018). The main indications for provided as Supplementary Material. rituximab in our cohort were the treatment of Epstein-Barr virus infection and hemolytic anemia. Weekly administrations of 1–3 doses were used and the principal ADR described during the DISCUSSION follow-up were infections. In studies published with rituximab use to prevent Epstein-Barr virus related post-transplant To our knowledge, this is the first prospective study in pediatric lymphoproliferative disorders in allogeneic hematopoietic stem oncohematological population assessing the incidence and cell transplantation, little information is provided on specific characteristics of ADR in real practice to specific drugs of side-effects of rituximab (Lindsay et al., 2020; Kiskaddon et al., interest in pharmacovigilance. This intensive 18 months 2021). In one study with 78 patients with Epstein-Barr virus follow-up study has shown that three out of four patients reactivation treated with rituximab, a higher 36-month treated had at least one ADR to a study drug considered cumulative incidence of bacterial, viral, and fungal infections suspicious, with a mean number of two episodes per patient. was described compared to a control group of patients not treated Pegaspargase and thioguanine have been the drugs most with rituximab, and a slower reconstitution of B cells was observed frequently involved (68%), and blood disorders and infections (Petropoulou et al., 2012). In other series of patients with were the most frequently reported ADR (78%). autoimmune hemolytic anemia, infective complications were Drugs used in cancer diseases are described as a risk factor of unusual, probably because patients were less exposed to ADR occurrence, but only very few pharmacovigilance studies concomitant cytotoxic drugs than those with hematological have assessed ADR in oncohematological pediatric patients. malignancies in whom Epstein-Barr virus was detected, and Methodology used in these studies is very heterogeneous and possibly because immunoglobulin replacement therapy was therefore, the results are also highly variable. Barret et al. administered. In a cohort of 61 children with autoimmune described a 14.4–23.5% annual rate incidence of adverse hemolytic anemia treated with weekly infusions of rituximab for events in a 7 years retrospective study by using databases four weeks, two cases of allergic reactions were described, one (Barrett et al., 2013). Queuille et al., in a prospective 50 days patient died due to agranulocytosis and sepsis probably related with study in inpatients, described at least one ADR in 65% of rituximab treatment, and 16 cases of hypogammaglobulinemia were inpatients (Queuille et al., 2001). In a retrospective study identified, and partially treated with immunoglobulin replacement carried out in pediatric cancer patients admitted at two in 13 of them for at least 5 months (Ducassou et al., 2017). In hospitals in Ethiopia during a 2 years period, ADR were another prospective study in 15 children with autoimmune described in 41.5% of patients, with a higher risk in patients hemolytic anemia, three moderate side effects during rituximab taking four or more chemotherapy agents and regimens based on infusion and one case of varicella zoster infection 2 months after etoposide, mercaptopurine, and doxorubicin (Workalemahu treatment were described (Zecca et al., 2003). In this study, although et al., 2020). In our study, a 72.2% (52 out of 72) ADR benefits and risks of this practice have not been proven, all children incidence was described, however our population was limited received intravenous immunoglobulin for 6 months to prevent to a list of specific drugs and follow-up time was different. Also, hypogammaglobulinemia after completing treatment. we have been able to assess incidence rates during time at risk of Nine out of ten ADR episodes were considered severe; no significant blood disorders or infections in clinical practice during other study has evaluated the severity of ADRs in pediatric treatment with specific drugs such as pegaspargase, rituximab, oncohematological patients taking into account current criteria and thioguanine even though, like in a clinical trial context, these of severity (European Medicines Agency, 2004). The proportion drugs were not given alone and other drugs can have contributed of episodes that required hospital admission, occurred during the to these incidence data. hospital stay, or on an outpatient basis was similar. In studies with Fifteen percent of the study drug treatments were used off- a different approach including only children with label, a smaller proportion than expected. The drugs administered oncohematological diseases admitted to the hospital, the off-label were different from the drugs used according to the proportion of ADR that caused admission during a specific labeling. For both reasons, the comparison of the incidence of period was 41% compared to 59% that occurred during ADR between the use of off-label and on-label drugs was hospitalization (Collins et al., 1974). If in our study we considered inappropriate. A similar trend was documented in exclude patients managed on an outpatient basis, these the study of Mascolo et al., which aimed to assess off-label use of percentages are quite similar: 51 and 49%, respectively. Only Frontiers in Pharmacology | www.frontiersin.org 6 May 2021 | Volume 12 | Article 670945 Amaro-Hosey et al. Drug Toxicity in Pediatric Oncohematology 9.5% of episodes of ADR were considered due to hypersensitivity useful to discuss the convenience to review protocols for most in our study compared to nearly 20% of hospital-acquired ADR in risky patients and the possibility to include new techniques such the study by Collins et al., with different drugs and methods. TMPT testing in our hospital. The ADR profile identified indicates a clear predominance of In conclusion, the main ADRs related to the study drugs in our hematologic and infectious reactions; also, some characteristic study are blood disorders and infections related to treatments that ADR such as hypertriglyceridemia due to pegaspargase has been include pegaspargase and thioguanine. The incidence data registered frequently. Gastrointestinal disorders have been less described can be useful to compare toxicity between different frequent with mucositis being the most common one (oral or anal centers and to assess new possibilities for action or prevention, in mucositis accounted for 61% of gastrointestinal disorders). Only order to reduce these risks and improve quality of life, always two cases of vomiting were identified (and no cases of nausea). maintaining efficacy. Neutrophil count decreases and alanine aminotransferase increases represented the most frequently identified ADR in a study focused on the identification of ADR with drug DATA AVAILABILITY STATEMENT combinations (Barrett et al., 2013). In an old study, Collins et al. reported vomiting as the principal ADR (Collins et al., The raw data supporting the conclusions of this article will be 1974). Drug characteristics can be in part responsible for this made available by the authors, without undue reservation. difference, but also the fact that chemotherapy-related nausea and vomiting prevention has improved substantially. On the other hand, prevention of hematological toxicity and infections are now ETHICS STATEMENT goals not fully addressed. Some thioguanine-related hematological ADR could possibly had been prevented by The studies involving human participants were reviewed and implementing thiopurine methyltransferase testing (TMPT) in approved by Vall d’Hebron Ethics Committee. Written informed patients with leukemia (Lennard, 2014; Lennard et al., 2015). This consent from the participants’ legal guardian/next of kin was not test for now is not performed in our center and when the results required to participate in this study in accordance with the sent to other centers are available the dose has already been national legislation and the institutional requirements. adjusted. Regarding bacterial infections, different antibiotic prophylaxis schemes (including ciprofloxacin, levofloxacin or amoxicillin-clavulanate) have shown in clinical trials a AUTHOR CONTRIBUTIONS reduction of episodes of fever and infections compared to placebo, but a significant reduction of mortality has not been ID, LV, GC and AA contributed to conception and design of the achieved (Castagnola et al., 2003; Laoprasopwattana et al., 2013; study. KA-H, LV, ID and GC organized the database. KA-H, LV, Alexander et al., 2018). The expected risk of an increase of BR, LA and LG contributed to the recording of the data. resistant strains prevents their recommendation as a routine. Substantial contributions to the analysis or interpretation of For this reason, antimicrobial prophylaxis (other than data for the work were made by KA-H, LV, GC, ID, and AA. trimetroprim/sulphamethoxazole to prevent Pneumocystis XV performed the statistical analysis. ID, KA-H, LV and GC jirovecii infection, acyclovir for herpes zoster prevention and wrote the first draft of the manuscript. All authors substantially antifungal in high-risk patients) is not used in our center contributed to the manuscript revision, read, and approved the either. Granulocyte colony-stimulating factor is only used submitted version. prophylactically in high-risk blocks (SEHOP/PETHEMA, 2013). One limitation of our study is the selection of patients treated with specific drugs. This has allowed a more intensive follow-up FUNDING of treated patients and the possibility to assess incidence data of patients treated with these drugs, but prevents a fair comparison Project PI16/02018, funded by Instituto de Salud Carlos III and with other studies. In order to improve the drug safety analysis in co-funded by European Union (ERDF/ESF). children with oncohematological diseases, guidelines would be a good tool not only to assess ADR but also to compare incidence and prevalence of ADR between centers to identify areas for ACKNOWLEDGMENTS improvement. During and after the study, the identification of adverse reactions in general in the pediatric oncohematology We thank Xavier Barroso (Catalan Institute of Pharmacology service has increased compared to previous years. Although the Foundation) for his technical support with the database. effect is expected to be temporary, we believe that more awareness has been raised about ADR and the desirability of looking for the most appropriate way to make their identification useful. SUPPLEMENTARY MATERIAL Multidisciplinary management and constant communication between teams should be encouraged in order to provide high The Supplementary Material for this article can be found online at: quality health assistance. On the other hand, the study was not https://www.frontiersin.org/articles/10.3389/fphar.2021.670945/ designed to assess the effect of measures of prevention but can be full#supplementary-material Frontiers in Pharmacology | www.frontiersin.org 7 May 2021 | Volume 12 | Article 670945 Amaro-Hosey et al. Drug Toxicity in Pediatric Oncohematology REFERENCES Focused Trigger Tool. Eur. J. Cancer 51, 427–435. doi:10.1016/j.ejca.2014. 12.002 Impicciatore, P., Choonara, I., Clarkson, A., Provasi, D., Pandolfini, C., and Bonati, Aaltonen, K. J., Joensuu, J. T., Virkki, L., Sokka, T., Aronen, P., Relas, H., et al. M. (2001). Incidence of Adverse Drug Reactions in Paediatric In/out-Patients: a (2015). Rates of Serious Infections and Malignancies Among Patients with Systematic Review and Meta-Analysis of Prospective Studies. Br. J. Clin. Rheumatoid Arthritis Receiving Either Tumor Necrosis Factor Inhibitor or Pharmacol. 52, 77–83. doi:10.1046/j.0306-5251.2001.01407.x Rituximab Therapy. J. Rheumatol. 42, 372–378. doi:10.3899/jrheum.140853 Kiskaddon, A. L., Landmesser, K., Carapellucci, J., Wisotzkey, B., and Asante- Agencia Española de Medicamentos y Productos Sanitarios. (2013). Medicamentos Korang, A. (2021). Expanded Utilization of Rituximab in Paediatric Cardiac Sujetos a Un Seguimiento Adicional. Available at:http://www.aemps.gob.es/vigilancia/ Transplant Patients. J. Clin. Pharm. Ther. doi:10.1111/jcpt.13346 medicamentosUsoHumano/seguimiento_adicional.htm#med_ssa (Accessed Laoprasopwattana, K., Khwanna, T., Suwankeeree, P., Sujjanunt, T., Tunyapanit, December 4, 2020). W., and Chelae, S. (2013). Ciprofloxacin Reduces Occurrence of Fever in Aguirre, C., and García, M. (2016). Evaluación de la causalidad en las Children with Acute Leukemia Who Develop Neutropenia during comunicaciones de reacciones adversas a medicamentos. Algoritmo del Chemotherapy. Pediatr. Infect. Dis. J. 32, e94–8. doi:10.1097/INF. Sistema Español de Farmacovigilancia. Med. Clin. (Barc.) 147, 461–464. 0b013e3182793610 doi:10.1016/j.medcli.2016.06.012 Lennard, L., Cartwright, C. S., Wade, R., and Vora, A. (2015). Thiopurine Alexander, S., Fisher, B. T., Gaur, A. H., Dvorak, C. C., Villa Luna, D., Dang, H., Methyltransferase and Treatment Outcome in the UK Acute Lymphoblastic et al. (2018). Effect of Levofloxacin Prophylaxis on Bacteremia in Children with Leukaemia Trial ALL 2003. Br. J. Haematol. 170, 550–558. doi:10.1111/bjh. Acute Leukemia or Undergoing Hematopoietic Stem Cell Transplantation. 13469 JAMA 320, 995–1004. doi:10.1001/jama.2018.12512 Lennard, L. (2014). Implementation of TPMT Testing. Br. J. Clin. Pharmacol. 77, Barrett, J. S., Patel, D., Dombrowsky, E., Bajaj, G., and Skolnik, J. M. (2013). Risk 704–714. doi:10.1111/bcp.12226 Assessment of Drug Interaction Potential and Concomitant Dosing Pattern on Lindsay, J., Yong, M. K., Greenwood, M., Kong, D. C. M., Chen, S. C. A., Rawlinson, Targeted Toxicities in Pediatric Cancer Patients. AAPS J. 15, 775–786. doi:10. W., et al. (2020). Epstein-Barr Virus Related Post-transplant 1208/s12248-013-9489-z Lymphoproliferative Disorder Prevention Strategies in Allogeneic Bellis, J. R., Kirkham, J. J., Thiesen, S., Conroy, E. J., Bracken, L. E., Mannix, H. L., Hematopoietic Stem Cell Transplantation. Rev. Med. Virol. 30, e2108. et al. (2013). Adverse Drug Reactions and Off-Label and Unlicensed Medicines doi:10.1002/rmv.2108 in Children: a Nested Case?control Study of Inpatients in a Pediatric Hospital. Makis, A., Kanta, Z., Kalogeropoulos, D., and Chaliasos, N. (2018). Anti-CD20 BMC Med. 11, 238. doi:10.1186/1741-7015-11-238 Treatment of Autoimmune Hemolytic Anemia Refractory to Corticosteroids Call, R. J., Burlison, J. D., Robertson, J. J., Scott, J. R., Baker, D. K., Rossi, M. G., et al. (2014). and Azathioprine: A Pediatric Case Report and Mini Review. Case Rep. Adverse Drug Event Detection in Pediatric Oncology and Hematology Patients: Using Hematol. 2018 2018, 1–7. doi:10.1155/2018/8471073 Medication Triggers to Identify Patient Harm in a Specialized Pediatric Patient Maratea, D., Bettio, M., Corti, M. G., Montini, G., and Venturini, F. (2016). The Population. J. Pediatr. 165, 447–452 e4. doi:10.1016/j.jpeds.2014.03.033 Efficacy and Safety of Rituximab in Treating Childhood Nephrotic Syndrome: Castagnola, E., Boni, L., Giacchino, M., Cesaro, S., De Sio, L., Garaventa, A., et al. an Italian Perspective. Ital. J. Pediatr. 42, 63. doi:10.1186/s13052-016-0271-6 (2003). Infectious Diseases Study Group of the Italian Association of Pediatric Mascolo, A., Scavone, C., Bertini, M., Brusco, S., Punzo, F., Pota, E., et al. (2020). Hematology and OncologyA Multicenter, Randomized, Double Blind Placebo- Safety of Anticancer Agents Used in Children: A Focus on Their Off-Label Use Controlled Trial of Amoxicillin/clavulanate for the Prophylaxis of Fever and through Data from the Spontaneous Reporting System. Front. Pharmacol 11, Infection in Neutropenic Children with Cancer. Pediatr. Infect. Dis. J. 22, 621. doi:10.3389/fphar.2020.00621 359–365. doi:10.1097/01.inf.0000061014.97037.a8 Medical Dictionary for Regulatory Activities (2020). Available at: https://www. Collins, G. E., Clay, M. M., and Falletta, J. M. (1974). A Prospective Study of the meddra.org/ (accessed September 22, 2020). Epidemiology of Adverse Drug Reactions in Pediatric Hematology and O’Connor, K., and Liddle, C. (2013). Prospective Data Collection of Off-Label Use Oncology Patients. Am. J. Hosp. Pharm. 31, 968–975. doi:10.1093/ajhp/31. of Rituximab in Australian Public Hospitals. Intern. Med. J. 43, 863–870. doi:10. 10.968 1111/imj.12206.Petropoulou Ducassou, S., Leverger, G., Fernandes, H., Chambost, H., Bertrand, Y., Armari- Petropoulou, A. D., Porcher, R., Peffault de Latour, R., Xhaard, A., Weisdorf, D., Alla, C., et al. (2017). Benefits of Rituximab as a Second-Line Treatment for Ribaud, P., et al. (2012). Increased Infection Rate after Preemptive Rituximab Autoimmune Haemolytic Anaemia in Children: a Prospective French Cohort Treatment for Epstein-Barr Virus Reactivation after Allogeneic Hematopoietic Study. Br. J. Haematol. 177, 751–758. doi:10.1111/bjh.14627 Stem-Cell Transplantation. Transplantation 94, 879–883. doi:10.1097/TP. European Medicines Agency (2004). Clinical Safety Data Management: 0b013e3182664042 Definitions and Standards for Expedited. Reporting. CPMP/ICH/3945/03. Posthumus, A. A. G., Alingh, C. C. W., Zwaan, C. C. M., van Grootheest, K. K., Available at: https://www.ema.europa.eu/en/documents/scientific- Hanff, L. L. M., Witjes, B. B. C. M., et al. (2012). Adverse Drug Reaction-Related guideline/international-conference-harmonisation-technical-requirements- Admissions in Paediatrics, a Prospective Single-Centre Study. BMJ Open 2, registration-pharmaceuticals-human-use_en-12.pdf (Accessed December 4, e000934. doi:10.1136/bmjopen-2012-000934 2020) Queuille, E., Bleyzac, N., Auray, J. P., Bertrand, Y., Souillet, G., Philippe, N., et al. European Medicines Agency (2021). MabThera: Summary of Product (2001). [A New Tool for Evaluation of Medication Errors Applied to Pediatric Characteristics. Available at: https://www.ema.europa.eu/en/documents/ Hematology]. Therapie 56, 775–783. product-information/mabthera-epar-product-information_en.pdf (Accessed Saiyed, M. M., Lalwani, T., and Rana, D. (2015). Is Off-Label Use a Risk Factor for April 13, 2021). Adverse Drug Reactions in Pediatric Patients? A Prospective Study in an Indian European Medicines Agency.(2013). Medicines under Additional Monitoring. Tertiary Care Hospital. Int. J. Risk Saf. Med. 27, 45–53. doi:10.3233/JRS-150642 Available at: https://www.ema.europa.eu/en/human-regulatory/post- SEHOP/PETHEMA (2013) Tratamiento de la Leucemia Aguda Linfoblástica de authorisation/pharmacovigilance/medicines-under-additional-monitoring Nuevo Diagnóstico (para niños mayores de 1 año y menores de 19 años). (Accessed December 4, 2020). Available at: https://www.sehh.es/images/stories/recursos/2014/documentos/ Gallagher, R. M., Bird, K. A., Mason, J. R., Peak, M., Williamson, P. R., Nunn, A. J., guias/LAL_SEHOP_PETHEMA_2013.pdf (Accessed November 17 2020). et al. (2011). Adverse Drug Reactions Causing Admission to a Paediatric Stausberg, J., and Hasford, J. (2011). Drug-related Admissions and Hospital- Hospital: a Pilot Study. J. Clin. Pharm. Ther. 36, 194–199. doi:10.1111/j. Acquired Adverse Drug Events in Germany: a Longitudinal Analysis from 1365-2710.2010.01194.x 2003 to 2007 of ICD-10-Coded Routine Data. BMC. Health Serv. Res. 11, 134. Gallagher, R. M., Mason, J. R., Bird, K. A., Kirkham, J. J., Peak, M., Williamson, P. doi:10.1186/1472-6963-11-134 R., et al. (2012). Adverse Drug Reactions Causing Admission to a Paediatric Thiesen, S., Conroy, E. J., Bellis, J. R., Bracken, L. E., Mannix, H. L., Bird, K. A., et al. Hospital. PLoS One 7, e50127. doi:10.1371/journal.pone.0050127 (2013). Incidence, Characteristics and Risk Factors of Adverse Drug Reactions Hébert, G., Netzer, F., Ferrua, M., Ducreux, M., Lemare, F., and Minvielle, E. in Hospitalized Children - a Prospective Observational Cohort Study of 6,601 (2015). Evaluating Iatrogenic Prescribing: Development of an Oncology- Admissions. BMC Med. 11, 237. doi:10.1186/1741-7015-11-237 Frontiers in Pharmacology | www.frontiersin.org 8 May 2021 | Volume 12 | Article 670945 Amaro-Hosey et al. Drug Toxicity in Pediatric Oncohematology Vaseghi, G., Abed, A., Jafari, E., Eslami, N., and Eshraghi, A. (2016). Assessment of Zhang, Y.-Y., Yang, Q.-S., Qing, X., Li, B.-R., Qian, J., Wang, Y., et al. (2020). Peg- Adverse Drug Reaction Due to Cancer Chemotherapy in a Teaching Oncology Asparaginase-Associated Pancreatitis in Chemotherapy-Treated Pediatric Hospital in Isfahan, Central of Iran. Rrct 11, 266–272. doi:10.2174/ Patients: A 5-Year Retrospective Study. Front. Oncol. 10, 538779. doi:10. 1574887110666150818112648 3389/fonc.2020.538779 Workalemahu, G., Abdela, O. A., and Yenit, M. K. (2020). Chemotherapy-Related Adverse Drug Reaction and Associated Factors Among Hospitalized Paediatric Conflict of Interest: The authors declare that the research was conducted in the Cancer Patients at Hospitals in North-West Ethiopia. Dhps Vol. 12, 195–205. absence of any commercial or financial relationships that could be construed as a doi:10.2147/DHPS.S254644 potential conflict of interest. Yun, H., Xie, F., Delzell, E., Chen, L., Levitan, E. B., Lewis, J. D., et al. (2015). Risk of Hospitalised Infection in Rheumatoid Arthritis Patients Receiving Copyright © 2021 Amaro-Hosey, Danés, Vendrell, Alonso, Renedo, Gros, Vidal, Biologics Following a Previous Infection while on Treatment with Anti- Cereza and Agustí. This is an open-access article distributed under the terms of the TNF Therapy. Ann. Rheum. Dis. 74, 1065–1071. doi:10.1136/annrheumdis- Creative Commons Attribution License (CC BY). The use, distribution or 2013-204011 reproduction in other forums is permitted, provided the original author(s) and Zecca, M., Nobili, B., Ramenghi, U., Perrotta, S., Amendola, G., Rosito, P., et al. the copyright owner(s) are credited and that the original publication in this journal is (2003). Rituximab for the Treatment of Refractory Autoimmune Hemolytic cited, in accordance with accepted academic practice. No use, distribution or Anemia in Children. Blood 101, 3857–3861. doi:10.1182/blood-2002-11-3547 reproduction is permitted which does not comply with these terms. Frontiers in Pharmacology | www.frontiersin.org 9 May 2021 | Volume 12 | Article 670945