Safety profile of chimeric antigen receptor T-cell immunotherapies (CAR-T) in clinical practice

Purpose Two chimeric antigen receptor T-cell (CAR-T) therapies have been approved in the United States (USA) in 2017 and Europe (EU) in 2018: axicabtagene ciloleucel and tisagenlecleucel. They contain the patient’s own T cells, which are extracted, genetically modified, and reinfused. Alongside the good efficacy results, the assessment of safety profile of these new therapies represents a great challenge. Our aim was to analyze the reports of the adverse drug reactions (ADR) after CAR-T administration as occurred in the real clinical setting. Methods We performed a retrospective observational study, collecting all the reports in EU (EudraVigilance, EV) and US (FAERS) databases of ADRs regarding axicabtagene ciloleucel and tisagenlecleucel. Both descriptive and statistical analyses were performed, the latter by using Reporting Odds Ratio (ROR). Results A total number of 1426 reports of suspected ADRs were retrieved in EudraVigilance and FAERS. Patients’ reported age reflected the age range for which the drugs are approved (18–64 years for axicabtagene ciloleucel and patients aged under 25 years for tisagenlecleucel). The most reported event was cytokine release syndrome (CRS), 185 events for tisagenlecleucel and 462 for axicabtagene ciloleucel in FAERS and 137 and 498, respectively, in EudraVigilance. A disproportionality was found comparing axicabtagene ciloleucel with tisagenlecleucel for the above-mentioned event: EV ROR 2.47, 95% CI 2.22–2.74, FAERS 1.89, 1.70–2.10. Conclusion CRS represents the major problem with the administration of CAR-T therapies. Our analysis has not revealed new ADRs; however, it supports the safety profile of CAR-T with new data from real clinical setting.


Introduction
In the last years, personalized medicine in immuno-oncology reached great goals, and, among these, the most important one was the advent of CAR-T therapies. The acronym CAR-T stands for chimeric antigen receptor T-cell therapy, which consists of patient's genetically modified white blood cells yielded capable to perform antitumor activity using a lentiviral vector encoding an anti-CD19 chimeric antigen receptor-CAR (axicabtagene ciloleucel) or by retroviral transduction to express a chimeric antigen receptor (tisagenlecleucel). After lymphodepletion, the modified T cells reinfused in the patients can attach a specific antigen on the tumor cells. Two CAR-T therapies have been approved in Europe (August 2018), i.e., axicabtagene ciloleucel (Yescarta®) [1,2] and tisagenlecleucel (Kymriah®) [3]. These therapies have been also approved in 2017 by the US Food and Drug Administration (US FDA) [4,5].
Axicabtagene ciloleucel is approved for two types of blood cancer: diffuse large B-cell lymphoma (DLBCL) and primary mediastinal large B-cell lymphoma (PMBCL) [6]. Tisagenlecleucel is approved for the treatment of B-cell acute lymphoblastic leukemia (ALL) refractory or in relapse (posttransplant, second or later relapse) in children and young adults up to 25 years and for adult patients with relapsed or refractory DLBCL after two or more lines of systemic therapy [7]. Two main studies were performed with tisagenlecleucel in B-cell ALL (92 children and young adults) [8] and in DLBCL (165 patients) [9]. Twelve months after treatment, the survival probability was 70% and 40.2%, respectively [10]. A main study in 111 patients with DLBCL and PMBCL was carried out for axicabtagene ciloleucel [11], showing a complete response in 47% of the patients and a partial response in the 19% [12]. Apart from the great efficacy results, the knowledge of the safety profile of these new therapies represents a great challenge. Both drugs must be administered in a qualified center for hematological malignancies with healthcare professionals trained in the management of possible adverse events. Patients undergo pre-medication, clinical assessment prior to infusion, and monitoring after infusion to minimize potential adverse events. Our aim was to evaluate all the adverse events reported following the administration of axicabtagene ciloleucel and tisagenlecleucel in the real clinical setting in Europe and in the USA.

Methods
Data were retrieved from the European database of suspected adverse drug reaction reports (EudraVigilance) using the online interface adrreports.eu [13] and from the US Food and Drug Administration Adverse Event Reporting system (FAERS) using a freely available and of public consultation dashboard. EudraVigilance is the system for the management and collection of suspected adverse events reported for medicines. It is maintained by the European Medicines Agency (EMA) on behalf of the European Union (EU). In November 2017, the new EudraVigilance system with enhanced functionalities was launched with an improvement of the collection of the individual case safety reports (ICSRs) from regulatory authorities and marketing authorization holders (MAHs) and of the detection of possible safety signals [14]. The FDA Adverse Event Reporting System (FAERS) is the database for the collection of all the reports of suspected ADRs submitted to FDA. All the reports are in a structured form that includes administrative information, patient's characteristics, suspected and concomitant drugs, and the adverse drug reactions. For our aim, we focused on the analysis of both anti-CD19 chimeric antigen receptor T-cell immunotherapies (CAR-T). We retrieved all the reports with axicabtagene ciloleucel or tisagenlecleucel reported as suspected drugs. Considering that both drugs received a marketing authorization valid throughout the European Union on 23 August 2018, we considered the period between September 2018 and 22 October 2019 for our analysis in EudraVigilance. For FAERS analysis, we considered the period between September 2017 (first CAR-T approval date) and 30 June 2019 (date of the most recent update of the dashboard).

Descriptive analysis
The extracted reports were identified by a unique EU Local Number in EudraVigilance or by a unique Case ID in FAERS. The information reported was the report type (spontaneous or from clinical studies), receipt date, primary source qualification (healthcare professional or non-healthcare professional), patient age group, patient sex, preferred term MedDRA, seriousness criteria, and suspect and concomitant drugs. The MedDRA is a standardized medical terminology that allows to report adverse events in the same way all over the world. It is used by both databases, and this allows an easy comparison of the data reported [15]. This dictionary has a hierarchic terminology: several preferred terms (PTs) are grouped in one System Organ Class (SOC) by etiology, manifestation site, or purpose.
We analyzed all the reports related to axicabtagene ciloleucel or tisagenlecleucel. A descriptive analysis was performed to evaluate all the reported adverse events. It should be noted that each report could contain one or more adverse events. For each drug, it was checked if the reported events were listed in the corresponding Summary of the Product Characteristics (SPCs) to ascertain the notoriety of the adverse reactions.

Statistical analysis
A case-non-case analysis was performed by using the Reporting Odds Ratio (ROR) with 95% confidence interval as statistical parameter. ROR allows a quantitative approach by the use of contingency tables. The aim is to compare the frequency of a drug-reaction pair with all the others in the database or with different therapeutic regimens. If ROR is >1, an increased frequency for the drug-reaction pair can be assumed. Considering the specificity of the treatments analyzed, the use of the whole database was considered not appropriate. Therefore, a disproportionality analysis was carried out by comparing the two treatments. Other articles in literature have used a similar approach when comparing different therapeutic regimens [16]. The proportions of the adverse drug reactions to axicabtagene ciloleucel were compared to those with tisagenlecleucel. The denominator was the total number of adverse drug reactions reported for both drugs. Table 1 shows the most reported adverse events for both CAR-T therapies in EudraVigilance and FAERS. For each event, it is also indicated whether the ADR is reported in the Summary of Product Characteristic of the corresponding CAR-T therapy (pointed out with *); it is not directly reported

EudraVigilance
We retrieved 683 reports of suspected adverse reactions referred to axicabtagene ciloleucel. One hundred ninety-five reports were related to female patients (28.6%) and 306 (44.8%) to males. For 182 reports (26.6%), sex of the patients was not reported. Table 2 shows the reports classified by age and sex in the two databases. Most of the patients (277; 40.6%) belonged to the 18-64 years age class. Only one report concerned a female patient of the age class 12-17 years and 2 patients, one male and one female, aged >85 years. The great majority of the ADRs (673, 98.5%) were reported by healthcare professionals. Overall, 2650 adverse events were reported for axicabtagene ciloleucel. The top five most reported were cytokine release syndrome (498 events, 18.8%), neurotoxicity (247, 9.3%), pyrexia (187, 7.1%), encephalopathy (94, 3.6%), and hypotension (75, 2.8%). The most reported concomitant therapies were acyclovir, levetiracetam, and ondansetron and, to a less extent, tocilizumab.

EudraVigilance
In the same period, we retrieved 284 reports of suspected adverse reactions referred to tisagenlecleucel of which 137

Discussion
To the best of our knowledge, this is the first study aimed at evaluating and comparing the safety profile of the CAR-T therapies with the use of data of real clinical practice. Postapproval pharmacovigilance is considered crucial for the evaluation of CAR-T safety profile [17], as it allows long-term follow-up in a large and uncontrolled population. From the present analysis, no new and unexpected ADRs emerged in daily clinical practice, and this is reassuring. The patients experiencing an ADR with either drug were different in term of age. Most of axicabtagene ciloleucel patients were aged 18-64 years or more, whereas those receiving tisagenlecleucel were under 25 years of age and mainly minors. This reflects the indications for which the drugs have been approved: tisagenlecleucel is the only CAR-T therapy specifically approved for the treatment of patients ≤25 years of age with R/ R B-cell acute lymphoblastic leukemia. Proportion of male patients was higher with both drugs and databases. This does not point out a greater predisposition of male patients to develop an ADR, but it is rather related to the prevalence of the disease to be treated. Both diffuse large Bcell lymphoma (DLBCL) and acute lymphocytic leukemia (ALL) have a slightly higher incidence in males than females. The US National Cancer Institute reported that the number of new cases per 100,000 persons in the period 2012-2016 were 6.7 in males and 4.6 in female for DLBCL and 1.9 and 1.5, respectively, in female for ALL [18,19].
A lower presence of consumers as reporters has been identified in EudraVigilance in comparison with FAERS, this confirming the lower tendency of patients to report ADRs in Europe [20]. Cytokine release syndrome (CRS) was the most reported ADR in both databases and for both drugs.
The pathophysiology of the syndrome is unclear, and recent studies showed that IL-6, IL-1, and nitric oxide produced by macrophages are involved in its course [21]. The vast majority of the others events reported in our study could be symptoms of this syndrome, which is characterized mainly by pyrexia and hypotension, the most reported events retrieved after CRS. Other ADRs associable with CRS are hypoxia, chills, cardiac adverse events (e.g., tachycardia, atrial fibrillation, and cardiac arrest), acute kidney injury, and also hepatic and musculoskeletal toxicities [22]. All these events have been reported for both drugs and in both databases. Different studies in literature account these events as the most common with CAR-T therapies [23][24][25]. In most cases, the trigger is the cytokine release syndrome that starts a subsequent cascade of events. The main goal is to prevent or at least limit the CRS to a low grade [26,27]. Some strategies in the management of CAR-T-related toxicities have been described such as pharmacological immunosuppression with tocilizumab or corticosteroids, suicide or elimination genes, and targeted activation [28,29]. Recently, Lee at al. have described new methods that could prevent or rather terminate within 3 h the CRS-like toxicity by using low molecular weight adapters [30]. N number of reports, ROR reporting odds ratio, CI_low lower bound of the 95% confidence interval, CI_up upper bound of the 95% confidence interval accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Funding Open access funding provided by Alma Mater Studiorum -Università di Bologna within the CRUI-CARE Agreement.
Data availability For this type of study, formal consent is not required.

Declarations
Conflict of interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.