This was a prospective observational study on the use of biotech drugs, carried out from April 2012 to December 2016, among Onco-hematology (OM), Hepato-gastroenterology (HG), Rheumatology (RT), Dermatology (DM), and Neurology (NE) Units in Campania Region (Italy) on a total of 775 patients. Data come from a large pharmacovigilance network that involved 9 clinical centers (hospital and/or Institute for Treatment and Research) located in different districts of this Region, being representative of the whole Campania Region. The study was approved by the ethic committee of the Coordinating Center of Università degli studi della Campania “L. Vanvitelli”. Patients, identified by the clinicians working within participating centers, were informed about the methods and aims of the project and agreed to participate. A written informed consent was obtained from patients. The study enrolled all patients who received for the first time (naïve patients) a biotech drugs. Pharmacological therapies were chosen only on the basis of clinical judgment and the follow-up visits were planned in accordance with the clinical routine. Follow-up visits consisted of objective examination, blood test, pharmacological treatment revaluation (in term of dose-adjustment, discontinuation, and switch to another drug), and assessment of any AE occurred during the drug therapy.

After patients' enrolment, which coincided with the first biotech administration, the number of follow-up visits per patient varied based on type of disease and related to pharmacological treatment regimen but also on patient's decision. In order to evaluate the occurrence of AEs associated to each biotech drug during the overall study period, we split follow-up period in 6 categories, considering also the injection time (the first biotech administration). The follow-up categories were: “at injection,” 1–30, 31–60, 61–90, 91–180, 181–360, >361 days.

Standardized monitoring form was supplied to the clinicians to collect demographic and clinical data at the time of the enrolment and at follow-up visits. The enrolment form included the following information: age, sex, clinical diagnosis, type of biotech drug use and exposure time, information on comorbidities; the follow-up form included: type of biotech drug use, in term of dose-adjustment, discontinuation and switch to another drug, AE occurrence (date, seriousness, and suspected drug).

When clinicians identified an AE likely associated with biotech drugs, a dedicated section of the standardized monitoring form was filled in. In this section symptoms and signs or diagnosis, date of occurrence, seriousness, and suspected drug were reported. As described in the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use, clinical safety data management: definitions and standards for expedited reporting E2A guideline (ICH-E2A, available on line at https://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E2A/Step4/E2A_Guideline.pdf), a serious AEs corresponds to any untoward medical occurrence that results in death, is life-threatening, requires inpatient hospitalization or prolongation of existing hospitalization, results in persistent or significant disability or incapacity, or results in a congenital anomaly/birth defect or clinically relevant conditions based on clinical judgments.

The reported AEs were coded using preferred terms from Medical Dictionary for Regulatory Activities (MedDRA) and grouped using the System Organ Class (SOCs) classifications of MedDRA. Once ADR was recorded in the dedicated section, according to the current European legislation on pharmacovigilance, clinicians filled in the suspected Adverse Drug Reaction (ADR) reporting form of the Italian Medicine Agency [Agenzia Italiana del Farmaco (AIFA)] and send it to qualified person responsible for pharmacovigilance (QPPV) of their respective health structures. Then, the QPPV recorded the ADR report into a nationwide spontaneous reporting database: the Italian Pharmacovigilance Network (Rete Nazionale di Farmacovigilanza) managed by AIFA. Of note, the decision to report suspected ADRs was exclusively taken by the managing clinicians.

A descriptive analysis of all AEs reported by the participating centers during the study period was performed. For continuous variables, descriptive statistics (mean, standard deviation, and frequencies) with percentages were calculated. Comparisons using Chi square test were performed (significance level was p < 0.05) for categorical variables. Data were analyzed using Microsoft Access and Excel programs.

Since our study was not designed to do a comparison between biotech drugs neither to find a specific association between biotech drugs and AEs, rather to describe all AEs occurring in routine clinical practice, we did not perform any sample size calculation.

We used Naranjo algorithm (Naranjo et al., 1981) in order to establish the strength of relationship between the biotech drug and suspected AEs. All scores ranged between possible and certain reports were considered reasonable for causality. Given the clinical impact of serious AEs and considering that all not serious ones were already expected, we decided to show Naranjo algorithm results exclusively for serious AEs.

Details on patients' clinical and demographic characteristics are reported in Table 1. Out of 840-screened patients, 775 were enrolled (mean age of 56; standard deviation – SD ± 15.2), of whom 53.81% were female. More than 40% of patients had at least one comorbidity, mainly represented by cardiovascular diseases (predominantly hypertension and hypertensive heart disease), diabetes mellitus, and hypercholesterolemia. The mean follow-up/patient was 3.48 (95% Confidence Interval 3.13–3.84).

The majority of the study population was enrolled in OM and RT Units (432 and 219 patients, respectively). In terms of gender distribution, a higher proportion of female patients was enrolled in RT Unit (148 females vs. 70 males), while the opposite situation was seen in HG Unit (29 females vs. 49 males; Table 1).

The most common reported diagnosis were hematological malignancies (N.155 pt; 20%), followed by rheumatoid arthritis (N. 130 pt; 16.77%), colorectal cancer (N. 125 pt; 16.13%), breast cancer (N. 88 pt; 11.35%) and psoriatic arthritis (N. 51 pt; 6.58%). Based on gender distribution, hematological malignancies and colorectal cancers were more common among male patients, while rheumatoid arthritis and breast cancer (as expected) were more frequent in females (Table 1).

The most commonly prescribed biotech drugs at the time of enrolment were rituximab (19.87%), bevacizumab (12.00%), infliximab (10.71%), trastuzumab (including trastuzumab emtansine; 8.90%), adalimumab (8.52%) and cetuximab (8.13%; data not shown). No biosimilar drugs were used in our population. Referring to the drug distribution by gender, trastuzumab and tocilizumab were more commonly used among female patients (15.11 vs. 1.68% and 8.39 vs. 0.28%, respectively), while rituximab and cetuximab were more frequently used among males (25.14 vs. 15.35% and 13.41 vs. 3.60%, respectively; Figure 1).

Some biotech drugs, such as, adalimumab, infliximab and rituximab, were utilized in more than one clinical Unit, thus they have a multiple therapeutic indication.

Out of 775 patients, 320 (41.29%) experienced at least one AE (mean of 4.2 AEs per patient, data not shown) with no gender differences (Table 2). Most of patients experienced at least one AE associated to cetuximab (68.25%), followed by rituximab (52.60%) and trastuzumab (47.83%). Adalimumab and etanercept were more frequently associated to AEs in female patients than in male ones (35.90% vs. 7.41%, p < 0.001; 27.59% vs. 10.00%, p = 0.023; Table 2).

Fifty patients discontinued the treatment with the biotech drug. Most of these cases were reported for abatacept, bevacizumab, etanercept, and infliximab. Progression disease, drug therapeutic failure and other AEs occurrence represented the main reasons reported of discontinuation (Supplementary Table 2).

As we reported, among 1311 AEs (occurred in 320 patients), 140 were serious. In our opinion the occurrence of such AEs is not surprising for two main reasons. First of all, all serious AEs occurred during the therapy with mAbs. It is well known that biotech drugs, particularly mAbs, can be frequently linked to serious AEs, including infusion reactions, infections, and autoimmune disorders (Tovey and Lallemand, 2011). Indeed, compared to other biotech drugs, mAbs have longer terminal half-lives and, consequently, a single dose of such drugs could lead to prolonged systemic exposure, with an increased risk of serious AEs. Moreover, chimeric mAbs, such as, brentuximab, rituximab, and cetuximab, are more frequently related to serious AEs. For such reason, their use is restricted to the treatment of clinical conditions with high morbidity and mortality (Tranter et al., 2013). Secondly, since traditional RCTs are designed with the aim to avoid risks to enrolled patients, the identification of serious AEs during RCTs is not a simple task. However, when new drugs become available for the use in clinical practice, they are commonly administered to patients not fully represented in RCTs and such patients are the ones more likely to experience serious AEs (Garcia-Doval et al., 2012). Therefore, since in “real life” patients, disease-related risks, comorbidities and concomitant immunosuppressant and chemotherapeutic agents can additionally contribute to the occurrence of AEs (Meadows and Hurwitz, 2012; Kotaka et al., 2016) and considering that almost 40% of our patients had at least one comorbidity, in our opinion, the occurrence of 140 serious AEs should not be considered an alarming figure, rather the consequence of the use of biotech drugs in a widely varied population with characteristics quite different to those of patients enrolled in the traditional RCTs. Nevertheless, scientific evidence is still controversial about risks related to biotech drugs, especially with regard to infections and malignancies risks. Thus, it is important to continue to closely monitor the use of these in clinical practice to improve the knowledge on their long -term safety.

Literature data suggest that treatment discontinuation with anti-TNF drugs occur approximately in 21–35% of patients (Combe et al., 2006). According to our findings, results of an internet-based survey by Bolge et al. revealed that lack of effectiveness was the primary reason for discontinuation in rheumatoid arthritis patients receiving etanercept, adalimumab, certolizumab, or golimumab, followed by other AEs occurrence (Bolge et al., 2015). Further clinical data confirm these findings (Weinblatt et al., 2011; Levin et al., 2014; Fafá et al., 2015; Nüßlein et al., 2015).

When biotech drugs are used in oncology setting, their combination with standard chemotherapy increase the risk of AE occurrence leading to more dose reductions or discontinuations (Oza et al., 2017). With this regard, a recent prospective observational cohort study, which enrolled 40 metastatic colorectal cancer patients treated with XELOX and bevacizumab, revealed that among all discontinuations 15 were related to disease progression and 7 to AEs occurrence (Burmester et al., 2013). Also the results of a recent multinational prospective single-arm study revealed that, among 1,021 ovarian cancer patients receiving bevacizumab and paclitaxel, discontinuation occurred in 33% of patients due to disease progression and in 17% of patients due to AEs (da Silva et al., 2014).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.