TuberOus SClerosis registry to increAse disease awareness (TOSCA) Post-Authorisation Safety Study of Everolimus in Patients With Tuberous Sclerosis Complex

This non-interventional post-authorisation safety study (PASS) assessed the long-term safety of everolimus in patients with tuberous sclerosis complex (TSC) who participated in the TuberOus SClerosis registry to increase disease Awareness (TOSCA) clinical study and received everolimus for the licensed indications in the European Union. The rate of adverse events (AEs), AEs that led to dose adjustments or treatment discontinuation, AEs of potential clinical interest, treatment-related AEs (TRAEs), serious AEs (SAEs), and deaths were documented. One hundred seventy-nine patients were included in the first 5 years of observation; 118 of 179 patients had an AE of any grade, with the most common AEs being stomatitis (7.8%) and headache (7.3%). AEs caused dose adjustments in 56 patients (31.3%) and treatment discontinuation in nine patients (5%). AEs appeared to be more frequent and severe in children. On Tanner staging, all patients displayed signs of age-appropriate sexual maturation. Twenty-two of 106 female (20.8%) patients had menstrual cycle disorders. The most frequent TRAEs were stomatitis (6.7%) and aphthous mouth ulcer (5.6%). SAEs were reported in 54 patients (30.2%); the most frequent SAE was pneumonia (>3% patients; grade 2, 1.1%, and grade 3, 2.8%). Three deaths were reported, all in patients who had discontinued everolimus for more than 28 days, and none were thought to be related to everolimus according to the treating physicians. The PASS sub-study reflects the safety and tolerability of everolimus in the management of TSC in real-world routine clinical practice.

This non-interventional post-authorisation safety study (PASS) assessed the long-term safety of everolimus in patients with tuberous sclerosis complex (TSC) who participated in the TuberOus SClerosis registry to increase disease Awareness (TOSCA) clinical study and received everolimus for the licensed indications in the European Union. The rate of adverse events (AEs), AEs that led to dose adjustments or treatment discontinuation, AEs of potential clinical interest, treatment-related AEs (TRAEs), serious AEs (SAEs), and deaths were documented. One hundred seventy-nine patients were included in the first 5

INTRODUCTION
Tuberous sclerosis complex (TSC) is a rare, genetic, multisystem disorder. TSC can affect almost any organ system, including the skin, central nervous system, kidneys, eye, heart, and lung. About 90% of the TSC patients experience neurological and renal abnormalities, which represent a major cause of morbidity and mortality (1,2). The clinical presentation of TSC is heterogeneous, and the degree of severity is highly variable between individuals, even among the family members (1). The onset of clinical manifestations of TSC also typically varies with age, which further adds to the complexity to the disease (3,4). These factors represent a significant challenge for the diagnosis and management of TSC. Current management guidelines are focused on early identification and close monitoring of lesion burden in combination with timely medical treatment of manifestations and early interventions for TSC-associated neuropsychiatric disorders (TANDs) (4). TSC is caused by pathogenic variants in either TSC1 or TSC2 genes, resulting in hyper-activation of the mammalian target of rapamycin (mTOR) signalling pathway and the subsequent development of hamartomatous lesions in patients with TSC (4).
Based on double-blinded, placebo-controlled, randomised, clinical trials that confirmed its safety and efficacy, the mTORinhibitor everolimus (Votubia R ) was approved in Europe in 2011 for the treatment of subependymal giant cell astrocytoma (SEGA) and renal angiomyolipoma (5)(6)(7)(8)(9)(10)(11)(12)(13). Randomised clinical trial studies are required as "gold standard" for product licensing. However, they fail to reflect the "real-world" scenarios, particularly in terms of AE representation. The randomised clinical trials have shown that everolimus was generally welltolerated in patients with TSC with manageable AEs, which were generally reversible and non-cumulative (14)(15)(16). However, since TSC is a rare disease, with a prevalence of 6.8-12.4 per 100,000 people (17), the three key registration trials included relatively small numbers of patients with TSC [ranging from 78 in EXIST-1 to 247 in EXIST-3 (5,9,12)].
The TuberOus SClerosis registry to increase disease Awareness (TOSCA) study was conducted to address existing lacunas in the diagnosis and management of TSC. Based on the request from European Medicines Agency (EMA) to use the TOSCA registry to collect data on long-term safety and reproductive abnormalities in patients taking everolimus for licensed indications, SEGA in children age 2-20 years, and angiomyolipoma in adults aged >18 years, the TOSCA postauthorisation safety study (PASS) was developed. Here, we report findings from this TOSCA sub-study.

Study Design and Participants and Data Collection
The TOSCA clinical study methodology has been published previously (15). In brief, TOSCA was a large-scale noninterventional study in patients with TSC. The study was designed with a core section, six ancillary research projects (with more detailed focus on SEGA, renal angiomyolipoma and lymphangiomyomatosis, genetics, TAND, epilepsy, and patient's quality of life), and a PASS sub-study (EU PASS Register Number EUPAS3247).
The TOSCA-PASS sub-study was aimed at collecting prospective long-term safety data of treatment with everolimus prescribed for the indications licensed in Europe at time of enrolment data on AEs, therapeutic drug monitoring data, and the long-term reproductive abnormalities within routine clinical practice were collected. The PASS sub-study was conducted in 11 European Union countries participating in the TOSCA registry.

Patients
Patients who participated in the TOSCA registry and received everolimus treatment in the licensed indications (for SEGA or renal angiomyolipoma) in the European Union were eligible for inclusion in the TOSCA PASS, after providing additional written informed consent.
The data collection cut-off was 10 August 2017 for the TOSCA PASS sub-study. As per EMA indication (EMA/CHMP/59467/2014, 20 February 2014), data collection on sexual maturation and fertility is to be continued for all paediatric patients until they reach Tanner stage 5, or age 16 years for females and age 17 years for males, whichever occurs first.
For the TOSCA PASS sub-study, being a non-interventional and observational study, all treatment-related decisions (dose adjustments, treatment discontinuation) were at the discretion of the treating physicians. No treatment protocol, diagnostic/therapeutic procedure, or a formal visit schedule was mandated by the TOSCA PASS study protocol. However, the recommended data collection per study schedule was at 3-monthly intervals, which most likely mirrors the patterns of routine clinical care of patients treated with everolimus. Detailed management of each individual's AEs was not collected; however, general guidelines were followed by investigators (18). Data were collected for all patients who achieved Tanner stage 5 by the cut-off date. For patients who discontinued the study prematurely, the reason for discontinuation was determined. All patients were instructed regarding possible AEs and their possible treatment (19).
In this manuscript, we present an interim analysis of patient data up to 10 August 2017. The long-term safety data of these patients will be reported after termination of the TOSCA PASS study.

Outcome Measures and Data Analysis
Incidence of AEs, AEs that lead to dose adjustment or discontinuation, everolimus treatment-related AEs (TRAEs) as per the investigator assessment, AEs of special interest (AESIs), serious AEs (SAEs), and deaths were documented during the treatment (from day of enrolling into the PASS study to 28 days after the last dose of everolimus). AESIs were those AEs that were of specific clinical interest in connection with everolimus treatment. Potential AEs sought included noninfectious pneumonitis, severe infections, hypersensitivity (anaphylactic reactions), stomatitis, wound healing complications, increased serum creatinine/proteinuria/renal failure, hyperglycaemia, new onset of diabetes mellitus, dyslipidaemia, hypophosphatemia, cardiac failure, cytopenias, haemorrhages, thrombotic and embolic events, female fertility (including secondary amenorrhea), pre-existing infections (reactivation/aggravation/exacerbation), safety in patients with hepatic impairment, postnatal developmental toxicity, pregnant or breast-feeding women, male infertility, and muscle wasting/muscle loss. The relationship of the incidence of AESIs with everolimus blood levels was also noted. SAEs were defined as AEs that are fatal or life-threatening, result in persistent or significant disability/incapacity, constitute a congenital anomaly/birth defect, are medically significant, require medical or surgical intervention, or require inpatient hospitalisation or prolonged existing hospitalisation.
Data on everolimus (dose, interruption, dose change, and duration) and concomitant medication were captured. Concomitant medications entered into the database were coded using the World Health Organization (WHO) drug reference list, which employs the anatomical therapeutic chemical (ATC) classification system.
The analysis set consisted of all patients who had at least one post-baseline safety assessment and were exposed to at least one dose of everolimus after the enrolment. The AEs and SAEs were summarised by system organ class and preferred term using the Medical Dictionary for Regulatory Activities (MedDRA) version 20.

Patient Disposition and Baseline Characteristics
A total of 179 patients were enrolled in the study. Of 179 patients, 16 patients (8.9%) had discontinued participation in the study, and 31 patients continued to be followed up as part of the ongoing paediatric PASS. Of the 16 patients who discontinued the study, 12 were lost to follow-up, three died, and one patient was withdrawn as per investigator's decision (Figure 1). Everolimus was initiated for 73 (40.8%) patients with SEGA, 122 (68.2%) with renal angiomyolipoma, and 17 (9.5%) for both SEGA and renal angiomyolipoma.
The demographic and clinical characteristics of the enrolled patients are summarised in Table 1

Safety
Overall, 118 of 179 patients (65.9%) had AEs of any grade, irrespective of its relationship with study drug ( Table 2). The most common grade 3/4 AE that occurred in >3% of patients was pneumonia observed in 2.8% of patients ( Table 3). The rate of AEs was higher in children compared with adults [75.4% (n = 61) vs. 61.0% (n = 72); P = 0.0541], and a decreasing trend on AE rate was noted with increase in age ( Table 2).
The most frequent TRAEs were stomatitis (6.7%), aphthous mouth ulcer (5.6%), and hypercholesterolaemia (5%). Everolimus dose adjustments due to AEs were reported in 56 patients (31.3%). The most common AEs that led to dose adjustments in at least two patients were diarrhoea (five patients); stomatitis, pneumonia, common cold, and urinary tract infection (three patients each); and metrorrhagia, pyrexia, pyelonephritis, sinusitis, influenza, and otitis (two patients each). Haemorrhagic events leading to dose adjustments were haemorrhage on the left side of the brain, bleeding in angiomyolipoma, and renal haemorrhage (one patient each). AEs leading to everolimus discontinuation were reported in nine patients (5%) and included fatigue and amenorrhea (1.1% each); and anaemia, mouth ulceration, empyema, pneumonia, hyperglycaemia, type I diabetes mellitus, flank pain, intestinal adenocarcinoma, seizure, and alopecia (0.6% each).
Three deaths were reported in the study. These deaths occurred after 28 days of everolimus discontinuation and were reported by study investigators as not related to everolimus treatment. Patient 1, male, aged 30 years, from the Netherland, died due to medically assisted death as per the local regulations on day 487 after commencement of everolimus administration. He had permanently discontinued everolimus treatment on day 205. Patient 2, male aged 52 years, died from influenza (grade 4) on day 1399; everolimus treatment was permanently discontinued on day 1359. Patient 3, male aged 46 years, died due to intestinal adenocarcinoma on day 74; everolimus was permanently discontinued on day 43. Autopsy was not performed for patients 1 and 2, whereas, this was unknown for patient 3.

Everolimus Dosage and Exposure
Data on everolimus dosage and exposure were available for 150 patients. The mean duration of the everolimus exposure was 302.4 ± 105.04 days (median, 365 days; range, 7-669 days). The mean and median daily doses and the most commonly administered dosage (5 mg throughout the study) are shown in Table 4. The mean everolimus blood level at baseline was 6.27 ng/ml (median, 4.9 ng/ml; range, 1.4-35.9 ng/ml), with the maximum concentration reported at third follow-up visit (mean, 6.6 ng/ml) and the least at fourth follow-up visit (mean 4.937 ng/ml). Median duration of exposure in <18 years (n = 59) and ≥18 (n = 91) was 365 days (P = 0.1735) with a significant difference in mean daily everolimus dose of 6.4 mg (range, 1-13 mg) in <18 years vs. 7.7 mg (range, 1-20 mg) in ≥18 years (P = 0.0144), respectively. Changes in everolimus dosage were reported in 53 patients (35.3%), with dose increase reported in 44 patients (83%), dose interruptions in 20 patients (13.3%), and dose reductions in 14 patients (9.3%) ( Table 4). The most common reasons for dose changes were side effects (21 patients, 14%); other reasons are specified in Table 4.
Of the 179 patients enrolled, 22 of 106 (20.8%) female patients had menstrual cycle disorders. Amenorrhea was reported in nine patients (8.5%) and other abnormal reproductive conditions in three patients (1.7%). In the initial analysis, three patients (1.7%, one male and two females) were reported to have abnormal onset of puberty. However, on further analysis, it was noted that one female child had precocious puberty, which was treated successfully 5 years before starting everolimus treatment. The second patient, a male child, had developed behavioural problems during puberty, which were thought to be secondary to oxcarbazepine and predated everolimus treatment. The third patient, an adult female patient, had abnormal puberty before the start of everolimus too. Thus, abnormal puberty was found to be not related to everolimus treatment.
Abnormal hormone levels of thyroid-stimulating hormone were reported in four patients (2.2%); testosterone levels were abnormal in two patients (1.1%); luteinising hormone, follicularstimulating hormone, and oestradiol were abnormal in one patient (0.6%) ( Table 6).   consistently demonstrated its efficacy and tolerability (6)(7)(8) which subsequently led to the approval of everolimus in the treatment of these TSC manifestations (14). Studies have also shown that even with prolonged treatment, no new toxicities or complications were observed (6,10,11,13,20). All these data were obtained from interventional controlled clinical trials.  In line with the previously reported everolimus in TSC studies (6)(7)(8), the most commonly reported AE was stomatitis, which    is effectively managed to minimise the occurrence and severity in TSC patients. The incidence of stomatitis and infections was relatively low in this PASS sub-study (7.8 and 34%) than in the previous EXIST studies (5,6,9,12). Rates of stomatitis in EXIST-1 and EXIST-2 were 31 and 48%, respectively (6,8). Overall, 55-68% of patients had stomatitis in EXIST-3, which included stomatitis, mouth, aphthous, lip, and tongue ulcerations; mucosal inflammation; and gingival pain (5,9,12,21). The decrease in AEs like stomatitis with longer follow-up could be attributed to better tolerability or better care or both (10). Additionally, the median dose of everolimus was similar to that in the EXIST interventional studies, whereas, the starting doses in the current study were lower, as suggested by 44 patients having their dose increased. In addition, outside of a strict trial protocol, physicians may have been able to interrupt and reduce the dose of everolimus to manage AEs. The rate of infections was reported in about 72% patients in EXIST-1 and 65% of patients in EXIST-2 (8,11). Similarly, the rate of infections was higher in the EXIST-3 study with nasopharyngitis in 14-23.8%, upper respiratory tract infection in 13-22.4%, and pharyngitis 1-10.2% of the patients (12,21). However, the incidence of diarrhoea (6.7%) was slightly higher than that reported in the EXIST-3 study (5%) but lower than that of the EXIST-1 and EXIST-2 studies (13% each) (6)(7)(8). Notably, PASS sub-study also showed a higher incidence of AEs in children than in the adults ( Table 2), as previously reported in the EXIST-1 study (16,22). It was unknown whether this was due to higher blood levels or increased susceptibility to everolimus.
No new safety signals were observed in the study. Most of the AEs observed in the study were manageable with dose adjustment and/or use of concomitant medication and were of modest severity, with grade 1 or 2 AEs observed in almost 42% of patients. The lower rate of stomatitis or aphthous mouth ulceration and some other AEs in this study compared with that expected from the literature could be due to several reasons including lower starting doses, early interruption or reduction of dose, better education and preparation of patients, or lower median blood concentration of everolimus in TOSCA PASS compared with those of the previous interventional trials (9,12,20). In general, there was likely to be a correlation between drug levels and AEs within individual patients as evidenced by the successful treatment of AEs by lowering the patients' dose ( Table 4), but it appears that the different individuals have different sensitivities to any particular blood level of everolimus causing AEs. Three deaths were reported during the study. All occurred in adult patients and were deemed by the study investigators as not related to everolimus treatment.
The data on menstrual irregularities concur with the previous findings with respect to clinical features but at lower frequency (9) and confirm that everolimus can cause amenorrhoea and other menstrual irregularities. In most patients, sexual maturation was not affected by everolimus.
In conclusion, the findings from this study are confirmed the manageable safety profile of everolimus in patients with TSC with no new safety signal. The long-term safety data will continue to be collected as per study protocol for the paediatric patients.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the study protocol and all amendments were reviewed and approved (if applicable) by independent ethics committee/institutional review board for each centre:

AUTHOR CONTRIBUTIONS
AJ, EB, MB, PC, MD, JF, MF, CH, SJ, JK, JL, AM, RN, VS, MS, RT, and BZ: designing the study, patient accrual, clinical care, data interpretation, and drafting, revising, final review, and approval of the manuscript. PdV, CF, Gd'A, TC, VC, FO'C, JQ, YT, and SY: designing the study, data interpretation, and drafting, revising, final review, and approval of the manuscript. LD'A: designing the study, trial management, data collection, data analysis, data interpretation, and drafting, revising, final review, and approval of the manuscript. RM: designing the study, data analysis, data interpretation, and drafting, revising, final review, and approval of the manuscript. SS: designing the study, trial statistician, data analysis, data interpretation, and drafting, revising, final review, and approval of the manuscript. KB: patient accrual, clinical care, and drafting, revising, final review, and approval of the manuscript. All authors contributed to the article and approved the submitted version.