Heart failure and atrial tachyarrhythmia on abiraterone: A pharmacovigilance study Marie Bretagne, Bénédicte Lebrun-Vignes, Antoine Pariente, Christian M Shaffer, Gabriel G Malouf, Pauline Dureau, Camille Potey, Christian Funck-Brentano, Dan M Roden, Javid J Moslehi, et al.

To cite this version:

Marie Bretagne, Bénédicte Lebrun-Vignes, Antoine Pariente, Christian M Shaffer, Gabriel G Mal- ouf, et al.. Heart failure and atrial tachyarrhythmia on abiraterone: A pharmacovigilance study. Archives of cardiovascular diseases, Elsevier/French Society of Cardiology, 2020, 113 (1), pp.9-21. ￿10.1016/j.acvd.2019.09.006￿. ￿hal-02874941￿

HAL Id: hal-02874941 https://hal.sorbonne-universite.fr/hal-02874941 Submitted on 19 Jun 2020

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Heart failure and atrial tachyarrhythmia on abiraterone: a pharmacovigilance study

Marie Bretagne, MD1; Bénédicte Lebrun-Vignes, MD1; Antoine Pariente, MD, PhD2; Christian M. Shaffer3; Gabriel G. Malouf, MD-PhD 4, Pauline Dureau, MD1; Camille Potey, MD5; Christian Funck- Brentano MD, PhD1; Dan M. Roden, MD3; Javid J. Moslehi, MD6, Joe-Elie Salem MD, PhD1,6

1 Sorbonne Université, INSERM CIC Paris-Est, AP-HP, ICAN, Regional Pharmacovigilance Centre, Pitié- Salpêtrière Hospital, Department of , F-75013 Paris, France.

2 Univ. Bordeaux, CHU de Bordeaux, INSERM, CIC-1401, Bordeaux Population Health Research Center, Team Pharmaco-epidemiology, Department of Pharmacology, UMR 1219, Bordeaux, France

3 Departments of Medicine, Pharmacology and Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN, USA.

4 Department of Hematology and Medical Oncology, CHRU Strasbourg, Hopital Civil, Strasbourg

5 Regional Pharmacovigilance Centre, Department of Pharmacology, Lille, France

6 Department of Medicine and Pharmacology, Cardio-oncology program, Vanderbilt University Medical Center, Nashville, Tennessee, USA.

* Correspondence to [email protected], [email protected]; Centre d'Investigation Clinique Paris-Est, Hôpital La Pitié-Salpêtrière, 47-83 Bld de l'hôpital, 75651 Paris Cedex 13, Secretariat: +33 1 42 17 85 31, Fax: +33 1 42 17 85 32 or Department of pharmacology, Cardio-oncology program, Vanderbilt University Medical Center, Nashville, Tennessee, USA, contact: +1 (615) 322-0067

NCT: 03243604

World count: 3000

Funding source: This study was supported by FRM “SPE20170336816”.

Disclosures: JJM has served on advisory boards at Bristol Myers, Pfizer, Novartis and Regeneron and has received research funding from Pfizer and Bristol Myers Squibb. The other authors have no conflict of interest to disclose. The supplied data from VigiBase come from a variety of sources. The likelihood of a causal relationship is not the same in all reports. The information does not represent the opinion of the World Health

Organization (WHO).

1

Key points.

Question: Abiraterone, a cytochrome-17 inhibitor decreasing production of adrenal , has become a critical component of deprivation treatment (ADT) for metastatic .

Early data from clinical trials suggested abiraterone was associated with hypermineralocorticism signs, including hypertension, hypokalemia and oedema. More recently, case reports of other cardiovascular toxicities have emerged. We searched the international pharmacovigilance database VigiBase for other cardiovascular toxicities known to be favored by increased aldosterone and decreased production and associated with abiraterone in a “real world” population.

Findings: We report the first large scale analysis linking atrial tachyarrhythmias (AT) and heart failure

(HF), reported here as associated with abiraterone using pharmacovigilance signal detection analysis.

Mean time to onset from start of treatment with abiraterone to AT and HF was within ~4-5 months.

Fatality rates ranged ~10%.

Meaning: Severe cardiac events related to cardiac AT and HF associated with other hyper- mineralocorticism signs (hypertension, hykalemia, oedema) occur commonly in patients exposed to abiraterone. These events should be considered in patient care and in clinical trial designs.

2

Abstract.

Importance: Abiraterone and are recently approved androgen deprivation therapies

(ADT) for metastatic prostate cancer. The cardiac safety profile for the two drugs is unknown.

Abiraterone, with its propensity to hyper-mineralocorticism on top of androgen deprivation, might carry an additional risk for atrial tachyarrhythmias (AT) and heart failure (HF) as compared to other

ADT.

Objective: To determine if abiraterone was associated with increased proportion of AT and HF reports among overall suspected adverse drug reactions (ADR) reported in several pharmacovigilance databases as compared to enzalutamide, other ADT and all other drugs.

Design: An observational, retrospective, pharmacovigilance study

Setting: The French pharmacovigilance database (01/24/1985-05/16/2017); the European pharmacovigilance database (12/01/2001-08/18/2017) and VigiBase, the World Health Organization’s pharmacovigilance database (10/01/1967-08/28/2017).

Participants: All individual case safety reports reported in men in these databases.

Main Outcomes and Measures: A disproportionality analysis using reporting odds-ratios (ROR).

Exposures: Exposure to abiraterone versus enzalutamide, or versus other ADT or versus entire database

Results: In the 5,759,781 ADR reports in men in VigiBase, 55,070 pertained to ADT. The ROR of atrial tachyarrhythmias (AT) for abiraterone were of 4.1[3.1-5.3] vs. enzalutamide, 3.7[3-4.5] vs. other

ADT, and 3.2[2.7-3.7] vs. entire database (p<0.0001 for all). The corresponding ROR for heart failure

(HF) were of 2.5[2-3], 1.5[1.3-1.7], and 2[1.7-2.3], respectively (p<0.0001 for all). These results were concordant when using the French and European pharmacovigilance databases. Mean time to AT and

HF onset was shorter on abiraterone (5.2±0.8 and 4.5±0.6 months, respectively) vs. other ADT

3

(13.3±3.2 and 9.2±1.1 months; p<0.05). Cases on abiraterone vs. other ADT were more frequently associated with ≥2 ADR-terms including AT, HF, hypokalemia, hypertension and edema (13.6% vs.

6%, p<0.0001). For abiraterone, age but not dose was associated with reporting of AT/HF vs. any other ADR.

Conclusions and Relevance: Compared to other ADT, abiraterone was associated with a higher reporting of AT and HF associated with hypokalemia, hypertension and edema. These findings are consistent with the hyper-mineralocorticism induced by abiraterone and not with other ADT.

Keywords: Abiraterone, enzalutamide, androgen deprivation therapy, heart failure, atrial fibrillation

4

Introduction

In metastatic prostate cancer, castration by bilateral orchiectomy or by androgen deprivation therapies

(ADT) is the standard of care.[1] First-line ADT mainly acts at the level of pituitary axis, using -releasing hormone (GnRH) agonists or antagonists (Fig-1). However, prostate cancer often becomes castration-resistant after several years.[2] It has recently been demonstrated that, by limiting resurgence of androgen activity at the prostate level, recently-developed compounds abiraterone-acetate and enzalutamide constitute breakthrough ADT that can be added in combination to first-line castration (generally GnRH agonists) to prolong overall survival.[3-6] Abiraterone-acetate is an inhibitor of androgen and corticosteroid biosynthesis at the adrenal level while enzalutamide is a potent 2ndgeneration antagonist (Fig-1). Low-dose corticosteroids are administered with abiraterone to inhibit any induced excessive adrenocorticotropic (ACTH) production (Fig-1).

This co-prescription is mandatory to limit signs of secondary hyper-mineralocorticism observed in phase-1 trials where abiraterone was given alone and hypertension, hypokalemia and fluid retention were observed.[2, 6] ADT has been associated with heart failure (HF), atrial tachyarrhythmia (AT) and

QT prolongation.[7-15] Abiraterone, with its propensity to hyper-mineralocorticism on top of androgen deprivation, might carry an additional risk for these cardiac conditions as compared to other

ADT.[11, 16-19] Here, we used the French pharmacovigilance database (BNPV, Base Nationale de

Pharmacovigilance) through 05/2017 to evaluate if abiraterone was associated with increased reporting for atrial tachyarrhythmias (AT) and heart failure (HF) compared to enzalutamide, and other

ADT. Enzalutamide was considered as the best control for abiraterone because both drugs had to be combined with another castration modality (generally GnRH agonists) and shared the same indication and target population at the time of analysis (metastatic castration-resistant prostate cancer, both approved in Europe and USA around the same time period: 2011-2012). Therefore, comparability of groups of prostate cancer patients on abiraterone vs. enzalutamide was expected to be maximal with no difference in background clinical characteristics and co-treatments. If signals were identified for these cardiac adverse drug-reactions (ADR) in the French pharmacovigilance database, it was planned

5 to expand our analysis seeking for signal confirmation using the European pharmacovigilance database and VigiBase, the World Health Organization’s (WHO) global database of individual safety case reports (ISCRs). We recently reported that these ADT or substantially increase risk of

QT prolongation and Torsade-de-pointes using such pharmacovigilance approach and results are detailed elsewhere.[20, 21]

Methods

We performed a disproportionality (case/non case) analysis that considered reports of suspected ADR in men contained in the French BNPV (01/24/1985-05/16/2017); in the European pharmacovigilance database (12/01/2001-08/18/2017); and in VigiBase (10/01/1967-08/28/2017),[22] to evaluate the reporting odds-ratio (ROR) of AT and HF for abiraterone compared to enzalutamide, other ADT and full database. An examination of the association of ADT-induced AT and HF with signs suggestive of hyper-mineralcorticism (e.g. hypokalemia, hypertension and edema) was also performed. Appendix-1 provides more details concerning the methodology used and these databases.[21-25]

A case/non-case analysis was performed first for abiraterone versus (vs.) enzalutamide because these drugs shared the same cancer indication and target population at the time of data extraction from the different pharmacovigilance databases.[23-25] Case/non-case analyses were also performed for abiraterone vs. the group of all other ADT and for abiraterone vs. the full database. This method compares the proportion of specific ADR reported for the case versus non-case groups. Reactions are based on the medical dictionary for regulatory activities (MedDRA) classification dictionary of terms for side effects. Disproportionality was estimated by calculating the ROR Chi-square (Supp.Table-

1).[21, 25] If the proportion of an ADR is greater in patients exposed to a drug (abiraterone) than in patients not exposed to this drug (enzalutamide, all ADT or full database selected as non-case, respectively), this suggests an association between the specific drug and the reaction and is a potential signal for safety. We searched for ADR related to suspected drug-induced AT, HF, hypokalemia,

6 hypertension and edema. Selected MedDRA Terms used for each of these ADR are detailed in

Supp.Table-2. Drugs considered as ADT are shown in Fig-1.

Characteristics of cases were described in terms of means (± standard deviation) or medians (with interquartile range) for quantitative variables, and in terms of effective and proportion for qualitative ones. Comparisons used, as appropriate, Student’s t-test or Mann-Whitney test for quantitative variables, Chi-2 or Fisher’s exact test for qualitative ones. Correlations (r) between quantitative variables were assessed using Spearman’s coefficient (Prism 7; GraphPad Software, Inc). Statistical significance was accepted for p-value<0.05.

7

Results

Disproportionality analyses

The numbers of total ISCRs reported on abiraterone, enzalutamide, other ADT and across the full database and the subset of those with AT, HF, hypertension, hypokalemia, and edema are detailed in

Table-1, for each pharmacovigilance database studied.

Analysis of the French pharmacovigilance database through 05/2017 revealed 127 ISCRs of suspected

ADR for abiraterone and 1552 for other ADT (including 71 for enzalutamide; Table-1). These ISCRs affected adult men treated for prostate cancer. Comorbidity and clinical characteristics of patients with

ISCRs related to abiraterone or enzalutamide were similar (Supp. Table 3). Abiraterone was associated with higher reporting rates of AT and HF vs. enzalutamide (ROR: 5.7 [1.3-25.3] and ROR: 4.3[1.4-

12.9]; respectively; p≤0.01 for both, Table-1) and vs. other ADT subgroup (Table 1). Results for hypertension, hypokalemia and edema are detailed in Table 1.

Analysis of the European-database of suspected ADR reports through 08/2017 revealed 4865

ISCRs for abiraterone and 25094 for other ADT (including 12953 for enzalutamide). Abiraterone was associated with higher reporting rates of AT, HF, hypertension, hypokalemia and edema vs. enzalutamide and vs. other ADT (Table-1).

Analysis of VigiBase reports through to 08/2017 revealed 9203 ISCRs for abiraterone, 45867 for other ADT (including 21189 for enzalutamide) and 5759781 for the full database from >120 countries.

Abiraterone was associated with higher reporting rates of AT, HF, hypertension, hypokalemia and edema vs. enzalutamide (ROR: 4.1[3.1-5.3], ROR: 2.5[2-3], ROR: 1.5[1.3-1.8], ROR: 14.5[10.8-19.8] and ROR: 1.9[1.6-2.2]; respectively; ; p<0.0001 for all, Table 1); vs. other ADT subgroup (ROR:

3.7[3-4.5], ROR: 1.5[1.3-1.7], ROR: 1.8[1.5-2], ROR: 16.4 [13-20.7] and ROR: 1.7[1.5-1.9]; p<0.0001 for all, Table 1); and vs. full database (ROR: 3.2[2.7-3.7], ROR: 2[1.7-2.3], ROR: 2.3[2-

2.6], ROR: 11.7[10.4-13.1] and ROR: 2.1[1.9-2.4]; p<0.0001 for all, Table 1). There was no over- reporting for AT, nor HF when analyzing enzalutamide vs. other ADT and when analyzing

8 enzalutamide vs. full database (Supp. Table 4). ISCRs with ADR suspected to be due to abiraterone overlapping with at least one other ADT suspected was extremely low (≤1.1%, Fig-2A).

Clinical characteristics of cases of AT and HF reported on abiraterone

Among the 127 ICSRs reported on abiraterone in the French pharmacovigilance database, 38 cases

(30%) were associated with combined AT and HF (n=6, 5%), HF (n=20, 16%) or AT (n=12, 10%); occurring most frequently within the first 3 months of abiraterone introduction (56%, Supp. Table 5).

These latter ADR were symptomatic and serious leading to hospitalization (n=31/38, 82%) or death

(n=4/38, 10%). Details concerning their clinical presentation, outcome and management are shown in

Supp. Table 5. Neurotoxicity and hepatotoxicity were the main other ADR reported on abiraterone

(n=15, 11.8% and n=14, 11%, respectively). Only one case of acute coronary syndrome was reported on abiraterone. Patients with HF and/or AT on abiraterone had more prevalent cardiovascular comorbidities, mainly hypertension and HF, than patients with other types of ADR on abiraterone

(60.5% vs. 36%, p=0.01 and 13.2% vs. 2.2%, p=0.01, respectively; Table 2). Thus, most patients with

HF on abiraterone had de novo HF (85%, n=22/26). Both HF with reduced ejection fraction and preserved ejection fraction were observed (respectively; n=8/12, 66% and n=4/12, 34% of patients with reported echocardiogram characteristics, Supp. Table 5). When available, brain natriuretic peptide was consistently increased in HF cases (n=7/7). All cases of AT were de novo, and often associated with documented hypokalemia (<3.5mmol/L) in 39% (n=7/18). Troponin increase was more frequently reported as being increased in patients with AT or HF on abiraterone vs. other types of ADR (10.5% vs. 1.1%, p=0.03, respectively), but the only case of acute coronary syndrome confirmed by angiography was found in the non-AT/HF subgroup of ADR on abiraterone. AT and/or

HF (AT/HF) on abiraterone were frequently associated with hypokalemia, hypertension and edema compared to other types of ADR on abiraterone (26.3% vs. 14.6%, p=0.11; 15.8% vs. 5.6%, p=0.06;

36.8% vs. 5.6%, p<0.0001; Table 2, respectively). To manage these AT/HF, abiraterone was

9 withdrawn in 85% of the patients in addition to AT/HF standard of care (Supp. Table 5). In the two cases where abiraterone was reintroduced, ADR recurred within few weeks.

Details concerning the number of ISCRs with AT, HF, edema, hypertension and hypokalemia on abiraterone (n=1079) vs. other ADT (n=2577) in VigiBase and how these conditions overlap, are shown in Fig-2A. ADR on abiraterone were more frequently associated with ≥2 terms including hypokalemia, hypertension, edema, AT and/or HF vs. other ADT (13.6% vs. 6%, p<0.0001). Mean time from treatment initiation to HF and AT was shorter on abiraterone (4.5±0.6 and 5.2±0.8 months, respectively) than with other ADT (9.2±1.1 and 13.3±3.2 months, respectively; both p<0.05). Most

AT and HF associated with abiraterone appeared within 3 months of introduction (58% and 61%, respectively, Fig-2B). Reports of AT, HF and hypokalemia with abiraterone were associated with age

(r=0.03-0.06, p<0.05) but not with dose of co-prescribed corticosteroids, or with dose of abiraterone

(Supp. Fig-1, Table 3). Reports of edema and hypertension were inversely associated with dose of co- prescribed corticosteroids (r=-0.07, and r=-0.12, respectively; both p<0.05).

10

Discussion

Using an analysis of three pharmacovigilance databases, we show that abiraterone was associated with a higher reporting rate for AT and HF as compared with enzalutamide and other ADT. AT and

HF on abiraterone were frequently associated with increases in cardiac biomarkers (BNP and troponin) and concurrent signs of hyper-mineralocorticism, i.e. hypokalemia, hypertension and edema. The onset of the reported AT and HF could be as short as one week; the event mostly occurred within 3 months after abiraterone start. AT and HF on abiraterone generally appeared de novo, but a history of hypertension and heart failure were identified as potential risk factors. The highest dose of corticosteroid (used in combination with abiraterone) was unrelated to reporting of AT or HF, and associated with decreased reporting for other signs of hyper-mineralcorticism as compared to other

ADR on abiraterone. This finding limits the potential confounding participation of corticosteroids as factors in the appearance of these side effects. Both HF with preserved and reduced ejection fraction were found. These AT and HF were serious ADR requiring hospitalization and they were mainly managed by abiraterone withdrawal and standard of care interventions to achieve resolution of AT and

HF.

Cardiovascular side effects induced by abiraterone in trials and real-life practice

ADT are major treatments to delay tumor development and improve survival in patients with prostate cancer.[1] However, several randomized controlled trials and observational studies have suggested that

ADT increase the risk of cardiovascular events, particularly AT and HF.[7-9, 26] It has been recently demonstrated that normalization of testosterone levels after supplementation in a cohort of 76 639 hypogonadic men was associated with a significant decrease in the incidence of AT,[8] suggesting that the underlying mechanism is on-target (testosterone suppression). In HF patients, testosterone supplementation was shown to improve peak-oxygen consumption, 6-minute and shuttle walk-test, independently from ejection fraction variation.[27-29] In addition to androgen deprivation, abiraterone is also responsible for hyper-mineralcorticism, a condition also known to increase incidence of HF and

11

AT.[8, 17-19] Therefore, abiraterone was likely to be even more toxic than other ADT in induction of

HF and AT. The comparison of abiraterone with enzalutamide in this study is particularly relevant since both drugs shared the same indication and population of patients (metastatic castration-resistant prostate cancer) at the time of ADR reporting covered in this work (through 05/2017 to 08/2017, for the different pharmacovigilance databases).[4] Notably, the indications for abiraterone have recently expanded because of data on overall survival and radiographic progression-free survival in men with newly diagnosed, metastatic, castration-sensitive prostate cancer.[30]

Our data are consistent with a meta-analysis of four randomized phase III trials with a mean follow- up of 25.2 months, showing that use of drugs inhibiting CYP-17 (abiraterone and orteronel) in combination with prednisone significantly increased the risk of cardiac disorders as compared to placebo plus prednisone (20% vs. 14%; Relative Risk: 1.47[1.27–1.7]) in patients with metastatic castration-resistant prostate cancer.[31] In patients with metastatic hormone-sensitive prostate cancer, a meta-analysis of two trials with a median follow-up (MFU) of at least 30 months also showed that

ADT plus abiraterone was associated with more cardiac disorders as compared to ADT alone (4% vs.

1.2%, Odds-Ratio: 2.93 [1.74-4.93]).[32] In these trials, the contribution of HF to these cardiac disorders was not described.[3, 5, 30, 33-35] AT incidence was higher in patients with more severe forms of prostate cancer; i.e. metastatic castration-resistant cases with previous chemotherapy as compared to patients with prostate cancer not previously treated with hormone therapy (incidence ranging from 1 to 5%; MFU: 12 to 40 months).[3, 5, 30, 35] Patients with clinically significant heart diseases or uncontrolled hypertension were excluded from these trials. Pre-existing cardiovascular conditions are known risk factors of HF and arrhythmias in patients on ADT.[19, 36, 37] Our study also suggests that patients with pre-existing hypertension and HT are particularly at risk of developing

AT/HF on abiraterone. Almost all ADR extracted from VigiBase were recorded during the post- marketing phase in patients with metastatic castration-resistant prostate cancer.

According to data extracted from the French pharmacovigilance database, troponin and BNP were increased in patients with HF and AT on abiraterone. In a post-marketing study of 17 unselected

12 patients with metastatic castration-resistant prostate cancer, the incidence of serious cardiovascular

ADR was higher than expected (n:5/17, 29.4%; time to onset: 2 to 8 months).[38] Patients all had pre- existing cardiac conditions. NT-proBNP and troponin levels at baseline and follow-up predicted a higher risk of cardiac ADR. Researches in larger cohorts are required to evaluate prognostic value of these biomarkers and to further estimate absolute and relative increased risk of cardiovascular ADR on abiraterone vs. other ADT or age-matched population control. It has also been shown that abiraterone trough concentration was associated to prostate cancer response.[39] Therefore, abiraterone therapeutic drug monitoring might have some value to preserve its efficacy avoiding some concentration-dependent ADR.

Cardiovascular toxicities and abiraterone mechanism of action

The anti-androgenic activity of abiraterone is mediated by decreased extra-gonadal androgen synthesis plus direct androgen receptor antagonism by its metabolite Δ4-abiraterone.[40] At the adrenal level, abiraterone induces 17alpha-hydroxylase/C(17,20)-lyase inhibition leading to decreased adrenal androgens and glucocorticoids synthesis.[2] There is no such effect on glucocorticoid metabolism with other ADT (Fig-1). Inefficient cortisol synthesis signals the hypothalamic-pituitary axis to increase ACTH. Consequently, the adrenal glands produce excess of mineralocorticoids, hormone precursors that do not require 17alpha-hydroxylase/C(17,20)-lyase processes, namely 11- deoxycroticosterone (DOC) and corticosterone (Fig-1).[2] These hormones lead to hypokalemia, hypertension and fluid overload. Supplementation of patients on abiraterone with corticosteroids is therefore required to prevent excessive ACTH increase.[2] However, ACTH suppression by this supplementation is variable, leading to inter-individual variability in residual secretion of mineralocorticoid hormone precursors and consequent clinical manifestations.[41] In our study, the highest dose of corticosteroids used in combination with abiraterone were associated with less edema and hypertension. This confirms the importance of corticosteroid supplementation to avoid any excess in the production of hormones with mineralocorticoid activity which are the main cause of

13 cardiovascular toxicities such as hypertension, AT and HF.[18, 42-47] Addition of aldosterone antagonist to abiraterone plus corticosteroids, salt deprivation and choosing a corticosteroid with no intrinsic mineralocorticoid activity, have been proposed to prevent abiraterone related cardiovascular toxicities and these strategies deserve further investigation.[41, 48, 49] Eplerenone is currently used in preference to as previous studies in vitro demonstrated that eplerenone did not bind and activate wild-type androgen-receptor (as opposed to spironolactone), but data concerning the clinical relevance of this finding are lacking.[6, 50] To date, prednisone, which does carry mineralocorticoid activity, is recommended in combination with abiraterone. It has been argued that dexamethasone with high glucocorticoid (repressing ACTH) and virtually no mineralocorticoid activity might be a better alternative. This is currently under investigation (NCT01867710). In almost all ADRs in our study

(>97%), steroids with mineralocorticoid activity were used (prednisone, prednisolone, methylprednisolone, hydrocortisone and cortisone).[49] Therefore, it was not possible to study whether the presence or absence of intrinsic mineralocorticoid activity of corticosteroids combined with abiraterone influenced the rates of AT/HF reports.

Limitations of the study

This study identified higher rates of reporting for AT, HF and signs suggestive of hyper- mineralocorticism on abiraterone as compared to other ADT in French pharmacovigilance database.

The level of causality attributed to abiraterone was plausible based on extrinsic data, chronology, symptomatology and known mechanisms of action of abiraterone.[51] Therefore, we expanded our analyses by seeking confirmation of these signals and better characterization of patients experiencing these ADR in other continental pharmacovigilance databases. At the time of our first analysis in

French pharmacovigilance database (05/2017), we found only one case of suspected acute coronary syndrome induced by abiraterone and one by enzalutamide. Moreover, Budoff et al. had recently showed that among older men with symptomatic hypogonadism, treatment with testosterone compared with placebo was associated with a significantly greater increase in coronary artery non calcified

14 plaque volume, as measured by coronary computed tomographic angiography.[52] Taken together, level of causality attributed to ADT and abiraterone to lead to cardiac ischemia was considered insufficient, precluding further analysis.

We used different databases with variable completeness of reporting of comorbid conditions and concomitant drugs, the likelihood that the suspected adverse reaction is drug-related is not the same in all cases, and there is limited possibility of verification of the rhythm abnormalities by means of electrocardiographic tracings. Another important limitation of this study is the expected under- reporting for AT and HF on ADT, since there is no particular mention to follow carefully for these side effects in ADT indication labels. The denominator and clinical characteristics of patients exposed to ADT cannot be evaluated, but in the present situation it was almost exclusively men with prostate cancer. The total number of ISCRs for each drug or group of drugs is used as denominator for this kind of disproportionality analysis in pharmacovigilance databases.[23-25] Clinical trials are mandatory but may not allow definite conclusions on drug safety, because they are built to evaluate efficacy in a specific population with strictly predefined inclusion criteria. Spontaneous notifications remain the cornerstone for ADR evaluation despite their limitations. Disproportionality analysis in pharmacovigilance databases remains a major method to detect signals in drug safety research and post-marketing surveillance.[53]

15

References.

[1] Parker C, Gillessen S, Heidenreich A, Horwich A, Committee EG. Cancer of the prostate: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2015;26 Suppl 5:v69-77. [2] Attard G, Reid AH, Auchus RJ, Hughes BA, Cassidy AM, Thompson E, et al. Clinical and biochemical consequences of CYP17A1 inhibition with abiraterone given with and without exogenous glucocorticoids in castrate men with advanced prostate cancer. J Clin Endocrinol Metab. 2012;97:507-16. [3] Ryan CJ, Smith MR, de Bono JS, Molina A, Logothetis CJ, de Souza P, et al. Abiraterone in Metastatic Prostate Cancer without Previous Chemotherapy. New England Journal of Medicine. 2013;368:138-48. [4] Scher HI, Fizazi K, Saad F, Taplin ME, Sternberg CN, Miller K, et al. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med. 2012;367:1187-97. [5] de Bono JS, Logothetis CJ, Molina A, Fizazi K, North S, Chu L, et al. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med. 2011;364:1995-2005. [6] Attard G, Reid AH, Yap TA, Raynaud F, Dowsett M, Settatree S, et al. Phase I clinical trial of a selective inhibitor of CYP17, , confirms that castration-resistant prostate cancer commonly remains hormone driven. J Clin Oncol. 2008;26:4563-71. [7] Martin-Merino E, Johansson S, Morris T, Garcia Rodriguez LA. Androgen deprivation therapy and the risk of coronary heart disease and heart failure in patients with prostate cancer: a nested case-control study in UK primary care. Drug Saf. 2011;34:1061-77. [8] Sharma R, Oni OA, Gupta K, Sharma M, Sharma R, Singh V, et al. Normalization of Testosterone Levels After Testosterone Replacement Therapy Is Associated With Decreased Incidence of Atrial Fibrillation. J Am Heart Assoc. 2017;6. [9] Bhatia N, Santos M, Jones LW, Beckman JA, Penson DF, Morgans AK, et al. Cardiovascular Effects of Androgen Deprivation Therapy for the Treatment of Prostate Cancer: ABCDE Steps to Reduce Cardiovascular Disease in Patients With Prostate Cancer. Circulation. 2016;133:537-41. [10] Abehsira G, Bachelot A, Badilini F, Koehl L, Lebot M, Favet C, et al. Complex Influence of and Sex Steroid Hormones on QT Interval Duration. J Clin Endocrinol Metab. 2016;101:2776-84. [11] Salem JE, Alexandre J, Bachelot A, Funck-Brentano C. Influence of steroid hormones on ventricular repolarization. Pharmacol Ther. 2016;167:38-47. [12] Barber M, Nguyen LS, Wassermann J, Spano JP, Funck-Brentano C, Salem JE. Cardiac arrhythmia considerations of hormone cancer therapies. Cardiovasc Res. 2019;115:878-94. [13] Salem JE, Yang T, Moslehi JJ, Waintraub X, Gandjbakhch E, Bachelot A, et al. Androgenic Effects on Ventricular Repolarization: A Translational Study From the International Pharmacovigilance Database to iPSC- Cardiomyocytes. Circulation. 2019. [14] Salem JE, Dureau P, Bachelot A, Germain M, Voiriot P, Lebourgeois B, et al. Association of Oral Contraceptives With Drug-Induced QT Interval Prolongation in Healthy Nonmenopausal Women. JAMA Cardiol. 2018;3:877-82. [15] Salem JE, Bretagne M, Lebrun-Vignes B, Waintraub X, Gandjbakhch E, Hidden-Lucet F, et al. Clinical characterization of men with long QT syndrome and torsades de pointes associated with hypogonadism: A review and pharmacovigilance study. Arch Cardiovasc Dis. 2019. [16] Kelly WK, Gomella LG. Androgen deprivation therapy and competing risks. JAMA. 2011;306:2382-3. [17] Zannad F, McMurray JJ, Krum H, van Veldhuisen DJ, Swedberg K, Shi H, et al. Eplerenone in patients with systolic heart failure and mild symptoms. N Engl J Med. 2011;364:11-21. [18] Milliez P, Girerd X, Plouin PF, Blacher J, Safar ME, Mourad JJ. Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism. J Am Coll Cardiol. 2005;45:1243-8. [19] O'Farrell S, Garmo H, Holmberg L, Adolfsson J, Stattin P, Van Hemelrijck M. Risk and timing of cardiovascular disease after androgen-deprivation therapy in men with prostate cancer. J Clin Oncol. 2015;33:1243-51. [20] Salem JE, Waintraub X, Courtillot C, Shaffer CM, Gandjbakhch E, Maupain C, et al. Hypogonadism as a Reversible Cause of Torsades de Pointes in Men. Circulation. 2018;138:110-3. [21] Grouthier V, Lebrun-Vignes B, Glazer AM, Touraine P, Funck-Brentano C, Pariente A, et al. Increased long QT and torsade de pointes reporting on tamoxifen compared with aromatase inhibitors. Heart. 2018;104:1859-63.

16

[22] Lindquist M. VigiBase, the WHO Global ICSR Database System: Basic Facts. Drug Information Journal. 2008;42:409-19. [23] van Puijenbroek E, Diemont W, van Grootheest K. Application of quantitative signal detection in the Dutch spontaneous reporting system for adverse drug reactions. Drug Saf. 2003;26:293-301. [24] De Bruin ML, Pettersson M, Meyboom RH, Hoes AW, Leufkens HG. Anti-HERG activity and the risk of drug- induced arrhythmias and sudden death. Eur Heart J. 2005;26:590-7. [25] Rothman KJ, Lanes S, Sacks ST. The reporting odds ratio and its advantages over the proportional reporting ratio. Pharmacoepidemiol Drug Saf. 2004;13:519-23. [26] Alexandre J, Moslehi J, Bersell KR, Funck-Brentano C, Roden DM, Salem JE. Anticancer drug-induced cardiac rhythm disorders: Current knowledge and basic underlying mechanisms. Pharmacol Ther. 2018. [27] Chung T, Kelleher S, Liu PY, Conway AJ, Kritharides L, Handelsman DJ. Effects of testosterone and on cardiac function: a randomized, placebo-controlled study. Clin Endocrinol (Oxf). 2007;66:235- 45. [28] Toma M, McAlister FA, Coglianese EE, Vidi V, Vasaiwala S, Bakal JA, et al. Testosterone supplementation in heart failure: a meta-analysis. Circ Heart Fail. 2012;5:315-21. [29] Salem JE, Nguyen LS, Hammoudi N, Preud'homme G, Hulot JS, Leban M, et al. Complex Association of Sex Hormones on Left Ventricular Systolic Function: Insight into Sexual Dimorphism. J Am Soc Echocardiogr. 2017. [30] Fizazi K, Tran N, Fein L, Matsubara N, Rodriguez-Antolin A, Alekseev BY, et al. Abiraterone plus Prednisone in Metastatic, Castration-Sensitive Prostate Cancer. N Engl J Med. 2017;377:352-60. [31] Roviello G, Sigala S, Danesi R, Re MD, Bonetta A, Cappelletti MR, et al. Incidence and relative risk of adverse events of special interest in patients with castration resistant prostate cancer treated with CYP-17 inhibitors: A meta-analysis of published trials. Crit Rev Oncol Hematol. 2016;101:12-20. [32] Rydzewska LHM, Burdett S, Vale CL, Clarke NW, Fizazi K, Kheoh T, et al. Adding abiraterone to androgen deprivation therapy in men with metastatic hormone-sensitive prostate cancer: A systematic review and meta-analysis. Eur J Cancer. 2017;84:88-101. [33] Fizazi K, Scher HI, Molina A, Logothetis CJ, Chi KN, Jones RJ, et al. Abiraterone acetate for treatment of metastatic castration-resistant prostate cancer: final overall survival analysis of the COU-AA-301 randomised, double-blind, placebo-controlled phase 3 study. The Lancet Oncology. 2012;13:983-92. [34] Ryan CJ, Smith MR, Fizazi K, Saad F, Mulders PFA, Sternberg CN, et al. Abiraterone acetate plus prednisone versus placebo plus prednisone in chemotherapy-naive men with metastatic castration-resistant prostate cancer (COU-AA-302): final overall survival analysis of a randomised, double-blind, placebo-controlled phase 3 study. The Lancet Oncology. 2015;16:152-60. [35] James ND, de Bono JS, Spears MR, Clarke NW, Mason MD, Dearnaley DP, et al. Abiraterone for Prostate Cancer Not Previously Treated with Hormone Therapy. N Engl J Med. 2017;377:338-51. [36] Haque R, UlcickasYood M, Xu X, Cassidy-Bushrow AE, Tsai HT, Keating NL, et al. Cardiovascular disease risk and androgen deprivation therapy in patients with localised prostate cancer: a prospective cohort study. Br J Cancer. 2017;117:1233-40. [37] Ziehr DR, Chen MH, Zhang D, Braccioforte MH, Moran BJ, Mahal BA, et al. Association of androgen- deprivation therapy with excess cardiac-specific mortality in men with prostate cancer. BJU Int. 2015;116:358- 65. [38] Campora S, Campazzi E, Zanardi S, Puntoni M, Piccininno M, Piccardo A, et al. Association of Biomarkers with Serious Cardiac Adverse Events during Abiraterone Acetate Treatment in Castration Resistant Prostate Cancer. Transl Oncol. 2016;9:600-5. [39] Carton E, Noe G, Huillard O, Golmard L, Giroux J, Cessot A, et al. Relation between plasma trough concentration of abiraterone and prostate-specific antigen response in metastatic castration-resistant prostate cancer patients. Eur J Cancer. 2017;72:54-61. [40] Li Z, Bishop AC, Alyamani M, Garcia JA, Dreicer R, Bunch D, et al. Conversion of abiraterone to D4A drives anti-tumour activity in prostate cancer. Nature. 2015;523:347-51. [41] Pia A, Vignani F, Attard G, Tucci M, Bironzo P, Scagliotti G, et al. Strategies for managing ACTH dependent mineralocorticoid excess induced by abiraterone. Cancer Treat Rev. 2013;39:966-73. [42] Mohammed SF, Ohtani T, Korinek J, Lam CS, Larsen K, Simari RD, et al. Mineralocorticoid accelerates transition to heart failure with preserved ejection fraction via "nongenomic effects". Circulation. 2010;122:370-8.

17

[43] Edelmann F, Tomaschitz A, Wachter R, Gelbrich G, Knoke M, Dungen HD, et al. Serum aldosterone and its relationship to left ventricular structure and geometry in patients with preserved left ventricular ejection fraction. Eur Heart J. 2012;33:203-12. [44] Reil JC, Hohl M, Selejan S, Lipp P, Drautz F, Kazakow A, et al. Aldosterone promotes atrial fibrillation. Eur Heart J. 2012;33:2098-108. [45] Boccanelli A, Mureddu GF, Cacciatore G, Clemenza F, Di Lenarda A, Gavazzi A, et al. Anti-remodelling effect of in patients with mild chronic heart failure (AREA IN-CHF study): final results. Eur J Heart Fail. 2009;11:68-76. [46] Mayyas F, Alzoubi KH, Van Wagoner DR. Impact of aldosterone antagonists on the substrate for atrial fibrillation: aldosterone promotes oxidative stress and atrial structural/electrical remodeling. Int J Cardiol. 2013;168:5135-42. [47] Schwarzl M, Alogna A, Zweiker D, Verderber J, Huber S, Manninger M, et al. A porcine model of early atrial fibrillation using a custom-built, radio transmission-controlled pacemaker. J Electrocardiol. 2016;49:124- 31. [48] Lammers C, Dartsch T, Brandt MC, Rottlander D, Halbach M, Peinkofer G, et al. Spironolactone prevents aldosterone induced increased duration of atrial fibrillation in rat. Cell Physiol Biochem. 2012;29:833-40. [49] Grossmann C, Scholz T, Rochel M, Bumke-Vogt C, Oelkers W, Pfeiffer AF, et al. Transactivation via the human glucocorticoid and mineralocorticoid receptor by therapeutically used steroids in CV-1 cells: a comparison of their glucocorticoid and mineralocorticoid properties. Eur J Endocrinol. 2004;151:397-406. [50] Luthy IA, Begin DJ, Labrie F. Androgenic activity of synthetic progestins and spironolactone in androgen- sensitive mouse mammary carcinoma (Shionogi) cells in culture. J Steroid Biochem. 1988;31:845-52. [51] Miremont-Salame G, Theophile H, Haramburu F, Begaud B. Causality assessment in pharmacovigilance: The French method and its successive updates. Therapie. 2016;71:179-86. [52] Budoff MJ, Ellenberg SS, Lewis CE, Mohler ER, 3rd, Wenger NK, Bhasin S, et al. Testosterone Treatment and Coronary Artery Plaque Volume in Older Men With Low Testosterone. JAMA. 2017;317:708-16. [53] Montastruc JL, Sommet A, Bagheri H, Lapeyre-Mestre M. Benefits and strengths of the disproportionality analysis for identification of adverse drug reactions in a pharmacovigilance database. Br J Clin Pharmacol. 2011;72:905-8.

18

Table 1. Disproportionality analysis comparing abiraterone vs. enzalutamide, abiraterone vs. any androgen deprivation therapy (ADT), abiraterone vs. full database on specific adverse drug reactions (ADR) in men using French, European and International (VigiBase) pharmacovigilance databases.

Type of ADR Number (%) of ADR ROR* [CI95] Abiraterone Enzalutamide Any ADT Full database Abiraterone vs. Abiraterone vs. Abiraterone vs. enzalutamide any ADT Full database French pharmacovigilance database (through 05/2017) Total 127 71 1552 NA Atrial tachyarrhythmias 18 (14%) 2 (3%) 7 (<1%) NA 5.7 [1.3-25.3] 36.5 [14.9-89.1] NA Heart failure 26 (20%) 4 (6%) 18 (1%) NA 4.3 [1.4-12.9] 21.9 [11.6-41.4] NA Hypertension 11 (9%) 4 (6%) 13 (<1%) NA 1.6 [0.5-5.2] 11.2 [4.9-25.6] NA Hypokalemia 23 (18%) 0 (0%) 8 (<1%) NA 32.1 [1.9-538.4] 42.7 [19.5-91.7] NA Edema 19 (15%) 5 (7%) 34 (2.2%) NA 2.3 [0.9- 6.5] 7.9 [4.3-13.9] NA European pharmacovigilance database (through 08/2017) Total 4865 12953 25094 NA Atrial tachyarrhythmias 115 (2.4%) 79 (0.6%) 158 (0.6%) NA 3.9 [3-5.3] 3.8 [3-4.9] NA Heart failure 222 (4.6%) 223 (1.7%) 600 (2.4%) NA 2.7 [2.3-3.3] 2 [1.7-2.3] NA Hypertension 169 (3.5%) 324 (2.5%) 554 (2.2%) NA 1.4 [1.2-1.7] 1.6 [1.4-1.9] NA Hypokalemia 250 (5.1%) 46 (0.4%) 93 (0.4%) NA 15.2 [11.1-20.9] 14.6 [11.5-18.5] NA Edema 197 (4%) 228 (1.8%) 489 (1.9%) NA 2.4 [1.9-2.9] 2.1 [1.8-2.5] NA VigiBase, the international pharmacovigilance database (through 08/2017) Total 9203 21189 45867 5759781 Atrial tachyarrhythmias 147 (1.6%) 84 (0.4%) 202 (0.4%) 29575 (0.5%) 4.1 [3.1-5.3] 3.7 [3-4.5] 3.2 [2.7-3.7] Heart failure 216 (2.3%) 205 (1%) 727 (1.6%) 69305 (1.2%) 2.5 [2-3] 1.5 [1.3-1.7] 2 [1.7-2.3] Hypertension 280 (3%) 426 (2%) 802 (1.7%) 77820 (1.4%) 1.5 [1.3-1.8] 1.8 [1.5-2] 2.3 [2-2.6] Hypokalemia 300 (3.3%) 49 (0.2%) 94 (0.2%) 16817 (0.3%) 14.5 [10.8-19.8] 16.4 [13-20.7] 11.7 [10.4-13.1] Edema 303 (3.3%) 377 (1.8%) 920 (2%) 90897 (1.6%) 1.9 [1.6-2.2] 1.7 [1.5-1.9] 2.1 [1.9-2.4] ROR: Reporting Odds-Ratio and its 95% confidence interval (CI95). Statistical significant results (p<0.05) are in bold. NA: not available Any ADT subgroup includes 11 ADT (Gonadotrophin-releasing hormone (GnRH) receptor agonists: , goserelin, buserelin, triptorelin; GnRH antagonists: degarelix; androgen receptor inhibitors: , , , enzalutamide; and steroidal androgen receptor inhibitors: , acetate) but excludes abiraterone. * The French pharmacovigilance database is smaller than the two other databases; therefore ROR estimates tend to be less precise with larger

CI95. 19

Table 2. Characteristics of men with prostate cancer and adverse drug reactions (ADR) induced by abiraterone related to atrial tachyarrhythmia (AT) and/or heart failure (HF) as compared to other type of ADR from the French Pharmacovigilance Database (through 05/2017).

Patients with AT Patients with other Characteristic and/or HF types of ADR p-value (n: 38) (n: 89) Age (mean ± SD) 76.3±8.6 73.7±15.8 0.08 Proportion of patient with at least one of the below cardiovascular past medical history, n (%) 27 (71.1) 44 (49.4) 0.02 Hypertension 23 (60.5) 32 (36.0) 0.01 Dyslipidemia 6 (15.8) 13 (14.6) 0.86 Diabetes 7 (18.4) 8 (9.0) 0.13 Myocardial infarction 4 (10.5) 10 (11.2) 0.91 Stroke 1 (2.6) 4 (4.5) 0.62 Peripheral artery disease 4 (10.5) 10 (11.2) 0.91 AT 6 (15.8) 7 (7.9) 0.18 HF 5 (13.2) 2 (2.2) 0.01 Proportion of patient with at least one of the below cardiovascular treatment, n (%) 19 (50) 49 (55.1) 0.60 Beta-blockers 6 (15.8) 22 (24.7) 0.27 Anti-arrhythmic class I (%), III (%), IV (%) 0 (0), 1 (2.6), 0 (0) 1 (1.1), 4 (4.5), 0 (0) 0.62 Antihypertensive drugs* 17 (44.7) 35 (39.3) 0.57 Lipid-lowering drugs 4 (10.5) 15 (16.9) 0.36 Antidiabetics drugs 4 (10.5) 8 (9.0) 0.79 Anticoagulant drugs 3 (7.9) 13 (14.6) 0.30 Antiplatelet drugs 6 (15.8) 17 (19.1) 0.66 Serious adverse event, n (%) 38 (100) 68 (76.4) 0.001 Hospitalization 31 (81.6) 45 (50.6) 0.001 Death 4 (10.5) 9 (10.1) 0.94 Characteristics of ADR; concurrent side effects Hypokalemia 10 (26.3) 13 (14.6) 0.12 Hypertension 6 (15.8) 5 (5.6) 0.06 Edema 14 (36.8) 5 (5.6) <0.0001 Troponin increase 4 (10.5) 1 (1.1) 0.01 Acute coronary syndrome (confirmed by angiogram) 0 (0) 1 (1.1) 0.51 Stroke 2 (5.3) 3 (3.4) 0.62 Cardio-pulmonary arrest 2 (5.3) 2 (2.2) 0.37 Neurotoxicity 1 (2.6) 14 (15.7) 0.04 Hepatotoxicity 1 (2.6) 13 (14.6) 0.05 Dermatotoxicity 0 (0) 6 (6.7) 0.10 Renal failure 2 (5.3) 9 (10.2) 0.37 Abbreviations: IQR, interquartile range; SD: Standard-deviation Statistics were performed by Fischer’s exact test for qualitative data and t-test or Mann-Whitney test for quantitative data, as needed. * In details, proportion of patients on antihypertensive drugs developing AT/HF vs. other types of ADR associated with abiraterone were: 6/38 (16%) vs. 11/89 (12%) for angiotensin converting enzyme inhibitors (p:0.6), 5/38 (13%) vs. 15/89 (17%) for angiotensin 2 receptor-blockers (p:0.6), 1/38 (3%) vs. 3/89 (3%) for anti-aldosterone drugs (p:0.8), 6/38 (16%) vs. 9/89 (10%) for loop diuretics, (p:0.4), 4/38 (11%) vs. 8/89 (9%) for thiazide diuretics (p:0.8).

20

Table 3. Details concerning association between clinical covariates and reporting for AT, HF, hypokalemia, edema and hypertension within overall adverse drug reaction on abiraterone (n: 9203) extracted from VigiBase (through 08/2017).

number AT HF Hypokalemia Edema Hypertension (available data) Age (y) 5989 r:0.03 r:0.05 r:0.06 r:0.04 r:0.00 p:0.02 p≤0.0001 p≤0.0001 p≤0.01 p:0.96 Dose of hydrocortisone 3828 r:0.00 r:0.02 r:0.00 r:-0.07 r:-0.12 equivalent (mg/day) p:0.63 p:0.19 p:0.68 p≤0.001 p≤0.0001 Dose of abiraterone 6285 r:0.00 r:0.02 r:0.02 r:0.00 r:0.02 (mg/day) p:0.48 p:0.23 p:0.08 p:0.60 p:0.14

In Supp. Fig 1 are represented distribution of reported dose of abiraterone and hydrocortisone equivalent among patients presenting an ADR on abiraterone in VigiBase.

Statistics: Correlation (r) were assessed by Pearson’s or Spearman’s test, as appropriate.

21

Figure Legends.

Figure 1. Mechanisms of action of the different androgen deprivation therapies (A), with a focus on abiraterone (B). 17alpha-hydroxylase/C(17,20)-lyase inhibition induced by abiraterone (n°1 et n°1’) lead to decreased adrenal glucocorticoids (n°2) and androgens (n°2’) synthesis. Inefficient cortisol synthesis signals the hypothalamic-pituitary axis to increase ACTH (n°3). Consequently, the adrenal glands produce an excess of mineralocorticoid hormone precursors that do not require 17alpha- hydroxylase/C(17,20)-lyase processes, namely 11-deoxycroticosterone (DOC) and corticosterone (n°4). These hormones lead to hypokalemia, hypertension and fluid overload, triggering renin decrease (n°5), explaining normal aldosterone levels. Supplementation of patients on abiraterone with corticosteroids is therefore required to prevent excessive ACTH increase. There is variability in ACTH suppression by this supplementation, leading to inter-individual variability in residual secretion of mineralocorticoid hormone precursors and consequently related clinical impact of hyper- mineralcorticism. Drugs considered as androgen deprivation therapies were: Abiraterone; gonadotropins-releasing hormone (GnRH) agonists (leuprorelin, goserelin, buserelin, triptorelin), GnRH antagonists (degarelix), nonsteroidal androgen receptor inhibitors (enzalutamide, bicalutamide, nilutamide, flutamide) and steroidal androgen receptor inhibitors (mifepristone, )

Figure 2A. Overlap between reporting for atrial tachyarrhythmias (AT), heart failure (HF), hypokalemia, edema and hypertension within each individual safety case report (ICSR) for abiraterone related adverse drug reactions (ADR) vs. other androgen deprivation therapy (ADT). Other ADT includes leuprorelin, goserelin, buserelin, triptorelin, degarelix, bicalutamide, nilutamide, flutamide, enzalutamide, mifepristone, and cyproterone acetate.

Figure 2B. Time to event onset for adverse drug reaction related to heart failure (HF) and atrial tachyarrhythmias (AT) on abiraterone as compared to other ADT, extracted from the international pharmacovigilance database. Other ADT includes leuprorelin, goserelin, buserelin, triptorelin, degarelix, bicalutamide, nilutamide, flutamide, enzalutamide, mifepristone, and cyproterone acetate.

22

Figure 1.

Abbreviations: ACTH: Adrenocorticotropic hormone, AR: androgen receptor, AS: Aldosterone synthase, DHEA: , DHT: , FSH: follicle stimulating hormone, DOC: 11-deoxycorticosterone, GnRH: Gonadotropin-releasing hormone, LH: Luteinizing hormone, 17 β-OHSD: 17 β-OH steroid dehydrogenase, 11β-h: 11 β-hydroxylase, 17 α-H: 17-α- hydroxylase, 17-20L: 17-20 Lyase, 21α-H: 21-α-hydroxylase, 3βHSD: 3-β-hydroxysteroid dehydrogenase

23

Figure 2A.

24

Figure 2B.

25