Eur J Clin Pharmacol DOI 10.1007/s00228-014-1700-3
PHARMACOEPIDEMIOLOGY AND PRESCRIPTION
α1-Adrenergic receptor antagonists and gynecomastia. A case series from the Italian spontaneous reporting system and VigiBase™ Ermelinda Viola & Sibilla Opri & Ugo Moretti & Roberto Leone & Maria Luisa Casini & Sara Ruggieri & Claudia Minore & Anita Conforti
Received: 24 January 2014 / Accepted: 20 May 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose The aim of this study was to analyze the cases of gynecomastia associated with α1A-adrenergic receptor antagonists (α1-ARAs) in the Italian spontaneous reporting system database (Rete Nazionale di Farmacovigilanza or RNF) and in the World Health Organization ICSRs database (VigiBase™), focusing on tamsulosin use. Methods We analyzed the spontaneous reports of gynecomastia related to the use of α1-ARAs and collected from the RNF and from VigiBase™ up to December 2012. Cases of gynecomastia have been defined as reports associated with gynecomastia according with Medical Dictionary for Regulatory Activities (MedDRA). Reporting odds ratio (ROR) and Information Component (IC) were calculated as measures of disproportionality in RNF and VigiBase™, respectively. Results Up to December 2012, about 186,000 reports were recorded in the RNF. Among these, 902 reports of adverse drug reaction (ADR) have been associated with the use of at least one α1-ARAs. Of these, in 15 cases, gynecomastia was a listed ADR: in 10, the suspected drug was tamsulosin (in eight, it was the sole suspect); in two, doxazosin and alfuzosin, respectively; and in one, terazosin. ROR for tamsulosin was 5.3 (95 % CI 1.8, 15.7). In VigiBase™, 84 reports of gynecomastia indicated tamsulosin as suspected drug. Tamsulosin-
Electronic supplementary material The online version of this article (doi:10.1007/s00228-014-1700-3) contains supplementary material, which is available to authorized users. E. Viola (*) : S. Opri : U. Moretti : R. Leone : A. Conforti Pharmacology Unit and Pharmacovigilance Centre of Veneto Region, University Hospital, P.le L.A. Scuro, 10, 37134 Verona, Italy e-mail: [email protected] M. L. Casini : S. Ruggieri : C. Minore Italian Medicines Agency, Pharmacovigilance Office, Agenzia Italiana del Farmaco - Via del Tritone, 181, 00187 Rome, Italy
associated gynecomastia showed the highest IC value within this class of drugs (IC 95 % 2.43). Conclusion In this study, we highlight a possible association between gynecomastia and tamsulosin use. To our knowledge, this association has not been described before and could represent a potential signal. Keywords α1-Adrenergic receptor antagonists . Gynecomastia . Adverse drug reactions . Spontaneous reporting system . Tamsulosin . Pharmacovigilance
Introduction Gynecomastia is a disorder characterized by the enlargement of male breast, caused by glandular proliferation and fat deposition. As regards its prevalence, gynecomastia is common in newborns and adolescents, but occurs also in male adults and especially in the elderly, where it is detected in up to 65 % of men. The etiology of gynecomastia is attributable to physiological factors, endocrine tumors or dysfunctions, non-endocrine diseases, drug use, or idiopathic causes [1]. Gynecomastia is caused by drugs in 10–25 % of all cases, mainly through an imbalance of the estrogen/androgen ratio, or by an enhancement of prolactin secretion [2]. Symptoms of benign prostatic hyperplasia (BHP) are commonly treated with two classes of drugs, 5-alpha reductase inhibitors (5-ARIs), i.e., finasteride, dutasteride, and α1-adrenergic receptor antagonists (α 1 -ARAs) that include doxazosin, terazosin, tamsulosin, alfuzosin, and the recently authorized silodosin. Gynecomastia is an adverse drug reactions (ADRs) known to be associated with the use of 5-ARIs. Alpha1-adrenergic receptors antagonists are prescribed to men with benign prostatic hyperplasia symptoms as they relax
Eur J Clin Pharmacol
prostate smooth muscle without directly interfering with sexual hormone balance. Tamsulosin, as well as alfuzosin and silodosin, is selective for the α1A-adrenergic receptor subtype, which is thought to be predominant in prostatic tissue and for this reason is used for the treatment of BHP only, while doxazosin and terazosin are used also as antihypertensive drugs [3]. To our knowledge, no cases of gynecomastia associated with tamsulosin or other α1A-ARAs are reported in scientific literature to date. Moreover, among the α1-ARAscontaining drugs, gynecomastia is mentioned only in the Summary of Product Characteristics (SPC) of doxazosin products. The aim of our study was to describe the cases of gynecomastia associated with the use of tamsulosin and of the other α1-ARAs recorded in the Italian ADRs database (RNF) and to study this association worldwide by means of the analysis of the data contained in VigiBase™, the pharmacovigilance database of the World Health Organization. The RNF database was set up in 2001 and currently contains more than 210,000 ADR reports sent from all the 20 Italian regions. VigiBase™ is the WHO Global Individual Case Safety Report database, developed and currently maintained by the Uppsala Monitoring Centre, the WHO Collaborating Centre for International Drug Monitoring. VigiBase™ contains over 8.04 million ADR reports coming from more than 110 countries participating in the WHO Programme for International Drug Monitoring, with more than 100,000 new reports added each quarter.
For the present analysis, a case by case survey of the Italian gynecomastia reports associated with the use of α1-ARAst was carried out up to 31 December 2012 in order to study patient characteristics and clinical features of the cases and to discuss causality and pharmacological hypotheses for this association. For each report, the causality assessment performed by the reporter was reviewed and, where missing, calculated by the authors using the Naranjo algorithm [7]. Statistics in RNF To evaluate the strength of gynecomastia-tamsulosin association, reporting odds ratio (ROR) and their confidence intervals were calculated for each drug within the α1-ARAs group taking in consideration the cases of gynecomastia, as previously defined. VigiBase™ data collection As a second step of the analysis, the reports of gynecomastia associated with at least one α1-ARAs were retrieved from VigiBase™ up to 31 December 2012 and analyzed as aggregate cases. The meaningful data collected from the spontaneous reports in VigiBase™ were the following: number of reports of gynecomastia on total number of reports for each α1-ARAs drug, number of reports in which the drug is the only drug suspected, time to onset of the reaction, number of positive re/de-challenge, average age of patients, and number of countries as source of reports.
Methods
Statistics in VigiBase™
Italian data collection
A statistical analysis was performed using the Information Component (IC) as a measure of disproportionality between observed and expected reporting of a specific drug-ADR combination. A positive IC value indicates that a particular drug-ADR combination has been reported more often than expected, based on all the reports in the database [8]. For each IC, we calculated the standard deviation (IC025) as a measure of the strength of the value. The IC025 is the value of the lower 95 % credibility interval for the IC and provides information about the stability of a particular IC value: the narrower the interval, the higher the stability. The IC does not give any information about the qualitative causality of a drug-ADR pair, but shows the quantitative dependencies based on the reports in the WHO database [9, 10].
All the cases of gynecomastia associated with α1-ARAs retrieved from the RNF database and VigiBase™ and recorded up to 31 December 2012 were analyzed. Drugs are classified according to the WHO Anatomical Therapeutic Chemical (ATC) classification system both in RNF and in VigiBase™, while ADRs are coded according to Medical Dictionary for Regulatory Activities (MedDRA) in both databases. Cases of gynecomastia have been defined as reports associated with at least one of the following MedDRA Preferred Terms (PTs): gynecomastia, breast enlargement, and breast pain. All these reactions are grouped in the System Organ Class (SOC) reproductive system and breast disorders. Summary of Product Characteristics of drugs [4], current agreed European Core Safety Profiles of the active ingredients considered [5], Micromedex® [6], and published scientific literature were reviewed to ascertain if gynecomastia was labeled or previously associated with α1-ARAs-containing drug use.
Results Up to 31 December 2012, in the whole database, 902 ADR reports (excluding those from literature) listed at least one α1-
Eur J Clin Pharmacol
ARAs as suspected drug, as shown in Fig. 1. One third of the reports were associated with doxazosin, 253 to tamsulosin, and about 150 to alfuzosin and terazosin. The most frequently reported ADRs concerned the nervous system (219 reports), the skin and subcutaneous tissue (121), and the reproductive system and breast (109). For this last SOC, the most frequent reactions included retrograde ejaculation (49 cases), erectile dysfunction (17 cases), and gynecomastia (15 cases, representing 13.7 % of all reports with an ADR coded within this SOC). The main characteristics of the cases of gynecomastia associated with α1-ARAs use are summarized in Table 1. The first report dates back to 2001; afterwards, one or more reports have been sent to the RNF each year, with the exception of the years 2004, 2005, and 2009, where no reports of gynecomastia were recorded. All reports were sent by health-care professionals from different Italian regions, mostly general practitioners. The causality assessment was defined as probable in six cases and possible in the remaining nine cases. The average age of patients was 68 years (range 39–81 years). The time to onset of the reaction showed a great variability, ranging from 10 days to 4 years from the start of treatment, consistently with the literature data of time to onset of gynecomastia for other drugs [11, 12]. In 10 cases, the suspected drug was tamsulosin (4 % of the total number of tamsulosin ADR reports in the RNF); in two, doxazosin and alfuzosin, respectively; and in one, terazosin. For three patients other ADRs than gynecomastia, e.g., erectile dysfunction (case no. 4), diarrhea, itching, dizziness (case no. 7), and erythema and edema (case no. 8) were also described. In eight patients, tamsulosin was the only suspected drug while in two cases, other drugs (aescin, dutasteride) were coreported as suspected. Where tamsulosin was the only suspected drug, in three cases, there were concomitant drugs
Fig. 1 Number of reports for each α1-ARAs in RNF up to December 2012. The total number of reports is higher than the effective total number of reports in RNF for this class because some reports could contain more than one α1-ARAs as suspected drug
reported as hydrochlorothiazide, nebivolol, cinnarizine, amiloride, and dutasteride. Gynecomastia is at present a known and labeled ADR only for this last drug. Data on de-challenge are often missing but focusing on patients with tamsulosin as only suspected drug; in two of them, gynecomastia recovered after withdrawal (cases no. 7 and 12), and in one (case no. 4), the reaction was not over even after tamsulosin withdrawal. Considering all the reports retrieved (see Table 1), gynecomastia resolved in 53 % of cases. Table 2 shows that gynecomastia was statistically significant associated with the use of tamsulosin (ROR 6.9; 95 % CI 3.6, 13.1) and, to a minor extent, with the other α1-ARAs. Data retrieval from VigiBase™ Up to 31 December 2012, VigiBase™ was containing about 26,350 reports of ADRs associated with α1-ARAs drugs. In Table 3, the cases of gynecomastia on the total number of ADR reports for each α1-ARAs are reported. Tamsulosin was the suspected drug in 9,066 cases, and gynecomastia was indicated as ADR in 84 reports, sent from 12 different countries. An accurate screening of the reports was performed; however, the possibility of duplicates cannot be excluded. The majority of these reports came from the USA (33), the UK (13), Spain (9), and the Netherlands (8). These reports concerned 83 males and one female with an average age of 70 years (range 40–91 years) calculated excluding all the reports where this data was not indicated (18). The latency time to the onset of the ADR varied from 31 days to 5 years. In five cases, a positive de-challenge was described. In six out of 84 cases, dutasteride was reported as suspected drug together with tamsulosin, and two cases mentioned the use of dutasteride as concomitant medication. A causality assessment was present in 13 out of 84 tamsulosinassociated cases. In two of them, the causality assessment was reported as “probable,” whereas “possible” was reported in seven cases, and “unlikely” was reported in the remaining four cases. The MedDRA PTs in the analyzed cases were gynecomastia in 68 cases, breast enlargement in seven cases, and breast pain in 14 (two or more PTs can be listed in the same report). Table 4 shows the Information Component (IC) and standard deviation (IC025) values, updated to October 2013, for α1ARAs, excluding silodosin, since the first two reports of gynecomastia and breast pain were recorded in VigiBase™ during 2013. IC values for PT gynecomastia were positive for all the α1-ARAs considered, and the highest value was detected for tamsulosin (2.43). The IC values for the other two PTs (breast enlargement and breast pain) were lower and >1 for tamsulosin only.
2001 2002 2002 2003 2003 2006 2007 2008 2008 2010 2010 2010
2011 2012 2012
1 2 3 4 5 6 7 8 9 10 11 12
13 14 15
71 64 76
65 39 78 73 73 68 79 42 73 81 81 64
Age (years)
Doxazosin Tamsulosin Tamsulosin
Tamsulosin Doxazosin (irbesartan) Terazosin Tamsulosin Tamsulosin Tamsulosin Tamsulosin Tamsulosin (aescin) Alfuzosin (felodipine) (ramipril) Alfuzosin (warfarin) (bisoprolol) Tamsulosin (dutasteride) Tamsulosin
Suspected α1-ARAs (other suspected drugs)
b
a
8 mg NS 0.4 mg
2 mg 4 mg 5 mg 0.4 mg 0.4 mg 0.4 mg 0.4 mg 0.4 mg 10 mg NS 0.4 mg 0.4 mg
α1-ARAs daily dose
Decrease in the dose NS NS
NS Withdrawn Withdrawn Withdrawn NS Withdrawn Withdrawn Withdrawn Withdrawn Withdrawn Not withdrawn Withdrawn
Actions
No No Lysine, carnitine No Cinnarizine No Dutasteride No No No No (Amiloride/HCT) (nebivolol) No No No
Concomitant drugs
Causality assessment was re-evaluated at the time of the present analysis for all of the reports listed in the table
Time between the start of the treatment and the clinical diagnosis of gynecomastia as recorded in the reports
NS not stated, HCT hydrochlorothiazide
Years
Case
2 months 17 days 33 months
5 months 31 months 25 months 6 months 2 years 20 days 3 months 10 days 21 months 19 months 6 months 4 years
Time onseta
NS Recovering Not recovered
Not recovered NS Recovered Not recovered Not recovered Unknown Recovered Recovered Recovered Recovered Recovering Recovering
Outcome
Probable Possible Possible
Possible Probable Probable Possible Possible Possible Possible Probable Possible Probable Possible Probable
Causalityb α1-ARAs
Table 1 Reports of gynecomastia associated with at least one α1-adrenergic receptor antagonist-containing drug registered in the RNF database up to 31 December 2012
General practitioner General practitioner General practitioner
General practitioner General practitioner General practitioner Hospital physician General practitioner General practitioner Pharmacist General practitioner General practitioner General practitioner General practitioner General practitioner
Sources
Eur J Clin Pharmacol
Eur J Clin Pharmacol Table 2 Reporting odds ratios and corresponding 95 % confidence intervals from the RNF for gynecomastia associated with the use of drugs administered for the treatment of benign prostatic hyperplasia Cases of gynecomastiaa
Drug
Exposed α1-ARAs
5-ARIs Positive control
RORs (CI 95 %) Non-exposed
Crude
Tamsulosin
10/244
372/62,939
6.9 (3.6, 13.1)
Doxazosin Alfuzosin Terazosin Finasteride Dutasteride Spironolactone
2/141 2/149 1/141 27/72 33/56 16/58
380/63,042 380/63,034 381/63,042 355/63,111 349/63,127 366/63,125
2.3 (0.5, 9.5) 2.2 (0.5, 9.0) 1.2 (0.1, 8.4) 66.7 (42.3, 105.0) 106.6 (68.4, 165.9) 47.6 (27.1, 83.5)
Spironolactone is considered as a positive control for gynecomastia RORs reporting odds ratios a
Interpreted as cases of gynecomastia in exposed and in non-exposed individuals, respectively; all reports in individuals receiving the drugs of interest are considered exposed; reports in individuals who were not receiving these drugs are considered non-exposed
Discussion Gynecomastia may have multiple potential etiologies, including chronic medical illnesses, medications, and the presence of tumors [13]. It is estimated that drugs account for 10–25 % of cases of gynecomastia in men; while the level of evidence is good for some drugs such as spironolactone, ketoconazole, anti-androgens or estrogens, and antiretrovirals, for many other drugs, the association between their use and the considered adverse reaction is based on retrospective studies, case– control studies or case series. Drug-induced gynecomastia has been recently reviewed in three scientific publications [11, 14, 15] but only one of them [11] mentions tamsulosin as the eighth reported drug as cause of gynecomastia in the US Food and Drug Administration Adverse Event Reporting System database. Gynecomastia is currently not labeled in the SPC of α1-ARAs, with the exception of doxazosin [16]. Starting from a routinely signal detection activity, the combination of gynecomastia and tamsulosin was filtered out from the Italian ADRs database, with the aim to evaluate if a possible association could be supported by the analysis of spontaneous reporting data present in RNF and VigiBase™. The analysis
was first performed in RNF and subsequently in the WHO database by sorting out for the entire class of α1-ARAs. For almost all the analyzed reports, a plausible temporal sequence between the administration of α1-ARAs and the onset of the pathology was found. Moreover, we did not observe any substantial increase in α1-ARAs utilization or other events in the last 6 years that could explain an increase in reporting gynecomastia. In 46.6 % of the Italian cases, gynecomastia resolved or improved after drug withdrawal (cases no. 3, 7, 8, 9, 10, 11, 12). We also found that tamsulosin was the most suspected α1ARA drug for gynecomastia in both database (66.6 % in RNF and 42.6 % in VigiBase™), as suggested by the signal detection result. It is relevant that tamsulosin was the only drug suspected in 46.6 % of RNF cases and in 28 % of VigiBase™ reports. Considering that these cases originated from two ADRs spontaneous reporting systems, the information provided by the reporters can be subjected to bias. It is important to note that the cases from VigiBase™ contain nine cases from the RNF database described in this article and unfortunately detailed case descriptions were not available for all cases.
Table 3 Reports of gynecomastia registered in VigiBase™ up to 31 December 2012 associated with at least one α1-adrenergic receptor antagonistcontaining drug Suspected drug
No. of reports for gynecomastia/total number of reports for drug
No. of countries
Reports in which the drug is the only drug suspected
Time to onset
Average age (years)
Alfuzosin Tamsulosin Terazosin Doxazosin
18/2,994 84/9,066 46/6,048 49/7,943
8 12 4 9
12 56 41 38
4 days–14 months 31 days–5 years 6 days–4 years 18 days–3 years
69.5 70.29 64.49 65.46
Eur J Clin Pharmacol Table 4 Information Component of α1-adrenergic receptor antagonist-containing drugs Drug
ICa (95 %) breast enlargement
IC025
ICa (95 %) breast pain
IC025
ICa(95 %) gynecomastia
IC025
Alfuzosin Doxazosin Tamsulosin Terazosin
−0.52 0.80 0.55 −1.32
−4.32 −0.37 −0.62 −5.12
−0.57 0.09 0.77 0.48
−3.16 −1.26 0.03 −0.62
1.71 1.58 2.43 2.11
0.92 1.08 2.10 1.62
a
The data are referred to 1 October 2013
Although a causal relationship cannot be established on a case series basis, some of the features of the present case series support the possible association between tamsulosin and gynecomastia. The mechanism by which tamsulosin could cause gynecomastia has not been clarified yet. The presence of α1adrenergic receptors in teat muscles and mammary gland of bovines has been demonstrated previously [17, 18]. On the level of mRNA expression, only two adrenergic receptor subtypes, the α1A(C)- and β2-adrenergic receptors, were detected in the bovine mammary gland [18]. Moreover, in a preclinical study, there are evidences that the doxazosininduced blockade of α1-adrenergic receptors significantly decreases circulating luteinizing hormone and androgens levels [19]. The results of this study showed that orally administered doxazosin imbalanced testosterone homeostasis and modified the transcriptional profile of steroidogenic machinery, cAMP/cGMP signaling pathway, and adrenergic receptors in Leydig cells of adult rats. From these assumptions, it is possible to suppose that the mechanism by which α1-ARAs could cause gynecomastia may involve the adrenergic receptors in Leydig cells. The current findings may alert health-care professionals to the possibility of tamsulosin-induced gynecomastia and may lead to additional evidence for this association.
Conclusions In this study, we examined gynecomastia as potential ADR of tamsulosin to confirm the signal filtered out by signal detection in the Italian ADRs database. To our knowledge, this association has not been described before in scientific literature and is not mentioned in the SPC of tamsulosin. The availability of this new information could be helpful for health-care practitioners in clinical practice, since tamsulosin could be administered with other drugs known to be able to cause gynecomastia, thus increasing the risk of developing this disease. The combination tamsulosin-gynecomastia should probably be highlighted in the product information of α1-ARAs.
Disclosures The opinions expressed by the authors in this article are personal and may not be construed or reported as those of AIFA and WHO. Conflicts of interest The authors declare that they have no conflict of interest.
References 1. Barros AC, Sampaio MC (2012) Gynecomastia: physiopathology, evaluation and treatment. Sao Paulo Med J 130(3):187–197 2. Eckman A, Dobs A (2008) Drug-induced gynecomastia. Expert Opin Drug Saf 7(6):691–702 3. Richardson CD, Donatucci CF, Page SO et al (1997) Pharmacology of tamsulosin: saturation-binding isotherms and competition analysis using cloned alpha 1-adrenergic receptor subtypes. Prostate 33(1): 55–59 4. Summary Product Information of Pamsvax® XL 400 micrograms capsules (Arrow) 5. European Core Safety Profiles, Tamsulosin, http://www.infarmed.pt/ portal/page/portal/INFARMED/MEDICAMENTOS_USO_ HUMANO/FARMACOVIGILANCIA/INFORMACAO_SEGURA NCA/Core%20Safety%20Profile%20(CSP)1/Tamsulosin_CSP_ NL_H_0014_PSUR_002_20130523.doc 6. MICROMEDEX® 2.0 and MICROMEDEX® 1.0 [homepage on the Internet]. Tamsulosin, Drug: Detailed evidence-based information (DRUGDEX®) updated 2013 November 14; cited 2013 November 15]. Available from: http://www.micromedexsolutions.com 7. Naranjo CA, Busto U, Sellers EM et al (1981) A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 30(2):239–245 8. Bate A, Lindquist M, Edwards IR, Olsson S, Orre S, Lansner A, De Freitas RM (1998) A Bayesian neural network method for adverse drug reaction signal generation. Eur J Clin Pharmacol 54:315–321 9. Bate A, Lindquist M, Orre R et al (2002) Data-mining analyses of pharmacovigilance signals in relation to relevant comparison drugs. Eur J Clin Pharmacol 58(7):483–490 10. Noren GN, Sundberg R, Bate A et al (2008) A statistical methodology for drug-drug interaction surveillance. Stat Med 27(16):3057– 3070 11. Bowman J, Kim H, Bustamante J (2012) Drug-induced gynecomastia. Pharmacotherapy 32(12):1123–1140 12. Jeunemaitre X, Chatellier G, Kreft-Jais C et al (1987) Efficacy and tolerance of spironolactone in essential hypertension. Am J Cardiol 60:820–825 13. Dickson G (2012) Gynecomastia. Am Fam Physician 85(7):716–722 14. Deepinder F, Braunstein GD (2012) Drug-induced gynecomastia: an evidence-based review. Expert Opin Drug Saf 11(5):779–795
Eur J Clin Pharmacol 15. Krause W (2012) Drug-inducing gynaecomastia—a critical review. Andrologia 44(Suppl):621–626. doi:10.1111/j.1439-0272.2011. 01240.x 16. Summary Product Information of Cardura® http://www.medicines. org.uk/emc/medicine/1456/spc#undesirable_effects. Accessed 28 August 2013 17. Roets E, Peeters G (1985) Identification and characterization of Prazosin binding to α1-adrenergic receptors in bovine teat muscles. Arch Int Pharmacodyn Ther 275:189–198
18. Wellnitz O, Zurbriggen A, Friis RR, Blum JW, Bruckmaier RM (2001) α1C- and β2-adrenergic receptor mRNA distribution in the bovine mammary gland detected by competitive RT-PCR. J Dairy Res 68:699–704 19. Stojkov NJ, Janjic MM, Kostic TS, Andric SA (2013) Orally applied doxazosin disturbed testosterone homeostasis and changed the transcriptional profile of steroidogenic machinery, cAMP/cGMP signalling and adrenergic receptors in Leydig cells of adult rats. Andrology 1(2):332–347
PHARMACOEPIDEMIOLOGY AND PRESCRIPTION
α1-Adrenergic receptor antagonists and gynecomastia. A case series from the Italian spontaneous reporting system and VigiBase™ Ermelinda Viola & Sibilla Opri & Ugo Moretti & Roberto Leone & Maria Luisa Casini & Sara Ruggieri & Claudia Minore & Anita Conforti
Received: 24 January 2014 / Accepted: 20 May 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose The aim of this study was to analyze the cases of gynecomastia associated with α1A-adrenergic receptor antagonists (α1-ARAs) in the Italian spontaneous reporting system database (Rete Nazionale di Farmacovigilanza or RNF) and in the World Health Organization ICSRs database (VigiBase™), focusing on tamsulosin use. Methods We analyzed the spontaneous reports of gynecomastia related to the use of α1-ARAs and collected from the RNF and from VigiBase™ up to December 2012. Cases of gynecomastia have been defined as reports associated with gynecomastia according with Medical Dictionary for Regulatory Activities (MedDRA). Reporting odds ratio (ROR) and Information Component (IC) were calculated as measures of disproportionality in RNF and VigiBase™, respectively. Results Up to December 2012, about 186,000 reports were recorded in the RNF. Among these, 902 reports of adverse drug reaction (ADR) have been associated with the use of at least one α1-ARAs. Of these, in 15 cases, gynecomastia was a listed ADR: in 10, the suspected drug was tamsulosin (in eight, it was the sole suspect); in two, doxazosin and alfuzosin, respectively; and in one, terazosin. ROR for tamsulosin was 5.3 (95 % CI 1.8, 15.7). In VigiBase™, 84 reports of gynecomastia indicated tamsulosin as suspected drug. Tamsulosin-
Electronic supplementary material The online version of this article (doi:10.1007/s00228-014-1700-3) contains supplementary material, which is available to authorized users. E. Viola (*) : S. Opri : U. Moretti : R. Leone : A. Conforti Pharmacology Unit and Pharmacovigilance Centre of Veneto Region, University Hospital, P.le L.A. Scuro, 10, 37134 Verona, Italy e-mail: [email protected] M. L. Casini : S. Ruggieri : C. Minore Italian Medicines Agency, Pharmacovigilance Office, Agenzia Italiana del Farmaco - Via del Tritone, 181, 00187 Rome, Italy
associated gynecomastia showed the highest IC value within this class of drugs (IC 95 % 2.43). Conclusion In this study, we highlight a possible association between gynecomastia and tamsulosin use. To our knowledge, this association has not been described before and could represent a potential signal. Keywords α1-Adrenergic receptor antagonists . Gynecomastia . Adverse drug reactions . Spontaneous reporting system . Tamsulosin . Pharmacovigilance
Introduction Gynecomastia is a disorder characterized by the enlargement of male breast, caused by glandular proliferation and fat deposition. As regards its prevalence, gynecomastia is common in newborns and adolescents, but occurs also in male adults and especially in the elderly, where it is detected in up to 65 % of men. The etiology of gynecomastia is attributable to physiological factors, endocrine tumors or dysfunctions, non-endocrine diseases, drug use, or idiopathic causes [1]. Gynecomastia is caused by drugs in 10–25 % of all cases, mainly through an imbalance of the estrogen/androgen ratio, or by an enhancement of prolactin secretion [2]. Symptoms of benign prostatic hyperplasia (BHP) are commonly treated with two classes of drugs, 5-alpha reductase inhibitors (5-ARIs), i.e., finasteride, dutasteride, and α1-adrenergic receptor antagonists (α 1 -ARAs) that include doxazosin, terazosin, tamsulosin, alfuzosin, and the recently authorized silodosin. Gynecomastia is an adverse drug reactions (ADRs) known to be associated with the use of 5-ARIs. Alpha1-adrenergic receptors antagonists are prescribed to men with benign prostatic hyperplasia symptoms as they relax
Eur J Clin Pharmacol
prostate smooth muscle without directly interfering with sexual hormone balance. Tamsulosin, as well as alfuzosin and silodosin, is selective for the α1A-adrenergic receptor subtype, which is thought to be predominant in prostatic tissue and for this reason is used for the treatment of BHP only, while doxazosin and terazosin are used also as antihypertensive drugs [3]. To our knowledge, no cases of gynecomastia associated with tamsulosin or other α1A-ARAs are reported in scientific literature to date. Moreover, among the α1-ARAscontaining drugs, gynecomastia is mentioned only in the Summary of Product Characteristics (SPC) of doxazosin products. The aim of our study was to describe the cases of gynecomastia associated with the use of tamsulosin and of the other α1-ARAs recorded in the Italian ADRs database (RNF) and to study this association worldwide by means of the analysis of the data contained in VigiBase™, the pharmacovigilance database of the World Health Organization. The RNF database was set up in 2001 and currently contains more than 210,000 ADR reports sent from all the 20 Italian regions. VigiBase™ is the WHO Global Individual Case Safety Report database, developed and currently maintained by the Uppsala Monitoring Centre, the WHO Collaborating Centre for International Drug Monitoring. VigiBase™ contains over 8.04 million ADR reports coming from more than 110 countries participating in the WHO Programme for International Drug Monitoring, with more than 100,000 new reports added each quarter.
For the present analysis, a case by case survey of the Italian gynecomastia reports associated with the use of α1-ARAst was carried out up to 31 December 2012 in order to study patient characteristics and clinical features of the cases and to discuss causality and pharmacological hypotheses for this association. For each report, the causality assessment performed by the reporter was reviewed and, where missing, calculated by the authors using the Naranjo algorithm [7]. Statistics in RNF To evaluate the strength of gynecomastia-tamsulosin association, reporting odds ratio (ROR) and their confidence intervals were calculated for each drug within the α1-ARAs group taking in consideration the cases of gynecomastia, as previously defined. VigiBase™ data collection As a second step of the analysis, the reports of gynecomastia associated with at least one α1-ARAs were retrieved from VigiBase™ up to 31 December 2012 and analyzed as aggregate cases. The meaningful data collected from the spontaneous reports in VigiBase™ were the following: number of reports of gynecomastia on total number of reports for each α1-ARAs drug, number of reports in which the drug is the only drug suspected, time to onset of the reaction, number of positive re/de-challenge, average age of patients, and number of countries as source of reports.
Methods
Statistics in VigiBase™
Italian data collection
A statistical analysis was performed using the Information Component (IC) as a measure of disproportionality between observed and expected reporting of a specific drug-ADR combination. A positive IC value indicates that a particular drug-ADR combination has been reported more often than expected, based on all the reports in the database [8]. For each IC, we calculated the standard deviation (IC025) as a measure of the strength of the value. The IC025 is the value of the lower 95 % credibility interval for the IC and provides information about the stability of a particular IC value: the narrower the interval, the higher the stability. The IC does not give any information about the qualitative causality of a drug-ADR pair, but shows the quantitative dependencies based on the reports in the WHO database [9, 10].
All the cases of gynecomastia associated with α1-ARAs retrieved from the RNF database and VigiBase™ and recorded up to 31 December 2012 were analyzed. Drugs are classified according to the WHO Anatomical Therapeutic Chemical (ATC) classification system both in RNF and in VigiBase™, while ADRs are coded according to Medical Dictionary for Regulatory Activities (MedDRA) in both databases. Cases of gynecomastia have been defined as reports associated with at least one of the following MedDRA Preferred Terms (PTs): gynecomastia, breast enlargement, and breast pain. All these reactions are grouped in the System Organ Class (SOC) reproductive system and breast disorders. Summary of Product Characteristics of drugs [4], current agreed European Core Safety Profiles of the active ingredients considered [5], Micromedex® [6], and published scientific literature were reviewed to ascertain if gynecomastia was labeled or previously associated with α1-ARAs-containing drug use.
Results Up to 31 December 2012, in the whole database, 902 ADR reports (excluding those from literature) listed at least one α1-
Eur J Clin Pharmacol
ARAs as suspected drug, as shown in Fig. 1. One third of the reports were associated with doxazosin, 253 to tamsulosin, and about 150 to alfuzosin and terazosin. The most frequently reported ADRs concerned the nervous system (219 reports), the skin and subcutaneous tissue (121), and the reproductive system and breast (109). For this last SOC, the most frequent reactions included retrograde ejaculation (49 cases), erectile dysfunction (17 cases), and gynecomastia (15 cases, representing 13.7 % of all reports with an ADR coded within this SOC). The main characteristics of the cases of gynecomastia associated with α1-ARAs use are summarized in Table 1. The first report dates back to 2001; afterwards, one or more reports have been sent to the RNF each year, with the exception of the years 2004, 2005, and 2009, where no reports of gynecomastia were recorded. All reports were sent by health-care professionals from different Italian regions, mostly general practitioners. The causality assessment was defined as probable in six cases and possible in the remaining nine cases. The average age of patients was 68 years (range 39–81 years). The time to onset of the reaction showed a great variability, ranging from 10 days to 4 years from the start of treatment, consistently with the literature data of time to onset of gynecomastia for other drugs [11, 12]. In 10 cases, the suspected drug was tamsulosin (4 % of the total number of tamsulosin ADR reports in the RNF); in two, doxazosin and alfuzosin, respectively; and in one, terazosin. For three patients other ADRs than gynecomastia, e.g., erectile dysfunction (case no. 4), diarrhea, itching, dizziness (case no. 7), and erythema and edema (case no. 8) were also described. In eight patients, tamsulosin was the only suspected drug while in two cases, other drugs (aescin, dutasteride) were coreported as suspected. Where tamsulosin was the only suspected drug, in three cases, there were concomitant drugs
Fig. 1 Number of reports for each α1-ARAs in RNF up to December 2012. The total number of reports is higher than the effective total number of reports in RNF for this class because some reports could contain more than one α1-ARAs as suspected drug
reported as hydrochlorothiazide, nebivolol, cinnarizine, amiloride, and dutasteride. Gynecomastia is at present a known and labeled ADR only for this last drug. Data on de-challenge are often missing but focusing on patients with tamsulosin as only suspected drug; in two of them, gynecomastia recovered after withdrawal (cases no. 7 and 12), and in one (case no. 4), the reaction was not over even after tamsulosin withdrawal. Considering all the reports retrieved (see Table 1), gynecomastia resolved in 53 % of cases. Table 2 shows that gynecomastia was statistically significant associated with the use of tamsulosin (ROR 6.9; 95 % CI 3.6, 13.1) and, to a minor extent, with the other α1-ARAs. Data retrieval from VigiBase™ Up to 31 December 2012, VigiBase™ was containing about 26,350 reports of ADRs associated with α1-ARAs drugs. In Table 3, the cases of gynecomastia on the total number of ADR reports for each α1-ARAs are reported. Tamsulosin was the suspected drug in 9,066 cases, and gynecomastia was indicated as ADR in 84 reports, sent from 12 different countries. An accurate screening of the reports was performed; however, the possibility of duplicates cannot be excluded. The majority of these reports came from the USA (33), the UK (13), Spain (9), and the Netherlands (8). These reports concerned 83 males and one female with an average age of 70 years (range 40–91 years) calculated excluding all the reports where this data was not indicated (18). The latency time to the onset of the ADR varied from 31 days to 5 years. In five cases, a positive de-challenge was described. In six out of 84 cases, dutasteride was reported as suspected drug together with tamsulosin, and two cases mentioned the use of dutasteride as concomitant medication. A causality assessment was present in 13 out of 84 tamsulosinassociated cases. In two of them, the causality assessment was reported as “probable,” whereas “possible” was reported in seven cases, and “unlikely” was reported in the remaining four cases. The MedDRA PTs in the analyzed cases were gynecomastia in 68 cases, breast enlargement in seven cases, and breast pain in 14 (two or more PTs can be listed in the same report). Table 4 shows the Information Component (IC) and standard deviation (IC025) values, updated to October 2013, for α1ARAs, excluding silodosin, since the first two reports of gynecomastia and breast pain were recorded in VigiBase™ during 2013. IC values for PT gynecomastia were positive for all the α1-ARAs considered, and the highest value was detected for tamsulosin (2.43). The IC values for the other two PTs (breast enlargement and breast pain) were lower and >1 for tamsulosin only.
2001 2002 2002 2003 2003 2006 2007 2008 2008 2010 2010 2010
2011 2012 2012
1 2 3 4 5 6 7 8 9 10 11 12
13 14 15
71 64 76
65 39 78 73 73 68 79 42 73 81 81 64
Age (years)
Doxazosin Tamsulosin Tamsulosin
Tamsulosin Doxazosin (irbesartan) Terazosin Tamsulosin Tamsulosin Tamsulosin Tamsulosin Tamsulosin (aescin) Alfuzosin (felodipine) (ramipril) Alfuzosin (warfarin) (bisoprolol) Tamsulosin (dutasteride) Tamsulosin
Suspected α1-ARAs (other suspected drugs)
b
a
8 mg NS 0.4 mg
2 mg 4 mg 5 mg 0.4 mg 0.4 mg 0.4 mg 0.4 mg 0.4 mg 10 mg NS 0.4 mg 0.4 mg
α1-ARAs daily dose
Decrease in the dose NS NS
NS Withdrawn Withdrawn Withdrawn NS Withdrawn Withdrawn Withdrawn Withdrawn Withdrawn Not withdrawn Withdrawn
Actions
No No Lysine, carnitine No Cinnarizine No Dutasteride No No No No (Amiloride/HCT) (nebivolol) No No No
Concomitant drugs
Causality assessment was re-evaluated at the time of the present analysis for all of the reports listed in the table
Time between the start of the treatment and the clinical diagnosis of gynecomastia as recorded in the reports
NS not stated, HCT hydrochlorothiazide
Years
Case
2 months 17 days 33 months
5 months 31 months 25 months 6 months 2 years 20 days 3 months 10 days 21 months 19 months 6 months 4 years
Time onseta
NS Recovering Not recovered
Not recovered NS Recovered Not recovered Not recovered Unknown Recovered Recovered Recovered Recovered Recovering Recovering
Outcome
Probable Possible Possible
Possible Probable Probable Possible Possible Possible Possible Probable Possible Probable Possible Probable
Causalityb α1-ARAs
Table 1 Reports of gynecomastia associated with at least one α1-adrenergic receptor antagonist-containing drug registered in the RNF database up to 31 December 2012
General practitioner General practitioner General practitioner
General practitioner General practitioner General practitioner Hospital physician General practitioner General practitioner Pharmacist General practitioner General practitioner General practitioner General practitioner General practitioner
Sources
Eur J Clin Pharmacol
Eur J Clin Pharmacol Table 2 Reporting odds ratios and corresponding 95 % confidence intervals from the RNF for gynecomastia associated with the use of drugs administered for the treatment of benign prostatic hyperplasia Cases of gynecomastiaa
Drug
Exposed α1-ARAs
5-ARIs Positive control
RORs (CI 95 %) Non-exposed
Crude
Tamsulosin
10/244
372/62,939
6.9 (3.6, 13.1)
Doxazosin Alfuzosin Terazosin Finasteride Dutasteride Spironolactone
2/141 2/149 1/141 27/72 33/56 16/58
380/63,042 380/63,034 381/63,042 355/63,111 349/63,127 366/63,125
2.3 (0.5, 9.5) 2.2 (0.5, 9.0) 1.2 (0.1, 8.4) 66.7 (42.3, 105.0) 106.6 (68.4, 165.9) 47.6 (27.1, 83.5)
Spironolactone is considered as a positive control for gynecomastia RORs reporting odds ratios a
Interpreted as cases of gynecomastia in exposed and in non-exposed individuals, respectively; all reports in individuals receiving the drugs of interest are considered exposed; reports in individuals who were not receiving these drugs are considered non-exposed
Discussion Gynecomastia may have multiple potential etiologies, including chronic medical illnesses, medications, and the presence of tumors [13]. It is estimated that drugs account for 10–25 % of cases of gynecomastia in men; while the level of evidence is good for some drugs such as spironolactone, ketoconazole, anti-androgens or estrogens, and antiretrovirals, for many other drugs, the association between their use and the considered adverse reaction is based on retrospective studies, case– control studies or case series. Drug-induced gynecomastia has been recently reviewed in three scientific publications [11, 14, 15] but only one of them [11] mentions tamsulosin as the eighth reported drug as cause of gynecomastia in the US Food and Drug Administration Adverse Event Reporting System database. Gynecomastia is currently not labeled in the SPC of α1-ARAs, with the exception of doxazosin [16]. Starting from a routinely signal detection activity, the combination of gynecomastia and tamsulosin was filtered out from the Italian ADRs database, with the aim to evaluate if a possible association could be supported by the analysis of spontaneous reporting data present in RNF and VigiBase™. The analysis
was first performed in RNF and subsequently in the WHO database by sorting out for the entire class of α1-ARAs. For almost all the analyzed reports, a plausible temporal sequence between the administration of α1-ARAs and the onset of the pathology was found. Moreover, we did not observe any substantial increase in α1-ARAs utilization or other events in the last 6 years that could explain an increase in reporting gynecomastia. In 46.6 % of the Italian cases, gynecomastia resolved or improved after drug withdrawal (cases no. 3, 7, 8, 9, 10, 11, 12). We also found that tamsulosin was the most suspected α1ARA drug for gynecomastia in both database (66.6 % in RNF and 42.6 % in VigiBase™), as suggested by the signal detection result. It is relevant that tamsulosin was the only drug suspected in 46.6 % of RNF cases and in 28 % of VigiBase™ reports. Considering that these cases originated from two ADRs spontaneous reporting systems, the information provided by the reporters can be subjected to bias. It is important to note that the cases from VigiBase™ contain nine cases from the RNF database described in this article and unfortunately detailed case descriptions were not available for all cases.
Table 3 Reports of gynecomastia registered in VigiBase™ up to 31 December 2012 associated with at least one α1-adrenergic receptor antagonistcontaining drug Suspected drug
No. of reports for gynecomastia/total number of reports for drug
No. of countries
Reports in which the drug is the only drug suspected
Time to onset
Average age (years)
Alfuzosin Tamsulosin Terazosin Doxazosin
18/2,994 84/9,066 46/6,048 49/7,943
8 12 4 9
12 56 41 38
4 days–14 months 31 days–5 years 6 days–4 years 18 days–3 years
69.5 70.29 64.49 65.46
Eur J Clin Pharmacol Table 4 Information Component of α1-adrenergic receptor antagonist-containing drugs Drug
ICa (95 %) breast enlargement
IC025
ICa (95 %) breast pain
IC025
ICa(95 %) gynecomastia
IC025
Alfuzosin Doxazosin Tamsulosin Terazosin
−0.52 0.80 0.55 −1.32
−4.32 −0.37 −0.62 −5.12
−0.57 0.09 0.77 0.48
−3.16 −1.26 0.03 −0.62
1.71 1.58 2.43 2.11
0.92 1.08 2.10 1.62
a
The data are referred to 1 October 2013
Although a causal relationship cannot be established on a case series basis, some of the features of the present case series support the possible association between tamsulosin and gynecomastia. The mechanism by which tamsulosin could cause gynecomastia has not been clarified yet. The presence of α1adrenergic receptors in teat muscles and mammary gland of bovines has been demonstrated previously [17, 18]. On the level of mRNA expression, only two adrenergic receptor subtypes, the α1A(C)- and β2-adrenergic receptors, were detected in the bovine mammary gland [18]. Moreover, in a preclinical study, there are evidences that the doxazosininduced blockade of α1-adrenergic receptors significantly decreases circulating luteinizing hormone and androgens levels [19]. The results of this study showed that orally administered doxazosin imbalanced testosterone homeostasis and modified the transcriptional profile of steroidogenic machinery, cAMP/cGMP signaling pathway, and adrenergic receptors in Leydig cells of adult rats. From these assumptions, it is possible to suppose that the mechanism by which α1-ARAs could cause gynecomastia may involve the adrenergic receptors in Leydig cells. The current findings may alert health-care professionals to the possibility of tamsulosin-induced gynecomastia and may lead to additional evidence for this association.
Conclusions In this study, we examined gynecomastia as potential ADR of tamsulosin to confirm the signal filtered out by signal detection in the Italian ADRs database. To our knowledge, this association has not been described before in scientific literature and is not mentioned in the SPC of tamsulosin. The availability of this new information could be helpful for health-care practitioners in clinical practice, since tamsulosin could be administered with other drugs known to be able to cause gynecomastia, thus increasing the risk of developing this disease. The combination tamsulosin-gynecomastia should probably be highlighted in the product information of α1-ARAs.
Disclosures The opinions expressed by the authors in this article are personal and may not be construed or reported as those of AIFA and WHO. Conflicts of interest The authors declare that they have no conflict of interest.
References 1. Barros AC, Sampaio MC (2012) Gynecomastia: physiopathology, evaluation and treatment. Sao Paulo Med J 130(3):187–197 2. Eckman A, Dobs A (2008) Drug-induced gynecomastia. Expert Opin Drug Saf 7(6):691–702 3. Richardson CD, Donatucci CF, Page SO et al (1997) Pharmacology of tamsulosin: saturation-binding isotherms and competition analysis using cloned alpha 1-adrenergic receptor subtypes. Prostate 33(1): 55–59 4. Summary Product Information of Pamsvax® XL 400 micrograms capsules (Arrow) 5. European Core Safety Profiles, Tamsulosin, http://www.infarmed.pt/ portal/page/portal/INFARMED/MEDICAMENTOS_USO_ HUMANO/FARMACOVIGILANCIA/INFORMACAO_SEGURA NCA/Core%20Safety%20Profile%20(CSP)1/Tamsulosin_CSP_ NL_H_0014_PSUR_002_20130523.doc 6. MICROMEDEX® 2.0 and MICROMEDEX® 1.0 [homepage on the Internet]. Tamsulosin, Drug: Detailed evidence-based information (DRUGDEX®) updated 2013 November 14; cited 2013 November 15]. Available from: http://www.micromedexsolutions.com 7. Naranjo CA, Busto U, Sellers EM et al (1981) A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 30(2):239–245 8. Bate A, Lindquist M, Edwards IR, Olsson S, Orre S, Lansner A, De Freitas RM (1998) A Bayesian neural network method for adverse drug reaction signal generation. Eur J Clin Pharmacol 54:315–321 9. Bate A, Lindquist M, Orre R et al (2002) Data-mining analyses of pharmacovigilance signals in relation to relevant comparison drugs. Eur J Clin Pharmacol 58(7):483–490 10. Noren GN, Sundberg R, Bate A et al (2008) A statistical methodology for drug-drug interaction surveillance. Stat Med 27(16):3057– 3070 11. Bowman J, Kim H, Bustamante J (2012) Drug-induced gynecomastia. Pharmacotherapy 32(12):1123–1140 12. Jeunemaitre X, Chatellier G, Kreft-Jais C et al (1987) Efficacy and tolerance of spironolactone in essential hypertension. Am J Cardiol 60:820–825 13. Dickson G (2012) Gynecomastia. Am Fam Physician 85(7):716–722 14. Deepinder F, Braunstein GD (2012) Drug-induced gynecomastia: an evidence-based review. Expert Opin Drug Saf 11(5):779–795
Eur J Clin Pharmacol 15. Krause W (2012) Drug-inducing gynaecomastia—a critical review. Andrologia 44(Suppl):621–626. doi:10.1111/j.1439-0272.2011. 01240.x 16. Summary Product Information of Cardura® http://www.medicines. org.uk/emc/medicine/1456/spc#undesirable_effects. Accessed 28 August 2013 17. Roets E, Peeters G (1985) Identification and characterization of Prazosin binding to α1-adrenergic receptors in bovine teat muscles. Arch Int Pharmacodyn Ther 275:189–198
18. Wellnitz O, Zurbriggen A, Friis RR, Blum JW, Bruckmaier RM (2001) α1C- and β2-adrenergic receptor mRNA distribution in the bovine mammary gland detected by competitive RT-PCR. J Dairy Res 68:699–704 19. Stojkov NJ, Janjic MM, Kostic TS, Andric SA (2013) Orally applied doxazosin disturbed testosterone homeostasis and changed the transcriptional profile of steroidogenic machinery, cAMP/cGMP signalling and adrenergic receptors in Leydig cells of adult rats. Andrology 1(2):332–347
