Evaluation of adverse event profiles for leuprolide and goserelin: Insights from the FDA adverse event reporting system following STROBE guidelines.

1. Introduction
Breast cancer (BC) is the most prevalent malignant tumor among women and a leading cause of cancer-related mortality worldwide.[1,2] According to GLOBOCAN 2020 data, BC has surpassed lung cancer as the cancer with the highest global incidence.[1,2] Among newly diagnosed cancers in women, BC accounts for approximately 24.5%, with related deaths comprising about 15.5%.[3,4] The risk of developing BC increases with age, typically peaking after 50 years; however, in Asia, a higher proportion of cases are diagnosed in patients under 40.[3,4] BC risk factors are categorized into non-modifiable and modifiable factors. Non-modifiable factors included gender, age, genetic predisposition (e.g., BRCA1 or BRCA2 mutations), and family history, while modifiable factors involved hormone exposure (e.g., early menarche, late menopause, infertility, or late first childbirth), long-term hormone replacement therapy, obesity, alcohol consumption, and physical inactivity.[5–8]
BC is classified based on its histological and molecular characteristics. The most prevalent pathological type is infiltrative BC, comprising 70% to 80% of all cases.[9] Molecular subtypes included Luminal A, Luminal B, HER2-enriched, and Basal-like (triple-negative BC), each exhibiting significant differences in treatment and prognosis.[10] For example, Luminal A generally has a favorable prognosis, while Basal-like BC is associated with a poorer prognosis but demonstrates good responsiveness to chemotherapy.
BC treatment encompassed surgery, radiotherapy, chemotherapy, endocrine therapy, and targeted therapy.[11,12] Early-stage patients often underwent breast-conserving surgery combined with radiotherapy, while late-stage patients primarily relied on systemic therapies. Targeted therapies, such as trastuzumab for HER2-positive patients, have significantly improved survival rates. Endocrine therapy was effective for hormone receptor-positive patients, particularly premenopausal women who often required ovarian function suppression (OFS) to reduce recurrence risk.[13–15] GnRH agonists initially activate GnRH receptors but subsequently induce their downregulation, resulting in a sustained reduction in estrogen production. Leuprolide and Goserelin, both GnRH agonists, are commonly used for OFS.
Leuprolide is a GnRH agonist that played a critical role in various sex hormone-related diseases by regulating gonadal hormone secretion.[11] Leuprolide was widely used in the treatment of prostate cancer, BC, endometriosis, and uterine fibroids.[11,16–18] In BC, Leuprolide was commonly used for OFS in premenopausal patients.[11] In the treatment of BC, combining Leuprolide with endocrine therapy has been shown to reduce the risk of recurrence, particularly in premenopausal patients with hormone receptor-positive BC.[11] Additionally, Leuprolide was used in assisted reproductive technology to regulate ovarian response and prevent premature ovulation.[19] Goserelin was also a GnRH agonist, and its mechanism of action and indications were similar to those of Leuprolide.[11] Recent studies indicate that although the 2 agents exert comparable therapeutic effects, their AE profiles may differ, underscoring the need for a rigorous comparative safety assessment.
Although several clinical studies have described the safety profiles of Leuprolide and Goserelin, much of the existing evidence is based on small cohorts or controlled trials, which may be insufficient to detect rare or long-term adverse events (AEs).[20–22] Real-world pharmacovigilance data from spontaneous reporting systems, such as FAERS, enable large-scale detection of safety signals through established methodologies, including the ROR, PRR, chi-square test, and BCPNN. However, to date, no study has systematically compared the AE profiles of Leuprolide and Goserelin using a comprehensive FAERS-based evaluation. Therefore, this study aimed to systematically compare the AE profiles of Leuprolide and Goserelin through analysis of FAERS data spanning 2004 to 2024, with the goal of identifying significant pharmacovigilance signals for each drug.

2. Methods

2.1. Source and population
This study utilized FAERS data spanning from the first quarter of 2004 to the third quarter of 2024, extracted with R (v 4.1.2) software. Managed by the FDA, FAERS is a publicly accessible database containing reports from healthcare professionals regarding cancer patients worldwide. The database compiles and analyzes AE reports for drugs and biologics. Each AE report is encoded with preferred terms (PTs) from the FAERS system’s Medical Dictionary for Regulatory Activities (MedDRA v 27.1), alongside the associated system organ classes. This study retrieved data on patient gender, age at the time of AE occurrence, report receipt date, report country, event, and outcome. FAERS data is voluntarily submitted by healthcare providers, consumers, drug manufacturers, and clinical researchers, which may introduce biases, including underreporting, misreporting, and duplicate submissions.

2.2. Drug identification and AEs
Due to the lack of standardization in drug naming within FAERS, both brand and generic names were employed to identify specific records of target drugs during data extraction. For reports involving 2 drugs, text strings such as “Camcevi KIT, Eligard KIT, Fensolva KIT, Lupaneta Pack, Lupron, Lupron Depot pet KIT, Lupron Depot, Viadur, Leuprolide acetate, Leuprolide acetate for depot suspension” and “Zoladex, Goserelin acetate” were used. Searches were performed using PTs, and records were counted based on individual safety reports, with AEs being the primary focus. To minimize bias, 2 researchers independently categorized AE reports and gathered clinical characteristics, including gender, age, and AE outcomes. To mitigate the risk of “indication bias” (where drug indications are reported as AEs, PTs associated with cancer-related signs and complications were excluded from analysis. An unexpected AE was defined as any significant AE not listed in the FDA’s drug prescribing information. According to the role classification in the drug table, only reports in which the drug was designated as the “primary suspect” were included in the analysis to enhance the specificity of AE attribution, while those listing the drug as a “secondary suspect” were excluded. To ensure data completeness and consistency, reports with missing or incomplete demographic variables were excluded from the final analysis.

2.3. Study design and statistical analysis
We utilized disproportionality analysis (DPA) to examine a 2 × 2 contingency table based on drug exposure and AEs. To evaluate the relationship between the exposed and nonexposed populations, we calculated the ROR and the PRR. Higher ROR and PRR values signified a stronger association between the AEs and the medication. To minimize the risk of false-positive AE signals, we validated the results using the BCPNN.[23,24] A signal was considered valid if the value of a was ≥3, the lower threshold of the 95% confidence interval (CI) for the ROR exceeded 1, the PRR was >2 with a corresponding χ2 value exceeding 4, and the lower bound of the 95% CI for the information component (IC) was above 0. Additionally, based on the IC 95% CI, we generated a time-trend chart of safety signals, which illustrated the temporal pattern of AEs associated with a drug in FAERS. If the time-trend chart exhibited a consistent upward trend and the 95% CI narrowed, the signal was considered stable, indicating a strong correlation between the drug and the AE. Using the BCPNN signal strength criteria, medium and strong signals were selected for analysis and discussion when the signal value, IC – 2SD, was ≥1.0.[25]
All analyses were conducted using R Studio (v 4.1.2), and all images were generated with the same software. Categorical data were presented as percentages, and the Breslow-Day statistical method was employed to compare the ROR of AEs between Leuprolide and Goserelin. A 2-tailed test was applied, with a significance threshold set at P <.05 (Fig. 1).

3. Results

3.1. General reporting trends
Between the first quarter of 2004 and the third quarter of 2024, a total of 12,418,989 AEs reports were collected from the United States FDA AE Reporting System (FAERS). Among these, 4253 were related to Goserelin, and 64,324 were related to Leuprolide. Goserelin was approved for use in 1989, while Leuprolide was first used in 1998. The majority of reports originated from the population aged 50 to 79. Male respondents submitted 46.65% of Goserelin AE reports and 66.19% of Leuprolide AE reports, mainly related to prostate cancer. Female respondents submitted 24.13% of Leuprolide AE reports (mainly for the endometriosis and uterine leiomyoma), and 15.42% of Goserelin AE reports (predominantly for BC) (Table 1). Figure 2 presented the annual AE report numbers for both drugs, highlighting a consistently low and fluctuating trend for Goserelin (<500/year) and a generally increasing trend for Leuprolide, peaking in 2024.
Table 2 presented the top 3 indications and top 5 concomitant drugs in AE reports for Leuprolide and Goserelin. Leuprolide was most commonly used to treat prostate cancer (38,422, 59.73%) and endometriosis (6673, 10.37%). Goserelin was most commonly used to treat prostate cancer (2148, 50.51%) and BC (287, 6.75%). Leuprolide was often used in combination with enzalutamide (3022, 1.95%) and bicalutamide (2226, 1.44%). Goserelin was frequently used in combination with bicalutamide (408, 4.14%), letrozole (164, 1.66%), and tamoxifen (148, 1.50%). Table 3 listed the top 20 drugs associated with the highest number of AE reports for Leuprolide and Goserelin.

3.2. Key signal detection results
Table 4 presents AEs meeting the reporting odds ratio (ROR), proportional reporting ratio (PRR), and Bayesian Confidence Propagation Neural Network (BCPNN) criteria (IC >1). Several AEs showed strong associations with both drugs, including elevated prostate-specific antigen (Leuprolide: χ2 = 84009.65, 95% CI lower bound = 52.04, IC-2Standard Deviation [SD] = 5.42; Goserelin: χ2 = 820.31, 95% CI lower bound = 14.44, IC-2SD = 3.38), increased blood testosterone (Leuprolide: χ2 = 22845.52, 95% CI lower bound = 68.07, IC-2SD = 5.56; Goserelin: χ2 = 290.23, 95% CI lower bound = 15.77, IC-2SD = 2.28), and abnormal PSA (Leuprolide: χ2 = 51615.98, 95% CI lower bound = 204.65, IC-2SD = 6.46; Goserelin: χ2 = 422.43, 95% CI lower bound = 25.63, IC-2SD = 2.16). In addition, metastatic prostate cancer (Leuprolide: χ2 = 35400.65, 95% CI lower bound = 69.22, IC-2SD = 5.65; Goserelin: χ2 = 900.27, 95% CI lower bound = 26.67, IC-2SD = 3.38) and hot flushes (Leuprolide: χ2 = 373221.78, 95% CI lower bound = 51.61, IC-2SD = 5.39; Goserelin: χ2 = 3439.89, 95% CI lower bound = 15.59, IC-2SD = 3.82) were also significantly associated.
Compared to Goserelin, Leuprolide was more likely to be associated with several adverse reactions, including pituitary apoplexy (χ2 = 1112.99, 95% CI lower = 38.38, IC-2SD = 3.4), terminal state (Leuprolide: χ2 = 16093.21, 95% CI lower = 29.79, IC-2SD = 4.68; Goserelin: χ2 = 178.7, 95% CI lower = 8.04, IC-2SD = 2.23), injection site abscess (Leuprolide: χ2 = 3881.22, 95% CI lower = 22.35, IC-2SD = 4.18; Goserelin: χ2 = 28.52, 95% CI lower = 3.39, IC-2SD = 0.5), and blood testosterone abnormalities (Leuprolide: χ2 = 38710.42, 95% CI lower = 149.61, IC-2SD = 6.21; Goserelin: χ2 = 113.84, 95% CI lower = 10.32, IC-2SD = 1.14), among others.
In contrast, Goserelin was more prone than Leuprolide to certain adverse reactions, including expulsion of medication (Leuprolide: χ2 = 15.6, 95% CI lower = 2.05, IC-2SD = 0.38; Goserelin: χ2 = 26970.08, 95% CI lower = 549.62, IC-2SD = 4.83), device use errors (χ2 = 1133.91, 95% CI lower = 45.17, IC-2SD = 3.09), arterial injury (χ2 = 1207.44, 95% CI lower = 55.89, IC-2SD = 2.84), bone metastases (Leuprolide: χ2 = 7170, 95% CI lower = 13.02, IC-2SD = 3.61; Goserelin: χ2 = 5433.75, 95% CI lower = 38.56, IC-2SD = 4.81), spine metastases (Leuprolide: χ2 = 1227.1, 95% CI lower = 10.69, IC-2SD = 3.22; Goserelin: χ2 = 654.25, 95% CI lower = 22.52, IC-2SD = 3.1), malignant neoplasm progression (χ2 = 8248.82, 95% CI lower = 21.36, IC-2SD = 4.28), and liver metastases (χ2 = 656.75, 95% CI lower = 11.84, IC-2SD = 3.17).

3.3. Comparative adverse events analysis
Time scans of safety signals illustrated the variation of AE–drug associations over time (Fig. 3), with error bars representing the 95% CI of the IC; intersections with zero indicate nonsignificant associations. Leuprolide showed strong correlations with several AEs, including night sweats, decreased and increased blood testosterone, bone and lung metastases, increased prostate-specific antigen, hot flushes, death, endometriosis, and injection site hemorrhage. Associations with night sweats and endometriosis peaked in 2015, while correlations with other AEs exhibited a fluctuating upward trend. For Goserelin, bone metastases showed a peak in 2018, hot flushes remained relatively stable, and death exhibited a bimodal fluctuation with peaks in 2015 and 2021.

3.4. Time-trend analysis
We compared the intensity of AE signals across 10 disease categories (Fig. 4), where dot color represents disease type, dot size reflects AE frequency, and position is determined by ROR and chi-square values, with higher positions indicating stronger drug–AE correlations. Leuprolide showed strong correlations with endometriosis (reproductive and breast disorders), prostate cancer metastases (neoplastic conditions), intercepted product preparation errors (injuries, poisoning, procedural complications), and palliative care (surgical and medical procedures). Goserelin was strongly correlated with pituitary hemorrhage in endocrine disorders.

4. Discussion
This study, based on the FAERS database, analyzed the AEs report data for Leuprolide and Goserelin from the first quarter of 2004 to the third quarter of 2024. The results indicated that AE reports for Leuprolide were primarily concentrated in the categories of “Reproductive system and breast disorders,” “Neoplastic benign, malignant, and unspecified,” “Injury, poisoning, and procedural complications,” and “Surgical and medical procedures.” The most frequent adverse reactions included “endometriosis,” “metastatic precursor cancer,” “intercepted product preparation errors,” and “palliative care.” In contrast, Goserelin exhibited a higher reporting frequency of AEs within the category of “Endocrine disorders,” with a particularly strong disproportionality signal for “pituitary hemorrhage.” Overall, more prominent and frequent safety signals were observed for Leuprolide compared with Goserelin. The findings of this study are consistent with those reported in previous research.[20–22] This study may still be influenced by residual confounding. Differences in disease severity, comorbidities, treatment strategies, and reporting practices across databases could have influenced the observed associations. Moreover, confounding by indication may exist, as patients with more advanced disease are more likely to receive these therapies. Consequently, some of the detected safety signals may partially reflect underlying clinical characteristics rather than the drug effects alone. It should be emphasized that the signals identified in this study represent disproportionality associations derived from spontaneous reporting data rather than evidence of causal relationships between the drugs and the reported AEs.
The mechanism of action of Leuprolide and Goserelin involved binding to the GnRH receptor in the anterior pituitary gland. Initially, this binding stimulated the secretion of gonadotropins, including Luteinizing Hormone and Follicle-Stimulating Hormone. However, sustained stimulation led to receptor desensitization, which ultimately inhibited the release of gonadotropins and reduced the production of gonadal hormones, such as estrogen and testosterone.[26] This pharmacological pathway underpins the efficacy of Leuprolide and Goserelin in the treatment of hormone-dependent diseases, including BC, prostate cancer, and endometriosis. Importantly, the endocrine suppression induced by GnRH agonists provides a mechanistic basis for several of the endocrine-related AEs identified in this study.
Leuprolide and Goserelin, both GnRH agonists, shared similar mechanisms of action, resulting in comparable adverse reactions. These adverse effects were primarily related to the initial transient increase in, followed by the long-term inhibition of, sex hormone production. During the early stages of treatment, the temporary surge in sex hormone levels might exacerbate existing symptoms. Common adverse reactions included hot flashes, alterations in sexual function, reduced bone density, headaches, and emotional fluctuations.[27] Hot flashes were one of the most frequently reported side effects, particularly during the early phase of treatment. As therapy progresses and sex hormone levels continued to decline, patients may experience symptoms associated with low hormone levels, such as hot flashes, sweating, and facial redness. Changes in sexual function, including reduced libido, erectile dysfunction, and menstrual irregularities or amenorrhea, were also common. These changes were mainly due to the suppression of sex hormone production, especially in women, where ovarian function was inhibited, potentially leading to disruptions in the menstrual cycle.[26,27] Prolonged use of Leuprolide or Goserelin may result in decreased bone density, increasing the risk of fractures. This side effect was particularly pronounced in patients receiving long-term treatment, necessitating regular bone density monitoring and consideration of preventive measures such as calcium and vitamin D supplementation.
Leuprolide and Goserelin exhibited similar mechanisms of action and indications but differed in their dosage forms and patient compliance considerations. Leuprolide offered various dosage forms, such as daily injections and long-acting injections, providing flexibility for patients with differing treatment needs. In contrast, Goserelin was primarily available as an implant, which was well-suited for patients requiring long-term, stable hormone suppression, although it may increase discomfort associated with local implantation. Leuprolide may have a slightly faster onset of action, while the long-acting formulation of Goserelin provided more consistent hormone suppression. Regarding adverse reactions, both drugs were associated with the inhibition of sex hormone levels. In clinical practice, selecting the most appropriate treatment regimen should be based on the patient’s specific needs, with regular monitoring of bone health and metabolic status.
This study, based on the FAERS database, provided a comprehensive analysis of the adverse reaction patterns of Leuprolide and Goserelin, particularly in relation to their impact across multiple organ systems. The data offered valuable insight into the adverse reaction profiles of these drugs in clinical settings. However, the study had several limitations. First, as the FAERS database relied on voluntary reporting, reporting bias was a concern, particularly regarding the underreporting of minor adverse reactions. In addition, the FAERS system does not allow confirmation of causality, and duplicate or incomplete reports may further affect the accuracy of signal detection. Second, because the data was sourced from diverse regions and institutions, variations in treatment protocols and patient severity may influence the incidence and types of adverse reactions observed. Despite these limitations, the newly identified endocrine-related and laboratory-based AEs observed in this study still provide meaningful safety signals that warrant further investigation.

5. Conclusion
This study analyzed adverse reaction data from the FAERS database to identify the adverse reaction profiles of Leuprolide and Goserelin. The present analysis highlights the distinct safety characteristics of the 2 GnRH agonists, emphasizing their differential profiles across key system organ classes. The findings indicated that the adverse reactions associated with Leuprolide were predominantly concentrated in categories such as “Reproductive system and breast disorders,” “Neoplastic benign, malignant, and nonspecific,” “Injury, poisoning, and procedural complexes,” and “Surgical and medical procedures.” In contrast, Goserelin was strongly correlated with adverse reactions related to “Endocrine disorders.”
These comparative differences suggest that clinicians may need to adopt tailored AE monitoring strategies when selecting between the 2 agents, particularly in hormone-dependent disease management. These results provided valuable data to support clinicians in making informed decisions when selecting and administering Leuprolide and Goserelin, ultimately contributing to the development of safer and more effective treatment strategies for BC patients. Nevertheless, the interpretation of these findings should consider inherent FAERS limitations, including reporting bias, inability to infer causality, and lack of demographic adjustment. Future research may benefit from validating these safety signals using real-world clinical cohorts or integrating traditional disproportionality analyses with machine-learning methods to enhance the robustness and predictive value of pharmacovigilance assessments.

Acknowledgments
We express our heartfelt appreciation to our mentors, study participants, loved ones, and colleagues, as well as to the academic community whose insightful contributions have profoundly shaped and enhanced this research. Your steadfast support, encouragement, and trust have been invaluable throughout this journey.

Author contributions
Conceptualization: Xiaoyi Ren.
Formal analysis: Ningning Ren.
Methodology: Ningning Ren.
Resources: Fangyu Zhou, Mengjie Yang, Wenzhong Ji, Jiamin Ma.
Supervision: Xiaoyi Ren.
Visualization: Fangyu Zhou, Mengjie Yang, Wenzhong Ji, Jiamin Ma.
Writing – original draft: Ningning Ren.
Writing – review & editing: Ningning Ren.