Analysis signals of disproportionate reporting associated with donanemab: A retrospective pharmacovigilance study using the FAERS database

Abstract

Objective

This study aims to identify signals of disproportionate reporting (SDR) associated with donanemab by analyzing data from the Food and Drug Administration’s Adverse Event Reporting System (FAERS) database.

Methods

A retrospective analysis was conducted on reports related to donanemab in the FAERS database from the third quarter of 2024 to the second quarter of 2025. Four signal detection methods were employed: reporting odds ratio (ROR), proportional reporting ratio (PRR), Multi-item gamma poisson shrinkage (MGPS), and Bayesian confidence propagation neural network (BCPNN). Additionally, descriptive analysis was performed on the time to onset of reports associated with donanemab.

Results

A total of 646 reports identified donanemab as the primary suspect drug. We identified common SDRs listed on the drug label, including amyloid related imaging abnormality-oedema/effusion, amyloid related imaging abnormality-microhaemorrhages and haemosiderin deposits, headache, and infusion-related reactions. In addition, some SDRs not previously listed on the drug label, such as back pain, feeling hot, and influenza like illness, were also detected. The median time to onset for donanemab-related adverse events was 31.5 days (interquartile range: 25–58 days).

Conclusion

This study provides a comprehensive analysis of SDRs associated with donanemab. These results are merely statistical indicators and require further validation through clinical trials.

Keywords

donanemab FAERS database adverse drug reactions alzheimer’s disease pharmacovigilance

1. Introduction

Alzheimer’s disease is the most prevalent form of dementia in the elderly, and with the aging global population, it has become one of the most significant threats to public health worldwide.^1–3 Patients with Alzheimer’s disease exhibit symptoms such as memory impairment, difficulty with language expression, and a decline in daily living activities, which impose substantial burdens on both families and society.⁴ Traditional treatment strategies primarily aim to increase the concentration of the neurotransmitter acetylcholine in the brain and regulate glutamate levels.^5,6 However, the efficacy of these therapies is limited, as they only alleviate symptoms without halting the progression of the disease.^7,8

The exact pathogenesis of Alzheimer’s disease remains unclear, with the prevailing hypothesis being the “β-amyloid hypothesis”.⁹ This theory posits that the accumulation of β-amyloid plaques in the brain triggers harmful cellular responses, ultimately leading to neuronal dysfunction and death.¹⁰ With advancements in medical technology, monoclonal antibody therapies targeting β-amyloid have emerged as promising treatments. Donanemab, a monoclonal antibody targeting β-amyloid, specifically binds to the N-terminal pyroglutamate site of β-amyloid, facilitating the clearance of β-amyloid plaques from the brain and potentially slowing disease progression.¹¹ Approved for use in the United States in July 2024 for patients with early-stage Alzheimer’s disease, donanemab was granted approval in China in December 2024.¹² Notably, it is the only monoclonal antibody therapy that allows for discontinuation following treatment success. Clinical studies have demonstrated that treatment with donanemab reduces the rate of disease progression by approximately 30% in Alzheimer’s disease patients.¹³ One study showed that donanemab significantly reduced the risk of disease progression, with patients achieving greater independence for extended periods.¹⁴

Despite the significant efficacy of Donanemab, its adverse events (AEs) have raised concerns among researchers. Commonly reported AEs include Amyloid-Related Imaging Abnormalities (ARIA), microhemorrhages, and headaches. Infusion-related reactions (IRRs) are the primary reasons for treatment discontinuation.^15,16 However, because donanemab has only recently entered clinical use, understanding and in-depth investigation of its associated AEs remain at an early stage. In addition, clinical trials are often limited by small sample sizes and suboptimal trial completion rates. Therefore, analyzing data from pharmacovigilance databases can help identify AE signals that may be underreported or unforeseen in clinical trials.

The FAERS database is a spontaneous reporting system and one of the largest pharmacovigilance databases globally.^17,18 It serves as a critical tool for the FDA in post-market drug safety monitoring. This system plays a pivotal role in identifying potential adverse event signals. The aim of this study is to leverage the FAERS database to assess the SDRs associated with donanemab, thereby providing healthcare professionals with more comprehensive and safer medication guidance.

2. Materials and methods

2.1. Data source, management, and study design

This study employs the FAERS database for a retrospective pharmacovigilance analysis. According to the drug marketing time, we retrieved AEs reports identifying donanemab as the primary suspect (PS) drug from the third quarter of 2024 to the second quarter of 2025. Due to the ongoing updates to the FAERS database, duplicate reports of adverse drug events may exist. To address this, we followed the FDA-recommended approach for handling duplicates.¹⁹ We selected the PRIMARYID, CASEID, and FDA_DT fields from the DEMO table. Reports were sorted by CASEID, FDA_DT, and PRIMARYID. For reports with the same CASEID, we retained the one with the largest FDA_DT value. If both CASEID and FDA_DT were identical, the report with the highest PRIMARYID value was retained. In addition, we conducted a manual review to ensure that there were no duplicate records. AEs involved in the reports were classified at the system organ class (SOC) and preferred term (PT) levels using the Medical Dictionary for Regulatory Activities (MedDRA, version 28.0). A detailed flowchart of this study’s methodology is presented in Figure 1. The study complied with the ethical standards of the Helsinki Declaration of 1975 as revised in 2024. All patient details were de-identified to ensure that individual patients could not be identified in any way. This study was exempt from ethical review and informed consent requirements under applicable guidelines, given its exclusive use of publicly accessible, anonymized data sourced from the FAERS database. The reporting of this study conforms to STROBE guidelines.²⁰

Figure 1.

Flowchart of the entire study.

2.2. Statistical analysis

This study used a case/non-case design similar to a case-control study. We studied AEs related to the study drug, rather than the disease itself. Disproportionality analysis is a widely used statistical method in the field of pharmacovigilance, playing a crucial role in identifying AE signals associated with drugs. In this study, we applied four commonly used algorithms: the reporting odds ratio (ROR), proportional reporting ratio (PRR), Multi-item gamma poisson shrinkage (MGPS), and Bayesian confidence propagation neural network (BCPNN) to detect SDRs related to donanemab.^21,22 The background reference group for this study consists of data from the FAERS database covering the period from the third quarter of 2024 to the second quarter of 2025. Supplemental Table 1 presents these four algorithms (all based on 2×2 contingency tables) along with their thresholds. To ensure the stability and reliability of the results, an AE was considered a positive signal only if it met the criteria of all four disproportionality analysis methods. A higher signal value indicates a greater likelihood of a statistical association between the drug and the specific AE. The time to onset (TTO) of donanemab-related AEs was defined as the interval between the start of donanemab therapy and the occurrence of the AE. All statistical analyses were conducted using R software (version 4.5.1).

3. Results

3.1. Descriptive analysis

From the third quarter of 2024 to the second quarter of 2025, a total of 1,610,006 AE reports were retrieved from the FAERS database. Among these, 646 reports identified donanemab as the primary suspect drug, accounting for 1,383 AEs. The clinical characteristics of donanemab-related AEs are summarized in Table 1. In the collected AE reports, the proportion of females (46.75%) was higher than that of males (27.40%). Regarding age distribution, the majority of patients were between 75 and 84 (21.67%). Among the reported outcomes, other serious medical events had the highest proportion (44.12%), followed by hospitalization (8.36%). Additionally, the majority of AE reports were from the United States (96.59%), with most reports submitted by consumers (89.78%).

Table 1.

Clinical characteristics of reports related to donanemab in the FAERS database.

Characteristics	Case number, n	Proportion, %
Overall	646
Sex
Female	302	46.75
Male	177	27.40
Missing	167	25.85
Age(year)
<18	0	0.00
18-64	14	2.17
65-74	94	14.55
75-84	140	21.67
≥85	24	3.72
Missing	374	57.89
Outcome
Hospitalization-Initial or Prolonged	54	8.36
Life-Threatening	16	2.48
Death	13	2.01
Required intervention to prevent permanent impairment/damage	6	0.93
Other Serious (Important Medical Event)	285	44.12
Missing	272	42.11
Report recorder
Consumer	580	89.78
Health Professional	41	6.35
Physician	6	0.93
Pharmacist	5	0.77
Missing	14	2.17
Reported countries
United States	624	96.59
Other countries	22	3.41

3.2. Disproportionality analysis

Donanemab-related AEs were distributed across 20 SOCs, with the number of AE reports for each related SOC presented in Table 2. The SOCs that met the thresholds of all four algorithms were nervous system disorders (n = 612, ROR 10.42, PRR 6.25, EBGM 6.24, IC 2.64) and vascular disorders (n = 74, ROR 2.96, PRR 2.85, EBGM 2.85, IC 1.51). The SOC with the highest number of reports was nervous system disorders, followed by injury, poisoning, and procedural complications.

Table 2.

Signal strength of SDRs associated with donanemab at the SOC level.

System organ class (SOC)	N	ROR (95% two-sided CI)	PRR (χ²)	EBGM (EBGM05)	IC (IC025)
Nervous system disorders	612	10.42(9.37-11.59)	6.25(2900.74)	6.24(5.61)	2.64(2.49)
Injury, poisoning and procedural complications	133	0.65(0.55-0.78)	0.69(22.2)	0.69(0.57)	-0.54(-0.8)
General disorders and administration site conditions	125	0.48(0.4-0.58)	0.53(62.64)	0.53(0.44)	-0.91(-1.18)
Vascular disorders	74	2.96(2.34-3.74)	2.85(90.76)	2.85(2.26)	1.51(1.13)
Gastrointestinal disorders	71	0.59(0.47-0.75)	0.61(18.99)	0.61(0.48)	-0.71(-1.05)
Investigations	68	0.85(0.67-1.09)	0.86(1.69)	0.86(0.67)	-0.22(-0.57)
Psychiatric disorders	64	1.09(0.85-1.4)	1.08(0.42)	1.08(0.84)	0.11(-0.25)
Skin and subcutaneous tissue disorders	57	0.68(0.52-0.89)	0.69(8.12)	0.69(0.53)	-0.53(-0.91)
Respiratory, thoracic and mediastinal disorders	46	0.68(0.51-0.91)	0.69(6.67)	0.69(0.52)	-0.53(-0.95)
Musculoskeletal and connective tissue disorders	39	0.54(0.4-0.75)	0.56(14.53)	0.56(0.4)	-0.85(-1.29)
Eye disorders	22	0.74(0.49-1.13)	0.75(1.92)	0.75(0.49)	-0.42(-1.01)
Immune system disorders	19	1.12(0.71-1.76)	1.11(0.23)	1.11(0.71)	0.16(-0.5)
Infections and infestations	13	0.14(0.08-0.25)	0.15(66.6)	0.15(0.09)	-2.73(-3.42)
Metabolism and nutrition disorders	11	0.39(0.22-0.71)	0.4(10.23)	0.4(0.22)	-1.33(-2.09)
Cardiac disorders	10	0.39(0.21-0.73)	0.4(9.39)	0.4(0.21)	-1.34(-2.13)
Ear and labyrinth disorders	6	1.06(0.48-2.36)	1.06(0.02)	1.06(0.47)	0.08(-1.02)
Renal and urinary disorders	6	0.29(0.13-0.65)	0.3(10.25)	0.3(0.13)	-1.76(-2.7)
Neoplasms benign, malignant and unspecified (incl cysts and polyps)	5	0.21(0.09-0.49)	0.21(15.32)	0.21(0.09)	-2.26(-3.24)
Congenital, familial and genetic disorders	1	0.32(0.04-2.26)	0.32(1.47)	0.32(0.04)	-1.65(-3.09)
Surgical and medical procedures	1	0.04(0.01-0.3)	0.04(21.8)	0.04(0.01)	-4.55(-5.65)

ROR, reporting odds ratio; CI, confidence interval; PRR, proportional reporting ratio; χ²,chi-squared; EBGM, empirical Bayesian geometric mean; EBGM 05, the lower limit of 95% CI of EBGM; IC, information component; IC025, the lower limit of 95% CI of the IC. N: Case numbers.

By integrating the four algorithms, a total of 40 AE signals related to donanemab were identified, spanning 10 different SOCs. This study lists the top 30 PTs based on the frequency of positive signals and the top 30 PTs based on the strength of the ROR signals, with the results presented in Tables 3 and 4. The three most frequent PTs were ARIA-oedema/effusion, ARIA-microhemorrhages and haemosiderin deposits, and headache. The top three PTs based on ROR signal strength were ARIA, ARIA-oedema/effusion, and cerebral microhaemorrhage. Additionally, some SDRs were not listed in the prescribing information, including back pain, disorientation, balance disorder, cognitive disorder, feeling hot, influenza like illness, cerebral atrophy, ischaemic disease, lacunar infarction, subdural haematoma, pallor, and cerebral infarction. Supplemental Table 2 lists all positive PT signals associated with donanemab.

Table 3.

The top 30 SDRs of donanemab ranked by the frequency at PTs levels.

PTs	N	ROR (95% two-sided CI)	PRR (χ²)	EBGM (EBGM05)	IC (IC025)
ARIA-oedema/effusion	117	1809.11(1441.85-2269.9)	1656.14(131165.67)	1122.67(894.76)	10.13(6.42)
ARIA-microhemorrhages and haemosiderin deposits	91	1419.15(1107.99-1817.7)	1325.84(87254.98)	960.5(749.9)	9.91(6.04)
Headache	78	6.91(5.5-8.68)	6.58(371.27)	6.57(5.22)	2.71(2.28)
Infusion related reaction	70	33.51(26.32-42.67)	31.87(2077.23)	31.59(24.81)	4.98(4.11)
ARIA	47	3682.97(2444.13-5549.74)	3557.84(82665.44)	1760.3(1168.19)	10.78(5.03)
Flushing	44	34.04(25.18-46.03)	32.99(1353.48)	32.69(24.18)	5.03(3.82)
Cerebral haemorrhage	39	70.02(50.77-96.56)	68.07(2529.03)	66.79(48.43)	6.06(4.19)
Cerebral microhaemorrhage	27	1598.15(1011.59-2524.81)	1566.97(29141.28)	1080.98(684.23)	10.08(4.14)
Confusional state	24	8.15(5.44-12.21)	8.03(147.66)	8.01(5.35)	3(2.06)
Brain oedema	22	92.88(60.62-142.3)	91.42(1917.42)	89.1(58.16)	6.48(3.59)
Cerebrovascular accident	22	9.49(6.23-14.47)	9.36(164.05)	9.33(6.12)	3.22(2.17)
Back pain*	21	4.75(3.09-7.32)	4.7(61.22)	4.69(3.05)	2.23(1.38)
Chills	20	8.85(5.69-13.76)	8.73(136.83)	8.71(5.6)	3.12(2.04)
Blood pressure increased	19	5.62(3.57-8.84)	5.55(71.02)	5.55(3.53)	2.47(1.52)
Erythema	16	4.03(2.46-6.61)	4(36.06)	4(2.44)	2(1.06)
Seizure	15	4.82(2.9-8.02)	4.78(44.82)	4.77(2.87)	2.25(1.22)
Blood pressure decreased	12	9.3(5.26-16.43)	9.23(87.87)	9.2(5.21)	3.2(1.69)
Hyperhidrosis	11	5.16(2.85-9.34)	5.12(36.52)	5.12(2.83)	2.36(1.09)
Tremor	11	4.06(2.24-7.35)	4.03(25.12)	4.03(2.23)	2.01(0.85)
Chest discomfort	9	4(2.08-7.71)	3.98(20.11)	3.98(2.06)	1.99(0.7)
Disorientation*	7	12.77(6.07-26.87)	12.71(75.27)	12.67(6.02)	3.66(1.34)
Balance disorder*	7	4.63(2.2-9.74)	4.62(19.82)	4.61(2.19)	2.21(0.64)
Magnetic resonance imaging abnormal	6	126.4(55.88-285.92)	125.86(717.31)	121.51(53.72)	6.92(1.62)
Ischaemic stroke	6	22.53(10.08-50.38)	22.44(122.15)	22.3(9.98)	4.48(1.37)
Feeling cold	6	11.64(5.21-25.99)	11.59(57.91)	11.56(5.18)	3.53(1.11)
Cognitive disorder*	6	6.35(2.84-14.16)	6.32(26.86)	6.31(2.83)	2.66(0.75)
Feeling hot*	6	5.54(2.48-12.36)	5.52(22.18)	5.51(2.47)	2.46(0.65)
Influenza like illness*	6	4.79(2.15-10.69)	4.78(17.9)	4.77(2.14)	2.25(0.54)
Blood pressure systolic increased	5	7.89(3.27-18.99)	7.86(29.88)	7.84(3.26)	2.97(0.69)
Cerebral atrophy*	4	59.69(22.19-160.61)	59.52(226.31)	58.54(21.76)	5.87(0.92)

*AEs, that are not mentioned in the drug label; ARIA: amyloid-related imaging abnormalities.

Table 4.

The top 30 signal strength of SDRs associated with donanemab ranked by the ROR at the PTs level.

PTs	N	ROR (95% two-sided CI)	PRR (χ²)	EBGM (EBGM05)	IC (IC025)
ARIA	47	3682.97(2444.13-5549.74)	3557.84(82665.44)	1760.3(1168.19)	10.78(5.03)
ARIA-oedema/effusion	117	1809.11(1441.85-2269.9)	1656.14(131165.67)	1122.67(894.76)	10.13(6.42)
Cerebral microhaemorrhage	27	1598.15(1011.59-2524.81)	1566.97(29141.28)	1080.98(684.23)	10.08(4.14)
ARIA-microhemorrhages and haemosiderin deposits	91	1419.15(1107.99-1817.7)	1325.84(87254.98)	960.5(749.9)	9.91(6.04)
Superficial siderosis of central nervous system	3	475.87(142.26-1591.75)	474.84(1248.3)	417.98(124.96)	8.71(0.44)
Cerebral small vessel ischaemic disease*	3	183.67(57.45-587.18)	183.27(516.64)	174.16(54.48)	7.44(0.48)
Magnetic resonance imaging abnormal	6	126.4(55.88-285.92)	125.86(717.31)	121.51(53.72)	6.92(1.62)
Lacunar infarction*	3	103.65(32.84-327.19)	103.43(295.55)	100.48(31.83)	6.65(0.49)
Infusion related hypersensitivity reaction	3	98.76(31.31-311.52)	98.55(281.72)	95.87(30.39)	6.58(0.48)
Brain oedema	22	92.88(60.62-142.3)	91.42(1917.42)	89.1(58.16)	6.48(3.59)
Cerebral haemorrhage	39	70.02(50.77-96.56)	68.07(2529.03)	66.79(48.43)	6.06(4.19)
Cerebral atrophy*	4	59.69(22.19-160.61)	59.52(226.31)	58.54(21.76)	5.87(0.92)
Flushing	44	34.04(25.18-46.03)	32.99(1353.48)	32.69(24.18)	5.03(3.82)
Infusion related reaction	70	33.51(26.32-42.67)	31.87(2077.23)	31.59(24.81)	4.98(4.11)
Subarachnoid haemorrhage	4	26.71(9.97-71.53)	26.63(97.94)	26.44(9.87)	4.72(0.82)
Ischaemic stroke	6	22.53(10.08-50.38)	22.44(122.15)	22.3(9.98)	4.48(1.37)
Subdural haematoma*	4	21.55(8.05-57.69)	21.49(77.7)	21.37(7.98)	4.42(0.78)
Dementia alzheimer’s type	3	15.28(4.91-47.56)	15.25(39.78)	15.19(4.88)	3.92(0.29)
Disorientation*	7	12.77(6.07-26.87)	12.71(75.27)	12.67(6.02)	3.66(1.34)
Feeling cold	6	11.64(5.21-25.99)	11.59(57.91)	11.56(5.18)	3.53(1.11)
Cerebrovascular accident	22	9.49(6.23-14.47)	9.36(164.05)	9.33(6.12)	3.22(2.17)
Blood pressure decreased	12	9.3(5.26-16.43)	9.23(87.87)	9.2(5.21)	3.2(1.69)
Chills	20	8.85(5.69-13.76)	8.73(136.83)	8.71(5.6)	3.12(2.04)
Pallor*	4	8.67(3.25-23.17)	8.65(27.01)	8.63(3.23)	3.11(0.48)
Confusional state	24	8.15(5.44-12.21)	8.03(147.66)	8.01(5.35)	3(2.06)
Blood pressure systolic increased	5	7.89(3.27-18.99)	7.86(29.88)	7.84(3.26)	2.97(0.69)
Cerebral infarction*	3	7.54(2.42-23.42)	7.52(16.93)	7.51(2.42)	2.91(0.07)
Retching	3	6.95(2.24-21.6)	6.94(15.22)	6.92(2.23)	2.79(0.03)
Headache	78	6.91(5.5-8.68)	6.58(371.27)	6.57(5.22)	2.71(2.28)
Cognitive disorder*	6	6.35(2.84-14.16)	6.32(26.86)	6.31(2.83)	2.66(0.75)

*AEs, that are not mentioned in the drug label; ARIA: amyloid-related imaging abnormalities.

3.3. Onset time of events

We identified a total of 49 reports with incorrect dates and 487 reports with missing dates. After excluding cases with missing or incorrect data, a total of 110 AE reports were included in the TTO analysis. The median onset time for donanemab-related AEs was 31.5 days (interquartile range: 25–58 days). As shown in Figure 2, half of the AEs occurred within the first month of initiating donanemab treatment (n = 55, 50.0%). Notably, AEs continued to occur even after 120 days of donanemab treatment. We further analyzed donanemab-related AEs by dividing them into two categories: AEs related to ARIA and other AEs, with the results shown in Supplemental Figure 1. AEs related to ARIA mainly occurred in the first two months after treatment and almost did not occur after three months of medication. Other AEs mainly occurred within the first month after medication and gradually decreased later.

Figure 2.

Time to onset of AEs associated with donanemab.

3.4. Subgroup analysis

Subgroup analysis was conducted to examine AEs displaying positive signals across different sexes. In both males and females, common SDRs included ARIA-microhemorrhages and haemosiderin deposits, headache, and infusion related reaction. However, certain SDRs were identified in specific sexes. Erythema, blood pressure decreased, and tremor were primarily identified in male patients, while confusional state, seizure, chest discomfort, and balance disorder were predominantly recognized in female patients. Detailed information on sex-stratified SDRs is provided in Supplemental Tables 3 and 4.

4. Discussion

Previous research on donanemab has primarily focused on its clinical efficacy and pharmacological mechanisms, with limited large-scale real-world safety studies.^23,24 This study utilizes the FAERS database to analyze donanemab-related AEs and identify associated SDRs. These findings provide valuable supplementary insights to existing safety research and offer new perspectives for both clinical practice and pharmacovigilance.

Our study demonstrates that the proportion of female patients in donanemab-related AE reports is higher than that of male patients, with the majority of reports involving patients aged 65 years or older. However, the results may be influenced by the substantial proportion of missing sex data (25.85%) and age data (57.89%). The predominance of reports from the United States may be attributed to Donanemab’s initial market approval and availability in the U.S. Among the reported outcomes, other serious medical events had the highest proportion. ‘Other serious medical events’ are defined by the FDA as events that do not meet other outcomes but may endanger the patient and may require medical or surgical intervention (treatment) to prevent the occurrence of other outcomes. This could be because some AEs related to ARIA are classified into this category, as they require monitoring, treatment interruption, or other therapeutic interventions.

Our study found that ARIA-oedema/effusion, ARIA-microhemorrhages and haemosiderin deposits, and headache were the most commonly reported SDRs. These findings are consistent with the drug label. This indirectly indicates the reliability of the research. Previous studies have indicated that monoclonal antibodies targeting aggregated forms of beta amyloid can induce ARIA, often accompanied by edema or hemosiderin deposition.²⁵ A multicenter, randomized, double-blind, placebo-controlled trial revealed that 24.0% of patients in the donanemab group developed ARIA-oedema/effusion.²⁶ ARIA typically occurs early in treatment and may be associated with severe and life-threatening events. Monitoring for ARIA is recommended during the first 24 weeks of donanemab therapy. If relevant symptoms occur, clinical evaluation is advised to determine whether dose adjustment or discontinuation is necessary. Notably, IRRs are also common SDRs. One study reported that 112 participants receiving donanemab discontinued treatment due to adverse events, with infusion related reactions being the most frequent cause of treatment interruption.²⁶ In the event of IRR, healthcare professionals may consider reducing the infusion rate or stopping the infusion entirely, while initiating appropriate treatment based on clinical symptoms. Prior to subsequent administrations, premedication with antihistamines, acetaminophen, or corticosteroids may be considered. Lecanemab is an amyloid-beta protofibril antibody and the first drug specifically designed for the targeted treatment of early Alzheimer’s disease. Comparing the AEs of donanemab and lecanemab enhances our understanding of the safety profiles of both drugs. Yan et al. previously conducted an analysis of the AEs associated with lecanemab.²⁷ Through this comparison, we identified both similarities and differences in the PT signals between the two treatments. For instance, SDRs such as nausea, pyrexia, nasopharyngitis, brain fog, and visual impairment showed positive signals exclusively in lecanemab. Conversely, cerebral haemorrhage, cerebral microhaemorrhage, confusional state, blood pressure increased, and erythema demonstrated positive signals only in donanemab. ARIA-oedema/effusion, ARIA-microhemorrhages and haemosiderin deposits, and headache were observed with relatively high frequency in both drugs. When using these two drugs in clinical practice, physicians should provide differentiated monitoring.

In terms of signal strength, ARIA, ARIA-oedema/effusion, and ARIA-microhemorrhages and haemosiderin deposits exhibited higher signal intensities, underscoring the importance of monitoring for ARIA. Additionally, cerebral microhaemorrhage ranked third in signal strength, which may be related to ARIA. A recent study has shown a substantial concordance between measures of disproportionality and relative risk estimates in analytical studies of medications with established use, suggesting that signal strength can provide an initial indication of clinical significance.¹⁸ Our analysis revealed that the ROR for SDRs related to ARIA significantly exceeded the threshold. Although the FDA approved donanemab for marketing, it issued a “black box” due to the significant risks associated with ARIA. This may have heightened public awareness of ARIA, leading to an increase in related reports in the FAERS database. This could introduce notoriety bias, inflating the number of reports and estimated RORs. In addition, previous clinical studies have shown that donanemab may cause AEs related to ARIA, with an incidence of 36.8%.¹⁶ Evidence indicates that patients who are ApoE ε4 homozygotes have a higher risk of developing ARIA during donanemab treatment than heterozygotes and noncarriers. Therefore, healthcare professionals may consider ApoE ε4 genotyping before treatment initiation and weigh the risks and benefits of therapy based on the test results.

We also identified SDRs not listed on the drug label, including back pain, disorientation, balance disorder, cognitive disorder, feeling hot, influenza like illness, cerebral atrophy, ischemic disease, lacunar infarction, subdural haematoma, pallor, and cerebral infarction. Some studies suggest that neuropathic pain (e.g., thalamic pain) may be associated with central nervous system damage (such as ischemia or hemorrhage).²⁸ Cerebral microhaemorrhage, induced by donanemab, could indirectly lead to referred pain or nerve root irritation. Influenza like illness could arise through several mechanisms, with some studies indicating that flu-like symptoms (e.g., fever, fatigue) may be linked to systemic inflammation following brain ischemia or hemorrhage.^29,30 Healthcare professionals should also consider whether influenza like illness is an infusion related reaction. Disorientation, balance disorder, cognitive disorder, and cerebral atrophy may be attributed to the underlying pathology of alzheimer’s disease itself. It is important to note that these SDRs are statistically significant only and require further investigation and validation. Since patients treated with donanemab are predominantly elderly and often have multiple comorbidities, a wide range of symptoms may occur, and some AEs may not necessarily be related to the drug. Comprehensive future studies are needed to clarify these causal relationships, considering not only the pharmacological properties of the drug but also individual patient differences.

Furthermore, our findings indicate that the median onset time for donanemab-related AEs was 31.5 days (interquartile range: 25–58 days). Although a significant proportion of AEs occurred within the first month of starting donanemab, this finding should be interpreted with caution. This is because the TTO analysis is based on only 110 reports with valid dates, and the study window is short because donanemab was newly marketed. Later events may be underrepresented simply because fewer patients had long follow-up by Q2 2025. The apparent decline over time could reflect reporting, follow-up, or data availability rather than true event timing.

Subgroup analysis revealed differences in the reports between males and females. For male patients, attention can be given to erythema, blood pressure decreased, and tremor. For female patients, monitoring for confusional state, seizure, chest discomfort, and balance disorder is recommended. However, the underlying reasons for these report differences remain unexplained and may be related to physiological factors, immune system differences, or hormonal levels.³¹ Healthcare professionals should remain vigilant in monitoring these SDRs when managing patients of different sexes to ensure the safety of the medication. However, due to the small number of other AEs besides ARIA and infusion reactions, caution should be exercised when interpreting gender differences, as the true distribution of gender subgroups may not be fully obtainable. Future prospective studies based on more comprehensive baseline data are crucial for improving the accuracy of analyses in specific populations and enhancing the reliability and applicability of conclusions.

However, this study acknowledges certain limitations inherent in the use of the FAERS database. As a spontaneous reporting system, FAERS may be subject to underreporting of AEs.³² Moreover, the quality of the reports is influenced by factors such as the expertise and subjective intentions of the reporters. The FAERS database does not encompass all AE reports related to donanemab or the total number of patients treated, making it impossible to accurately calculate the true incidence of AEs. Disproportionality analysis indicates only a statistical association but cannot completely rule out false positives, particularly when some signals are reported in only 3 or 4 reports and the signal strength is extremely high. Sparse cells can produce inflated estimates with wide intervals. Additionally, the database lacks detailed clinical information about patients, such as comorbidities and concurrent medications, which makes it challenging to control for confounding factors. These limitations underscore the need for future research that incorporates more comprehensive clinical data, such as those from randomized controlled trials, to further validate the safety profile of donanemab.

5. Conclusion

This study utilized the FAERS database to analyze donanemab-related AEs, revealing various SDRs associated with its use in real-world settings. In addition to identifying common SDRs listed on the drug label, we also found some SDRs not included on the label, such as back pain, feeling hot, and influenza like illness. In addition, adverse events continued to occur even 120 days after treatment with donanemab, which may underscore the importance of long-term monitoring.

Supplemental material

Supplemental material - Analysis signals of disproportionate reporting associated with donanemab: A retrospective pharmacovigilance study using the FAERS database

Supplemental material for Analysis signals of disproportionate reporting associated with donanemab: A retrospective pharmacovigilance study using the FAERS database by Yujian Zhang, Qineng Gong, Jin Yu, Linlin Zhang, Ye Hu in Science Progress.

Supplemental material

Supplemental material - Analysis signals of disproportionate reporting associated with donanemab: A retrospective pharmacovigilance study using the FAERS database

Footnotes

ORCID iD

Ye Hu

Ethical considerations

Since the FAERS database is publicly available and patient records are anonymized, ethical approval and informed consent were not required for this study.

Author contributions

YZ: Writing the article, data collection, conception and design. QG: Writing the article, analysis and interpretation. JY: Conception and design, analysis and interpretation. LZ: Formal analysis, original draft, critical revision of the article. YH: Writing the article, conception and design. All authors contributed to the article and approved the submitted version.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a grant from Scientific Research Project of Health Commission of Yancheng City (YK2024118).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

No datasets were generated or analysed during the current study.*

Supplemental material

Supplemental material for this article is available online.

References

Phillips

Schwartz

. Nasal lymphatic obstruction of CSF drainage as a possible cause of Alzheimer's disease and dementia. Frontiers in aging neuroscience 2024; 16: 1482255. https://doi.org/10.3389/fnagi.2024.1482255

Yang

Qiu

. Research Progress on the Pathogenesis, Diagnosis, and Drug Therapy of Alzheimer's Disease. Brain sciences 2024; 14(6): 590. https://doi.org/10.3390/brainsci14060590

Feng

Sun

, et al. Global, regional, and national burden of Alzheimer's disease and other dementias, 1990-2019. Frontiers in aging neuroscience 2022; 14: 937486. https://doi.org/10.3389/fnagi.2022.937486

Karger

. Emotional experience in patients with advanced Alzheimer's disease from the perspective of families, professional caregivers, physicians, and scientists. Aging & mental health 2018; 22(3): 316–322. https://doi.org/10.1080/13607863.2016.1261797

Jiang

Liu

Zou

, et al. Discovery of 2-(cyclopropanecarboxamido)-N-(5-((1-(4-fluorobenzyl)piperidin-4-yl)methoxy)pyridin-3-yl)isonicotinamide as a potent dual AChE/GSK3β inhibitor for the treatment of Alzheimer's disease: Significantly increasing the level of acetylcholine in the brain without affecting that in intestine. European journal of medicinal chemistry 2021; 223: 113663. https://doi.org/10.1016/j.ejmech.2021.113663

Conway

. Alzheimer's disease: targeting the glutamatergic system. Biogerontology 2020; 21(3): 257–274. https://doi.org/10.1007/s10522-020-09860-4

Sharma

Rudrawar

Bharate

, et al. Recent advancements in the therapeutic approaches for Alzheimer's disease treatment: current and future perspective. RSC medicinal chemistry 2025; 16(2): 652–693. https://doi.org/10.1039/d4md00630e

Smith

Ownby

. Disease-Modifying Treatments and Their Future in Alzheimer's Disease Management. Cureus 2024; 16(3): e56105. https://doi.org/10.7759/cureus.56105

Daly

. A philosophy of science approach to the amyloid hypothesis of Alzheimer's disease. The European journal of neuroscience 2024; 60(5): 4707–4722. https://doi.org/10.1111/ejn.16500

10.

Tan

Carpio

AMA

Heller

, et al. Behavioral and Neuronal Characterizations, across Ages, of the TgSwDI Mouse Model of Alzheimer's Disease. Genes 2023; 15(1): 47. https://doi.org/10.3390/genes15010047

11.

Espay

. Donanemab in Early Alzheimer’s Disease. The New England journal of medicine 2021; 385(7): 666–667.

12.

Harris

. FDA Green-Lights Second Alzheimer Drug, Donanemab. Jama 2024; 332(7): 524. https://doi.org/10.1001/jama.2024.13386

13.

Gueorguieva

Chow

Chua

, et al. Donanemab exposure-response in early symptomatic Alzheimer's disease. Alzheimer's & dementia : the journal of the Alzheimer's Association 2025; 21(7): e70491. https://doi.org/10.1002/alz.70491

14.

Loeffler

. Antibody-Mediated Clearance of Brain Amyloid-β: Mechanisms of Action, Effects of Natural and Monoclonal Anti-Aβ Antibodies, and Downstream Effects. Journal of Alzheimer's disease reports 2023; 7(1): 873–899. https://doi.org/10.3233/adr-230025

15.

Gueorguieva

Willis

Chua

, et al. Donanemab Population Pharmacokinetics, Amyloid Plaque Reduction, and Safety in Participants with Alzheimer's Disease. Clinical pharmacology and therapeutics 2023; 113(6): 1258–1267. https://doi.org/10.1002/cpt.2875

16.

Kang

. Donanemab: First Approval. Drugs 2024; 84(10): 1313–1318. https://doi.org/10.1007/s40265-024-02087-4

17.

Huang

Chen

, et al. Adverse event profile of lorazepam: a real-world pharmacovigilance study using the FDA adverse event reporting system database. Frontiers in pharmacology 2024; 15: 1465245. https://doi.org/10.3389/fphar.2024.1465245

18.

Xing

Zhang

Wang

, et al. Post-marketing safety concerns with lecanemab: a pharmacovigilance study based on the FDA Adverse Event Reporting System database. Alzheimer's research & therapy 2025; 17(1): 15. https://doi.org/10.1186/s13195-024-01669-4

19.

Xie

Chen

, et al. Psychiatric disorders associated with fluoroquinolones: a pharmacovigilance analysis of the FDA adverse event reporting system database. Frontiers in pharmacology 2024; 15: 1435923. https://doi.org/10.3389/fphar.2024.1435923

20.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet (London, England) 2007; 370(9596): 1453–1457. https://doi.org/10.1016/S0140-6736(07)61602-X

21.

Gao

Cao

Liu

. Analysis and mining of Dupilumab adverse events based on FAERS database. Scientific reports 2025; 15(1): 8597. https://doi.org/10.1038/s41598-025-92330-z

22.

Chen

Jiang

Zhu

, et al. Mining and evaluation of adverse event signals for capmatinib based on the FAERS database. Frontiers in pharmacology 2024; 15: 1417661. https://doi.org/10.3389/fphar.2024.1417661

23.

Ayton

. Brain volume loss due to donanemab. European journal of neurology 2021; 28(9): e67–e68. https://doi.org/10.1111/ene.15007

24.

Ross

Weinberg

Arnold

. Cost-effectiveness of Aducanumab and Donanemab for Early Alzheimer Disease in the US. JAMA neurology 2022; 79(5): 478–487. https://doi.org/10.1001/jamaneurol.2022.0315

25.

Salloway

Pain

Lee

, et al. TRAILBLAZER-ALZ 4: A phase 3 trial comparing donanemab with aducanumab on amyloid plaque clearance in early, symptomatic Alzheimer's disease. Alzheimer's & dementia : the journal of the Alzheimer's Association 2025; 21(5): e70293. https://doi.org/10.1002/alz.70293

26.

Sims

Zimmer

Evans

, et al. Donanemab in Early Symptomatic Alzheimer Disease: The TRAILBLAZER-ALZ 2 Randomized Clinical Trial. Jama 2023; 330(6): 512–527. https://doi.org/10.1001/jama.2023.13239

27.

Yan

Zhang

, et al. A real-world disproportionality analysis of FDA adverse event reporting system (FAERS) events for lecanemab. Frontiers in pharmacology 2025; 16: 1559447. https://doi.org/10.3389/fphar.2025.1559447

28.

Huang

Gao

, et al. Fgr contributes to hemorrhage-induced thalamic pain by activating NF-κB/ERK1/2 pathways. JCI insight 2020; 5(20). https://doi.org/10.1172/jci.insight.139987

29.

Yoosuf

Alawneh

Hashim

, et al. Dengue fever with bilateral thalamic diffusion restriction lesions: a case report. BMC neurology 2025; 25(1): 185. https://doi.org/10.1186/s12883-025-04197-2

30.

Liddle

Dirks

Almekhlafi

, et al. An Ambiguous Role for Fever in Worsening Outcome After Intracerebral Hemorrhage. Translational stroke research 2023; 14(2): 123–136. https://doi.org/10.1007/s12975-022-01010-x

31.

Liu

Yin

Fan

, et al. A real-world pharmacovigilance study of Certolizumab pegol based on FAERS database. Scientific reports 2025; 15(1): 28529. https://doi.org/10.1038/s41598-025-13502-5

32.

Wan

Zhou

Chen

, et al. Analysis and mining of brodalumab adverse events based on FAERS database. Scientific reports 2025; 15(1): 18175. https://doi.org/10.1038/s41598-025-03192-4

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.14 MB

0.41 MB