Abstract
Background
Hypoxia-inducible factor prolyl hydroxylase inhibitors (HIF-PHIs) play an important role in the treatment of renal anemia. However, some studies suggest a potential link between HIF-PHIs and increased thrombosis risk, though these findings are inconsistent and lack large-scale clinical support. We aim to summarize embolic and thrombotic events associated with HIF-PHIs in different populations in real world, especially among high-risk patients.
Methods
Using the Japan Adverse Drug Event Report (JADER) database from January 1, 2020, to September 30, 2023, a disproportionality analysis was employed to identify embolic and thrombotic events associated with HIF-PHIs using the reporting odds ratios (ROR) and information component (IC). We also evaluated the time to onset among different populations.
Results
From January 2020 to September 2023, the JADER database reported a total of 253,599 cases, including 1,222 cases of embolic and thrombotic events, which represented 30.44% of all HIF-PHIs reported. Embolic and thrombotic events (ROR = 6.68) related to HIF-PHIs is positive signal, with the strongest signal observed for vessel type unspecified and mixed arterial and venous (ROR = 7.97). The signal intensity is higher in females than in males (p = 0.008) and also greater in the non-dialysis population compared to the dialysis population (p < 0.001). The median onset time was shorter in the dialysis population than in the non-dialysis population (days: 27 vs. 47, p = 0.016).
Conclusion
Attention to embolic and thrombotic events associated with HIF-PHIs is essential, with careful selection of specific types based on underlying diseases, sex, age, and indications.
Keywords: Hypoxia-inducible factor prolyl hydroxylase inhibitors, renal anemia, embolic and thrombotic events, JADER
1. Introduction
Renal anemia is a prevalent complication among patients with chronic kidney disease (CKD), significantly impacting their quality of life and survival rates. Approximately 23% of individuals with CKD experience anemia in the early stages, with the incidence soaring to nearly 90% in those with end-stage renal disease [1]. The current treatment landscape primarily includes erythropoiesis-stimulating agents and iron supplementation; however, a subset of patients does not respond adequately and continues to struggle with anemia [2]. As research progresses regarding the safety and efficacy of these treatments, there is a growing interest in new therapeutic options, particularly hypoxia-inducible factor prolyl hydroxylase inhibitors (HIF-PHIs). Hypoxia-inducible factor (HIF) is a key transcription factor that modulates gene expression to boost erythropoiesis and to control metabolism and angiogenesis when oxygen levels are low. HIF-PH inhibitors prevent the destabilization and degradation of HIFs, allowing the transcription factors to drive hypoxia programming and generate new red blood cells [3,4]. Currently, HIF-PHIs (daprodustat, enarodustat, molidustat, roxadustat, and vadadustat) have been widely used to treat renal anemia in Japan.
Despite the remarkable achievement in boosting hemoglobin levels, the unclear embolic and thrombotic adverse events (AEs) pose a great challenge to the clinical application. Preliminary studies have assessed the safety of HIF-PHIs through clinical trials, yet these trials often employ stringent enrollment criteria that may not accurately represent the broader patient population. Some clinical trials indicate an increased incidence of embolic and thrombotic events among patients receiving HIF-PHIs, but these findings are frequently limited by small sample sizes and short follow-up periods [5,6]. Additionally, many clinical trials have not adequately examined the long-term cardiovascular safety of HIF-PHIs, highlighting the need for more comprehensive data to validate these findings [7]. However, the safety analysis of HIF-PHIs using real-world data focused only on the risk of hypothyroidism and gastrointestinal hemorrhage [8,9], and data are lacking regarding the comparative analysis of real-world safety of HIF-PHIs used for embolic and thrombotic events comprehensively.
Therefore, we conducted large-scale and long-term study in real world using the latest data in the Japan Adverse Drug Event Report (JADER) database from 2020 to 2023 to provide an in-depth and comprehensive understanding of embolic and thrombotic events associated with HIF-PHIs and to provide a useful reference for clinical practices.
2. Materials and methods
2.1. Data source
We conducted a pharmacovigilance study on embolic and thrombotic events associated with HIF-PHIs based on the JADER database, a publicly available database containing approximately 900,000 adverse event (AE) cases reported since 2004 [10]. The PMDA website (https://www.pmda.go.jp/index.html) provides access to the JADER database data, which is available in four CSV files: ‘Demo’, ‘Drug’, ‘Reac’, and ‘Hist’. Demographic information such as age, sex, height, weight, and reporting year is included in the ‘Demo’ file. The ‘Drug’ file provides information on drug names, administration dates, and classifications related to adverse events, labeling drugs as ‘suspected’, ‘concomitant’, or ‘interaction’. The ‘Reac’ file documents adverse events, noting onset dates and clinical outcomes. Finally, the patient’s medical history is kept in the ‘Hist’ file. Data from these files can be combined using identification numbers to create an extensive dataset.
2.2. Drug selection and adverse reaction definition
This study extracted all reported cases of HIF-PHIs, including roxadustat, daprodustat, vadadustat, enarodustat, and molidustat, in the JADER database from January 1, 2020, to September 30, 2023. The inclusion was based on the date each drug was approved in Japan (Table S1). Only drugs categorized under the role code ‘suspected drug’ were considered for this analysis. Adverse events in the JADER database are categorized by Preferred Terms (PTs) as per the Medical Dictionary for Regulatory Activities (MedDRA), version 26.1 [11]. We screened the HIF-PHIs reports using standardized MedDRA queries (SMQs) specifically for PTs labeled ‘Embolic and thrombotic events’ (SMQ code: 20000081), which encompasses 416 PTs. This includes PTs for ‘Embolic and thrombotic events, arterial’ (20000082), ‘Embolic and thrombotic events, vessel type unspecified and mixed arterial and venous’ (20000083), and ‘Embolic and thrombotic events, venous’ (20000084) (Tables S2-4) [12]. The screening process is detailed in Figure 1.
Figure 1.
Flow chart of the analysis process in JADER. A detailed description of the selection process of embolic and thrombotic events for hypoxia-inducible factor prolyl hydroxylase inhibitors (HIF-PHIs) in the Japan Adverse Drug Event Report (JADER) database. Embolic and thrombotic events included embolic and thrombotic events, arterial (20000082), embolic and thrombotic events, vessel type unspecified and mixed arterial and venous (20000083) and embolic and thrombotic events, venous (20000084).
2.3. Statistical analysis
One of the most vital methods in pharmacovigilance research is disproportionality analysis, which is utilized for the quantitative assessment of data in pharmacovigilance databases. More robust results were achieved by calculating reporting odds ratios (ROR), Bayesian confidence propagation neural networks of information component (IC), and their 95% credibility intervals (CrI) using the shrinkage transformations statistical formula [13] (Tables S5-6). An AE was classified as a positive signal when the lower limit of the 95% CrI of ROR (ROR025) exceeded 1 and the number of cases was no less than three. The threshold for a positive signal is defined by the lower limit of the 95% CrI of IC (IC025) being greater than 0 [14]. Conversely, a negative signal, indicating that the target drug is largely irrelevant to the target AE, will be generated if the ROR025, IC025, or the number of target AE reports does not meet these specified parameters. Higher values of ROR and IC suggest stronger imbalances and signal strength, indicating that the target drugs are more likely to cause the target AEs than other drugs in the JADER database.
Subgroup analyses were conducted across various categories such as gender, age groups, indications, and underlying diseases to enhance the reliability and stability of the research results. Differences in ROR and IC related to sex and age for each AE were tested and quantified as the ratio of reporting odds ratios (RROR) and ratio of information components (RIC), respectively. The z-test was employed to assess differences among the subgroups using the following formula:
where ê1 and ê2 are the natural logarithms of RORs or ICs, and Sê1 and Sê2 are their respective standard errors [15].
We used the Kaplan-Meier method to estimate the event free probabilities for the time to onset of HIF-PHIs-related embolic and thrombotic events. The Wilcoxon test and Kruskal-Wallis test were used to compare the median time to onset between different groups. In this study, p < 0.05 was considered statistically significant, and all statistical tests were two-tailed. All statistical analyses were conducted using R 4.4.1 and Microsoft Excel 2021. Ethical approval and informed consent were not required for this study as the JADER database is publicly accessible and patient identity information is kept confidential.
3. Results
3.1. Basic characteristics
From January 2020 to September 2023, a total of 253,599 cases (including 4,015 HIF-PHIs cases) were recorded in the JADER database after the ‘suspected drug’ were applied (Figure 1). Embolic and thrombotic events accounted for 30.44% of all HIF-PHIs reported in JADER. Roxadustat had the highest total reported embolic and thrombotic events with 936 (76.60%) followed by daprodustat with 193 (15.79%), while enarodustat and molidustat had the lowest with 13 (1.06%) AEs respectively. The clinical characteristics of patients with HIF-PHIs-related embolic and thrombotic events were described in Table 1. Specifically, embolic and thrombotic events induced by HIF-PHIs were generally reported more frequently in male (n = 636, 52.05%) and 80 to 89 years (n = 364, 29.79%). In addition, there was a higher proportion of cases reported in 2021 (n = 401, 32.82%). Embolic and thrombotic events are fatal or life threatening, with 7.14% of patients showing unrecovered, 6.25% of patients showing sequelae, and 9.47% of patients dying because of HIF-PHIs-induced embolic and thrombotic events.
Table 1.
Characteristics of patients with HIF-PHIs-related embolic and thrombotic events in JADER.
| Roxadustat (N = 936) | Daprodustat (N = 193) | Vadadustat (N = 67) | Enarodustat (N = 13) | Molidustat (N = 13) | Overall HIF-PHIs (N = 1,222) | |
|---|---|---|---|---|---|---|
| Sex | ||||||
| Female | 372 (39.74%) | 81 (41.97%) | 38 (56.72%) | 4 (30.77%) | 6 (46.15%) | 501 (41.00%) |
| Male | 517 (55.24%) | 80 (41.45%) | 26 (38.81%) | 9 (69.23%) | 4 (30.77%) | 636 (52.05%) |
| Missing | 47 (5.02%) | 32 (16.58%) | 3 (4.48%) | 0 (0.00%) | 3 (23.08%) | 85 (6.96%) |
| Age (years) | ||||||
| <10 | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) |
| 10 to 19 | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) |
| 20 to 29 | 2 (0.21%) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) | 1 (7.69%) | 3 (0.25%) |
| 30 to 39 | 6 (0.64%) | 0 (0.00%) | 1 (1.49%) | 0 (0.00%) | 0 (0.00%) | 7 (0.57%) |
| 40 to 49 | 32 (3.42%) | 1 (0.52%) | 0 (0.00%) | 1 (7.69%) | 0 (0.00%) | 34 (2.78%) |
| 50 to 59 | 63 (6.73%) | 5 (2.59%) | 6 (8.96%) | 0 (0.00%) | 0 (0.00%) | 74 (6.06%) |
| 60 to 69 | 147 (15.71%) | 19 (9.84%) | 5 (7.46%) | 2 (15.38%) | 1 (7.69%) | 174 (14.24%) |
| 70 to 79 | 279 (29.81%) | 32 (16.58%) | 21 (31.34%) | 3 (23.08%) | 2 (15.38%) | 337 (27.58%) |
| 80 to 89 | 264 (28.21%) | 65 (33.68%) | 26 (38.81%) | 5 (38.46%) | 4 (30.77%) | 364 (29.79%) |
| 90 to 99 | 76 (8.12%) | 35 (18.13%) | 7 (10.45%) | 2 (15.38%) | 2 (15.38%) | 122 (9.98%) |
| ≥100 | 0 (0.00%) | 0 (0.00%) | 1 (1.49%) | 0 (0.00%) | 0 (0.00%) | 1 (0.08%) |
| Missing | 67 (7.16%) | 36 (18.65%) | 0 (0.00%) | 0 (0.00%) | 3 (23.08%) | 106 (8.67%) |
| Year | ||||||
| 2020 | 257 (27.46%) | 21 (10.88%) | 3 (4.48%) | 0 (0.00%) | 0 (0.00%) | 281 (23.00%) |
| 2021 | 305 (32.59%) | 66 (34.20%) | 22 (32.84%) | 2 (15.38%) | 6 (46.15%) | 401 (32.82%) |
| 2022 | 270 (28.85%) | 56 (29.02%) | 29 (43.28%) | 1 (7.69%) | 4 (30.77%) | 360 (29.46%) |
| 2023 | 104 (11.11%) | 50 (25.91%) | 13 (19.40%) | 10 (76.92%) | 3 (23.08%) | 180 (14.73%) |
| Clinical outcome | ||||||
| Recovery | 498 (44.86%) | 70 (28.69%) | 16 (22.22%) | 6 (40.00%) | 1 (6.25%) | 591 (40.56%) |
| Improvement | 168 (15.14%) | 41 (16.80%) | 22 (30.56%) | 4 (26.67%) | 5 (31.25%) | 240 (16.47%) |
| Sequelae | 56 (5.05%) | 21 (8.61%) | 13 (18.06%) | 0 (0.00%) | 1 (6.25%) | 91 (6.25%) |
| Unrecovered | 86 (7.75%) | 11 (4.51%) | 5 (6.94%) | 2 (13.33%) | 0 (0.00%) | 104 (7.14%) |
| Death | 110 (9.91%) | 19 (7.79%) | 6 (8.33%) | 1 (6.67%) | 2 (12.50%) | 138 (9.47%) |
| Missing | 192 (17.30%) | 82 (33.61%) | 10 (13.89%) | 2 (13.33%) | 7 (43.75%) | 293 (20.11%) |
JADER, Japan Adverse Drug Event Report; HIF-PHIs, Hypoxia-Inducible Factor Prolyl Hydroxylase Inhibitors.
3.2. Disproportionality analysis
The ROR (95% credibility interval [Crl]) for HIF-PHIs was 6.68 (6.28, 7.09) for embolic and thrombotic events, 4.31 (3.76, 4.94) for arterial, 6.04 (5.35, 6.81) for venous and 7.97 (7.41, 8.57) for vessel type unspecified and mixed arterial and venous, as shown in Figure 2. The IC (95% Crl) for HIF-PHIs was 2.74 (2.65, 2.80) for embolic and thrombotic events, 2.99 (2.89, 3.07) for arterial, 2.59 (2.40, 2.73) for venous and 2.11 (1.89, 2.27) for vessel type unspecified and mixed arterial and venous, respectively (Figure 3). The signal of arterial thrombosis events was strongest in molidustat (ROR = 5.44, IC = 2.44), followed by roxadustat (ROR = 4.52, IC = 2.18), but enarodustat (IC = 1.71, 95% Crl [-0.36, 2.92]) had a positive signal in the ROR but not in the IC. The signal of venous thrombotic events in roxadustat (ROR = 6.17, IC = 2.63) was the strongest, the signal of enarodustat (ROR = 4.33, IC = 2.11) was the weakest, but molidustat (ROR025 = 0.62, IC025 = −1.25) had a negative signal in the ROR and IC. There were positive signals in mixed thrombotic events with all HIF-PHIs. Overall, the signal intensity in mixed thrombotic events is greater than that in venous thrombotic events, which is higher than that in arterial thrombotic events.
Figure 2.
Signals detection at the SMQ level for embolic and thrombotic events by ROR. Forest plot shows the ROR of HIF-PHIs-related embolic and thrombotic events. (a) Embolic and thrombotic events (SMQ); (b) Embolic and thrombotic events, arterial (SMQ); (c) Embolic and thrombotic events, venous (SMQ); (d) Embolic and thrombotic events, vessel type unspecified and mixed arterial and venous (SMQ). A signal was considered present when the lower limit of the 95% CrI of the calculated ROR exceeded 1. The percentage represents the proportion of specific adverse events associated with the target drug out of all adverse events.
Figure 3.
Signals detection at the SMQ level for embolic and thrombotic events by IC. Forest plot shows the IC of HIF-PHIs-related embolic and thrombotic events. (a) Embolic and thrombotic events (SMQ); (b) Embolic and thrombotic events, arterial (SMQ); (c) Embolic and thrombotic events, venous (SMQ); (d) Embolic and thrombotic events, vessel type unspecified and mixed arterial and venous (SMQ). A signal was considered present when the lower limit of the 95% CrI of the calculated IC exceeded 0. The percentage represents the proportion of specific adverse events associated with the target drug out of all adverse events.
Peripheral artery occlusion (ROR = 14.05, IC = 3.81) was the strongest positive signal in arterial thrombotic events for HIF-PHIs. Deep vein thrombosis (ROR = 12.61, IC = 3.66) was the strongest positive signal in venous thrombotic events and shunt occlusion (ROR = 58.34, IC = 5.87) was the strongest positive signal in mixed thrombotic events (Table 2). Additionally, roxadustat had the highest reported shunt occlusion (n = 306) and the strongest positive signal (ROR = 74.79, IC = 6.22).
Table 2.
Disproportionality analysis of HIF-PHIs at PT level in JADER.
| SMQ | PT | Daprodustat |
Enarodustat |
Molidustat |
Roxadustat |
Vadadustat |
Overall HIF-PHIs |
||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n | ROR (95% Crl) | IC (95% Crl) | n | ROR (95% Crl) | IC (95% Crl) | n | ROR (95% Crl) | IC (95% Crl) | n | ROR (95% Crl) | IC (95% Crl) | n | ROR (95% Crl) | IC (95% Crl) | n | ROR (95% Crl) | IC (95% Crl) | ||
| Embolic and thrombotic events, arterial | Acute myocardial infarction | 9 | 4.71 (2.43, 9.13) | 2.24 (1.10, 3.00) | 0 | NA | NA | 1 | 2.63 (0.36, 19.01) | 1.40 (–2.39, 3.08) | 27 | 4.97 (3.37, 7.33) | 2.31 (1.67, 2.77) | 7 | 7.96 (3.75, 16.89) | 2.99 (1.69, 3.84) | 43 | 5.70 (4.17, 7.80) | 2.51 (2.00, 2.87) |
| Arterial occlusive disease | 0 | NA | NA | 1 | 2.97 (0.40, 21.94) | 1.57 (–2.21, 3.26) | 0 | NA | NA | 5 | 6.37 (2.48, 16.35) | 2.67 (1.11, 3.66) | 0 | NA | NA | 6 | 6.40 (2.67, 15.35) | 2.68 (1.26, 3.59) | |
| Lacunar infarction | 5 | 5.60 (2.29, 13.66) | 2.48 (0.92, 3.47) | 1 | NA | NA | 1 | 2.87 (0.40, 20.83) | 1.52 (–2.26, 3.21) | 14 | 6.90 (3.98, 11.94) | 2.79 (1.88, 3.41) | 1 | 2.34 (0.33, 16.76) | 1.23 (–2.56, 2.91) | 21 | 7.76 (4.91, 12.28) | 2.96 (2.22, 3.47) | |
| Mesenteric arterial occlusion | 0 | NA | NA | 2 | NA | NA | 0 | NA | NA | 3 | 5.41 (1.52, 19.16) | 2.43 (0.37, 3.64) | 0 | NA | NA | 3 | 4.94 (1.39, 17.50) | 2.30 (0.23, 3.51) | |
| Myocardial infarction | 5 | 1.98 (0.82, 4.79) | 0.99 (–0.57, 1.97) | 3 | NA | NA | 0 | NA | NA | 28 | 3.54 (2.42, 5.17) | 1.82 (1.19, 2.27) | 6 | 5.59 (2.49, 12.55) | 2.48 (1.07, 3.39) | 39 | 3.53 (2.55, 4.87) | 1.82 (1.29, 2.20) | |
| Peripheral arterial occlusive disease | 3 | 4.36 (1.38, 13.78) | 2.13 (0.06, 3.33) | 4 | NA | NA | 1 | 2.92 (0.40, 21.25) | 1.55 (–2.24, 3.23) | 20 | 13.64 (8.36, 22.26) | 3.77 (3.02, 4.29) | 1 | 2.55 (0.35, 18.33) | 1.35 (–2.43, 3.04) | 25 | 13.28 (8.45, 20.86) | 3.73 (3.06, 4.20) | |
| Peripheral artery occlusion | 3 | 4.87 (1.53, 15.47) | 2.28 (0.21, 3.49) | 5 | NA | NA | 2 | 4.90 (1.18, 20.42) | 2.29 (–0.30, 3.68) | 16 | 13.44 (7.72, 23.40) | 3.75 (2.91, 4.33) | 0 | NA | NA | 21 | 14.05 (8.47, 23.30) | 3.81 (3.08, 4.32) | |
| Peripheral artery thrombosis | 1 | 2.48 (0.34, 18.17) | 1.31 (–2.47, 3.00) | 6 | NA | NA | 0 | NA | NA | 11 | 13.62 (6.67, 27.83) | 3.77 (2.74, 4.46) | 0 | NA | NA | 12 | 12.66 (6.30, 25.45) | 3.66 (2.68, 4.33) | |
| Transient ischemic attack | 3 | 3.74 (1.19, 11.76) | 1.90 (–0.17, 3.11) | 11 | NA | NA | 0 | NA | NA | 10 | 5.40 (2.84, 10.26) | 2.43 (1.36, 3.16) | 0 | NA | NA | 13 | 5.30 (3.00, 9.37) | 2.41 (1.47, 3.05) | |
| Embolic and thrombotic events, venous | Deep vein thrombosis | 29 | 8.76 (6.03, 12.72) | 3.13 (2.51, 3.57) | 1 | 2.32 (0.32, 16.70) | 1.21 (–2.57, 2.90) | 0 | NA | NA | 134 | 13.41 (11.15, 16.14) | 3.75 (3.46, 3.95) | 12 | 9.35 (5.24, 16.66) | 3.22 (2.25, 3.89) | 176 | 12.61 (10.69, 14.88) | 3.66 (3.41, 3.84) |
| Embolism venous | 5 | 6.69 (2.72, 16.43) | 2.74 (1.18, 3.73) | 0 | NA | NA | 1 | 2.91 (0.40, 21.19) | 1.54 (–2.24, 3.23) | 4 | 2.86 (1.05, 7.77) | 1.52 (–0.25, 2.60) | 0 | NA | NA | 10 | 5.20 (2.71, 9.98) | 2.38 (1.30, 3.11) | |
| Pulmonary embolism | 11 | 3.40 (1.87, 6.18) | 1.77 (0.74, 2.46) | 2 | 3.86 (0.95, 15.77) | 1.95 (–0.64, 3.34) | 0 | NA | NA | 46 | 4.62 (3.43, 6.23) | 2.21 (1.72, 2.56) | 4 | 3.36 (1.25, 9.01) | 1.75 (–0.02, 2.83) | 63 | 4.52 (3.50, 5.85) | 2.18 (1.76, 2.48) | |
| Pulmonary infarction | 2 | 4.00 (0.97, 16.58) | 2.00 (–0.59, 3.39) | 0 | NA | NA | 0 | NA | NA | 2 | 2.74 (0.66, 11.33) | 1.45 (–1.14, 2.84) | 1 | 2.80 (0.38, 20.38) | 1.48 (–2.30, 3.17) | 5 | 5.07 (1.99, 12.90) | 2.34 (0.78, 3.33) | |
| Pulmonary thrombosis | 1 | 2.31 (0.32, 16.77) | 1.21 (–2.57, 2.90) | 0 | NA | NA | 0 | NA | NA | 1 | 1.51 (0.21, 10.96) | 0.60 (–3.19, 2.28) | 2 | 4.60 (1.11, 19.00) | 2.20 (–0.39, 3.59) | 4 | 3.76 (1.35, 10.45) | 1.91 (0.15, 2.99) | |
| Retinal vein occlusion | 4 | 5.14 (1.90, 13.92) | 2.36 (0.60, 3.44) | 0 | NA | NA | 0 | NA | NA | 6 | 3.72 (1.64, 8.44) | 1.89 (0.48, 2.81) | 0 | NA | NA | 10 | 4.63 (2.43, 8.83) | 2.21 (1.13, 2.94) | |
| Venous thrombosis | 1 | 2.21 (0.31, 15.99) | 1.15 (–2.64, 2.83) | 0 | NA | NA | 0 | NA | NA | 2 | 2.29 (0.56, 9.39) | 1.20 (–1.40, 2.59) | 0 | NA | NA | 3 | 2.62 (0.82, 8.37) | 1.39 (–0.68, 2.60) | |
| Venous thrombosis limb | 2 | 2.60 (0.64, 10.50) | 1.38 (–1.22, 2.77) | 1 | 2.86 (0.40, 20.75) | 1.52 (–2.26, 3.21) | 0 | NA | NA | 10 | 5.16 (2.72, 9.80) | 2.37 (1.29, 3.09) | 1 | 2.36 (0.33, 16.92) | 1.24 (–2.54, 2.93) | 14 | 5.43 (3.13, 9.40) | 2.44 (1.54, 3.06) | |
| Embolic and thrombotic events, vessel type unspecified and mixed arterial and venous | Atrial thrombosis | 0 | NA | NA | 2 | 4.94 (1.17, 20.93) | 2.31 (–0.29, 3.70) | 0 | NA | NA | 5 | 6.30 (2.46, 16.13) | 2.65 (1.09, 3.64) | 0 | NA | NA | 7 | 7.29 (3.21, 16.56) | 2.87 (1.56, 3.72) |
| Brain stem infarction | 2 | 3.25 (0.80, 13.21) | 1.70 (–0.89, 3.09) | 0 | NA | NA | 0 | NA | NA | 8 | 6.09 (2.95, 12.60) | 2.61 (1.39, 3.41) | 1 | 2.59 (0.36, 18.65) | 1.37 (–2.41, 3.06) | 11 | 6.51 (3.46, 12.23) | 2.70 (1.68, 3.40) | |
| Cardiac ventricular thrombosis | 1 | 2.09 (0.29, 15.02) | 1.06 (–2.72, 2.75) | 0 | NA | NA | 1 | 2.94 (0.40, 21.48) | 1.56 (–2.23, 3.24) | 3 | 2.85 (0.90, 9.06) | 1.51 (–0.56, 2.72) | 0 | NA | NA | 5 | 3.59 (1.45, 8.91) | 1.85 (0.28, 2.83) | |
| Cerebellar infarction | 0 | NA | NA | 0 | NA | NA | 0 | NA | NA | 6 | 4.38 (1.92, 10.01) | 2.13 (0.72, 3.04) | 1 | 2.56 (0.36, 18.38) | 1.35 (–2.43, 3.04) | 7 | 3.96 (1.84, 8.55) | 1.99 (0.68, 2.84) | |
| Cerebral infarction | 73 | 7.55 (5.95, 9.59) | 2.92 (2.53, 3.20) | 4 | 4.62 (1.68, 12.68) | 2.21 (0.44, 3.29) | 5 | 5.93 (2.38, 14.79) | 2.57 (1.01, 3.55) | 152 | 4.89 (4.15, 5.78) | 2.29 (2.02, 2.49) | 26 | 8.30 (5.57, 12.36) | 3.05 (2.40, 3.51) | 259 | 5.91 (5.19, 6.73) | 2.56 (2.36, 2.71) | |
| Device occlusion | 0 | NA | NA | 0 | NA | NA | 0 | NA | NA | 5 | 6.92 (2.65, 18.08) | 2.79 (1.23, 3.78) | 0 | NA | NA | 5 | 5.99 (2.29, 15.66) | 2.58 (1.02, 3.57) | |
| Disseminated intravascular coagulation | 2 | 0.83 (0.21, 3.33) | −0.27 (–2.86, 1.12) | 0 | NA | NA | 0 | NA | NA | 15 | 1.76 (1.05, 2.93) | 0.81 (–0.06, 1.41) | 0 | NA | NA | 17 | 1.42 (0.88, 2.30) | 0.51 (–0.31, 1.08) | |
| Embolic cerebral infarction | 1 | 2.25 (0.31, 16.29) | 1.17 (–2.61, 2.86) | 0 | NA | NA | 0 | NA | NA | 8 | 8.09 (3.83, 17.11) | 3.02 (1.80, 3.82) | 0 | NA | NA | 9 | 7.42 (3.64, 15.15) | 2.89 (1.75, 3.65) | |
| Embolic stroke | 5 | 5.63 (2.31, 13.74) | 2.49 (0.93, 3.48) | 0 | NA | NA | 0 | NA | NA | 8 | 4.08 (2.00, 8.31) | 2.03 (0.82, 2.83) | 0 | NA | NA | 13 | 4.92 (2.79, 8.69) | 2.30 (1.36, 2.94) | |
| Embolism | 12 | 11.86 (6.59, 21.34) | 3.57 (2.59, 4.24) | 0 | NA | NA | 0 | NA | NA | 43 | 18.60 (13.22, 26.18) | 4.22 (3.71, 4.58) | 1 | 2.27 (0.32, 16.22) | 1.18 (–2.60, 2.87) | 56 | 18.20 (13.30, 24.92) | 4.19 (3.74, 4.50) | |
| Hemiplegia | 4 | 3.26 (1.21, 8.74) | 1.70 (–0.06, 2.78) | 0 | NA | NA | 0 | NA | NA | 3 | 1.02 (0.33, 3.18) | 0.03 (–2.04, 1.24) | 0 | NA | NA | 7 | 1.61 (0.76, 3.42) | 0.69 (–0.61, 1.54) | |
| Monoplegia | 4 | 4.77 (1.76, 12.89) | 2.25 (0.49, 3.33) | 0 | NA | NA | 0 | NA | NA | 0 | NA | NA | 0 | NA | NA | 4 | 1.74 (0.64, 4.70) | 0.80 (–0.97, 1.88) | |
| Paresis | 3 | 4.99 (1.57, 15.90) | 2.32 (0.25, 3.53) | 0 | NA | NA | 0 | NA | NA | 0 | NA | NA | 0 | NA | NA | 3 | 2.42 (0.76, 7.70) | 1.27 (–0.79, 2.48) | |
| Shunt occlusion | 17 | 11.05 (6.77, 18.03) | 3.47 (2.65, 4.03) | 0 | NA | NA | 0 | NA | NA | 306 | 74.79 (56.61, 98.79) | 6.22 (6.04, 6.36) | 3 | 4.29 (1.37, 13.41) | 2.10 (0.03, 3.31) | 326 | 58.34 (41.97, 81.10) | 5.87 (5.68, 6.00) | |
| Shunt thrombosis | 0 | NA | NA | 1 | 2.97 (0.40, 22.16) | 1.57 (–2.21, 3.26) | 0 | NA | NA | 27 | 35.03 (8.33, 147.34) | 5.13 (4.49, 5.58) | 0 | NA | NA | 28 | 31.53 (4.29, 231.81) | 4.98 (4.35, 5.43) | |
| Thrombosis | 17 | 6.61 (4.07, 10.72) | 2.72 (1.91, 3.29) | 0 | NA | NA | 1 | 2.50 (0.35, 18.07) | 1.32 (–2.46, 3.01) | 26 | 3.47 (2.34, 5.14) | 1.79 (1.14, 2.26) | 3 | 3.10 (0.99, 9.69) | 1.63 (–0.44, 2.84) | 47 | 4.48 (3.33, 6.02) | 2.16 (1.68, 2.51) | |
| Thrombosis in device | 1 | 1.83 (0.26, 13.12) | 0.87 (–2.91, 2.56) | 0 | NA | NA | 0 | NA | NA | 78 | 50.26 (32.96, 76.66) | 5.65 (5.28, 5.92) | 0 | NA | NA | 79 | 39.67 (25.91, 60.76) | 5.31 (4.94, 5.58) | |
| Thrombotic cerebral infarction | 1 | 1.87 (0.26, 13.42) | 0.90 (–2.88, 2.59) | 0 | NA | NA | 0 | NA | NA | 10 | 6.99 (3.64, 13.43) | 2.80 (1.73, 3.53) | 0 | NA | NA | 11 | 5.99 (3.20, 11.20) | 2.58 (1.56, 3.28) | |
Abbreviations: JADER, Japan Adverse Drug Event Report; HIF-PHIs, Hypoxia-Inducible Factor Prolyl Hydroxylase Inhibitors; NA, not applicable.
Cerebral infarction was reported to be most common (n = 73) with using daprodustat (ROR = 7.55, IC = 2.92). However, the strongest positive signal is embolism in daprodustat (ROR = 11.86, IC = 3.57). Cerebral infarction (ROR = 4.62, IC = 2.21) is the only positive signal both calculated by ROR and IC in enarodustat (ROR = 4.62, IC = 2.21) and molidustat (ROR = 5.93, IC = 2.57). Among vadadustat, cerebral infarction (ROR = 8.30, IC = 3.05) has the highest number of reports (n = 26).
3.3. Subgroup analysis
Additionally, given the potential confounding effect of baseline information on the results of the disproportionality analyses, sensitivity analyses for embolic and thrombotic events in HIF-PHIs incorporating sex, age, different indications, and underlying diseases were performed to bolster result confidence in Table 3 and Figure 4. The signal intensity is higher in female than in male (ROR 7.32 vs 6.17, p = 0.008). The signal intensity was higher in the non-dialysis population than in the dialysis population (ROR 6.85 vs 2.53, p < 0.001) (Table 3). Late elderly has a lower risk of embolic and thrombotic events than early elderly (ROR 4.11 vs 7.01, p < 0.001) as same as adult (ROR 4.11 vs 8.39, p < 0.001) (Figure 4a). Hyperlipidemia (ROR = 5.28, IC = 2.40) has the strongest signal than other underlying diseases. The signal intensity was higher in the diabetes mellitus population than in the heart failures population (ROR 5.15 vs 4.00, p = 0.046). The signal intensity was higher in the hypertension population than in the heart failures population (IC 2.31 vs 2.00, p = 0.031) (Figure 4b). This important assessment provides key insights that help doctors tailor treatments based on the specific characteristics of different patient groups.
Table 3.
Interaction analysis of subgroup in different sex, age, indications and underlying diseases for embolic and thrombotic events in HIF-PHIs.
| Subgroup | Overall HIF-PHIs | All other drugs | ROR (95% Crl) | p value for ratio of ROR | IC (95% Crl) | p value for ratio of IC |
|---|---|---|---|---|---|---|
| Sex | ||||||
| Male | 22.15% (3,431) | 3.28% (211,053) | 6.17 (5.67, 6.71) | 0.008 | 2.63 (2.51, 2.71) | 0.002 |
| Female | 22.72% (2,663) | 2.86% (231,479) | 7.32 (6.66, 8.04) | 2.87 (2.74, 2.97) | ||
| Age | ||||||
| Younger | 0.00% (1) | 1.36% (15,175) | NA | NA | −0.04 (-10.36, 1.94) | NA |
| Adult | 23.40% (577) | 2.70% (120,031) | 8.39 (6.90, 10.20) | 3.03 (2.74, 3.23) | ||
| Early elderly | 24.70% (2,567) | 3.21% (181,685) | 7.01 (6.38, 7.69) | 2.81 (2.68, 2.90) | ||
| Late elderly | 20.52% (2,826) | 4.34% (68,473) | 4.11 (3.72, 4.53) | 2.04 (1.90, 2.14) | ||
| Indication | ||||||
| Dialysis | 30.41% (1,003) | 4.82% (2,574) | 2.53 (2.02, 3.17) | < 0.001 | 1.34 (1.15, 1.48) | < 0.001 |
| Non-dialysis | 25.37% (4,257) | 3.31% (239,369) | 6.85 (6.37, 7.37) | 2.78 (2.68, 2.85) | ||
| Underlying diseases | ||||||
| Hypertension | 27.61% (2,050) | 4.57% (45,664) | 4.95 (4.45, 5.50) | NA | 2.31 (2.17, 2.41) | NA |
| Diabetes mellitus | 26.01% (1,238) | 4.16% (29,987) | 5.15 (4.48, 5.91) | 2.36 (2.18, 2.50) | ||
| Hyperlipidemia | 28.80% (191) | 4.58% (6,747) | 5.28 (3.78, 7.36) | 2.40 (1.95, 2.72) | ||
| Heart failures | 20.85% (729) | 4.05% (10,343) | 4.00 (3.27, 4.91) | 2.00 (1.73, 2.20) | ||
| Hyperuricaemia | 24.57% (635) | 3.36% (6,217) | 4.56 (3.63, 5.72) | 2.19 (1.92, 2.38) | ||
Crl, credibility interval; IC, information components; NA, not applicable; ROR, reporting odds ratios.
Figure 4.
Interaction analysis of age and underlying diseases for ROR and IC. (a) the p value of the interaction between different age groups, the ROR p value (bottom left), and the IC value p value (top right). (b) the p value of the interaction between different underlying diseases, the ROR p value (bottom left), and the IC value p value (top right). NA, not applicable.
3.4. Time to onset analysis
To determine the timing of HIF-PHIs-related embolic and thrombotic events in more detail, we analyzed the Time to onset (TTO) by subgroups in Figure 5. The median onset time was significantly shorter in the dialysis population than in the non-dialysis population (days: 27 vs. 47, p = 0.016). Different age, sex, and typing may not influence the median onset time of HIF-PHIs-related embolic and thrombotic events (p > 0.05). Notably, the median onset time was highest under non-dialysis population at 47 days (IQR 16–112); the shortest median onset time was adult at 12 days (IQR, 33–111).
Figure 5.
Cumulative distribution curves demonstrate the onset time of HIF-PHIs-related embolic and thrombotic events in different subgroups. The influence of typing (a), sex (b), age group (c), and indication (d) on the cumulative incidence of embolic and thrombotic after HIF-PHIs use.
4. Discussion
HIF-PHIs are currently a focus for clinic, despite worries about potential side effects. While clinical trial outcomes are generally favorable, the small number of individuals in these trials limits the capacity to extrapolate the safety of HIF-PHIs. As a result, real-world research on HIF-PHIs is critical. Our study included five HIF-PHIs that have been officially launched internationally. To our knowledge, this is the largest real-world pharmacovigilance study of embolic and thrombotic events associated with HIF-PHIs from JADER database. Our research results showed an increasing trend in HIF-PHIs-related embolic and thrombotic events reports since the drug’s market launch, reaching its peak in 2021, followed by a gradual decline. This pattern may be partially explained by the Weber effect, a well-documented phenomenon in pharmacovigilance where the introduction of new therapeutics is often associated with an initial surge in adverse event reporting due to increased awareness and vigilance among healthcare professionals and the public [16,17].
Previous studies have explored the mechanisms of HIF-PHIs and found that they may impact blood rheological properties and coagulation by regulating hypoxia-related genes. This insight provides a theoretical basis for further research into the relationship between HIF-PHIs and thrombosis [18]. Current clinical data suggest that HIF-PHIs may increase the risk of thrombotic events in some patients. However, a systematic risk assessment model has not been established due to the limited sample sizes and populations in the available clinical trials [19]. In this study, we analyze real-world data to better understand the underlying trends and patterns. Deep vein thrombosis was the strongest positive signal in venous thrombotic events and shunt occlusion was the strongest positive signal in mixed thrombotic events. In addition, this study revealed two key findings: first, it identified arterial thrombotic events such as myocardial infarction and cerebral infarction listed in the HIF-PHIs’ instructions; second, it found that peripheral artery occlusion emerged as the strongest positive signal among these events, although it was not mentioned in the instructions.
Although recent studies affirm the safety of these drugs [20], it is vital to recognize the thrombotic events associated with each and avoid them in patients with similar preexisting conditions. In addition to the above-mentioned overall HIF-PHIs, our study perform disproportionality analyses on daprodustat, enarodustat, molidustat, roxadustat, and vadadustat individually. The signal of arterial thrombosis events was strongest in molidustat (ROR = 5.44, IC = 2.44), followed by roxadustat (ROR = 4.52, IC = 2.18). The signal of venous thrombotic events (ROR = 6.17, IC = 2.63) and mixed thrombotic events (ROR =8.79, IC =3.14) in roxadustat was the strongest. Based on the data above, healthcare providers should evaluate the benefits and risks of HIF-PHIs for patients with conditions such as peripheral artery thrombosis or deep vein thrombosis. For those at high risk of thrombotic events, alternative therapies should be considered.
Studies indicate that roxadustat does not influence platelet production or activation [21]. The global pivotal study of roxadustat found that, while the rates of major adverse cardiovascular events (such as myocardial infarction, stroke, and all-cause mortality) were similar in both the roxadustat and control groups, patients receiving roxadustat had higher rates of vascular access thrombosis and deep vein thrombosis [22]. The real-world data analysis revealed that the most significant adverse reactions associated with roxadustat related to embolic and thrombotic events were shunt occlusion and device-related thrombosis. Consequently, the impact of roxadustat on embolic and thrombotic events continues to raise clinical concerns. Clinicians carefully observe patients receiving roxadustat for any indications of embolic or thrombotic events.
Current literature does not provide adequate correlation analysis between patient characteristics and thrombotic events. This includes factors such as underlying diseases and age groups. The ongoing concerns regarding the long-term use of HIF-PHIs, including the potential for increased cardiovascular risks and other complications, have led to calls for more comprehensive safety assessments in diverse populations [23,24]. According to the characteristics of patients with HIF-PHIs-related embolic and thrombotic events in JADER, male presented a larger proportion of embolic and thrombotic events than female (52.05 vs. 41.00%). However, after further disproportionality analysis, we found that female had a higher signal intensity than the male (ROR 7.32 vs 6.17, p = 0.008). Although direct evidence linking HIF-PHIs to sex-specific thrombosis remains limited, emerging data from chronic kidney disease populations, a key demographic for HIF-PHIs, provide valuable insights. For example, a case-control study of 683 diabetic patients with renal impairment found that females exhibited significantly higher rates of proteinuria and elevated systolic blood pressure compared to males [25]. These sex-specific risk factors are clinically significant because proteinuria promotes endothelial dysfunction through oxidative stress, while hypertension exacerbates shear-induced platelet activation. Clinicians pay close attention to female patients, as they may be at a higher risk of thrombotic events when treated with HIF-PHIs.
While our study found a lower intensity of adverse thrombotic reactions in adults aged 80 and above compared to those aged 60-79, this result may reflect statistical anomalies or unmeasured confounding factors. Moreover, individuals aged 60 to 79 also show lower intensity signals compared to adults aged 20 to 59. This finding contradicts the conventional theory that the risk of thrombosis increases with age [26]. Several factors may explain this observation. First, older patients may have more comorbidities, and thrombotic events might be underreported or attributed to other conditions. Second, older patients may receive lower doses of HIF-PHIs or be more likely to use antithrombotic medications, which could reduce the risk of thrombotic events. Third, age-related changes in coagulation systems or vascular pathology might alter the mechanisms of thrombosis. Finally, the JADER database may not fully capture thrombotic events in older patients due to underreporting or limited healthcare access. Further studies are needed to explore the relationship between age and thrombotic risk in patients treated with HIF-PHIs. Given that the elderly population are more likely to have other risk factors, and the old age itself is a risk factor for venous thrombosis [27], physicians should be more cautious when prescribing.
Hyperlipidemia (ROR = 5.28, IC = 2.40) has the strongest signal than other underlying diseases. Research has identified oxidized low-density lipoprotein (oxLDL) as a significant factor in the development of atherosclerosis and hyperlipidemia. It plays a crucial role in promoting dysfunction in endothelial cells, triggering inflammatory responses, and creating a procoagulant environment. Furthermore, the study revealed that a high-fat diet leads to the exposure of phosphatidylserine in the body, a process that is closely associated with heightened pro-coagulant activity, thereby increasing the risk of thrombosis [28]. Both the diabetic and hypertensive populations exhibit higher signal intensity compared to those with heart failure. Hypertension or diabetes is closely related to endothelial dysfunction. Endothelial cells produce important compounds such as nitric oxide and prostacyclin, which are released in response to different stimuli, including physical forces, hormones, and substances released from platelets. These compounds are essential for promoting vascular relaxation and inhibiting platelet activity, thereby contributing to the overall maintenance of vascular health and function [29]. Hypertension and diabetes can promote endothelial dysfunction, characterized by a shift in the endothelium’s functions that results in reduced vasodilation, an increased inflammatory response, and a greater tendency for thrombosis [30].
Japanese guidelines for HIF-PHIs are approved for both dialysis and non-dialysis patients. However, the FDA has not approved daprodustat for non-dialysis patients, and enarodustat’s guidelines in China are not approved for dialysis patients. The signal intensity was higher in the non-dialysis population than in the dialysis population (ROR 6.85 vs 2.53, p < 0.001) by the study. Even though there is research, the safety of enarodustat has been evaluated in non-dialysis CKD patients, with findings indicating a favorable safety profile but also raising awareness about possible side effects that warrant further investigation [31]. Clinicians should carefully assess a patient’s dialysis status when prescribing HIF-PHIs, as patients who are not on dialysis may be at an increased risk of thrombotic events.
Our research performed an analysis on the time to onset of embolic and thrombotic events. The median onset time was significantly shorter in the dialysis population than in the non- dialysis population (Days: 27 vs. 47, p = 0.016). The shortest median onset time was adult at 12 days (IQR, 33 – 111). Nevertheless, owing to the scarcity of records on certain types of HIF-PHIs as well as the key variable, therapy start date, this conclusion requires more clinical evidence. During the initial weeks of HIF-PHI therapy, clinicians should closely monitor patients, especially those who are dialysis-dependent, as thrombotic events may develop sooner in this group.
However, our study may have certain limitations also. First, the JADER database is a spontaneous reporting system, which comes with inherent limitations, including incomplete information, underreporting, and the inability to accurately calculate the incidence of adverse events. Furthermore, because the data is collected spontaneously, there is a risk of reporting biases. The database primarily includes data from Japan, and as such, findings may not be fully applicable to other regions due to differences in healthcare systems, genetic factors, drug prescribing patterns, and reporting practices. For example, certain adverse events might be more prevalent in specific populations due to genetic variations or regional differences in medication use. Therefore, it is important to acknowledge that our findings may not fully represent the global safety profile of HIF-PHIs. To address these limitations, future research should incorporate data from multiple sources, including international pharmacovigilance databases such as VigiBase and the FAERS database. This approach would enable cross-regional comparisons and offer a more comprehensive understanding of the safety profile of HIF-PHIs on a global scale. Secondly, spontaneous reporting systems are primarily designed to identify potential signals and associations, rather than establish causal relationships. As such, this exploratory retrospective analysis based on observational data cannot definitively establish a causal link between the drug and AEs. We emphasize the need for future randomized controlled trials or prospective cohort studies to confirm these causal connections, as these designs would provide more robust evidence. Thirdly, the JADER database contains limited information on drug doses, which prevented us from analyzing the potential relationship between specific doses of HIF-PHI drugs and thrombotic events. Fourth, we were unable to collect or analyze information on antithrombotic therapy, which may affect the interpretation of thrombotic events associated with HIF-PHI drugs. Fifth, the study lacked detailed information on arteriovenous fistulas and specific hemodialysis treatment modalities, which could potentially influence the occurrence of thrombotic events in patients with CKD. Sixth, the limited number of thrombotic events reported for enarodustat and molidustat represents a limitation of our study. This is likely due to the more recent market introduction of these drugs compared to other HIF-PHIs, which has resulted in a shorter post-marketing surveillance period and fewer reported cases. To better assess the safety profiles of these newer agents, further studies with longer follow-up periods are necessary. Seventh, due to limited data on ERAs in the JADER database, we were unable to use ERAs as a control group. Lastly, it is impossible to rule out the interference of other drugs during the analysis. Despite these limitations, our study offers valuable insights into the embolic and thrombotic events associated with HIF-PHIs in the Japanese population, contributing to the growing body of evidence regarding its safety profile. And disproportionality analysis remains an important method for monitoring drug safety and identifying rare signals [32,33]. In summary, analyzing real-world data will address gaps by providing representative population characteristics and long-term efficacy data, thereby giving clinicians a stronger basis for risk assessment with HIF-PHIs.
5. Conclusion
We analyzed the real-world data on embolic and thrombotic events associated with HIF-PHIs using the JADER database. Our findings indicate a significant association between HIF-PHIs and increased risks of thrombotic events, with specific risks varying by drug, sex, and patient condition. The results underscore the importance of careful patient selection and monitoring, particularly for high-risk populations such as those with dialysis dependency or underlying vascular conditions. Further investigation is needed to refine the safety profile of HIF-PHIs, especially in diverse patient populations.
Supplementary Material
Acknowledgment
The technical support by Shu-xin Jiao, Guo-hao Cai, and Ming-dao Lin is greatly appreciated.
Funding Statement
This manuscript was supported by [Joint Program on Health Science & Technology Innovation of Hainan Province] under Grant [No. WSJK2025MS145]; [Project supported by the Education Department of Hainan Province] under Grant [No. Hnjg2025-93]; [the Innovational Fund for Scientific and Technological Personnel of Hainan Province] under Grant [No. KJRC2023D27]; [the Independent Scientific Research Project of Ordos Central Hospital] under Grant [No. EY2024QN07]; and [the Inner Mongolia Academy of Medical Sciences Public Hospital Scientific Research Joint Fund Project] under Grant [No. 2024GLLH1288].
Ethics statement
Written informed consent was not obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article because the data in the article uses de-identified data and does not involve patient privacy. Since the JADER database is accessible to the public and patient records are anonymized and de-identified, ethical clearance and informed consent are not required for this study.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
Publicly available datasets were analyzed in this study. This data can be found here: (https://www.pmda.go.jp/index.html).
References
- 1.Lamerato L, James G, van Haalen H, et al. Epidemiology and outcomes in patients with anemia of CKD not on dialysis from a large US healthcare system database: a retrospective observational study. BMC Nephrol. 2022;23(1):166. doi: 10.1186/s12882-022-02778-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Felaco P, Felaco M, Franceschelli S, et al. Erythropoietin induces miRNA-210 by JAK2/STAT5 signaling in PBMCs of end-stage renal disease patients. FEBS J. 2020;287(23):5167–5182. doi: 10.1111/febs.15302. [DOI] [PubMed] [Google Scholar]
- 3.Gupta N, Wish JB.. Hypoxia-inducible factor prolyl hydroxylase inhibitors: a potential new treatment for anemia in patients with CKD [published correction appears in Am J Kidney Dis. 2017;69(6):815–826. doi: 10..1053/j.ajkd.2017.04.001]. Am J Kidney Dis. 2017;69(6):815–826. doi: 10.1053/j.ajkd.2016.12.011. [DOI] [PubMed] [Google Scholar]
- 4.Mullard A. FDA approves first hypoxia-inducible factor prolyl hydroxylase inhibitor. Nat Rev Drug Discov. 2023;22(3):173–173. doi: 10.1038/d41573-023-00028-6. [DOI] [PubMed] [Google Scholar]
- 5.Wang Y, Yu X.. Stabilizing hypoxia-inducible factor to manage anemia in chronic kidney disease: from Basic Theory to Clinical Study. Kidney Dis (Basel). 2024;10(2):132–142. doi: 10.1159/000536039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zhao H, Li P, Zhang HL, et al. An updated meta-analysis on the efficacy and safety of hypoxia-inducible factor prolyl hydroxylase inhibitor treatment of anemia in nondialysis-dependent chronic kidney disease. Ren Fail. 2023;45(2):2258986. doi: 10.1080/0886022X.2023.2258986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Odawara M, Nishi H, Nangaku M.. A spotlight on using HIF-PH inhibitors in renal anemia. Expert Opin Pharmacother. 2024;25(10):1291–1299. doi: 10.1080/14656566.2024.2378903. [DOI] [PubMed] [Google Scholar]
- 8.Tanaka H, Tani A, Onoda T, et al. Hypoxia-inducible factor prolyl hydroxylase inhibitors and hypothyroidism: an analysis of the japanese pharmacovigilance database. In Vivo. 2024;38(2):917–922. doi: 10.21873/invivo.13519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zhou Y, Ren Q, Hu R, et al. Does HIF-PHI increased risk of gastrointestinal hemorrhage in patients with renal anemia: a review of cases reported to the U.S. Food and drug administration adverse event reporting system. Ren Fail. 2021;43(1):1170–1171. doi: 10.1080/0886022X.2021.1952881. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pharmaceuticals and Medical Devices Agency . [Japanese adverse drug event report database]. [accessed 2024 August 10]. Available from: https://www.pmda.go.jp/index.html
- 11.MedDRA . Introductory guide: medDRA version 26.0. 2023. [accessed 2024 August 16]. Available from: https://admin.meddra.org/sites/default/files/guidance/file/intguide_26_0_English.pdf
- 12.Introductory guide for standardised MedDRA queries (SMQs) Version 26.1. 2023. [accessed 2024 August 16]. Available from: https://admin.meddra.org/sites/default/files/guidance/file/SMQ_intguide_26_1_English.pdf
- 13.Norén GN, Hopstadius J, Bate A.. Shrinkage observed-to-expected ratios for robust and transparent large-scale pattern discovery. Stat Methods Med Res. 2013;22(1):57–69. doi: 10.1177/0962280211403604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.van Puijenbroek EP, Bate A, Leufkens HG, et al. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions. Pharmacoepidemiol Drug Saf. 2002;11(1):3–10. doi: 10.1002/pds.668. [DOI] [PubMed] [Google Scholar]
- 15.Altman DG, Bland JM.. Interaction revisited: the difference between two estimates. BMJ. 2003;326(7382):219–219. doi: 10.1136/bmj.326.7382.219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bate A, Evans SJ.. Quantitative signal detection using spontaneous ADR reporting. Pharmacoepidemiol Drug Saf. 2009;18(6):427–436. doi: 10.1002/pds.1742. [DOI] [PubMed] [Google Scholar]
- 17.Harpaz R, DuMouchel W, LePendu P, et al. Performance of pharmacovigilance signal-detection algorithms for the FDA adverse event reporting system. Clin Pharmacol Ther. 2013;93(6):539–546. doi: 10.1038/clpt.2013.24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Maruno S, Tanaka T, Nangaku M.. [Renal anemia and hypoxia-inducible factor prolyl hydroxylase inhibitor]. Rinsho Ketsueki. 2021;62(5):371–377. doi: 10.11406/rinketsu.62.371. [DOI] [PubMed] [Google Scholar]
- 19.Chen D, Niu Y, Liu F, et al. Safety of HIF prolyl hydroxylase inhibitors for anemia in dialysis patients: a systematic review and network meta-analysis. Front Pharmacol. 2023;14:1163908. doi: 10.3389/fphar.2023.1163908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ren S, Zhao Y, Wu J, et al. Hypoxia-inducible factor-prolyl hydroxylase inhibitors for treatment of anemia in chronic kidney disease: a systematic review and network meta-analysis. Front Pharmacol. 2024;15:1406588. doi: 10.3389/fphar.2024.1406588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Zhao JX, Xu YY, Xie JY, et al. Roxadustat does not affect platelet production, activation, and thrombosis formation. Arterioscler Thromb Vasc Biol. 2021;41(10):2523–2537. doi: 10.1161/ATVBAHA.121.316495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Provenzano R, Szczech L, Leong R, et al. Efficacy and cardiovascular safety of roxadustat for treatment of anemia in patients with non-dialysis-dependent CKD: pooled results of three randomized clinical trials. Clin J Am Soc Nephrol. 2021;16(8):1190–1200. doi: 10.2215/CJN.16191020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Nakanishi T, Kuragano T.. Growing concerns about using hypoxia-inducible factor prolyl hydroxylase inhibitors for the treatment of renal anemia. Clin Kidney J. 2024;17(3):sfae051. doi: 10.1093/ckj/sfae051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Mima A. Hypoxia-inducible factor-prolyl hydroxylase inhibitors for renal anemia in chronic kidney disease: advantages and disadvantages. Eur J Pharmacol. 2021;912:174583. doi: 10.1016/j.ejphar.2021.174583. [DOI] [PubMed] [Google Scholar]
- 25.Murkamilov IT, Aitbaev KA, Fomin VV.. [Prevalence, age and gender features of chronic kidney disease in patients with diabetes mellitus]. Ter Arkh. 2023;95(6):481–486. doi: 10.26442/00403660.2023.06.202242. [DOI] [PubMed] [Google Scholar]
- 26.Mahé I, Caulin C, Bergmann JF.. L’âge, un facteur indépendant de risque de thrombose. Données épidémiologiques [Age, an independent risk factor for thrombosis. Epidemiologic data]. Presse Med. 2005;34(12):878–886. doi: 10.1016/s0755-4982(05)84068-0. [DOI] [PubMed] [Google Scholar]
- 27.Gregson J, Kaptoge S, Bolton T, et al. Cardiovascular Risk Factors Associated With Venous Thromboembolism. JAMA Cardiol. 2019;4(2):163–173. doi: 10.1001/jamacardio.2018.4537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Yan M, Wang Z, An Y, et al. OxLDL enhances procoagulant activity of endothelial cells by TMEM16F-mediated phosphatidylserine exposure. Cell Biol Int. 2024;48(6):848–860. doi: 10.1002/cbin.12150. [DOI] [PubMed] [Google Scholar]
- 29.Rajendran P, Rengarajan T, Thangavel J, et al. The vascular endothelium and human diseases. Int J Biol Sci. 2013;9(10):1057–1069 doi: 10.7150/ijbs.7502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Gimbrone MA, Jr, García-Cardeña G.. Endothelial cell dysfunction and the pathobiology of atherosclerosis. Circ Res. 2016;118(4):620–636. doi: 10.1161/CIRCRESAHA.115.306301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Mima A. Enarodustat treatment for renal anemia in patients with non-dialysis chronic kidney disease. In Vivo. 2023;37(2):825–829. doi: 10.21873/invivo.13148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Li GF, Yu G.. Drug-induced liver injury with ritonavir-boosted nirmatrelvir: evidence from coronavirus disease 2019 emergency use authorization adverse event reporting system. Gastroenterology. 2023;165(1):305–306. doi: 10.1053/j.gastro.2023.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Gao G, Yu G, Li GF.. Control compounds for drug-induced liver injury evaluation: evidence from database of adverse event notifications. J Hepatol. 2025;82(4):e206–e208. doi: 10.1016/j.jhep.2024.12.002. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Publicly available datasets were analyzed in this study. This data can be found here: (https://www.pmda.go.jp/index.html).