A real-world disproportionality analysis of FDA adverse event reporting system (FAERS) events for avatrombopag Authors: Zhong Xue, Maohua Chen, Mingzuo Wang, Fan Zhang, Zhaoshuo Chen Published: 2024 ABSTRACT Avatrombopag is a next-generation thrombopoietin receptor agonist approved for the treatment of thrombocytopenia in patients with chronic liver disease scheduled to undergo surgery and thrombocytopenia in patients with chronic immune thrombocytopenia. However, realistic data on its post-marketing long-term safety and tolerability in large sample populations are incomplete. The adverse event (AE) reports of avatrombopag were analyzed based on the FAERS database. By extracting large-scale data from the FAERS database, this study used various signal quantification techniques such as reporting odds ratios, proportional reporting ratios, Bayesian confidence propagation neural network, and multi-item gamma Poisson shrinker to calculate and evaluate the ratio and association between avatrombopag and specific AEs. In the FAERS database, data from the second quarter of 2018 to the fourth quarter of 2023 were extracted for this study, a total of 1217 avatrombopag-related AEs reports were included for analysis. Based on four calculation methods, this study identified and 32 preferred terms associated with avatrombopag. Common AEs in the product label such as thrombosis, headache, contusion, petechiae, and gingival bleeding were detected. AEs not mentioned in the product label, such as photopsia and ear discomfort were also detected. The first 30 days after initiation of medication were the most common period for both serious and non-serious AEs. This study demonstrates the common AEs and the potential AEs associated with avatrombopag in real-world practice, which could provide valuable cautionary evidence for clinicians and pharmacists to manage safety issues with avatrombopag.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-024-80067-0. Introduction Thrombocytopenia is defined as a platelet count below 150 × 109/L1, which is common in patients with chronic liver disease (CLD) and immune disorder, with two to five persons per 100,000 and with 15–90%, respectively2–4. It usually increases in severity as the degree of disease worsens, and increases the risk of excessive bleeding particularly after surgery, which poses a huge economic burden and even death5. Platelet transfusion is one of the primary treatment methods for thrombocytopenia patients and severe thrombocytopenia who require invasive procedures, but its usage is limited by associated side effects6. The drugs with thrombopoietin (TPO) activity are offering a convenient and non-invasive approach to increasing platelet levels7. Avatrombopag is an oral, second-generation thrombopoietin receptor agonist (TPO-RA) that was approved by the FDA in 2018 for the treatment of thrombocytopenia in patients with chronic liver disease scheduled to undergo surgery and thrombocytopenia in patients with chronic immune thrombocytopenia8. It binds specifically to the transmembrane domain of the TPO receptor on humans and induces the proliferation and differentiation of megakaryocytes and progenitor cells. It also works synergistically with endogenous TPO to enhance the effect of increasing platelet counts9. Previous randomized controlled trials showed that avatrombopag was superior to placebo in reducing the need for platelet transfusions or any hemostatic rescue treatment10. Common adverse events (AEs) associated with avatrombopag include abdominal pain (2.90–10.30%), diarrhea (3.5–4.5%), nausea (4.5–8.6%), pyrexia (7.0–15.7%), dizziness (3.0%), and headache (2.9–10.3%). Notably, in patients with grade three or four AEs, headache (6.3%), vomiting (6.3%) and platelet count decreased (3.1%) have been observed. AEs can lead to drug discontinuation, affecting the disease treatment. Furthermore, some serious AEs of avatrombopag can cause life-threatening risks or even death2,3. Given the widespread clinical use of avatrombopag11, a comprehensive real-world study of its AEs is warranted. The FDA Adverse Event Reporting System (FAERS) is a database that houses numerous reports related to AEs, providing a rich data source for the study of AEs12. This study aims to comprehensively analyze the association of avatrombopag with AEs by conducting a deep analysis of FAERS data, which could offer comprehensive information for its clinical decision-making. Results Basic information on AE reports This study extracted data from the FAERS database from the second quarter of 2018 to the fourth quarter of 2023, with a total of 10,118,235 AE reports. After data cleaning and drug screening, this study included a total of 1217 AE reports related to avatrombopag for analysis. The clinical characteristics of avatrombopag-associated AEs were described in Table 1. Overall AE reports for which data were available, a higher proportion of females was recorded, accounting for 58.42%. Although 55.79% of the reports did not mention the specific age of the patients. However, in the section where age was reported, the largest group of patients was reported in the 18–65 age group with a percentage of 51.67%. Weight was reported in 15.37% of the reports, with the highest number of patients weighing < 80 kg at 39.04%. The vast majority of AE reports (88.22%) came from the United States, followed by Spain (2.14%), Italy (1.81%), and other country (7.83%). Immune thrombocytopenia was the most frequently reported indication (56.73%), followed by thrombocytopenia (25.44%), immune thrombocytopenic purpura (2.73%) and others (15.11%). Co-medication was reported in 25.55% of cases, the most frequent being prednisone (12.22%), others being spironolactone (6.43%), eltrombopag (6.11%), romiplostim (5.79%), and omeprazole (5.14%) in that order. Serious AEs were reported in 404 (33.20%) AEs, including 76 (18.81%) deaths, 10 (2.48%) life-threatening events, 198 (49.01%) hospitalisations, 3 (0.74%) disabilities, and 273 (67.57%) other serious outcomes. In terms of reporting sources, healthcare professionals reported more AEs at 57.94%. Consumers reported 42.06% of AEs. In terms of the year of reporting, excluding Q2–Q4 of 2018, the most reported year was 2023 (23.17%), followed by 2022 (20.38%), 2021 (21.36%), 2020 (21.53%), 2019 (9.04%). As shown in Fig. 1, the first 30 days after initiation of medication were the most common period for both serious and non-serious AEs, accounting for 71.89% and 84.07%, respectively. The time to onset of serious AEs was different from that of non-serious AEs [20.5 (6–59d) vs. 1 (0–9d), P = 0.03 (Student t-test, two tailed)]. Table 1 Basic information on AE reports for avatrombopag in the FAERS database (April 2018 to December 2023).Characteristics Case number, nCase proportion, %Number of events 1217Gender Female 65958.42% Male 46941.58% Unknown 897.31%Age (years) < 1861.12% 18 ≤ and ≤ 6527851.67% > 6525447.21% Unknown 67955.79% Median (IQR) 64.5 (52–73)-Weight (Kg) < 807339.04% 80 ≤ and ≤ 1006635.29% > 1004825.67% Unknown 103084.63% Median (IQR) 82.26 (69.5-100.84)-Reported countries United States 107188.22% Spain 262.14% Italy 221.81% Other country 957.83% Unknown 30.25%Indications Immune thrombocytopenia 58256.73% Thrombocytopenia 26125.44% Immune thrombocytopenic purpura282.73% Others 15515.11% Unknown 19115.69%Combination drugs Prednisone 3812.22% Spironolactone 206.43% Eltrombopag 196.11% Romiplostim 185.79% Omeprazole 165.14% Unknown 90674.45%Outcomes Non-serious Outcome81366.80% Serious Outcome40433.20% Death7618.81% Life-threatening102.48% Hospitalization19849.01% Disability 30.74% Other serious outcomes27367.57%Time-to-onset (days) 21817.91% Median (IQR)6 (1–40)Reporters Health professional70457.94% Consumer51142.06% Unknown20.16%Reporting year 202328223.17% 202224820.38% 202126021.36% 202026221.53% 20191109.04% 2018Q2-Q4554.52%AEs, adverse event; FAERS, FDA adverse event reporting system; IQR, interquartile range; n, number; Q2, second quarter; Q4, fourth quarter. Table 1 Basic information on AE reports for avatrombopag in the FAERS database (April 2018 to December 2023).Characteristics Case number, nCase proportion, %Number of events 1217Gender Female 65958.42% Male 46941.58% Unknown 897.31%Age (years) < 1861.12% 18 ≤ and ≤ 6527851.67% > 6525447.21% Unknown 67955.79% Median (IQR) 64.5 (52–73)-Weight (Kg) < 807339.04% 80 ≤ and ≤ 1006635.29% > 1004825.67% Unknown 103084.63% Median (IQR) 82.26 (69.5-100.84)-Reported countries United States 107188.22% Spain 262.14% Italy 221.81% Other country 957.83% Unknown 30.25%Indications Immune thrombocytopenia 58256.73% Thrombocytopenia 26125.44% Immune thrombocytopenic purpura282.73% Others 15515.11% Unknown 19115.69%Combination drugs Prednisone 3812.22% Spironolactone 206.43% Eltrombopag 196.11% Romiplostim 185.79% Omeprazole 165.14% Unknown 90674.45%Outcomes Non-serious Outcome81366.80% Serious Outcome40433.20% Death7618.81% Life-threatening102.48% Hospitalization19849.01% Disability 30.74% Other serious outcomes27367.57%Time-to-onset (days) 21817.91% Median (IQR)6 (1–40)Reporters Health professional70457.94% Consumer51142.06% Unknown20.16%Reporting year 202328223.17% 202224820.38% 202126021.36% 202026221.53% 20191109.04% 2018Q2-Q4554.52%AEs, adverse event; FAERS, FDA adverse event reporting system; IQR, interquartile range; n, number; Q2, second quarter; Q4, fourth quarter. Basic information on AE reports for avatrombopag in the FAERS database (April 2018 to December 2023). AEs, adverse event; FAERS, FDA adverse event reporting system; IQR, interquartile range; n, number; Q2, second quarter; Q4, fourth quarter. Fig. 1Time to onset of avatrombopag-related AEs. AEs, adverse event; IQR, interquartile range. Time to onset of avatrombopag-related AEs. AEs, adverse event; IQR, interquartile range. Results of risk signal mining The occurrence of AEs related avatrombopag were distributed across 26 system organ class (SOC), as detailed in Table 2. Using the most stringent EBGM method to rank the AE signals, the five most prominent categories were investigations [number (n) = 360, reporting odds ratios (ROR) 3.10, proportional reporting ratios (PRR) 2.78, information component (IC) 1.47, empirical Bayesian geometric mean (EBGM) 2.78], hepatobiliary disorders (n = 40, ROR 1.76, PRR 1.75, IC 0.8, EBGM 1.75), social circumstances (n = 22, ROR 1.56, PRR 1.55, IC 0.63, EBGM 1. 55), nervous system disorders (n = 239, ROR 1.44, PRR 1.39, IC 0.48, EBGM 1.39) and vascular disorders (n = 72, ROR 1.28, PRR 1.28, IC 0.35, EBGM 1.28). Table 2 Signal strength of AE reports of avatrombopag at the SOC level in FAERS database.SOCNumberROR (95% CI) PRR (χ2) IC (IC025) EBGM (EBGM05) General disorders and administration site conditions4921.23 (1.12–1.36)a1.19 (17.26)0.25 (0.22)a 1.19 (1.07)Investigations3603.10 (2.77–3.46)a2.78 (433.24)a 1.47 (1.32)a 2.78 (2.48)a Injury, poisoning and procedural complications2991.02 (0.90–1.15)1.02 (0.08)0.02 (0.02)a 1.02 (0.90)Nervous system disorders2391.44 (1.26–1.64)a1.39 (28.39)0.48 (0.42)a 1.39 (1.22)Gastrointestinal disorders1871.15 (0.99–1.34)1.14 (3.44)0.19 (0.16)a 1.14 (0.98)Musculoskeletal and connective tissue disorders1091.05 (0.87–1.28)1.05 (0.29)0.07 (0.06)a 1.05 (0.87)Respiratory, thoracic and mediastinal disorders940.93 (0.76–1.14)0.93 (0.49)-0.10 (-0.12)0.93 (0.76)Infections and infestations840.61 (0.49–0.76)0.63 (19.62)-0.67 (-0.83)0.63 (0.51)Skin and subcutaneous tissue disorders820.63 (0.50–0.78)0.64 (17.63)-0.65 (-0.80)0.64 (0.51)Vascular disorders721.28 (1.02–1.62)a1.28 (4.38)0.35 (0.28)a1.28 (1.01)Psychiatric disorders500.43 (0.32–0.57)0.44 (37.64)-1.19 (-1.57)0.44 (0.33)Blood and lymphatic system disorders430.95 (0.71–1.29)0.96 (0.09)-0.07 (-0.09)0.96 (0.71)Hepatobiliary disorders401.76 (1.29–2.40)a1.75 (12.86)0.80 (0.59)a1.75 (1.28)Cardiac disorders 310.58 (0.41–0.83)0.59 (9.24)-0.77 (-1.10)0.59 (0.41)Eye disorders 300.69 (0.48–0.99)0.70 (4.06)-0.52 (-0.75)0.70 (0.49)Metabolism and nutrition disorders280.51 (0.35–0.74)0.51 (13.24)-0.96 (-1.40)0.51 (0.35)Immune system disorders 250.64 (0.43–0.95)0.65 (4.89)-0.63 (-0.93)0.65 (0.44)Renal and urinary disorders240.47 (0.32–0.71)0.48 (13.88)-1.06 (-1.59)0.48 (0.32)Social circumstances 221.56 (1.02–2.37)a1.55 (4.34)0.63 (0.42)a1.55 (1.02)Surgical and medical procedures170.40 (0.25–0.64)0.40 (15.37)-1.31 (-2.12)0.40 (0.25)Product issues 140.27 (0.16–0.46)0.28 (26.82)-1.85 (-3.12)0.28 (0.16)Neoplasms benign, malignant and unspecified (incl cysts and polyps)140.14 (0.08–0.23)0.14 (76.69)-2.83 (-4.78)0.14 (0.08)Reproductive system and breast disorders110.63 (0.35–1.15)0.64 (2.32)-0.65 (-1.18)0.64 (0.35)Ear and labyrinth disorders90.73 (0.38–1.40)0.73 (0.90)-0.45 (-0.87)0.73 (0.38)Congenital, familial and genetic disorders20.30 (0.08–1.21)0.30 (3.20)-1.72 (-6.87)0.30 (0.08)Pregnancy, puerperium and perinatal conditions10.10 (0.01–0.73)0.10 (7.82)-3.27 (-23.26)0.10 (0.01)a Indicates statistically significant signals in algorithm.AE, adverse event; CI, confidence interval; SOC, System Organ Class; ROR, reporting odds ratio; PRR, proportional reporting ratio; χ2, chi-squared; IC, information component; IC025, the lower limit of the 95 two-sided CI of the IC; EBGM, empirical Bayesian geometric mean; EBGM05, the lower 95 two-sided CI of EBGM. Table 2 Signal strength of AE reports of avatrombopag at the SOC level in FAERS database.SOCNumberROR (95% CI) PRR (χ2) IC (IC025) EBGM (EBGM05) General disorders and administration site conditions4921.23 (1.12–1.36)a1.19 (17.26)0.25 (0.22)a 1.19 (1.07)Investigations3603.10 (2.77–3.46)a2.78 (433.24)a 1.47 (1.32)a 2.78 (2.48)a Injury, poisoning and procedural complications2991.02 (0.90–1.15)1.02 (0.08)0.02 (0.02)a 1.02 (0.90)Nervous system disorders2391.44 (1.26–1.64)a1.39 (28.39)0.48 (0.42)a 1.39 (1.22)Gastrointestinal disorders1871.15 (0.99–1.34)1.14 (3.44)0.19 (0.16)a 1.14 (0.98)Musculoskeletal and connective tissue disorders1091.05 (0.87–1.28)1.05 (0.29)0.07 (0.06)a 1.05 (0.87)Respiratory, thoracic and mediastinal disorders940.93 (0.76–1.14)0.93 (0.49)-0.10 (-0.12)0.93 (0.76)Infections and infestations840.61 (0.49–0.76)0.63 (19.62)-0.67 (-0.83)0.63 (0.51)Skin and subcutaneous tissue disorders820.63 (0.50–0.78)0.64 (17.63)-0.65 (-0.80)0.64 (0.51)Vascular disorders721.28 (1.02–1.62)a1.28 (4.38)0.35 (0.28)a1.28 (1.01)Psychiatric disorders500.43 (0.32–0.57)0.44 (37.64)-1.19 (-1.57)0.44 (0.33)Blood and lymphatic system disorders430.95 (0.71–1.29)0.96 (0.09)-0.07 (-0.09)0.96 (0.71)Hepatobiliary disorders401.76 (1.29–2.40)a1.75 (12.86)0.80 (0.59)a1.75 (1.28)Cardiac disorders 310.58 (0.41–0.83)0.59 (9.24)-0.77 (-1.10)0.59 (0.41)Eye disorders 300.69 (0.48–0.99)0.70 (4.06)-0.52 (-0.75)0.70 (0.49)Metabolism and nutrition disorders280.51 (0.35–0.74)0.51 (13.24)-0.96 (-1.40)0.51 (0.35)Immune system disorders 250.64 (0.43–0.95)0.65 (4.89)-0.63 (-0.93)0.65 (0.44)Renal and urinary disorders240.47 (0.32–0.71)0.48 (13.88)-1.06 (-1.59)0.48 (0.32)Social circumstances 221.56 (1.02–2.37)a1.55 (4.34)0.63 (0.42)a1.55 (1.02)Surgical and medical procedures170.40 (0.25–0.64)0.40 (15.37)-1.31 (-2.12)0.40 (0.25)Product issues 140.27 (0.16–0.46)0.28 (26.82)-1.85 (-3.12)0.28 (0.16)Neoplasms benign, malignant and unspecified (incl cysts and polyps)140.14 (0.08–0.23)0.14 (76.69)-2.83 (-4.78)0.14 (0.08)Reproductive system and breast disorders110.63 (0.35–1.15)0.64 (2.32)-0.65 (-1.18)0.64 (0.35)Ear and labyrinth disorders90.73 (0.38–1.40)0.73 (0.90)-0.45 (-0.87)0.73 (0.38)Congenital, familial and genetic disorders20.30 (0.08–1.21)0.30 (3.20)-1.72 (-6.87)0.30 (0.08)Pregnancy, puerperium and perinatal conditions10.10 (0.01–0.73)0.10 (7.82)-3.27 (-23.26)0.10 (0.01)a Indicates statistically significant signals in algorithm.AE, adverse event; CI, confidence interval; SOC, System Organ Class; ROR, reporting odds ratio; PRR, proportional reporting ratio; χ2, chi-squared; IC, information component; IC025, the lower limit of the 95 two-sided CI of the IC; EBGM, empirical Bayesian geometric mean; EBGM05, the lower 95 two-sided CI of EBGM. Signal strength of AE reports of avatrombopag at the SOC level in FAERS database. a Indicates statistically significant signals in algorithm. AE, adverse event; CI, confidence interval; SOC, System Organ Class; ROR, reporting odds ratio; PRR, proportional reporting ratio; χ2, chi-squared; IC, information component; IC025, the lower limit of the 95 two-sided CI of the IC; EBGM, empirical Bayesian geometric mean; EBGM05, the lower 95 two-sided CI of EBGM. This study identified a total of 32 significantly preferred terms (PTs) that simultaneously met the criteria of all four algorithms (Table 3). In our study, platelet count abnormalities (including platelet count abnormal, platelet count increased, platelet disorder, thrombocytosis, and platelet count decreased) reported the highest number of cases, and also the higher signal values. Some common AEs were found in our study, including thrombosis, headache, contusion, petechiae, and gingival bleeding. In addition, some previously unreported AEs were found in our results, such as liver function abnormalities, photopsia, ear discomfort, and others. Table 3 Signal strength of AE reports of avatrombopag at the PT level in FAERS database.PTNumberROR (95%CI)PRR (χ2) IC (IC025)EBGM (EBGM05)Platelet count decreased 16432.52 (27.78–38.07)30.82 (4723.10)4.94 (4.22)30.71 (26.23)Headache1324.74 (3.98–5.64)4.58 (372.37)2.19 (1.84)4.58 (3.84)Platelet count increased96148.29 (120.81-182.02)143.65 (13372.90)7.14 (5.82)141.25 (115.07)Platelet count abnormal50152.41 (114.97-202.05)149.92 (7267.62)7.20 (5.43)147.31 (111.12)Contusion214.84 (3.15–7.43)4.81 (63.41)2.27 (1.47)4.81 (3.13)Pulmonary embolism205.96 (3.84–9.25)5.92 (81.87)2.57 (1.65)5.92 (3.81)Thrombosis174.54 (2.82–7.31)4.52 (46.62)2.18 (1.35)4.52 (2.80)Adverse drug reaction173.45 (2.14–5.56)3.44 (29.42)1.78 (1.11)3.44 (2.13)Deep vein thrombosis167.94 (4.85–12.97)7.90 (96.38)2.98 (1.82)7.89 (4.83)Rhinorrhoea123.60 (2.04–6.35)3.59 (22.44)1.84 (1.05)3.59 (2.04)Full blood count abnormal115.81 (3.21–10.51)5.79 (43.64)2.53 (1.40)5.79 (3.20)Portal vein thrombosis1076.92 (41.23-143.52)76.67 (740.14)6.25 (3.35)75.99 (40.73)Thrombocytosis948.39 (25.11–93.27)48.25 (414.10)5.58 (2.90)47.98 (24.89)Drug effect less than expected912.36 (6.42–23.79)12.32 (93.54)3.62 (1.88)12.31 (6.39)Cerebral haemorrhage 95.92 (3.08–11.40)5.91 (36.70)2.56 (1.33)5.91 (3.07)Laboratory test abnormal84.84 (2.42–9.69)4.83 (24.28)2.27 (1.13)4.83 (2.41)Platelet disorder749.46 (23.51-104.06)49.35 (329.64)5.62 (2.67)49.06 (23.32)Petechiae 715.16 (7.22–31.85)15.13 (92.20)3.92 (1.86)15.10 (7.19)Embolism617.81 (7.99–39.72)17.78 (94.84)4.15 (1.86)17.75 (7.96)Immune thrombocytopenia612.59 (5.65–28.06)12.57 (63.78)3.65 (1.64)12.55 (5.63)Gingival bleeding610.91 (4.90-24.32)10.89 (53.86)3.44 (1.55)10.88 (4.88)Graft versus host disease516.02 (6.66–38.54)15.99 (70.15)4.00 (1.66)15.96 (6.63)Liver function test abnormal56.59 (2.74–15.86)6.58 (23.67)2.72 (1.13)6.58 (2.74)Renal vein thrombosis4141.29 (52.56–379.80)141.11 (547.26)7.12 (2.65)138.79 (51.63)Ammonia increased419.43 (7.28–51.85)19.40 (69.65)4.27 (1.60)19.36 (7.25)Cerebral venous sinus thrombosis342.16 (13.55-131.17)42.12 (119.83)5.39 (1.73)41.91 (13.47)Antiphospholipid syndrome339.52 (12.70-122.92)39.48 (111.99)5.30 (1.70)39.30 (12.63)Photopsia 312.20 (3.93–37.87)12.19 (30.76)3.61 (1.16)12.17 (3.92)Autoimmune hepatitis38.72 (2.81–27.06)8.71 (20.45)3.12 (1.01)8.70 (2.80)General physical condition abnormal37.59 (2.44–23.55)7.58 (17.12)2.92 (0.94)7.57 (2.44)Hepatic encephalopathy37.57 (2.44–23.50)7.57 (17.08)2.92 (0.94)7.56 (2.44)Ear discomfort36.35 (2.05–19.72)6.35 (13.51)2.67 (0.86)6.34 (2.04)AE, adverse event; CI, confidence interval; SOC, System Organ Class; ROR, reporting odds ratio; PRR, proportional reporting ratio; PTs, preferred terms; χ2, chi-squared; IC, information component; IC025, the lower limit of the 95 two-sided CI of the IC; EBGM, empirical Bayesian geometric mean; EBGM05, the lower 95 two-sided CI of EBGM. Table 3 Signal strength of AE reports of avatrombopag at the PT level in FAERS database.PTNumberROR (95%CI)PRR (χ2) IC (IC025)EBGM (EBGM05)Platelet count decreased 16432.52 (27.78–38.07)30.82 (4723.10)4.94 (4.22)30.71 (26.23)Headache1324.74 (3.98–5.64)4.58 (372.37)2.19 (1.84)4.58 (3.84)Platelet count increased96148.29 (120.81-182.02)143.65 (13372.90)7.14 (5.82)141.25 (115.07)Platelet count abnormal50152.41 (114.97-202.05)149.92 (7267.62)7.20 (5.43)147.31 (111.12)Contusion214.84 (3.15–7.43)4.81 (63.41)2.27 (1.47)4.81 (3.13)Pulmonary embolism205.96 (3.84–9.25)5.92 (81.87)2.57 (1.65)5.92 (3.81)Thrombosis174.54 (2.82–7.31)4.52 (46.62)2.18 (1.35)4.52 (2.80)Adverse drug reaction173.45 (2.14–5.56)3.44 (29.42)1.78 (1.11)3.44 (2.13)Deep vein thrombosis167.94 (4.85–12.97)7.90 (96.38)2.98 (1.82)7.89 (4.83)Rhinorrhoea123.60 (2.04–6.35)3.59 (22.44)1.84 (1.05)3.59 (2.04)Full blood count abnormal115.81 (3.21–10.51)5.79 (43.64)2.53 (1.40)5.79 (3.20)Portal vein thrombosis1076.92 (41.23-143.52)76.67 (740.14)6.25 (3.35)75.99 (40.73)Thrombocytosis948.39 (25.11–93.27)48.25 (414.10)5.58 (2.90)47.98 (24.89)Drug effect less than expected912.36 (6.42–23.79)12.32 (93.54)3.62 (1.88)12.31 (6.39)Cerebral haemorrhage 95.92 (3.08–11.40)5.91 (36.70)2.56 (1.33)5.91 (3.07)Laboratory test abnormal84.84 (2.42–9.69)4.83 (24.28)2.27 (1.13)4.83 (2.41)Platelet disorder749.46 (23.51-104.06)49.35 (329.64)5.62 (2.67)49.06 (23.32)Petechiae 715.16 (7.22–31.85)15.13 (92.20)3.92 (1.86)15.10 (7.19)Embolism617.81 (7.99–39.72)17.78 (94.84)4.15 (1.86)17.75 (7.96)Immune thrombocytopenia612.59 (5.65–28.06)12.57 (63.78)3.65 (1.64)12.55 (5.63)Gingival bleeding610.91 (4.90-24.32)10.89 (53.86)3.44 (1.55)10.88 (4.88)Graft versus host disease516.02 (6.66–38.54)15.99 (70.15)4.00 (1.66)15.96 (6.63)Liver function test abnormal56.59 (2.74–15.86)6.58 (23.67)2.72 (1.13)6.58 (2.74)Renal vein thrombosis4141.29 (52.56–379.80)141.11 (547.26)7.12 (2.65)138.79 (51.63)Ammonia increased419.43 (7.28–51.85)19.40 (69.65)4.27 (1.60)19.36 (7.25)Cerebral venous sinus thrombosis342.16 (13.55-131.17)42.12 (119.83)5.39 (1.73)41.91 (13.47)Antiphospholipid syndrome339.52 (12.70-122.92)39.48 (111.99)5.30 (1.70)39.30 (12.63)Photopsia 312.20 (3.93–37.87)12.19 (30.76)3.61 (1.16)12.17 (3.92)Autoimmune hepatitis38.72 (2.81–27.06)8.71 (20.45)3.12 (1.01)8.70 (2.80)General physical condition abnormal37.59 (2.44–23.55)7.58 (17.12)2.92 (0.94)7.57 (2.44)Hepatic encephalopathy37.57 (2.44–23.50)7.57 (17.08)2.92 (0.94)7.56 (2.44)Ear discomfort36.35 (2.05–19.72)6.35 (13.51)2.67 (0.86)6.34 (2.04)AE, adverse event; CI, confidence interval; SOC, System Organ Class; ROR, reporting odds ratio; PRR, proportional reporting ratio; PTs, preferred terms; χ2, chi-squared; IC, information component; IC025, the lower limit of the 95 two-sided CI of the IC; EBGM, empirical Bayesian geometric mean; EBGM05, the lower 95 two-sided CI of EBGM. Signal strength of AE reports of avatrombopag at the PT level in FAERS database. AE, adverse event; CI, confidence interval; SOC, System Organ Class; ROR, reporting odds ratio; PRR, proportional reporting ratio; PTs, preferred terms; χ2, chi-squared; IC, information component; IC025, the lower limit of the 95 two-sided CI of the IC; EBGM, empirical Bayesian geometric mean; EBGM05, the lower 95 two-sided CI of EBGM. Discussions In this study, we employed a pharmacovigilance methodology based on the FAERS database to probe the complex association between avatrombopag and AEs, assess their post-marketing safety profile. Common AEs were found in our study, including thrombosis, headache, contusion, petechiae, and gingival bleeding, which is consistent with previous studies10,13. Notably, some previously unreported AEs were found in our results, such as liver function abnormalities, photopsia, ear discomfort, and others. The association liver function abnormalities and avatrombopag has not been clear yet, while eltrombopag (another TPO-RA drug) has been noted with hepatotoxicity14. This is also may caused by liver disease of patients, which needs to be explored in more and more rigorous studies in the future. It is recommend that regular monitoring of liver function levels during treatment. There are also some PTs may be associated with disease, such as platelet count decreased, and the resulting petechiae, gingival bleeding and contusion. Furthermore, photopsia and ear discomfort was also reported in this study, but the exact effects and the mechanisms of the potential associations have not been fully explored and require further clinical studies. In addition, our study showed that PTs such as platelet count abnormal (n = 50) and platelet count increased (n = 96) ranked in the top 2 AE signal intensities. It has been noted that in previous studies that platelet count may increase excessively with avatrombopag overdose, and it can result in thrombotic or thromboembolic complications15,16. In previous Phase 3 study, thrombocytopenia patients treated with avatrombopag were observed thromboembolic event (9.40%)13. Thrombotic events were also found in our study, including renal vein thrombosis (n = 4), portal vein thrombosis (n = 10), cerebral venous sinus thrombosis (n = 3), embolism (n = 6), deep vein thrombosis (n = 16), pulmonary embolism (n = 20), thrombosis (n = 17). Thrombotic events, when they occur, can cause serious consequences. Hence, our findings supported that enhance platelet testing during the use of avatrombopag17, assessing platelet counts weekly until a stable platelet count greater than or equal to 50 × 109/L, and then obtain platelet counts monthly thereafter. In addition, it is necessary that obtain platelet counts weekly for at least 4 weeks after discontinuation of avatrombopag18. Monitor platelet counts closely would help intervene and treat thromboembolic events timely. This study found that a large number of AEs occurred within 30 days of drug administration and the median time to occurrence of non-serious AEs was 1 day, which is consistent with previous study1. It suggests that healthcare professionals should be alert to the occurrence of adverse events as soon as the drug is administered. Differences in the time to onset to serious AE versus non-serious AE may be related to continued medication use as well as the result of superimposed disease progression, with continued medication treatment, vigilance should be exercised for the occurrence of serious AEs. Four algorithms, including ROR, PRR, bayesian confidence propagation neural network (BCPNN) and multi-item gamma poisson shrinker (MGPS), were used in our study. If the event count for a specific drug AE pair was three or higher, the PRR was two or higher, and the corresponding chi-square value was four or higher, then the pair was considered a signal. The ROR is determined by comparing an event report for a particular medicine to all other event reports and then comparing those report ratios to all other drug reports19. The PRR measures how much more adverse drug-related events are reported disproportionately when compared to the same events reported for all or a few other medicines16. It was defined if the lower limit of the 95% two-sided confidence interval (CI) of ROR exceeded 1. The logarithm of the ratio between the observed and predicted rates of drug-AE combination reporting is used to calculate the IC17. The AE signal indicates when the lower bound of the 95% two-sided confidence interval of the IC (IC025) value exceeds 017. The EBGM values are adjusted estimates of relative reporting ratios, observed reporting rate/expected reporting rate, after Bayesian multi-item gamma Poisson shrinker20. Signals were identified if EBGM and its lower limit of the 95% confidence interval (EBGM05) values were 2 or more. A drug associated with a higher risk of a specific adverse event, compared to other drugs, typically exhibits increased disproportionality, resulting in a higher score. To ensure the stability of the detected signals, the AE was considered a statistically significant signal when all the disproportionality measures (ROR, PRR, IC, and EBGM) had crossed the significance threshold. The formulas and thresholds of the four algorithms are shown in Table S1. Due to the limitations of the data of the spontaneous reporting system and the different conditions and principles of limiting signal generation in various algorithms, the number of signals generated and the false positive results are different21. For instance, ROR can estimate relative risk to reduce bias; PRR has high sensitivity but lower specificity, making it prone to detect false positive signals; MGPS is the only method that uses stratified analysis, which is less affected by confounding factors such as demographic data and allows for associations between drug use characteristics and AEs. When there are few reports, EBGM is more stable than PRR. IC has lower quality requirements for AE reporting data but can detect strong correlations in AEs22. In practice, pharmacovigilance systems often use a combination of these methods to increase the robustness of signal detection. The above AE reveals possible risks of avatrombopag for thrombocytopenia. However, this study has several limitations. Firstly, the reporting of AEs recorded in the FAERS database is inherently spontaneous and is not limited to healthcare professionals but also to consumers; false, underreported, inaccurate, incomplete, and late reports are difficult to address and inevitably result in reporting bias. In addition, only cases presenting with AE were included in the FAERS database, making it impossible to calculate the incidence of AE associated with avatrombopa. Notably, although this study identified new potential AEs associated with avatrombopa use, disproportionate analyses provide only an estimate of signal strength, which is only statistically significant and neither quantifies risk nor establishes causality. They may be influenced by a variety of factors, including baseline patient characteristics, comorbidities, and concomitant medications. Further studies are therefore needed to establish a causal relationship between these adverse events and avatrombopag. However, despite some limitations of the FAERS database in pharmacovigilance studies, there is still a role for it when large amounts of data are involved23. The comprehensive characterisation of avatrombopag AE signals and the discovery of some unexpected AE signals may provide a basis for further clinical studies of avatrombopag and guide us to use it more safely and effectively. Conclusions Utilizing the FAERS database, a total of 32 significantly PTs were identified after conducting a disproportionate analysis of post-marking AEs associated with avatrombopa. Platelet count abnormalities, thrombosis, headache, contusion, petechiae, and gingival bleeding should be of high concern in the study. Unexpected new adverse events such as photopsia and ear pain may also occur. The first 30 days after initiation of medication were the most common period for AEs. Overall, the results of this study may provide guidance for drug monitoring and safety assessment in clinical practice, and provide basis for drug regulatory authorities and healthcare providers to formulate appropriate management strategies. Materials and methods Study design and data source This study conducted a retrospective pharmacovigilance review using the FAERS database to investigate the association between avatrombopag and potential AEs. FAERS is an important US Food and Drug Administration database that collects information on AEs reported by patients and healthcare professionals during drug use. For this study, we extracted data from the FAERS database from the second quarter of 2018 to the fourth quarter of 2023 to obtain the most recent information on drug AEs. To ensure the specificity and accuracy of the research, this study restricted the search to ‘avatrombopag’ and ‘doptelet’ as the primary suspect drugs. Data were collected, pre-processed and cleaned using software tools (SAS) for MySQL to ensure accuracy and completeness24. Following the FDA recommended method for removing duplicate reports, we select the PRIMARYID, CASEID and FDA_DT fields from the DEMO table. We sort by CASEID, FDA_DT and then PRIMARYID. For reports with the same CASEID, we keep the one with the largest FDA_DT value. Second, for reports where both CASEID and FDA_DT are the same, we retain the one with the largest PRIMARYID value. Since the second quarter of 2018, each quarterly data package includes a list of reports that have been removed. After data deduplication, we remove reports based on the CASEID listed in the list of deleted reports. The details can be found in Fig. 2. Fig. 2The flow chart of the data analysis. The flow chart of the data analysis. The data was then mapped to the RxNorm and MedDRA concept libraries. In addition, we performed standardisation and translation of preferred terms (PTs) to ensure consistency of the data. In order to present the data, this study merged AE reports with the same PTs. At the same time, the PTs were categorised and organised using the SOC method to better summarise and analyse the characteristics of AEs. Statistical analysis In this study, various statistical analysis techniques were employed to evaluate the relationship between avatrombopag and potential AEs. The calculation methods are shown in Tables S1 and S2, starting with a descriptive statistical analysis of the AE data related to avatrombopag. Then, using disproportionality methods including ROR25, PRR26, BCPNN27, and MGPS algorithms28, possible associations and anomalies were detected, aiding in determining whether the AEs related to avatrombopag have statistical significance. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1