A Real-World Pharmacovigilance Study of Ceftazidime/avibactam: Data Mining of the FDA Adverse Event Reporting System (FAERS) Database

doi:10.21203/rs.3.rs-3802796/v1

Download PDF

Research Article

A Real-World Pharmacovigilance Study of Ceftazidime/avibactam: Data Mining of the FDA Adverse Event Reporting System (FAERS) Database

https://doi.org/10.21203/rs.3.rs-3802796/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Introduction

Ceftazidime/avibactam (CAZ/AVI) is a combination of a well-known third-generation, broad-spectrum cephalosporin with a new beta-lactamase inhibitor that has been approved for the treatment of various infectious diseases (especially MDR-GNB infections) by the FDA.

Aim

The present study extensively assessed real-world CAZ/AVI-related adverse events (AEs) through data mining of the FDA Adverse Event Reporting System (FAERS) database to better understand toxicities.

Methods

The signals of CAZ/AVI-related AEs were quantified using disproportionality analyses, including the reporting odds ratio (ROR), the proportional reporting ratio (PRR), the Bayesian confidence propagation neural network (BCPNN) and the multi-item gamma Poisson shrinker (MGPS) algorithms. System organ classifications (SOCs) and preferred terms (PTs) from the Medical Dictionary for Regulatory Activities (MedDRA) were used in the definition.

Results

A total of 628 instances of CAZ/AVI-related AEs were identified among 10,114,815 records gathered from the FAERS database. A total of 61 PTs with significant disproportionality that simultaneously met the criteria of all four algorithms were retained. Several unexpected safety signals may also occur, including melaena, hypernatraemia, depressed level of consciousness, brain oedema, petechiae, delirium, and shock haemorrhagic. The median onset time for AEs associated with CAZ/AVI was 4 days, with nearly half cases occurring within 3 days after CAZ/AVI initiation.

Conclusions

Some of our research findings were consistent with the information described in drug labels and monographs, and we also discovered potential novel and unexpected AE signals associated with CAZ/AVI. Future clinical investigations are needed to validate our findings and establish their relationship. Our findings might serve as important supporting data for future CAZ/AVI safety investigations.

ceftazidime/avibactam

MDR-GNB

pharmacovigilance

data mining

FAERS

Our study, using the FDA Adverse Event Reporting System (FAERS), revealed 628 instances of Ceftazidime/avibactam (CAZ/AVI)-related adverse events, with 61 significant disproportionality signals. Notably, unexpected safety signals, such as melaena, hypernatraemia, and brain oedema, were identified. These findings highlight the importance of continuous monitoring for novel adverse events associated with CAZ/AVI in real-world settings.

Our analysis showed that the median onset time for CAZ/AVI-related adverse events was 4 days, with nearly half occurring within 3 days of CAZ/AVI initiation. This information is crucial for healthcare practitioners to be vigilant during the early stages of treatment. Recognizing the rapid onset of adverse events can facilitate timely intervention and improve patient safety in the clinical use of CAZ/AVI.

Our study not only confirmed known adverse events associated with CAZ/AVI but also revealed potential novel signals. The identification of unexpected adverse events emphasizes the need for further clinical investigations to validate our findings and establish causal relationships. Our research provides valuable insights that can guide future safety investigations of CAZ/AVI, serving as a foundation for evidence-based decision-making in clinical practice.

A significant global public health concern is the increasing prevalence of Gram-negative bacteria (GNB) that are multidrug resistant (MDR; resistant to at least one agent in three or more drug classes). These infections are more likely to result in significant morbidity and mortality, prolonged hospital stays, and higher costs than infections due to non-MDR organisms[1, 2]. Several processes might lead to antimicrobial resistance in GNBs, but one of the most prevalent ones is the production of beta-lactamases[3]. Effective antimicrobial therapy must be initiated promptly for infections caused by this group of bacteria.

The increasing prevalence of extended-spectrum beta-lactamase (ESBL)-producing pathogens has driven increased use and reliance on carbapenems[4]. Thus, the growth and spread of carbapenem-resistant pathogens are of special concern and have brought attention to the urgent need for novel antimicrobial medicines[5, 6]. Ceftazidime/avibactam (CAZ/AVI), an antibiotic on the market with the broadest registered indications and activity against MDR-GNB, was given FDA approval to be marketed in the US in February 2015[7]. This antibiotic combines a well-known third-generation, extended-spectrum cephalosporin and a novel beta-lactamase inhibitor that retains efficacy against MDR-GNB-producing class A (ESBLs and KPCs [K. pneumoniae carbapenemases]), class C (cephalosporinases; AmpCs), and some class D (such as OXA-48-oxacillinases) β-lactamases[3, 8]. CAZ/AVI is recommended for the treatment of complicated urinary tract infections (cUTIs, including pyelonephritis), complicated intra-abdominal infections (cIAIs), complicated hospital-acquired pneumonia (HAP), complicated ventilator-associated pneumonia (VAP), and other infections caused by aerobic GNBs in adults and children who are at least 3 months old and have limited treatment options[9].

In accordance with the product description and early assessments of postmarketing safety, liver transaminase elevations and gastrointestinal symptoms were the most frequent adverse drug reactions (ADRs) associated with CAZ/AVI[10]. Most of the recent safety studies on CAZ/AVI have only been reported in clinical trials and a few meta-analyses due to the drug's relatively recent availability[10–12]. Numerous uncommon adverse events (AEs), including hepatobiliary, renal, hematological, and neurological side effects, have been reported as a result of the increased use of CAZ/AVI[13]. However, studies on CAZ/AVI-related AE signals based on global databases and real-world data are rare[14].

With its features of a broad monitoring range and earlier detection of potential ADR signals, the spontaneous reporting system has emerged as the primary information source for researching postmarketing drug safety[15, 16]. The Food and Drug Administration Adverse Event Reporting System (FAERS), which covers tens of millions of case reports of AEs submitted by physicians, pharmacists, manufacturers, and others, is the largest and most comprehensive publicly accessible postmarketing safety surveillance database in the world. It was created to support the FDA's postmarketing surveillance for all approved drugs and therapeutic biologic products[17]. In the present study, AEs reported with CAZ/AVI from the second quarter of 2015 to the second quarter of 2022 were retrospectively investigated using data mining of FAERS, and all CAZ/AVI-related reports were quantitatively evaluated by signal detection utilizing four disproportionality analytical methods to provide a reference for clinical monitoring and risk identification.

Aim

The present study extensively assessed real-world CAZ/AVI-related adverse events (AEs) through data mining of the FDA Adverse Event Reporting System (FAERS) database to better understand toxicities.

Ethics approval

Ethics approval was not required for a study of this nature.

Study Design and Data Sources

The FAERS, the largest and most comprehensive publicly accessible postmarketing safety surveillance database in the world, served as the source of the data. Each quarter, FAERS files are released by the FDA. The database comprises seven datasets: patient demographic and administrative information (file descriptor DEMO), drug and biologic information (DRUG), adverse events (REAC), patient outcomes (OUTC), report sources (RPSR), start and end dates of drug therapy (THER), and indications for use/diagnosis (INDI). A relational database was created to connect these seven datasets with the specific identification numbers found in each FAERS report. This real-world, retrospective pharmacovigilance study used the FAERS database to conduct a disproportionality analysis. We gathered data from reports received between the second quarter of 2015 (when the FDA approved CAZ/AVI) and the second quarter of 2022 (when the FAERS database had undergone its most recent upgrade at the time the study was conducted)[15, 17].

Data Cleaning

Both brand names and generic names were used to identify data related to the target medicine CAZ/AVI because FAERS has two variables related to drug names (DRUGNAME and PROD_AI) and permits the submission of arbitrary drug names. In this study, the search terms used were ceftazidime/avibactam (or other generic names with nonstandard formats), Avycaz (the brand name in the U.S.), and Zavicefta (the brand name in non-U.S. regions). The preferred term (PT) from the Medical Dictionary for Regulatory Activities (MedDRA) is used to code AEs in the FAERS. The highest level of MedDRA is the system organ class (SOC), and each SOC can be categorized into more general PT subcategories[18, 19]. Based on MedDRA SOC and PT levels that had more than three entries in the FAERS, we extracted all reports including CAZ/AVI-related AEs in order to describe the spectrum of toxicities. To eliminate duplicate reports submitted by different people and institutions, we chose the most recent FDA_DT when the PRIMARYIDs were the same and the higher PRIMARYID where the FDA_DT and the CASEID were the same, in accordance with FDA instructions. The role code for AEs had been assigned by reporters, including primary suspect drug, secondary suspect drug, concomitant, and interacting. Death (DE), life-threatening (LT), hospitalization-initial or prolonged (HO), disability (DS), congenital anomaly (CA), or other important medical event (OT) were all considered serious patient outcomes[20]. Clinical information of CAZ/AVI-related AEs was gathered, including age, gender, reporter, reporting area, reporting time and outcomes. We also evaluated the time to onset (the interval between the beginning of drug use and the date an AE first occurred) of AEs brought on by CAZ/AVI by removing those with inaccurate or incomplete records. Figure 1 depicts the study's flow diagram.

Statistical Analysis

Demographics and characteristics of all AE reports related to CAZ/AVI were displayed using descriptive analysis, including gender, age, indications, serious outcomes, reported countries, reporting persons, and reporting years. Disproportionality analysis, which is frequently used in pharmacovigilance studies, was employed to identify potential signals for AEs reported regarding body changes associated with CAZ/AVI. As shown in Table 1, four major specific indices that were calculated using common formulas were employed to assess potential relationships between CAZ/AVI and AEs, including the reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN), and multi-item gamma Poisson shrinker (MGPS), which are widely used and currently employed by the Medicines and Healthcare products Regulatory Agency (MHRA) in the UK, the Netherlands Pharmacovigilance Centre, the WHO, and the FDA, respectively. Several studies have compared data mining algorithms; different algorithms have slightly different properties, and none are universally superior to the others[21–26]. In our investigation, we considered only signals with at least three target AE reports. Any of the four algorithms meeting the criteria should be considered a positive signal of drug-associated AEs. For this investigation, we selected AE signals that simultaneously met all four algorithm criteria. Using MySQL 8.0 and Excel 2016, all the statistical analyses and data processing were carried out. We identified “new signals” by referring to pharmaceutical information from regulatory agencies, including the FDA and the European Medicines Agency (EMA)[9, 27].

Table 1

Four major algorithms used to assess potential associations between CAZ/AVI and AEs.
Algorithms	Equation	Criteria
Reporting odds ratio (ROR)	ROR = ad/b/c 95%CI = e^{ln(ROR)±1.96(1/a+1/b+1/c+1/d)^0.5}	lower limit of 95% CI > 1, N ≥ 3
Proportional reporting ratio (PRR)	PRR = a(c + d)/c/(a + b) χ2=[(ad-bc)^2](a + b + c + d)/[(a + b)(c + d)(a + c)(b + d)]	PRR ≥ 2, χ2 ≥ 4, N ≥ 3
Bayesian confidence propagation neural network (BCPNN)	IC = log2a(a + b + c + d)/((a + c)(a + b)) 95%CI = E(IC) ± 2V(IC)^0.5	IC025 > 0
Multi-item gamma Poisson shrinker (MGPS)	EBGM = a(a + b + c + d)/(a + c)/(a + b) 95%CI = e^{ln(EBGM)±1.96(1/a+1/b+1/c+1/d)^0.5}	EBGM05 > 2

Notes: (a) number of reports containing both the target drug and target AE; (b) number of reports containing other AEs of the target drug; (c) number of reports containing the target AE of other drugs; (d) number of reports containing other drugs and other AEs.

Abbreviations: 95% CI, 95% confidence interval; N, the number of reports; χ2, chi-squared; IC, information component; IC025, the lower limit of 95% CI of the IC; E(IC), the IC expectations; V(IC), the variance of IC; EBGM, empirical Bayesian geometric mean; EBGM05, the lower limit of 95% CI of EBGM.

General Characteristics

After excluding duplicates, a total of 10,114,815 AE reports were submitted to the FAERS database between April 2015 and June 2022. Among these, 628 cases involved CAZ/AVI as the primary suspect drug, and Table 2 presents the demographics and characteristics of these 628 reports. From 2015 to 2022, the number of reported AEs essentially exhibited a gradually increasing trend. Males made up a larger proportion of all AEs (38.06% vs. 28.03% females). Patients aged between 18 and 65 years composed the largest percentage of reports (13.85%), and senior individuals (those older than 65 years) also represented a significant portion (6.53%). Klebsiella infection was the most frequently reported indication (7.64%), followed by pneumonia (7.32%), infection (4.78%), Pseudomonas infection (4.62%), and Pneumonia Klebsiella (3.50%). Physicians (58.76%), pharmacists (17.36%), and other health professionals (7.17%) made up the majority of the sources of reports. The country that reported the most cases was America (18.31%), followed by France (12.10%), Italy (9.08%), China (8.28%) and Britain (5.89%). Death (38.54%) was the most frequently reported severe outcome, followed by other serious (important medical events) and hospitalization–initial or prolonged events, occurring in 195 (31.05%) and 102 (16.24%) cases, respectively. The high rate of death incidents may be more correlated with the progression of diseases.

Table 2

Demographics and characteristics of AE reports with CAZ/AVI from the FAERS database (April 2015 - June 2022)
Characteristics	Case Number, n	Case Proportion, %
Number of events	628
Gender
Female	176	28.03%
Male	239	38.06%
Unknown	213	33.92%
Age (years)
< 18	6	0.96%
18 ≤ and ≤ 65	87	13.85%
> 65	41	6.53%
Unknown	494	78.66%
Indications (top five)
Klebsiella infection	48	7.64%
Pneumonia	46	7.32%
Infection	30	4.78%
Pseudomonas infection	29	4.62%
Pneumonia Klebsiella	22	3.50%
Serious outcomes
Death (DE)	242	38.54%
Life-threatening (LT)	17	2.71%
Hospitalization–initial or prolonged (HO)	102	16.24%
Disability (DS)	3	0.48%
Congenital anomaly (CA)	1	0.16%
Other serious (important medical event) (OT)	195	31.05%
Reported countries (top five)
America(US)	115	18.31%
France(FR)	76	12.10%
Italy(IT)	57	9.08%
China(CN)	52	8.28%
Britain(GB)	37	5.89%
Reported persons
Health profession
Physician (MD)	369	58.76%
Pharmacist (PH)	109	17.36%
Other health-professional (OT)	45	7.17%
Non-healthcare professional
Consumer (CN)	100	15.92%
Unknown	5	0.80%
Reporting years
2022Q1-2	56	8.92%
2021	129	20.54%
2020	121	19.27%
2019	130	20.70%
2018	76	12.10%
2017	73	11.62%
2016	30	4.78%
2015Q2-4	13	2.07%

Notes: 2022Q1-2, the first two quarters of 2022.

2015Q2-4, the last three quarters of 2015.

Signal Detected at the SOC Level

Table 3 displays signal reports for CAZ/AVI at the SOC level. The significant SOCs were “infections and infestations (SOC: 10021881)”, “investigations (SOC: 10022891)”, “renal and urinary disorders (SOC:10038359)”, “hepatobiliary disorders (SOC: 10019805)”, and “congenital, familial and genetic disorders (SOC: 10010331)”.

Table 3

Signal strength of AEs of CAZ/AVI at the system organ class (SOC) level in the FAERS database
System Organ Class (SOC)	CAZ/AVI Cases Reporting SOC	ROR (95% Two-Sided CI)	PRR (95% Two-Sided CI)	χ2	IC (IC025)	EBGM (EBGM05)
General disorders and administration site conditions	348	2.18 (1.93–2.46)	1.89 (1.73–2.07)	167.50	0.92 (0.73)	1.89 (1.67)
Injury, poisoning and procedural complications	192	2.34 (2.01–2.72)	2.16 (1.89–2.46)	127.13	1.11 (0.87)	2.16 (1.85)
Infections and infestations	191	5.41 (4.64–6.30)	4.82 (4.22–5.50)	594.25	2.27 (2.01)	4.82 (4.14)
Nervous system disorders	124	1.81 (1.51–2.18)	1.74 (1.47–2.06)	41.08	0.80 (0.52)	1.74 (1.45)
Investigations	119	2.58 (2.14–3.12)	2.45 (2.06–2.91)	105.86	1.29 (1.00)	2.45 (2.03)
Renal and urinary disorders	59	3.20 (2.47–4.15)	3.11 (2.42–3.99)	85.59	1.64 (1.21)	3.11 (2.40)
Hepatobiliary disorders	54	8.22 (6.26–10.79)	7.95 (6.12–10.32)	329.31	2.99 (2.42)	7.94 (6.05)
Skin and subcutaneous tissue disorders	54	1.81 (1.38–2.38)	1.78 (1.37–2.31)	18.90	0.83 (0.42)	1.78 (1.36)
Gastrointestinal disorders	48	0.61 (0.45–0.81)	0.62 (0.47–0.82)	11.91	-0.69 (-1.1)	0.62 (0.46)
Respiratory, thoracic and mediastinal disorders	45	1.38 (1.02–1.85)	1.36 (1.02–1.82)	4.47	0.45 (0.01)	1.36 (1.01)
Blood and lymphatic system disorders	42	2.67 (1.96–3.63)	2.62 (1.95–3.53)	42.60	1.39 (0.89)	2.62 (1.93)
Cardiac disorders	30	2.17 (1.51–3.12)	2.15 (1.51–3.06)	18.55	1.10 (0.53)	2.15 (1.50)
Psychiatric disorders	27	1.40 (0.96–2.05)	1.40 (0.96–2.03)	3.08	0.48 (-0.08)	1.40 (0.95)
Metabolism and nutrition disorders	22	2.80 (1.84–4.26)	2.77 (1.83–4.19)	25.02	1.47 (0.76)	2.77 (1.82)
Vascular disorders	20	1.12 (0.72–1.74)	1.11 (0.72–1.72)	0.24	0.16 (-0.48)	1.11 (0.72)
Immune system disorders	10	0.76 (0.41–1.41)	0.76 (0.41–1.41)	0.78	-0.40 (-1.23)	0.76 (0.41)
Musculoskeletal and connective tissue disorders	9	0.56 (0.29–1.07)	0.56 (0.29–1.07)	3.18	-0.84 (-1.68)	0.56 (0.29)
Eye disorders	8	1.11 (0.55–2.22)	1.11 (0.56–2.21)	0.09	0.15 (-0.82)	1.11 (0.55)
Neoplasms benign, malignant and unspecified (incl cysts and polyps)	7	1.72 (0.82–3.61)	1.72 (0.82–3.59)	2.09	0.78 (-0.35)	1.72 (0.82)
Ear and labyrinth disorders	5	1.12 (0.46–2.69)	1.12 (0.47–2.68)	0.06	0.16 (-1.03)	1.12 (0.46)
Product issues	5	1.22 (0.51–2.93)	1.22 (0.51–2.92)	0.19	0.28 (-0.93)	1.22 (0.51)
Congenital, familial and genetic disorders	4	12.95 (4.85–34.58)	12.92 (4.85–34.39)	43.98	3.69 (0.66)	12.91 (4.84)
Pregnancy, puerperium and perinatal conditions	2	1.79 (0.45–7.15)	1.79 (0.45–7.13)	0.69	0.84 (-1.14)	1.79 (0.45)
Surgical and medical procedures	1	0.46 (0.07–3.30)	0.47 (0.07–3.30)	0.62	-1.10 (-2.66)	0.47 (0.07)

Note: Bolded words indicate statistically significant signals in the four algorithms.

Abbreviations: ROR, reporting odds ratio; CI, confidence interval; PRR, proportional reporting ratio; χ2, chi-squared; IC, information component; EBGM, empirical Bayesian geometric mean.

Signal Detected at the PT Level

Table 4 lists all 61 significant PTs of interest that simultaneously complied with all four methods. In this study, Candida infection (PT: 10074170), acute kidney injury (PT: 10069339), blood creatinine increased (10005483), blood alkaline phosphatase increased (10059570), eosinophilia (PT: 10014950), platelet count decreased (10035528), toxic epidermal necrolysis (PT: 10044223), epilepsy (PT: 10015037), encephalopathy (PT: 10014625), status epilepticus (PT: 10041962), myoclonus (PT: 10028622), and haemolytic anaemia (PT: 10018916) were detected in data mining, in line with the drug's directions and warnings[9]. It should be noted that patients receiving CAZ/AVI have reported unanticipated AEs, including melaena (PT: 10027141), hypernatraemia (PT: 10020679), depressed level of consciousness (10012373), brain oedema (10048962), petechiae (10034754), delirium (10012218), shock haemorrhagic (10049771), and so on, were uncovered in the label for CAZ/AVI. Vomiting (PT:10047700), abdominal pain (PT: 10000081), constipation (PT: 10010774), injection site phlebitis (PT:10022090), hypokalaemia (PT: 10021015), dysgeusia (PT: 10013911), nephrolithiasis (PT:10029148), urticaria (PT:10046735), lymphocytosis (PT:10025280), increased blood lactate dehydrogenase (10005630), prothrombin time prolonged (PT: 10037063), vulvovaginal inflammation (PT: 10079372), angioedema (PT: 10002424), and erythema multiforme (PT: 10015218), listed on the prescription label, failed to satisfy the requirements of at least one of the four algorithms we examined.

Table 4

List of 61 significant PTs associated with CAZ/AVI identified by four pharmacovigilance methods simultaneously in the FAERS database
SOC	PT	PT/N	ROR (95% Two-Sided CI)	PRR (95% Two-Sided CI)	χ2	IC (IC025)	EBGM (EBGM05)
General disorders and administration site conditions	Death	119	6.15 (5.10–7.42)	5.72 (4.82–6.80)	470.38	2.52 (2.18)	5.72 (4.74)
	Drug resistance	24	33.76 (22.54–50.56)	33.21 (22.33–49.40)	748.94	5.05 (3.28)	33.16 (22.14)
	Multiple organ dysfunction syndrome	17	21.21 (13.15–34.23)	20.97 (13.07–33.65)	323.22	4.39 (2.64)	20.95 (12.99)
	Treatment failure	16	6.49 (3.97–10.63)	6.43 (3.95–10.47)	73.48	2.68 (1.59)	6.43 (3.93)
	Drug ineffective for unapproved indication	7	4.86 (2.31–10.22)	4.84 (2.31–10.14)	21.35	2.28 (0.70)	4.84 (2.30)
Injury, poisoning and procedural complications	Off label use	94	4.64 (3.76–5.72)	4.40 (3.62–5.35)	250.43	2.14 (1.78)	4.40 (3.57)
	Product storage error	8	5.04 (2.51–10.09)	5.01 (2.51–10.01)	25.74	2.33 (0.84)	5.01 (2.50)
	Incorrect product administration duration	6	10.06 (4.51–22.43)	10.02 (4.51–22.27)	48.71	3.32 (1.06)	10.01 (4.49)
	Incorrect drug administration rate	4	46.23 (17.31-123.47)	46.10 (17.31–122.80)	176.11	5.52 (0.93)	46.00 (17.22)
	Circumstance or information capable of leading to medication error	4	6.91 (2.59–18.45)	6.90 (2.59–18.35)	20.17	2.79 (0.39)	6.89 (2.58)
	Product administered to patient of inappropriate age	4	9.17 (3.44–24.48)	9.15 (3.44–24.35)	29.03	3.19 (0.53)	9.15 (3.43)
	Product name confusion	3	92.12 (29.60-286.69)	91.93 (29.61-285.41)	268.63	6.52 (0.54)	91.53 (29.41)
Infections and infestations	Pathogen resistance	49	210.84 (158.35-280.73)	203.63 (154.43–268.50)	9785.65	7.66 (4.92)	201.66 (151.46)
	Septic shock	21	22.86 (14.85–35.18)	22.54 (14.74–34.46)	432.01	4.49 (2.90)	22.51 (14.63)
	Sepsis	12	4.79 (2.71–8.45)	4.76 (2.71–8.35)	35.64	2.25 (1.09)	4.75 (2.69)
	Pseudomonas infection	6	35.50 (15.91–79.20)	35.35 (15.90-78.61)	199.96	5.14 (1.51)	35.29 (15.82)
	Klebsiella infection	5	45.10 (18.72-108.62)	44.94 (18.72-107.91)	214.37	5.49 (1.27)	44.85 (18.62)
	Pneumonia Klebsiella	5	131.59 (54.54-317.52)	131.13 (54.51-315.44)	641.66	7.03 (1.37)	130.31 (54.01)
	Osteomyelitis	4	9.53 (3.57–25.43)	9.51 (3.57–25.30)	30.44	3.25 (0.55)	9.50 (3.56)
	Encephalitis	3	20.09 (6.47–62.40)	20.05 (6.47–62.13)	54.25	4.32 (0.38)	20.03 (6.45)
	Candida infection	3	6.57 (2.12–20.40)	6.56 (2.12–20.31)	14.13	2.71 (0.04)	6.56 (2.11)
Nervous system disorders	Seizure	14	4.02 (2.38–6.81)	3.99 (2.37–6.73)	31.49	2.00 (1.00)	3.99 (2.36)
	Encephalopathy	11	21.54 (11.90–39.00)	21.38 (11.87–38.54)	213.58	4.42 (2.16)	21.36 (11.80)
	Myoclonus	7	28.84 (13.72–60.64)	28.70 (13.70-60.13)	186.95	4.84 (1.68)	28.67 (13.63)
	Epilepsy	7	10.95 (5.21–23.02)	10.90 (5.21–22.83)	62.95	3.45 (1.28)	10.90 (5.18)
	Status epilepticus	5	20.03 (8.32–48.22)	19.96 (8.32–47.91)	89.99	4.32 (1.10)	19.94 (8.28)
	Depressed level of consciousness	5	6.33 (2.63–15.24)	6.31 (2.63–15.15)	22.37	2.66 (0.58)	6.31 (2.62)
	Altered state of consciousness	5	10.65 (4.42–25.62)	10.61 (4.42–25.46)	43.53	3.41 (0.87)	10.61 (4.41)
	Neurotoxicity	4	9.66 (3.62–25.78)	9.64 (3.62–25.64)	30.95	3.27 (0.55)	9.63 (3.61)
	Nervous system disorder	4	8.90 (3.33–23.74)	8.87 (3.33–23.62)	27.95	3.15 (0.52)	8.87 (3.32)
	Generalised tonic-clonic seizure	3	6.21 (2.00-19.28)	6.20 (2.00-19.20)	13.08	2.63 (0.01)	6.20 (2.00)
	Brain oedema	3	11.69 (3.77–36.32)	11.67 (3.77–36.16)	29.26	3.54 (0.25)	11.67 (3.76)
	Partial seizures	3	26.31 (8.47–81.74)	26.26 (8.47–81.38)	72.81	4.71 (0.43)	26.23 (8.44)
	Petit mal epilepsy	3	30.03 (9.66–93.28)	29.96 (9.67–92.87)	83.88	4.90 (0.44)	29.92 (9.63)
Investigations	Platelet count decreased	12	5.03 (2.85–8.88)	5.00 (2.84–8.78)	38.42	2.32 (1.14)	5.00 (2.83)
	Blood creatinine increased	7	5.18 (2.46–10.88)	5.16 (2.46–10.80)	23.48	2.37 (0.76)	5.16 (2.45)
	White blood cell count increased	5	6.39 (2.66–15.39)	6.37 (2.66–15.29)	22.66	2.67 (0.59)	6.37 (2.65)
	Blood bilirubin increased	5	9.67 (4.02–23.26)	9.64 (4.02–23.12)	38.69	3.27 (0.82)	9.63 (4.00)
	Blood alkaline phosphatase increased	3	7.04 (2.27–21.87)	7.03 (2.27–21.78)	15.52	2.81 (0.07)	7.03 (2.26)
Renal and urinary disorders	Acute kidney injury	21	4.28 (2.78–6.59)	4.23 (2.77–6.47)	52.06	2.08 (1.27)	4.23 (2.75)
Renal and urinary disorders	Renal failure	15	4.49 (2.70–7.47)	4.46 (2.69–7.37)	40.30	2.16 (1.16)	4.46 (2.68)
Hepatobiliary disorders	Cholestasis	13	32.11 (18.59–55.47)	31.83 (18.52–54.70)	387.70	4.99 (2.55)	31.78 (18.40)
	Hepatocellular injury	7	14.58 (6.94–30.64)	14.51 (6.93–30.39)	88.03	3.86 (1.43)	14.50 (6.90)
	Drug-induced liver injury	4	5.44 (2.04–14.51)	5.43 (2.04–14.44)	14.44	2.44 (0.26)	5.42 (2.03)
	Hepatic failure	4	6.66 (2.50-17.78)	6.65 (2.50-17.69)	19.20	2.73 (0.37)	6.65 (2.49)
	Hepatitis	4	7.76 (2.91–20.70)	7.74 (2.91–20.59)	23.46	2.95 (0.45)	7.73 (2.90)
	Hepatic cytolysis	3	25.97 (8.36–80.69)	25.92 (8.36–80.33)	71.79	4.69 (0.42)	25.89 (8.33)
Skin and subcutaneous tissue disorders	Toxic epidermal necrolysis	4	13.73 (5.15–36.65)	13.70 (5.15–36.46)	47.06	3.77 (0.68)	13.69 (5.13)
Skin and subcutaneous tissue disorders	Petechiae	3	13.40 (4.32–41.62)	13.38 (4.32–41.44)	34.33	3.74 (0.29)	13.37 (4.30)
Blood and lymphatic system disorders	Thrombocytopenia	11	4.67 (2.58–8.46)	4.64 (2.58–8.37)	31.49	2.21 (1.01)	4.64 (2.56)
	Eosinophilia	5	13.39 (5.56–32.24)	13.35 (5.56–32.03)	57.10	3.74 (0.97)	13.34 (5.54)
	Haemolytic anaemia	4	22.03 (8.25–58.81)	21.97 (8.25–58.49)	79.99	4.46 (0.81)	21.95 (8.22)
	Haemolysis	3	18.12 (5.83–56.27)	18.08 (5.84–56.02)	48.37	4.18 (0.36)	18.07 (5.82)
Gastrointestinal disorders	Melaena	5	9.77 (4.06–23.51)	9.74 (4.06–23.37)	39.20	3.28 (0.83)	9.73 (4.04)
Respiratory, thoracic and mediastinal disorders	Respiratory failure	13	8.42 (4.88–14.54)	8.35 (4.86–14.35)	84.19	3.06 (1.69)	8.35 (4.84)
Psychiatric disorders	Mental status changes	5	9.68 (4.02–23.30)	9.65 (4.02–23.16)	38.77	3.27 (0.82)	9.65 (4.01)
Psychiatric disorders	Delirium	4	5.63 (2.11–15.02)	5.61 (2.11–14.94)	15.17	2.49 (0.28)	5.61 (2.10)
Vascular disorders	Shock haemorrhagic	3	15.64 (5.04–48.59)	15.61 (5.04–48.37)	41.01	3.96 (0.33)	15.60 (5.02)
Cardiac disorders	Cardiac arrest	7	4.50 (2.14–9.45)	4.48 (2.14–9.38)	18.94	2.16 (0.64)	4.48 (2.13)
Metabolism and nutrition disorders	Hypernatraemia	5	49.21 (20.43-118.54)	49.04 (20.42-117.77)	234.78	5.61 (1.29)	48.93 (20.31)
Metabolism and nutrition disorders	Metabolic acidosis	4	5.93 (2.22–15.84)	5.92 (2.22–15.76)	16.36	2.57 (0.31)	5.92 (2.22)

Abbreviations: SOC, system organ class; PT, preferred term; ROR, reporting odds ratio; CI, confidence interval; PRR, proportional reporting ratio; χ2, chi-squared; IC, information component; EBGM, empirical Bayesian geometric mean.

Onset Time of AEs

The onset times of CAZ/AVI-associated AEs were extracted from the database. In total, 180 CAZ/AVI-related AEs recorded both the start date and event date. According to the findings, the number of cases in which AEs occurred within 1 day (n = 45, 7.17%) or within 1–3 days (n = 44, 7.07%) following the start of CAZ/AVI accounted for nearly half of all cases that had recorded both the start date and event date, as shown in Fig. 2.

RCTs are the gold standard for evaluating the effectiveness of medications, however, they are not the best method for identifying uncommon safety signals[28, 29]. To complement RCTs and enhance safety assessments, real-world observational studies and pharmacovigilance data mining are utilized[30]. In this study, we conducted an analysis using a comprehensive postmarketing safety surveillance database to explore any potential relationship between CAZ/AVI and its AEs, adopting a pharmacovigilance approach to collect and analyze extensive safety data.

This study illustrated that the indication ranking first was Klebsiella infection, possibly because CAZ/AVI has advantages in the treatment of Klebsiella infection[7]. Health experts (83.28%) contributed the majority of the reports; these experts are usually considered more trustworthy sources of reports. In total, 89.18% of all cases experienced severe outcomes. It should be emphasized that the serious outcomes may be correlated with the infection itself rather than with CAZ/AVI-related AEs. However, it is not clear, and it is not possible to clearly distinguish between infection-related outcomes and outcomes caused by AEs due to CAZ/AVI administration.

The disproportionality analysis revealed that “infections and infestations”, “investigations”, “renal and urinary disorders”, “hepatobiliary disorders”, and “congenital, familial and genetic disorders” were the most common and significant AEs at the SOC level. The majority of AEs associated with infection were probably brought on by infected patients' illness development rather than by CAZ/AVI therapy per se, including pathogen resistance and Candidal infections[9]. Our study identified positive signals related to liver and kidney functions, as well as to disorders of the blood and lymphatic systems. The majority of these AEs corresponded with information already documented in the medication's product label. However, when it comes to "congenital, family, and generic disorders," no positive signals met the standard at the PT level, likely due to a restricted number of case reports.

In our analysis, we also detected significant signals of AEs related to the central nervous system (CNS), including seizure, encephalopathy, epilepsy, myoclonus, status epilepticus, depressed level of consciousness, altered state of consciousness, nervous system disorder, neurotoxicity, petit mal epilepsy, generalised tonic-clonic seizure, brain oedema, and partial seizures. These AEs were not extensively mentioned in the CAZ/AVI instructions. However, our analysis results showed that AEs related to the CNS were a significant category of AEs for the drug, with a large number of related case reports being identified. These events should garner clinical attention. A study investigating the CNS adverse events associated with the use of CAZ/AVI revealed that CAZ/AVI exhibited a relatively stronger signal for nervous system disorders than did meropenem, ceftazidime, and ceftriaxone in real-world data[14]. Previous safety research revealed that the single drug ceftazidime can easily penetrate the blood‒brain barrier; however, most of the ADRs in the CNS caused by ceftazidime are transient[31]. Previous research has shown that the main risk factors for CNS-related ADRs in patients receiving antibacterial drugs include kidney failure (ceftazidime and avibactam are primarily excreted through the kidneys in their prototype), potential brain abnormalities, and CNS infections[32]. Additionally, the emerged positive signals “mental status changes” and “delirium” may be associated with the neurotoxicity of the medication or potentially linked to the patient's frailty following illness.

In our safety analysis of CAZ/AVI, we also detected several hemorrhagic AE signals, including petechiae, melaena, and shock haemorrhagic. However, our study did not find any positive signals in terms of coagulation-related test indicators. We speculate that the occurrence of bleeding AEs associated with this product may be related to thrombocytopenia to some extent. However, further studies are needed to better understand the safety characteristics related to bleeding.

Notably, hypernatremia is another newly detected significant signal, which suggests that this sodium-containing preparation may lead to excessive sodium intake in clinical practice. This medication includes approximately 146 mg of sodium per bottle (2.5 g), which accounts for 7.3% of the WHO-recommended adult daily sodium requirement of 2 g. That is, the drug (2.5 g iv q8h) will be responsible for 21.9% of the maximum daily sodium intake of adults recommended by the WHO under normal dosage. Regardless of whether the data are reliable, excessive sodium intake should undoubtedly be considered in patients on a sodium-controlled diet.

According to the study's findings, nearly half of the AE cases occurred within 1 day (n = 45, 7.17%) or within 1–3 days (n = 44, 7.07%) following the start of CAZ/AVI. The median onset time was found to be 4 days. The longest occurrence period was 55 days in 9 of the AE cases that occurred a month later. Therefore, a longer follow-up period is required to monitor the AEs of CAZ/AVI in upcoming clinical studies.

The FAERS database is considered one of the most important sources of data, and the study of spontaneous reporting systems is a valuable technique for discovering potential signals. However, our study has several limitations. First, since all the information was submitted voluntarily, the credibility of the findings may be inconsistent. Second, the data analysis did not consider several unmeasured variables that could impact AEs, such as potential drug‒drug interactions, drug combinations, and comorbidities. Third, the FAERS does not contain information about patients who took the drug without experiencing any AEs. Therefore, it is impossible to deduce the actual incidence of reported AEs from FAERS data. Fourth, we were unable to establish a precise causal link. The disproportionality study only provided an estimation of signal strength, which was statistically significant. It did not quantify risk or identify causality. Prospective clinical trials are still needed to confirm any causal connection[33]. Nevertheless, the FAERS database has been successfully utilized to analyze postmarketing pharmacovigilance studies in the past. Despite these drawbacks, these data can provide a new update for medical staff who are constantly monitoring patients who receive CAZ/AVI.

In conclusion, the time to AE onset, safety signal spectrum, and possible risks associated with CAZ/AVI treatment were all quantitatively evaluated in this investigation using the FAERS pharmacovigilance database. Unexpected and novel severe AEs may also manifest, including melaena, hypernatraemia, depressed level of consciousness, brain oedema, petechiae, delirium, and shock haemorrhagic. Our investigation might offer important supporting data for CAZ/AVI safety in clinical research and practice. Cohort studies and long-term data are still needed to corroborate these results and to further understand the safety profile of CAZ/AVI because this study was based on a spontaneous reporting system database, which inherently involves possible biases.

Acknowledgments

None.

Funding

This work was supported by the National Natural Science Foundation of China (No. 31570361) and the Yichang Medical and Health Research Project (A20-2-010).

Conflicts of interest

The authors declare that they have no conflicts of interest.

Bassetti M, Vena A, Giacobbe DR, et al. Management of infections caused by multidrug-resistant Gram-negative pathogens: recent advances and future directions. Arch Med Res. 2021;52(8):817-27. https://doi.org/10.1016/j.arcmed.2021.09.002
Mills JP, Marchaim D. Multidrug-resistant Gram-negative bacteria. Infect Dis Clin North Am. 2021;35(4):969-94. https://doi.org/10.1016/j.idc.2021.08.001
Shirley M. Ceftazidime-avibactam: a review in the treatment of serious Gram-negative bacterial infections. Drugs. 2018;78(6):675-92. https://doi.org/10.1007/s40265-018-0902-x
Aslan AT, Akova M. Extended spectrum β-lactamase producing Enterobacteriaceae: carbapenem sparing options. Expert Rev Anti Infect Ther. 2019;17(12):969-81. https://doi.org/10.1080/14787210.2019.1693258
Jean S-S, Lee W-S, Lam C, et al. Carbapenemase-producing Gram-negative bacteria: current epidemics, antimicrobial susceptibility and treatment options. Future Microbiol. 2015;10(3):407-25. https://doi.org/10.2217/fmb.14.135
Doi Y. Treatment options for carbapenem-resistant Gram-negative bacterial infections. Clin Infect Dis. 2019;69(Supplement_7):S565-S75. https://doi.org/10.1093/cid/ciz830
Tumbarello M, Raffaelli F, Giannella M, et al. Ceftazidime-avibactam use for klebsiella pneumoniae carbapenemase–producing K. pneumoniae infections: a retrospective observational multicenter study. Clin Infect Dis. 2021;73(9):1664-76. https://doi.org/10.1093/cid/ciab176
Yahav D, Giske CG, Grāmatniece A, et al. New β-lactam–β-lactamase inhibitor combinations. Clin Microbiol Rev. 2020;34(1):1-61. https://doi.org/10.1128/cmr.00115-20
https://nctr-crs.fda.gov/fdalabel/services/spl/set-ids/d9c2803f-dc9c-4b19-b4a3-8303bc8c15fd/spl-doc
Cheng K, Newell P, Chow JW, et al. Safety profile of ceftazidime–avibactam: pooled data from the adult phase II and phase III clinical trial programme. Drug Saf. 2020;43(8):751-66. https://doi.org/10.1007/s40264-020-00934-3
Sternbach N, Leibovici Weissman Y, Avni T, et al. Efficacy and safety of ceftazidime/avibactam: a systematic review and meta-analysis. J Antimicrob Chemoth. 2018;73(8):2021-9. https://doi.org/10.1093/jac/dky124
Zhong H, Zhao X, Zhang Z, et al. Evaluation of the efficacy and safety of ceftazidime/avibactam in the treatment of Gram-negative bacterial infections: a systematic review and meta-analysis. Int J Antimicrob Agents. 2018;52(4):443-50. https://doi.org/10.1016/j.ijantimicag.2018.07.004
Dietl B, Martinez LM, Calbo E, et al. Update on the role of ceftazidime-avibactam in the management of carbapenemase-producing Enterobacterales. Future Microbiol. May 2020;15:473-84. https://doi.org/10.2217/fmb-2020-0012
Guo X, Guo M, Li J, et al. Central nervous system adverse events of ceftazidime/avibactam: A retrospective study using Food and Drug Administration Adverse Event Reporting System. J Clin Pharm Ther. Dec 2022;47(12):2369-72. https://doi.org/10.1111/jcpt.13796
Inácio P, Cavaco A, Airaksinen M. The value of patient reporting to the pharmacovigilance system: a systematic review. Br J Clin Pharmacol. 2017;83(2):227-46. https://doi.org/10.1111/bcp.13098
Patadia VK, Coloma P, Schuemie MJ, et al. Using real-world healthcare data for pharmacovigilance signal detection – the experience of the EU-ADR project. Expert Rev Clin Pharmacol. 2014;8(1):95-102. https://doi.org/10.1586/17512433.2015.992878
Sakaeda T, Tamon A, Kadoyama K, et al. Data mining of the public version of the FDA Adverse Event Reporting System. Int J Med Sci. 2013;10(7):796-803. https://doi.org/10.7150/ijms.6048
Brown EG. Using MedDRA: implications for risk management. Drug Saf. 2004;27(8):591-602. https://doi.org/10.2165/00002018-200427080-00010
Brown EG, Wood L, Wood S. The medical dictionary for regulatory activities (MedDRA). Drug Saf. Feb 1999;20(2):109-17. https://doi.org/10.2165/00002018-199920020-00002
https://ctep.cancer.gov/protocoldevelopment/electronic_applications/docs/aeguidelines.pdf#search=%22CTCAE%22
DuMouchel W, Smith ET, Beasley R, et al. Association of asthma therapy and churg-strauss syndrome: an analysis of postmarketing surveillance data. Clin Ther. 2004;26(7):1092-104. https://doi.org/10.1016/s0149-2918(04)90181-6
Hauben M, Reich L, Chung S. Postmarketing surveillance of potentially fatal reactions to oncology drugs: potential utility of two signal-detection algorithms. Eur J Clin Pharmacol. 2004;60(10):747-50. https://doi.org/10.1007/s00228-004-0834-0
van Puijenbroek EP, Bate A, Leufkens HGM, 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. https://doi.org/10.1002/pds.668
Ang PS, Chen Z, Chan CL, et al. Data mining spontaneous adverse drug event reports for safety signals in Singapore – a comparison of three different disproportionality measures. Expert Opin Drug Saf. 2016;15(5):583-90. https://doi.org/10.1517/14740338.2016.1167184
Shu Y, Ding Y, Dai B, et al. A real-world pharmacovigilance study of axitinib: data mining of the public version of FDA Adverse Event Reporting System. Expert Opin Drug Saf. 2021;21(4):563-72. https://doi.org/10.1080/14740338.2022.2016696
Bate A, Evans SJ. Quantitative signal detection using spontaneous ADR reporting. Pharmacoepidemiol Drug Saf. Jun 2009;18(6):427-36. https://doi.org/10.1002/pds.1742
https://www.ema.europa.eu/en/medicines/human/EPAR/zavicefta
Singh S, Loke YK. Drug safety assessment in clinical trials: methodological challenges and opportunities. Trials. Aug 20 2012;13:138. https://doi.org/10.1186/1745-6215-13-138
Grossman J, Mackenzie FJ. The randomized controlled trial: gold standard, or merely standard? Perspect Biol Med. 2005;48(4):516-34. https://doi.org/10.1353/pbm.2005.0092
Baniasadi S, Habibi M, Haghgoo R, et al. Increasing the number of adverse drug reactions reporting: the role of clinical pharmacy residents. Iran J Pharm Res. Winter 2014;13(1):291-7. https://doi.org/10.22037/ijpr.2014.1442
Grill MF, Maganti RK. Neurotoxic effects associated with antibiotic use: management considerations. Br J Clin Pharmacol. Sep 2011;72(3):381-93. https://doi.org/10.1111/j.1365-2125.2011.03991.x
Deshayes S, Coquerel A, Verdon R. Neurological adverse effects attributable to β-lactam antibiotics: a literature review. Drug Saf. 2017;40(12):1171-98. https://doi.org/10.1007/s40264-017-0578-2
Shu Y, He X, Liu Y, et al. A real-world disproportionality analysis of olaparib: data mining of the public version of FDA Adverse Event Reporting System. Clin Epidemiol. 2022;14:789-802. https://doi.org/10.2147/clep.s365513

Download PDF

Reviewers agreed at journal
11 Jan, 2024
Reviewers invited by journal
06 Jan, 2024
Editor invited by journal
28 Dec, 2023
First submitted to journal
27 Dec, 2023
Editor assigned by journal
26 Dec, 2023

You are reading this latest preprint version

A Real-World Pharmacovigilance Study of Ceftazidime/avibactam: Data Mining of the FDA Adverse Event Reporting System (FAERS) Database

Status:

Version 1

Abstract

Introduction

Aim

Methods

Results

Conclusions

Figures

Impact statements

Introduction

Aim

Ethics approval

Methods

Study Design and Data Sources

Data Cleaning

Statistical Analysis

Results

General Characteristics

Signal Detected at the SOC Level

Signal Detected at the PT Level

Onset Time of AEs

Discussion

Conclusion

Declarations

References

Status:

Version 1