Body Mass Index-related Trends in Adverse Events with Enfortumab Vedotin: Analysis of a Pharmacovigilance Database and Medical Records

Motoki Kei 1,2, Shiori Sato 2, Ryosuke Yamaga 2, Satoko Karaki 2, Yoshihiro Uesawa 1*, Toshimasa Ito 2

1. Department of Medical Molecular Informatics, Meiji Pharmaceutical University, Tokyo 204–8588, Japan;

2. Department of Pharmacy, Tokyo Women’s Medical University Adachi Medical Center, 4-33-1 Kohoku, Adachi-ku, Tokyo 123–8558, Japan;

*Corresponding author: Yoshihiro Uesawa　uesawa@my-pharm.ac.jp

Motoki Kei　kei.motoki@twmu.ac.jp

Shiori Sato sato.shiori_1@twmu.ac.jp

Ryosuke Yamaga Yamaga.ryosuke@twmu.ac.jp

Satoko Karaki tada.satoko@twmu.ac.jp

Toshimasa Ito ito.toshimasa@twmu.ac.jp

Abstract

Background

Enfortumab vedotin (EV) is an antibody–drug conjugate used to treat advanced urothelial carcinoma. Although EV therapy is effective, adverse events—particularly skin disorders—pose a challenge to optimal treatment. Previous reports suggest that body mass index (BMI) may influence the risk of such adverse events, but large-scale analyses and clinical validation are limited. This study aimed to investigate the association between BMI and EV-related adverse events using both a national pharmacovigilance database and single-center medical records.

Methods

Adverse event reports were extracted from the Japanese Adverse Drug Event Report (JADER) database (2004 Q2–2024 Q3), stratified by BMI (< 22 vs ≥ 22 kg/m²). Reporting odds ratios (RORs) and 95% confidence intervals were calculated for each event, and z-tests were used to compare frequencies between BMI groups. Volcano plots visualized disproportionate reporting.

A retrospective medical record review was conducted at a single institution, including 17 patients receiving EV therapy. Patient demographics, treatment dosing, and adverse events were collected. Fisher’s exact test was used for categorical comparisons, and descriptive statistics were reported.

Results

In JADER, patients with BMI ≥ 22 kg/m² had higher reporting frequencies of skin disorders, pruritus, hyperglycemia, and toxic epidermal necrolysis (p < 0.05), whereas low-BMI patients showed increased reports of pyelonephritis (p = 0.007). Retrospective review confirmed that patients who developed skin toxicity had significantly higher BMI and body weight than those without (mean BMI 25.1 vs 17.7 kg/m², p = 0.004; mean weight 63.9 vs 45.8 kg, p = 0.020). These patients also received higher doses relative to ideal body weight (108.3% vs 79.9%, p = 0.007), suggesting a potential contribution of relative overdose to skin toxicity.

Conclusions

High BMI is associated with an increased risk of EV-related skin disorders, potentially exacerbated by dosing based on actual rather than ideal body weight. Conversely, low BMI may increase susceptibility to infections such as pyelonephritis. These findings support individualized risk assessment and dosing strategies based on patient physique, contributing to safer and more effective EV therapy.

Trial Registration

Not applicable

Keywords

Enfortumab vedotin

adverse events

skin toxicity

body mass index

pharmacovigilance

urothelial carcinoma

retrospective analysis

drug dosing

medical record review

volcano plot

JADER database

Background

Mechanism of Action of Enfortumab vedotin and Nectin-4 Expression in Skin

Enfortumab vedotin (EV) is an antibody–drug conjugate (ADC) composed of a human IgG1 monoclonal antibody targeting Nectin-4, conjugated to a microtubule polymerization inhibitor. It is primarily indicated for patients with advanced urothelial carcinoma who have previously received both chemotherapy and immune checkpoint inhibitor therapy (1, 2). Nectin-4, a cell adhesion molecule, is highly expressed in many epithelial malignancies, including bladder cancer, but is generally limited in normal tissues (3). In a study by Challita-Eid et al., Nectin-4 positivity was detected in approximately 69% of tumors, with moderate-to-high expression confirmed in 60% of bladder cancer cases (3). Analysis of 294 normal tissue samples in the same study showed weak to moderate Nectin-4 expression in keratinocytes of the epidermis, sweat glands, hair follicles, bladder epithelium, salivary glands, and the esophagus (3). Because Nectin-4 is also present in normal skin, EV may exert pharmacological effects on these tissues, potentially leading to adverse cutaneous events such as rash and dermatitis (1). Notably, skin disorders emerged as a prominent adverse effect during EV clinical trials. The drug label carries a warning stating, “Caution should be exercised regarding the onset of skin disorders during the first cycle” (4).

Incidence of Skin Disorders During EV Therapy

Rash and other cutaneous reactions are among the most common adverse events associated with EV, occurring in more than half of treated patients. In a pooled safety analysis of the EV-201 trial conducted in the United States, skin disorders were reported in 55% of patients, with Grade ≥ 3 events observed in 13% (5). Similarly, in the EV-301 trial, skin toxicities of any grade were reported in 47% of patients, and Grade ≧ 3 events occurred in 15% (6). In an initial clinical report from Japan, Nishihara et al. observed skin disorders in 12 of 16 patients (75%) receiving EV, with treatment discontinuation required in 2 cases due to severe rash—including one case of Stevens–Johnson syndrome (SJS) (7). These findings highlight that cutaneous adverse events are frequent and sometimes severe during EV therapy, underscoring the importance of early recognition and management.

Skin rashes typically appear approximately 9 days after the first EV dose (range, 5–18 days), with the highest incidence occurring during the first treatment cycle (7). Accordingly, close monitoring of skin condition is essential early in the course of therapy. Multiple case reports have documented severe EV-associated cutaneous adverse events, including SJS and toxic epidermal necrolysis (TEN). A postmarketing safety analysis by the US Food and Drug Administration identified at least eight cases of EV-related SJS/TEN in the first year after the drug’s approval (7). The median time to onset of these severe reactions was 11 days (range, 9–21 days); all patients required hospitalization, and four deaths were reported (8). This incidence is markedly higher than the background occurrence of SJS/TEN in the general population (approximately one to seven cases per million), representing an important—though rare—safety risk associated with EV (8). Although the precise mechanism remains unclear, Nectin-4 expression in normal skin is thought to contribute to the pathogenesis (8). Emerging case reports have begun to provide pathological insights. Viscuse et al. described two distinct EV-associated reactions: one case of rapidly progressive, fatal SJS/TEN approximately 2 weeks after the initial dose and another of erythema multiforme-like rash that responded to corticosteroids and supportive care (1). In Japan, several patients with urothelial carcinoma developed diffuse erythema, high fever, and mucosal lesions around day 9 of EV therapy, consistent with a TEN-like course (9). These observations highlight that while fatal severe skin reactions are uncommon, cutaneous adverse events during EV therapy are frequent, and early detection and prompt management are critical to ensure patient safety and allow continuation of treatment.

Risk Factors for EV-Associated Skin Disorders

Recent retrospective studies have identified several potential risk factors for the development of EV-related skin disorders. One notable factor is a history of cutaneous reactions during prior immune checkpoint inhibitor (ICI) therapy. A multicenter retrospective study conducted in Kyushu, Japan, showed that patients who experienced skin disorders during ICI treatment had a significantly higher incidence of skin reactions after initiating EV. Specifically, 88.9% of patients with prior ICI-related rash developed skin disorders during EV therapy, compared with 36.7% of those without such a history (p = 0.008) (10). These findings suggest that patient-specific factors, such as inherent cutaneous susceptibility or immune hyperreactivity, may increase the likelihood of EV-associated skin events.

Recent studies have also investigated the association between patient physique—particularly body weight and body mass index (BMI)—and the risk of drug-related adverse events. BMI is a widely used clinical measure of obesity and is closely linked to the risk of obesity-related and atherosclerotic diseases. Epidemiological studies in Japanese populations have shown that the lowest risk for many diseases occurs at a BMI of approximately 22 kg/m², which is considered the ideal body weight for this population (11). In the context of EV therapy, a retrospective analysis by Vlachou et al. (Johns Hopkins Hospital, USA) involving 56 patients receiving EV monotherapy demonstrated a significant association between higher baseline body weight/BMI and the occurrence of skin manifestations. The median weight in the lesion group was 80.9 kg, significantly higher than 69.4g in the no-lesion group, and the median BMI was 26.5 compared with 23.3, respectively (p < 0.01). These findings suggest that higher body weight and elevated BMI may be risk factors for EV-associated skin lesions (12). The potential benefit of prophylactic steroid premedication in preventing such reactions has also been investigated. Furubayashi et al. compared the incidence of skin disorders between a cohort receiving dexamethasone premedication before EV (n = 6) and a cohort without steroid premedication (n = 22). Skin disorders occurred in 33.3% of patients in the steroid premedication group versus 45.5% in the non-premedication group, indicating a modest reduction with steroid use. Furthermore, the incidence of severe rash was 16.7% with steroid premedication compared with 36.4% without.

However, because of the small sample size, these differences did not reach statistical significance (13). Importantly, no significant differences in treatment efficacy (e.g., tumor response rates or survival outcomes) were observed between patients who did and did not receive steroid premedication. While these findings suggest a possible role for corticosteroid prophylaxis in reducing EV-associated skin toxicity, the authors emphasized that further studies are needed to confirm its effectiveness.

Correlation Between Treatment Effectiveness and Skin Disorder Manifestation

Cutaneous toxicity associated with EV has been increasingly recognized not only as an adverse effect but also as a potential biomarker of treatment efficacy. Although the Japanese UROKYU study found no significant differences in response rates or survival between patients who did and did not develop skin toxicity during EV therapy (10), several retrospective studies from other countries have reported contrasting results. For instance, a retrospective analysis from Johns Hopkins Hospital demonstrated a markedly higher objective response rate (68.3%) in patients who experienced skin toxicity compared with those without such events (20.7%) (14). Similarly, the disease control rate was substantially higher in the skin-toxicity group (approximately 83%) than in the no-toxicity group (48%). Furthermore, overall survival was significantly longer among patients with skin reactions, with a reported hazard ratio of 0.48 (p = 0.0235) (14). These findings suggest that the occurrence of skin disorders during EV therapy may be associated with improved tumor control and clinical outcomes in certain patient populations.

Purpose and Research Design of This Study

Although EV-associated skin disorders occur frequently, their risk factors remain poorly understood. Previous studies suggest that the presence or absence of these adverse events may influence treatment efficacy (10, 14), underscoring the need to clarify factors that contribute to their development and to establish effective management strategies that support patient adherence and treatment continuity. In particular, evidence from adverse event databases on EV-related skin toxicity is limited, and no prior analyses have focused on body weight or BMI. While some clinical studies using medical records have indicated that BMI may be a risk factor for skin disorders (12), none have examined whether BMI-related parameters—such as weight-based drug dosage—are associated with the risk of these reactions. To address this knowledge gap, we investigated the relationship between body weight, BMI, and the occurrence of EV-related skin disorders. Specifically, we used the Pharmaceuticals and Medical Devices Agency’s (PMDA) Japan Adverse Drug Event Report Database (JADER) to detect potential signals of adverse events, and we conducted a retrospective analysis of single-institution medical record data to validate the influence of body weight and BMI on skin disorder occurrence. Given that EV has only recently entered clinical practice, case numbers from individual institutions or regions remain limited, making it challenging to draw robust conclusions from single-center studies. To address this limitation, we utilized the JADER—a publicly accessible pharmacovigilance database that compiles spontaneous adverse event reports nationwide—to conduct a broad, hypothesis-generating analysis at the national level. However, spontaneous reporting systems such as JADER have inherent methodological limitations, including reporting, selection, and information biases, as well as the absence of denominators (i.e., total drug exposure), which preclude accurate incidence estimation and causal inference based on signal detection alone.

To overcome these constraints, our study was designed in two complementary stages: (1) initial signal detection using JADER to identify potential risk associations between body weight, BMI, and EV-related skin disorders and (2) subsequent validation through more controlled analysis of single-institution medical record data, aiming to minimize confounding factors and enhance the robustness of our findings.

Methods

Database Analysis Using JADER

Database

Japan maintains a national spontaneous adverse reaction reporting system known as the JADER, administered by the PMDA (15). JADER comprehensively integrates data on drugs used in Japan with reported adverse reactions and relevant patient information. Since the PMDA’s establishment in April 2004, adverse event reports have been continuously collected and made publicly available through JADER. The database is updated monthly, and all datasets can be freely accessed and downloaded by researchers. For this study, we obtained JADER datasets covering the period from the first quarter of 2004 (April–June) through the third quarter of 2024 (October–December), comprising a total of 949,124 reports.

Creation of Analysis Data Tables

In JADER, each adverse event report is divided into four tables according to content: the demographic information table (DEMO), the drug information table (DRUG), the adverse reaction table (REAC), and the medical history table (HIST). Figure 1 provides an overview of the information contained in each table. For this analysis, we used the DEMO, DRUG, and REAC tables. No restrictions were placed on the role of the drug; EV was included as a suspect, concomitant, or interacting drug, and all entries were treated equally.

We performed data cleaning on DEMO before analysis. (1) Cases lacking the height and weight data required to calculate BMI were excluded. BMI was calculated as weight (kg)/[height (m)]². (2) Cases with extreme BMI values (< 10 kg/m² or > 100 kg/m²) were excluded as likely data entry errors. To minimize potential digit preference bias in the reporting of height and weight, we applied a uniform correction by adding 5 kg to all recorded weights and 5 cm to all recorded heights (e.g.,, a recorded weight of 50 kg was treated as 55 kg). After data cleaning, the resulting DEMO dataset (termed “mDEMO”) contained 297,976 cases. These were stratified into two groups based on BMI: ≧22 kg/m² and < 22 kg/m², using 22 kg/m² as the cutoff. This threshold was selected based on epidemiological evidence indicating the lowest overall health risk at a BMI of approximately 22 kg/m² in Japanese populations (11).

Next, we merged the mDEMO, DRUG, and REAC tables using unique case identification numbers. Duplicate entries corresponding to the same case (e.g., follow-up reports) were consolidated into a single record to avoid double-counting. Using the case ID as a key, we linked DRUG and REAC information to each entry in mDEMO, thereby constructing two analytical datasets: one for the BMI ≧ 22 kg/m² group (Table A) and one for the BMI < 22 kg/m² group (Table B). For each dataset, 2 × 2 contingency tables were generated to cross-classify exposure to EV (yes/no) against the presence or absence of specific adverse events. Based on these tables, we calculated the reporting odds ratio (ROR) and corresponding p-value using Fisher’s exact test to assess the association between EV exposure and each adverse event within each BMI group. To account for zero cell counts in the contingency tables, Haldane’s correction (adding 0.5 to all cell values) was applied (16, 17). Furthermore, to prevent overestimation due to multiple reports of the same event in a single patient, each unique adverse event (Preferred Term [PT]) was counted only once per case. Figure 2 presents an overview of the data processing and analytical workflow.

Figure 1 Structure of the JADER

Schematic diagram illustrating the four information tables comprising JADER: the Case List Table (DEMO), Drug Information Table (DRUG), Adverse Reaction Table (REAC), and Underlying Disease Table (HIST). The figure shows the name and primary data elements contained in each table.

Fig. 2

Flowchart for the analytical dataset construction process

Flowchart outlining the steps used to create the analytical dataset. After data cleaning to exclude cases with missing or extreme BMI values, the DEMO, DRUG, and REAC tables were merged using case IDs. The resulting dataset was stratified into BMI ≥ 22 kg/m² and BMI < 22 kg/m² groups for contingency table analysis and signal detection.

Adverse Event Terminology and Scope

We restricted the analysis to adverse events reported in association with the investigational drug EV. All adverse events recorded for EV in the JADER were included without additional filtering. Event terminology was standardized according to the Medical Dictionary for Regulatory Activities (MedDRA), version 27.1 (18), ensuring uniform classification and facilitating reproducibility of the analysis.

Relationship Between EV and Adverse Events (Signal Detection)

For each BMI-stratified dataset (Table A, BMI < 22 kg/m²; Table B, BMI ≧ 22 kg/m²), we calculated the ROR as a disproportionality measure, along with the corresponding 95% confidence intervals (CIs) and p-values, to assess the association between EV and each adverse event. The ROR was defined as (a/b)/(c/d), where “a” is the number of EV cases with the event, “b” is the number of EV cases without the event, “c” is the number of non-EV cases with the event, and “d” is the number of non-EV cases without the event. Two-tailed p-values for each 2 x 2 contingency table were calculated using Fisher’s exact test (Table 1). Results were visualized using volcano plots, with the natural logarithm of the ROR (ln ROR) plotted on the x-axis and the − log10(p-value) plotted on the y-axis (19–25). Figure 3-a and 3-b present the volcano plots for the BMI ≧ 22 kg/m² and BMI < 22 kg/m² groups, respectively. In these plots, each point represents an adverse event PT; points located toward the right with higher − log10(p-value) indicate events with higher RORs (signals) and stronger statistical significance. Volcano plots, widely used in fields such as genomics for differential expression analyses, enable intuitive visualization and identification of disproportionality signals in pharmacovigilance datasets (24).

Table 1
2 ×2 contingency table used for signal detection analysis
	Specific adverse event	Other adverse events
EV	a	b
Non-EV	c	d

2 × 2 contingency table illustrating the classification of cases based on exposure to EV and the presence or absence of a specific adverse event. a, b, c, and d represent the number of cases in each cell. The ROR is calculated as (a/b)/(c/d)

ROR, reporting odds ratio = (a/b)/(c/d)

Statistical Analysis (JADER)

All computational analysis for signal detection—including calculation of the ROR, p-values, and 95% CIs—were performed using Python version 3.12.7. Volcano plots were generated with JMP Pro version 18.0 (SAS Institute Inc., Cary, NC, USA). A p-value of < 0.05 was considered statistically significant for signal detection. For direct comparisons between the BMI ≧ 22 kg/m² and BMI < 22 kg/m² groups, differences in proportions were assessed using a two-tailed z-test, with |z| >1.96 as the threshold for statistical significance (corresponding to p < 0.05). The z-test was based on the standard normal approximation for two independent proportions, consistent with the method proposed by Lu et al. for evaluating host–factor interactions in spontaneous reporting databases (26).

The formula used for the z-test is shown below:

$\:\text{Z\:=\:}\frac{{\text{l}\text{n}(\text{R}\text{O}\text{R}}_{1})-{\text{l}\text{n}(\text{R}\text{O}\text{R}}_{2})}{\sqrt{{\text{S}\text{E}}_{1}^{2}-{\text{S}\text{E}}_{2}^{2}}}$

$\:\text{SE\:=\:}\frac{\text{ln}\left(95\text{\%}\text{C}\text{I}\:\text{u}\text{p}\text{p}\text{e}\text{r}\:\text{b}\text{o}\text{u}\text{n}\text{d}\right)-\text{l}\text{n}(\text{95\%\:CI\:lower\:bound)}}{2\:\times\:\:1.96}$

Retrospective Medical Record Review

Subjects

We conducted a retrospective review of medical records at our institution, including patients who received EV therapy between November 2021 and August 2024. A total of 17 patients met the inclusion criteria. Patients were excluded if they were younger than 20 years at the time of treatment, did not complete at least one full cycle of EV therapy, or participated in a clinical trial as part of prior treatment.

Collected Variables

We collected comprehensive baseline and treatment data from patient records, including age at EV initiation, sex, height, weight at each administration, the actual EV dose administered at each infusion, the EV dose calculated based on ideal body weight (“ideal body weight dose”), the ideal body weight dosage rate (%), Eastern Cooperative Oncology Group Performance Status at EV initiation, primary tumor site (e.g., bladder or upper urinary tract), presence of metastases, prior treatment history, and key baseline laboratory values. The ideal body weight dosage rate* was defined as the percentage of the actual administered dose relative to the dose that would be given based on the patient’s ideal body weight. The formula for this rate is:

$\:\text{I}\text{d}\text{e}\text{a}\text{l}\:\text{W}\text{e}\text{i}\text{g}\text{h}\text{t}\:\text{D}\text{o}\text{s}\text{a}\text{g}\text{e}\:\text{R}\text{a}\text{t}\text{e}（\text{％}）=\frac{\text{A}\text{c}\text{t}\text{u}\text{a}\text{l}\:\text{a}\text{d}\text{m}\text{i}\text{n}\text{i}\text{s}\text{t}\text{e}\text{r}\text{e}\text{d}\:\text{d}\text{o}\text{s}\text{e}}{\text{D}\text{o}\text{s}\text{e}\:\text{b}\text{a}\text{s}\text{e}\text{d}\:\text{o}\text{n}\:\text{i}\text{d}\text{e}\text{a}\text{l}\:\text{b}\text{o}\text{d}\text{y}\:\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}}\:\times\:\:100$

The ideal body weight-based dose was calculated using a standard formula for ideal body weight.

Evaluated Outcomes

In the medical record analysis, we assessed the occurrence of key adverse events during EV therapy, including skin disorders (e.g., rash or dermatitis), hyperglycemia, peripheral neuropathy, neutropenia, and the number of EV treatment cycles completed. Adverse events were identified and graded according to the same MedDRA v27.1 terminology used in the JADER analysis to ensure consistency across datasets.

Statistical Analysis (Medical Records)

The retrospective clinical data were analyzed using JMP Pro 18.0, with a p-value of < 0.05 considered statistically significant. Categorical variables (e.g., occurrence of specific adverse events in high-BMI vs. low-BMI patients) were compared using Fisher’s exact test. Given the small sample size, the analysis primarily reports descriptive comparisons and exact p-values, without multivariate adjustment.

Ethical Considerations

The use of anonymized data from the open-access JADER database did not require specific ethical approval, as it involved secondary use of publicly available information.

For the retrospective medical record review, ethical approval was obtained from the Tokyo Women’s Medical University Ethics Review Committee (Approval No. 2024 − 0114) prior to data collection.

The study was conducted in accordance with the principles of the Declaration of Helsinki and relevant regulatory guidelines.

Given the retrospective observational design, the requirement for written informed consent was waived by the Ethics Review Committee.

Results

z-test Results (BMI-Stratified JADER Analysis)

Using the BMI-stratified JADER datasets, we identified five adverse events with significantly different reporting frequencies between the high-BMI ( ≧ 22 kg/m²) and low-BMI (< 22 kg/m²) groups based on the z-test. These results are summarized in Table 2. Specifically, pruritus, skin disorders (composite term for rash/dermatitis), hyperglycemia, and TEN were reported significantly more frequently in the high-BMI group, whereas pyelonephritis occurred significantly more often in the low-BMI group (all p < 0.05). For instance, pyelonephritis had a z-score of − 2.70(p = 0.007), indicating a signal disproportionately associated with the low-BMI group. In contrast, pruritus, skin disorders, hyperglycemia, and TEN had positive z-scores (2.14–2.46) with p-values ranging from 0.014 to 0.047, indicating higher reporting rates in the high-BMI group. These findings suggest that the adverse event profile of EV may differ according to patient BMI, with dermatologic events more common among patients with higher BMI and certain infections (e.g., pyelonephritis) more frequent in those with lower BMI.

Table 2
z-test results comparing adverse event frequencies between high-BMI (≥ 22 kg/m²) and low-BMI (< 22 kg/m²) groups
Adverse event	BMI ≧ 22kg/ m²	BMI < 22kg/ m²	Z-score	p-value (two-tailed test)
Adverse event	ROR(95%CI)	ROR(95%CI)	Z-score	p-value (two-tailed test)
Pyelonephritis	5.37(2.11–13.7)	24.4(13.6–43.5)	-2.70	0.007
pruritus	11.4(6.90–18.7)	2.11(0.61–7.34)	2.46	0.014
skin disorder	96.0(61.4-150.2)	36.7(18.7–71.7)	2.34	0.019
Hyperglycemia	18.4(12.1–27.9)	7.32(3.53–15.2)	2.14	0.032
Toxic Epidermal Necrolysis	15.8(10.2–24.6)	7.31(3.94–13.6)	1.99	0.047

The high-BMI group showed significantly higher reporting frequencies of pruritus, skin disorders, hyperglycemia, and TEN, whereas the low-BMI group showed a significantly higher reporting frequency of pyelonephritis (all p < 0.05).

Volcano Plot Analysis

To visually assess disproportionality signals, we generated volcano plots for each BMI group (high-BMI and low-BMI) (Figs. 3A and 3B). In these plots, labeled points correspond to adverse events with statistically significant z-test results. In the BMI ≥ 22 kg/m² group (Fig. 3A), the points for pruritus, skin disorders, hyperglycemia, and TEN are positioned toward the right side of the plot with elevated − log10(p-values), indicating strong signals associated with EV in this subgroup. In contrast, in the BMI < 22 kg/m² group (Fig. 3B), pyelonephritis is depicted as an outlying point with a high ln(ROR) and − log10(p-value), confirming its disproportionately higher reporting frequency in patients with lower BMI. Taken together, these volcano plots provide a clear visual complement to the z-test findings: EV-related dermatologic events and hyperglycemia are prominent in high-BMI patients, whereas pyelonephritis stands out as a notable signal in low-BMI patients.

Figure 3 Volcano plots of EV-associated adverse events by BMI group

(a) BMI ≥ 22 kg/m² group: Each point represents an adverse event (PT); the x-axis shows ln(ROR) and the y-axis shows − log₁₀(p-value). Pruritus, skin disorders, hyperglycemia, and toxic epidermal necrolysis are highlighted as signals with significantly higher RORs (toward the right) and strong statistical significance (toward the top). (b) BMI < 22 kg/m² group: This plot similarly maps ln(ROR) against − log₁₀(p-value). Pyelonephritis is highlighted as a notable signal in this group, with a markedly higher ROR and statistical significance in low-BMI patients. Skin disorders, hyperglycemia, and TEN also appear with elevated RORs but primarily reflect signals observed in the high-BMI group, allowing visual comparison between subgroups

Patient Baseline Characteristics in the Retrospective Study (Grouped by Skin Disorder)

We next examined the baseline characteristics of the 17 patients included in the retrospective medical record review, comparing those who developed skin toxicity during EV therapy (n = 12) with those who did not (n = 5). Table 3 summarizes these findings. There were no significant differences between the two groups with respect to age, sex distribution, cancer stage, or prior treatment history. However, patients who experienced skin toxicity had a significantly higher median BMI and body weight than those without skin events (mean BMI 25.1 vs. 17.7, p = 0.004; mean weight 63.9 vs. 45.8 kg, p = 0.020). They also received a greater number of EV treatment cycles on average (mean 11.6 vs. 2.6 cycles, p = 0.008), reflecting longer treatment duration in this group. Additionally, the initial dose per ideal body weight was substantially higher among patients with skin toxicity (mean 108.3% vs. 79.9% of the ideal-weight dose, p = 0.007), indicating relative overdosing compared with the nonskin-toxicity group. Although the actual initial dose per body weight was numerically higher in the skin-toxicitygroup, the difference did not reach statistical significance (p = 0.095). Collectively, these findings suggest that higher body mass and dosing relative to ideal body weight are associated with an increased risk of skin toxicity during EV treatment, consistent with the BMI-related signal observed in the JADER analysis.

Table 3
Baseline characteristics of patients receiving EV therapy, stratified by presence or absence of skin toxicity
Characteristics	Non-skin toxicity group (N = 5)	Skin toxicity group (N = 12)	All Patients (N = 17)	p-value
Age,years	66.6 ± 4.02	69.8 ± 2.60	68.8 ± 8.83	0.260
Female sex,n(%)	2	3	5	0.600
Initial cancer stage,n(%)
Ⅰ	0	0	0	1.000
Ⅱ	0	0	0	1.000
Ⅲ	1	1	2	1.000
Ⅳ	3	9	12	1.000
Unknown	1	2	3	1.000
Prior chemotherapy,n(%)
Gemcitabine + Platinum preparation	5	12	17	1.000
Avelumab	4	9	13	1.000
Pembrolizumab	1	2	3	1.000
Nivolumab	0	2	2	1.000
BMI,kg/ m²	17.7 ± 2.11	25.1 ± 1.31	22.5 ± 5.44	0.004
Weight,kg	45.8 ± 6.40	63.9 ± 4.11	58.6 ± 16.2	0.020
Number of chemotherapy cycles	2.60 ± 2.45	11.6 ± 1.60	8.94 ± 6.78	0.008
Initial dose per body weight,%	88.5 ± 5.73	97.9 ± 3.70	95.1 ± 13.2	0.095
Initial dose per IBW, %	79.9 ± 8.49	108.3 ± 5.48	99.9 ± 22.7	0.007

A total of 17 patients were included in the retrospective analysis (5 without skin toxicity and 12 with skin toxicity). There were no significant differences between the two groups with respect to age, sex, cancer stage at baseline, or prior chemotherapy. In contrast, patients who developed skin toxicity had significantly higher BMI and body weight and received a greater number of EV treatment cycles compared with those without skin toxicity. The skin-toxicity group also had a significantly higher initial dose relative to ideal body weight, indicating a higher effective dose in heavier patients.

Consistent with these findings, Fig. 4 shows that the BMI distribution was significantly higher in the skin-toxicity group than in the no-toxicity group (p = 0.004). Similarly, Fig. 5 demonstrates that the ratio of actual dose to ideal-body-weight dose was significantly greater in patients with skin toxicity (p = 0.007). Taken together with Table 3, these results highlight a clear trend: patients with higher BMI, who consequently received a larger relative dose—were more likely to develop EV-related skin adverse events.

Figure 4 Relationship between BMI and skin toxicity during EV therapy

This box plot compares BMI values between patients who developed skin toxicity and those who did not. The median BMI was significantly higher in the skin-toxicity group than in the no-toxicity group (p = 0.004)

Fig. 5

Relationship between dose (relative to ideal body weight) and skin toxicity

This box plot compares the initial EV dosage as a percentage of the ideal bod -weight-based dose between patients with and without skin toxicity. The dosage relative to ideal weight was significantly higher in the skin-toxicity group (mean ≈ 110%) than in the no-toxicity group (mean ≈ 80%) (p = 0.007), indicating that patients with higher BMI effectively received an excessive dose, which may have contributed to the development of skin toxicity

Discussion

Studies Using JADER

Comparisons between BMI groups using z-tests for each adverse event revealed statistically significant differences for skin disorders, pruritus, pyelonephritis, hyperglycemia, and TEN (all p < 0.05). These patterns were consistent with the volcano plot analyses. In the BMI ≥ 22 kg/m² group, skin disorders, pruritus, pyelonephritis, hyperglycemia, and TEN were associated with significantly higher reported odds ratios and strong statistical signals. Similarly, the BMI < 22 kg/m² group showed elevated odds ratios for skin disorders, pyelonephritis, hyperglycemia, and TEN. Overall, these findings suggest that the adverse event profile of EV varies partly according to patient BMI, with overweight patients more prone to skin-related toxicities and underweight patients showing a higher risk of infections.

The increased risk of skin disorders in patients with high BMI observed in this analysis is consistent with previous reports. For example, Vlachou et al. found that among patients receiving EV therapy, the median baseline BMI at treatment initiation was significantly higher in those who developed skin disorders than in those who did not (26.5 vs. 23.3, p < 0.01), suggesting that elevated BMI may be a risk factor for EV-related skin toxicity (12). Our JADER analysis, based on postmarketing data from Japan, supports and extends these findings beyond a single-center cohort.

Conversely, previous studies have demonstrated a U-shaped relationship between BMI and infection risk, with low BMI associated with a a higher incidence of infections such as respiratory tract infections and sepsis (27). Systematic reviews have further shown that low BMI is linked to an increased risk of pneumonia (28), likely reflecting impaired immune function due to malnutrition (29). This aligns with our observation of increased infection-related adverse events in patients with lower BMI

Furthermore, low BMI has also been identified as a potential risk factor for pyelonephritis. A large Swedish cohort study involving more than one million women demonstrated that those with a BMI < 18.5 had a significantly higher risk of developing pyelonephritis than women of normal weight (HR 1.13, 95% CI 1.06–1.21), suggesting that low BMI may independent;y increase susceptibility to urinary tract infections, particularly pyelonephritis (30). In line with these observations, our adverse event database analysis indicates that BMI may influence the risk profile of adverse events in patients receiving EV. These findings have important clinical implications for preventive and monitoring strategies. For patients with high BMI, early implementation of interventions to prevent skin disorders is warranted at the initiation of EV therapy. Conversely, for underweight patients, heightened vigilance for infection risk—especially pyelonephritis—is necessary given their potentially compromised immunity. During EV therapy, prompt recognition and management of symptoms such as fever or back pain should be prioritized. Overall, this study underscores the importance of individualized adverse event management based on patient physique, which may help enhance the safety of EV therapy.

Investigation Using Medical Records

Based on the trends observed in the JADER database, we hypothesized that “patients with high BMI are more likely to develop skin disorders and pruritus” in real-world clinical settings. To test this, we examined the reproducibility of BMI-related differences in adverse event frequency using medical records from a single institution. Our retrospective analysis confirmed that among patients receiving EV, those who developed skin disorders had significantly higher body weight and BMI than those who did not. Specifically, the mean BMI was notably higher in the skin disorder group, and their actual body weight generally fell within the overweight range. Importantly, when comparing the ratio of actual weight-based dosing to the target dose calculated using ideal body weight, the skin disorder group ,showed a significantly higher value (mean 110.7%) than the nondisorder group (mean 79.9%). This finding indicates that overweight patients were effectively receiving excess EV doses relative to their ideal weight. Thus, in real-world clinical practice, patients with higher BMI appear more prone to developing EV-related skin disorders, and this may be partly explained by relative overdosing linked to weight-based dosing. Notably, the observation of “higher BMI in the skin disorder group” in our hospital data is consistent with both the JADER analysis and previous international reports, supporting the generalizability of BMI’s influence on skin disorder risk during EV therapy.

Furthermore, this analysis revealed a previously unreported association between the actual administered dose and the risk of skin disorders. Specifically, patients in the BMI ≥ 22 kg/m² group received, on average, 110.7% of the dose calculated based on ideal body weight, suggesting that this relative overdose may have contributed to the development of skin toxicity. These findings imply that increased drug exposure resulting from excessive dosing relative to ideal weight—rather than high BMI itself—may be a key underlying factor. This provides a novel and clinically relevant insight from a dosing perspective. Whereas previous studies have primarily focused on patient-related risk factors, such as BMI and medical history (10, 12), our results emphasize the importance of also considering drug-related factors, particularly dose intensity, in understanding and preventing EV-related skin adverse events.

Although this was a single-center study, the use of real-world clinical data provides findings that are directly applicable to routine practice. For patients with high BMI, initiating proactive management to prevent skin disorders at the start of EV therapy is essential. Our results suggest that obese patients may experience higher drug exposure than expected based on standard body weight-based dosing. Accordingly, strategies such as setting upper dose limits when appropriate and implementing preventive skincare interventions from the first treatment cycle are both practical and clinically relevant. Furthermore, recent studies indicate that the occurrence of skin disorders may correlate with EV treatment efficacy, with emerging research exploring the use of skin symptoms as potential biomarkers of treatment response (14). These observations highlight the possibility of relative overdose in patients with high BMI, suggesting that future clinical strategies may need to include revising dosing regimens—such as adopting body surface area-based calculations—to optimize safety and efficacy in this population.

Limitation

This study has several limitations. First, JADER is a spontaneous adverse event reporting database and is inherently susceptible to reporting bias arising from underreporting, missing data, and reporting errors. Furthermore, because the total number of drug users (the denominator) is unknown, it is not possible to calculate absolute incidence rates or accurately assess risk. To enhance the reliability of our analysis, we excluded cases with missing or erroneous data and restricted the study to drugs with a sufficient number of reports. Nonetheless, several reporting biases may still influence the results. These include the “bad publicity effect” (an increase in reports of high-profile adverse events), the “spillover effect” (increased reporting for drugs within the same class or indication), the “Weber effect” (a reporting peak shortly after drug launch followed by a decline over time), and the “masking effect” (underreporting of specific events due to their association with other drugs) (31). Such biases can lead to over- or underestimation of adverse event frequencies and should be considered when interpreting the findings. Additionally, when multiple drugs are administered concomitantly, it is often difficult to identify the specific causative agent responsible for an adverse reaction (32, 33). Although not included in the present analysis, factors such as the underlying disease, concomitant medications, total number of drugs administered, route of administration, and duration of therapy may also act as potential confounders influencing the occurrence of adverse events. The medical record-based component of this study carries its own limitations.

The sample size was relatively small (17 patients), and the retrospective design inherently limits causal inference and control of confounding variables. Furthermore, because the data were obtained from a single institution, potential biases related to treatment practices and reporting accuracy cannot be excluded. Therefore, caution is warranted when generalizing these findings to other populations or clinical settings. Moreover, while this analysis compared groups using a BMI threshold of 22 kg/m², the appropriateness of this cutoff is debatable. Statistically, a BMI of 22 kg/m² is often regarded as the level associated with the lowest population risk, whereas clinically, obesity is typically defined as BMI ≥ 25 kg/m², reflecting differing criteria depending on the context. Given these considerations and the aforementioned limitations, the present findings should be interpreted as hypothesis-generating rather than definitive. Validation through larger prospective studies will be essential to confirm these observations and further clarify the clinical relevance of BMI in EV-associated adverse events.

Integrative Considerations

Findings from the two distinct data sources demonstrated broadly consistent trends. In particular, the association between higher BMI and an increased incidence of EV-related skin disorders was observed in both the JADER database analysis and the medical record review, indicating strong concordance between these datasets. By contrast, while the JADER analysis suggested a significantly higher risk of pyelonephritis in patients with low BMI, this association could not be confirmed in the medical record analysis, underscoring the need for further investigation in larger clinical cohorts. Overall, both data sources supported the conclusion that BMI influences the adverse event profile of EV therapy, with each offering complementary insights. The large-scale database analysis revealed broad population-level patterns—namely,increased dermatologic events in patients with high BMI and increased infection risk in those with low BMI—whereas the institutional medical record review provided real-world clinical validation of the skin disorder association and introduced a novel dosing perspective. Importantly, no major discrepancies emerged between these approaches; rather, integrating the two enhanced the robustness and clinical relevance of the findings.

Future Prospects

The findings of this study warrant further investigation and development. To elucidate the causal relationship between BMI and EV-related adverse events, prospective studies and multicenter collaborative research are essential. Larger patient cohorts are needed to confirm the incidence of adverse events across BMI categories, ideally accompanied by pharmacokinetic analyses to clarify whether drug exposure is increased in obese patients. Determining optimal dosing strategies represents another key challenge. Clinical trials comparing actual body weight-based versus ideal body weight-based dosing in obese patients could help test the overdose hypothesis and identify an optimal balance between efficacy and safety. Similarly, detailed investigation of the relationship between infection risk in low-BMI patients and their nutritional or immune status may enable early identification of high-risk populations and the development of targeted preventive interventions. Furthermore, the role of supportive therapies, such as steroid premedication, in modifying BMI-related adverse events remains an open research question—for example, whether steroid premedication reduces skin toxicity in obese patients. Importantly, the issues highlighted in this study extend beyond EV and are relevant to dosing considerations for other anticancer agents. Future research may ultimately contribute to the establishment of personalized dosing and management strategies tailored to patient factors, including BMI.

Conclusions

This study investigated the relationship between BMI and adverse drug reactions in patients receiving EV, using two complementary data sources: the JADER database and a single-center medical record review. Patients with high BMI exhibited increased risks of skin disorders (including pruritus), TEN, and hyperglycemia, whereas those with low BMI showed a potential increased risk of pyelonephritis. These findings were largely consistent across both datasets, reinforcing high BMI as a risk factor for EV-related skin toxicities. Importantly, this study demonstrated for the first time that excess drug exposure due to relative overdosing in obese patients may contribute to the development of skin disorders. These results highlight the importance of individualized risk assessment and management strategies tailored patient physique and nutritional status.

Future prospective studies are warranted to clarify the underlying causal mechanisms and inform evidence-based dosing guidelines. BMI should be recognized as a key clinical factor influencing adverse event risk during EV therapy and considered in optimizing treatment strategies for individual patients.

List of abbreviations

BMI Body mass index

CI Confidence intervals

EV Enfortumab vedotin

DEMO Demographic table

DRUG Drug Information Table

REAC Reaction table

HIST History Table

ICI Immune checkpoint inhibitor

JADER Japanese Adverse Drug Event Report

PMDA Pharmaceuticals and Medical Devices Agency

PT Preferred Term

ROR Reporting odds ratio

TEN Toxic epidermal necrolysis

Declarations

Ethics approval and consent to participate

The use of anonymized data from the open-access JADER database did not require specific ethical approval, as it involved secondary use of publicly available information. For the retrospective medical record review, ethical approval was obtained from the Tokyo Women’s Medical University Ethics Review Committee (Approval No. 2024 − 0114) prior to data collection. The study was conducted in accordance with the principles of the Declaration of Helsinki and relevant regulatory guidelines. Given the retrospective observational design, the requirement for written informed consent was waived by the Ethics Review Committee.

Consent for publication

Not applicable

Data Availability

Data from the Japanese Adverse Drug Event Report (JADER) database are publicly available from the Pharmaceuticals and Medical Devices Agency (PMDA) website (https://www.pmda.go.jp).The medical record data used in this study were obtained from a single medical institution and contain personal health information. Therefore, these data are not publicly available due to privacy restrictions. Aggregated or de-identified data may be made available from the corresponding author upon reasonable request.

Competing interests

The authors declare that they have no competing interests

Funding

This work was partially supported by the Grants-in-Aid for Scientific Research (KAKENHI) from the Japan Society for the Promotion of Science (JSPS), Grant Number 22K06707.

Author Contribution

Conceptualization, Y.U.; methodology, Y.U.; software, Y.U.; validation, M.K. and Y.U.; formal analysis, M.K. and Y.U.; investigation, M.K. , S.S. , R.Y., S.K. and Y.U.; resources, M.K. and Y.U.; data curation, M.K. and Y.U.; writing—original draft preparation, M.K.; writing—review and editing, M.K. and Y.U.; visualization, M.K. and Y.U.; supervision, T.I. and Y.U.; project administration, Y.U.; funding ac-quisition, Y.U. All authors have read and agreed to the published version of the manuscript.

References

Viscuse PV, Marques-Piubelli ML, Heberton MM, Parra ER, Shah AY, Siefker-Radtke A, Gao J, Goswami S, Ivan D, Curry JL, Campbell MT. Case report: enfortumab vedotin for metastatic urothelial carcinoma: a case series on the clinical and histopathologic spectrum of adverse cutaneous reactions from fatal Stevens-Johnson syndrome/toxic epidermal necrolysis to dermal hypersensitivity Reaction. Front Oncol. 2021;11:621591. org:10.3389/fonc.2021.621591.

Maguire WF, Lee D, Weinstock C, Gao X, Bulik CC, Agrawal S, Chang E, Hamed SS, Bloomquist EW, Tang S, Pazdur R, Kluetz PG, Amiri-Kordestani L, Suzman DL. FDA Approval Summary: Enfortumab Vedotin plus Pembrolizumab for Cisplatin-Ineligible Locally Advanced or Metastatic Urothelial Carcinoma. Clin Cancer Res. 2024;30(10):2011–6. https://doi.org/10.1158/1078-0432.CCR-23-3738.

Challita-Eid PM, Satpayev D, Yang P, An Z, Morrison K, Shostak Y, Raitano A, Nadell R, Liu W, Lortie DR, Capo L, Verlinsky A, Leavitt M, Malik F, Aviña H, Guevara CI, Dinh N, Karki S, Anand BS, Pereira DS, Joseph IB, Doñate F, Morrison K, Stover DR. Enfortumab Vedotin Antibody-Drug Conjugate Targeting Nectin-4 Is a Highly Potent Therapeutic Agent in Multiple Preclinical Cancer Models. Cancer Res. 2016;76(10):3003–13. https://doi.org/10.1158/0008-5472.CAN-15-1313.

Astellas Pharma Inc. (2024, September). PADCEV® for intravenous infusion 30 mg, package insert (8th ed.). https://www.pmda.go.jp/PmdaSearch/bookSearch/01/04987233746362

Yu EY, Petrylak DP, O'Donnell PH, Lee JL, van der Heijden MS, Loriot Y, Stein MN, Necchi A, Kojima T, Harrison MR, Hoon Park S, Quinn DI, Heath EI, Rosenberg JE, Steinberg J, Liang SY, Trowbridge J, Campbell M, McGregor B, Balar AV. Enfortumab vedotin after PD-1 or PD-L1 inhibitors in cisplatin-ineligible patients with advanced urothelial carcinoma (EV-201): a multicentre, single-arm, phase 2 trial. Lancet Oncol. 2021;22(6):872–82. https://doi.org/10.1016/S1470-2045(21)00094-2.

Lacouture ME, Patel AB, Rosenberg JE, O'Donnell PH. Management of Dermatologic Events Associated With the Nectin-4-directed Antibody-Drug Conjugate Enfortumab Vedotin. Oncologist. 2022;27(3):e223–32. https://doi.org/10.1093/oncolo/oyac001.

Nishihara K, Kurose H, Ito N, Ohnishi S, Hirano T, Suekane H, Watanabe K, Chikui K, Ueda K, Uemura KI, Nakiri M, Suekane S, Igawa T, [EFFICACY, AND SAFETY OF ENFORTUMAB VEDOTIN IN ADVANCED UROTHELIAL CARCINOMA TREATMENT. AN INITIAL EXPERIENCE IN A SINGLE INSTITUTION]. Nihon Hinyokika Gakkai Zasshi. 2024;115(1):21–7. https://doi.org/10.5980/jpnjurol.115.21. Japanese.

Nguyen MN, Reyes M, Jones SC. Postmarketing Cases of Enfortumab Vedotin-Associated Skin Reactions Reported as Stevens-Johnson Syndrome or Toxic Epidermal Necrolysis. JAMA Dermatol. 2021;157(10):1237–9. https://doi.org/10.1001/jamadermatol.2021.3450.

Suda F, Yamada T, Muto M, Korenaga Y. A case of toxic epidermal necrolysis-like cutaneous adverse reaction caused by enfortumab vedotin. Rinsho Hifuka (Japanese J Clin Dermatology). 2024;78(7):489–93. https://imis.igaku-shoin.co.jp/journal/412/78/7/1412207337/.

10.

Furubayashi N, Minato A, Tomoda T, Masaoka H, Hori Y, Kiyoshima K, Negishi T, Haraguchi Y, Koga T, Song Y, Harada K, Kuroiwa K, Seki N, Fujimoto N, Nakamura M. Uro-Oncology Group in Kyushu (UROKYU). Cutaneous and Renal Toxicities of Enfortumab Vedotin for Advanced Urothelial Carcinoma: The UROKYU Study. Anticancer Res. 2024;44(7):3025–32. https://doi.org/10.21873/anticanres.17115.

11.

Matsuzawa Y, Tokunaga K, Kotani K, Keno Y, Kobayashi T, Tarui S. Simple estimation of ideal body weight from body mass index with the lowest morbidity. Diabetes Res Clin Pract. 1990;10 DOI not available.

12.

Vlachou E, Matoso A, McConkey D, Jing Y, Johnson BA, Hahn NM, Hoffman-Censits J. Enfortumab Vedotin-related Cutaneous Toxicity and Radiographic Response in Patients with Urothelial Cancer: A Single-center Experience and Review of the Literature. Eur Urol Open Sci. 2023;49:100–3. https://doi.org/10.1016/j.euros.2023.01.002.

13.

Furubayashi N, Mochida M, Kijima A, Fujimoto Y, Nakamura M, Negishi T. Steroid Premedication Impact on Efficacy and Cutaneous Toxicity of Enfortumab Vedotin for Advanced Urothelial Carcinoma. In Vivo. 2025 May-Jun;39(3):1607–14. https://doi.org/10.21873/invivo.13961

14.

Vlachou E, Johnson BA, McConkey D, Jing Y, Matoso A, Hahn NM, Hoffman-Censits J. Enfortumab vedotin-related cutaneous toxicity correlates with overall survival in patients with urothelial cancer: a retrospective experience. Front Oncol. 2024;14:1377842. https://doi.org/10.3389/fonc.2024.1377842.

15.

Pharmaceuticals and Medical Devices Agency. (2017). Information on suspected adverse drug reaction reports. PMDA. Retrieved November 5, 2017.

16.

Watanabe H, Matsushita Y, Watanabe A, Maeda T, Nukui K, Ogawa Y, Sawa J, Maeda H. Early detection of important safety information. Recent methods for signal detection. Jpn J Biomet. 2004;25:37–60. https://doi.org/10.5691/jjb.25.37.

17.

Greenland S, Schwartzbaum JA, Finkle WD. Problems due to small samples and sparse data in conditional logistic regression analysis. Am J Epidemiol. 2000;151:531–9. https://doi.org/10.1093/oxfordjournals.aje.a010240.

18.

(MedDRA Japanese Maintenance Organization. Available online: https://www.meddra.org/ (accessed on 16 Sep 2025).&#12290.

19.

Hosoya R, Uesawa Y, Ishii-Nozawa R, Kagaya H. Analysis of factors associated with hiccups based on the Japanese Adverse Drug Event Report database. PLoS ONE 2017, 12, e0172057– https://doi.org/10.1371/journal.pone.0172057

20.

Kawabe A, Uesawa Y. Analysis of corticosteroid-induced glaucoma using the Japanese adverse drug event reporting database.Pharmaceuticals 2023, 16, 948. https://doi.org/10.3390/ph16070948

21.

Okunaka M, Kano D, Matsui R, Kawasaki T, Uesawa Y. Comprehensive analysis of chemotherapeutic agents that induce infectious neutropenia. Pharmaceuticals. 2021;14:681. https://doi.org/10.3390/ph14070681.

22.

Kan Y, Nagai J, Uesawa Y. Evaluation of antibiotic-induced taste and smell disorders using the FDA adverse event reporting system database. Sci Rep. 2021;11:9625. https://doi.org/10.1038/s41598-021-88958-2.

23.

Nakao Y, Asada M, Uesawa Y. Comprehensive study of drug-induced pruritus based on adverse drug reaction report database.Pharmaceuticals 2023, 16, 1500. https://doi.org/10.3390/ph16101500

24.

Chen JJ, Wang SJ, Tsai CA, Lin CJ. Selection of differentially expressed genes in microarray data analysis. Pharmacogenom J 2007, 7, 212–20. https://doi.org/10.1038/sj.tpj.6500412

25.

Liu Q, Cui Z, Deng C, Yang C, Shi T. A real-world pharmacovigilance analysis of adverse events associated with irbesartan using the FAERS and JADER databases. Front Pharmacol. 2024;15:1485190. https://doi.org/10.3389/fphar.2024.1485190.

26.

Lu Z, Suzuki A, Wang D. Statistical methods for exploring spontaneous adverse event reporting databases for drug-host factor interactions. BMC Med Res Methodol. 2023;23(1):71. https://doi.org/10.1186/s12874-023-01885-w.

27.

Dobner J, Kaser S. Body mass index and the risk of infection - from underweight to obesity. Clin Microbiol Infect. 2018;24(1):24–8. DOI not available.

28.

Phung DT, Wang Z, Rutherford S, Huang C, Chu C. Body mass index and risk of pneumonia: a systematic review and meta-analysis. Obes Rev. 2013;14(10):839–57. https://doi.org/10.1111/obr.12055.

29.

Bourke CD, Berkley JA, Prendergast AJ. Immune Dysfunction as a Cause and Consequence of Malnutrition. Trends Immunol. 2016;37(6):386–98. https://doi.org/10.1016/j.it.2016.04.003.

30.

Jansåker F, Forsberg PO, Li X, Sundquist K. The relation of body mass index, body height, and parity to pyelonephritis: A nationwide population-based cohort study of over one million parous women (1997–2018). Int J Infect Dis. 2022;125:67–73. https://doi.org/10.1016/j.ijid.2022.10.022.

31.

Wang HW, Hochberg AM, Pearson RK, Hauben M. An experimental investigation of masking in the US FDA adverse event reporting system database. Drug Saf. 2010;33:1117–33. https://doi.org/10.2165/11584390-000000000-00000.

32.

Pariente A, Avillach P, Salvo F, Thiessard F, Miremont-Salamé G, Fourrier-Reglat A, Haramburu F, Bégaud B. Moore, N.;Association Française des Centres Régionaux de Pharmacovigilance (CRPV). Effect of competition bias in safety signal generation:Analysis of a research database of spontaneous reports in France. Drug Saf. 2012;35:855–64. https://doi.org/10.1007/BF03261981.

33.

Poleksic A, Xie L. Database of adverse events associated with drugs and drug combinations. Sci Rep. 2019;9:20025. https://doi.org/10.1038/s41598-019-56525-5.

Yes