ABSTRACT
Background and Aims
Apalutamide (APA) and enzalutamide (ENZ) are androgen receptor pathway inhibitors widely used for prostate cancer. Skin rash is a characteristic adverse event of APA, but how often it occurs in real‐world practice, the factors related to it, and possible differences by disease stage are not well understood.
Methods
Using a Japanese nationwide claims database (Medical Data Vision), we identified patients with prostate cancer treated with APA (240 mg/day) or ENZ (160 mg/day) between April 2008 and August 2021. A skin‐related adverse event was defined as a case with both a skin‐related diagnosis and treatment for the skin event during therapy. Patients were classified as mCSPC or CRPC. We compared the rate and timing of skin rash between the two drugs and between disease stages, and examined associated clinical factors.
Results
A total of 5524 patients were included (1264 APA and 4260 ENZ). The rate of skin rash was higher in the APA group than in the ENZ group (25.3% vs. 9.95%; p < 0.001), confirmed by Kaplan–Meier analysis. Within each drug, the rate did not differ significantly between mCSPC and CRPC. Most events occurred within the first month (median 29 days for APA and 22 days for ENZ). Age, body weight, BMI, and BSA were not associated with skin rash. After adjustment for age, body weight, BMI, and BSA, the risk remained about 2.6‐fold higher with APA (adjusted HR: 2.56, 95% CI: 2.18–3.01).
Conclusion
Skin‐related adverse events requiring treatment were more common with APA than with ENZ, but they did not differ by disease stage for either drug and were not predicted by body size. Careful skin monitoring and early intervention are important for all patients from the start of APA or ENZ treatment, regardless of disease stage.
Keywords: apalutamide, enzalutamide, prostate cancer, real‐world database, skin rash, skin‐related adverse events
1. Introduction
Prostate cancer is one of the most common cancers in men, and its incidence is increasing in Japan [1]. Advanced prostate cancer, which is mainly treated with drug therapy, can affect survival, and as the number of patients grows, its treatment is becoming more important [2]. In recent years, drug therapy for advanced prostate cancer has improved greatly, and the arrival of androgen receptor pathway inhibitors (ARPIs) has expanded treatment options. Among ARPIs, enzalutamide (ENZ) and apalutamide (APA) are widely used in disease stages such as metastatic castration‐sensitive prostate cancer (mCSPC) and castration‐resistant prostate cancer (CRPC) [3]. The two drugs have similar chemical structures and share the same mechanism of action, blocking the androgen receptor pathway, but some differences in their side‐effect profiles have been reported [4]. One feature of APA is the high rate of skin rash. In the phase III trials of APA, skin rash was reported in 27.1% of patients in the TITAN trial and 23.8% in the SPARTAN trial [5, 6]. In Japanese patients, this is even more notable, and the rate of skin rash is said to reach about 50% [7].
Several reports on APA‐related skin rash have also been published in Japan. Some reported that low body weight (BW) or low body mass index (BMI) were related factors [8], while other studies could not identify any related factor [9], so the results are not consistent. Therefore, the predictive factors for skin‐related adverse events in real‐world practice are still unclear. In addition, large‐scale real‐world data that include elderly or frail patients—who are often excluded from clinical trials—are scarce, and reports that use a large database to compare the rate of skin rash between CSPC and CRPC, or to compare APA with other ARPIs, are also limited.
Skin rash with APA is a clinically important adverse event that can affect dose adjustment, treatment continuation, and patients' quality of life (QOL). However, its occurrence and related factors in real‐world practice have not been studied enough.
Therefore, in this study, we used a large Japanese database to compare the rate of skin rash between patients treated with APA and ENZ, and to examine how skin rash occurs according to disease status (CSPC and CRPC).
2. Materials and Methods
2.1. Ethical Approval
This study was approved by the Institutional Review Board of Yokohama City University Medical Center (approval number F240500030) and was conducted in accordance with the Declaration of Helsinki. Because this was a retrospective analysis of anonymized administrative claims data, the requirement for written informed consent was waived, and an opt‐out approach was used. This observational study was reported in accordance with the STROBE and RECORD guidelines for studies using routinely collected health data.
2.2. Data Source and Study Population
This study used a Japanese nationwide healthcare claims database that covers about 25.6% of Diagnosis Procedure Combination (DPC) hospitals in Japan. The data were provided by Medical Data Vision (MDV; Tokyo, Japan). Between April 2008 and August 2021, we identified 413,427 patients diagnosed with prostate cancer based on ICD‐10 codes and treated with ARPIs or bicalutamide. From these, we then selected patients treated with APA or ENZ. Patients treated with bicalutamide or darolutamide were not included.
The MDV database is one of the largest hospital‐based administrative claims databases in Japan. It compiles standardized diagnosis, procedure, and prescription records from acute‐care hospitals that use the DPC system, and at the time of analysis, it covered more than 20 million patients from over 400 hospitals across all regions of Japan. Diagnoses are coded with ICD‐10 together with standardized Japanese disease names, and drugs are recorded with national receipt (claims) codes. Because the database is anonymized and hospital‐based, patients cannot be followed after transfer to a non‐participating hospital.
In total, 1781 patients received APA, and 17,254 received ENZ. To reduce the effect of dose modification at treatment initiation, only patients who started treatment at the standard approved dose were included. In other words, patients who started APA at 240 mg/day and ENZ at 160 mg/day were analyzed. Finally, 1264 APA‐treated patients and 4260 ENZ‐treated patients were included in this study (Figure 1).
Figure 1.
Flow chart of patient selection.
Patients were then classified according to disease status. Patients with a diagnosis of CRPC were classified as the CRPC group. Patients without a diagnosis of CRPC but with metastatic prostate cancer or bone metastasis were classified as the mCSPC group.
2.3. Definition of Skin‐Related Adverse Events
Skin‐related adverse events were defined as cases in which both a skin‐related diagnosis and treatment for the skin event were recorded during APA or ENZ therapy. Treatments for skin‐related adverse events included antihistamines, topical corticosteroids, and systemic corticosteroids (oral or injectable). Because claims databases do not show which disease a drug was prescribed for, combining diagnosis records with treatment records was used to more reliably identify skin‐related adverse events that required treatment.
The onset date of a skin‐related adverse event was defined as the earlier of two dates: the date of the skin‐related diagnosis and the date on which treatment for the skin event began, provided that both events occurred during the treatment period. Systemic corticosteroids were counted as treatment for skin‐related adverse events only when they were started on or after the date of the skin‐related diagnosis.
The specific skin‐related diagnoses (with their ICD‐10 classifications) and the medications used to define treatment for a skin‐related event are listed in Supporting Information S1: Table S1 and Supporting Information S2: Table S2. Throughout this paper, we use the term “skin‐related adverse event” for any event captured by this claims‐based definition, because the definition may include dermatologic events other than rash; the term “skin rash” is used only when referring specifically to rash.
In the primary analysis, patients who had skin disease before treatment were not excluded. To examine whether pre‐existing skin disease affected the results, we performed a sensitivity analysis that excluded patients with any skin‐related diagnosis recorded within 180 days before starting APA or ENZ.
2.4. Study Variables and Endpoints
The following variables were evaluated: age, height, BW, BMI, body surface area (BSA), the rate of skin‐related adverse events, use of antihistamines, topical corticosteroids, and systemic corticosteroids, time to the start of each treatment, and the need for inpatient treatment.
The primary endpoint was the rate of skin‐related adverse events in patients treated with APA and ENZ. Secondary endpoints included a comparison of skin rash rates between mCSPC and CRPC, assessment of the timing of rash onset, and analysis of clinical factors related to skin rash.
2.5. Statistical Analyses
Continuous variables were analyzed using the Mann–Whitney U‐test, and categorical variables were compared using the chi‐square test. Time to skin rash onset was analyzed using Kaplan–Meier curves and compared using the log‐rank test. Statistical analyses were performed using GraphPad Prism software (GraphPad Software, La Jolla, CA, USA). p values < 0.05 were considered statistically significant.
Baseline characteristics were compared using the Mann–Whitney U‐test for continuous variables and the chi‐square test for categorical variables. To judge the clinical relevance of baseline differences despite the large sample size, we also calculated the standardized mean difference (SMD) for each continuous covariate; an absolute SMD below 0.1 was regarded as negligible. The cumulative incidence of skin‐related adverse events was estimated with the Kaplan–Meier method and compared with the log‐rank test; patients were censored at the end of the treatment period (the last date of APA or ENZ supply), so that the differing observation periods between groups and between disease stages were accounted for in the time‐to‐event analysis. To evaluate the association between treatment (APA vs. ENZ) and skin‐related adverse events with adjustment for baseline covariates, we fitted multivariable Cox proportional hazards models including age, BW, BMI, and BSA. Because BW, BMI, and BSA are mathematically interdependent, these body‐size measures were entered both separately (each with age) and together, and all models are reported. Hazard ratios (HRs) are presented with 95% confidence intervals (CIs). The comparison of skin‐related adverse events between APA and ENZ was pre‐specified; the subgroup analyses by disease stage (mCSPC vs. CRPC) and the multivariable and sensitivity analyses were exploratory. All tests were two‐sided, and p < 0.05 was considered statistically significant. p values are reported following standard conventions (p < 0.001; to the nearest thousandth between 0.001 and 0.01; and to the nearest hundredth for values ≥ 0.01). Analyses were performed with GraphPad Prism (version 10; GraphPad Software, La Jolla, CA, USA) and Python (version 3.12; SciPy and lifelines packages). Reporting followed the SAMPL guidelines.
3. Results
3.1. Patients' Cohort
A total of 5524 patients were included in this study, of whom 1264 received APA and 4260 received ENZ (Figure 1). The baseline characteristics of the two treatment groups are summarized in Table 1. Patients treated with APA were younger than those treated with ENZ; the median (mean ± SD) age was 73 years (73.2 ± 7.7) in the APA group and 75 years (74.3 ± 7.8) in the ENZ group. BMI and BSA were also slightly higher in the APA group than in the ENZ group. The rate of skin rash was higher in the APA group than in the ENZ group (25.3% vs. 9.95%). Within each treatment group, the rate was similar between patients with mCSPC and those with CRPC (27.4% vs. 24.6% in the APA group and 7.0% vs. 10.5% in the ENZ group). Systemic corticosteroids were given more often in the APA group than in the ENZ group (6.01% vs. 2.49%). Skin rash usually appeared within the first month after treatment initiation, with a median time to onset of 29 days in the APA group and 22 days in the ENZ group (Table 2). Although several baseline differences reached statistical significance, all absolute SMDs were 0.14 or less, indicating that the differences were small in magnitude (Table 1). The between‐group differences in the rate of skin rash and in systemic corticosteroid use were statistically significant (both p < 0.001; Table 2).
Table 1.
Baseline characteristics of patients treated with APA and ENZ.
| Variable | APA (n = 1264) | ENZ (n = 4260) | p value | SMD |
|---|---|---|---|---|
| Age (years) | 73 (73.2 ± 7.7) | 75 (74.3 ± 7.8) | < 0.001 | 0.14 |
| Height (m) | 1.64 (1.64 ± 0.07) | 1.64 (1.64 ± 0.07) | 0.047 | 0.00 |
| Weight (kg) | 62.8 (63.8 ± 10.6) | 61.9 (62.3 ± 10.8) | < 0.001 | 0.14 |
| BMI (kg/m2) | 23.4 (23.6 ± 3.4) | 23.1 (23.2 ± 3.4) | < 0.001 | 0.12 |
| BSA (m2) | 1.69 (1.69 ± 0.15) | 1.67 (1.67 ± 0.15) | < 0.001 | 0.13 |
Note: Values are median (mean ± SD). Continuous variables were compared using the Mann–Whitney U‐test. SMD was calculated as the difference in means divided by the pooled standard deviation; an absolute SMD < 0.1 indicates a negligible imbalance. Although several differences reached statistical significance owing to the large sample size, all absolute SMDs were ≤ 0.14, indicating that between‐group differences were small in magnitude.
Abbreviations: APA, apalutamide; BMI, body mass index; BSA, body surface area; ENZ, enzalutamide; SD, standard deviation; SMD, standardized mean difference.
Table 2.
Incidence of skin‐related adverse events, treatment, and timing with APA and ENZ.
| APA All (n = 1264) | APA CSPC (n = 325) | APA CRPC (n = 939) | ENZ All (n = 4260) | ENZ CSPC (n = 661) | ENZ CRPC (n = 3599) | p value (APA vs. ENZ, All) | |
|---|---|---|---|---|---|---|---|
| Skin‐related adverse event | 320 (25.3%) | 89 (27.4%) | 231 (24.6%) | 424 (9.95%) | 46 (7.0%) | 378 (10.5%) | < 0.001 |
| Systemic corticosteroid for skin event | 76 (6.01%) | 21 (6.46%) | 55 (5.86%) | 106 (2.49%) | 11 (1.66%) | 95 (2.64%) | < 0.001 |
| Time to onset (days) | 29 (55.8 ± 67.9) | 36 (55.5 ± 58.5) | 28 (55.8 ± 70.9) | 22 (56.8 ± 97.0) | 23 (48.0 ± 79.4) | 22 (57.9 ± 98.8) | — |
| Observation period (days) | 203 (240 ± 192) | 154 (175 ± 140) | 224 (264 ± 203) | 210 (257 ± 213) | 99.5 (175 ± 191) | 230 (273 ± 214) | — |
Note: Values are n (%) or median (mean ± SD). Proportions were compared using the chi‐square test; continuous variables were compared using the Mann–Whitney U‐test. p values for the categorical rows compare APA (All) versus ENZ (All).
Abbreviations: APA, apalutamide; CRPC, castration‐resistant prostate cancer; CSPC, castration‐sensitive prostate cancer; ENZ, enzalutamide.
3.2. Incidence and Timing of Skin Rash
Kaplan–Meier analysis showed that the cumulative rate of skin rash was significantly higher in the APA group than in the ENZ group (Figure 2). In contrast, within each treatment group, there was no significant difference in the cumulative rate of skin rash between patients with mCSPC and those with CRPC. In both treatment groups, skin rash occurred mainly during the first month after treatment initiation (Table 2).
Figure 2.
Kaplan–Meier curves for skin rash‐free probability in the apalutamide (APA) and enzalutamide (ENZ) groups, shown separately for the mCSPC and CRPC subgroups. Hazard ratios and log‐rank p values are shown for the comparison between APA and ENZ and for the comparisons between mCSPC and CRPC within each treatment group.
3.3. Multivariable and Sensitivity Analyses
To determine whether the higher risk of skin‐related adverse events with APA simply reflected baseline differences in age or body size, we performed multivariable Cox proportional hazards analyses (Table 3). In the unadjusted model, the hazard of a skin‐related adverse event was about 2.7‐fold higher with APA than with ENZ (HR: 2.705, 95% CI: 2.339–3.128, p < 0.001). After adjustment for age and body size, the estimate was essentially unchanged, whether age was combined with BW, BMI, or BSA separately (HR: 2.56 in each model) or with all three together (HR: 2.562, 95% CI: 2.183–3.006, p < 0.001). In the fully adjusted model, none of age, BW, BMI, or BSA was significantly associated with skin‐related adverse events (Table 4). Thus, the higher risk with APA was independent of age and body size.
Table 3.
Multivariable Cox proportional hazards analyses for skin‐related adverse events (APA vs. ENZ).
| Model | Covariates included | Adjusted HR (APA vs. ENZ) | 95% CI | p value |
|---|---|---|---|---|
| Unadjusted | Treatment only | 2.705 | 2.339–3.128 | < 0.001 |
| Model 1 | Age + body weight | 2.560 | 2.183–3.002 | < 0.001 |
| Model 2 | Age + BMI | 2.566 | 2.187–3.010 | < 0.001 |
| Model 3 | Age + BSA | 2.564 | 2.185–3.008 | < 0.001 |
| Model 4 | Age + body weight + BMI + BSA | 2.562 | 2.183–3.006 | < 0.001 |
Note: Reference group: ENZ. After adjustment for age and body‐size measures, the hazard of skin‐related adverse events remained about 2.6‐fold higher with APA than with ENZ, essentially unchanged from the unadjusted estimate.
Abbreviations: CI, confidence interval; HR, hazard ratio.
Table 4.
Fully adjusted model (covariate estimates).
| Variable | Adjusted HR | 95% CI | p value |
|---|---|---|---|
| Apalutamide versus enzalutamide | 2.562 | 2.183–3.006 | < 0.001 |
| Age, per 1 year | 0.999 | 0.988–1.010 | 0.82 |
| Body weight, per 1 kg | 0.970 | 0.842–1.118 | 0.68 |
| BMI, per 1 kg/m2 | 1.035 | 0.866–1.238 | 0.70 |
| BSA, per 1 m2 | 4.557 | 0.005–4073.926 | 0.66 |
Note: Body weight, BMI, and BSA are mathematically interdependent; entering them simultaneously (Model 4) introduces collinearity, which widens the confidence intervals for these individual coefficients (notably BSA). The estimate for treatment (APA vs. ENZ) is stable across all models (Models 1–4), indicating that the association is robust and not explained by differences in age or body size.
In the sensitivity analysis excluding patients with a skin‐related diagnosis within 180 days before treatment (APA, 12; ENZ, 65), the higher risk with APA persisted (HR: 2.936, 95% CI: 2.528–3.410, p < 0.001; fully adjusted HR: 2.785, 95% CI: 2.361–3.284, p < 0.001), consistent with the primary analysis (Table 5).
Table 5.
Sensitivity analysis excluding pre‐existing skin disease (180‐day washout).
| Analysis | APA, n | ENZ, n | Events | HR (APA vs. ENZ) | 95% CI | p value |
|---|---|---|---|---|---|---|
| Primary cohort | 1264 | 4260 | 744 | 2.705 | 2.339–3.128 | < 0.001 |
| 180‐day washout | 1252 | 4195 | 695 | 2.936 | 2.528–3.410 | < 0.001 |
Note: Patients with any skin‐related diagnosis recorded within 180 days before starting APA or ENZ were excluded (APA, 12; ENZ, 65). In the fully adjusted model (age + body weight + BMI + BSA), the washout cohort gave an adjusted HR of 2.785 (95% CI: 2.361–3.284, p < 0.001). Results were consistent with the primary analysis.
3.4. Association Between Clinical Factors and Skin Rash
Age, BW, BMI, and BSA were not related to the rate of skin rash in either the APA or ENZ group. Similar results were observed in the mCSPC and CRPC subgroups (Figures 3 and 4). No clear relationship was seen between skin rash and age, BW, BMI, or BSA in either treatment group or in the mCSPC and CRPC subgroups (Figures 3 and 4).
Figure 3.
Box plots of age, body weight, body mass index (BMI), and body surface area (BSA) in patients with (rash+) and without (rash−) a skin‐related adverse event in the APA group, shown for (a) all patients, (b) the mCSPC subgroup, and (c) the CRPC subgroup. Boxes indicate the interquartile range, center lines the median, and whiskers the range; open circles indicate the mean and error bars the 95% confidence interval of the mean. p values are from the Mann–Whitney U‐test.
Figure 4.
Box plots of age, body weight, BMI, and BSA in patients with (rash+) and without (rash−) a skin‐related adverse event in the ENZ group, shown for (a) all patients, (b) the mCSPC subgroup, and (c) the CRPC subgroup. Plot conventions are as in Figure 3.
4. Discussion
In this study, we used a nationwide claims database of Japanese prostate cancer patients to examine skin‐related adverse events with APA and ENZ. We found that the rate of skin‐related adverse events was significantly higher in the APA group than in the ENZ group, and this trend was consistent in both CSPC and CRPC stages. On the other hand, in both the APA and ENZ groups, there was no significant difference in the rate of skin‐related adverse events between CSPC and CRPC, and patient factors such as age, BW, BMI, and BSA were not related to onset. In addition, most skin‐related adverse events occurred early after the start of treatment. An important finding of this study is that it used a large database to show the features of APA‐ and ENZ‐related skin adverse events in Japanese real‐world practice. Importantly, the higher risk of skin‐related adverse events with APA remained after adjustment for age, BW, BMI, and BSA in multivariable Cox models, and also after excluding patients with pre‐existing skin disease, indicating that the difference was associated with the drug itself rather than with baseline differences between the groups.
Skin‐related adverse events during cancer drug therapy are important side effects that affect treatment continuation and patients' QOL. Although most APA‐related skin adverse events are Grade 1 or 2, if they are found or managed late and become severe, they may lead to treatment interruption, dose reduction, or even treatment discontinuation [10]. In fact, in the TITAN trial, skin rash led to treatment interruption in 8.4%, dose reduction in 5.3%, and discontinuation in 2.3% of patients [5]. In a real‐world report of 303 patients, 6.6% needed treatment interruption, 2.3% dose reduction, and 1.7% discontinuation [11]. Furthermore, skin adverse events during cancer treatment are known to lower patients' QOL not only through physical discomfort but also through changes in appearance that reduce body image and self‐esteem and affect social activities [12]. Therefore, from the viewpoints of both treatment continuation and QOL, APA‐related skin adverse events are important events that require proper monitoring and early intervention.
In this study, the rate of skin‐related adverse events in the APA group was 25.3%, which was roughly the same as in the TITAN trial (27.1%) and the SPARTAN trial (23.8) [5, 6]. In contrast, Japanese subgroup analyses have reported a higher rate of about 50% [7, 13]. One reason why the rate in Japanese patients was lower in our study than in these reports may be the difference in the definition of skin‐related adverse events. In this study, we defined a skin‐related adverse event as a case in which a skin‐related diagnosis was recorded, and drug treatment for the skin event was given. In claims data, it is difficult to know exactly which disease a drug was prescribed for from the prescription record alone. In fact, antihistamines and corticosteroids are used for many diseases other than skin diseases, so prescription records alone may not identify skin adverse events related to APA or ENZ. Therefore, we combined the skin‐related diagnosis with the treatment record to make it more likely that a case reflected treatment for a skin adverse event. However, with this definition, mild cases that were observed without drug treatment, and cases that had only a diagnosis or only drug treatment, were not included. Therefore, this study did not evaluate all skin‐related adverse events; the results should be interpreted as mainly reflecting cases that were diagnosed and required drug treatment in real‐world practice. In addition, systemic corticosteroids were needed in about 6% of the APA group and about 2.5% of the ENZ group. The rate of systemic corticosteroid use in the APA group was close to the rate of Grade ≥ 3 skin rash reported in the TITAN and SPARTAN trials (5.2%–6.3%). Although we could not directly evaluate the severity of skin adverse events, cases that needed systemic corticosteroids may reflect relatively severe skin adverse events. However, because the severity of skin events cannot be graded from claims data, this interpretation should be made with caution.
In this study, the median time to onset of APA‐related skin adverse events was 29 days, which was somewhat earlier than in previous reports. In the SPARTAN and TITAN trials, the median time to skin rash onset was 82 and 81 days, respectively [5, 6], and Japanese real‐world studies have reported about 2 months [14, 15]. In contrast, in our study, most cases occurred within the first month after the start of treatment, and most were seen within 100 days. Although differences in the definition of onset date and in analysis methods may have had an effect, the trend that APA‐related skin adverse events occur mostly within 3 months after the start of treatment was consistent with previous reports. Therefore, it is important to watch skin symptoms carefully for at least 3 months after starting APA, especially during the first month, and to start early intervention when needed.
In this study, we analyzed only patients who started APA at 240 mg/day, but age, BW, BMI, and BSA were not related to skin adverse events. In contrast, Sasaki et al. [8], Katsuta et al. [14], and Urabe et al. [15] reported that low BW, low BMI, and older age, respectively, may be risk factors for APA‐related skin adverse events. On the other hand, Tohi et al. [9] did not identify any significant risk factor, so the results are not consistent. The reasons why our results differ from previous reports may include differences in the patient population and sample size, as well as differences in the definition of skin adverse events. Our study focused on skin adverse events that required drug treatment, so we did not necessarily compare the same population as previous reports that evaluated all skin adverse events, including mild cases. Therefore, it is difficult to predict the risk of skin adverse events based on body size alone, and it is important to watch skin symptoms carefully in all patients from early after the start of treatment.
In this study, the rate of skin adverse events in the APA group was 27.4% in CSPC and 24.6% in CRPC, with no significant difference between the two groups. In the ENZ group as well, there was no significant difference between CSPC and CRPC. This result suggests that the occurrence of skin adverse events may depend more on the drug itself than on the disease status. In fact, all‐grade rash was reported in 27.1% in the TITAN trial (mCSPC) and 23.8% in the SPARTAN trial (nmCRPC); although the target diseases were different, the rates were fairly close [5, 6]. Moreover, in a pooled analysis of the SPARTAN, TITAN, and phase I trials in Japanese patients, APA‐related skin adverse events were reported in 56% of the SPARTAN Japanese subgroup and 50% of the TITAN Japanese subgroup, showing a high rate in both nmCRPC and mCSPC [13]. Japanese real‐world studies also reported high rates regardless of disease stage: 38.7% in the nmCRPC study by Katsuta et al. [14], 40% in the mCSPC study by Urabe et al. [15], and 56% in the study by Sasaki et al. [8], which included nmCRPC and mCSPC. Our results agree with these reports and support the idea that APA‐related skin adverse events do not depend on disease stage but are events that should be watched in all patients receiving APA. In addition, the same trend was seen for ENZ in our study, and no difference by disease stage was found for ENZ either. Taken together, these findings suggest that the occurrence of skin adverse events may depend more on the drug itself than on disease stage, and that proper skin monitoring is important from early after the start of treatment for both APA and ENZ, regardless of disease stage.
This study has several limitations. First, because it was based on claims data, we could not obtain detailed clinical information such as PSA level, Gleason score, TNM classification, site and volume of metastasis, and performance status, so we could not fully adjust for background factors that may affect skin adverse events. Second, we could not directly evaluate the severity, shape, or extent of the skin rash. Third, because we included only cases that met both a diagnosis and drug treatment, mild cases that did not need drug treatment, and cases with only a diagnosis or only drug treatment were not included. Therefore, the results of this study should be interpreted as reflecting skin‐related adverse events that were diagnosed and required drug treatment in real‐world practice.
5. Conclusion
Using a large Japanese real‐world database, we found that the rate of skin‐related adverse events requiring treatment was significantly higher with APA than with ENZ. For both drugs, the rate did not differ between mCSPC and CRPC, and patient factors such as age, BW, BMI, and BSA were not related to onset. Most events occurred early, mainly within the first month after starting treatment.
Author Contributions
Akihito Hashizume: conceptualization, methodology, writing – original draft. Takashi Kawahara: conceptualization, methodology, writing – original draft, funding acquisition. Masanobu Yamazaki: investigation, data curation. Daiji Takamoto: investigation. Teppei Takeshima: Investigation. Shinnosuke Kuroda: investigation. Hiroji Uemura: writing – review and editing. Kazuhide Makiyama: writing – review and editing. Jun‐Ichi Teranishi: writing – review and editing. All authors have read and approved the final version of the manuscript. Takashi Kawahara had full access to all of the data in this study and takes complete responsibility for the integrity of the data and the accuracy of the data analysis.
Funding
The authors have nothing to report.
Ethics Statement
This study was approved by the Institutional Review Board of Yokohama City University Medical Center (approval number F240500030) and was conducted in accordance with the Declaration of Helsinki. Because this was a retrospective study using anonymized administrative claims data, the requirement for written informed consent was waived, and an opt‐out approach was adopted.
Conflicts of Interest
The authors declare no conflicts of interest.
Use of Generative AI and AI‐Assisted Technologies
During the preparation of this work, the authors used ChatGPT (OpenAI), Claude (Anthropic), and DeepL for English language editing. In addition, ChatGPT (OpenAI) was used to assist in writing the code for the multivariable Cox proportional hazards analysis and figures. After using these tools, the authors reviewed and verified all outputs and edited the content as needed, and take full responsibility for the content of the publication.
Transparency Statement
The lead author, Takashi Kawahara, affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained.
Supporting information
Table S1: Skin‐related diagnoses and their ICD‐10 classifications used to define skin‐related adverse events.
Table S2: Medications used to define treatment for skin‐related adverse events.
Acknowledgments
This research received no specific grant from any funding agency in the public, commercial, or not‐for‐profit sectors. As there was no funding source, no funder had any role in the study design; the collection, analysis, or interpretation of data; the writing of the report; or the decision to submit the report for publication.
Data Availability Statement
The data that support the findings of this study were obtained from Medical Data Vision (MDV) under a data use agreement and are therefore not publicly available. Derived data supporting the tables and figures of this study are available from the corresponding author upon reasonable request.
References
- 1. Sasaki T., Higashi T., and Inoue T., “Urological Cancer Statistics on Incidence From 1975 to 2019 and Mortality From 1958 to 2022 in Japan,” International Journal of Clinical Oncology 29, no. 8 (2024): 1088–1095. [DOI] [PubMed] [Google Scholar]
- 2. Siegel D. A., O'Neil M. E., Richards T. B., Dowling N. F., and Weir H. K., “Prostate Cancer Incidence and Survival, by Stage and Race/Ethnicity – United States, 2001–2017,” MMWR. Morbidity and Mortality Weekly Report 69, no. 41 (2020): 1473–1480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Spratt D. E., Srinivas S., Adra N., et al., “Prostate Cancer, Version 3.2026, NCCN Clinical Practice Guidelines in Oncology,” Journal of the National Comprehensive Cancer Network 23, no. 11 (2025): 469–493. [DOI] [PubMed] [Google Scholar]
- 4. Ji C., Guha M., Zhu X., et al., “Enzalutamide and Apalutamide: In Vitro Chemical Reactivity Studies and Activity in a Mouse Drug Allergy Model,” Chemical Research in Toxicology 33, no. 1 (2020): 211–222. [DOI] [PubMed] [Google Scholar]
- 5. Chi K. N., Agarwal N., Bjartell A., et al., “Apalutamide for Metastatic, Castration‐Sensitive Prostate Cancer,” New England Journal of Medicine 381, no. 1 (2019): 13–24. [DOI] [PubMed] [Google Scholar]
- 6. Smith M. R., Saad F., Chowdhury S., et al., “Apalutamide Treatment and Metastasis‐Free Survival in Prostate Cancer,” New England Journal of Medicine 378, no. 15 (2018): 1408–1418. [DOI] [PubMed] [Google Scholar]
- 7. Uemura H., Arai G., Uemura H., et al., “Apalutamide for Metastatic, Castration‐Sensitive Prostate Cancer in the Japanese Population: A Subgroup Analysis of the Randomized, Double‐Blind, Placebo‐Controlled Phase 3 TITAN Study,” International Journal of Urology 28, no. 3 (2021): 280–287. [DOI] [PubMed] [Google Scholar]
- 8. Sasaki D., Hatakeyama S., Tanaka T., Okamoto T., Yoneyama T., and Ohyama C., “Impact of Body Size on Skin‐Related Adverse Events in Advanced Prostate Cancer Treated With Apalutamide: A Multicenter Retrospective Study,” International Journal of Urology 29, no. 7 (2022): 772–773. [DOI] [PubMed] [Google Scholar]
- 9. Tohi Y., Kato T., Fukuhara H., et al., “Real‐World Analysis of Apalutamide‐Associated Skin Adverse Events in Japanese Patients With Advanced Prostate Cancer: A Multi‐Institutional Study in the Chu‐Shikoku Japan Urological Consortium,” International Journal of Clinical Oncology 27, no. 8 (2022): 1348–1355. [DOI] [PubMed] [Google Scholar]
- 10. Birtle A. J., Formisano L., Descamps V., Weisenseel P., and Vilaseca A., “Early Identification and Management of Patients With Rash on Apalutamide,” Oncology and Therapy 12, no. 3 (2024): 609–620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Pan A., Reingold R. E., Zhao J. L., et al., “Dermatological Adverse Events in Prostate Cancer Patients Treated With the Androgen Receptor Inhibitor Apalutamide,” Journal of Urology 207, no. 5 (2022): 1010–1019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Belzer A., Pach J. J., Valido K., and Leventhal J. S., “The Impact of Dermatologic Adverse Events on the Quality of Life of Oncology Patients: A Review of the Literature,” American Journal of Clinical Dermatology 25, no. 3 (2024): 435–445. [DOI] [PubMed] [Google Scholar]
- 13. Uemura H., Koroki Y., Iwaki Y., et al., “Skin Rash Following Administration of Apalutamide in Japanese Patients With Advanced Prostate Cancer: An Integrated Analysis of the Phase 3 SPARTAN and TITAN Studies and a Phase 1 Open‐Label Study,” BMC Urology 20, no. 1 (2020): 139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Katsuta M., Kimura T., Tashiro K., et al., “Low Body Weight as a Risk Factor for Apalutamide‐Related Cutaneous Adverse Events,” Anticancer Research 42, no. 4 (2022): 2023–2028. [DOI] [PubMed] [Google Scholar]
- 15. Urabe F., Onuma H., Tashiro K., et al., “Apalutamide‐Induced Rash and Relative Dose Intensity in Metastatic Castration‐Sensitive Prostate Cancer: A Multicenter Japanese Cohort Study,” Translational Andrology and Urology 14, no. 11 (2025): 3554–3563. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Skin‐related diagnoses and their ICD‐10 classifications used to define skin‐related adverse events.
Table S2: Medications used to define treatment for skin‐related adverse events.
Data Availability Statement
The data that support the findings of this study were obtained from Medical Data Vision (MDV) under a data use agreement and are therefore not publicly available. Derived data supporting the tables and figures of this study are available from the corresponding author upon reasonable request.