Epidemiology and Treatment Outcomes of Voriconazole-Induced Invasive Cutaneous Squamous Cell Carcinoma in Japanese Patients with Pulmonary Aspergillosis

Abstract

Objectives:

Voriconazole is a broad-spectrum antifungal agent known to be associated with cutaneous squamous cell carcinoma (cSCC) in White organ transplant recipients; however, such cases are rarely reported in Asian populations. This study aimed to examine the epidemiology and treatment outcomes of Japanese patients without a history of organ transplantation who developed invasive cSCC during voriconazole therapy.

Methods:

This retrospective observational study analyzed data from six Japanese male patients who were treated at our hospital between the years 2013 and 2023. All patients had Fitzpatrick skin type IV and were undergoing long-term voriconazole therapy for pulmonary aspergillosis. The diagnosis of invasive cSCC was confirmed via skin biopsy. Clinical, pathological, and treatment data were collected and analyzed descriptively.

Results:

Fifteen facial lesions were identified (nine at initial presentation and six during follow-up). The median duration of voriconazole treatment before cSCC diagnosis was 66.5 months. Tumors had a median diameter of 17 mm, with an estimated growth rate of 6 mm per month. Most lesions demonstrated deep invasion (Clark level IV or V), and some exhibited perineural or perivascular infiltration. All lesions were surgically excised, except for one that was treated with radiotherapy. No regional recurrence or distant metastasis was observed during a median follow-up period of 9 months.

Conclusions:

Prolonged voriconazole therapy may increase the risk of aggressive cSCC in Japanese patients, even in the absence of prior transplantation. Early identification of phototoxic skin changes and prompt intervention are crucial to prevent deep tissue invasion and improve clinical outcomes.

Introduction

Voriconazole is a broad-spectrum antifungal agent belonging to the triazole class. It is widely used to treat invasive fungal infections caused by Aspergillus, Candida, Fusarium, and Scedosporium species. Clinical trials have shown that voriconazole is more efficacious and less toxic than amphotericin B for the treatment of invasive aspergillosis^1,2). Moreover, it is commonly used as a prophylactic antifungal therapy in patients undergoing hematopoietic stem cell or solid organ transplantation^3-6). Voriconazole is absorbed within 2 hours of oral administration, with an oral bioavailability exceeding 90%, allowing for flexible switching between oral and intravenous formulations as clinically appropriate⁷⁾.

An association has been reported between voriconazole use and the development of skin cancer, particularly among organ transplant recipients (OTRs), who are at increased risk due to immunosuppression^8-21). In this population, squamous cell carcinoma (SCC), and basal cell carcinoma account for approximately 95% of skin cancers, approximately 75% of which occur in sun-exposed areas such as the head, neck, and hands¹¹⁾. The presence of actinic keratoses, seborrheic keratoses, or human papillomavirus-related lesions further elevates the risk of cutaneous SCC (cSCC)¹⁶⁾.

However, most studies of voriconazole-associated cSCC have focused on White populations in Western countries, with limited data available for Asian populations. In a study of 412 OTRs, only six lesions in 5 of 27 Asian patients (1.5%) were diagnosed as skin cancer compared to the diagnosis of 370 lesions in 64 of 154 White patients (95.1%)²¹⁾. Asian individuals, particularly those with Fitzpatrick skin types III or IV, typically have higher melanin content, which provides some protection against ultraviolet (UV) radiation-induced skin damage, and may contribute to a lower overall incidence of skin cancer compared to that in Whites²¹⁾. Furthermore, Japan's low deceased organ donor rate (1.18 donors per million population in 2023, compared to 49.38 and 48.04 per million in Spain and the United States, respectively) may reflect the rarity of descriptions regarding voriconazole-induced cSCC²²⁾. Owing to its rarity, voriconazole-induced skin cancer can be misdiagnosed or overlooked, particularly in transplant recipients, wherein lesions may be mistaken for graft-versus-host disease²³⁾. In countries where organ transplantation is rare, such as Japan, physicians may unknowingly continue voriconazole therapy without recognizing its cutaneous side effects. Consequently, it remains underrepresented in Japan, with most studies being limited to investigations of actinic keratosis, noninvasive carcinoma, or keratoacanthoma^24-26). Given the scarcity of data on voriconazole-induced cSCC in Asian populations, especially among non-transplant-recipient patients, this study aimed to retrospectively evaluate the epidemiology, clinical features, and treatment outcomes of Japanese patients with voriconazole-induced invasive cSCC in the absence of organ transplantation.

Methods

Ethics statement

This study has been approved by the Institutional Review Board of our institution (approval number 4661). Informed consent for this observational study was obtained through an opt-out procedure. The research was conducted in accordance with the principles outlined in the Declaration of Helsinki and its later amendments.

Study design and setting

This was a retrospective observational study conducted at our institution. Clinical characteristics and treatment outcome data were collected from all patients who developed invasive cSCC potentially associated with voriconazole therapy between 2013 and 2023.

Participants and eligibility criteria

We included all Japanese patients who were treated for pulmonary aspergillosis with voriconazole therapy, either at our institution or at other hospitals, and who were subsequently treated for voriconazole-induced invasive cSCC at our hospital. The eligibility criteria included patients with cSCC exhibiting invasion to the dermis based on a histopathologic diagnosis, with no history of skin cancer (other than the cSCC in question) or organ transplantation (either hematopoietic stem cell or allogeneic bone marrow transplants) and no previous diagnosis of photosensitive disorders such as xeroderma pigmentosum or porphyria. Patients who were currently receiving long-term treatment with other phototoxic medications or immunosuppressive treatments (including steroids) and those undergoing anticancer treatment were excluded from the study.

Diagnosis and outcome assessment

All cases of cSCC were confirmed by pathologists via skin biopsy, which was performed within 2 weeks of the initial visit. Tumors were classified according to the criteria outlined in the eighth edition of the American Joint Committee on Cancer and the Union for International Cancer Control staging manual²⁷⁾. Accordingly, eyelid lesions were classified separately from other facial lesions based on the different criteria in the manual. Key outcome measures included changes in tumor size, the level of histopathologic invasion (Clark level), the presence of perineural or perivascular invasion, and tumor, node, metastasis (TNM) classifications, as these factors reflect the tumor characteristics and treatment outcomes of cSCC. The estimated tumor growth rate was calculated by dividing the diameter of the major axis by the time interval between each patient's initial awareness of the tumor nodule with elevation and the hospital visit. Recurrence, metastasis, and mortality were also evaluated based on periodic visits or diagnostic imaging conducted every 3 months during follow-up after the initial cSCC treatment.

Data collection and variables

Patient demographic data, clinical history, treatment details, and outcomes were obtained from electronic medical records. Relevant variables included the patients' age, skin type (Fitzpatrick classification²⁸⁾) (Table 1), duration and dosage of voriconazole treatment, tumor characteristics, surgical margins, and reconstruction methods. Imaging results from computed tomography (CT) or magnetic resonance imaging (MRI) were reviewed to guide treatment planning.

Table 1.

Fitzpatrick Classification of Skin Phototypes.

Type I	Type II	Type III	Type IV	Type V	Type VI
Always burn, never tan	Usually burn, tan less than average (with difficulty)	Sometimes mild burn, tan about average	Rarely burn, tan more than average (with ease)	Brown skin	Black skin

Treatment and bias control

Treatment decisions were based on tumor pathology, radiologic findings, and patient functional status based on the American Society of Anesthesiologists Physical Status (ASA-PS) classification criteria. All treatments for cSCC were administered within 1 month of histopathologic diagnosis to minimize delays and potential bias that could affect the outcomes due to treatment timing.

Sample size and statistical considerations

The study included all eligible patients identified during the defined study period. Due to the expected small sample size and the study's retrospective design, the ability to conduct formal statistical analyses was limited. Quantitative variables such as tumor size and drug exposure duration were summarized using medians and interquartile ranges (IQRs). Subgroup or interaction analyses were not conducted. Missing data were minimal and handled through manual verification of clinical records.

Results

Six patients met the inclusion criteria and were enrolled in this retrospective study. No patients were excluded during the study period, and complete demographic, clinical, and outcome data were available for all individuals. The patient case list is presented in the Supplementary Material. Key demographic and clinical characteristics are summarized in Table 2.

Table 2.

Demographic and Clinical Characteristics of the Study Participants.

Age, median (IQR), years	78 (73.5-78.8)
Sex, n (%)
Male	6 (100)
Female	0 (0)
Fitzpatrick skin type, n (%)
I	0 (0)
II	0 (0)
III	0 (0)
IV	6 (100)
V	0 (0)
VI	0 (0)
ASA-PS score, n (%)
I	0 (0)
II	3 (50)
III	3 (50)
IV	0 (0)
V	0 (0)
Duration of voriconazole use, median (IQR), months	66.5 (51-73.8)
Dose of voriconazole, median (IQR), mg	400 (325-400)
Single/multiple cSCC lesions at the first visit, n (%)
Single	4 (66.7)
Multiple	2 (33.3)
Other skin tumors at the first visit, n (%)
Multiple actinic keratoses	4 (66.7)
Bowen’s disease	1 (16.7)
Follow-up period, median (IQR), months	9 (6.5-26.5)

ASA-PS: American Society of Anesthesiologists Physical Status; cSCC: cutaneous squamous cell carcinoma; IQR: interquartile range

All patients were Japanese men with Fitzpatrick skin type IV. The median age at the time of the first visit was 78 years (IQR: 73.5-78.8 years). The ASA-PS scores were evenly distributed, with three patients classified as grade II and three as grade III. The median duration of voriconazole therapy before presentation was 66.5 months (IQR: 51-73.8 months), with a median dose of 400 mg per day (IQR: 325-400 mg per day).

Following cSCC diagnosis, voriconazole therapy was discontinued in three patients. The remaining three patients continued treatment owing to the necessity of managing pulmonary aspergillosis and their overall poor clinical condition. At the initial presentation, two patients had multiple cSCC lesions. Four patients also presented with multiple actinic keratoses and one had Bowen's disease on the dorsum of the hand. The median follow-up duration was 9 months (IQR: 6.5-26.5 months).

The tumor characteristics are summarized in Table 3. At the first visit, nine cSCC lesions were identified, and six additional lesions were detected during follow-up, resulting in a total of 15 tumors treated during the study period. Lesions were most commonly located on the scalp (five patients), followed by the cheek (three patients), eyelid, ear, temple (two patients each), and lip (one patient).

Table 3.

Tumor Characteristics of cSCC.

Clinical characteristics of cSCC (n = 15)
Region, n (%)
Scalp	5 (33.3)
Cheek	3 (20)
Eyelid	2 (13.3)
Ear	2 (13.3)
Temple	2 (13.3)
Lip	1 (6.7)
Size, median (IQR), mm
Major axis	17 (9-29)
Minor axis	14 (5.5-24)
Estimated growth rate, median (IQR), mm/month	6 (5-9.5)
≥4 mm/month, n (%)	12 (80)
<4 mm/month, n (%)	3 (20)
Histopathologic feature of excised cSCC (n = 14)
Clark level, n (%)
I	0 (0)
II	0 (0)
III	0 (0)
IV	12 (85.7)
V	2 (14.3)
Degree of differentiation, n (%)
Well-differentiated	13 (92.9)
Moderately-differentiated	1 (7.1)
Poorly-differentiated	0 (0)
Invasion, n (%)
Perineural	2 (14.3)
Perivascular	1 (7.1)
T grade based on the TNM classification, n (%)
pT1	9 (64.3)
pT2	1 (7.1)
pT3	3 (21.4)
pT4	1 (7.1)

cSCC: cutaneous squamous cell carcinoma; IQR: interquartile range; TNM: tumor, node, metastasis

The median tumor size (diameter of major axis) was 17 mm (IQR: 9-29 mm), whereas the median diameter of minor axis was 14 mm (IQR: 5.5-24 mm). The median growth rate was 6 mm/month (IQR: 5-9.5 mm/month), with 12 lesions (80%) exhibiting growth rates of ≥4 mm/month (Figure 1).

Figure 1.

Box plot showing the estimated growth rate of cutaneous squamous cell carcinoma (cSCC) in patients treated with voriconazole. The estimated tumor growth rate was calculated by dividing the diameter of major axis of the lesion by the time interval between the patient’s awareness of the tumor nodule with elevation and the clinical visit. The median growth rate was 6 mm/month (interquartile range [IQR]: 5-9.5 mm/month), with 12 lesions (80%) exhibiting growth rates of ≥4 mm/month.

Twelve lesions were classified as Clark level IV and two as level V, indicating deep dermal invasion. Most tumors were well-differentiated (92.9%). Aggressive histologic features, including perineural invasion (n = 2) and perivascular invasion (n = 1), were observed in three lesions. Tumor invasion depth was staged pathologically as T1 in nine lesions, T2 in one lesion, T3 in three lesions, and T4 in one lesion due to cartilage involvement of the ear.

Treatment information is summarized in Table 4. Fourteen lesions were surgically excised, and histopathologic analysis was subsequently performed. Thirteen of these lesions were managed with surgery alone, whereas one was treated with adjuvant radiotherapy (60 Gy) owing to an insufficient surgical margin. An additional lesion involving extensive parotid and subcutaneous neck invasion was treated with radiotherapy alone (60 Gy) based on imaging findings consistent with a clinical T3 classification, despite the lack of confirmatory histopathology.

Table 4.

Treatment Details and Outcomes.

Type of treatment, n (%)
Surgical excision	13 (86.7)
Surgical excision with adjuvant radiotherapy (60 Gy)	1 (6.7)
Radiotherapy (60 Gy)	1 (6.7)
Horizontal surgical margin, median (IQR), mm	6 (4-6)
Reconstructive procedure after excision, n (%)
FTSG	8 (57.1)
Direct closure	4 (28.6)
Combination with skin flap and FTSG in two-stage procedure	1 (7.1)
Open treatment	1 (7.1)
Treatment outcome, n (%)
Regional recurrence
Yes	0 (0)
No	15 (100)
Lymph node or distant metastasis
Yes	0 (0)
No	6 (100)
Final staging based on TNM classification
I	1 (16.7)
II	0 (0)
III	4 (66.7)
IV	1 (16.7)
Survival status at follow-up
Survived	2 (33.3)
Deceased	4 (66.7)

FTSG: full-thickness skin graft; IQR: interquartile range; TNM: tumor, node, metastasis

The median horizontal surgical margin was 6 mm (IQR: 4-6 mm). Deep tissue barriers, including the fascia or periosteum, were resected en bloc in all cases. Full-thickness skin grafting was performed for eight lesions. Four lesions were closed directly, one required both skin flap and full-thickness skin grafting, and one was managed by open healing. In a separate case, a two-stage procedure was performed using an artificial dermis, followed by flap and graft reconstruction to ensure complete tumor excision (Figure 2).

Figure 2.

Two-stage surgical procedure using artificial dermis following the excision of cutaneous squamous cell carcinoma (cSCC). A. A narrow 4-mm horizontal surgical margin is designed to preserve the eyeball during cSCC excision. B. The periosteum beneath the tumor is removed with a deep margin, exposing the orbital bone (yellow arrowhead). C. An artificial dermis is applied over the defect to allow for the histopathologic confirmation of an adequate surgical margin. D. After confirming complete excision of the cSCC, a glabellar flap is designed to cover the exposed bone in the second-stage procedure. E. The eyelid defect is reconstructed using a full-thickness skin graft. F. Postoperative appearance. Photographs were only collected after 1 month, as the patient died shortly after the second surgery.

No regional recurrences or metastases were observed during the follow-up period. The final TNM staging was stage I in one patient, stage III in four patients, and stage IV in one patient. There were no tumor-related deaths; however, four patients died due to the progression of pulmonary aspergillosis during follow-up. Of the four patients who died during the follow-up period, two were able to discontinue voriconazole therapy, whereas the other two were not.

Discussion

This study examined the epidemiology and treatment outcomes of voriconazole-induced cSCC in Japanese patients with pulmonary aspergillosis, a setting wherein this condition is rarely reported²⁵⁾. The patients, all of whom had Fitzpatrick skin type IV, developed multiple aggressive lesions, with most exhibiting rapid growth and deep invasion. Despite the presence of high-risk histopathological features such as Clark level IV or V and perineural/perivascular invasion, there were no regional recurrences or metastases during the follow-up period. These findings suggest that early detection and complete surgical excision may result in favorable outcomes, even in clinically aggressive cases.

The mechanism of voriconazole-induced carcinogenesis is believed to involve a multistep process, beginning with acute and chronic phototoxicity, progressing through actinic keratosis, and ultimately leading to cSCC^19,29). Voriconazole is primarily metabolized in the liver by cytochrome P450 (CYP) enzymes, primarily CYP2C19 and, to a lesser extent, CYP2C9 and CYP3A4⁷⁾. Interestingly, these enzymes have also been identified in human keratinocytes, which suggests that peripheral metabolism of voriconazole may occur in the skin^30,31). The main metabolite of voriconazole is voriconazole N-oxide (VNO), which has minimal to no antifungal activity⁷⁾. Studies of UV absorption spectra have indicated that VNO, rather than voriconazole itself, may be responsible for phototoxic reactions, as it absorbs light in the UV B range (<310 nm)³²⁾. Moreover, VNO and its photoproducts under UV B exposure can act as UV A sensitizers, generating reactive oxygen species that lead to oxidative DNA damage, thereby supporting its role in both phototoxicity and photocarcinogenesis³³⁾. Additionally, voriconazole has been shown to activate the aryl hydrocarbon receptor and upregulate the levels of cyclooxygenase-2 (COX-2), a key enzyme in the development of UV-related tumors, including skin cancers³⁴⁾. Notably, the activity of CYP2C19 is determined by genetic polymorphisms, with a higher frequency of poor metabolizer alleles reported in Asian populations than in White, Black, and Arab populations³⁵⁾. These alleles may result in elevated blood concentrations of voriconazole in Asian individuals, potentially increasing their susceptibility to phototoxicity and subsequent skin cancer. As the rate of organ transplantation continues to rise in Japan, clinicians should remain vigilant for the development of voriconazole-induced skin neoplasms and ensure timely recognition and treatment when these lesions occur.

In the present study, the median age of the patients with voriconazole-induced cSCC was 78 years, which is slightly younger than the typical peak age of cSCC diagnosis of 80-89 years that has been reported for sporadic cases in the Japanese population^36,37). Furthermore, a sex disparity was also observed. In Japan, 63% of cSCC cases are reported to originate in sun-exposed areas, such as the head, neck, and hands³⁸⁾. This rate is relatively low compared to the 70% rate reported in White populations³⁹⁾. Several risk factors, including residing in areas with high ambient UV radiation, being male, living to an older age, having a history of skin cancer, and prolonged use of immunosuppressive therapy, are associated with the development of keratinocyte carcinoma in OTRs²⁰⁾. Additionally, the incidence of skin cancer is significantly lower in non-White OTRs than in their White counterparts, particularly among individuals with Fitzpatrick skin types I-III²¹⁾. In the Rotterdam Study, male participants were found to have over 10 times more actinic keratoses than female participants in a cohort of 2,061 individuals with a mean age of 72 years⁴⁰⁾. Similarly, our findings showed a male predominance among patients with voriconazole-induced cSCC, which may be partially explained by behavioral differences, such as greater sun exposure and lower use of sunscreen among men, as well as biological factors, including the photoprotective effects of estrogen-induced melanin production in women. Given these findings, careful consideration should be given before initiating long-term voriconazole treatment in male patients, even those of a relatively younger age, particularly in regions with high UV exposure.

The patients in this series, all of whom had Fitzpatrick skin type IV, received voriconazole at a median dose of 400 mg/day for a median duration of 5.5 years before the onset of cSCC. This treatment duration, with a similar dose among patients who were not on immunosuppressive therapy, including OTRs, exceeds those reported in most OTR studies. Feist et al.¹⁴⁾ reported that the median duration of voriconazole therapy in OTRs who developed skin cancer was 3.4 years, which was significantly longer than the 1.7 years observed in those without skin cancer. Moreover, extended therapy was associated with more aggressive disease, with an odds ratio of 2.1 for skin cancer development based on logistic regression analysis. Similarly, in a large population-based study of 9,599 non-Hispanic White lung transplant recipients, D'Arcy et al.²⁰⁾ found that the patients exposed to voriconazole for more than 15 months had a threefold higher risk of developing cSCC compared to that of unexposed individuals. These findings support the hypothesis that prolonged voriconazole exposure may contribute to the development of phototoxicity-induced preneoplastic lesions that eventually progress to invasive skin cancer. These findings also highlight the fact that even in non-transplant-recipient patients, long-term voriconazole exposure may dramatically increase the risk of cSCC development. Moloney et al.⁴¹⁾ further suggested that reducing immunosuppression in OTRs with aggressive SCC may prolong survival times compared to those in patients whose immunosuppressive regimens remain unchanged. In this cohort, no patient had a life-threatening comorbidity or immunosuppressive condition based on the ASA-PS classification. Nevertheless, four of six patients died from progressive pulmonary disease. These findings suggest that, beyond voriconazole exposure and skin type, poorer respiratory conditions may also influence cSCC development and outcomes. Even in the absence of immunosuppression, underlying chronic illness may limit the ability of a patient to recognize or respond to early tumor changes, emphasizing the need for proactive monitoring and early dermatologic assessment.

In this study, all patients had multiple cSCC lesions with aggressive characteristics, including deep invasion beyond Clark level IV, rapid tumor growth (≥4 mm/month), and perineural or perivascular invasion in 3 of 14 lesions, although none of the tumors were poorly differentiated. Cutaneous SCCs with a growth rate of >4 mm/month have been shown to be associated with a higher risk of nodal progression and a shorter progression time to lymph node metastasis compared to that of cSCCs with a growth rate of <4 mm/month⁴²⁾. Treatment outcomes for voriconazole-induced cSCC are rarely described in the literature, although its epidemiology and mechanisms of carcinogenesis have attracted increasing interest. Voriconazole use is associated with a 65-fold increase in cSCC incidence among OTRs compared to that in the general population; additionally, the risk of metastasis rises to approximately 8% in OTRs versus 0.5%-5% in the general population¹⁶⁾. In a French study involving 188 OTRs, 66% of patients developed a second cSCC within 5 years of their initial diagnosis⁴³⁾. Lindelöf et al.¹⁸⁾ reported a 52-fold increased risk of disease-specific death in OTRs compared to that in the general population, which they attributed to the higher frequency and more aggressive behavior of multiple tumors in these patients. Notably, Clark level IV was the most commonly reported depth of invasion (36.8%), and nearly half of the tumors were <10 mm in size. Despite the high-risk features in the present study, no regional recurrences or lymphatic or distant metastases were observed during follow-up. These favorable outcomes may be attributed to complete surgical excision with adequate margins, based on preoperative histopathology and CT/MRI findings, following established guidelines for high-risk areas⁴⁴⁾. In contrast, prior reports have noted that incomplete excisions are most frequently seen in lesions of the ear, nose, and cheek, which often involve deep surgical margins⁴⁵⁾. Mohs micrographic surgery, an effective technique that facilitates real-time histopathologic assessment, is a globally-recognized treatment for high-risk non-melanoma skin cancers⁴⁶⁾. However, it is rarely performed in Japan, primarily due to limited physician training and reimbursement barriers; these factors contribute to a general reluctance to adopt the procedure⁴⁷⁾. As an alternative, our institution often employs a two-stage excision strategy using artificial dermis after the initial tumor ablation. This method facilitates precise histopathologic margin assessment and ensures complete tumor resection. It is especially valuable in functionally and cosmetically sensitive areas where narrower margins may be preferred. Radiotherapy was also successfully applied in cases with inadequate surgical margins and produced favorable outcomes, consistent with previous studies on conventional cSCC treatment⁴⁸⁾. Nevertheless, the potential adverse effects of radiotherapy, including cataracts, osteoradionecrosis, and severe contracture, must be carefully considered, particularly in cases involving facial lesions.

Several studies have reported a significant regression of preneoplastic lesions in patients who discontinued voriconazole treatment⁴⁹⁾. However, determining whether to continue or discontinue voriconazole therapy remains challenging, as the antifungal benefits often outweigh the potential risks associated with long-term use. Alternative treatment options include other azole antifungals, such as itraconazole and posaconazole, as well as echinocandins, including caspofungin, micafungin, and anidulafungin, which may be administered following the discontinuation of voriconazole. Nonetheless, few studies have conducted comparative analyses regarding the efficacy of these antifungal agents⁵⁰⁾. Furthermore, caution should be exercised when transitioning between antifungal treatments, particularly in patients with comorbidities, due to the potential for drug-drug interactions and the development of resistance among fungal species⁵¹⁾. Patients with relatively stable respiratory conditions may initially be advised to discontinue voriconazole therapy. However, significant exacerbation of pulmonary aspergillosis may adversely affect the patient's prognosis more than cSCC associated with voriconazole, as demonstrated by the current study. Therefore, healthcare providers should carefully deliberate on the continuation or discontinuation of voriconazole therapy and ensure that informed consent is obtained from patients. Given the complexities associated with discontinuing voriconazole, it is imperative to educate patients on the importance of protective behaviors against sun exposure and to encourage regular self-assessment for skin changes to facilitate timely interventions. Additionally, routine dermatological evaluations by an expert physician are recommended. Prompt identification and intervention can markedly improve outcomes for patients with progressive pulmonary disease who are receiving long-term antifungal therapy.

This study has some limitations. It was conducted at a single center, with a small sample size and a retrospective design, which may limit the generalizability of the findings and could have introduced selection bias. Although significant clinical and histopathologic features of voriconazole-induced cSCC were identified in Japanese patients with pulmonary aspergillosis, the relatively short follow-up period may have limited the ability to assess long-term outcomes such as regional recurrence or distant metastasis. A large-scale, prospective multicenter study is needed to obtain more robust data and clarify the epidemiology of voriconazole-induced cSCC in this population.

In conclusion, the epidemiology and treatment outcomes of voriconazole-induced cSCC in Japanese patients with Fitzpatrick skin type IV and pulmonary aspergillosis were investigated. Due to the rarity of this condition in Asian populations, physicians may be less familiar with its potential adverse effects and may continue to administer voriconazole therapy without recognizing early cutaneous signs of malignancy. In this cohort, multiple cSCC lesions often exhibited aggressive features and posed significant treatment challenges. However, early detection and timely surgical management resulted in favorable outcomes. As phototoxic symptoms associated with voriconazole can often be mitigated by dose adjustment or discontinuation, clinicians should remain vigilant for signs of cutaneous malignancy and initiate treatment before deep tissue invasion occurs.

Author Contributions: Y.A. conceived and designed the study, analyzed and interpreted the data, drafted the manuscript, and was accountable for all aspects of the work. K.I. acquired the data and edited the manuscript. K.M., Y.Y., M.N., M.M., and K.Ik. treated the patients and analyzed the outcomes. I.H. performed the histopathological review, critically revised the manuscript, and approved the final version.

Conflicts of Interest: There are no conflicts of interest.

Ethical Approval: This study was approved by the Research Ethics Board of Tokushima University (approval no. 4661). The research was conducted in accordance with the principles outlined in the Declaration of Helsinki and its later amendments.

Consent to Participate: Informed consent for participation was obtained using an opt-out method appropriate for this observational study.

Consent for Publication: Informed consent for publication was obtained using an opt-out method appropriate for this observational study.