Abstract
Introduction
Vulvar Paget’s disease (VPD) is a rare condition, and a comprehensive analysis of its clinical and pathological characteristics, treatment approaches, and prognosis has not yet been conducted.
Methods
We conducted a retrospective population-based cohort study using publicly available, de-identified data from the Surveillance, Epidemiology, and End Results (SEER) database. Patients diagnosed with VPD between 2004 and 2021were included. Kaplan-Meier and Cox models assessed impacts on overall and cancer-specific survival. Propensity score matching (PSM) balanced surgical and non-surgical groups.
Results
1,778 patients with VPD were included. Most patients were women aged ≥50 years (96.5%) and of White race (81.7%). The labium majus was the most commonly specified tumor site (203/1,778, 11.4%), although most cases were recorded as vulva, NOS. A total of 1,374 patients underwent surgical treatment. After PSM, surgery remained associated with improved survival. A generally similar trend was observed across subgroups, although not all subgroup comparisons reached statistical significance.
Conclusion
Based on the SEER database, this study found that surgery was beneficial for improving the prognosis of VPD patients, especially for those who were elderly.
Introduction
Extramammary Paget’s disease (EMPD) is a rare type of malignant skin tumor that predominantly occurs in areas rich in apocrine glands, particularly the genital and perianal skin. 1 As the most common type of EMPD, vulvar Paget’s disease (VPD) only accounts for approximately 1-2% of all malignant tumors in the vulvar region. 2 VPD is most commonly found in postmenopausal white women. 2 It is an intraepithelial disease characterized by vulvar scaly plaques with pruritus or burning as a common but nonspecific feature.1,3 Therefore, at the initial clinical visit, VPD is often not recognized and is misdiagnosed as eczema or a fungal infection. 4 Early skin biopsy is of significant importance for diagnosis. The presence of typical Paget cells in the epidermis, which are large atypical cells with abundant, clear cytoplasm that is sometimes eosinophilic, serves as the basis for diagnosing VPD.4,5 However, the diagnosis of VPD remains difficult due to its latent and indolent course and the nonspecific symptoms and clinical manifestations.
Currently, surgery remains a major treatment approach for VPD, particularly when invasive disease is suspected or confirmed.6-8 However, there is still controversy over whether to perform extensive resection. Some studies indicate that positive surgical margins are associated with an increased risk of recurrence in patients, but there are also studies reporting no correlation between positive surgical margins and overall patient survival rates.9-11 Due to tissue and function damage caused by surgery, there is an increasing focus on alternative treatment methods for VPD. Common alternatives include topical medications, such as imiquimod (self-applied cream), radiation therapy, chemotherapy, photodynamic therapy (a form of phototherapy using photosensitizing compounds that selectively become toxic to targeted cancer cells and other diseased cells when exposed to light), laser therapy or a combination of these methods.9,12,13 Among these alternatives, topical imiquimod has shown encouraging results in selected patients, although the effectiveness of non-surgical treatments overall still requires further validation. 14 In addition, the high recurrence rate of VPD is also a therapeutic challenge that needs to be addressed currently.
In order to understand the course of this rare disease and optimize current treatment plans, this study conducts a comprehensive analysis of the clinical and pathological characteristics, treatment patterns, and prognostic features of VPD.
Method
Study Design and Ethics
This was a retrospective population-based cohort study based on publicly available, de-identified data from the Surveillance, Epidemiology, and End Results (SEER) database. This study was reported in accordance with the STROBE guideline.15,16 Eligible patients were selectively identified from the SEER database according to predefined inclusion and exclusion criteria. All patient details were de-identified, and no identifiable private information was available to the investigators. Therefore, Institutional Review Board/Ethics Committee approval and informed consent were not required.
Patient Selection
Patients with a pathological diagnosis of VPD in the SEER database between 2004 and 2021 were retrospectively identified. Cases were screened using ICD-O-3 morphology codes 8542/2 and 8542/3, corresponding to preinvasive/in situ and invasive Paget’s disease, respectively, together with vulvar primary site codes C51.0, C51.1, C51.2, C51.8, and C51.9. To minimize the inclusion of secondary Paget’s disease and preferentially identify primary VPD, only patients with sequence number indicating one primary tumor were retained. Patients were excluded if they lacked pathological confirmation, had incomplete surgical information, or had unclear or missing clinicopathological characteristics or follow-up data. A total of 1,778 potentially eligible cases were initially identified. The patient selection flowchart is shown in Figure 1. Flowchart of inclusion and exclusion of patients in this study
Statistical Analysis
Categorical variables were expressed as numbers (percentages). The Kaplan-Meier method and Cox proportional hazards model were used to evaluate the effects of clinicopathological characteristics and treatment patterns on overall survival (OS). OS was defined as the time from diagnosis to death from any cause. Propensity score matching (PSM) was performed to reduce baseline confounding between the surgical and non-surgical groups. A 1:3 nearest-neighbor matching strategy was applied using a caliper width of 0.2 without replacement. The propensity score model included the following baseline covariates: age, race, marital status, year of diagnosis, primary tumor site, tumor size, the International Federation of Gynecology and Obstetrics (FIGO) stage (2018) and time from diagnosis to treatment. Standardized mean differences (SMDs) were used to assess covariate balance before and after matching, and Love plots were generated to visualize the balance of covariates after matching. Statistical significance was defined as a two-tailed p-value <0.05. All statistical analyses were conducted using R (version 4.3.2).
In accordance with the journal’s guidelines, we will provide our data for independent analysis by a selected team by the Editorial Team for the purposes of additional data analysis or for the reproducibility of this study in other centers if such is requested.
Results
Patient Demographics and Clinicopathological Characteristics
Patient Demographics and Baseline Characteristics
Distribution of the number of cases of vulvar Paget’s disease from 2004 to 2021
Regarding tumor location, 203 (11.4%) cases involved the labium majus, 34 (1.9%) the labium minus, 6 (0.3%) the clitoris, and 101 (5.7%) overlapping vulvar lesions, whereas 1,434 (80.7%) were recorded as vulva, NOS. Tumor size was <20 mm in 222 (12.5%) patients, 20–40 mm in 269 (15.1%), and >40 mm in 263 (14.8%), while 1,024 (57.6%) had unknown tumor size. In the combined stage variable, 557 (31.3%) patients were classified as Tis, 910 (51.2%) as localized, 122 (6.9%) as regional, and 21 (1.2%) as distant disease. According to the FIGO stage (2018) variable, 390 (21.9%) patients were classified as Tis, 675 (38.0%) as stage I, 110 (6.2%) as stage II, 27 (1.5%) as stage III, and 41 (2.3%) as stage IV, while 535 (30.1%) had unknown stage.
In terms of treatment, 1,374 (77.3%) patients underwent surgery and 404 (22.7%) did not. Only 42 (2.3%) patients received radiotherapy and 43 (2.4%) received chemotherapy.
Survival Analysis Before PSM
Kaplan-Meier Estimates of 1.3.5-Year Survival (95% Confidence Intervals) Before PSM
1Log-rank test.
Kaplan–Meier analyses before matching showed significant differences in OS according to age (p<0.0001), FIGO/stage-related categories (p<0.0001), and surgical status (p<0.0001), whereas the difference according to primary site did not reach statistical significance (p=0.056). The corresponding Kaplan–Meier curves are shown in Figure 3. Kaplan-Meier curves for OS before PSM stratified by age (A), SEER-recorded FIGO/stage-related categories (B), primary tumor site (C), and surgical status (D)
Baseline Covariates of the Surgical and Non-surgical Groups Before and After PSM
Baseline Covariates of the Surgical and Non-surgical Groups Before and After Matching
1Standardized Mean Difference.
After 1:1 nearest-neighbor PSM, 380 surgical patients and 148 non-surgical patients were retained. Overall covariate balance improved after matching, although residual imbalance remained in some variables, particularly those with a high proportion of unknown values. Detailed pre- and post-match covariate distributions and SMDs are presented in Table 3, and the post-match balance is further illustrated in Supplementary Figure 1.
Survival Analysis After PSM
Median Survival Time of Surgical and Non-surgical Groups After PSM
Kaplan-Meier estimates
Kaplan-Meier Estimates of 1.3.5-Year Survival (95% Confidence Intervals) After PSM
1Log-rank test.
In the matched non-surgical group, the median OS was 90.0 months (95% CI: 72.0–NA), whereas in the matched surgical group, the median OS was 179.0 months (95% CI: 146.0–NA). The corresponding 1-, 3-, and 5-year OS rates were 93.4%, 78.9%, and 64.5% in the non-surgical group, compared with 97.8%, 90.6%, and 85.8% in the surgical group. The difference in OS between the matched surgical and non-surgical groups remained statistically significant (log-rank p<0.001). The Kaplan–Meier curves after matching are shown in Figure 4. Kaplan-Meier curves for OS after PSM stratified by age (A), SEER-recorded FIGO/stage-related categories (B), primary tumor site (C), and surgical status (D)
Stratified Survival Analysis According to Invasion Status
Because the cohort included both preinvasive/in situ and invasive VPD, additional Kaplan-Meier analyses stratified by invasion status were performed. As shown in Supplementary Figure 2, surgery was associated with better OS in both the in situ and invasive subsets before and after PSM. In patients with in situ disease, the difference between the surgical and non-surgical groups was significant both before PSM (p<0.0001) and after PSM (p=0.024). In patients with invasive disease, the difference was also significant both before PSM (p<0.0001) and after PSM (p=0.00019). These analyses were performed to distinguish the survival patterns of biologically distinct disease subsets and to avoid interpreting tumors in situ together with invasive stage I–IV disease.
Discussion
VPD is a rare and slowly developing vulvar epithelial neoplastic lesion, and there is no clear diagnosis and treatment guideline at present. Difficulty in diagnosis, high recurrence rate, and tissue defects and functional damage caused by surgery are the main problems encountered.1,17 We utilized the SEER database to gain further insights into VPD with the aim of refining current treatment plans and enhancing patient outcomes.
Consistent with previous studies, we found that VPD occurred more frequently in elderly white women. Labium majus was the most common site of involvement.17,18 In our study, we found that elderly and advanced stages VPD patients tended to have poorer OS. This may be related to the generally poorer physical condition of elderly patients. Patients at late stages may have lymph node or distant metastases, and the tumor may often penetrate the basement membrane to infiltrate into the dermis and subcutaneous fat layers, leading to rapid progression of the disease through lymphatic or blood pathways. Previous studies also concluded that the depth of invasion and lymph node metastasis were significant factors contributing to a poorer prognosis.6,19 Van der Linden et al also showed that VPD had a lower survival rate after infiltration occurs. 4 This further illustrated the importance of specifying the depth of VPD infiltration. Because preinvasive and invasive vulvar Paget’s disease differ in biological behavior and prognosis, distinguishing these two entities is clinically important when interpreting treatment patterns and survival outcomes.
Our study suggested that surgery was associated with improved survival in this cohort; however, this finding should be interpreted in the context of the heterogeneous biological spectrum of VPD. According to the size and location of the lesion, the surgical procedures that can be chosen include radical vulvectomy, radical partial vulvectomy, and simple partial vulvectomy. 20 Due to the multi-focal nature and microscopic dissemination characteristics of VPD, the positive surgical margin rate is typically high.17,21 Whether surgery should pursue negative margins is currently a matter of debate. Sopracordevole 22 and Nasioudis 17 et al believe that the surgical margin status of VPD is not correlated with disease recurrence and prognosis, while a recent retrospective study suggests that positive margins are associated with an increased risk of local recurrence and all-cause mortality. 21 Therefore, the value of negative surgical margins in VPD still requires further research confirmation.
For patients with severe comorbidities or extensive metastases who cannot tolerate surgery or have recurrent disease after surgery, non-invasive treatment options such as imiquimod therapy, radiotherapy, carbon dioxide laser ablation, photodynamic therapy (PDT), and chemotherapy is available. However, our study indicated that chemotherapy and radiotherapy didn’t improve the prognosis. In addition to the reason that VPD is not sensitive to chemotherapy and radiotherapy, it may also be led to the number of patients included in our study who underwent chemotherapy and radiotherapy was relatively small.23,24 Topical application of 5% imiquimod cream can avoid tissue damage caused by surgery and has been proven to be effective for both primary and recurrent disease. 25 However, the specific target population for its use still needs further research.
This study is the largest known real-world study of VPD to date. The SEER database widely represents the US population and has a high degree of integrity and accuracy in disease diagnosis and follow-up, making the trends observed in this study reliable. Nevertheless, our study had certain limitations. Firstly, retrospective studies have their inevitable inherent biases. Secondly, the SEER database lacks specific surgical procedures and other variables, which are also important for improving the prognosis of VPD patients. Thirdly, subgroup analyses were exploratory in nature, were not adjusted for multiple comparisons, and should be interpreted with caution, especially in small subgroups. Further prospective studies are highly necessary.
Conclusion
Based on the SEER database, this study conducted an in-depth investigation into the clinical pathological characteristics, treatment, and prognosis of VPD. The findings indicated that surgery was beneficial for improving the prognosis of VPD patients. Patients with VPD who underwent surgery had improved survival in this cohort; however, findings from subgroup analyses, especially those related to treatment timing, should be interpreted cautiously.
Supplemental Material
Supplemental Material - Survival and Treatment Patterns in Vulvar Paget’s Disease: Evidence From SEER
Supplemental Material for Survival and Treatment Patterns in Vulvar Paget’s Disease: Evidence From SEER by Lele Chang, Yaxin Kang, Jing Liu, Cuibo Lin and Qin Xu by Cancer Control.
Supplemental Material
Supplemental Material - Survival and Treatment Patterns in Vulvar Paget’s Disease: Evidence From SEER
Supplemental Material for Survival and Treatment Patterns in Vulvar Paget’s Disease: Evidence From SEER by Lele Chang, Yaxin Kang, Jing Liu, Cuibo Lin and Qin Xu by Cancer Control.
Footnotes
Ethical Considerations
This study was conducted using publicly available, de-identified data from the SEER database. Therefore, Institutional Review Board/Ethics Committee approval was not required for this study.
Consent to Participate
Informed consent was also not required because no identifiable private information was involved.
Consent for Publication
All the authors agreed to publish this work.
Author Contributions
LC and YK proposed the main idea and drafted the initial version of the manuscript. JL conducted the data analysis, while JL were responsible for data collection and validation. QX and CL provided guidance on the methodology and structure of the article. QX and JL contributed to funding acquisition. QX and LC had carefully polished and revised the overall article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Fujian Cancer Hospital High-Level Talent Training Program (2023YNG09), Joint Funds for the innovation of science and Technology, Fujian province (Grant number: 2023Y9449), Natural Science Foundation of Fujian Province (Grant number: 2024J011099 and 2023J011273).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The datasets used and analysed during the current study available from the corresponding author on reasonable request.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
