Article Text
PDF
Abstract
Introduction Vulvar squamous cell carcinoma develops through two separate pathways, associated with the presence or absence of high-risk human papilloma virus (HPV). The objective of this study was to evaluate treatment response and clinical outcomes in women with HPV-associated versus HPV-independent vulvar squamous cell carcinoma treated with primary radiation therapy, in order to determine the ability to use HPV status as a predictor of response to radiation therapy.
Methods This was a retrospective cohort study combining data from British Columbia Cancer, Canada and Duke University, USA. Patients were included who had been treated with radiation therapy but excluded if they had received major surgical interventions. Immunohistochemistry for p16 (as a surrogate for high-risk HPV infection) and p53 was performed. We analyzed the univariable association between p16 status and clinico-pathological features and performed univariable survival analysis for p16.
Results Forty-eight patients with vulvar squamous cell carcinoma treated with primary radiation therapy were identified: 26 p16 positive/HPV-associated patients and 22 p16 negative/HPV-independent patients. p16 positive vulvar squamous cell carcinoma demonstrated a significantly improved overall survival (HR 0.39, p=0.03) and progression-free survival (HR 0.35, p=0.02). In women treated with definitive radiation therapy, p16 positivity was associated with improved overall survival (HR 0.29, p<0.01) and progression-free survival (HR 0.21, p<0.01). Among patients who received sensitizing chemotherapy, a significant association was observed with p16 positive tumors and overall survival (HR 0.25, p=0.03) and progression-free survival (HR 0.09, p<0.01).
Conclusion This study suggests that HPV status in vulvar squamous cell carcinoma has both prognostic and predictive implications, with increased radiosensitivity demonstrated in HPV-associated vulvar squamous cell carcinoma. Implications may include radiation dose de-escalation for HPV-associated vulvar squamous cell carcinoma and increased surgical aggressiveness for HPV-independent vulvar squamous cell carcinoma.
- radiotherapy
- vulva
- vulvar neoplasms
- pathology
- radiation oncology
Statistics from Altmetric.com
HIGHLIGHTS
p16-associated vulvar squamous cell carcinoma demonstrated a significantly improved overall survival and progression-free survival.
In women treated with definitive radiotherapy, p16 positivity was associated with improved overall survival and progression-free survival.
Among patients who received sensitizing chemotherapy, p16 positivity was associated with improved overall survival and progression free-survival.
Introduction
Vulvar squamous cell carcinoma is the most common type of vulvar cancer, accounting for approximately 3–5% of all gynecological malignancies.1 Vulvar squamous cell carcinoma develops through two separate etiopathogenic pathways, relative to the presence or absence of high-risk human papillomavirus (HPV). HPV-associated tumors often present in younger women with similar risk factors as for cervical squamous cell carcinoma. Morphology is typically non-keratinizing basaloid or warty and progression is through a precursor lesion termed high-grade squamous intraepithelial lesion, also known as VIN2/3.2 3 In contrast, HPV-independent vulvar squamous cell carcinoma commonly arises in older women with morphology that is typically well-differentiated and keratinizing and an associated precursor termed differentiated vulvar intraepithelial neoplasia (VIN). Differentiated VIN is associated with rapid recurrence and progression to invasive disease.4 HPV-independent tumors commonly exhibit TP53 gene mutations and are increasingly being recognized as having a more aggressive clinical phenotype with worse disease-specific survival.5 6
The prevalence of HPV-associated vulvar squamous cell carcinoma ranges from 10–70% across different studies.7 HPV status can be determined by nucleic acid-based testing or immunohistochemical staining for a surrogate marker of oncogenic HPV infection: the cyclin-dependent kinase inhibitor p16.4 p16 positivity through immunohistochemistry has been validated as a reliable marker of HPV-associated vulvar squamous cell carcinoma,8 and can be used as an inexpensive and readily available method for determining the HPV status of vulvar tumors.
In squamous cell carcinomas of the head and neck, p16 positive/HPV-associated tumors have consistently shown improved response to radiation resulting in superior locoregional control and survival, as compared with p16 negative/HPV-independent disease.9–11 In vulvar squamous cell carcinoma, however, the impact of HPV positivity in women treated with primary radiotherapy has not yet been established. Our objective was to evaluate outcomes and responses to treatment in women with HPV-associated versus HPV-independent locally advanced vulvar squamous cell carcinoma treated with primary/definitive radiation therapy in order to evaluate not only the prognostic implication, but also the ability to use HPV status as a predictor of response to radiation therapy.
Methods
This is a dual-center retrospective cohort study of locally advanced vulvar squamous cell carcinoma receiving up front radiation therapy ± chemotherapy. Data were abstracted from British Columbia Cancer in Vancouver, British Columbia, Canada as well as Duke University, Durham, North Carolina, USA. Clinical information, pathological parameters, treatment details, and patient outcomes were collected from medical records.
Cohort—British Columbia Cancer Agency
Following Institutional Review Board approval, we identified all cases of locally advanced vulvar squamous cell carcinoma treated at the British Columbia Cancer agency between January 1, 1980 and December 1, 2016. Patients were included who had been treated with radiation therapy (either for locally advanced, surgically unresectable disease or for advanced stage at presentation), or chemotherapy plus radiation therapy, but were excluded if they had received major surgical interventions (eg, radical vulvectomy) or had received chemotherapy alone (no radiation therapy). Patients with both curative and palliative intent radiation therapy were included. Inclusion also required having tissue available (formalin-fixed paraffin-embedded archival material) for immunohistochemistry and adequate clinico-pathological and outcome data for analyses.
Although vulvar squamous cell carcinoma is a surgically staged disease, in this series stage was assigned based on treating physician clinical examination ± imaging according to the International Federation of Gynecology and Obstetrics (FIGO) 2009 classification. Tumor dimensions before treatment had been carefully recorded from clinical exam and again measured 6 weeks post-completion of therapy. Follow-up and surveillance was uniform and in accordance with the guidelines set forth by British Columbia Cancer (every 3–4 months for the first 2 years, and then every 6 months for a total of 5 years).
Cohort—Duke University
Women with vulvar squamous cell carcinoma treated with upfront, definitive chemoradiation for locally advanced, non-surgical disease between January 6, 1994 and January 6, 2015 were identified from a previously published cohort,12 and clinical cases with available tissue blocks for evaluation were included for a combined analysis (n=12). The same inclusion and exclusion criteria were used for this collaborating center, with independent Institutional Review Board approval.
Immunohistochemistry for p16 and p53
Immunohistochemistry and scoring for all tissue blocks (British Columbia Cancer, and Duke) was performed after obtaining Institutional Review Board approval at one center (Vancouver General Hospital). Immunohistochemistry was performed on 4 µm thick FFPE (formalin-fixed paraffin-embedded) whole tissue sections. Immunohistochemical stains were performed using the Ventana Discovery XT and Ventana Benchmark XT systems (Ventana Medical Systems, Tucson, AZ) as per the manufacturer's instructions. Sections were cut and placed onto charged glass slides, air dried for 10 min and baked at 60°C for 10 min. Cell conditioning solution CC1 (Ventana, Tris–based EDTA buffer, pH 8.0), heat induced antigen retrieval (37°C for 32 min), and Ventana XT Optiview DAB detection kit were used for all antibodies. The p53 antibody clone DO-7 (mouse monoclonal, Dako, Burlington, ON, Canada) and the p16 antibody clone E6H4 (mouse monoclonal, Roche, Westborough, MA) were used.
Immunohistochemistry interpretation was performed by two pathologists (LH, ET) according to previously published criteria.13 p16 was scored as positive if there was strong continuous block-like staining in at least the lower one-third of the lower squamous epithelium thickness. Any lesser staining was scored as negative. Cases were classified as HPV-associated if p16 immunohistochemistry was strongly positive (as above) and HPV-independent if p16 immunohistochemistry was classified as negative.
p53 immunohistochemistry was scored as abnormal when any one of the following three staining patterns were observed: (1) strong continuous nuclear staining of basal cells with strong parabasal staining in at least one-third of the epithelial thickness (overexpression type); (2) complete absence of nuclear staining in the basal cell layer (null type); (3) continuous cytoplasmic staining involving the basal and parabasal layers (cytoplasmic). Cases showing patchy basal p53 staining of variable intensity were scored as normal (wild-type).
Statistical Analysis
All statistical analyses for this project were done using R project for statistical computing (R 3.5.2). We considered the univariable association between p16 immunohistochemistry and other predictors using a χ2 test for binary and categorical variables and the Kruskal-Wallis test for continuous variables.
The primary outcome was clinical response of the vulvar tumor, as determined by the treating radiation oncologist. Complete clinical response was designated if there was no visible tumor present at radiation completion. Hazard ratios were calculated using univariable Cox regression models. Estimates of progression-free survival, overall survival, and disease-specific survival were assessed graphically using the Kaplan-Meier plots and compared using the log-rank test. Statistical significance was set at 0.05. Multivariable Cox regression models were used to correct for the key clinico-pathological parameters of interest as well as to assess the interaction effect between p16 and treatment. Cases with missing values were removed from analyses and only complete cases were considered.
Results
Cohort Characteristics British Columbia Cancer Agency
Thirty-six complete cases were included in the analysis. Median age of the cohort was 71.5 years (IQR 60–83). Approximately one third of patients presented with FIGO stage II disease (36.1%), and one third with stage III disease (33.3%). Average size (largest diameter) of the vulvar tumor at presentation was 69.4±5.4 mm and 58.3% had clinically or radiologically positive nodes. Thirty-one patients had definitive radiation therapy, and five patients received palliative radiation therapy (n=1 previous definitive radiotherapy, n=4 advanced stage, poor performance status). Overall mean radiation dose to the vulva was 55.5±2.1 Gy given in 30 fractions (mean 29.4, SD 10.9). Mean radiation dose to the vulva in patients treated with definitive radiation was 60.7±4.56 Gy given in standard fractionation, while the mean palliative dose to the vulva was 29.63±7.87 Gy in varied fractionations. Fifty percent (n=18) of women received concurrent radio-sensitizing chemotherapy, of which only two contained a non-cisplatin based regimen.
Cohort Characteristics Duke University
Twelve complete cases with tissue blocks were available from Duke University. Women in the Duke cohort were younger (median age 64 years, IQR 49–68) than the Vancouver cohort. Mean size of the vulvar tumor on presentation was 53±9 mm; three patients (25%) presented at FIGO stage II and eight (75%) presented at FIGO stage III or beyond. Fifty percent of this cohort had clinically or radiologically positive lymph nodes. Ninety-two percent (n=11) received concurrent radio-sensitizing chemotherapy with a mean radiation dose of 62±2 Gy given in standard fractionation to the vulva.
Pooled Cohort Characteristics
In total, 48 unique cases with complete clinico-pathological, molecular, and outcome data were included in the pooled analysis. Cohort characteristics are shown in Table 1. Online supplemental Figure 1 demonstrates a flow diagram of participants from both cohorts.
Supplemental material
Cohorts’ clinical and pathological characteristics
The median age of the pooled cohort was 67 years (IQR 60–79) with 28 women (58.3%) presenting with clinically/radiologically assessed FIGO stage III or IV disease. The median tumor size at presentation was 60 mm (IQR 44–80) and 56% of the cohort was node positive. A total of 60% of patients (n=29) received radio-sensitizing chemotherapy and the median dose of radiation given was 61.4 Gy (IQR 59.4–66.0). Median definitive radiotherapy dose was 61.2 Gy (IQR 59.4–65.0) given in standard fractionation, and median palliative intent radiation dose was 27.5 Gy (IQR 25.0–35.9). Median time to recurrence within the entire cohort was 12 months and median overall survival was 23 months.
HPV Status and Patient Outcomes Pooled Analysis
Of the pooled cohort, 26 patients were p16 positive/HPV-associated and 22 patients were p16 negative/HPV-independent. Of the p16 positive tumors, 24/26 (92%) had wild-type (normal) p53 staining, 1/26 (4%) had abnormal p53 staining (overexpression pattern), and 1/26 had uninterpretable p53 (this was a fine needle aspiration cytology specimen). Of the p16 negative tumors, 20/22 (91%) had abnormal p53 staining (15 overexpression pattern, 4 null pattern, 1 cytoplasmic pattern) and 2/22 had wild-type (normal) p53 staining.
Women with p16 positive tumors were significantly younger than women with p16 negative tumors (64 vs 74, p=0.02), but there was no significant difference seen in FIGO stage (p=0.62), presenting tumor size (p=0.71), or dose of radiation therapy (p=0.81) received between the groups (Table 2).
Univariable association of p16 status and clinico-pathological parameters
The observed percentage of cases with a complete clinical response was higher among p16 positive tumors (72.2%) compared with p16 negative tumors (42.9%); however, this analysis was only available for the British Columbia Cancer cohort and the result did not reach statistical significance (p=0.19). On univariate analysis, when compared with patients with p16 negative tumors, p16 positive vulvar squamous cell carcinoma demonstrated a significantly improved overall survival (HR 0.39, 95% CI 0.17 to 0.92; p=0.03) and a significantly improved progression-free survival (HR 0.35, 95% CI 0.14 to 0.85; p=0.02) (Figure 1A,B). p53 wild-type tumors also demonstrated a significantly improved overall survival (HR 0.25, 95% CI 0.10 to 0.61; p<0.001) and a significantly improved progression-free survival (HR 0.42, 95% CI 0.17 to 1.00; p=0.04) when compared with the p53 abnormal tumors.
Kaplan-Meier curves showing (left) overall survival (OS) and (right) progression-free survival (PFS) for tumors stratified by p16 immunohistochemistry (IHC).
In women treated with definitive radiation therapy, a significant association was observed with p16 positive tumors and overall survival (HR 0.29, p<0.01) and progression-free survival (HR 0.21, p<0.01) (Figure 2A,B). Additionally, among patients who received sensitizing chemotherapy with radiation therapy, a significant association was observed with p16 positive tumors and overall survival (HR 0.25, p=0.03) and progression-free survival (HR 0.09, p<0.01) (Figure 3A,B).
Kaplan-Meier curves showing (left) overall survival (OS) and (right) progression-free survival (PFS) for tumors stratified by p16 immunohistochemistry (IHC) among patients who received definitive radiation therapy.
Kaplan-Meier curves showing (left) overall survival (OS) and (right) progression-free survival (PFS) for tumors stratified by p16 immunohistochemistry (IHC) among patients who received definitive radiation therapy with sensitizing chemotherapy.
In multivariable analysis on the whole cohort, after adjusting for age, tumor size, nodal status, stage, and treatment type, none of the clinico-pathological parameters maintained a significant association with the outcome. We did, however, observe a significant interaction between p16 and treatment (sensitizing chemotherapy with radiation as compared with radiation alone: overall survival, p=0.015; disease-specific survival, p=0.034; progression-free survival, p=0.001), suggesting that the prognostic effect of p16 is dependent on the type of treatment received.
Discussion
In this study, we showed a significant improvement in both overall survival and progression-free survival in p16 positive (HPV-associated) tumors. This finding was confirmed among patients who received both definitive doses of radiotherapy (curative intent) as well as among those who received sensitizing chemotherapy. In the multivariate analysis, none of the key clinico-pathological parameters maintained a significant association with the outcome; however, patient numbers were low and we were likely underpowered to detect a difference. Additionally, after observing a visual difference in the Kaplan-Meier plots between those treated with radiotherapy alone versus those treated with radiation and sensitizing chemotherapy (Figures 2 and 3), we elected to perform an interaction analysis and were able to confirm a signification interaction between p16 and treatment received.
The use of HPV to stratify head and neck squamous cell carcinomas into two separate groups has long been appreciated to have both prognostic and predictive implications. When combined with other factors, such as stage of disease, comorbidity and smoking status, HPV status helps inform prognosis as well as guides treatment decisions on surgery as well as radiotherapy14 because HPV-associated head and neck squamous cell carcinomas have been shown to have an increased sensitivity to radiation.9 15 The biological mechanisms for this is not yet fully understood, but is believed to be secondary to impaired DNA repair and increased cellular apoptosis in HPV-associated tumors, rendering these cells more susceptible to radiation and less likely to respond with re-population following radiotherapy.11 16 17 These observations have led to the development of numerous trials of radiotherapy de-escalation within the HPV-associated cohort.18 19
Similar to head and neck squamous cell carcinomas, vulvar squamous cell carcinoma can be divided into two disease entities based on the presence or absence of HPV. Emerging literature is showing that HPV-associated tumors have improved prognosis with greater locoregional control and survival.13 20–22 The reasons for the improved outcomes are likely multifactorial. The precursor lesion of HPV-independent disease, differentiated VIN, occurs in older women and is more commonly associated with positive surgical margins despite attempts at curative resection. Additionally, studies have demonstrated a more aggressive clinical course for differentiated VIN with shorter latent phase and a median time to progression to squamous cell carcinoma of 1.9 years.4 This is in contrast to the HPV-associated precursor lesion, high-grade squamous intraepithelial lesion, which often has a protracted clinical course and a long latent period.23 The prognostic implications of HPV status on vulvar cancer, however, extend beyond differentiated VIN and high grade squamous intraepithelial lesion. Studies stratifying vulvar squamous cell carcinoma by HPV status have shown that HPV-independent vulvar squamous cell carcinoma has significantly worse survival outcomes, specifically in the era post-radical en-bloc vulvar resections. These data suggest that HPV-independent vulvar squamous cell carcinoma may require more aggressive surgery, but also raises the question about the radiosensitivity of HPV-independent tumors. This was the impetus behind our study—to help elucidate whether HPV status can be used not only to prognosticate, but also to predict response to radiotherapy, as is the case for head and neck squamous cell carcinomas.
Other series that have examined radiation response according to HPV status have been in women who have undergone both surgical resection followed by radiation, with only a subset having received primary radiation therapy alone.13 20 22 By using populations where radiation was used in the neoadjuvant or adjuvant setting, these studies have not been able to account for the impact of surgery (radical vulvectomy or lymph node dissection) on outcomes. Our study is unique in that we selected a precisely defined population of patients treated with primary radiation or chemo/radiation alone, thereby controlling for the effects of surgery on outcomes. In addition to confirming the prognostic implication of p16 positivity, we were able to answer the study’s primary question: HPV positivity is associated with significantly better outcomes in patients treated with definitive radiotherapy for advanced vulvar squamous cell carcinoma.
Strengths of our study include our strictly defined study population, rigorous pathology review, and uniform classification of cases with p16 immunohistochemistry by two gynecologic pathologists. Limitations of our study include its retrospective nature, small study numbers, and the reliance on the availability of archival tissue specimens over a 30 year period. In addition, although the management and surveillance of patients was guided by our cancer center’s guidelines, the guidelines changed over time as radiotherapy techniques improved and evidence emerged for the addition of sensitizing chemotherapy.
As research in this area expands, we anticipate that this knowledge will be used to guide treatment decisions—with increased surgical aggressiveness for the less radio-responsive, and more clinically aggressive HPV-independent tumors and/or consideration of radiotherapy dose de-escalation for HPV-associated vulvar squamous cell carcinoma. We hope that, as with head and neck squamous cell carcinomas, treatment algorithms will eventually be stratified by HPV status from time of first diagnosis. This will guide surgical aggressiveness, adjuvant therapy, and surveillance strategies in these biologically diverse disease entities.
Our research adds to the growing body of evidence suggesting that HPV status in vulvar squamous cell carcinoma has both prognostic and predictive implications, with increased radiosensitivity demonstrated in HPV-associated disease.
References
Footnotes
Contributors LP, LH, CBG and JNM contributed to conceptualization and writing of manuscript. All authors were involved in the editing and approval of the final manuscript and/or contribution of tissues.
Funding Division of Gynecologic Oncology at the University of British Columbia for funding support.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information.