Article Text
PDF
Abstract
Background In squamous cell vulvar cancer, sentinel node biopsy is accepted as standard treatment in well-defined patient groups and has reduced surgical morbidity considerably. Currently, due to the lack of evidence, it cannot be offered to patients with tumors of 4 cm diameter or greater or with multifocal tumors, or in local recurrences.
Primary objective This study is primarily a pilot and feasibility trial, aiming to evaluate if the prerequisites concerning detection rate and negative predictive value are satisfactory before the implementation of a multinational trial.
Study hypothesis Sentinel node biopsy has an acceptable negative predictive value and detection rate in the study cohort.
Trial design This study is planned as a prospective, national, multicenter interventional trial. Participating patients will undergo a sentinel node biopsy in addition to an inguinofemoral lymphadenectomy.
Inclusion and exclusion criteria Inclusion criteria: for women in group 1, a primary tumor ≥4 cm in diameter; in group 2, a multifocal primary tumor; in group 3, a local recurrence without previous inguinofemoral lymphadenectomy or radiation to the groins; in group 4, a local recurrence, with previous inguinofemoral lymphadenectomy and/or radiation to the groins.
Primary endpoint The primary endpoints are the detection rate and the negative predictive value of the sentinel node procedure.
Sample size In each of the four study arms, recruitment of 20–30 patients is planned.
Estimated dates for completing recruitment and presenting results Recruitment will take place between November 2019 and October 2021. Results will be available in December 2021.
Trial registration The trial is registered at “ClinicalTrials.gov” (ID: NCT04147780).
- vulvar and vaginal cancer
- sentinel lymph node
Statistics from Altmetric.com
Introduction
Vulvar cancer is a rare gynecological cancer, constituting about 5% of all cancers in the female genital tract.1 In Sweden, about 150 women are diagnosed with vulvar cancer every year. About 90% of these are squamous cell carcinomas and primary therapy is, in most cases, surgical, consisting of removal of the vulvar tumor and an inguinofemoral lymph node dissection. In advanced disease, either chemoradiation alone or surgery followed by chemoradiation are recommended.2 Recent studies have shown local recurrence rates of 25% and 36% after 5 years and 10 years of follow-up, respectively, in FIGO (International Federation of Gynecology and Obstetrics) stage I-II disease.3 Local recurrences are predominantly treated by surgery.4
Treatment-related morbidity is a matter of concern after vulvar cancer treatment, particularly regarding the morbidity caused by the inguinofemoral lymphadenectomy. Recent studies have described postoperative complications in up to 75% of patients undergoing inguinofemoral lymphadenectomy, for example lymphedema, lympho-cysts, infections, recurrent erysipelas and wound breakdown.5 As a result, the sentinel node concept gained attention in the treatment of vulvar cancer, aiming to reduce the high morbidity caused by inguinofemoral lymphadenectomy. In 2008 and 2012, two large multicenter studies demonstrated that sentinel node biopsy (SNB) was safe in early-stage squamous cell vulvar cancer.6 7 Comparison of the treatment-related morbidity of inguinofemoral lymphadenectomy and sentinel node biopsy showed significant advantages for the latter.8 Today, sentinel node biopsy is regarded as standard treatment of the groins in cases of unifocal, squamous cell vulvar cancers of less than 4 cm maximum diameter and without clinical or diagnostic imaging signs of inguinal lymph node involvement.2 Likewise, inguinofemoral lymphadenectomy is regarded as the standard treatment for large tumors and multifocal disease, or in cases of local recurrence when no treatment of the inguinal lymph nodes has been performed previously.2
Because subgroup analysis in previous studies showed a higher groin recurrence rate in patients with tumors of 4 cm diameter or greater, or with multifocal tumors, these cases are currently not considered suitable for a sentinel node biopsy.6 7 9 Until now, only a few studies have described the use of this technique in locally recurrent vulvar cancer.10 11 Thus, it is still not clear whether the status of local lymph nodes in locally recurrent vulvar cancer is predictive of survival outcome, or if it might act as a determinant for subsequent adjuvant therapy.
The aim of this study is to investigate the feasibility of sentinel node biopsy in patients with squamous cell vulvar cancer who are currently not regarded as suitable for receiving the sentinel node technique (that is, patients with tumors of 4 cm diameter or greater, or with multifocal tumors or locally recurrent disease). In addition, we want to investigate whether it is possible to identify a sentinel node after previous inguinofemoral lymphadenectomy or radiotherapy of the groins. A positive result from this pilot study might constitute the basis for a full-scale multinational trial in the future.
Methods
Trial design
This prospective, non-randomized interventional cohort study will investigate the feasibility of sentinel node biopsy in women with newly diagnosed primary or recurrent vulvar cancer, who currently do not qualify for the sentinel node technique. Eligible patients will undergo sentinel node biopsy as part of their surgical treatment (in most cases in addition to an inguinofemoral lymphadenectomy). The study was approved by the Swedish Ethical Review Authority (ID: 2019–04647).
Setting
The study is planned as a nationwide study in Sweden. Since 2017, the treatment of vulvar cancer patients has been accredited to four tertiary-referral university hospitals in Sweden: the Sahlgrenska University Hospital in Gothenburg, the Linköping University Hospital, the Skåne University Hospital in Lund, and the Karolinska University Hospital in Stockholm. Each center treats about 75 patients with primary or recurrent vulvar cancer, annually. All new patients with primary or recurrent vulvar cancer are discussed at a weekly video-linked national, multidisciplinary team conference, in which all four accredited centers participate. Eligibility for study inclusion will be evaluated at these conferences, and the participant informed consent and recruitment procedures will be conducted at the treating center.
Participants
The study comprises four patient groups, and in each of these groups the feasibility and reliability of the sentinel node biopsy will be investigated (Figure 1).
RecruitmentCanno flow chart.
The first and second groups consist of women with newly diagnosed primary squamous cell vulvar cancer, without clinical or diagnostic imaging signs of inguinal lymph node involvement or distant metastasis, and who are considered appropriate for primary surgical treatment (either with a tumor of 4 cm diameter or greater (group 1) or a multifocal tumor (group 2)). The third group consists of women with a first local vulvar cancer recurrence, who previously either underwent a sentinel node biopsy as part of their primary treatment or who did not have any treatment of the groins at all. Patients in these three groups will undergo surgical treatment comprising local tumor resection or vulvectomy and inguinofemoral lymphadenectomy in one or both groins, and an additional sentinel node biopsy. The fourth group consists of patients with a first local vulvar cancer recurrence who previously received surgery (comprising vulva resection or vulvectomy), inguinofemoral lymphadenectomy and/or radiation of the groins as primary treatment. These patients will undergo a sentinel node biopsy, provided that lymph scintigraphy reveals a sentinel node in the groin. If the scintigraphy does not reveal any lymph drainage, the groins will not be explored. If the scintigraphy reveals drainage to pelvic lymph nodes, no surgical excision of these will be performed.
Pre-operative work-up, surgical treatment of the tumor and possible adjuvant treatment will be performed according to the Swedish national guidelines for vulvar cancer (to be published in 2020). The sentinel node biopsy procedure will be conducted as described in the Swedish national guidelines and according to a study-specific protocol. Sentinel lymph node detection is facilitated by a radioactive tracer or a combination of radioactive tracer and blue dye (at the surgeon’s discretion). The pathological analysis of the sentinel nodes comprises standard hematoxylin-eosin staining and ultra-staging of the specimen, including keratin-specific immunohistochemical staining, according to the method described in the GROINSS-V-I protocol.6
Participating women will receive adjuvant treatment by radiotherapy (if possible, in combination with chemotherapy) to the groins and lower pelvis (in case of inguinal lymph node metastases), according to national guidelines. Adjuvant treatment to the vulva will be considered according to national guidelines, depending on risk factors such as tumor size, resection margins, lympho-vascular-space-invasion (LVSI) and multifocality, and determined at the post-operative national, multidisciplinary team conference.
Follow-up
All participants will be followed up clinically with visits, according to the national guidelines, every fourth month during the first 2 years and thereafter every sixth month (up to 5 years following treatment).
Endpoints
The primary endpoints are the sentinel node detection rate and the negative predictive value of the sentinel node biopsy (that is, the rate of patients undergoing the procedure and having a false-negative sentinel node).
Secondary endpoints are the number of retrieved sentinel lymph nodes and the proportion of metastases diagnosed by routine pathological work-up and by so-called ultra-staging, and the percentage of lymph nodes with micro-metastases or isolated tumor cells.
Sample size
The sample size estimation is based on the primary endpoint – that is, the negative predictive value for the sentinel node biopsy. Each of the three patient groups (that is, patients with tumors of 4 cm diameter or greater, patients with multifocal tumors and patients with a local recurrence) will be analyzed separately, aiming for a specific risk estimation for each of these three groups. A negative predictive value of 95% is regarded as an acceptable value, in accordance with the considerations and results described in the Gynecologic Oncology Group (GOG−173) study by Levenback et al.7 The fourth group (patients with a local recurrence, for whom primary treatment included inguinofemoral lymphadenectomy and/or radiation of the groins) is not included in the sample size estimation as only the detection rate, but not the negative predictive value, can be calculated in this case.
A sample size of at least 123 patients in each of groups 1–3 is needed to show a negative predictive value of 95% for the sentinel node biopsy, setting the limit for a type 1 error at 0.05 and a type 2 error at 0.20. In Sweden, about 60 patients are diagnosed with a local recurrence each year. Approximately 20–25 of these patients will be eligible for the study. Assuming an inclusion rate of 75%, it would take 7–8 years to recruit the necessary sample of 123 patients. Around 15–20 patients with a primary tumor ≥4 cm and 10–15 patients with multifocal primary disease will be eligible per year. Assuming the same inclusion rate of 75%, it would take 9–14 years to complete the study, thus implying a high risk of not reaching clinically relevant results within an appropriate time.
Given these constrains, this study is planned as a pilot and feasibility study that aims to evaluate whether a future multinational approach, with a full-scale trial, is feasible. We plan to include at least 20–30 patients in each of the four groups within 2 years. Recruitment will start in November 2019.
Statistical methods
Determination of the detection rate will be performed per groin and per patient. The negative predictive value is a function of the true negative and false negative sentinel lymph nodes (that is, the probability not to miss a lymph node metastasis) and thus it depends on both the sensitivity of the method and the risk of metastasis.
Discussion
With this study we intend to show that sentinel node biopsy is feasible even in larger primary tumors, in multifocal primary disease and in cases of local recurrence, even though the detection rate might be lower in recurrent than in primary disease. For large and multifocal primary tumors, we expect the same detection rate as documented for unifocal tumors up to 4 cm in diameter (87% per groin according to a recent meta-analysis).9 Studies in breast cancer and malignant melanoma have shown a lower detection rate and a higher percentage of atypical localization of the sentinel nodes in recurrent disease. On the other hand, it was still considered possible to gain further information from sentinel node mapping to predict prognosis and to optimize adjuvant treatment.12 13
To our knowledge only two published studies have investigated the sentinel node technique in local recurrence of vulvar cancer.10 11 In 2001 de Hullu et al described, in a case report, a woman with a local recurrence (after squamous cell carcinoma of the vulva) that was treated by a unilateral inguinofemoral lymphadenectomy and radiotherapy of the groins at primary diagnosis. Bilateral sentinel node biopsy was successfully performed at the time of the local recurrence. No follow-up data were provided.10 In 2016 van Doorn et al showed, in a highly selected sample of 27 patients, a detection rate of 84% per groin and 87% per patient, which were almost identical to values reported in an earlier meta-analysis on sentinel node biopsy in primary vulvar cancer.9 There were no data on the negative predictive value in van Doorn’s study, and no inguinal recurrences were observed during a median follow-up of 27 months.11 Since a false-negative sentinel node (that is, a sentinel node without metastasis when there is another lymph node in the groin with metastatic disease) almost inevitably leads to a later inguinal recurrence and a poor prognosis, both a high accuracy and a high negative predictive value for the sentinel node procedure are crucial.4
Previous secondary data analysis on the oncological safety of the sentinel node biopsy in larger or multifocal tumors have shown a higher rate of inguinal recurrences compared with unifocal tumors of less than 4 cm diameter, but most studies have suffered from a low number of events in the subgroups and from very heterogeneous or insufficiently-defined study cohorts.6 7 9 14 In some studies, patients with larger or multifocal tumors were included but not specifically documented or analyzed.9 14 To the best of our knowledge, no published studies have investigated (as a primary focus) the negative predictive value of sentinel node mapping in these patient groups. The 23 patients with multifocal tumors included in van der Zee et al’s GROINSS-V-I study showed a higher rate of inguinal recurrence, but no sample size or power calculations were performed for this specific subgroup. The endpoint of their study was the groin recurrence rate for the whole cohort, comprising both unifocal and multifocal tumors.6 In the Gynecologic Oncology Group (GOG−173) study, the 149 patients with tumors 4–6 cm in diameter showed (in a post-hoc subgroup analysis) a higher false-negative predictive value.7 In that study, neither pre-operative imaging nor specific surgical skills or ultra-staging of the sentinel nodes were mandatory, and the procedure was, in the beginning, performed partly without a radiotracer. Although Levenback and his colleagues reached an almost similar false-negative rate for tumors of less than 4 cm diameter to that reported in the GROINSS-V-I study,6 the above-noted circumstances might have influenced the results for larger tumors (with a higher prevalence of lymph node metastases) more than the results for smaller tumors. In addition, the sample size and power calculations were based on the whole cohort, comprising tumors both smaller and larger than 4 cm.
Our study will add further evidence, either confirming Levenback’s results or challenging them. It will also add important information on how the sentinel node can be used in cases of recurrence or prior groin treatment, and in multifocal disease. The sentinel node technique reduces surgical morbidity considerably, which is highly warranted and justifies another trial investigating the same patient group with a more stringent protocol. Taking these conditions into consideration, we chose both the detection rate and the negative predictive value as primary endpoints in the present study. We expect that, with the accuracy and proficiency offered by dedicated vulvar cancer treatment centers, we will obtain a satisfactory negative predictive value even for these patient groups. Furthermore, there is evidence that the sentinel node technique leads to an improved identification and resection rate for lymph nodes with very small metastases, leading to a higher proportion of correctly-staged patients who will therefore receive adequate adjuvant treatment. Analyzing the sentinel node with a specified protocol (ultra-staging) adds further accuracy.6 15 Our secondary endpoint will provide information about occult micro-metastases and will thus improve the accuracy of staging and treatment.
If the sentinel node biopsy procedure in this study is shown to be feasible and has acceptable safety in patients with tumors larger than 4 cm or with multifocal tumors, or in recurrent disease, it might promote a future multi-national study with the necessary sample size and power to confirm our findings.
Footnotes
Contributors Conception and study design: D Zach, P Kjölhede, L Moberg, K Stenström Bohlin, P Kannisto; Statistical calculations, data interpretation: D Zach, P Kjölhede; Manuscript writing: D Zach, P Kjölhede; Manuscript editing and reviewing: D Zach, P Kjölhede, K Stenström Bohlin, L Moberg, P Kannisto; Final approval of manuscript: D Zach, P Kjölhede, K Stenström Bohlin, L Moberg, P Kannisto.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.