Sentinel Node Biopsy for Vulvar/Cervical Cancer

The hypothesis of sentinel node biopsy is that, if one or more sentinel lymph nodes are negative for malignancy, the remaining regional lymph nodes will also be negative. At least in theory, if the sentinel node shows no evidence of metastatic involvement, a complete node dissection can be avoided.

Vulvar cancer

Vulvar cancer accounts for less than 5% of all gynecologic malignancies. ^[1] The American Cancer Society estimated that, in 2011, 4340 women would be diagnosed with vulvar cancer, with approximately 940 deaths. ^[2]

Overall, 30% of patients with vulvar cancer are diagnosed with metastatic disease to the inguinal or pelvic lymph nodes. ^[1] Lymph node status is the most important prognostic factor for survival in vulvar cancer. Independent risk factors for inguinal lymph node metastases include clinical lymph node status, tumor grade, lymph vascular space invasion, depth of invasion, and age. ^[1]

Unfortunately, assessment with clinical palpation of the inguinal lymph nodes is unreliable in identifying nodal disease. Of patients with nodes deemed clinically normal, 16%-24% have metastatic disease, while 24%-41% of those with nodes deemed to be involved clinically have negative histologic examination results. ^{[3, 4]}

The inaccuracy of physical examination in detecting inguinal node metastases has led to the use of various radiologic imaging techniques for assessment. However, no single imaging technique has been consistently reliable in the detection of node involvement. ^[3]

A systematic review of the medical literature from 1979-2004 compared the results of noninvasive and minimally invasive radiologic modalities versus histology from an inguinofemoral lymphadenectomy to determine groin node status. Index tests included ultrasonography with or without fine-needle aspiration, CT scanning, MRI, positron-emission tomography (PET) scanning, and sentinel node identification using blue dye and^99m Tc nanocolloid. Noninvasive tests were less accurate than invasive tests in predicting the groin node status; MRI was the most accurate, with a sensitivity of 86% and a negative likelihood ratio (LR) of 0.4. Sentinel node biopsy using^99m Tc nanocolloid had a pooled sensitivity of 97% and a negative LR of 0.21. This procedure is the most accurate test reviewed to date. ^[3]

Although sentinel node biopsy is accepted as a feasible alternative, many oncologists consider complete inguinofemoral dissection as the criterion standard for staging and diagnostic identification of metastatic disease. Surgical staging of vulvar cancer has changed dramatically since en bloc dissections were performed in the 1970s, which included resection of the vulva, mons pubis, and bilateral inguinofemoral and pelvic nodes in one large specimen. This radical dissection resulted in significant postoperative morbidity in many patients.

The current surgical standard of care for clinical stage I/II disease with more than 1-mm depth of invasion confined to the vulva involves radical wide local incision or complete radical vulvectomy with unilateral or bilateral inguinofemoral lymphadenectomy.

A retrospective analysis of 330 patients with vulvar cancer treated at the Mayo Clinic between 1955-1990 assessed disease-free interval and sites of failure in patients with squamous cell carcinoma treated with either unilateral or bilateral complete inguinofemoral node dissections. Approximately 34% of patients had involvement of the inguinofemoral nodes, while approximately 65% did not, which is consistent with other similar studies.

During the study period, 192 patients died—17.9% of disease and 3.3% of treatment complications, and the remainder from other or unknown causes. Treatment failures, defined as recurrences or metastases, were more common in patients with inguinal metastasis at the time of primary surgery. The recurrence rate during the first 2 years of follow-up in this group was 32.7%, as opposed to 5.1% in patients with no evidence of inguinal lymph node metastasis. No failures were detected in the groin after 24 months of follow-up; however, 67% of patients with groin recurrences died within the first 2 years of follow-up. Overall, 35.2% of the patients who experienced treatment failure presented 5 years after initial treatment. ^[5]

This retrospective study highlights the importance of recognizing groin involvement early in the disease process and its implications for adjuvant treatment options. Many other studies have also supported the conclusion that the most important prognostic factor in vulvar carcinoma is the status of the regional lymph nodes and the grave significance of groin recurrence. Therefore, the complete inguinofemoral node dissection, with its very low false-negative rate for identification of metastases, provides a criterion standard to which new surgical techniques can be compared.

The impetus to develop an alternative surgical technique is the significant morbidity associated with a complete inguinofemoral nodal dissection. The primary morbidities include lymphedema (25%-35%), infection (40%-60%), and wound breakdown (15%-25%). ^[1] However, any new procedure that reduces morbidity must do so without compromising detection of metastatic disease or increasing the risk of groin recurrence.

Clinical research in vulvar cancer has focused on validating the accuracy of sentinel node identification in the inguinal lymph nodes. Sentinel node biopsy is a well-established technique for breast cancer and melanoma in the literature. ^[6] Recent case series have recently suggested sentinel lymph biopsy as an alternative to complete inguinofemoral lymphadenectomy for stage I/II vulvar cancer.

In a US cost-effectiveness and outcomes analysis between sentinel node biopsy and inguinofemoral nodal dissection in early-stage vulvar cancer, Erickson et al reported lower treatment costs ($65.2 million vs $76.8 million, respectively) and complications costs ($23,711 vs $31.198, respectively) with similar outcomes (3-y inguinofemoral recurrence-free survival: 96.9% vs 97.3%, respectively) for sentinel node biopsy. ^[7]

Cervical cancer

Over 12,000 women were expected to be diagnosed with cervical cancer in the United States in 2011. ^[2]

Lymph node metastasis is the most important prognostic factor for recurrence and death and is a major determinant of treatment. ^[8] The standard surgical lymph node assessment in early cervical cancer is systemic pelvic lymphadenectomy, which is associated with significant short- and long-term morbidity, including prolonged operating time, increased blood loss, lymphocyst formation (20%), lymphedema (10%-15%), and, rarely, neural or vascular injury. ^[8]

In operable stage IB1 cervical cancer, the prevalence of lymph node metastases is estimated to be 15%-20%, which suggests that approximately 85% of patients who undergo a complete pelvic lymphadenectomy receive no direct benefit but are at risk of associated morbidities. ^[8] Use of sentinel node biopsy in patients with cervix cancer was developed to avoid a complete pelvic lymphadenectomy and its associated sequelae. ^[9] As discussed below, evidence suggests that sentinel node biopsy is a feasible method of lymph node assessment with a high detection rate and low false-negative rate. ^{[10, 11]}

Vulvar cancer

Appropriate candidates for sentinel node biopsy include patients with a squamous cell carcinoma of the vulva with more than 1 mm of invasion, a tumor of less than 4 cm in diameter, and no obvious metastatic disease on physical examination or imaging studies. ^{[12, 13]} Surgeons may consider some form of preoperative imaging to rule out grossly involved lymph nodes, the presence of which negates the effectiveness of sentinel node biopsy.

Patients with vulvar melanoma are also good candidates for sentinel node biopsy, and the specific selection criteria are the same as those for sentinel node biopsy in patients with cutaneous melanoma in other locations. ^[12]

The expert panel statement from the International Sentinel Node Society Meeting in 2008 recommended that sentinel node biopsy procedures for vulvar cancer be performed by a skilled multidisciplinary team. A gynecologic oncologist trained in the procedure, a nuclear medicine specialist, and surgical pathologist familiar with the concept of ultrastaging should be recruited in order to provide the highest likelihood of successful sentinel node dissection.

The expert panel statement recommends that each oncologist perform at least 10 consecutive cases with successful sentinel node identification and no false-negative results with at least 5-10 patients/year before performing the procedure without lymphadenectomy. ^{[12, 13]} Oncologists with limited annual exposure to patients with vulvar cancer should consider referral to physicians who care for a high volume of patients who are candidates for sentinel node biopsy. ^[12]

Vulvar cancer

Consider excluding patients who have undergone excisional biopsies, who have undergone previous resection of the primary lesion, or who have large or grossly infected tumors, as the lymphatic channels are often distorted. Similarly, the lymphatic channels of grossly metastatic lymph nodes may be obstructed by tumor cells, which may contribute to false-negative results.

Vulvar cancer

Establishment of sentinel node biopsy has evolved based on outcome data accumulated through review of several case series, which have indicated that the recurrence rate of tumor in the inguinal nodes is very low if the sentinel node(s) are negative for metastasis. ^{[1, 12, 13, 14, 15]} As is common with rare tumors, large randomized controlled trials comparing treatment strategies are not feasible.

Published data of groin relapse rates associated with sentinel node procedures versus complete inguinofemoral dissection vary. The panelists from the 2008 International Sentinel Node Society Meeting agree that the groin relapse rate is between 0% and 7%-8% for sentinel node biopsy, compared to the 0%-4% relapse rate for complete lymphadenectomy. ^[12] However, among patients with false-negative sentinel node biopsy findings, even when the false-negative result is detected early, the mortality rate ranges from 67%-75%. ^[5]

The largest study to date, by Van der Zee et al, reported a false-negative rate of 0%-2% among sentinel node procedures performed in 403 patients and 623 groins. In 259 patients with unifocal vulvar disease and negative sentinel node(s), 6 groin recurrences (referred to interchangeably as "false-negative results") were diagnosed (2.3%) and resulted in death in all 6 patients. The overall survival rate at 3 years was 97%. ^[13]

This study reported that both short-term and long-term morbidity were significantly decreased in patients who underwent only sentinel node biopsy. ^[13]

Short-term morbidity included wound breakdown or a wound infection requiring antibiotics. The wound breakdown rate (11.7%) and wound infection rate (4.5%) in patients who had undergone sentinel node dissection were significantly lower than in patients who underwent complete inguinofemoral lymphadenectomy (34% and 21.3%, respectively). Long-term morbidity was defined as lymphedema present at 2 visits more than 1 year after primary therapy and/or recurrent erysipelas of one episode requiring antibiotics. Patients in the sentinel node biopsy group experienced less long-term morbidity than those in the complete lymphadenectomy group (erysipelas, 0.4% vs 16.2%; lymphedema, 1.9% vs 25.2%). ^[13]

These rates of short- and long-term morbidity are comparable to those of other published results. This study indicates that patients with negative sentinel node biopsy findings experience minimal treatment-related morbidity and a low groin recurrence rate and have excellent survival rates. ^[13]

The Gynecology Oncology Group (GOG), understanding the limitations of offering sentinel node biopsy to all appropriate patients, recently published preliminary results of the GOG 173 trial, which was designed to estimate the sensitivity of sentinel node biopsy in a community-based setting. The sentinel node detection rate was 96%, but a false-negative rate of 9%-10% was noted. Morbidity and other outcome data have not been reported. ^[16]

Oonk et al reported no difference in quality of life (QOL) between patients who underwent sentinel node biopsy alone and those who underwent inguinofemoral lymphadenectomy. Patients who underwent inguinofemoral dissection reported more discomfort in groins, vulva, and legs and a higher rate of the need to wear stockings to control lymphedema. Patients who underwent sentinel node biopsy were more content with their treatment overall, but this did not translate into an improved QOL. There was no difference in sexual activity between the two groups, although the rates of sexual activity in patients with vulvar cancer are generally lower than that of the general population. This small study does not support the idea that a decrease in long-term morbidity also translates into an improved overall QOL among patients with vulvar cancer undergoing sentinel node biopsy. ^[17]

Interestingly, in the 2001 de Hullu et al study, patients were queried about their recommendations to other patients about sentinel node biopsy versus complete inguinofemoral lymphadenectomy. They were asked if they would recommend sentinel node biopsy to a relative or close friend if the sentinel node procedure had a 10% false-negative rate. Only about half of patients who underwent complete lymphadenectomy would recommend sentinel node biopsy to a relative or close friend, while 84% of patients who underwent sentinel node biopsy only would recommend it. The authors concluded that the patients who had a metastatic sentinel node and underwent a full inguinofemoral lymphadenectomy were less likely to advise sentinel node biopsy if there was a chance of missing metastatic disease. These patients were more willing to accept the complications of complete lymphadenectomy than to trust a new procedure that could potentially underestimate the extent of their disease. ^{[14, 17]}

Cervical cancer

The application of sentinel node biopsy in cervical cancer has been studied since 1999. Over the last decade, the technique has significantly improved, resulting in high sentinel node identification rates and high negative predictive values. ^[18] In approximately 800 patients who have undergone sentinel node mapping reported in the literature, the sentinel node detection rate has been quoted at 90% with a 92% sensitivity for identification of metastatic disease and a 97% negative predictive value. ^{[19, 20]}

In 2010, Gortzak-Uzan et al noted a higher rate of pelvic lymph node metastases in patients who underwent sentinel node biopsy (87 patients) versus a matched control group of 1277 patients (17% vs 7%). None of the cases with negative sentinel nodes showed malignant cells in the subsequent permanent sections, yielding a negative predictive value of 100%. There were 5 recurrences in the sentinel node group (mean follow-up of 13 mo) compared to 16 recurrences in the control group (median follow-up of 59 mo), suggesting that the detection rate of lymph node metastases with sentinel node biopsy is more than double that of complete lymphadenectomy. The higher rate of metastases noted in sentinel node procedures may be secondary to the practice of ultrastaging, identification and removal of sentinel nodes in aberrant locations, or by ensuring complete removal of positive sentinel nodes at the time of the primary procedure. ^[15]

The SENTICOL study, a large prospective study of 139 patients in France, found that 16.9% of patients with stage IA1 to IB1 disease had sentinel node metastases. Using the dual tracer technique, a sensitivity of 92% was observed for the identification of nodal metastases by positive sentinel nodes, while a negative predictive value of 98.2% and a false-negative rate of 8% were noted. When analyzing patients with sentinel nodes extracted bilaterally, the sensitivity for metastatic node detection was 100%. These sensitivity results are higher than the commonly quoted sensitivities of PET scan, MRI, or CT scanning cited in patients with cervical cancer. ^{[21, 22]}

Vulvar cancer

Several studies have cited the painful peritumoral injections of radiocolloid in the vulva several hours to one day prior to surgery as a disadvantage of the current sentinel node technique.

Crane et al (2011) published a case study investigating the use of intraoperative near-infrared fluorescence imaging as a new technique for sentinel node detection in which a fluorescence agent is injected in the routine fashion after anesthesia is induced. Upon excitation with a laser beam, the agent emits light, which is captured and processed by a fluorescence camera. Real-time images can then be displayed on a monitor in the operating room. The superficial localization of most inguinal lymph nodes may enable transcutaneous lymphatic mapping in the groin; however, this method may be limited in obese patients with a greater distance between the skin and the sentinel node bed.

In vivo, 26 of 29 radioactive sentinel nodes could be detected with fluorescence alone; however, the addition of radiocolloid and a gamma probe allowed for detection of a sentinel node in every groin. ^[23] This is an excellent example of how additional injectable agents may improve the recognition of sentinel nodes in vulvar cancer.

A retrospective study by Katz et al (2003) revealed that radiation therapy alone or in combination with lymphadenectomy in patients with positive nodes was highly effective in preventing groin recurrence. The extent of the lymph node dissection ranged from node excision to complete inguinofemoral dissection. ^[24]

An ongoing study from van der Zee and others, the Groningen International Study on Sentinel Nodes in Vulvar Cancer-II (GOINSS-V-II), will observe outcomes among patients with a metastatic sentinel node who receive radiotherapy instead of inguinofemoral lymphadenectomy. There is optimism this treatment plan will further reduce the morbidity of treating node-positive patients with both a complete lymphadenectomy and radiation. ^{[13, 24]} The proposed end date has been extended to 2017. ^[25]

Cervical cancer

Cormier et al (2011) have proposed a surgical algorithm to evaluate patients with cervical cancer based on the outcome results in their case series. The algorithm includes the following:

• Sentinel nodes are removed and, if routine hematoxylin and eosin staining results are negative, they are submitted for ultrastaging.

• Any suspicious lymph nodes are removed, regardless of mapping results.

• If there is no mapping in a hemipelvis, a side-specific lymphadenectomy including interiliac or subaortic nodes is performed, along with an en bloc parametrectomy with resection of the cervical tumor.

If the algorithm above had been applied, 6.6% of patients with failed mapping would have required complete lymphadenectomy, and 18.9% of patients with a unilateral sentinel node identified would have undergone a contralateral complete lymphadenectomy. Most patients (74.6%) with optimal mapping would have avoided complete bilateral pelvic lymphadenectomy. Theoretically, the false-negative rate would be 0% and the sensitivity and negative predictive value would have been 100%. This algorithm needs verification in a prospective study to ensure its reliability prior to application in the general population. ^[8]

Although several case series and prospective studies have addressed this issue in sub-analyses, the outcomes have been variable, and the question remains whether each pelvic nodal basin (right and left) should be considered as one or two separate units. For example, some series have reported proceeding to a pelvic lymphadenectomy when a sentinel node failed to map to that side, while some series claim that not enough evidence supports proceeding with full dissection. Hauspy et al (2007) argue that the sentinel node should be evaluated per side, and, following this protocol, the false-negative rate can be minimized. ^[26] Conversely, other investigators believe that fewer patients will benefit from the less-radical procedure if each side is evaluated individually.