2331
Vulvar Carcinoma The Price of Less Radical Surgery
Joanne A. de Hullu, Ph.D.1 Harry Hollema, Ph.D.2 Sietske Lolkema, M.D.1 Marike Boezen, Ph.D.3 Henk Boonstra, Ph.D.4 Matthe P. M. Burger, Ph.D.5 Jan G. Aalders, Ph.D.1 Marian J. E. Mourits, Ph.D.1 Ate G. J. van der Zee, Ph.D.1
BACKGROUND. The objective of this study was to determine whether modifications in the treatment of patients with vulvar carcinoma influence the rates of recurrence and survival. METHODS. Between 1982 and 1997, 253 patients with T1 and T2 invasive squamous cell carcinoma of the vulva were treated by essentially the same team of gynecologic oncologists, and 168 patients (Group I) underwent radical vulvectomy with en bloc inguinofemoral lymphadenectomy. Standard therapy was changed in 1993, and 85 patients (Group II) underwent wide local excision with inguinofemoral lymphadenectomy through separate incisions. The rates of recurrence and survival were compared between both groups.
1
RESULTS. In Group II, the overall recurrence rate (33.3%) within 4 years was
Department of Gynecologic Oncology, University Hospital Groningen, Groningen, The Netherlands.
2
Department of Pathology, University Hospital Groningen, Groningen, The Netherlands.
3
Department of Epidemiology and Statistics, University Groningen, Groningen, The Netherlands.
4
Department of Gynecologic Oncology, University Hospital St. Radboud, Nijmegen, Nijmegen, The Netherlands.
5
Department of Gynecologic Oncology, Academic Medical Center and The Netherlands Cancer Institute, Amsterdam, The Netherlands.
increased compared with Group I (19.9%; P ⫽ 0.03). In Group II, 5 of 79 patients (6.3%) developed fatal groin or skin bridge recurrences compared with 2 of 159 patients (1.3%) in Group I (P ⫽ 0.029); this did not result in a difference in overall survival. In Group II, 40 of 79 patients had tumor free margins measuring ⱕ 8 mm, resulting in 9 local recurrences; whereas 39 of 79 patients had tumor free margins measuring ⬎ 8 mm, resulting in no local recurrences (P ⫽ 0.002). CONCLUSIONS. The current study showed that fatal recurrences in either the groin or the skin bridge were more frequent after wide local excision and inguinofemoral lymphadenectomy through separate incisions; however, probably due to lack of power, this did not result in shorter survival. In 40 of 79 patients, the histologic margins measured ⱕ 8 mm, resulting in a high risk of local recurrences. Therefore, the authors recommend obtaining surgical margins of 2 cm for the local treatment of patients with vulvar carcinoma. Cancer 2002;95:2331– 8. © 2002 American Cancer Society. DOI 10.1002/cncr.10969
KEYWORDS: vulvar carcinoma, separate incisions, skin bridge recurrence, surgery, recurrence rate, survival.
C Address for reprints: J. A. de Hullu, Ph.D., Department of Gynecologic Oncology, University Hospital Groningen, P.O. Box 30001, 9700 RB Groningen, The Netherlands; Fax: 011-31-503611806; E-mail:
[email protected] Received May 24, 2002; revision received June 21, 2002; accepted July 9, 2002. © 2002 American Cancer Society
arcinoma of the vulva is a rare disease and accounts for only 3–5% of all gynecologic malignancies.1 About 90% of carcinomas of the vulva are of the squamous cell type.2,3 The primary way vulvar carcinoma is spread, is lymphogenic to the inguinofemoral lymph nodes. Surgery is the first choice for the treatment of patients with squamous cell carcinoma of the vulva. Overall survival figures rose from 20% to over 60% after the introduction of radical vulvectomy with en bloc bilateral inguinofemoral and pelvic lymphadenectomy instead of simple local excision.4,5 However, although overall survival figures were excellent, the short-term and long-term morbidity associated with this procedure were substantial.3 Complications were related either to radical vulvectomy or to lymphadenectomy. Postoperative wound breakdown and infection occurred with a frequency of
2332
CANCER December 1, 2002 / Volume 95 / Number 11
up to 85%, often prolonging hospitalization.2 Psychosexual disturbance and lymphedema of the legs were major long-term problems, especially in young patients.6,7 In an attempt to decrease the incidence of complications, a search was made for modifications of the surgical procedure without compromising the prognosis.3,8 The replacement of radical vulvectomy by less radical wide local excision is one of these modifications. Hacker et al. summarized 11 studies and found no significant increase in recurrence or mortality rates after wide local excision in patients with T1 tumors (greatest dimension ⱕ 2 cm).3 However, that review contained only retrospective studies on selected groups of patients, which may have lead to bias. The introduction of inguinofemoral lymphadenectomy through separate incisions as alternative to the en bloc approach was another modification and was introduced by Byron et al.9 The rationale was that the dissemination of disease to the inguinofemoral lymph nodes does not occur in continuity but by embolization.10 Preservation of a skin bridge resulted in a lower frequency of postoperative wound breakdown and a lower risk of developing lymphedema of the legs.3 However, seroma and lymphedema of the legs still were observed frequently, whereas only occasional tumor recurrences in the skin bridge have been reported.11–16 The Cochrane Collaboration proposed five levels of evidence (Levels 1–5) for ranking the validity of different types of studies.17 No prospective, randomized studies (Level 1) have ever been performed that compared radical surgery with modified radical surgery for patients with vulvar carcinoma. However, in the recent Cochrane review on surgical interventions for patients with early-stage squamous cell carcinoma of the vulva, the previous conclusion of Hacker et al. that wide local excision is safe was confirmed based on observational studies. No studies with respect to the use of separate incisions met the minimum criteria set by the Cochrane Collaboration.18 Because the skin bridge recurrence rate is the main issue with respect to the safety of this treatment modification, all patients who underwent the procedure with separate incisions were pooled, and, with a skin bridge recurrence rate of 1% to date, the use of separate incisions also was deemed safe.18 A possible alternative to a randomized, controlled trial to evaluate the impact of changes in therapy is to compare therapy outcomes in a series of patients who were treated according to an individualized, less radical protocol with therapy outcomes in a series of patients who were treated according to a more classic radical protocol (Level 3), preferably from the same
institution and by the same group of gynecologic oncologists. In a previous study from our institute, we reported on the outcome of patients with vulvar carcinoma who were diagnosed between 1982 and 1992 with primary T1 and T2 tumors and who were treated with standard, radical vulvectomy with en bloc inguinofemoral lymphadenectomy.19 In 1993, a radical change to more individualized treatment was made by essentially the same team of gynecologic oncologists. The objective of the current study was to analyze the consequences of these treatment modifications with respect to recurrence rates and survival.
MATERIALS AND METHODS Patients Clinical and histopathologic data on all consecutive patients with vulvar carcinoma in the northeast of The Netherlands who are referred to the Groningen University Hospital for treatment are stored prospectively in a computerized data base. For the current study, all patients with primary invasive squamous cell carcinoma of the vulva who were referred between January 1982 and October 1997 were identified. Between 1982 and 1992, the standard treatment for patients with vulvar carcinoma consisted of radical vulvectomy with en bloc bilateral inguinofemoral lymphadenectomy (Group I). Standard treatment (by essentially the same team of gynecologic oncologists) was changed in 1993 (Group II), when wide local excision became the standard treatment, and a modified radical vulvectomy was performed only in patients with multifocal tumors and/or abnormal aspects of the remainder of the vulva. Unilateral inguinofemoral lymphadenectomy through separate incisions was performed in patients with lateralized tumors (lesions that did not cross the midline with the medial margin of the tumor measuring ⬎ 1 cm from the midline structures20), whereas all other patients underwent bilateral inguinofemoral lymphadenectomy through separate incisions. Modified radical vulvectomy with en bloc inguinofemoral lymphadenectomy was performed in patients with suspicious lymph nodes in the groin areas upon palpation. In 1988, the staging classification system of the International Federation of Gynecology and Obstetrics (FIGO) for vulvar carcinoma was changed from clinical staging to surgicopathologic staging.21 To be able to compare both groups in this study, all patients were staged in retrospect based on final histopathologic examination. During both periods, postoperative external beam radiotherapy was delivered to the vulva, groins, and pelvis up to a total dose of 45 grays (Gy; 25 fractions of 1.8 Gy daily) when two or more positive lymph nodes or one lymph node metastasis with ex-
Less Radical Surgery in Vulvar Carcinoma/de Hullu et al.
tranodal growth were identified. When invasive tumor was found in the margins, a local reexcision was performed unless postoperative radiotherapy already had been planned based on lymph node metastases.
Surgical Techniques Radical vulvectomy with en bloc inguinofemoral lymphadenectomy was performed with classic butterfly incisions and was similar to the technique described by Monaghan.22 For a wide local excision, the incision on the vulva was carried down to the inferior fascia of the urogenital diaphragm. The intention was to obtain surgical tumor free, lateral margins of at least 1 cm.3 Modified radical vulvectomy was performed as described by Burke et al.23: an intermediate resection between the classic radical vulvectomy and wide local excision. Two elliptic incisions were made: one incision placed at the vaginal introitus and an outer incision placed at the labiocrural folds and brought across the mons pubis and perineal body. This modified radical vulvectomy preserved the radicality of the vulvar resection while retaining skin over the groins. Separate incisions for inguinofemoral lymphadenectomy were made parallel to the inguinal ligament; then, the lymph node-bearing fat pad was removed. The extent of the dissection was the inguinal ligament cephalad, the adductor longus muscle medially, and the sartorius muscle inferolaterally. After opening the cribiform fascia, all lymph node-bearing fatty tissue medial from the femoral vein was removed as well.3
2333
Diagnosis and Treatment of Recurrences All lesions suspect for recurrence were confirmed histologically, when possible. The treatment of choice for recurrent disease was dependent on previous therapy and the site of recurrence. In general, patients with locoregional recurrences underwent surgery with or without radiotherapy, whereas patients with pelvic or distant metastases were treated with chemotherapy (5-FU and carboplatin) if they were in good general condition.
Histopathology For the current study, all slides of tumors were reviewed by an experienced gynecopathologist (H.H.). The tumors were reexamined for: greatest tumor dimension (mm), depth of invasion (with a calibrated micrometer, depth of invasion was measured in mm from the stratum granulosum to the deepest point of the tumor), degree of differentiation (good, moderate, or poor; when the percentage of undifferentiated cells was ⬍ 5%, the tumor was scored as well differentiated; tumors with 5–50% undifferentiated cells were scored moderately differentiated; and tumors with ⬎ 50% undifferentiated cells were scored poorly differentiated),25 and the presence of lymph/vascular invasion (defined as the presence of tumor emboli in endothelial-lined spaces). In Group I, only sections of the margins from the specimen were taken to exclude tumor in these margins. In Group II, sections were taken such that the exact lateral and basal tumor free margins (in mm) could be determined. The pathologist was not aware of the results of treatment and follow-up of the patients.
Follow-Up Standard follow-up surveillance consisted of a clinical history and physical examination. For follow-up, each patient visited our outpatient clinic at gradually increasing intervals up to at least 10 years. For the current study, all data that were stored in the data base were checked in the case notes. If necessary, follow-up data were completed by contacting the general practitioner, the patient herself, or one of the relatives of the patient. The final date of the analysis was October, 2001. Recurrences were registered as local, groin, skin bridge, pelvic, or distant. To be capable of comparing both groups with respect to recurrence rates, two check points of evaluation of recurrences were used for this study: 2 years after primary treatment (because, based on the literature, most nonlocal recurrences develop within 2 years19,24) and 4 years after primary treatment (because the last patient from Group II who was analyzed for the current study was treated 4 years before the date of the final analysis).
Statistical Analysis Data analysis was performed using SPSS software (release 10.0; SPSS Inc., Chicago, IL). For the analyses of recurrence rates within 2 years and 4 years respectively, patients with follow-up ⬍ 24 months and ⬍ 48 months, respectively, and patients who died from intercurrent disease were excluded from the analyses. Survival was estimated in months from the date of surgery to the date of either death or last follow-up. Four-year survival was calculated using the Kaplan– Meier method. Intercurrent deaths were censored. Differences between characteristics were tested with two-sided chi-square tests using a Yates correction for small numbers, when necessary. Differences in continuous characteristics, like age, were tested with a t test. P values ⱕ 0.05 were considered significant. Logistic regression models were used to estimate the associations between histopathologic and/or clin-
2334
CANCER December 1, 2002 / Volume 95 / Number 11
ical factors and the presence of lymph node metastases, recurrences, and 4-year survival, respectively.
RESULTS Patient Characteristics Between January, 1982 and October, 1997, 333 patients with primary invasive squamous cell carcinoma of the vulva were referred to the Gynecologic Oncology Department. Fifty of 333 patients (36 patients in Group I and 14 patients in Group II) were treated with primary radiotherapy (with or without chemotherapy) because of extensive disease and/or tumors that were not confined to the vulva (T3 or T4 according to the TNM classification system). Thirty of 283 patients with T1 tumors (measuring ⱕ 2 cm in greatest dimension) and T2 tumors (measuring ⬎ 2 cm in greatest dimension; 17 patients in Group I and 13 patients in Group II) did not undergo inguinofemoral lymphadenectomy: 11 patients with microinvasive disease (depth of invasion ⱕ 1 mm), 17 patients with poor general health, 1 patient with a past history of multiple primary epithelial tumors, and 1 patient who had been treated previously for cervical carcinoma with radical surgery and postoperative radiotherapy. These patients were excluded from the analysis in the current study. The study population included 253 patients with primary T1 and T2 invasive squamous cell carcinomas who were appropriate candidates for surgery: 168 patients in Group I and 85 patients in Group II. The median age of all 253 patients at the time of diagnosis was 71 years (range, 28 –94 years).
Treatment All 168 patients in Group I underwent radical vulvectomy; whereas, in Group II, 45 patients underwent wide local excision, and 30 patients underwent modified radical vulvectomy. In Group I, radical vulvectomy was combined with en bloc bilateral inguinofemoral lymphadenectomy in 159 of 168 patients and with en bloc unilateral lymphadenectomy in 9 patients. In Group II, 2 patients underwent en bloc bilateral lymphadenectomy because of suspicious lymph nodes; 58 patients underwent bilateral lymphadenectomy, and 25 patients underwent unilateral lymphadenectomy through separate incisions. In Group I, 51 of 168 patients (30%) had lymph node metastases, a rate that was comparable to the rate for Group II, with 31 of 85 patients (37%) who had lymph node metastases (P ⫽ 0.32). No differences were found between Group I and Group II with respect to age, greatest tumor dimension, lymph/vascular invasion, depth of invasion, or grade, as shown in Table 1. Table 2 shows that all histopathologic factors
TABLE 1 Frequency of Clinicopathologic Characteristics in Both Groupsa Characteristic
Group I
Group II
P value
Total patients Tumor status T1 (ⱕ 2 cm) T2 (⬎ 2 cm) Lymph/vascular invasionb No Yes Gradec Well differentiated Moderately/poorly differentiated Depth of invasion (mm)d Median Range Age (yrs)d Median Range
168
85
—
53 115
25 60
— 0.728
133 29
70 13
— 0.660
67 99
33 52
— 0.814
5 1–40
5 1–27
— 0.75
71 28–94
71 34–94
— 0.66
a
Group I included 168 patients who underwent radical vulvectomy with en bloc inguinofemoral lymphadenectomy. Group II included 85 patients who underwent wide local excision with inguinofemoral lymphadenectomy through separate incisions. b Data on lymph/vascular invasion were missing for eight patients. c Data on tumor grade were missing for two patients. d Variables for depth of invasion and age are continuous.
were related independently to the presence of lymph node metastases. Four patients in Group I did not have tumor free margins: One patient died postoperatively of an intercurrent disease, one patient died due to poor general health and complications shortly after the operation before starting the radiotherapy, and two patients also had bilateral lymph node metastases and underwent postoperative radiotherapy (one patient developed a local and groin recurrence, and one patient is without evidence of disease). In Group II, two patients had no tumor free margins: One patient had a second microinvasive tumor in the lateral margin and did not develop a recurrence during close follow-up; and the other patient also had bilateral lymph node metastases, underwent postoperative radiotherapy, and died of distant metastases. In all other 247 patients, margins were free. In Group II, the exact tumor free margins were measured: In 40 patients, both tumor free margins were ⬎ 8 mm; whereas, in 45 patients, either the lateral margins or the basal margins were ⱕ 8 mm.
Follow-Up General data For all 253 patients, the median follow-up was 110 months (range, 3–220 months). Four patients who were without disease after 15 months, 26 months, 28 months, and 30 months, respectively, were lost to
Less Radical Surgery in Vulvar Carcinoma/de Hullu et al.
2335
TABLE 2 Logistic Regression Model for Histopathologic Factors in Relation to the Presence of Lymph Node Metastases and Histopathologic Factors and Treatment Group in Relation to Recurrence and Survival of Patients with Vulvar Carcinoma: Odds Ratios with 95% Confidence Intervals Risk of lymph node metastases
Overall recurrences within 2 yrs
Risk of death within 4 yrs related to vulvar carcinoma
Overall recurrences within 4 yrs
Histopathologic or clinical factor
OR
95% CI
OR
95% CI
OR
95% CI
OR
95% CI
Tumor sizea Depth of invasion (continuous) Lymph/vascular invasionc Graded Treatment groupe
3.73 1.08 2.23 2.82 —
1.64–8.51b 1.01–1.15b 1.05–4.74b 1.45–5.48b —
6.50 1.05 4.96 3.03 2.29
1.40–30.17b 0.97–1.13 1.99–12.37b 1.10–8.34b 1.00–5.28b
1.66 1.06 3.74 3.66 2.272
0.67–4.15 0.99–1.13 1.62–8.64b 0.18–1.70 1.11–4.67b
10.94 1.06 5.20 4.37 1.98
1.34–89.11b 0.97–1.15 2.03–13.32b 1.33–14.39b 0.82–4.80
OR: odds ratio; 95%CI: 95% confidence interval. a Seventy-eight patients had T1 tumors (measuring ⱕ 2 cm in greatest dimension), and 175 patients had T2 tumors (measuring ⬎ 2 cm in greatest dimension). b P ⬍ 0.05. c There were 203 patients without lymph/vascular invasion and 42 patients with lymph/vascular invasion. d One hundred patients had well-differentiated tumors, and 151 patients had poorly or moderately differentiated tumors. e Group I included 168 patients who underwent radical vulvectomy with en bloc inguinofemoral lymphadenectomy, and Group II included 85 patients who underwent wide local excision with inguinofemoral lymphadenectomy through separate incisions.
follow-up. Within 4 years after primary treatment, 34 patients died of intercurrent disease. The follow-up for all other patients was at least 4 years.
Recurrences within 2 years Fifteen of 253 patients were excluded from this analysis of recurrences within 2 years (1 patient was lost to follow-up, and 14 patients died of intercurrent disease without evidence of disease within 24 months after primary surgery). Thus, 238 patients were eligible for this analysis. An overview of the recurrences within 2 years in both groups is given in Table 3. In Group I, 21 of 159 (13.2%) patients developed recurrent disease; whereas, in Group II, 17 of 79 patients (21.5%) developed recurrent disease (P ⫽ 0.09). In multivariate analysis, after correction for tumor dimension, depth of invasion, presence or absence of lymph/vascular invasion, and grade, the treatment for Group II was associated with a higher risk of developing recurrences (P ⫽ 0.05). For the associations of histopathologic factors and treatment (Group II vs. Group I) with respective recurrence rates and survival, see Table 2. Two of 159 patients in Group I (1.3%) developed groin recurrences, whereas in 5 of 79 patients in Group II (6.3%) developed groin or skin bridge recurrences (P ⫽ 0.029). All seven patients who developed recurrences died of disease. No groin recurrences in contralateral groins were diagnosed in 34 patients after they underwent unilateral lymphadenectomy. Twelve 12 of 159 patients in Group I (7.5%) patients and 9 of 79 patients in Group II (11.4%) devel-
TABLE 3 Site of 38 Recurrences within 2 Years in Both Groups of Patientsa Site(s) of recurrence
Group I
Group II
Total no. of patients
Local Groin Skin bridge Pelvis Distant Local/groin Local/distant Skin bridge/distant Groin/pelvis Total
11 1 — 5 3 1 — — — 21
7 1 1 2 2 1 1 1 1 17
18 2 1 7 5 2 1 1 1 38
a
Group I included 168 patients who underwent radical vulvectomy with en bloc inguinofemoral lymphadenectomy. Group II included 85 patients who underwent wide local excision with inguinofemoral lymphadenectomy through separate incisions.
oped local recurrences (P ⫽ 0.32). An analysis of the exact tumor free margins in Group II showed that no patient with histologic tumor free margins measuring ⬎ 8 mm (n ⫽ 39 patients) developed a local recurrence, whereas 9 of 40 patients (22.5%) with at least 1 tumor free margin measuring ⱕ 8 mm developed local recurrences (P ⫽ 0.002).
Recurrences within 4 years For an analysis of recurrences within 4 years, 38 of 253 patients were excluded (4 patients were lost to followup, and 34 patients died of intercurrent disease within 48 months after primary surgery). Thus, 215 patients were eligible for this analysis. Of 215 patients, 52 pa-
2336
CANCER December 1, 2002 / Volume 95 / Number 11
tients (24.2%) developed any recurrence. In Group I, 29 of 146 patients (19.9%) developed recurrent disease compared with Group II, in which 23 of 69 patients (33.3%) developed recurrent disease (P ⫽ 0.03). The additional 14 detected recurrences between 2 years and 4 years after primary surgery all were local recurrences (8 patients in Group I and 6 patients in Group II [3 patients after undergoing wide local excision and 3 patients after undergoing modified radical vulvectomy]) The results of all other analyses were comparable with recurrences after 2 years.
Survival Off all 66 patients who developed recurrences, 37 died of disease, with a median follow-up of 5 months (range, 1–142 months) from the date of first recurrence until the date of death from disease. Eighteen patients died of intercurrent disease, and 11 patients are alive without evidence of disease at a median follow-up of 110 months (range, 50 –210 months). There was no difference in overall survival between Group I and Group II.
DISCUSSION Although randomized studies have never been performed to investigate whether wide local excision with inguinofemoral lymphadenectomy through separate incisions is as safe as radical vulvectomy with en bloc inguinofemoral lymphadenectomy in the treatment of patients with early-stage squamous cell carcinoma of the vulva, these modifications generally are considered safe.26 –28 The current study is the first large, single-institute study that has analyzed the consequences of an abrupt change in the treatment of patients with vulvar carcinoma who were treated by essentially the same team of gynecologic oncologists. Our study shows a small but significant increase in skin bridge and groin (regional) recurrence rates after the separate incisions approach compared with the en bloc approach. For the current study, patients with groin recurrences and skin bridge recurrences were analyzed together, because both are true regional recurrences, and patients in both categories have a high likelihood of dying of disease. In 1981, Hacker et al. published results from 100 patients after separate incisions and observed no groin recurrences and 2 skin bridge recurrences, both in patients with lymph node metastases.11 In addition, in patients without lymph node metastases, incidental skin bridge recurrences have been reported, whereas groin recurrences are very rare.3,12–16 Despite salvage therapy, all seven patients with skin bridge and groin recurrences in our study died of disease. However, the small increase in the number of fatal skin bridge and
groin recurrences did not result in overall survival differences between the two groups, probably due to a lack of power. To show survival differences of approximately 5% in, for example, randomized trials, ⬎ 500 patients in each arm would be necessary. The current study illustrates that the skin bridge recurrence, a new type of nonlocal recurrence since the introduction of separate incisions, is difficult to treat and compromises prognosis. Compared with our experience, others have reported that the early diagnosis and removal of skin bridge recurrences in combination with radiotherapy may salvage these patients.15,16 It is important to identify the patients who are at particularly high risk of developing recurrent disease in the tissue bridge between the vulva and the groin. It has been suggested that, in patients with metastatic lymph nodes largely replaced by tumor, stasis of lymph flow may lead to retrograde tumor permeation; however, until now, there has been no clear explanation for skin bridge recurrences in patients with negative lymph nodes. In our study, two patients developed skin bridge recurrences 4 months and 5 months after their primary treatment, which consisted of wide local excision and unilateral inguinofemoral lymphadenectomy through separate incisions (negative lymph nodes). Due to the low incidence of skin bridge recurrences, it is not possible currently to preoperatively select those patients with a higher risk for skin bridge recurrences other than patients with macroscopically involved, metastatic lymph nodes in the groin(s). In these patients, radical vulvectomy with en bloc inguinofemoral lymphadenectomy generally is accepted as the standard treatment. Despite the increase in the combined skin bridge and groin recurrence rate among the patients in Group II, it has to be realized that the combined incidence of skin bridge and groin recurrences remains very low and concerns an increase in the recurrence rate from 1% up to 6%. A reduction in complications (wound breakdown, leg edema) and length of hospitalization after less radical treatment has been reported in different studies of less radical therapy in patients with vulvar carcinoma.29,30 Among our patient population, treatment morbidity and quality of life were not registered, although our clinical impression also points to less morbidity after the change in treatment. Thus, considering the pros (less morbidity) and cons (small increase in number of skin bridge and groin recurrences) of the separate incisions, in our opinion, the current treatment policy should be continued. Two other changes in the treatment of patients with vulvar carcinoma may reduce morbidity in the future: Zhang et al. recently reported a ⬎ 30% reduc-
Less Radical Surgery in Vulvar Carcinoma/de Hullu et al.
tion in lymphedema after routinely preserving the saphenous vein during inguinofemoral lymphadenectomy.31 Moreover, the sentinel lymph node procedure seems to be a good diagnostic tool with which to select patients for whom it is safe to omit complete inguinofemoral lymphadenectomy.32 In both groups, all nonlocal recurrences were diagnosed within 2 years; whereas, after 2 years, only local recurrences were observed. The risk of local recurrence was increased in patients who had tumor free margins measuring ⱕ 8 mm. These results are comparable to those from the study by Heaps et al., who found a 50% local recurrence rate in patients with margins measuring ⬍ 8 mm.33 In our study, 50% of patients had histologic tumor free margins measuring ⱕ 8 mm despite the intended surgical margins of 1 cm. The high percentage of patients with histologic margins ⱕ 8 mm may be related to shrinkage due to fixation. In earlier morphometric studies in the cervix, we found that shrinkage of 15% of the specimen had to be taken into consideration.34 Based on our current results, we have decided to change the intended surgical margins from 1 cm to 2 cm. This single-institution study of 253 patients with early-stage vulvar carcinoma showed that wide local excision and unilateral or bilateral inguinofemoral lymphadenectomy through separate incisions leads to a small but significant increase in the overall recurrence rate compared with the en bloc approach. In particular, the number of combined skin bridge and groin recurrences was higher after individualized treatment, although the power of this study was not sufficient to prove that the modifications in treatment may compromise overall survival. This study also showed that surgical margins measuring 1 cm resulted in histologic margins measuring ⱕ 8 mm in 50% of patients, leading to a higher local recurrence rate. Therefore, surgical margins of 2 cm should be considered to minimize the incidence of local recurrences.
7. 8. 9.
10. 11.
12.
13.
14.
15.
16. 17. 18.
19.
20.
21.
22.
REFERENCES 1. 2.
3.
4. 5. 6.
Hunter DJS. Carcinoma of the vulva: a review of 361 patients. Gynecol Oncol. 1975;3:117–123. Sturgeon SR, Brinton LA, Devesa SS, Kurman RJ. In situ and invasive vulvar cancer incidence trends (1973 to 1987). Am J Obstet Gynecol. 1992;166:1482–1485. Hacker NF. Vulvar cancer. In: Berek JS, Hacker NF, editors. Practical gynecologic oncology. Baltimore: Williams & Wilkins, 2000:553–596. Taussig FJ. Cancer of the vulva: an analysis of 155 cases. Am J Obstet Gynecol. 1940:40;764 –779. Way S. Carcinoma of the vulva. Am J Obstet Gynecol. 1960; 79:692– 697. DiSaia PJ, Creasman WT, Rich WM. An alternative approach to early cancer of the vulva. Am J Obstet Gynecol. 1979;133: 825– 830.
23.
24.
25.
26.
2337
Andersen BL, Hacker NF. Psychosexual adjustment after vulvar surgery. Obstet Gynecol. 1983;62:457– 462. Ghurani GB, Penalver MA. An update on vulvar cancer. Am J Obstet Gynecol. 2001;185:294 –299. Byron SC, Lamb EJ, Yonemoto RH, Kase S. Radical inguinal node dissection in the treatment of cancer. Surg Gynecol Obstet. 1962;114:401– 408. Willis RA. The spread of tumours in the human body, 3rd ed. London: Butterworth, 1973. Hacker NF, Leuchter RS, Berek JS, Castaldo TW, Lagasse LD. Radical vulvectomy and bilateral inguinal lymphadenectomy through separate groin incisions. Obstet Gynecol. 1981; 58:574 –579. Schulz MJ, Penalver M. Recurrent vulvar carcinoma in the intervening tissue bridge in early invasive stage I disease treated by radical vulvectomy and bilateral groin dissection through separate incisions. Gynecol Oncol. 1989;35:383–386. Grimshaw RN, Murdoch JB, Monaghan JM. Radical vulvectomy and bilateral inguinal-femoral lymphadenectomy through separate incisions– experience with 100 cases. Int J Gynecol Cancer. 1993;3:18 –23. Christopherson W, Buchsbaum HJ, Voet R, Lifschitz S. Radical vulvectomy and bilateral groin lymphadenectomy utilizing separate groin incisions: report of a case with recurrence in the intervening skin bridge. Gynecol Oncol. 1985; 21:247–251. Rose PG. Skin bridge recurrences in vulvar cancer: frequency and management. Int J Gynecol Cancer. 1999;9:508 – 511. Hopkins MP, Reid GC, Morley GW. Radical vulvectomy. The decision for the incision. Cancer. 1993;72:799 – 803. Reviewers’ handbook. Cochrane Collaboration, version 4.1.4. Oxford: Cochrane, 2001. Ansink A, van der Velden J. Surgical interventions for early squamous cell carcinoma of the vulva (Cochrane review). Cochrane library, issue 1. Oxford: Update Software, 2002. Burger MP, Hollema H, Emanuels M, Krans M, Pras E, Bouma J. The importance of the groin node status for the survival of T1 and T2 vulval carcinoma patients. Gynecol Oncol. 1995;57:327–334. Burger MPM, Hollema H, Bouma J. The side of groin node metastases in unilateral vulvar carcinoma. Int J Gynecol Cancer. 1996;6:318 –322. Shepherd JH. Staging announcement. FIGO staging of gynecologic cancers: cervical and vulva. Int J Gynecol Cancer. 1995;5:319. Monaghan JM. Complications in the surgical management of gynaecological and obstetrical malignancy. 1st ed. London: Bailliere Tindall, 1989. Burke TW, Eifel PJ, McGuire WP, Wilkinson EJ. The vulva. In: Hoskins WJ, Perez CA, Young RC, editors. Principles of practice of gynecologic oncology. Philadelphia: Williams & Wilkins, 2000:795–796. Maggino T, Landoni F, Sartori E, et al. Patterns of recurrence in patients with squamous cell carcinoma of the vulva. Cancer. 2000;89:116 –122. Kabulski Z, Frankman O. Histologic malignancy grading in invasive squamous cell carcinoma of the vulva. Int J Obstet Gynecol. 1978;16:233–237. Farias-Eisner R, Cirisano FD, Grouse D, et al. Conservative and individualized surgery for early squamous carcinoma of the vulva: the treatment of choice for Stage I and II disease. Gynecol Oncol. 1994;53:55–58.
2338
CANCER December 1, 2002 / Volume 95 / Number 11
27. Helm CW, Hatch K, Austin JM, et al. A matched comparison of single and triple incision techniques for the surgical treatment of carcinoma of the vulva. Gynecol Oncol. 1992;46:150–156. 28. Siller BS, Alvarez RD, Conner WD, et al. T2/3 vulva cancer: a case-control study of triple incision versus en bloc radical vulvectomy and inguinal lymphadenectomy. Gynecol Oncol. 1995;57:335–339. 29. Lin JY, DuBeshter B, Angel C, Dvoretsky PM. Morbidity and recurrence with modifications of radical vulvectomy and groin dissection. Gynecol Oncol. 1992;47:80 – 86. 30. Gould N, Kamelle S, Tillmanns T, et al. Predictors of complications after inguinal lymphadenectomy. Gynecol Oncol. 2001;82;329 –332. 31. Zhang SH, Sood AK, Sorosky JI, Anderson B, Buller RE.
Preservation of the saphenous vein during inguinal lymphadenectomy decreases morbidity in patients with carcinoma of the vulva. Cancer. 2000;89:1520 –1525. 32. De Hullu JA, Hollema H, Piers DA, et al. Sentinel lymph node procedure is highly accurate in squamous cell carcinoma of the vulva. J Clin Oncol. 2000;18:2811–2816. 33. Heaps JM, Fu YS, Montz FJ, Hacker NF, Berek JS. Surgicalpathologic variables predictive of local recurrence in squamous cell carcinoma of the vulva. Gynecol Oncol. 1990;38: 309 –314. 34. Boonstra H, Oosterhuis JW, Oosterhuis AM, Fleuren GJ. Cervical tissue shrinkage by formaldehyde fixation, paraffin wax embedding, section cutting and mounting. Virchows Arch A Pathol Anat Histopathol. 1983;402:195–201.
