johnson2020
johnson2020
johnson2020
Int J Gynecol Cancer: first published as 10.1136/ijgc-2020-001516 on 7 July 2020. Downloaded from http://ijgc.bmj.com/ on July 26, 2020 at University of N S Wales 1247645. Protected by
Vulvar sarcoma outcomes by histologic
subtype: a Surveillance, Epidemiology, and
End Results (SEER) database review
Sarah Johnson,1 Malte Renz,1 Lindsay Wheeler,2 Elisabeth Diver,1 Oliver Dorigo,1 Babak Litkouhi,1
Kian Behbakht,2 Brooke Howitt,3 Amer Karam1
copyright.
University School of Medicine, Results (SEER) database was searched for women and clinical outcomes for vulvar cancer is focused
Stanford, USA; amkaram@ diagnosed with vulvar sarcoma between 1973 and on vulvar squamous cell carcinomas. Considering
stanford.edu 2018. We identified 315 patients and reviewed their primary vulvar sarcomas exclusively, there is a dearth
demographic, clinicopathologic, surgical, and survival of disease-specific evidence to guide management.
SJ and MR contributed equally. information. Statistical analyses included χ2 and t-tests, Treatment recommendations are loosely based on
Kaplan–Meier survival, and Cox regression analyses. the treatment of vulvar squamous cell carcinomas
Received 29 April 2020 Results The most common histopathologies of vulvar and sarcomas from other anatomic sites. The clin-
Revised 19 June 2020 sarcomas were dermatofibrosarcomas (85/315, 27%) and
ical presentation is similar to squamous carcinomas
Accepted 23 June 2020 leiomyosarcomas (72/315, 22.9%). Rhabdomyosarcomas
(18/315, 5.7%), liposarcomas (16/315, 5.1%), and of the vulva; however, patients tend to be younger at
malignant fibrous histiocytomas (16/315, 5.1%) were the time of diagnosis.3 Surgery remains the mainstay
less frequent. The majority of patients underwent surgery of treatment, but there is limited evidence to guide
(292/315, 92.7%), which included lymph node dissections the extent of surgery, the role of lymphadenectomy.5
in 21.6% (63/292). Survival and lymph node involvement Based on limited data, some authors recommend
varied significantly with histologic subtype. The 5-year adjuvant radiation particularly for high-grade lesions
disease-specific survival for dermatofibrosarcomas, or in the setting of positive or near positive surgical
liposarcomas, and fibrosarcomas was 100% and only margins, with limited data to guide management.2 3 6
60.3% and 62.5% for malignant fibrous histiocytomas Existing data indicate that, for vulvar sarcomas,
and rhabdomyosarcomas, respectively. None of the
lesion size, spread of disease, and mitotic activity
patients with (dermato)fibrosarcomas, liposarcomas, or
are the primary drivers of prognosis. It has been
leiomyosarcomas had positive lymph nodes, in contrast to
rhabdomyosarcomas and malignant fibrous histiocytomas suggested that lesions greater than 5 cm in diam-
with 77.8% and 40% positive lymph nodes, respectively. eter with infiltrating margins, extensive necrosis, and
© IGCS and ESGO 2020. No The 5-year disease-specific survival for women with >5 mitotic figures per 10 high power fields are more
commercial re-use. See rights
positive lymph nodes was 0%. likely to recur after surgical resection.2 4 5 However,
and permissions. Published by
BMJ. Conclusions Vulvar sarcomas are heterogeneous with the foundation for these data stem from urologic
survival highly dependent on the histopathologic subtype. sarcomas. Subsequent literature focusing on vulvar
To cite: Johnson S, Renz M, While surgical excision is the mainstay of treatment for
Wheeler L, et al. Int J Gynecol
and vaginal sarcomas found that only tumor grade
all vulvar sarcomas, staging lymphadenectomy should be seemed to predict outcome7 and inadequate surgical
Cancer Published Online First:
deferred for (dermato)fibrosarcomas, liposarcomas, and margins.5 In general, vulvar sarcomas are thought
[please include Day Month
leiomyosarcomas as there were no cases of lymph nodes
Year]. doi:10.1136/ijgc-2020- to be aggressive, but available data are sparse
metastases.
001516 and inconsistent.5 8–12 The majority of the literature
Int J Gynecol Cancer: first published as 10.1136/ijgc-2020-001516 on 7 July 2020. Downloaded from http://ijgc.bmj.com/ on July 26, 2020 at University of N S Wales 1247645. Protected by
involves case reports and histology- specific literature reviews Table 1 Vulvar sarcoma histologic subtypes
about specific vulvar sarcoma subtypes.9 10 13–17 These case reports
Histologic subtype n=315 %
and reviews of the literature hint at histology-based trends, such
as improved outcomes among liposarcomas and the potentially Dermatofibrosarcoma 85 27
more aggressive disease course among epithelioid sarcomas.8 10 In Leiomyosarcoma 72 22.9
2015, Chokoeva et al reviewed the clinical presentation, pathologic Undifferentiated sarcomas 30 9.5
diagnostic criteria, and prognosis for the most common histologic
Epithelioid sarcoma 25 7.9
subtypes.11 12
Given the relative paucity of data on vulvar sarcomas, we sought Other 24 7.6
to identify clinical and pathologic factors associated with improved Rhabdomyosarcoma 18 5.7
survival. We calculated survival outcomes of various subtypes of Malignant fibrous histiocytoma 16 5.1
vulvar sarcoma and examined factors that affect overall survival Liposarcoma 16 5.1
and may guide treatment.
Fibrosarcoma 14 4.4
Spindle cell sarcoma 10 3.2
Ewing sarcoma 5 1.6
METHODS
The Surveillance, Epidemiology, and End Results (SEER) data-
base of the National Cancer Institute was searched from 1973 to 4.4%) were less frequent, while spindle cell sarcomas (10/315,
2019 to identify women with vulvar cancer. A total of 315 cases 3.2%) and Ewing sarcomas 5/315 (1.6%) were rare (Table 1).
of vulvar sarcoma were identified. Demographic, clinicopathologic, Table 2 lists the demographic and epidemiologic features of
and survival information were reviewed. Unadjusted associations patients with vulvar sarcoma. The patients had a mean age of
of variables of interest were evaluated using Pearson’s χ2 test and 45.9±1.1 years at initial diagnosis and the majority were white
t-test for the comparison of means. The primary endpoint of this (72.4%), with black women representing a significant minority
study was 5-year disease-specific survival, with patients censored (19.7%). Vulvar sarcomas were predominantly lesions with a mean
either at death from disease or at date of last follow-up. Disease- size of 49.9 (±2.25) mm at presentation. Most patients had local
specific survival was calculated from the time of original diagnosis disease (232/315, 73.7%), while grade was unknown in over 50%.
and was assessed using the Kaplan–Meier survival analysis, and the Lymph nodes, surgically assessed in 21.6% of cases, were nega-
Mantel–Cox log-rank test was used to assess the statistical signif- tive in 63%. Chemotherapy and radiation were used in a minority of
copyright.
icance of the differences between survival curves. Both univariate cases (11.4% and 19%, respectively).
and multivariate Cox proportional hazards models were used to
calculate hazard ratios (HRs) and their 95% confidence intervals Vulvar sarcomas: outcome by epidemiologic and
(CIs) for the analysis of individual variables with respect to overall histopathologic features
survival. For multivariate models, forward stepwise procedures The 5-year disease-specific survival for primary sarcomas of the
were chosen in an a priori fashion, where variables were entered vulva overall was 77.8% (Table 3 and Figure 1). We performed
into the model with p<0.05 and removed if the significance of that univariate analyses to assess factors affecting outcome (Table 3).
variable subsequently exceeded p=0.10. The validity of the propor- Stage (Figure 1C), tumor size (Figure 1B), and lymph node metas-
tional hazards assumption was tested for each variable incorpo- tasis at the time of diagnosis affected outcome. For patients with
rated into the final Cox models using log–log survival curves. No localized vulvar sarcoma, the 5- year disease- specific survival
instances in which this assumption was not met were identified. was 93.6%. For patients with locoregional and distant spread,
SEER*STAT Version 8.1.5 (Surveillance Research Program, NCI, the survival rates were 66.5% and 0%, respectively (Table 3 and
Bethesda, Maryland, USA) was used to extract case level data from Figure 1C; p<0.001). The 5-year disease-specific survival was 92%
the SEER public-use databases. All analyses were conducted using for tumors measuring ≤5 cm and 75% for tumors measuring >5
the Statistical Package for the Social Sciences (SPSS, V24.0). cm (p<0.008). Lymph node involvement had the greatest impact on
survival. If lymph nodes were negative, the 5-year disease-specific
survival for women with vulvar sarcoma was 91%, but fell to 0%
with positive lymph nodes (p<0.001). There was a statistically
RESULTS
significant difference in 5-year disease-specific survival on univar-
Vulvar sarcomas: epidemiologic and histopathologic features iate analysis for low-grade (grades 1 and 2) versus high-grade
A total of 315 patients with vulvar sarcoma were analyzed in this (grades 3 and 4) sarcomas (p<0.001).
retrospective analysis of the SEER data from over 45 years (1973–
2018). We grouped vulvar sarcomas into 11 categories according Vulvar sarcomas: outcome by intervention
to their histologic features. A list of the groups is given in Table 1. Surgical excision was associated with an improved 5-year disease-
The most common histopathology of vulvar sarcoma was found specific survival (Table 3, Figure 1D). The 5-year disease-specific
to be dermatofibrosarcomas (85/315, 27%), followed by leiomyo- survival was 89% for vulvar sarcomas undergoing surgical excision
sarcomas (72/315, 22.9%). Sarcomas of uncertain differentiation and 63% for those patients who did not undergo surgery (p<0.001).
(30/315, 9.5%), epithelioid sarcomas (25/315, 7.9%), rhabdomyo- Lymphadenectomy was associated with a statistically significant
sarcomas (18/315, 5.7%), liposarcomas (16/315, 5.1%), malignant decrease in survival; 5-year disease-specific survival was 77% for
fibrous histiocytomas (16/315, 5.1%), and fibrosarcomas (14/315, patients who underwent lymphadenectomy versus 88% for patients
Int J Gynecol Cancer: first published as 10.1136/ijgc-2020-001516 on 7 July 2020. Downloaded from http://ijgc.bmj.com/ on July 26, 2020 at University of N S Wales 1247645. Protected by
Table 2 Demographics and clinical presentation of Table 3 Univariate analysis of 5-year disease-specific
patients with vulvar sarcoma survival (DSS) by demographic and clinical features, and
Characteristic Sarcoma histologic subtype of vulvar sarcomas
Five-year DSS,
Race Clinical feature %
White 228 (72.4%)
Race (p=0.2)
Black 62 (19.7%)
White 85
Other 19 (6%)
Black 91
Unknown 6 (1.9%)
Other 93.8
Mean age at diagnosis, years (±SE) 45.9±1.1
Stage (p<0.001)
Median age at diagnosis, years (range) 45 (0–98) I Local 93.6
Surgery II Regional 66.5
Yes 292 (92.7%) IV Distant 0
No 21 (6.7%) Size (p<0.008)
Unknown 2 (0.6%) <5 cm 92.2
Mean size, mm (±SE) 49.9±2.25 ≥5 cm 74.9
Median size, mm (range) 45 (5–150) Grade (p<0.001)
Grade 1 95.7
1 26 (8.3%) 2 100
2 49 (15.6%) 3 54.5
3 26 (8.3%) 4 73.5
4 32 (10.2%) Unknown 89.3
Unknown 182 (57.8%) Lymph nodes (p<0.001)
Stage Positive 0
copyright.
Localized 232 (73.7%) Negative 91.1
Regional 35 (11.1%) Surgery (p<0.001)
Distant 17 (5.4%) Yes 88.6
Unstaged 31 (9.9%) No 62.5
Lymphadenectomy Lymphadenectomy (p=0.006)
Yes 63 (20%) Yes 77.1
No 252 (80%) No 88.4
Int J Gynecol Cancer: first published as 10.1136/ijgc-2020-001516 on 7 July 2020. Downloaded from http://ijgc.bmj.com/ on July 26, 2020 at University of N S Wales 1247645. Protected by
copyright.
Figure 1 Kaplan–Meier overall survival curves for vulvar sarcomas. (A) Overall survival based on race. (B) Overall survival
by size of tumor. (C) Overall survival by stage. (D) Overall survival based on surgery for the primary tumor. (E) Overall survival
based on performance of radiation. (F) Overall survival by sarcoma subtype.
Outcome of vulvar sarcomas by histologic subtypes nodes by histology and the number of patients per histology who
The 5-year disease-specific survival for primary sarcomas of the underwent lymphadenectomy are shown in Online supplementary
vulva was 77.8% (Table 3). When each vulvar sarcoma subtype table 1. The 5-year disease-specific survival for women with posi-
was considered separately, women with fibrosarcoma or liposar- tive lymph nodes was 0% (Table 3).
coma had a 100% 5-year disease-specific survival and patients
with dermatofibrosarcoma had a 5-year disease-specific survival Multivariate analysis
of 98.7%. In contrast, women with malignant fibrous histiocytomas On multivariate analysis (Table 4), age (HR 1.04 for each additional
and rhabdomyosarcoma had survival rates of 60.3% and 62.5%, year of age, p<0.001) and stage (HR 3.99 and 16.4 for regional and
respectively (Table 3 and Figure 1F). distant spread compared with local disease, p=0.001 and p<0.001,
Only 20% of patients underwent lymph node dissections (Table 1). respectively) were significantly associated with an increased risk of
None of the patients with dermatofibrosarcoma, leiomyosarcoma, death. Surgical excision was associated with a significant survival
or liposarcoma who underwent lymphadenectomy had positive advantage (HR 0.39, p=0.05); conversely, the use of radiation
lymph nodes. Rhabdomyosarcomas and malignant fibrous histiocy- therapy was not associated with a survival advantage (HR 1.55,
tomas were more likely to show positive lymph nodes (77.8% and 95% CI 0.73 to 3.31, p=0.26). Numbers were too low to assess for
40%, respectively, p<0.001 for each). The rate of positive lymph the effect of grade and lymphadenectomy on multivariate analysis.
Int J Gynecol Cancer: first published as 10.1136/ijgc-2020-001516 on 7 July 2020. Downloaded from http://ijgc.bmj.com/ on July 26, 2020 at University of N S Wales 1247645. Protected by
Table 4 Multivariate analysis of vulvar sarcoma (310/315
Chokoeva and colleagues have provided a thoughtful histology-
patients) specific overview, review of management, and guidelines for vulvar
sarcoma,11 12 the current study is one of the largest to examine
HR for
side-by-side survival data among the various histologies.
death from
Clinical feature disease P value 95% CI When 5-year disease-specific survival was directly compared
among sarcoma subtypes, it was evident that the prognosis for
Age at diagnosis 1.04 <0.001 1.022 to 1.056 liposarcoma, fibrosarcoma, and dermatofibrosarcomas stands in
in years
stark contrast to rhabdomyosarcoma and malignant fibrous histio-
Race cytomas with 100% or close to 100% survival at 5 years compared
White 1.0 with 77.8% and 40% survival at 5 years, respectively. Furthermore,
Black 0.57 0.22 0.23 to 1.4 as opposed to malignant fibrous histiocytomas and rhabdomyosar-
Other 0.21 0.13 0.03 to 1.59 comas, lymph node metastases were not identified in liposarcomas,
fibrosarcomas, dermatofibrosarcomas, and leiomyosarcomas.
Stage
These data indicate that the diagnosis of a sarcoma of the vulva
Local 1.0 is nuanced and that the histologic subtype should be taken into
Regional 4.95 <0.001 2.14 to 11.44 consideration given the effect of histopathology on sarcoma spread
Distant 20.56 <0.001 8.07 to 52.39 and prognosis.
Surgery The role of lymphadenectomy necessitates further investiga-
tion considering the rarity of lymph node metastases for certain
No 1.0
subtypes of sarcoma, however this dataset suggests that regional
Yes 0.39 0.05 0.68 to 3.22 lymphadenectomy may be omitted safely for fibrosarcomas, lipo-
Radiation therapy sarcomas, and leiomyosarcomas. It is unclear if adjuvant therapy,
Yes 1.0 in particular radiation therapy, has a therapeutic role except for
No 1.02 0.69 0.92 to 1.14 those patients who are not candidates for surgery. These data are
likely biased in this study as it is likely that women with tumors
Sarcoma group 0.018
with a worse prognosis were administered adjuvant therapy more
commonly. The role of adjuvant chemotherapy or radiation could
not be evaluated in this study and treatment decisions need to be
copyright.
However, consistent with the univariate analysis also in the multi- individualized and data taken into account from sarcoma treat-
variate analysis, the histologic subtype was significantly associated ments at other anatomic sites.
with risk of death, with spindle cell sarcomas, epitheloid sarcomas, The results of this study are clinically relevant as they infer that
malignant fibrous histiocytomas, leiomyosarcomas, rhabdomyo- cases of liposarcoma, dermatofibrosarcoma, and fibrosarcoma
sarcomas and Ewing sarcomas faring worse than fibrosarcomas, may be adequately treated by surgery alone. Similarly, the data
dermatofibrosarcomas and liposarcomas (p=0.02). presented here suggest that histopathology may help guide the use
of lymphadenectomy in cases of vulvar sarcoma.
The primary limitation of this study is the limited number of
DISCUSSION sarcoma cases due to rarity of the disease. Unfortunately, with
This SEER review highlights the fact that vulvar sarcomas are a rare diseases such as vulvar sarcoma, the options for investiga-
varied group of malignancies with survival highly dependent tion are limited. However, this review supplements the existing
on the histopathologic subtype. Utilization of the SEER database histology-specific case reports and literature reviews with the goal
enables comparison of demographics and patient outcomes, with of providing further context for evaluation and management. Addi-
an average long-term follow-up of 10 years. This SEER review is tionally, the SEER database, while large, is incomplete and retro-
consistent with existing data suggesting that, for sarcomas of the spective and specifically does not capture all the details regarding
vulva, age at the time of diagnosis, tumor size, stage and grade adjuvant therapy, in particular chemotherapy. Understanding adju-
of disease have a statistically significant impact on survival. This vant treatments would have allowed for a more comprehensive
study highlights the importance of surgical excision in the manage- review. Finally, the SEER database lacks a central pathology review,
ment of these tumors, as surgical excision remains one of the most so the risk of misclassification of histology cannot be excluded,
important predictors of long-term survival. Our data suggest that which may be of relevance in these rare and difficult to classify
there was no race-based difference in outcomes in sarcomas of cancers.
the vulva.
Literature comparing outcomes among histologic subtypes of
vulvar sarcoma is limited. Curtin and colleagues found that when
looking at both vulvar and vaginal sarcomas (including nine cases CONCLUSION
of vulvar sarcoma), there was no statistically significant difference This SEER database review of 315 cases of primary vulvar sarcoma
between leiomyosarcomas and non-leiomyosarcomas.7 Nirenberg collected over 45 years represents a large series of an extremely
and colleagues reported in their small series and on review of rare disease. It shows that stage, size, and grade are prognostica-
the existing literature that aggressive angiomyxoma and derma- tors of outcome, but also that the specific histologic subtype of a
tofibrosarcoma were associated with a good prognosis.4 While vulvar sarcoma is predictive based on univariate and multivariate
Int J Gynecol Cancer: first published as 10.1136/ijgc-2020-001516 on 7 July 2020. Downloaded from http://ijgc.bmj.com/ on July 26, 2020 at University of N S Wales 1247645. Protected by
analysis. Furthermore, the histologic subtype may be predictive of 6 Holloway CL, Russell AH, Muto M, et al. Synovial cell sarcoma
of the vulva: multimodality treatment incorporating preoperative
lymph node involvement. While surgical excision of the primary external-beam radiation, hemivulvectomy, flap reconstruction,
lesion is the mainstay of treatment, lymph node dissection appears interstitial brachytherapy, and chemotherapy. Gynecol Oncol
to be indicated only in select histologic subtypes. Ideally, these 2007;104:253–6.
7 Curtin JP, Saigo P, Slucher B, et al. Soft-tissue sarcoma of the
findings will be corroborated in future studies, which will remain vagina and vulva: a clinicopathologic study. Obstet Gynecol
challenging given the rarity of primary vulva sarcomas. 1995;86:269–72.
8 Nucci MR, Fletcher CD. Liposarcoma (atypical lipomatous tumors)
of the vulva: a clinicopathologic study of six cases. Int J Gynecol
Contributors All listed authors fulfill authorship criteria. AK contributed to the Pathol 1998;17:17–23.
conceptualization, data collection and writing of the manuscript, SJ, MR and LW 9 Chiyoda T, Ishikawa M, Nakamura M, et al. Successfully treated
contributed to the data collection and writing of the manuscript, OD, ED, BL, BH and case of epithelioid sarcoma of the vulva. J Obstet Gynaecol Res
KB contributed to the writing of the manuscript. 2011;37:1856–9.
10 Moore RG, Steinhoff MM, Granai CO, et al. Vulvar epithelioid
Funding The authors have not declared a specific grant for this research from any sarcoma in pregnancy. Gynecol Oncol 2002;85:218–22.
funding agency in the public, commercial or not-for-profit sectors. 11 Chokoeva AA, Tchernev G, Cardoso JC, et al. Vulvar sarcomas: short
Competing interests None declared. guideline for histopathological recognition and clinical management.
Part 2. Int J Immunopathol Pharmacol 2015;28:178–86.
Patient consent for publication Not required. 12 Chokoeva AA, Tchernev G, Cardoso JC, et al. Vulvar sarcomas: short
guideline for histopathological recognition and clinical management.
Provenance and peer review Not commissioned; externally peer reviewed. Part 1. Int J Immunopathol Pharmacol 2015;28:168–77.
Data availability statement Data are available upon reasonable request. 13 González-Bugatto F, Añón-Requena MJ, López-Guerrero MA, et al.
Vulvar leiomyosarcoma in Bartholin's gland area: a case report and
literature review. Arch Gynecol Obstet 2009;279:171–4.
14 McCluggage WG, Sumathi VP, Nucci MR, et al. Ewing family of
REFERENCES tumours involving the vulva and vagina: report of a series of four
1 Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA cases. J Clin Pathol 2007;60:674–80.
Cancer J Clin 2012;62:10–29. 15 Sumathi VP, Fisher C, Williams A, et al. Synovial sarcoma of the
2 Magné N, Pacaut C, Auberdiac P, et al. Sarcoma of vulva, vagina vulva and vagina: a clinicopathologic and molecular genetic study of
and ovary. Best Pract Res Clin Obstet Gynaecol 2011;25:797–801. 4 cases. Int J Gynecol Pathol 2011;30:84–91.
3 Ulutin HC, Zellars RC, Frassica D. Soft tissue sarcoma of the vulva: 16 McKenzie M, Pintilie H, Wilkinson N, et al. A rare case of vulval
a clinical study. Int J Gynecol Cancer 2003;13:528–31. leiomyosarcoma: management and an updated review of the
4 Nirenberg A, Ostör AG, Slavin J, et al. Primary vulvar sarcomas. Int J literature. J Obstet Gynaecol 2011;31:675–6.
Gynecol Pathol 1995;14:55–62. 17 Boldorini R, Riboni F, Cristina S, et al. Primary vulvar Ewing's
5 Aartsen EJ, Albus-Lutter CE. Vulvar sarcoma: clinical implications. sarcoma/primitive neuroectodermal tumor in a post-menopausal
Eur J Obstet Gynecol Reprod Biol 1994;56:181–9. woman: a case report. Pathol Res Pract 2010;206:476–9.
copyright.