Carcinoma of the vulva in young women

Carcinoma

of the Vulva

MARK J. MESSING, MD, AND DONALD

in Young Women G. GALLUP, MD

Objective: To determine if young women with carcinoma of the vulva have a different risk factor history and outcome compared with older women. Methods: We conducted a retrospective review of the medical records of 78 women treated at the Medical College of Georgia for squamous carcinoma of the vulva during 1979-1993. Women younger than 45 years were compared with those 45 and over for historic risk factors, treatment modality, and outcome. Results: Over the study interval, the average presenting age of these patients decreased from 69 to 55 years. Women under 45 were found to have a stronger history of condyloma (P < .OOl, 95% confidence interval [CI] 3.69-87.96). There was no significant difference by age in the duration of symptoms before presentation, smoking history, or tumor size. Women 45 and over were more likely to have advancedstage disease (International Federation of Gynecology and Obstetrics [FIG01 stage III or IV) (P = .03,95X CI 0.43-0.91). Treatment did not differ significantly with age. In a univa-

riate analysis, advanced FIG0 stage, presence of metastases, and tumor size were associated with shorter survival. There was no detected difference in survival for women in either age group. Conclusion: There appears to be a trend in our patient population toward younger women presenting with squamous carcinoma of the vulva. Human papillomavirus infection appears to be more common in younger women with vulvar carcinoma. There may be a difference in the etiologies producing squamous carcinomas of the vulva. Education encouraging the early detection and prevention of sexually

transmitted disease

diseases might alter the rising incidence

in younger

women.

(Obstet

Gynecol

of this

1995;86:51-4)

Carcinoma of the vulva is generally regarded as a disease of older women. The mean age of women with squamous carcinoma of the vulva is between 65-80 years, with the incidence peaking by age 80.’ In contrast, carcinoma in situ of the vulva has its highest incidence rate by age 60. In one epidemiologic study, only 75% of vulvar cancers were detected before age 50.’

Another

extended

series

only 3.3% of cases younger

VOL.

86,

NO.

1, JULY

covering

than 35.>

30 years

revealed

We have been concerned by the young ages of women in our patient population who have developed carcinoma of the vulva. These young women do not appear to be clinically immunocompromised, although many do give a history of condyloma or previous surgeries for vulvar problems. Vulvar carcinoma has been proposed to exist as two distinct groups.“,” The women of one group are younger and have risk factors such as human papillomavirus (HPV) infection, cervical dysplasia, and a basaloid or poorly differentiated tumor. In the other group, the women are older, lack infectious risk factors, and have well-differentiated and keratinizing tumors. We undertook a retrospective review of our last 15 years of treating carcinoma of the vulva to determine if young women with this condition have a different risk factor history and outcome compared with older women. Particular attention was paid to any history or clinical evidence of HPV infection.

Materials and Methods All

cases

of carcinoma

of the vulva

were

identified

from

the tumor registry at the Medical College of Georgia. The medical record was reviewed, and follow-up information was obtained from the referring physician, when available. There were 98 cases of vulvar carcinoma identified for the period 1979-1993. Eighteen cases of nonsquamous histology were excluded, as were two squamous cases with insufficient information. Pathology was reviewed on all patients at the time of initial diagnosis or referral, and subsequently at the multidisciplinary Gynecologic Oncology Tumor Board. The data were analyzed by comparing patients under age 45 with those age 45 and older. This study has a power of 80% to detect a difference in proportions of 0.30.

Analysis

of the

data

was

by

2

or Fisher

exact

test

for categoric variables. Two-tailed probability results are given with odds ratios and confidence intervals (0. Survival

analysis

was

by

the

Kaplan-Meier

test

and

comparison of survival curves by the log-rank test. Cox proportional hazards survival model was used to test for a relationship between survival and presumed risk

1995

SSDI

0029.7844/‘35/$9.50 (~02Y-7H44(Y5)(10lOl-V

51

Table

1. Comparison

of Prognostic

Risk

Factors

Risk factor

(45 y (n = 18)

a45 y (II = 60)

P

Condyloma Smoking Alcohol Metastases Tumor >2 cm Symptoms >6 mo FIG0 stage >I1

10 10 5 2 15 h 3

4 lb 5 20 40 20 29

c.001 .04 .05 NS NS NS .03

OR = odds ratio; FIG0 = International

CI = confidence interval; Federation of Gynecology

OR 17 3 4

0.21

by Age

CI 3.69-87.96 0.99-I 1.57 0.86-20.21

0.43-0.91

NS = not significant; and Obstetrics.

factors. Statistical analysis was performed using a computer software package, True Epistat (Epistat Services, Richardson, TX).

Results Seventy-eight cases of primary squamous carcinoma of the vulva were available for analysis from January 1979 through December 1993. The mean age was 60.7 years (median 63, range 29-91). Two age peaks were noted at ages 50 and 70. Eighteen (23%) of the cases were in women younger than 45 years of age. There was a trend toward women presenting at a younger age. The average age was 69 in 197991984,67 in 1984-1989, and 55 in 198991993. Duration of symptoms was available for 61 patients. The median duration of symptoms before seeking medical care was 6 months. Patients presented with complaints of a lesion, lump, or pain in 70% of cases. There was no significant difference in the duration of symptoms for younger versus older women. Table 1 describes the comparison of risk factors for younger and older women. A history of condyloma was given in 18% of cases. Thirty-four percent of women were smokers and 13% had some history of alcohol intake. A history of condyloma was significant (P < .OOl) for women younger than 45. Smoking history and history of alcohol consumption, although having significant P values, both had CIs that included 1. An analysis of stage at presentation revealed International Federation of Gynecology and Obstetrics (FIGO) stage I, 32%; stage II, 27%; stage III, 19%; and stage IV, 22%. Table 1 demonstrates that older women were more likely to present with advanced local disease (stage III or IV), but not metastatic disease. The median tumor size at presentation was 4 cm (range O-27). Primary lesion size was not significantly different between age groups. However, lesion size over 2 cm was significantly associated with the presence of metastatic disease (P < ,001). A univariate analysis of prognostic factors for sur-

52

Messing

and

Gallup

Carcirmma

of the V~rlzn

viva1 was performed. The following were associated with decreased survival: FIG0 stage IV (P < .OOl, 95% CI 1.6-5.1), presence of metastases (P < .OOl, 95% CI 1.5-3.6), and tumor size greater than 2 cm (P = .002, 95% CI 0.09-0.34). Treatment consisted of surgery alone in 57 cases, a combination of radiation with or without adjuvant chemotherapy or surgery in 19, and no treatment in two. A partial radical vulvectomy was performed in 31 cases and a complete radical vulvectomy in 22. Fifteen nonradical excisions were performed, for a total of 68 primary vulvar excisions. Forty-four patients also underwent inguinal lymph node dissection, with 16 having only superficial nodes sampled and 28 having superficial and deep nodes evaluated. The modality of primary treatment appeared to be more affected by a changing philosophy of care (toward less radical vulvar resection) over time, rather than by the patient’s age. Twenty-two patients age 45 or over had disease recurrence. Sites of recurrence included the vulva in seven; groin, seven; vagina, four; lung, four; brain, one; and pelvis, two. One patient younger than age 45 was never disease-free and one had a vulvar recurrence. Disease status at this time includes 45% disease-free and 26% dead of disease. Eighteen percent are alive with disease and 12% are dead of other causes. The median survival follow-up interval was 16 months (maximum 87) for those younger than age 45 and 31 months (maximum 162) for those 45 and older. Median survival is 47 months for stage I, 51 months for stage II, 41 months for stage III, and 12 months for stage IV. Decreased survival was significantly associated with advanced stage (P < .OOl) (Figure 1). A significant difference in survival by age could not be demonstrated.

Discussion Squamous carcinoma comprises most vulvar carcinomas. The typical patient is 65-80 years old. Reports of women younger than age 35 with vulvar cancer are unusual.2,“-Y Although some of these patients have underlying diseases that would contribute to immunosuppression, this is not uniformly present. Roman et al5 reported on three women age 25 or younger with stage II disease. None had documented immunologic suppression. Wilkinson et al7 reported on two women, both age 22, with Fanconi anemia and vulvar carcinomas in association with HPV infection. Carter et al9 listed a variety of medical immunosuppressive illnesses in a group of women younger than age 40 with vulvar carcinoma. Over the last 15 years, there appears to be a trend in our patient population toward younger women presenting with squamous carcinoma of the vulva. Al-

Obstetrics

b Gynecology

S

Figure 1. Survival analysis comparing International Federation of Gynecology and Obstetrics stage. P < .OOl. Circles = stage I cancer; spmes = stage II cancer; frian~les = stage III cancer; diamonds = stage IV cancer.

r

”

i v

a I I

46

though we are a referral center, we have no reason to suspect that the referral pattern has changed, because most obstetrician-gynecologists do not treat vulvar cancer. In contrast with our data, other reportslO,” have not noted any change in the age of women with vulvar cancer. Sturgeon et al” did not find any changes in the incident age of vulvar cancer in the period 1973-1987, with the exception of a slight decline in women older than 55. Hording et al” also found no difference in age trends over the two lo-year periods, from 1971-1990. In contrast with invasive vulvar cancer, preinvasive lesions have been shown to be on the increase. Women under age 35 have a nearly doubled incidence of vulvar intraepithelial neoplasia III from 1.1 to 2.1 per 100,OOO.‘” A history of smoking was not significantly different between age groups in our study. A power calculation for this study reveals that there is a 20% chance of missing a difference as large as 30%). Smoking and a history of genital warts have been shown to increase the relative risk (RR) of vulvar neoplasia compared with controls.‘* Risk appears higher for current smokers and, together, these produce an RR 35 times that of controls.12 Neither tumor size nor time from onset of symptoms differed by age in our study. This is surprising because younger patients could be expected to have sexual partners or be more comfortable with selfexamination. Older women in our series more often had advanced-stage disease, suggesting that age may play a role in the disease course.

VOL.

86, NO.

1, JULY

1995

68

88

lee

128

148

168

166

We were able to identify a history or current presence of condyloma significantly more frequently in younger women with squamous carcinoma of the vulva. Because this was a retrospective review, we cannot exclude a bias in reporting for older patients and those from the earlier years of the study period. Evidence for infection with HPV has been identified in approximately 17-50% of vulvar carcinomas.‘“-lh The presence of HPV infection has been reported to occur in association with intraepithelial-like growth patterns, dysplasia, and the presence of multifocal disease.13 Human papillomavirus-positive tumors have been found to occur more often in younger patients than HPV-negative tumors. ‘1,14m17 However, some reports show no statistically significant age difference in HPV or non-HPV cases.13’18 Younger women and those who smoke have an increased incidence of HPV infection.15*16 Human papillomavirus 16 appears to be the major subtype identified in higher grade dysplastic vulvar lesions as well as in invasive tumors.1’~15~17~1y Age alone has not clearly been shown to worsen the prognosis of women with vulvar cancer. The shorter follow-up time and sample size for our younger patients may have limited our ability to determine if age was a significant predictor for survival. Rutledge et al” did not find age to be a prognostic variable for survival or recurrence, although there was a nonsignificant trend toward worse survival in older patients. Choo* also found no difference in prognosis by age. Significant

Messing

and Gallup

Cavcinonza

of the Vulva

53

predictors of survival and recurrence have included tumor size, clinical stage, the presence of positive regional lymph nodes, therapeutic intent, and the presence of positive margins.20 Although the study by Park et alI7 suggested that there was a higher rate of metastases in HPV-positive cancer, HPV status in other studies has not been shown to be a prognostic factor for recurrence or survival.‘5,‘6 Our analysis showed that survival was significantly affected by advanced stage, presence of metastases, and tumor size. The type of therapy did not appear to affect survival and was more a reflection of the increasing use of less radical vulvar surgery for unilateral lesions. The presence of carcinoma of the vulva in younger women may result from either the natural history of the disease being hastened or the development of a different disease with different risk factors and natural history. Young women with vulvar cancer may constitute a unique group sharing a common diagnosis but have a different disease than their older counterparts. The presence of HI’V infection may identify the etiologic agent involved. Despite these factors, the ultimate treatment and prognosis appears to be unaffected by age. Early detection will lead to low-stage disease and result in better survival. Education of women in the prevention of sexually transmitted diseases, smoking cessation, and self-examination might alter the rising incidence of this disease in younger women and improve their survival.

8. Mabuchi vulva:

D, Tarone

R. An

epidemiologic

study

of cancer

of the

1993;51:307-10. 10. Sturgeon SR, Brinton invasive vulvar cancer Gynecol 1992;166:1482-5.

3. Crum CP. Carcinoma sis. Obstet Gynecol 4. Kruman RJ, Trimble

of the vulva: Epidemiology and 1992;79:448-54. CL, Shah KV. Human papillomavirus

pathogenesis of vulvar carcinoma. 1992;4:582-5. 5. Roman LD, Mitchell MF, Burke invasive squamous carcinoma of Gynecol 6. Leibowitch changes

Curr

Opin

TW, Silva the vulva

pathogeneand

Obstet

the

Gynecol

EG. Unsuspected in young women.

Oncol 1991;41:182-5. M, Neil1 S, Pelisse M, Moyal-Baracco associated with squamous cell carcinoma

M. The epithelial of the vulva:

review of the clinical, Br J Obstet Gynaecol 7. Wilkinson EJ, Morgan

histolgical and viral findings 1990;97:1135-9. LS, Friedrich EG. Association

in 78 women.

anemia and squamous tract with condyloma Med 1984;29:447-53.

cell carcinoma of the lower acuminata: A report of two

female genital cases. J Reprod

54

Messing

and

Gallup

of Fanconi’s

Carcinoma of the Vulva

A

Kessler II. Epidemilogy study. Cancer 1985;55:1843-8.

of cancer

of the

tumors in young Gynecol Oncol

LA, Devesa SS, Kurman RJ. In situ and incidence trends (1973 to 1987). Am J Obstet

11. Hording U, Junge J, Daugaard S, Lundvall F, Paulsen H, Bock JE. Vulvar squamous cell carcinoma and papillomaviruses: Indications for two different etiologies. Gynecol Oncol 1994;52:241-6. 72. Brinton LA, Nasca PC, Mallin K, Baptiste MS, Wilbanks G, Richart R. Case-control 1990;75:859%66. 13. Follen-Mitchell MC, Crum Cl’.

study

of cancer

of

the

vulva.

Obstet

M, Prasad CJ, Silva EG, Rutledge Second genital primary squamous

Gynecol

FN, McArthur neoplasms

in

vulvar carcinoma: Viral and histopathologic correlates. Obstet Gynecol 1993;81:13-8. T, Kurman RJ, Park JS, Kessis T, Daniel RW, Shah KV. 14. Toki Probable nonpapillomavirus etiology of squamous cell carcinoma of the vulva in older women: A clinicopathologic study using in situ hybridization and polymerase chain reaction. Int J Gynecol Path01 1991;10:107-25. 15. Bloss JD, Liao S, Wilczynski features of vulvar carcinomas Evidence that squamous cell

SP, et al. Clinical and histologic analyzed for papillomavirus status: carcinoma of the vulva has more than

one etiology. Hum Path01 1991;22:711-8. 16. Andersen WA, Franquemont DW, Williams J, Taylor Vulvar squamous cell carcinoma and papillomaviruses: arate entities? Am J Obstet Gynecol 1991;165:320-36.

PT, Crum Two

Cl’. sep-

17. Park JS, Jones RW, McLean MR, et al. Possible etiologic heterogeneity of vulvar intraepithelial neoplasia. Cancer 1991;67:1599-607. Gj, Delvenne I’, MacConnell I’, Chalas E, Neto C, Mann WJ. 18. Nuovo Correlation of histology DNA in vulvar cancers. 19. Buscema J, Naghashfar

20.

and detection of human papillomavirus Gynecol Oncol 1991;43:275-80. Z, Sawada E, Dame1 R, Woodruff JD, Shah

K. The predominance of human papillomavirus neoplasia. Obstet Gynecol 1988;71:601-6. Rutledge F’N, Follen-Mitchell M, Munsell indicators for 1991;42:239-44.

cervix, vagina, and vulva based on the Third National Cancer Survey in the United States. Am J Obstet Gynecol 1979;129:525-32. 2. Choo YC. Invasive squamous carcinoma of the vulva in young patients. Gynecol Oncol 1982;13:158-64.

DS,

9. Carter J, Carlson J, Fowler J, et al. Invasive vulvar women: A disease of the immunosuppressed?

References 1. Henson

K, Brass A case-control

invasive

carcinoma

of the

type MF, vulva.

16 in vulvar

et al. Prognostic Gynecol

Oncol

Address reprint requests to: Mark 1. Messing, MD Section of Gynecologic Oncology Departtnetlt of Obstetrics and Gyr~ecology The Medical College of Georgia, CK-166 Augustn, GA 30912

Received October 24, 1994. Received in revised form March 6, 1995 Accepted March 23, 1995. Copyright Gynecologists.

0

1995

by

The

American

College

of Obstetricians

and

Obstetrics b Gynecology