The purpose of this paper is to bring to the attention of the gynecologic community eight cases of recently described, unilateral soft tissue masses of the labium majus in prepubertal girls. Despite the rapid growth, lack of palpable borders, and the impossibility of a complete removal, the histologic pattern and clinical behavior are those of a benign fibrous lesion.
MATERIALS AND METHODS
The cases were diagnosed between 1990 and 2006 in girls who had come to medical attention for unilateral, asymptomatic swelling of the labium majus.
The first four of this series were operated upon by A.A. and a pediatric surgeon using a direct vertical incision over the labium majus swelling and excising an overlying wedge of slightly indurated, tan, peau d’orange skin. Because of concern about the nature of the mass, at the same time an inguinal exploration was done, as well as vaginoscopy, proctoscopy, and microsurgical biopsies of the adjacent cervix and vagina. The additional four cases were done by different pediatric surgeons and a pediatric urologist.
Perineal and pelvic ultrasonography and magnetic resonance image (MRI) studies were performed on our four patients and were reviewed by a pediatric radiologist (K.N.). A fifth patient was evaluated by pelvic sonogram. Clinical data and follow-up information were obtained from clinical charts and the records of A.A. and the other surgeons.
The surgical specimens were sampled generously and/or entirely. Tissue was fixed in 10% formalin, embedded in paraffin, sectioned at a thickness of 5 microns, and stained with hematoxylin and eosin. Glass slides were reviewed, and paraffin blocks were obtained for the cases not resected at Mount Sinai Hospital. Trichrome and elastic (von Gieson) stains were performed on all cases. Immunohistochemistry was performed on formalin-fixed, paraffin-embedded tissues of all cases, with appropriate controls. Staining was performed on an automated immunostainer NexES (Ventana Medical Systems, Tucson, AZ). The following antibodies were used: vimentin (Ventana), CD34 (Ventana), muscle specific actin (MSA, Ventana), smooth muscle actin (SMA; BioGenex, San Ramon, CA), S100 (Ventana), ER/PR (Ventana), myoglobin (Ventana), desmin (Ventana), CD31 (Dako, Goldstrup, Denmark), CD68 (Dako), HAM56 (Ventana), synaptophysin (Ventana), and c-kit (CD117; Oncogene, Cambridge, MA). In five cases, tissue was immediately fixed in 3% glutaraldehyde with a 0.2 mol/L sodium cacodylate buffer at pH 7.4. The tissue was then processed for embedding in EMbed 812 (EMS, Hatfield, PA) via graded steps of ethanol and propylene oxide. One-micrometer plastic sections were cut, stained with methylene blue and azure II, and observed by light microscopy. Representative areas were chosen for ultrathin sectioning. The ultrathin sections were stained with uranyl acetate and lead citrate. The sections were viewed with a Hitachi H-7000 transmission electron microscope (Hitachi, Tokyo, Japan). In three cases, fresh tissue was immediately placed in Roswell Park Memorial Institute (Buffalo, NY) solution (RPMI) and transported to the Tumor Cytogenetics Laboratory for tissue culture and karyotype analysis.
Control tissue from the labium majus was obtained from two autopsy controls, ages 7 months and 3 years, and evaluated by the full profile of histochemical and immunohistochemical stains. The gynecologic and pathologic literature for vulvar lesions was reviewed for cases that appeared to be related clinically and pathologically to our cases. The study was approved by the Institutional Review Board of Mount Sinai School of Medicine.
The patients ranged in age from 5 to 7 years (Table 1). In all cases the swelling developed rapidly within 3 months, and then the growth subsided. After reaching a certain size, ranging from 3 to 6 cm, there seemed to be no further growth, but the surgeon was not consulted in several cases up to 3½ years (Fig. 1, A and B). In two cases the lesion was on the left, while in six cases it was on the right (Table 1). The overlying skin showed a slight peau d’orange appearance with a tan color. There was nothing unusual in the pregnancy history, no history of trauma, and no family history. There was consanguinity in one case, with one set of great grandparents. Seven of the eight patients were Ashkenazi Jewish from a closed observant community living in the same borough in New York City, and one of them had lived there previously. One was Chinese who lived nearby.
None of the children had pubic hair or breast development. Two cases had short (up to 5 mm) hairs on the skin overlying the masses. It was difficult to see because of the tan colored peau d’orange skin with slight induration. In case 8 the hair was found by careful examination of the pediatric endocrine consultant, but not noticed by the surgeon. In case 6 it was discovered by colposcopy during surgery. All the cases had dark scalp hair. In two cases preoperative serum analysis showed normal prepubertal hormone levels. Karyotype of the peripheral blood in four cases was normal 46,XX. Preoperative diagnoses included lymphangioma, lipoma, and hemangioma.
Ultrasound examination of our four Mount Sinai patients showed a poorly defined soft tissue mass of the labium majus, which did not have a high velocity flow by color Doppler. Magnetic resonance imaging (Fig. 2) showed a lesion in the labium infiltrating deep toward the adjacent levator pelvic floor, with the T1 weighted MRI signal characteristic of muscle and fascia. One patient had a negative MRI of the head and a pelvic sonogram demonstrating a normal prepubertal uterus and ovaries.
In all the personal cases, it was recognized that the unusual tissue extended into all the adjacent tissue and could not be removed surgically without resulting in mutilation. An inguinal incision did not permit adequate exposure of the mass. The operative notes of the different pediatric surgeons doing three of the additional cases indicated that the entire lesion was removed. In the eighth case, the operating pediatric urologic surgeon felt that the lesion could not be completely removed. At the time of surgery on our cases inguinal exploration, vaginoscopy, proctoscopy, and microsurgical biopsies of the adjacent cervix and vagina were negative.
With our first personal case, because of uncertain interpretation, there was overly cautious resection, and a small amount of the mass was left below the incision. A few months later it was thought to be a recurrence, but in fact it represented incomplete removal in a cosmetically sensitive area and did not grow further. In the three remaining personal cases, a cautious but more extensive resection resulted in “no recurrence.” Postoperative MRIs of four Mount Sinai patients showed extension into the adjacent pelvic floor and “possible residual lesions.” By clinical examination, however, subsequently the lesions appeared to stop growing.
Follow-up was done by the referring pediatrician. In follow-up over 4 months to 9 years, there has been no evidence of recurrence by physical examination performed by the referring pediatricians in our cases or by report in the other cases.
The surgical specimens consisted of skin and underlying unencapsulated, poorly circumscribed masses that ranged in greatest diameter from 3 to 6 cm. The excised mass was composed grossly of intertwining bands of shiny fibrous tissue surrounding small nests of fat.
Histologic examination revealed subepidermal hypocellular fibrous tissue extending deep into the subcutaneous fat tissue, surrounding mature adipose tissue, blood vessels, and nerves. The lesions were not circumscribed and merged with the surrounding tissues (Fig. 3A). The infiltrating tissue consisted of dense collagenous tissue, rich in elastic fibers and containing fibroblasts displaying regular, uniform oval nuclei, with no pleomorphism and no mitotic activity (Fig. 4). The composition of the resected tissue was similar to normal vulvar stromal tissue (Fig. 3B), except that the fibrous tissue component was increased in amount and displayed infiltrative growth deep into the subcutaneous fat. Immunochemical analysis confirmed the mesenchymal nature of the tissue. The fibroblastic cells were diffusely positive for vimentin (eight of eight), and CD34 (five of eight). There was focal positive staining for estrogen receptors in four of eight. Progesterone receptors and all other immunostains were negative. Control vulvar stromal tissue in two cases showed a similar immunohistochemical profile, except for a negative result for estrogen receptors. Electron microscopy confirmed the fibroblastic nature of cells embedded in the dense collagenous stroma.
In three of our personal cases, cytogenetic analysis of the lesions showed the normal female karyotype 46,XX. In two cases the addition of dehydroepiandrosterone sulfate (DHEAS) to the tissue culture speeded mitotic activity. The concentration was not determined.
A variety of benign and malignant lesions may present with swelling of the labium majus, such as Bartholin’s duct cyst, hemangioma, lymphangioma, leiomyoma, rhabdomyoma, neurofibroma, lipoma, myofibromatosis, angiofibroblastoma, and embryonal rhabdomyosarcoma.1,2 Within the prepubertal age group, however, such disorders are rare. In recent years, however, there have been several reports of a clinical-pathologic entity that is not generally recognized and is not described in textbooks of pathology, gynecology, pediatrics, and pediatric and adolescent gynecology, despite the fact that it may comprise as much as 20% of soft tissue masses of the vulva in children.3
The literature on this entity is scant. We have reviewed the literature of vulvar lesions to identify cases that appear to be similar to ours. An abstract published in 1999 reported three girls (aged 4–8 years), two with soft tissue masses in the labium majus and one in the groin (Boyd TK, Perez-Atayde A. Myxoid vulvar fibromatosis: a recurrent vulvar lesion of young girls. Pediatr Dev Pathol 1999;2:196). The excised tissue was composed of poorly circumscribed, infiltrative spindle cells, with no evidence of cytologic atypia or mitoses, with a variably myxoid background in the subcutis and overlying skin. The lesions were reported as recurring in two cases due to incomplete excision. It was named “myxoid vulvar fibromatosis.” In 2000, our group had an abstract published of our work with (at that time) seven patients, based on a presentation at The U.S. and Canadian Academy of Pathology in March 2000, which at the time we called a “Novel Fibrous Proliferation of the Vulva” (Magid MS, Walsh MM, Greco RE, Gordon RE, Yee H, Deligdisch L, Altchek A. Mod Pathol 2000;13:205A). In that same year, a case report appeared of an 8-year-old girl with a left labial mass, which was termed a “vulvar hamartoma.”2 Although the histology of this lesion was not illustrated, by description it seems consistent with our cases. There was no follow-up. In 2004, in a report of 11 prepubertal girls (aged 4–12 years), a “previously unrecognized” mesenchymal vulvar tumor was named “prepubertal vulva fibroma.” It was based on consultation histologic slide reviews. The hypocellular lesions were composed of bland spindle cells in a variably collagenous, edematous, or myxoid stroma infiltrating into normal vulvar soft tissues. Three patients developed locally recurrent tumor within months, and one had a second recurrence. The authors considered this lesion to be “a potentially recurrent neoplasm” rather than a hamartoma or stromal hyperplasia; nevertheless, it had a benign course “except for local recurrence when incompletely excised.”1 In 2005, there was a report of 14 girls (aged 3.9–13.2 years) who had surgery for “expansion” of the labium majus without definable borders. Histologic examination showed proliferating fibroblasts in a myxoid-collagenous matrix, which were encircling the usual vulvar soft tissue components. Recurrence was found in seven cases, but in one case there was regression without re-excision. The authors called the lesion “childhood asymmetric labium majus enlargement (CALME)” and thought it to be a distinct clinicopathologic entity of prepuberty and early puberty in response to hormonal surges.3
In all these reports, the pathologic descriptions and clinical behavior appear to be similar, although the names assigned to this lesion differ. The reports describe the same age group affected (prepubertal girls), and the same painless, usually unilateral, poorly circumscribed, and infiltrative vulvar mass. The masses histologically feature bland fibroblasts, with similar immunohistochemical profiles to each other and to normal vulvar tissues, deeply infiltrating around normal vulvar stromal tissues. There is no evidence of neural, muscular, or myofibroblastic differentiation. In our cases the masses developed within approximately 3 months and then stabilized. In two cases (case 7 and 8) this continued for 3 and 3 1/2 years before surgery confirming its natural history of cessation of growth.
We believe that the tumor is due to fibroblastic overgrowth, probably responsive to the physiologic surge of adrenal hormone. Abnormal hormone levels have not been described in these prepubertal patients. Vargas and colleagues3 suggested that the vulvar enlargement reflects asymmetric physiologic growth of otherwise normal tissue, analogous to gynecomastia or unilateral breast hypertrophy. Alternatively, it may mimic the sequela of pathologic exposure to hormones during development, such as estrogenic activation of vaginal neoplasia after fetal exposure to diethylstilbestrol. We and others found that the lesional fibroblasts are often estrogen receptor positive.3 Vargas and colleagues reported positive staining for estrogen receptors in a control patient,3 but our two controls failed to stain for estrogen receptors. Previous studies have documented the presence of estrogen receptors in only a small proportion of the dermal fibroblasts of the labia majora in adults.4 An appropriately controlled study of the expression of estrogen receptors in prepubertal vulvar tissue has not yet been reported. Thus, if this lesion is hormonally responsive, it is not clear whether it arises as an idiosyncratic sensitivity to hormonal stimulation in vulvar tissue that normally contains estrogen receptors or as the result of premature/abnormal receptor expression in prepubertal vulvar tissue. We noted with interest that the addition of DHEAS to cell cultures of the lesion enhanced cell proliferation. Dehydroepiandrosterone sulfate is a weakly androgenic steroid product of the adrenal gland, which is secreted in increasing amounts beginning in the middle of the first decade in girls (adrenarche), preceding the production of ovarian sex steroids that heralds puberty.5 Although the mitotic effect of DHEAS in our cell cultures was not a controlled observation, it promotes speculation about the role of hormones in the growth of this lesion. Previous studies have described the mitogenic effect of DHEA on estrogen receptor–positive breast cancer cells in culture.6 The role of prepubertal hormones, including the presence of estrogen receptors, in this lesion is an interesting focus for future investigation. We do not favor classifying this lesion as a neoplasm because it appears to be a self-limited hormonally driven process, despite its infiltrative growth. In our cases, the ethnic and geographic clustering suggests that there may be a selective vulnerability to hormonal effect. It is possible, however, that the fact that most of our patients were of Ashkenazi Jewish extraction from a restricted geographic area may represent a selection bias because this group comprises 10% of the first author’s practice, and New York City has a large Jewish population. Other reports have not indicated ethnicity or religion.
The surgeon must be aware of the possibility of this being a benign lesion despite rapid growth, lack of palpable border, and infiltration into the surrounding tissue. The overlying tan peau d’orange may be a clinical clue to the diagnosis (Poster presentation at the 54th Annual Clinical Meeting of the American College of Obstetricians and Gynecologists [ACOG] May 6–10, 2006, Washington, DC, Altchek A, Deligdisch L, Norton KI, Gordon R, Greco MA, Magid MS. Prepubertal unilateral soft fibrous tumor of labium majus without palpable border in girls. Obstet Gynecol 2006;107[4 suppl]:20S). Clinical and pathology consultation is advisable. Although previous authors1 have implied that the surgeon should do a complete removal to prevent recurrence, it is not possible and would be mutilating if attempted. Our experience and that of others3 have shown that, despite the initial rapid growth, the lesions stop enlarging and do not appear to recur after surgery, despite worrisome surgical and MRI observations of infiltrative growth. We believe that reports of “recurrence” probably represent incomplete removal and local edema. Metastasis has not been described. Surgery should be done to confirm the diagnosis and for cosmetic reasons, because the mass is very apparent and emotionally disturbing to the child. The surgeon has to make a judgment regarding the amount of tissue to be removed for cosmetic reasons versus the extent of the surgery. The pathologist should be alerted to this lesion to avoid overreaction to incomplete removal when assessing the margins of the lesion.
Because published reports used different terminologies to define lesions that seem to represent a similar pathologic process, we favor a unified term that would exclude the notion of this lesion being a neoplasm to avoid excessive, unnecessary, and possibly mutilating surgical procedures. The main histopathologic feature is that of a diffuse growth of benign fibroblastic tissue that traps mature adipose tissue, blood vessels, and nerves without displaying cellular changes of malignancy. The process halts after initial growth and does not appear to recur after surgery or to metastasize. Previously employed nomenclature does not adequately describe the pathologic and clinical nature of this lesion. The term “fibroma” designates a circumscribed benign neoplasm; not only does this term denote neoplasia, but it also does not reflect the diffuse infiltration of surrounding tissues seen in these cases. Fibromatosis is a term that encompasses a variety of lesions that are typically characterized by locally invasive growth by benign-appearing cells, similar to what we observe in our cases. Fibromatoses, however, generally are considered to be neoplasms, usually of myofibroblasts (as opposed to fibroblasts), and these lesions commonly exhibit aggressive behavior with repeated recurrences,7 not the natural history we are seeing in our cases. Hamartoma, representing a disorganized overgrowth of tissues native to a particular site, does not describe this predominantly fibroblastic process. Similarly, the term childhood asymmetric labium majus enlargement (CALME) does not indicate the proliferating cell of origin. We propose, as a term to accurately reflect cell of origin, pathologic process (with its implied self-limited growth), and clinical behavior—prepubertal labial fibrous hyperplasia. Further epidemiological investigation is indicated to determine if there are identifiable genetic or environmental predispositions.