FIBROADHNOMAS HAVE ALWAYS BEEN considered benign tumors of the breast. They are the most frequent benign tumor of the breast after fibrocystic disease. Histologically, they consist of both epithelial and stromal components.1 It is a well-accepted clinical practice for these lesions to be observed after histopathological confirmation.1- 2 Fibroadenomas have not been considered to significantly increase the risk of developing subsequent breast cancer. We present a case of malignant transformation of a fibroadenoma to cystosarcoma phyllodes after 5 years of radiologic stability.
CASE History
A 45-year-old female, G^sub 2^P^sub 0^, was referred to our breast center by her primary care physician in October 2003 for evaluation of a rapidly growing right breast mass at the site of prior fibroadenoma at the 2 o’clock location. She was found to have an 0.9-cm nodule in this area on her baseline mammogram in April 1997 that was occult on sonography (Fig. 1). Stereotactically guided fine-needle aspiration (FNA) and core biopsy of this nodule revealed a fibroadenoma (Fig. 2). The nodule remained radiographically stable from April 1997 to June 2002.
On physical examination, the patient was found to have a 3-cm firm mobile mass in the 2 o’clock position of the right breast without evidence of any lymphadenopathy. Mammogram in October 2003 showed that the palpable mass corresponded with a lobulated dense, partially circumscribed noncalcified 4 3-cm mass, which represented a major interval change (Fig. 3). A focused right breast ultrasound corresponded with a lobulated heterogenous solid mass at the 2 o’clock axis, 2 cm from the nipple and 1 cm deep to the skin, measuring 2.8 2.0 2.7 cm (Fig. 4). This corresponded to the site of prior 0.9-cm nodule.
A fine-needle aspiration in October 2003 was consistent with atypical cells, with mixed features of cellular biphasic tumor and cystic papillary changes. The patient underwent wide local excision. Pathology revealed malignant cystosarcoma phyllodes, 2.5 cm in maximum diameter, with 1-cm margins (Figures 5-7).
Discussion
Most women diagnosed with fibroadenomas are treated with observation. Fibroadenomas are typically diagnosed in young women with the average age of patients being 20-30 years old.1 Due to the fact that it is acceptable to observe fibroadenomas, it is crucial to identify risk factors for malignant transformation of these lesions.
Cystosarcoma phyllodes, similar to fibroadenoma, typically present as a firm, seldom painful, smooth, well-circumscribed and mobile mass.1,3 It demonstrates continuous growth, but it may also grow rapidly resulting in shiny stretched skin with visible underlying veins. As a result of this, the skin can ulcerate due to ischemia. Differentiation between malignant versus benign cystosarcoma phyllodes should not be based on the presenting symptoms such as pain, retraction of the nipple and skin fixation, as they can occur in both entities.3
It is quite challenging to differentiate fibroadenoma and cystosarcoma phyllodes by either mammography or sonography.1,3 A confirmed diagnosis occurs in only 32 per cent of mammograms.3 In approximately 10 per cent of cases, cystosarcoma phyllodes are seen mammographically in the absence of a palpable mass. Sonography may demonstrate small cystic areas within an otherwise benign appearing mass, which would require a biopsy.4
The average age of patients with cystosarcoma phyllodes is 40 years.1,4 Seventy per cent of these tumors occur between ages 30 and 55 years. However, they have been reported in girls as young as age 10 and into the ninth decade. The incidence of these tumors in women younger than age 25 may approach that of breast carcinoma. They occur rarely in men.4 Although a cystosarcoma phyllodes tumor contains epithelial and stromal components with similar histological characteristics to fibroadenoma, the stroma in phyllodes has increased cellularity, pleomorphism, nuclear atypia, and mitotic figures.1, 3-5 When compared to fibroadenomas, the natural history of cystosarcoma phyllodes is quite different. Fibroadenomas tend to grow slowly and often reach a maximum diameter of 2-3 cm in size. A significant percentage of fibroadenomas regress spontaneously, whereas cystosarcoma phyllodes tumors grow rapidly and can attain sizes of 10 cm or more.1
The diagnosis of cystosarcoma phyllodes tumor cytologically is challenging, with significant overlap with fibroadenoma. Large size, low epithelial/stromal ratio, epithelial atypia, columnar stromal cells with visible cytoplasm, and stromal giant cells favors a phyllodes tumor rather than a fibroadenoma.6 The lesion can be categorized from benign to borderline to malignant based on whether there are infiltrative borders, the degree of stromal overgrowth, the presence of cellular atypia, and the proportion of mitotic figures.4, 7, 8 According to Norris and Taylor, 0-4 mitoses/10 hpf with minimal atypia was classified as benign, 5-9 mitoses/10 hpf with moderate atypia as borderline, and >10 mitoses/10 hpf with marked atypia as malignant.3 Azzopardi and Salvador! used the same histological classification to categorize these lesions.9 Fine- needle aspiration by an experienced cytopathologist is a highly reliable tool in diagnosing cystosarcoma phyllodes tumors10, 11 with accuracy rate of 92.8 per cent.10 To avoid diagnostic error, multiple-site aspiration of the mass is recommended.3 However, only a “positive” biopsy, revealing the features of a malignant phyllodes tumor or the benign features of a fibroadenoma can be relied on. All other findings would require excisional biopsy.4
Cystosarcoma phyllodes account for 0.3-0.9 per cent of all breast tumors1 3-5, 7, 9 and only 2-3 per cent of all mammary fibroepithelial neoplasms.3, 7 The incidence rate is 1 in 100,000 with an overall risk of malignancy of 2.1 per 1 million women. There are no predisposing factors, and the etiology is unknown. One study reported Latin women at higher risk than other ethnic groups. There has been no documented relationship between menopausal status or use of oral contraceptives and development of these tumors. A strong correlation between development of these tumors and nulliparity was demonstrated in one study, in which all the patients with malignant cystosarcoma phyllodes tumor were nulliparous.3
The right breast is the more common site for this tumor. In one study, 62.5 per cent of tumors were in the right breast, with the remaining 37.5 per cent in the left breast. Bilaterality was as high as 30 per cent in one study. Of these tumors, 32-36 per cent occur in the upper outer quadrant, 14-17 per cent in the upper inner quadrant, 11-25 per cent in the lower outer quadrant, and 6-8 per cent in the lower inner quadrant. In the same study, 37 per cent of the tumors presented in multiple sites.3
FIG. 1. Initial mammogram demonstrating an 0.9-cm nodule, biopsy proven to be a fibroadenoma.
These tumors range from 1 to 41 cm, with median size of 5 cm. In one study, 73 per cent of benign phyllodes tumors were 7 cm.3 The term cystosarcoma phyllodes was first described and coined by Johannes Muller in 1838, which he clearly stated to be a benign tumor.3-5, 7 They are commonly benign, considering this fact, the World Health Organization coined the term “phyllodes tumor” in 1981.3 However, malignant types exist and can metastasize.3, 12 Lee and Pack reported the first case of metastatic phyllodes tumor in 1931.3,5
High-grade malignant phyllodes tumor (MPT) forms approximately 25- 35 per cent of all phyllodes tumors.4, 9 Approximately one-third to one-half of the malignant tumors will metastasize.4 However, the overall metastatic rate is
FIG. 2. Core biopsy of the breast lesion performed in 1997. Histologic features are those of a fibroadenoma. Note the hypocellular stroma (H&E, 20).
FIG. 3. Mammogram in 2003, demonstrating 4 3-cm mass corresponding to the palpable mass.
FIG. 4. Focused breast ultrasound in 2003, demonstrating a tabulated heterogenous solid mass, measuring 2.8 2.0 2.7 cm.
FIG. 5. Biopsy of the same lesion performed in 2003. Note the complex epithelium-lined spaces characteristic of a phyllodes tumor (H&E, 20).
Haagensen was one of the early advocates for wide excision instead of mastectomy.5 Wide local excision with 1-2 cm margins remains the mainstay of treatrnent.3-5, 8, 9, 12, 14 Simple mastectomy is preferred for larger tumors, in cases where appropriate margin clearance can not be achieved or for repeated local recurrencesdespite adequate margins.4, 12 Due to the fact that these tumors do not metastasize via lymphatic drainage, sentinel lymph node biopsy or axillary lymph node dissection is not recommended unless they are clearly involved with tumor.3-5, 7, 9 Palpable axillary lymph nodes is reported in 20 per cent of patients with phyllodes tumor5; however, only 5 per cent of all patients with palpable nodes have histological proof of malignancy,3, 5 and less than 1 per cent of all patients with MPT had positive nodes.9 Because in the setting of even malignant phyllodes tumors palpable lymphadenopathy is almost always reactive3, 4, 9 from tumor necrosis or secondary infection of ulcerating lesions,9 the presence of metastatic disease in the nodes should be confirmed prior to nodal dissection.3, 4
FIG. 6. Malignant phyllodes tumor. High-power magnification showing a cellular stroma consisting of spindle cells with nuclear atypia and abnormal mitoses (H&E, 400).
FIG. 7. Malignant phyllodes tumor. Residual area with hypocellular stroma and pericanalicular pattern of fibroadenoma (H&E, 40).
The role of adjuvant chemotherapy or radiation therapy is not well established due to lack of randomized trials secondary to rarity of this tumor.3, 5, 12, 15 Currently, there is no consensus regarding the use of adjuvant treatment.3, 8, 15 Although it has been shown that 20-40 per cent of phyllodes tumors are estrogen receptor positive and all exhibit the presence of progesterone receptors,3, 9 hormone therapy has no established role in their treatment.3-5, 9 However, in selected cases radiotherapy may be useful,3, 8, 13 such as inability to attain negative margins.14 There is anecdotal evidence suggesting a possible role for radiotherapy for borderline or malignant phyllodes tumors in the setting of breast conservation and when treated with mastectomy when the local failure risk is high. Chaney et al. reported on their experience with adjuvant radiation therapy for 8 patients with localized phyllodes tumors (2 benign, 1 borderline, 5 malignant; size 3.5-15 cm). With follow up of 22-84 months, they reported no local or distant failures. They concluded that adjuvant radiation therapy may be useful at doses of 50-60 Gy for patients with positive, 10 cm, or following resection of recurrent disease.8 Pandey et al. reported that adjuvant RT is thought to decrease local recurrence (LR) with no significant effect on survival.7 Furthermore, adjuvant radiation therapy should be considered in patients demonstrating stromal overgrowth in tumors >5 cm.3 The role of chemotherapy seems to be limited to the treatment of metastasis and for palliation of unresectable local recurrence.7
Local recurrence occurs 5-15 per cent for benign tumors and 20- 30 per cent for malignant cases.3, 14 Kapiris et al.9 reported the LR of MPT at 44 per cent and DM at 27 per cent. They concluded that both tumor size and margin status are important factors in local recurrence and distant metastasis. They found a 7-fold increase in LR for tumors greater than 10 cm and a 4-fold increase in LR when the surgical margins were involved. In their study, they also demonstrated LR to be an important predictor of future metastatic disease, with 85 per cent of patients who ultimately developed metastases, had suffered a previous LR.9 August et al. suggested that if radiation therapy is used after wide local excision of high risk lesions (tumor >5 cm, presence of stromal overgrowth, >10 mitoses/ hpf, infiltrating margins), for all comers local recurrence would likely be avoided in 90 per cent or more of patients.4 Chaney et al. reported stromal overgrowth as the strongest predictor of distant metastasis and ultimate outcome. Due to this fact, they feel that patients with stomal overgrowth feature, and specifically those with tumors >5 cm, should be considered for adjuvant chemotherapy.14
Pandey et al. reported the 5-year overall survival for patients with malignant phyllodes tumors at 74.2 per cent and the 5-year disease-free survival at 59.6 per cent.7 Chaney et al. reported 5- year and 10-year survival rates of 82 per cent and 42 per cent, respeclively.14 Pandey et al. and Shabahang et al. reported that the margin of surgical excision was the only independent prognostic factor for long-term survival.7- 16 However, it was felt that patients with larger tumors, >5 cm, had increased hazard, whereas those undergoing radiotherapy, married women, and age >35 showed a decreased hazard.7 Mortality is secondary to direct extension of local recurrences into the chest or from metastasis to lung, bone, or both.5
In a study by Noguchi et al.1 investigating the clonal analysis of fibroadenoma and phyllodes tumor it was shown that in fibroadenoma both epithelial cells and stromal cells were polyclonal. However, phyllodes tumor consists of polyclonal epithelial cells and monoclonal stromal cells. Therefore, phyllodes tumor can be regarded as stromal cells neoplasm. Because phyllodes tumor is often preceded or accompanied by fibroadenoma and these two entities have some histological similarity, they speculated that phyllodes tumor begins as fibroadenoma and subsequently a single stromal cell undergoes mutation and develops into a phyllodes tumor composed mainly of monoclonal stomal cells and partially of polyclonal epithelial cells. These data may prove to be clinically useful to distinguish fibroadenoma from phyllodes tumor by clonal analysis on small DNA samples obtained by FNA cytology.1
In a 1995 study, Noguchi et al.2 reported three cases of fibroadenoma that were diagnosed by excisional biopsy that recurred as benign phyllodes. Clonal analysis showed that all three fibroadenomas were monoclonal in origin. These results were inconsistent with their previous findings that fibroadenoma was polyclonal. Hence, it was suggested that from the clonal point of view there are two types of fibroadenoma, polyclonal and monoclonal, but they could not be differentiated histologically. It is speculated that the incidence of monoclonal fibroadenoma is quite low, therefore most of the usual fibroadenomas that can be cured by enucleation are polyclonal in origin and without tendency to recur locally. The percentage of monoclonal fibroadenomas against total fibroadenomas remains to be determined. Furthermore, the recurrent phyllodes tumors showed inactivation of the X chromosome-linked androgen receptor (AR) gene, which was also demonstrated in the primary fibroadenoma in each case. Hence, based on these results, they speculated that the recurrent phyllodes tumors were true recurrences of the primary fibroadenoma. The assumption was that the recurrence was due to residual tumor secondary to inadequate excision of the initial fibroadenoma. The authors therefore believe that fibroadenoma can progress to phyllodes tumor. Thus, they speculated that monoclonal fibroadenoma starts as a polyclonal fibroadenoma and due to a stromal cell line undergoing mutation, it forms a monoclonal fibroadenoma, which subsequently can progress to phyllodes tumor. Therefore, it was suggested that monoclonal fibroadenoma should be treated like phyllodes tumor with wide local excision.2
Follow-up after definitive surgery for malignant phyllodes tumor consists of biannual physical exam for 5 years, then annual, baseline unilateral mammogram 3 months postexcision with/without radiation if breast was conserved, then annual bilateral screening mammography. Biannual CT of the chest should be performed for 2-5 years for high-risk lesions.4
Conclusion
Based on the data from the literature and the case presented in this paper, it can be concluded that there are monoclonal fibroadenomas that can progress to phyllodes tumor. Therefore, it is important to differentiate between monoclonal and polyclonal fibroadenoma,as the former is treated surgically and the later is monitored clinically. Because the PCR method for clonal analysis can be performed on FNA sample, this distinction can be achieved without surgery.2 However, because performing clonal analysis on all fibroadenomas is time consuming and not cost effective, it may be a potentially valuable tool in evaluating rapidly growing fibroadenomas.
REFERENCES
1. Noguchi S, Motomura K, Inaji H, et al. Clonal analysis of fibroadenoma and phyllodes tumor of the breast. Cancer Res 1993; 53:4071-4.
2. Noguchi S, Yokouchi H, Aihara T, et al. Progression of fibroadenoma to phyllodes tumor demonstrated by clonal analysis. Cancer 1995;76:1779-85.
3. Guerrero MA, Ballard BR, Grau AM. Malignant phyllodes tumor of the breast: review of the literature and case report of stromal overgrowth. Surg Oncol 2003; 12:27-37.
4. August DA, Kearney T. Cystosarcoma phyllodes: mastectomy, lumpeetomy, or lumpectomy plus irradiation. Surg Oncol 2000;9:49- 52.
5. Geisler DP, Boyle MJ, Malnar KF, et al. Phyllodes tumors of the breast: a review of 32 cases. Am Surg 2000;66:360-6.
6. Tse GM, Ma TK, Pang LM, dieting H. Fine needle aspiration cytologie features of mammary phyllodes tumors. Acta Cytol 2002;46:855-63.
7. Pandey M, Mathew A, Kattoor J, et al. Malignant phyllodes tumor. Breast J 2001;7:411-6.
8. Chaney AW, Pollack A, McNeese M, Zagars GK. Adjuvant radiotherapy for phyllodes tumor of breast. Radiat Oncol Investig 1998;6:264-7.
9. Kapiris I, Nasiri N, A’Hern R, Healy V, Gui GP. Outcome and predictive factors of local recurrence and distant metastases following primary surgical treatment of high-grade malignant phyllodes tumours of the breast. Eur J Surg Oncol 2001;27:723-30.
10. Jayaram G, Sthaneshwar P. Fine-needle aspiration cytology of phyllodes tumors. Diagn Cytopathol 2002;26:222-7.
11. Scolyer RA, McKenzie PR, Achmed D, Lee CS. Can phyllodes tumours of the breast be distinguished from fibroadenomas using fine needle aspiration cytology? Pathology 2001;33:437-43.
12. Shabbir J, O’Sullivan JB, Mahmood S, Byrnes G. Phyllodes tumor of breast. J CollPhysicians Surg Pak 2003; 13:170-1.
13. Harada S, Fujiwara H, Hisatsugu T, Sugihara H. Malignant cystosarcoma phyllodes with lymph node metastasis; case report. Jpn J Surg 1987;17:174-7.
14. Chaney AW, Pollack A, McNeese MD, et al. Primary treatment of cystosarcoma phyllodes of the breast. Cancer 2000;89:1502-11.
15. Khan SA, Badve S. Phyllodes tumors of the breast. Curr Treat Options Oncol 2001;2:139-47.
16. Shabahang M, Franceschi D, Sundaram M, et al. Surgical management of primary breast sarcoma. Am Surg 2002;68:673-7.
EDNA K. VALDES, M.D., SUSAN K. BOOLBOL, M.D., JEAN-MARC COHEN, M.D., SHELDON M. FELDMAN, M.D., F.A.C.S.
From the Louis Venet Comprehensive Breast Service, Beth Israel Medical Center, New York, New York
Address correspondence and reprint requests to Sheldon M. Feldman, M.D., F.A.C.S., Phillips Ambulatory Care Center, 10 East Union Square, Suite 4E, New York, NY 10003.
Copyright The Southeastern Surgical Congress Apr 2005
Comments