The clinical features and prognosis of phyllodes tumors:a single institution experience in Taiwan

The clinical features and prognosis of phyllodes

tumors: a single institution experience in Taiwan

Chin-Chan Lin1_{,Hui-Wen Chang}2_{, Chen-Yuan Lin}2_{, Chang-Fang Chiu}2_,

Shin-Pen Yeh2

1. Division of Hematology and Oncology, Department of Internal Medicine, China Medical University Hospital, Taichung, Taiwan 2. Department of Pathology, China Medical University Hospital,

Taichung, Taiwan

Corresponding author:

Chang-Fang Chiu, Ph.D.

China Medical University Hospital, Taichung, Taiwan 2, Yude Road, Taichung,

Taiwan 404

Tel: +886-4-22052121 ext 3484 Fax: +886-4-22034421

Abstract:

Background and Objectives

The aim of this study was to document the clinical and pathological features of a single institutional series of Asian origins with phyllodes tumors, and to determine the prognosis, the adequate management and predictive histologic features.

Methods

between 2003 and 2010

.

Results

phyllodes tumors recurred, but those with malignant phyllodes tumors had high recurrence rate(41％). The 5 year disease-free survival (DFS)

was 59％. The 5 year overall survival was 81％. The width of surgical

margin is not related to disease recurrence and the stromal overgrowth is the only prognostic factor in terms of disease-free survival and

overall survival.

Conclusions

The phyllodes tumors of borderline and malignant classification in Asian patients had high recurrent rate. Clinical and pathologic factors except stromal overgrowth can’t predict disease recurrence. Further molecular researches are warranted.

Mini-abstract:

Breast; Phyllodes tumors; Pathology

Introduction:

Phyllodes tumors of the breast are rare fibroepithelial neoplasm that account for 0.3％ to 0.9％ of all breast tumors in women [1, 2].

Although most phyllodes tumors occur in women between the ages of 35 and 55 years, adolescent and elderly women are also affected. Clinical presentation and behavior are largely similar to those of

fibroadenoma of the breast and the diagnosis is often only made after histological findings of increased stromal cellularity with leafy fronds [3]. However, unlike the more common fibroadenoma, Between 10％

and 40％ of phyllodes tumors have a malignant biology and a tendency

for local recurrence and general dissemination [4].

Many controversial issues existed. In the report of WHO, the overall recurrence rate of phyllodes tumors was 21 percent. However several

studies showed varied recurrence rates. It should be noted that Asian women seemed to recur more often compared with Western

population. In terms of managment, most authors recommend excision of phyllodes tumors with clear margin, but the optimal margin width is unknown [5].

In addition, to date, still controversy as to whether histological features, such as histological grade and the degree of stromal

overgrowth, could reliably predict local recurrence [6,7]. In most series, mortality associated with phyllodes tumors invariably occurs after metastatic spread from malignant tumors, which often have a very poor response to chemotherapy and radiotherapy. It therefore becomes important to understand the factors that increase the likelihood of local recurrence, and more importantly, the factors that may predict for a malignant recurrence. While still fraught with many questions and occasionally conflicting results, advances in

immunohistochemical and molecular methods have shed light on the biological nature of this neoplasm.

This retrospective study analyzed the clinicopathological features of a cohort of phyllodes tumors of Asian origins presenting to a tertiary

hospital in Taichung, Taiwan since 2003 to 2010. It is our aim to decide the prognosis, adequate management and correlation with histologic characteristics.

Materials and Methods

Patients and clinical data

The data base from the department of pathology at China Medical University Hospital, Taichung, Taiwan was utilized to identify 33

patients diagnosed with phyllodes tumors by histopathology between 2003 and 2010.

Clinical data was obtained from the Medical Records Department of China Medical University Hospital, Taichung, Taiwan. Each individual case history was reviewed to obtain demographic and clinical data which included: age at diagnosis, sex, data of diagnosis, surgery type, treatment outcome, recurrence, lymph node involvement, distant metastasis, length of follow-up, dead or alive, and other breast tumors/diseases. Patients’ characteristics, pathologic variables, and surgical procedures were investigated as predictors of local recurrence, distant metastasis, and survival. Two types of surgical procedures—

partial mastectomy and mastectomy were performed, depending on tumor size, margin involvement, and patient or surgeon preference. Tumors were subdivided into three groups according to their maximum measured diameter: <5 cm, 5 to 10 cm, and >10 cm. Treatment failure were classified as local recurrence (involving the breast, chest

wall/axilla, or surgical bed) or distant metastasis

Pathological diagnosis

All of the original slides of hematoxylin and eosin were reviewed to confirm the diagnosis. Pathologic data obtained from the macroscopic reports and histopathological reviews included: stromal overgrowth

(Stromal overgrowth was defined as stromal proliferation to the point

where epithelial elements were absent in at least one low-power field(X40)present or absent; defined as at least one lower power

field(40× total magnification) composed entirely of stroma without

epithelium), stromal atypia (mild, moderate, severe), stromal mitoses (defined as mitotic counts per 10 high per field (hpf)), margin (pushing with compression of adjacent tissues, or infiltrative), margin

involvement (close, and involved), and size.

malignant phyllodes tumors. The phyllodes tumors were classified as benign if they had low cellularity, no stromal overgrowth, mild stromal cell atypia, pushing margins and mitotic counts less than 2 per 10 hpfs. They were classified as malignant phyllodes tumors if they had stromal overgrowth, infiltrative margins and mitotic counts more than 5 per 10 hpfs. The phyllodes tumors that did not meet the criteria for benign or malignant phyllodes tumors were classified as borderline tumors.

Stastistics

Life tables were constituted with the Kaplan-Meier method, and survival curves were compared by using the log-rank test. Univariate and multivariate Cox regression analysis were used to assess the relationship between the clinical, surgical, and pathologic factors and outcomes of local recurrence, distant metastasis, and survival. Hazard ratios were constructed by using Cox regression analysis. P value less than 0.05 was significant.

Results

Patient Demograhics/Presentations

67 years (mean: 46 year-old, median: 49 year-old). Nine patients (27％)

were postmenopausal, and 24 (73％) were premenopausal. Seventeen

patients(52％) had the tumors in the left breast, and Sixteen(48％)

patients in the right breast. No patients had bilateral diseases. Seven patients(21％) reported a history of benign breast masses and 5 of

these 7 patients were diagnosed to have fibroadenoma.

Histologically, of the 33 phyllodes tumors, 8 were benign tumors, 13 borderline and 12 malignant. The maximal dimension ranged from 1.5 cm to 28 cm (mean: 7.4 cm, median: 6.0 cm). Thirteen patients (39％)

had tumors less than 5cm in their greatest dimension; 13

patients(39％) had tumors between 5 to 10 cm, and 7 patients(22％)

had tumors more than 10 cm.

Primary tumor treatment

had received mastectomy and 26 patients (78％; 7 benign, 11

boderline, and 8 malignant) had partial mastectomy. Axillary dissection was performed in 9 patients (27％) but no one had pathologically

proven nodal involvement. Neither adjuvant chemotherapy nor radiotherapy were administered.

Clinical outcome

Nine patients had recurrence (6 patients with local recurrence, 1 distant metastases, and 2 both with local recurrence and distant metastases). Among 13 patients with borderline phyllodes tumors, 3 patients(23％) had recurrence. Among 12 patients classified as

malignant phyllodes tumors, 5 patients(41％) had recurrence. No

patients classified as benign phyllodes tumors had local recurrence or distant metastasis. (Table 1)

With a median follow-up of 21 months (range:0.4-78 months), local recurrence occurred in 8 patients (24％) at an average of 9.5 months

after diagnosis (median: 7.5 months; range: 2-21 months). The local recurrent rate after partial mastectomy and mastectomy were 39％

and 17％, respectively (P=0.5). All six patients with local recurrence involved chest wall or surgical bed. No one had locoregional lymph node metastasis. Of the 7 patients who had a local recurrence after partial mastectomy, only 1 patient(14％) developed distant metastasis.

Only one of 7 patients receiving mastectomy had local recurrence after mastectomy and later she developed distant metastasis.

In univariate analysis, local recurrence was significantly increased by the presence of stromal overgrowth (p=0.016), but he other variables, including surgical margins, classification, surgical types, tumor borders, mitosis counts, all showed no significance in terms of local recurrence rates. In multivariate analysis, the presence of stromal overgrowth also had significance in terms of local recurrent rates.(Table 2)

Among 7 patients with benign phyllodes tumors, there were 3 patients with closed margins and 4 with free margins (defined as margin more than 1 centimeter). Among 12 patients with borderline tumors, 5 patients have closed margins and 7 free margins. Three of 13 patients with borderline tumors have local recurrence, all of whom have free margins. Among 12 patients with malignant tumors, 8

patients have closed margins and 4 free margins. Five of 12 patients with malignant tumors have local recurrence. Two of these 5 patients had closed margins and three patients had free margins.

Distant metastasis occurred in 3 (9％)of 33 patients at a mean of

22.5 months after the primary operation (median:17 months; range: 8-20 months). One patient had lung, brain, and small intestine

third patient had lung metastasis. Two of the 3 patients with distant metastases also had a local recurrence before the diagnosis of distant metastases. One patient received chemotherapy and another patient received chemotherapy and palliative radiotherapy. These 2 patients died 5 months and 1 month later after the diagnosis of distant

metastases, respectively. The third patient with lung, brain and small intestine metastasis had poor response to chemotherapy. She had received several cycles of tumor resection and survived 17 months after diagnosis of metastases.All these three patients had received chemotherapy due to distant metastasis. The chemotherapy regimens were composed of doxorubicin and ifosfamide, epirubicin and

paclitaxel, docetaxel and gemcitabine, respectively. However, all three patients had disease progression after chemotherapy.

Survival

Follow-up data were available on all 33 patients. The duration of follow-up ranged from 0.4 month to 78 months (mean=27 months; median=21 months). The overall 5 year disease-free survival (DFS) was 59％(Figure 1). In terms of events of local recurrence and distant

respectively (Figure 1 and 2).The 5 year overall survival was 81％

(Figure 32). No patients with only local recurrence died until last follow-up time. However all three patients developing distant metastasis passed away.

In univariate analysis, stromal overgrowth had significant effect on patient survival (p=0.047). Other variables such as classification, surgical margins , tumor borders, mitosis counts , and stromal atypia had no statistically significant impact on survival rates.

Discussion

In the current study, 7 (21％)Patients reported a history of benign

breast masses and 5 of these 7 patients were diagnosed to have fibroadenoma. This results in a challenge of differential diagnosis between these two diseases. The more commonly benign phyllodes tumor’s mild stromal hypercellularity can histologically overlap with the cellular fibroadenoma . Frequently, they are morphologically indistinguishable on limited tissues like need core biopsies [8]. Whilst phyllodes tumors are typically thought to be larger than fibroadenoma.

to 28 cm) with a larger mean of 7.4 cm and median of 6 cm.The size range overlaps considerably with that of fibroadenoma, emphasizing that size alone cannot be used as a criterion to distinguish between phyllodes tumors and fibroadenoma [5, 9-10].

Phyllodes tumor is a rare tumor that behaves unpredictably. Surgical treatment remains the mainstay of therapy for patients with phyllodes tumor. In most patients, axillary lymph nodes are not palpable at presentation, because metastatic spread of these tumors is primarily hematogenous. Seven patients (22％) had received mastectomy in this

analysis and 9 patients (27％) had received axillary dissection. The

reason to perform mastectomy and axillary dissection is mostly due to huge mass of primary tumors. Six of seven patients had primary tumor larger than 8.5 centimeter (range 8.5-28 centimeter). One patient with the primary tumor sized 4.5 centimeter had received mastectomy

because of herself’s discretion. No lymph node metastasis was noted in

our patient who receiving axillary dissection series. Lymphadenectomy is indicated only for suspicious lymph nodes in women with phyllodes tumors.

an important risk factor for local recurrence and have recommended surgical margin should be more than 1 centimeter in width [11,12]. In our series, no relationship between surgical margin and tumor

recurrence was noted. Not all patients with margin involvement had recurrence, and malignant phyllodes tumors may recur and metastasize despite of excision or mastectomy with adequate margins [6,11]. Some institutions also are questioning the need for the 1cm clearance in phyllodes tumors [7, 13].

It is well established that there is a wide range in the age at diagnosis of phyllodes tumors but it is most frequent in middle age with a median age at presentation of 45 year. Several authors have found that phyllodes tumors present at an earlier age in those of Asian and Latino White origin and recur often [5, 9,10,14]. In the current study including 33 Asian patients, the median age of diagnosis was 49 years old, which were not different to the presentation age of western population. However, higher recurrence rate of 27％ compared to 21％

in the report of WHO was noted [5]. Particularly, in patients classified borderline or malignant phyllodes tumors, the recurrent rate was up to 23％ and 41％, respectively. The reason of high recurrence rate in our

series is not clear.

Many authors, including those of the WHO, suggest that local

recurrence rates increased with classification but most studies have not shown a statistically significant difference[5, 15-16].In our series, no patient classified as benign phyllodes tumors had tumor recurrence. Although the limitation of small sample sizes need to be concerned, simple excision with negative margin may be adequate for benign phyllodes tumors. Regard to malignant phyllodes tumors, adjuvant therapy in the context of clinical trials is warranted because of unacceptable high recurrent rate.

Compounding treatment decisions is the lack of reliable histologic indicators that predict recurrence. Different studies have regarded stromal overgrowth, infiltrating margins, high mitotic rate, and degree of stromal atypia as important predictors of recurrence and/or

prognosis ,while othesr have disagreed with these findings [17-20]. In our series, stromal overgrowth is significantly related to local

recurrence rate, distant metastatic rate, and overall survival. The pathogenic mechanisms involved in the progression are complex. The current evidence support a model where initiation

involves epithelial stromal interactions, and tumor progression indicate the loss of stromal epithelial interdependence. The morphology of stromal overgrowth represented a critical step that the stroma becomes autonomous of the cell signaling massages from the epithelium, which involves cell cycle markers, p53, c-kit/CD117, angiogenic factors, adhesion molecules and matrix proteins [21].

In summary, this is a single institution experience of a rare tumor. The high local recurrence rate was noted in the category of borderline and malignant phyllodes tumors. Stromal overgrowth was the only

parameter related significantly to outcome. In contrast to some studies in the literature, no relationship was found between surgical margin width and subsequent recurrence. High recurrence rate of malignant phyllodes tumors warranted further researches in the pathogenesis and clinical intervention.

Acknowlegements:

This study is supported from the research grant of China Medical University Hospital (DMR-100-170)

Conflict of interest statement

The authors declare no conflict of interest

References

1.Chaney AW, Pollack A, McNeese MD, et al. (2000) Primary treatment of cystosarcoma phyllodes of the breast. Cancer 89: 1502-11.

2.Rowell MD, Perry RR, Hsiu JG, Barranco SC. (1993) Phyllodes tumors. Am J Surg 165: 376-9.

3.Bennett IC, Khan A, De Freitas R, et al. (992) Phyllodes tumours: a clinicopathological review of 30 cases. Aust N Z J Surg 62: 628-33.

4.Asoglu O, Ugurlu MM, Blanchard K, et al. (2004) Risk factors for recurrence and death after primary surgical treatment of malignant phyllodes tumors. Ann Surg Oncol 11: 1011-7.

5.Karim RZ, Gerega SK, Yang YH, et al. (2009) Phyllodes tumours of the breast: a clinicopathological analysis of 65 cases from a single institution. Breast 18: 165-70. 6.Barth RJ, Jr. (1999) Histologic features predict local recurrence after breast conserving therapy of phyllodes tumors. Breast Cancer Res Treat 57: 291-5.

7.Reinfuss M, Mitus J, Duda K, et al. (1996) The treatment and prognosis of patients with phyllodes tumor of the breast: an analysis of 170 cases. Cancer 77: 910-6. 8.Jara-Lazaro AR, Tan PH. (2009) Molecular pathogenesis of progression and recurrence in breast phyllodes tumors. Am J Transl Res 1: 23-34.

9.Parker SJ, Harries SA. Phyllodes tumours. (2001) Postgrad Med J 77: 428-35. 10.Jacklin RK, Ridgway PF, Ziprin P, et al. (2006) Optimising preoperative diagnosis in phyllodes tumour of the breast. J Clin Pathol 59: 454-9.

11.Kapiris I, Nasiri N, A'Hern R, et al. (2001) 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 27: 723-30. 12.Chua CL, Thomas A, Ng BK. (1989) Cystosarcoma phyllodes: a review of surgical options. Surgery 105: 141-7.

13.Salvadori B, Cusumano F, Del Bo R, et al. (1989) Surgical treatment of phyllodes tumors of the breast. Cancer 63: 2532-6.

14.Bernstein L, Deapen D, Ross RK. (1993) The descriptive epidemiology of malignant cystosarcoma phyllodes tumors of the breast. Cancer 71: 3020-4.

15.Mangi AA, Smith BL, Gadd MA, et al. (1999) Surgical management of phyllodes tumors. Arch Surg 134: 487-92; discussion 92-3.

16.Mokbel K, Price RK, Mostafa A, et al. (1999) Phyllodes tumour of the breast: a retrospective analysis of 30 cases. Breast 8: 278-81.

17.Hawkins RE, Schofield JB, Fisher C, et al. (1992) The clinical and histologic criteria that predict metastases from cystosarcoma phyllodes. Cancer 69: 141-7. 18.Tse GM, Putti TC, Kung FY, et al. (2002) Increased p53 protein expression in malignant mammary phyllodes tumors. Mod Pathol 15: 734-40.

19.Pietruszka M, Barnes L. (1978) Cystosarcoma phyllodes: a clinicopathologic analysis of 42 cases. Cancer 41: 1974-83.

20.Contarini O, Urdaneta LF, Hagan W, et al. (1982) Cystosarcoma phylloides of the breast: a new therapeutic proposal. Am Surg 48: 157-66.

21.Karim RZ, Scolyer RA, Tse GM, et al. (2009) Pathogenic mechanisms in the initiation and progression of mammary phyllodes tumours. Pathology 41: 105-17.

Figure Legends

Figure 1. Time to local recurrence during follow up

Figure 2. Time to ditant metastasis during follow up Figure 3. Time to death during follow up