Japanese Journal of Clinical Oncology Advance Access originally published online on October 11, 2007
Japanese Journal of Clinical Oncology 2007 37(10):730-736; doi:10.1093/jjco/hym099
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2007 Foundation for Promotion of Cancer Research
Phyllodes Tumor of the Breast: Stromal Overgrowth and Histological Classification are Useful Prognosis-predictive Factors for Local Recurrence in Patients with a Positive Surgical Margin
1 Departments of Surgery, National Hospital Organization, National Shikoku Cancer Center, 160 Kou Minamiumemoto-machi, Matsuyama-city, Ehime 791-1288, Japan
2 Departments of Pathology, National Hospital Organization, National Shikoku Cancer Center, 160 Kou Minamiumemoto-machi, Matsuyama-city, Ehime 791-1288, Japan
For reprints and all correspondence: N. Taira, Dentistry Pharmaceutical Sciences, Okayama University Graduate School of Medicine, 2-5-1 Shikata, Okayama 700-8558, Japan; E-mail: ntaira{at}md.okayama-u.ac.jp (Present address)
Received March 23, 2007; accepted June 16, 2007
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Background: The local recurrence rate of phyllodes tumors is high and ensuring a sufficient surgical margin is considered important for local control. However, the preoperative diagnosis rate of phyllodes tumors is low and we often encounter cases in which a sufficient surgical margin is not achieved, since in routine medical practice the lesion may not be diagnosed as phyllodes tumor until postoperative biopsy of a mammary mass. Furthermore, there are no established therapeutic guidelines for surgical stump-positive phyllodes tumors. We reviewed the outcomes of excision of phyllodes tumors to investigate factors involved in local recurrence and to determine the indication for re-excision in stump-positive cases.
Methods: The subjects were 45 patients treated for phyllodes tumors at our institution from January 1980 to July 2005. Age, tumor size, surgical method, stromal cellular atypia, mitotic activity, stromal overgrowth, histological classification and surgical stump status were analyzed.
Results: Median age was 45 years old (range 28–75) and tumor size was 1–17 cm (median 3.5 cm). Pathologic diagnoses were benign, borderline and malignant in 31, five and nine cases, respectively, and the surgical stump was negative in 27 lesions and positive in 15. Median follow-up was 101 months (range 1–273), with local recurrence in six cases and distant metastasis in one. The local recurrence-free rate was 88, 88 and 84% and the disease-free rate was 85, 85 and 81% after 5, 10 and 15 years, respectively. Overall 10-year survival was 97%. In univariate analysis, a positive surgical margin, stromal overgrowth and histological classification were predictive factors for local recurrence after breast-conservation surgery (P = 0.0034, 0.0003, 0.026). A positive surgical stump was the only independent predictor of local recurrence in multivariate analysis (RR 0.086; 95% CI 0.01–0.743, P = 0.012). Stromal overgrowth was a predictive factor for local recurrence in cases with a positive surgical margin (P = 0.0139).
Conclusion: Wide excision is the preferred therapy for phyllodes tumor and preoperative diagnosis is important for good local control. Re-excision is recommended in cases with a positive surgical margin and stromal overgrowth and malignancy.
Key Words: phyllodes tumor • cystosarcoma phyllodes • breast tumor
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Surgery has been the primary mode of treatment of phyllodes tumor, with mastectomy being the common treatment of choice. However, surgery in which the breast is conserved has become increasingly common in management of phyllodes tumors for cosmetic reasons. There is a relatively high incidence of local recurrence associated with phyllodes tumors, and age, tumor size, surgical approach, mitotic activity, stromal overgrowth and surgical margin have been reported as prognosis-predictive factors related to local recurrence (1–11). Among these, the most important factor may be the surgical margin. Therefore, a goal of surgical treatment is to ensure a negative surgical margin after excision. However, the preoperative diagnostic accuracy of phyllodes tumor is often poor because mammary phyllodes tumors are rare and their clinical, imaging, cytology and histology characteristics are similar to those of fibroadenomatous breast tumor, which has a high incidence (12–17). Thus, lesions may not be diagnosed as phyllodes tumors until postoperative pathological examination in routine clinical practice. Furthermore, criteria for handling cases with a positive surgical margin have not been established. The main biological feature of phyllodes tumors is latent malignancy and the course is benign in most cases. Based on these features, we followed the course of cases of phyllodes tumor in which re-excision was not performed, even though the surgical stump was positive. Here, we review the outcomes of these cases to investigate the necessity of re-excision in stump-positive cases and to clarify factors associated with local recurrence.
|
PATIENTS AND METHODS |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Medical records of 45 patients who underwent initial excision and were histologically diagnosed with phyllodes tumor between January 1980 and July 2005 at the Shikoku Cancer Center were investigated retrospectively. The data extracted included the duration of illness, tumor size, tumor localization, mammography findings, echography findings, fine needle aspiration cytology findings, biopsy diagnosis, preoperative diagnosis, age at the time of surgery and surgical procedure. Regarding the follow-up period, the last day that a patient visited hospital within 1 year after surgery (until November 2006) was regarded as the final follow-up date. Patients not examined within 1 year of surgery were surveyed by telephone and the day of this interview was regarded as the final follow-up date.
All excised mammary masses were fixed in formalin and sectioned into 5 mm slices (10 mm thickness for larger masses), regardless of the benign or malignant status of the tumor. Paraffin blocs were prepared and HE-stained for diagnosis, thereby allowing a detailed pathological investigation of the phyllodes tumors. All slides of phyllodes tumors were reviewed by two pathologists without knowledge of the original diagnosis and clinical outcome. Numerous studies have attempted to determine which histologic features of phyllodes tumors are useful in predicting clinical behavior (1–4,18–25). Most pathologists use a combination of several histologic features to classify phyllodes tumors, with diagnosis based on the three most widely accepted classification features: (1) the degree of stromal cellular atypia; (2) mitotic activity per 10 high-power fields (hpfs); and (3) presence or absence of 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) (22). Based on these three features, the phyllodes tumors were classified as benign, borderline and malignant (Table 1). Increasingly greater importance has been attached to the condition of the tumor margin (circumscribed or infiltrative) as a subclassification factor for phyllodes tumors. Since many patients in the current study underwent tumor extirpation, accurate judgment of the tumor margin was difficult in all patients. Accordingly, the tumor margin condition was not included as a subclassification factor. All slides were examined for evaluation of surgical stumps, and the surgical margin was judged to be positive when at least one tumor cell was exposed at the margin.
|
The clinicopathological backgrounds of the 45 patients are shown in Table 2. The age of the patients ranged from 28 to 75 years old, with a median age of 45 years old. The chief complaint was a mammary mass in all patients. The earliest and latest times of visiting hospital after the patient noticed the mass were 1 day and 10 years, respectively, and the median time was 3 months. The largest dimension of the tumor ranged from 1 to 17 cm, with a median size of 3.5 cm. Twenty-eight patients (62%) had tumors of <5 cm in size (largest dimension) and 17 patients (38%) had tumors
5 cm in size. Twenty-five patients (56%) presented with left-sided lesions and 20 patients (44%) had right-sided lesions.
|
Fine needle aspiration cytology was performed in 41 of the 45 patients, and the diagnosis was benign in 33 patients, borderline in seven and malignant in one. Preoperative biopsy was performed in 10 patients, core-needle biopsy in eight, mammotome biopsy with echo guiding in one, and incisional biopsy in one. The histological diagnosis of the eight cases that underwent core-needle biopsy was phyllodes tumor in four patients, fibroadenoma in three and mastopathy in one. All cases that underwent echo-guided mammotome biopsy and incisional biopsy were histologically diagnosed as phyllodes tumor. The preoperative diagnosis based on palpation, mammography, mammary echography, cytology and biopsy histology was phyllodes tumor in 14 patients, fibroadenoma in 14, suspected breast cancer in seven, breast cancer in one, cyst in one and unclear in eight. The preoperative diagnosis rate of phyllodes tumor was 31%. The histologic type was benign in 31 patients (69%), borderline in five (11%), and malignant in nine (20%).
The surgical procedure was mastectomy in three patients and breast-conservation surgery in 42 patients. Axillary dissection was not performed in any patients. The breast-conservation surgery included various types ranging from enucleation of the tumor to wide excision to ensure a negative tumor margin. In the breast-conservation surgery group, the surgical margin was negative in 27 patients (64%) and positive in 15 (36%). None of the surgical stump-positive cases underwent re-excision. Treatment failures were classified as local (involving the breast or chest wall/axilla surgical bed) or distant. The local recurrence-free period was defined as the period between the day of surgery and the day of diagnosis of local recurrence. Univariate actuarial curves were plotted using the Kaplan–Meier method and statistical comparisons were accomplished using a log-rank test. Multivariate analysis was performed using the Cox proportional hazards regression model.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The median duration of follow-up was 101 months (range 1–273 months). Local recurrence was noted in six patients and distant metastasis occurred in one patient during the follow-up period. The 5-, 10-, and 15-year cumulative local recurrence-free rates were 88, 88 and 84%, respectively, and the cumulative disease-free rates were 85, 85 and 81%, respectively. Death occurred only in the patient with distant metastasis, and the 10-year survival rate was 97%.
LOCAL RECURRENCE
The median time of local recurrence was 18.5 months (range 1–180 months). All six cases of local recurrence involved relapse of the tumor in the mammary tissue near the wound of the initial excision following breast-conservation surgery, and the median size of the recurrent tumors was 3.3 cm (range 1–5.5 cm). No patients developed regional lymph node metastasis in the axillary, supraclavicular or infraclavicular lymph nodes. No local recurrence occurred in the mastectomy group. Thus, factors associated with local recurrence were analyzed in 42 cases in which breast-conservation surgery was performed, excluding mastectomized cases. In univariate analysis, the surgical margin (P = 0.0034), stromal overgrowth (P = 0.0003) and histological classification (P = 0.026) were significant predictive factors for local recurrence (Fig. 1, Table 3). In multivariate analysis, the surgical margin was the only independent predictive factor for local recurrence (risk ratio, 0.086; 95% confidence interval, 0.01–0.743; P = 0.026). Local recurrence was detected in five of the 15 surgical stump-positive cases. In the stump-positive cases, stromal overgrowth and histological malignancy were noted in two cases showing local recurrence. Factors involved in local recurrence were investigated in the stump-positive cases, and stromal overgrowth was found to be a significant prognosis-predictive factor (P = 0.0139, Table 4). Histological malignancy was excluded as an investigative factor in this analysis because it was found in stromal overgrowth-positive cases.
|
|
|
DISTANT METASTASIS
Distant metastasis was noted in one patient during the follow-up period and this was the only fatal case. The patient was a 75-year-old female who underwent breast-conservation surgery for a 5 cm tumor in the left breast. Regarding the pathological characteristics of this tumor, stromal cell atypia was marked, mitotic activity was 3/HPF, stromal overgrowth was present and the tumor margin was positive. Shadows of multiple lung nodules were detected 3 years after surgery and lung metastasis from phyllodes tumor was histologically diagnosed by trans-bronchial lung biopsy. At the time of the diagnosis of lung metastasis the general condition was poor, and the patient died of respiratory failure before initiation of anticancer drug therapy 2 months after diagnosis of lung metastasis.
TREATMENT AFTER LOCAL RECURRENCE
Cases of local recurrence tended to be repeated recurrence. Five of the six patients underwent recurrent tumor resection twice and one patient underwent this procedure three times during the follow-up period, but no patients have required mastectomy to date. A comparison of histopathology between the initially excised and recurrent tumors indicated no marked changes in any patient; i.e. no histological aggravation.
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Phyllodes tumors are rare, accounting for less than 1% of all cases of mammary tumors (1,26). The biological features of these tumors include latent malignancy and the tumors show various clinical findings. Although a wide range of histological findings are classified as phyllodes tumor, the pathological findings are not necessarily consistent with the clinical course. There have been many reports concerning pathological prognostic factors (1–4,18–25).
Since phyllodes tumors tend to recur at the local site, local treatment requires wide excision to ensure inclusion of the tumor margin. Mastectomy is the most reliable procedure with regard to local control, but breast-conservation surgery is currently selected in most cases (excluding cases with very large tumors) for aesthetic reasons. The frequency of local recurrence has varied from 8 to 46% in previous reports, and age, tumor size, surgical approach, mitotic activity, stromal overgrowth and surgical margin have been reported as prognosis-predictive factors related to local recurrence (1–11). Among these, the most important factor may be the surgical margin. Pandey et al. reported that a negative surgical margin independently predicts improved disease-free survival and decrease local recurrence (7). Mangi et al. reported that the frequency of local recurrence was correlated with the status of the surgical margin, and the incidence of local recurrence was very low in cases with 1 cm or wider margins (5). In a study of a large series of 101 patients, Chaney et al. found no significant predictive factors for local recurrence, in part because the 10-year local recurrence rate was only 8%, a very good surgical outcome (10). These good therapeutic results were suggested to be due to ensuring a negative surgical margin during excision (the stump positivity rate was 1%).
In our patients, the surgical stump was an independent prognosis-predictive factor related to local recurrence, supporting the previous reports. To achieve excision ensuring a negative surgical margin, the extent to which phyllodes tumor is diagnosed before surgery is very significant. Triple assessment by clinical, radiological and cytological or histological examinations forms the fundamental basis for evaluation of all breast lumps. In patients with phyllodes tumors, all three methods individually have low sensitivity and, even in combination, the diagnostic accuracy is often poor because mammary phyllodes tumors rarely develop and their clinical, imaging and cytology and histology findings are similar to those of fibroadenomatous breast tumor, which has a high incidence (12–17). In our patients, phyllodes tumor was diagnosed before surgery in 14 cases (31%) and the stump positivity rate was 33%; this high rate may have been due to the low rate of preoperative diagnosis. The preoperative diagnosis rate of phyllodes tumors may be increased by more frequent application of core biopsy and echo-guided mammotome biopsy.
Cases of pathological diagnosis of phyllodes tumors are encountered unexpectedly after excision of mammary tumors in routine medical practice. Our data raise a question regarding the handling of such cases that are stump-positive. The local recurrence rate in stump-positive cases and cases with tumors localized near the stump has been reported to be about 20% (27), and in our patients the local recurrence rate was 33% in stump-positive cases. Many reports have recommended immediate excision to ensure a negative surgical margin in stump-positive cases because of the high local recurrence rate, whereas a ‘wait-and-watch’ policy has been proposed as acceptable for benign lesions because the local recurrence rate of benign and borderline phyllodes tumors is lower than that of malignant cases (8,28). We have followed the courses of stump-positive cases in which re-excision was not performed because the phyllodes tumor was latent malignancy and the course was benign in most cases. The subsequent local recurrence rate in stump-positive cases was 33%, and the presence of stromal overgrowth and a malignant tumor should lead to consideration of re-excision in such cases. In univariate analysis, the presence of stromal overgrowth and malignancy were significant predictive factors for local recurrence, and in the 15 stump-positive cases in our series, local recurrence was noted in both cases judged positive for stromal overgrowth and malignant tumor (100%), but in only three of the 13 cases judged negative for stromal overgrowth and with a benign or borderline tumor (23%). Stromal overgrowth was a significant prognosis-predictive factor in stump-positive cases. These results suggest that the presence or absence of stromal overgrowth and the histological classification are useful reference factors in consideration of re-excision. We recommend ensuring inclusion of the tumor margin by re-excision in cases with a positive surgical margin and stromal overgrowth and malignancy. Cases with negative stromal overgrowth and a benign/borderline tumor have a local recurrence rate of 23%, and we propose that a ‘wait-and-watch’ policy is acceptable in these cases.
The role of radiation therapy for phyllodes tumors remains unclear, since clinical data supporting the use of adjuvant radiation is based on anecdotal case reports (7,29), rather than on large patient series. No series has shown radiation therapy to be of benefit in the primary treatment of phyllodes tumors, and the M.D. Anderson series did not support the use of adjuvant radiotherapy for patients with adequately resected disease (10).
Distant metastases rates of up to 17% have been reported (1–11). In our series of 45 patients, one patient (2.2%) developed distant metastases during a mean follow-up period of 101 months, but we have an insufficient number of cases to evaluate the prognostic factors for distant metastases. However, we note that most distant metastases develop from borderline or malignant tumors. Many histological prognostic factors for distant metastases have been evaluated, including stromal overgrowth, infiltrating margins, mixed mesenchymal components, high mitotic rate and stromal atypia (1–11). The influence of local recurrence on survival is unclear, but sufficient local control may be important in cases judged to be malignant.
In conclusion, in excision of phyllodes tumors it is important to achieve a negative surgical margin for good local control. A phyllodes tumor with a positive surgical margin diagnosed based on excisional biopsy is likely to have a high local recurrence rate, particularly in cases with stromal overgrowth and a malignant tumor. Thus, in such cases we recommend ensuring inclusion of the tumor margin by re-excision.
Conflict of interest statement
None declared.
|
References |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
1 Reinfuss M, Mitus J, Duda K, Stelmach A, Rys J, Smolak K. The treatment and prognosis of patients with phyllodes tumor of the breast: an analysis of 170 cases. Cancer (1996) 77:910–6.[CrossRef][Web of Science][Medline]
2 Pietruszka M, Barnes L. Cystosarcoma phyllodes: a clinicopathologic analysis of 42 cases. Cancer (1978) 41:1974–83.[CrossRef][Web of Science][Medline]
3 Hines JR, Murad TM, Beal JM. Prognostic indicators in cystosarcoma phylloides. Am J Surg (1987) 153:276–80.[CrossRef][Web of Science][Medline]
4 Cohn-Cedermark G, Rutqvist LE, Rosendahl I, Silfversward C. Prognostic factors in cystosarcoma phyllodes. A clinicopathologic study of 77 patients. Cancer (1991) 68:2017–22.[CrossRef][Web of Science][Medline]
5 Mangi AA, Smith BL, Gadd MA, Tanabe KK, Ott MJ, Souba WW. Surgical management of phyllodes tumors. Arch Surg (1999) 134:487–92. [Discussion 492–3.].
6 de Roos WK, Kaye P, Dent DM. Factors leading to local recurrence or death after surgical resection of phyllodes tumours of the breast. Br J Surg (1999) 86:396–9.[CrossRef][Web of Science][Medline]
7 Pandey M, Mathew A, Kattoor J, Abraham EK, Mathew BS, Rajan B, et al. Malignant phyllodes tumor. Breast J (2001) 7:411–6.[CrossRef][Medline]
8 Zurrida S, Bartoli C, Galimberti V, Squicciarini P, Delledonne V, Veronesi P, et al. Which therapy for unexpected phyllode tumour of the breast? Eur J Cancer (1992) 28:654–7.[CrossRef][Medline]
9 Chen WH, Cheng SP, Tzen CY, Yang TL, Jeng KS, Liu CL, et al. Surgical treatment of phyllodes tumors of the breast: retrospective review of 172 cases. J Surg Oncol (2005) 91:185–94.[CrossRef][Web of Science][Medline]
10 Chaney AW, Pollack A, McNeese MD, Zagars GK, Pisters PW, Pollock RE, et al. Primary treatment of cystosarcoma phyllodes of the breast. Cancer (2000) 89:1502–11.[CrossRef][Web of Science][Medline]
11 Ben Hassouna J, Damak T, Gamoudi A, Chargui R, Khomsi F, Mahjoub S, et al. Phyllodes tumors of the breast: a case series of 106 patients. Am J Surg (2006) 192:141–7.[CrossRef][Web of Science][Medline]
12 Yilmaz E, Sal S, Lebe B. Differentiation of phyllodes tumors versus fibroadenomas. Acta Radiol (2002) 43:34–9.[Web of Science][Medline]
13 Liberman L, Bonaccio E, Hamele-Bena D, Abramson AF, Cohen MA, Dershaw DD. Benign and malignant phyllodes tumors: mammographic and sonographic findings. Radiology (1996) 198:121–4.
14 Chao TC, Lo YF, Chen SC, Chen MF. Sonographic features of phyllodes tumors of the breast. Ultrasound Obstet Gynecol (2002) 20:64–71.[CrossRef][Web of Science][Medline]
15 Krishnamurthy S, Ashfaq R, Shin HJ, Sneige N. Distinction of phyllodes tumor from fibroadenoma: a reappraisal of an old problem. Cancer (2000) 90:342–9.[CrossRef][Web of Science][Medline]
16 Ng WK. Fine needle aspiration cytology of fibroadenoma with multinucleated stromal giant cells. A review of cases in a six-year period. Acta Cytol (2002) 46:535–9.[Web of Science][Medline]
17 Shet T, Rege J. Cystic degeneration in phyllodes tumor. A source of error in cytologic interpretation. Acta Cytol (2000) 44:163–8.[Web of Science][Medline]
18 Hawkins RE, Schofield JB, Fisher C, Wiltshaw E, McKinna JA. The clinical and histologic criteria that predict metastases from cystosarcoma phyllodes. Cancer (1992) 69:141–7.[CrossRef][Web of Science][Medline]
19 Grimes MM. Cystosarcoma phyllodes of the breast: histologic features, flow cytometric analysis, and clinical correlations. Mod Pathol (1992) 5:232–9.[Web of Science][Medline]
20 Treves N, Sunderland DA. Cystosarcoma phyllodes of the breast: a malignant and a benign tumor; a clinicopathological study of seventy-seven cases. Cancer (1951) 4:1286–332.[CrossRef][Web of Science][Medline]
21 Norris HJ, Taylor HB. Relationship of histologic features to behavior of cystosarcoma phyllodes. Analysis of ninety-four cases. Cancer (1967) 20:2090–9.[CrossRef][Web of Science][Medline]
22 Ward RM, Evans HL. Cystosarcoma phyllodes. A clinicopathologic study of 26 cases. Cancer (1986) 58:2282–9.[CrossRef][Web of Science][Medline]
23 Inoshita S. Phyllodes tumor (cystosarcoma phyllodes) of the breast. A clinicopathologic study of 45 cases. Acta Pathol Jpn (1988) 38:21–33.[Medline]
24 Murad TM, Hines JR, Beal J, Bauer K. Histopathological and clinical correlations of cystosarcoma phyllodes. Arch Pathol Lab Med (1988) 112:752–6.[Web of Science][Medline]
25 Kario K, Maeda S, Mizuno Y, Makino Y, Tankawa H, Kitazawa S. Phyllodes tumor of the breast: a clinicopathologic study of 34 cases. J Surg Oncol (1990) 45:46–51.[Web of Science][Medline]
26 Dyer NH, Bridger JE, Taylor RS. Cystosarcoma phylloides. Br J Surg (1966) 53:450–5.[CrossRef][Web of Science][Medline]
27 Hajdu SI, Espinosa MH, Robbins GF. Recurrent cystosarcoma phyllodes: a clinicopathologic study of 32 cases. Cancer (1976) 38:1402–6.[CrossRef][Web of Science][Medline]
28 Chua CL, Thomas A, Ng BK. Cystosarcoma phyllodes: a review of surgical options. Surgery (1989) 105:141–7.[Web of Science][Medline]
29 Eich PD, Diederich S, Eich HT, Micke O, Wagner W. Diagnostic radiation oncology: malignant cystosarcoma phylloides. Strahlenther Onkol (2000) 176:192–5.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||