Fibroepithelial lesions Fibroepithelial lesions may give rise to a number of diagnostic problems but the most important concern phyllodes tumour. At the benign end of the diagnostic spectrum difficulty is occasionally encountered in distinguishing a cellular fibroadenoma from a benign phyllodes tumour. At the malignant end the criteria for the separation of benign and malignant phyllodes tumour are often misinterpreted with consequent overdiagnosis of malignant phyllodes tumour. Classically fibroadenomas are usually encountered in women under the age of 30 years whilst phyllodes tumours are seen much more frequently in middle aged or older women. However, there is considerable overlap between the two entities at both ends of the age range and a significant minority of phyllodes tumours occurs in adolescents and young women. 31 Size alone is of no value as a distinguishing feature; although most fibroadenomas are small, measuring less then 20 mm in diameter and phyllodes tumours tend to be larger than this, there is marked variation in the size of both lesions. We have certainly encountered a number of phyllodes tumours measuring no more than 10 mm in diameter. Nevertheless, most problems are encountered with relatively large fibroadenomas in younger women when the presence of an intracanalicular pattern with leaf-like foci and cellular stroma raise the possibility of benign phyllodes tumour. Relative uniformity of stromal nuclei and a lack of mitoses favour a diagnosis of fibroadenoma. Further blocks should always be examined but it may be very difficult to make a clear distinction in some cases. It may be prudent to issue a cautious report in such cases, especially if the lesion extends to resection margins, since risk of local recurrence in phyllodes tumour is related to incomplete excision. In the past phyllodes tumour gained an unwarranted aggressive reputation largely based on data from tertiary referral centres which, by their very nature, tend to acquire the more difficult cases for treatment. In addition there has been a general lack of agreement on the morphological criteria which should be used to designate malignancy in an individual tumour. As a result some pathologists, in an understandable attempt to avoid under-diagnosis, are inclined to err on the malignant side in assessing the histological appearances. It is now accepted that the majority of phyllodes tumours are, in fact, entirely benign; in our own community based study in Nottingham two thirds of the cases were classified as benign on morphological grounds and only 16 per cent had malignant features. 32 We therefore prefer to use the term 'phyllodes tumour' rather than cystosarcoma phyllodes and believe that the latter, with its innate implication of malignancy, should be abandoned. Based on semiquantitative criteria 33, 34 we divide phyllodes tumour into three categories, benign, borderline and malignant. 32 Benign tumours have a pushing margin, minimal stromal overgrowth, cellularity and nuclear pleomorphism with stromal mitotic counts less than 10 per 10 fields (field diameter 0.152 mm2). In malignant lesions the margin is infiltrative, there is marked stromal overgrowth, cellularity and nuclear pleomorphism and mitotic counts are greater than 10 per 10 fields. Cases are placed in the borderline category if they fulfil some, but not all, of the criteria for malignancy, eg, a pushing margin, but moderate nuclear pleomorphism and an intermediate mitotic count. Interestingly we found that none of these morphological features was useful in predicting local recurrence, which was strongly related to completeness of local excision. Metastasis is very uncommon in phyllodes tumour and only occurs in tumours with a malignant phenotype. It is our policy to excise benign phyllodes tumours with a clear margin of at least 10 mm; re-excision may be appropriate in some cases if completeness of excision is not achieved at the first operation. Mastectomy is advisable for patients with malignant phyllodes tumours and is also offered to women with very large benign lesions to avoid a poor cosmetic result. The borderline category is useful in preventing over-diagnosis of malignancy and therefore potential over-treatment. Optimum local therapy depends on the size of the tumour. Local excision is imperative, but only if this cannot be achieved with a satisfactory cosmetic result need mastectomy be considered. |
Copyright 2000, The Author(s) and/or The Publisher(s)
| Organisation: FORPATH asbl  |
Coordination: Dr Bernard Van den Heule |
Host: Labo CMP |