Tumour histological grade may progress between primary and recurrent invasive mammary carcinoma. (Original Article).

Aims: To assess the constancy of the histological grade of invasive breast carcinomas by comparing primary tumours with their axillary metastases and local or regional recurrences.

Methods: Eighty four recurrent invasive breast carcinomas with a primary tumour or previous recurrence were available for histological review from the period 1980 to 2000. These and any further recurrences were graded by one observer.

Results: Nine, 24, and 51 tumours with grades 1, 2, and 3, respectively, recurred. Grade 1, 2, and 3 tumours recurred within a median time of 88, 42, and 23 months, respectively. The intraobserver reproducibility of the histological grade was goad ([kappa] = 0.66], and the grades of the primary tumours and their axillary metastases or next recurrence also exhibited good agreement. However, when further (second to sixth) recurrences were included in the analysis, the agreement between the grade of the tumours and their last recurrence was only moderate ([kappa] = 0.48). Only two of the nine grade 1 and 15 of the 24 grade 2 tumours retained their grade in their last recurrence.

Conclusions: Low grade carcinomas require a longer follow up. These long term data support the possibility of a transition from low grade invasive breast carcinomas to higher grade tumours. It is suggested that low grade (well differentiated) breast carcinomas are not a single entity: some may progress to high grade tumours, whereas others appear not to progress.

**********

Although several steps of carcinogenesis have already been explored, the development of malignant tumours is not fully understood. The currently available data demonstrate that the development of malignant tumours requires the presence of several genetic lesions and/or epigenetic factors, and their development involves progression. In breast cancer, the morphologically identifiable steps of this progression include atypical ductal hyperplasia (with a fivefold increase in relative risk of breast cancer), (1) ductal carcinoma in situ (DCIS) (a lesion considered an obligate precursor of invasive cancer, with at least a 10 fold increase in relative risk of the development of the latter), (1) and invasive carcinoma. Earlier steps may include typical ductal hyperplasia, especially its florid variant, (2) which does not usually progress to cancer, or certain histologically unidentifiable lesions, which already carry genetic predisposing lesions, as suggested by loss of heterozygosity (LOH) studies. (3) Great care mu st be taken in the interpretation of these results because normal epithelium of the breast also seems monoclonal by LOH analysis. (4) At the other end of the spectrum, there are also unanswered questions relating to the progression of invasive cancers.

The systemic theory formulated by Fisher postulates that invasive breast cancer is a systemic disease from its beginning. (5) However, the opposing locoregional theory put forward earlier by Halsted (6) also has some truth. Both the mortality decrease resulting from early detection as a consequence of breast cancer screening, (7 8) and the relatively high proportion of patients with breast cancer cured with locoregional treatment alone, (9-11) favour the spectrum theory (12) and the progressive nature of breast cancer. It is well accepted that breast cancer is not a single disease; as an example, well differentiated carcinomas have better outcomes than poorly differentiated carcinomas of the same size and nodal status. It remains unclear whether high grade (poorly differentiated) carcinomas develop from low grade carcinomas, as suggested by some studies, (13-15) or are derived directly from high grade DCIS, as suggested by others. (16-18) In our present study, invasive carcinomas with subsequent recurrences w ere studied with respect to their histological grade, to evaluate the constancy of this prognostic parameter.

METHODS

Locally or regionally recurrent invasive breast tumour specimens assessed between January 1980 and December 2000 were identified from the hard copy (1980-1997) and electronic (1998-2000) files at the department of pathology of the Bacs-Kiskun County Teaching Hospital. The basis of the analysis was formed by those recurrent invasive carcinomas for which either the primary tumour or (in the absence of a primary tumour assessed between 1980 and 2000) a previous recurrence was available for histopathological review from the same period. Only ipsilateral recurrences were taken into consideration. Several of the patients had multiple metachronous recurrences. Recurrences were ranked according to the time of their appearance. For the few cases with simultaneous local and axillary recurrence, the local recurrence was thought to precede the regional …