TY - JOUR
T1 - A new pathological system for grading DCIS with improved prediction of local recurrence: results from the UKCCCR/ANZ DCIS trial
AU - Pinder, S. E.
AU - Duggan, C.
AU - Ellis, I. O.
AU - Cuzick, J.
AU - Forbes, J. F.
AU - Bishop, H.
AU - Fentiman, I. S.
AU - George, W. D.
PY - 2010/6/29
Y1 - 2010/6/29
N2 - BACKGROUND: There is no consensus agreement regarding optimal management of locally excised ductal carcinoma in situ (DCIS) or features of greatest assistance in predicting disease behaviour. Cases in the UKCCCR/ANZ DCIS trial have been histologically reviewed to determine the features of prognostic importance.
METHOD: A total of 72% of 1694 cases entered into the UKCCCR/ANZ DCIS trial had full pathological review. A large number of histological features were assessed, blinded to outcome and compared regarding ability to predict ipsilateral recurrence, as either DCIS or progression to invasive carcinoma.
RESULTS: Pathological features associated with ipsilateral recurrence in univariate analysis included high cytonuclear grade, larger lesion size, growth pattern, presence of necrosis or chronic inflammation, incompleteness (or uncertainty of completeness) of excision and smaller margin width. Receipt of post-operative radiotherapy was also a strong prognostic factor. We report a novel sub-division of the large group of high-grade lesions, which enables identification of a very poor prognosis subgroup; namely, DCIS that is of high cytonuclear grade, predominantly (>50%) solid architecture, bearing extensive comedo-type necrosis (>50% of ducts). In addition, we found little difference in ipsilateral recurrence rates between low-and intermediate-grade groups. Hazard ratios for low, intermediate, high and the new, very high, grade were 0.42, 0.33, 0.62 and 1.00, respectively, for ipsilateral in situ or invasive recurrence.
CONCLUSION: We present a novel pathological classification for DCIS with substantially better prognostic discrimination for ipsilateral recurrence than the classical categorisation based on cytonuclear grade alone. British Journal of Cancer (2010) 103, 94-100. doi:10.1038/sj.bjc.6605718 www.bjcancer.com Published online 1 June 2010 (C) 2010 Cancer Research UK
AB - BACKGROUND: There is no consensus agreement regarding optimal management of locally excised ductal carcinoma in situ (DCIS) or features of greatest assistance in predicting disease behaviour. Cases in the UKCCCR/ANZ DCIS trial have been histologically reviewed to determine the features of prognostic importance.
METHOD: A total of 72% of 1694 cases entered into the UKCCCR/ANZ DCIS trial had full pathological review. A large number of histological features were assessed, blinded to outcome and compared regarding ability to predict ipsilateral recurrence, as either DCIS or progression to invasive carcinoma.
RESULTS: Pathological features associated with ipsilateral recurrence in univariate analysis included high cytonuclear grade, larger lesion size, growth pattern, presence of necrosis or chronic inflammation, incompleteness (or uncertainty of completeness) of excision and smaller margin width. Receipt of post-operative radiotherapy was also a strong prognostic factor. We report a novel sub-division of the large group of high-grade lesions, which enables identification of a very poor prognosis subgroup; namely, DCIS that is of high cytonuclear grade, predominantly (>50%) solid architecture, bearing extensive comedo-type necrosis (>50% of ducts). In addition, we found little difference in ipsilateral recurrence rates between low-and intermediate-grade groups. Hazard ratios for low, intermediate, high and the new, very high, grade were 0.42, 0.33, 0.62 and 1.00, respectively, for ipsilateral in situ or invasive recurrence.
CONCLUSION: We present a novel pathological classification for DCIS with substantially better prognostic discrimination for ipsilateral recurrence than the classical categorisation based on cytonuclear grade alone. British Journal of Cancer (2010) 103, 94-100. doi:10.1038/sj.bjc.6605718 www.bjcancer.com Published online 1 June 2010 (C) 2010 Cancer Research UK
U2 - 10.1038/sj.bjc.6605718
DO - 10.1038/sj.bjc.6605718
M3 - Article
SN - 1532-1827
VL - 103
SP - 94
EP - 100
JO - BJC: British Journal of Cancer
JF - BJC: British Journal of Cancer
IS - 1
ER -
