In a Canadian study reported in the Journal of Clinical Oncology, Hahn et al evaluated the impact of certain predictive models for local and invasive local recurrence in patients undergoing breast-conserving surgery with or without adjuvant radiotherapy for pure ductal carcinoma in situ (DCIS).
Study Details
The study used data from the Ontario DCIS cohort on patients with pure DCIS treated with breast-conserving surgery with or without adjuvant radiotherapy between 1994 and 2003. The models for predicting 10-year local or invasive local ipsilateral recurrence evaluated in the analysis were clinicopathologic predictive factors (CPF) alone or with the 12-gene DCIS Score (DS) or the 21-gene Recurrence Score (RS). Model accuracy was assessed by C-statistic, Akaike information criterion (AIC; lower = better), and -2log likelihood estimate (-2LogL; lower = better).
Key Findings
The cohort included 1,226 women treated with breast-conserving surgery, with 712 (58%) receiving radiotherapy. Ipsilateral local recurrence as a first event occurred in 194 women (15.8%), while invasive recurrence occurred in 112 (9.1%).
For instances of local recurrence, the two molecular-based models—DS plus CPF (C-statistic = 0.70, AIC = 2628.27, -2LogL = 2608.27) and RS plus CPF (C-statistic = 0.70, AIC = 2628.83, -2LogL = 2610.83)—performed better in predicting 10-year risk vs the CPF model alone (C-statistic = 0.66, AIC = 3570.06, -2LogL = 3548.06).
For instances of invasive local recurrence, DS plus CPF (C-statistic = 0.68, AIC = 1670.05, -2LogL = 1650.05) and RS plus CPF (C-statistic = 0.69, AIC = 1675.16, -2LogL = 1651.16) performed better in predicting 10-year risk vs the CPF model alone (C-statistic = 0.68, AIC = 1680.59, -2LogL = 1654.59.); the RS plus CPF model did not exhibit better performance than the DS plus CPF model.
The investigators concluded, “Models incorporating the DS or RS more accurately predicted the 10-year risk of local recurrence and invasive local recurrence after breast-conserving surgery compared with models on the basis of CPF alone. Inclusion of the RS, compared with DS, did not improve the prediction of the 10-year risk of invasive local recurrence.”
Eileen Rakovitch, MD, FRCPC, of the Department of Radiation Oncology, Sunnybrook Health Sciences Centre, Toronto, is the corresponding author for the Journal of Clinical Oncology article.
Disclosure: The study was supported by ICES (Institute for Clinical Evaluative Sciences), which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. For full disclosures of the study authors, visit ascopubs.org.
