Long-term outcomes of ductal carcinoma in situ of the breast: a systematic review, meta-analysis and meta-regression analysis

BACKGROUND: To summarize data on long-term ipsilateral local recurrence (LR) and breast cancer death rate (BCDR) for patients with ductal carcinoma in situ (DCIS) who received different treatments. METHODS: Systematic review and study-level meta-analysis of prospective (n = 5) and retrospective (n = 21) studies of patients with pure DCIS and with median or mean follow-up time of ≥10 years. Meta-regression was performed to assess and adjust for effects of potential confounders – the average age of women, period of initial treatment, and of bias – follow-up duration on recurrence- and death-rates in each treatment group. LR and BCDR rates by local treatment used were reported. Outside of randomized trials, remaining studies were likely to have tailored patient treatment according to the clinical situation. RESULTS: Nine thousand four hundred and four DCIS cases in 9391 patients with 10-year follow-up were included. The adjusted meta-regression LR rate for mastectomy was 2.6 % (95 % CI, 0.8–4.5); breast-conserving surgery with radiotherapy (RT), 13.6 % (95 % CI, 9.8–17.4); breast-conserving surgery without RT, 25.5 % (95 % CI, 18.1–32.9); and biopsy-only (residual predominately low-grade DCIS following inadequate excision), 27.8 % (95 % CI, 8.4–47.1). RT + tamoxifen (TAM) in conservation surgery (CS) patients resulted in lower LR compared to one or no adjuvant treatments: LR rate for CS + RT + TAM, 9.7 %; CS + RT(no TAM), 14.1 %; CS + TAM(no RT), 24.7 %; CS(alone), 25.1 % (linear trend for treatment P < 0.0001). Compared to CS + RT + TAM, a significantly higher invasive LR was observed for CS(alone), odds ratio (OR) 2.61 (P < 0.0001); CS + TAM(no RT), OR 2.52 (P = 0.001); CS + RT(no TAM), OR 1.59 (P = 0.022). BCDR was similar for mastectomy, breast-conserving surgery with or without RT (1.3–2.0 %) and non-significantly higher for biopsy-only (2.7 %). Additionally, the 15-year follow-up was reported where all like-studies had ≥ 15-year data sets; the biopsy-only patients had a meta-analysed total LR rate of 40.2 % and the invasive LR rate was 28.1 %. The biopsy-only patients had a ≥ 15-year BCDR (that included women with metastatic disease) of 17.9 %; the ≥ 15-year BCDR was 55.2 % for those with invasive LR. CONCLUSIONS: More local intervention was associated with greater local control for patients with DCIS at long-term follow-up. For patients undergoing breast-conservation, invasive LR was significantly lower when two rather than one adjuvant treatment modalities were given. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12885-015-1904-7) contains supplementary material, which is available to authorized users.