Efficacy
The DCIS Cochrane review by Goodwin et al. included 4 randomised control trials: NSABP B-17, EORTC10853, UKCCCR and SweDCIS.r The review compared the addition of radiation therapy to breast conserving surgery (BCS) (n = 3925 women). They reported a statistically significant benefit from the addition of radiation therapy on all ipsilateral breast events (HR 0.49, 95% CI: 0.41-0.59, p <0.00001) and ipsilateral recurrence (HR 0.64, 95% CI: 0.41 to 1.01, p = 0.05).r The addition of radiation therapy reduced the risk of recurrence of either DCIS or invasive cancer in the treated breast by 51%.
The 15-year combined analysis of the NSABP B-17 and B-24 trials showed the reduction in risk of local recurrence appeared to persist in the long term.r The ipsilateral breast recurrence rate was reduced from 19.4% to 8.9% with radiation therapy at 15 years. The overall survival (OS) rate was similar, 83% versus 84%.
Further evidence supporting this protocol was provided by a second meta-analysis by Correa et al.r This meta-analysis reviewed individual patient data for the same 4 trials (n = 3925 women) and also compared BCS alone to BCS with adjuvant radiation therapy. The 10-year ipsilateral breast recurrence rate with surgery alone was statistically significant, 28% compared to 13% for BCS with radiation therapy. This benefit was greater in women >50 years old (28% versus 11%) compared with women <50 years old (29% versus 18%). Women with small, low-grade tumours and negative surgical margins still benefited (30% versus 12%, statistically significant). There was no effect on breast cancer specific mortality (~4%), other-cause mortality (~5%), and all-cause mortality (~8%).
A meta-analysis conducted by Stuart et al. included 9391 patients with follow-up at 10 years reported adjuvant radiation therapy after BCS was associated with greater local control when compared to patients who were managed with BCS and biopsy alone.r At 10 years, OS was similar in patients treated with mastectomy or breast conserving surgery with or without radiation therapy.
Bijker et al. reported the inclusion of radiation therapy resulted in a 48% reduction in the 10-year local recurrence (LR) rate in DCIS (14% with BSC versus 7% with BSC and radiation therapy, statistically significant). Factors that increased the risk for LR were age <40 years old, grade 2 or 3, cribriform or solid growth pattern, doubtful margin, and local excision alone. Size was not a prognostic factor. There was no difference reported in OS or distant metastases.r
In 20-year follow-up the sweDCIS trial showed the absolute risk reduction of ipsilateral breast recurrence in the radiation therapy arm was 12.0% at 20 years (95% CI: 6.5-17.7), with a relative risk reduction of 37.5%.r Absolute reduction was 10.0% (95% CI: 6.0-14.0) for DCIS. The cumulative incidence of invasive ipsilateral recurrences was found to continuously rise over the 20-year period before plateauing.
Role of hypofractionation and boost
The role of hypofractionated radiation therapy and breast boost are examined in a meta-analysis of observational trials published in 2015.r No difference in the risk of local recurrence was observed between the patients that received boost versus no boost in the general cohort (12 studies, 6943 patients; Odds Ratio [OR]: 0.91, 95% CI: 0.77-1.08, very low level evidence). There was a reduced risk for local recurrence when a boost was administered in patients with positive margins compared to no boost (6 studies, 811 patients; OR: 0.56, 95% CI: 0.36-0.87, very low level evidence). No difference in local recurrence rate between patients who received hypofractionated versus standard radiation therapy was observed (4 studies, 2534 patients; OR: 0.78, 95% CI: 0.58-1.03, low level evidence). These studies provide a low level of evidence.
Moran et al. examined the role of breast boost in a retrospective analysis of 4131 patients from 10 institutions.r The delivery of a breast boost was associated with lower ipsilateral breast tumour recurrence (IBTR) (HR 0.73) and IBTR-free survival at 5, 10 and 15 years. On multivariate analysis, Moran et al. found grade, presence of necrosis, margin status, age (<50 or ≥50 years), tumour size and a lack of tamoxifen use to be significantly contributory to IBTR.
Data presented in abstract form from the TROG 07.01 trial reported in patients with non-low risk DCIS treated with BCS, the addition of tumour bed boost following conventional or hypofractionated whole breast irradiation reduced local recurrence rates.rThe 5-year free from local recurrence rates were 93% in the no boost group and 97% in the boost group (HR 0.47; 95% CI: 0.31-0.72, p <0.001). No significant differences were observed in the 5-year free from local recurrence rates between the conventionally fractionated and the hypofractionated groups.
Optimal margins
Although it is widely accepted that positive margins definitively increase the risk of locoregional recurrence (LRR) in patients who undergo BCS even with adjuvant radiation therapy, the exact optimal margin threshold is not yet defined at an international level.rrr
ASTRO guidelines and other eminent groups recommend a >2 mm margin.rrrrr Other groups infer smaller margins may be acceptable in selected cases.rrr
Tadros et al. recently published a retrospective analysis of a cohort of 1491 ‘contemporary’ patients with DCIS.r They reported no statistically significant difference in locoregional control between patients with <2 mm and ≥2 mm negative margins who underwent adjuvant radiation therapy. For patients who did not undergo adjuvant radiation therapy however, those with margins <2 mm were significantly more likely to develop LRR than those with margins ≥2 mm.
Van Zee et al. echoed the above in their retrospective analysis of 2996 patients.r They reported no significant association between negative margin width and risk of LRR for patients who received adjuvant radiation therapy, however, in patients with smaller negative margins who did not receive radiation therapy there was a significant risk of LRR.
The 2016 ASTRO guidelines noted that clinical judgement must be used when deciding on the need for re-excision in patients with >0-1 mm margins. Factors to consider include residual calcifications on post-excision mammography, extent of DCIS in proximity to margin, which margin is close (superficial/deep versus radial), cosmetic impact of re-excision, and overall life expectancy.r
The reference committee consensus is that clear margins ≥2 mm is optimal, however; narrower margins may be acceptable in selected cases.
Low-risk subgroups
Omission of radiation therapy may be considered in some low-risk patients such as those who were eligible for RTOG 9804 Good-Risk DCIS trial.r However, with longer follow-up (median 13.9 years), a steady increase of 1% per year of IBR was observed up to at least 15 years.rIn this population, the use of radiation therapy reduced IBR with a HR of 0.36.
The RTOG 9804 eligibility criteria included:
- mammogram detected DCIS or incidental finding of DCIS in tissue of an otherwise benign biopsy
- unicentric disease
- low or intermediate nuclear grade
- size <2.5 cm
- margin ≥3 mm to ink
- negative post-excision mammogram.
Although the benefit of the addition of radiation therapy has been demonstrated in all sub-groups of patients, absolute benefit in these patients is small (1% risk of recurrence per year is halved by the addition of radiation therapy).rThe known and increasing long-term risk of a local recurrence must be considered in the context of a patient's longevity and wellbeing.
Consider the use of a DCIS recurrence nomogram to assist in decision making.r
Chemoprevention
COSA provides a clinician guide for medications to lower the risk of breast cancer.