Background There is uncertainty as to which lymph node regions should be irradiated following breast cancer surgery. Systematic review of radiation dosimetry indicates that in randomised trials of nodal radiation therapy (RT) versus not, radiation delivery was qualitatively better in modern trials compared to older trials.

Methods We undertook an individual patient data meta–analysis of randomised trials assessing the benefits and risks of RT to different lymph node regions including the axilla, supraclavicular fossa (SCF) and internal mammary chain (IMC). Eligible studies started before 2009, and included a randomisation, or pseudo–randomisation (by left–versus–right sided tumours), in which the only difference between treatment groups was the use, or extent, of nodal irradiation. Surgery/RT to the breast was the same in both arms. Analyses used standard log–rank methods, and were stratified by study, age, nodal status and year of follow–up. – Studies were categorised according to estimated mean heart dose in the nodal RT arm and whether regimens were likely to have delivered ≥85% of prescribed dose to target nodal regions.

Results Information was available on 13,132 women in 14 comparisons of nodal RT versus not. There were 3260 recurrences, 2545 deaths from breast cancer and 4147 deaths overall. Eight trials starting 1961–1978, with median follow–up 9.2 (interquartile [IQR] range 3.4–17.5) years, had estimated >8 Gy mean heart dose and likely nodal dose <85% in the nodal RT arm. In these older trials, including 2178 women, nodal RT had no effect on breast cancer recurrence [Rate ratio (RR)=0.98, 95% CI 0.85–1.13, p=0.83] or breast cancer mortality (RR=1.05, 0.91–1.21, p=0.54), but increased non–breast cancer mortality (RR=1.44, 1.20–1.73, p<0.0001), leading to a net increase in any death (RR=1.18, 1.06–1.32, p=0.004).

Six studies starting 1989–2003, with a mean follow–up 9.1 [IQR 7.0–11.0] years, had likely nodal dose ≥85%, and estimated mean heart dose <8 Gy in the nodal RT arm. In these more recent studies, including 10,954 women, nodal RT reduced breast cancer recurrence (RR=0.86, 95% CI 0.79–0.94, p=0.0006), breast cancer mortality (RR=0.81, 0.74–0.90, p<0.0001) and overall mortality (RR=0.86, 0.80–0.93, p=0.0002). No excess of non–breast cancer mortality was apparent (RR=0.96, 0.79–1.18, p=0.71). Recurrence rate ratios did not vary significantly according to nodal region(s) irradiated (axilla/SCF/IMC), or the use of adjuvant chemotherapy.

Conclusions RT to regional lymph nodes in older (1961–78) studies increased the overall risk of death, probably explained by radiation exposure of the lungs and heart. Nodal RT in more recent (1989–2003) studies reduced breast cancer recurrence, breast cancer mortality and overall mortality without increasing non–breast cancer mortality. The proportional benefits from today's RT may be larger. Absolute benefits for individual women will depend on their absolute recurrence and breast cancer mortality risks.

Citation Format: Dodwell D, Taylor C, McGale P, Coles C, Duane F, Gray R, Kühn T, Hennequin C, Oliveros S, Wang Y, Overgaard J, Poortmans P, Whelan T. Regional lymph node irradiation in early stage breast cancer: An EBCTCG meta-analysis of 13,000 women in 14 trials [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr GS4-02.