A three-week course of locoregional radiotherapy did not increase lymphoedema or compromise recurrence outcomes compared with the conventional five-week regimen in a large phase III breast cancer trial.
Women with high-risk breast cancer could safely spend two fewer weeks undergoing locoregional radiotherapy, according to results from a large international randomised trial.
The Danish Breast Cancer Group Skagen Trial 1 found that moderately hypofractionated radiotherapy delivered as 40Gy in 15 fractions over three weeks was non-inferior to the conventional 50Gy in 25 fractions over five weeks for the primary endpoint of arm lymphoedema.
The phase 3 trial included 2908 women with early, high-risk breast cancer who required locoregional radiotherapy after lumpectomy or mastectomy. Participants were randomly assigned to 50Gy in 25 fractions or 40Gy in 15 fractions. The median age was 57 years.
Results were published in the Journal of Clinical Oncology.
At three years, lymphoedema was reported in 9.4% of women receiving conventional fractionation and 8.0% receiving hypofractionation, comfortably meeting the trial’s prespecified non-inferiority criterion. The odds ratio was 0.84 (95% CI 0.62-1.14). The cumulative incidence analysis also supported non-inferiority, with a five-year hazard ratio for lymphoedema of 0.99 (95% CI 0.82-1.19).
The researchers noted that the shorter regimen did not appear to add risk among women undergoing more extensive axillary surgery.
Among the 2122 women who underwent axillary lymph node dissection, three-year lymphoedema rates were 12.3% with 50Gy and 10.1% with 40Gy. Among the 713 women treated with sentinel node biopsy alone, the corresponding rates were 2.1% and 2.4%.
Other normal-tissue effects were generally similar or less severe with hypofractionation. At five years, telangiectasia was reported in 16% of patients receiving 40Gy compared with 22% receiving 50Gy, while overall patient satisfaction was 77% and 71%, respectively.
Cancer control was also similar between the groups. There were 78 locoregional recurrences overall, including 40 in the conventional-fractionation group and 38 in the hypofractionated group. This produced an eight-year hazard ratio of 0.96 (95% CI 0.62-1.51).
Distant recurrence occurred in 11.0% of the 50Gy group and 12.0% of the 40Gy group at five years, with an eight-year hazard ratio of 1.10 (95% CI 0.89-1.37).
The researchers observed a higher five-year estimate of breast cancer mortality with hypofractionation (7.2% compared with 5.1% for conventional treatment, an absolute difference of 2.1 percentage points).
However, they said the difference was not significant over the longer eight-year analysis, with a hazard ratio of 1.25 (95% CI 0.93-1.66), prompting an unplanned exploratory subgroup analysis.
There was also no significant difference in all-cause mortality, with an eight-year hazard ratio of 1.08 (95% CI 0.85-1.36).
The researchers said their findings add substantially to the evidence for shortening radiotherapy in higher-risk patients, a group in whom hypofractionation has been adopted more cautiously because of concerns about morbidity associated with regional nodal irradiation. Previous evidence for shorter schedules had been stronger in women receiving breast-only radiotherapy.
The researchers said the results confirmed that moderately hypofractionated locoregional radiotherapy did not increase lymphoedema during the first five years, including among women treated with sentinel node biopsy or axillary dissection.
An accompanying Journal of Clinical Oncology editorial said the findings provided “level one evidence” supporting the safety and effectiveness of moderate hypofractionation for regional node irradiation.
Authors, Canadian radiation oncologist and researcher Dr Elysia Donovan and Canadian oncologist and researcher Professor Timothy Whelan, noted that clinicians had been more reluctant to use hypofractionation for regional nodes than for whole-breast irradiation because of concerns about late toxicities, particularly lymphoedema and brachial neuropathy.
They said the latest findings, together with results from the French HypoG-01 trial (published in The Lancet), provided strong evidence supporting moderate hypofractionation in this setting.
Dr Donovan and Professor Whelan highlighted the trial’s international participation, inclusion of patients with locally advanced disease treated with neoadjuvant chemotherapy, use of modern radiotherapy techniques, and rigorous quality assurance as factors supporting the generalisability of its findings.
“A strength of the Danish trial is that it compared risks of lymphedema using multiple methods: prevalence rates, odds ratios, and cumulative risks,” they wrote.
“An important distinction observed in the trial is that cumulative risk of lymphedema (16.5% at three years) may overestimate cross-sectional risk (8.9% at three years) as lymphedema in some instances may be temporary and recede over time.
“This may, in part, explain the differences in lymphedema rates reported at three years by the Danish and French trials.
“The Danish trial also confirmed several other important observations. It showed that most patients (80%) who developed lymphedema did it by three years.
“It also observed that the incidence of lymphedema was much less after sentinel lymph node biopsy compared with axillary dissection (on average 2.3% v 11.2%, respectively) highlighting that with modern surgery and radiotherapy, the risk of lymphedema has been remarkably reduced, which is important to communicate to patients with breast cancer receiving locoregional treatment.”
Related
The authors also addressed an “anomaly observed in the Danish trial”, that the five-year rate of breast cancer mortality appeared to be increased in patients treated with hypofractionation.
“However, no difference between groups was observed for the hazard ratio of breast cancer mortality over the whole eight years of follow-up,” they wrote.
“This finding is also not consistent with the observation in the trial of no significant differences in locoregional or distant recurrence.
“Of note, in the similar French trial, moderate hypofractionation was associated with an increase in distant disease-free and breast cancer–specific survival, suggesting that the difference in breast cancer mortality observed in the Danish trial may be due to challenges with attribution of the cause of death or random error.”
They said questions remained, particularly around women undergoing breast reconstruction after mastectomy and whether ultra-hypofractionation could reduce regional node treatment further, from three weeks to one. Early evidence was encouraging, but trials examining ultra-hypofractionation that includes the internal mammary nodes were ongoing.
For now, Dr Donovan and Professor Whelan said the trials supported moderate hypofractionation for regional node irradiation, with the shorter approach reducing the treatment burden for patients, improving convenience and lowering healthcare costs.
“These findings will provide confirmation for centres that have already adopted moderate hypofractionation and should lead to change in practice in centres that are still using conventional fractionation,” they concluded.
Journal of Clinical Oncology, July 2026 (research)
Journal of Clinical Oncology, July 2026 (editorial)
