Long-Term Outcomes After INTRABEAM IORT and ASTRO 2024 Risk Stratification

Bratt and colleagues described long-term, real-world follow-up after single-session intraoperative radiotherapy delivered with the Zeiss INTRABEAM IORT system in a multicenter retrospective report of women with early-stage invasive breast cancer treated during breast-conserving surgery.

Alongside mature local control and survival outcomes, the investigators analyzed recurrence patterns using a 2024 guideline-based risk framework to compare outcomes across lower- versus higher-risk groups. The report focuses on a single-fraction IORT approach delivered at the time of lumpectomy and summarizes overall and risk-stratified findings over extended follow-up.

The authors reported a multicenter retrospective cohort comprising 358 IORT-treated breasts in 356 patients treated between 2014 and 2018, with IORT prescribed as a single 20 Gy dose to the applicator surface and delivered intraoperatively. Median follow-up was 7.1 years. Local recurrence was observed in 14 cases (3.9%), with a median time to local recurrence of 5.2 years. Kaplan–Meier estimates for local recurrence-free survival were 98.3% at 5 years and 94.8% at 8 years; overall survival was 99.4% and 97.7% at 5 and 8 years, respectively; and the 8-year mastectomy-free survival estimate was 98.2%. The investigators also noted that more than half of local recurrences occurred beyond 5 years, highlighting recurrence timing within the long-term follow-up profile.

To assess whether guideline-defined risk categories differentiated outcomes, the investigators applied the ASTRO 2024 partial-breast irradiation risk classification (i.e., the guideline’s candidate-category stratification) to classifiable cases (n = 341). They reported 220 Suitable, 71 Conditional (pooling “conditionally recommended” and “conditionally not recommended”), and 50 Unsuitable, with 17 unclassified due to missing data. In a Cox model comparing Conditional versus Suitable groups, the authors reported a higher hazard for local recurrence in the Conditional group (HR 3.25, 95% CI 1.05–10.08; p = 0.041).

In an exploratory post hoc analysis motivated by the guideline’s comments on multiple conditional features, cases meeting ≥2 conditional criteria were contrasted with those meeting 0–1 criteria; the authors reported a higher local recurrence contrast and a Firth-penalized hazard ratio of 8.26 (95% CI 2.06–26.06; p = 0.005).

The investigators interpreted these stratified results as suggesting that, within this retrospective IORT-treated cohort, the 2024 classification may help distinguish a higher-recurrence subgroup, including in the exploratory post hoc analysis based on the accumulation of conditional criteria.

Implementation details centered on intraoperative delivery with a 50 kV X-ray source positioned within spherical applicators (1.5–5.0 cm) selected to fit the surgical cavity, with treatment duration reported as 20–50 minutes depending on applicator size, after tumor excision and negative sentinel lymph node assessment on frozen section.

Downstream interventions included re-excision in six cases for close margins and supplemental whole-breast irradiation in 29 cases, attributed to involved lymph nodes (n = 17), positive margins (n = 5), or unknown reasons (n = 7). The authors reported that three patients underwent mastectomy during follow-up (two after local recurrence and one bilateral mastectomy following recurrence by patient choice). Complications were described as early in 51/358 cases (14.2%) and late in 95/358 cases (26.5%), with reported event types spanning infection, seroma/hematoma, pain or sensory symptoms, skin toxicity, wound complications, cosmetic changes, and less common events such as burns, allergic reactions, and pleural effusion, forming the paper’s overall picture of long-term follow-up events after intraoperative delivery.

Key Takeaways:

• In this multicenter retrospective series, the authors reported 14 local recurrences (3.9%) over a median 7.1-year follow-up, with Kaplan–Meier estimates of local recurrence-free survival of 98.3% at 5 years and 94.8% at 8 years and overall survival of 99.4% at 5 years and 97.7% at 8 years.
• When the cohort was stratified by ASTRO 2024 risk categories, the authors reported higher modeled local recurrence risk in Conditional versus Suitable groups, and an exploratory analysis linked multiple conditional criteria with a higher recurrence signal.
• The report also described additional procedures and adjunct whole-breast irradiation in a minority of cases, along with early and late complication rates and a range of complication types observed over follow-up.