Risk of radiation necrosis with concurrent antibody-drug conjugates and radiotherapy in HER2-positive breast cancer with brain metastases: A meta-analysis.

Abstract

1039 Background: Antibody-drug conjugates (ADCs) have transformed the outcomes of HER2-positive breast cancer (BC), particularly in patients with brain metastases (BM) due to the use of ADCs like trastuzumab emtansine (T-DM1) and trastuzumab deruxtecan (T-DXd), which have demonstrated intracranial efficacy. Radiotherapy (RT), especially stereotactic radiosurgery (SRS), remains a cornerstone for BM management. However, combining ADCs with RT may increase the risk of symptomatic radiation necrosis (SRN). This meta-analysis evaluates SRN outcomes in patients receiving concurrent (C-ADC) versus non-concurrent ADCs with RT (NC-ADC). Methods: A systematic search was performed in January 2025 across PubMed, Cochrane, and conference proceedings from ASCO, SNO, ESMO, and SABCS. Eligible studies included randomized controlled trials and cohort studies (CS) of C-ADC and NC-ADC in HER2-positive BC patients with BM. Studies reporting SRN rates or related outcomes were included. A random-effects model was used to calculate pooled proportions and risk ratios (RR) with 95% confidence intervals (CIs). Heterogeneity across studies was assessed using the I² statistic, with values of 0–25% considered low, 26–50% moderate, and greater than 50% high. Results: Out of 884 studies screened, 9 CS (N = 421) were included. The median age was 56.3 years (IQR: 49.8–57). Patients receiving prior intracranial RT was 41.28%, 19.28% underwent SRS, and 19.58% received prior whole brain radiotherapy (WBRT). The median time of C-ADC was 8.75 days (IQR: 8.0–18.0) and NC-ADC was 273.5 days (IQR: 225.75–327.75). The pooled proportion of SRN in the C-ADC group was 19.5% (95% CI: 9.2%–29.8%; I² = 39.19%, τ² = 0.0061, P = 0.1382) indicating moderate heterogeneity. NC-ADC group experienced a pooled SRN proportion of 6.9% (95% CI: 2.5%–11.2%; I² = 51.98%, τ² = 0.0014, P = 0.0089), signifying high heterogeneity. The pooled RR showed a significantly increased SRN risk in the C-ADC group (RR = 2.726, 95% CI: 1.454–5.109, P = 0.002). Heterogeneity for the pooled RR was negligible (I² = 0.0%, Q = 0.20, P = 0.977), indicating consistent findings across studies. Conclusions: C-ADC is associated with a significantly higher risk of SRN. This risk is concerning but must be balanced against potential improvements in local control and efficacy outcomes. Prospective studies are needed to optimize treatment schedule and sequences to minimize toxicity and optimize survival. Meta-analytical findings. Analysis/Measure Effect Size 95% CI I² P-Value Notes Pooled Proportion (Overall) 0.110 0.050–0.169 76.53% 0.0003 High heterogeneity (τ²=0.0047) C-ADC Proportion 0.195 0.092–0.298 39.19% 0.0002 Moderate heterogeneity (τ²=0.0061) NC-ADC Proportion 0.069 0.025–0.112 51.98% 0.0021 High heterogeneity (τ²=0.0014) Risk Ratio (C-ADC vs. NC-ADC) 2.726 1.454–5.109 0.0% 0.002 No heterogeneity (τ²=0.0000)