2283 - Radiosensitization and Active Antitumor Effect of Antibody-Drug Conjugates on Brain Metastases after Whole Brain Radiation
Presenter(s)
Y. Yu1, T. Treechairusame1,2, B. S. Imber3, L. R. G. Pike3, M. Aristophanous4, N. Y. Lee1, S. Modi5, and D. G. Hsu4; 1Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 2Division of Radiation Oncology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand, 3Memorial Sloan Kettering Cancer Center, New York, NY, 4Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, 5Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, NY
Purpose/Objective(s): HER2-Targeting Antibody-drug conjugates (ADCs) are increasingly used in patients with brain metastases. We hypothesized concurrent ADCs could radiosensitize tumors during whole brain radiation therapy (WBRT). We analyzed local control after WBRT and modeled failure timing to distinguish radiosensitization from drug-mediated tumor suppression.
Materials/Methods:
We queried our institutional database for patients treated with WBRT for brain metastases who also received HER2-targeted ADC therapy (T-DM1 or T-DXd) in any sequence. Brain metastases were tracked longitudinally on MRIs (q3 months) using METRO, a deep learning brain metastasis segmentation program, which measured longest axial diameter (LAD). Local failures were manually verified. We defined local failure as LAD exceeding baseline LAD + 2 mm on two consecutive post-WBRT scans, and restricted the evaluable cohort to lesions with baseline LAD >= 5 mm. ADC was classified as concurrent if any individual dose was administered within 21 days before to 7 days after WBRT start. Competing risks regression was performed per brain metastasis with death as a competing event and clustering by patient. To distinguish radiosensitization from active drug effect, we performed three analyses: (1) concurrent ADC vs no concurrent ADC; (2) time-varying ADC exposure (on/off) via cause-specific Cox regression; and (3) comparison of new brain metastasis rates during ADC-active vs ADC-inactive person-time.Results:
Thirty-six patients received WBRT and HER2-targeted ADC therapy (breast, n=29; lung, n=6; salivary, n=1), with 1,300 brain metastases tracked sequentially. Of these, 640 lesions in 29 patients met the >= 5 mm baseline LAD threshold for evaluation. Six patients had concurrent ADC (T-DM1, n=4; T-DXd, n=2), of whom 4 had evaluable lesions (58 lesions); 25 patients (582 lesions) had no concurrent ADC. Overall, 31 of 640 evaluable lesions (4.8%) developed local failure. No local failures occurred among concurrent ADC lesions (0/58) compared to 31/582 (5.3%) among controls. On a per-patient level, no concurrent ADC patient (0/4) experienced local failure compared to 10/25 (40%) without concurrent ADC (24-month cumulative incidence: 0% vs 41.4%). In the time-varying analysis, no local failures occurred during active ADC exposure (0/16 failures during ADC-active intervals) despite ADC-active person-time comprising 35.9% of total person-time. Local control remained durable after discontinuation of ADC: no local failures occurred in any concurrent patient after ADC completion. New brain metastases developed at similar rates during ADC-active and ADC-inactive intervals (0.75 vs 0.78/month, p=0.58).Conclusion: No local failures were observed among brain metastases treated with concurrent HER2-targeted ADC and WBRT. Local control remained durable after discontinuation of ADC. This result suggests concurrent HER2 targeted ADC may radiosensitize tumors to WBRT. Validation in larger cohorts is warranted.