1123 - Prediction of Severe Early Toxicities with pATM Blood Assay: Results from a Prospective Real-Life Study

Purpose/Objective(s): We aimed to evaluate the diagnostic performance of the previously reported Ataxia-Telangiectasia Mutated protein (pATM) blood assay on the occurrence of severe early radiotherapy (RT)-induced toxicities in a real-life cohort from the Luxembourgish National Radiotherapy center (APRILUX).

Materials/Methods: 200 consecutive patients managed in curative intent for any primary were enrolled (NCT05433974). Total blood was sampled before the initiation of RT and pATM assay was performed by the neighboring independent laboratory - previously trained. We set a pATM concentration cutoff of 57.8 ng/mL to differentiate patients prone to grade <2 vs. grade?2 toxicity (ELISA). Patients were assessed in blind weekly during RT and 3 months after completion of RT. The maximal early toxicity endpoint was recorded using the Common Terminology Criteria for Adverse Events (CTCAE v5.0). We also collected clinical and dosimetric parameters with a possible impact on toxicity severity. The correlation was evaluated with univariate analysis and logistic regression. The discrimination power of the pATM assay was evaluated through the Area Under the Receiver Operator Characteristics Curve (AUC-ROC).

Results: The 200 patients were included between October 2023 and May 2024. Among the 198 analyzable patients, 61.1%, 14.1% and 8.6% were treated on breast, prostate and brain, respectively. The mean delivered dose was 51.5 Gy (24.0 – 70.0) with fractionation ranging between 1.80 and 7.25 Gy. 68 out 198 (34,3 %) patients exhibited a grade=2 toxicity event (any system) - including 29 patients with dermatitis; Specifically, in the breast subpopulation, 29 out of 121 patients (25,6 %) experienced grade=2 toxicity event including 17 dermatitis.

Grade =2 events were associated with pATM assay results, phototype, cardiovascular comorbidities and mean dose to heart (p=0.06). On the whole cohort, a combined biological/clinical model including pATM assay result (OR=3.2) and phototype (OR=1,57) could predict any grade =2 events with an AUC of 0.68. In the breast subpopulation, the AUC reached 0.73 (OR=7.6 and 1.5, respectively). Focusing on grade =2 dermatitis in the breast subpopulation, the AUC reached 0.76 with variables including pATM assay (OR=3.9), mean dose to heart (OR=1,1) and cardiovascular comorbidities (OR=3,7).

Conclusion: APRILUX is the first real-life prospective study to demonstrate the value of a pre-RT pATM blood assay - performed in a neighboring laboratory - associated with clinical/dosimetric parameters to accurately predict grade =2 early toxicity in any system, any primary. The performance increased when focusing on dermatitis and breast cancer patients. Correlation with late toxicity will be reported after a longer follow-up.