Wait or treat? managing asymptomatic brain metastases in oncogene-mutated NSCLC: Results of a phase-III randomized controlled trial.

Abstract

8624 Background: The timing of brain radiation therapy (RT) in patients with oncogene-mutated NSCLC with asymptomatic brain metastases (ABM) has been a matter of debate. No level 1 evidence supports delayed brain RT in ABM. This phase III, open-label, RCT (NCT05236946) evaluates Upfront Cranial RT (U-CRT) versus Delayed CRT (D-CRT) in ABM of oncogene-mutated NSCLC. Methods: Eligible patients (≥18 years, ECOG ≤2), with EGFR mutation or ALK gene rearrangement and completely ABM, were randomized to U-CRT vs D-CRT at intracranial progression (ICP), stratified by GPA score (0-2 vs > 2) and BM presentation (Synchronous vs Metachronous). Key exclusions include symptomatic BM, brainstem metastases, and prior cranial RT. All pts received standard systemic therapy (TKI ± CT). The primary endpoint is intracranial PFS with death treated as a competing event. Secondary endpoints included OS, PFS, and treatment toxicity. All patients underwent MRI brain every 3 months for 1 st year and 6 months thereafter, unless indicated clinically. Cumulative incidence functions were estimated, and subdistribution hazards were compared using the Fine and Gray model. With a planned enrollment of 208 patients and 139 target events, the study had 80% power to detect a hazard ratio of 0.62 at a two-sided alpha of 0.05. The icPFS was analyzed in intention-to-treat populations using Kaplan-Meier estimates and log-rank tests. Results: A total of 208 patients were randomized (105-U-CRT; 103-D-CRT). Baseline characteristics were well-balanced in both arms. The median age was 53 years (range, 26-83). EGFR and ALK mutations were seen in 87.6% and 12.4% in U-CRT and 79.6% & 20.4% in D-CRT arm. 1 st /2 nd generation TKIs ± CT were received by 58% & 72% pts, respectively and rest received 3 rd generation TKI. In U-CRT, 44.8% received SRS/SRT, 52.4% received WBRT± boost, and 3 pts did not receive RT. The median follow-up was 27.4 months (95%CI 26.2 -33). The cumulative incidence of ICP at 1 yr were 8.7% (95%CI 2.9%, 14.5%) & 25.7% (95%CI 16.8%, 34.7%), and at 2-yrs were 21.7% (95%CI 12.6%, 30.8%) & 50% (95%CI 39.2%, 60.9%) with sub-HR = 0.35 (95%CI 0.21, 0.59), p < 0.001. The median icPFS were 18 mo (95%CI 15.8, 23.5) and 14.3 mo (95%CI 12.6, 19.6) with HR-0.84 (95%CI 0.59, 1.20), p = 0.35. There was no difference in median PFS of 11.7 mo (95%CI 9.2, 14.1) & 12.0 mo (95% CI 9.8, 14.9), p = 0.37 and in median OS of 23.3 mo (95% CI 17.7-28.8) & 28.7 mo (95%CI 17.7, 39.7), p = 0.06, respectively. In the D-CRT arm, 39/47 who had ICP received salvage CRT. The incidence of grade ≤2 radiation necrosis was 25.4% and grade > 3 in 5.8% in upfront RT arm. Conclusions: In asymptomatic BM of oncogene-mutated NSCLC pts, upfront cranial RT significantly reduced the incidence of ICP; however, it did not improve the survival outcomes. The difference in survival may become more apparent as the OS data matures. Data on QOL, neurocognition, and molecular analysis will be reported later. Clinical trial information: NCT05236946 .