Background: The JCOG1212 trial seeks to evaluate the safety and efficacy of the superselective intra-arterial infusion of high-dose cisplatin with concomitant radiotherapy (RADPLAT) for patients (pts) with T4aN0M0 or T4bN0M0 maxillary sinus squamous cell carcinomas (MS-SCC). T4aN0M0 MS-SCC requires radical surgery with or without complete resection of the orbital contents as a standard treatment, which results in disfigurement and functional impairment. We herein report the results of the efficacy confirmation phase in the T4a cohort. Methods: Eligible pts with T4aN0M0 MS-SCC were registered and received cisplatin 100 mg/m² intra-arterially weekly for 7 weeks with concomitant radiotherapy (total 70 Gy) as determined by the results of the previous dose-finding phase. The trial investigated the 3-year overall survival (3yr OS) of RADPLAT to demonstrate its non-inferiority to surgery, the current standard of care, and its superiority to IV-CRT, which is used for pts refusing surgery. From the results of observational study undertaken by our group, the historical control for the 3yr OS was set at 80% for surgery and 65% for IV-CRT. A margin of 15% was set to demonstrate its non-inferiority to surgery and superiority to IV-CRT. Thus, when the lower boundary of the two-sided 90% confidence interval (CI) exceeds the threshold of 65% for 3yr OS, RADPLAT will be confirmed as the new standard. A sample size of 62 was required to achieve 80% power with a one-sided 5% significance level. Results: From April 2014 to August 2018, 65 pts were registered in the T4a cohort from 18 institutions, consisting of 54 males and 11 females with a median age of 64 years (range, 40-78 years) and ECOG PS 0/1 (58/7). After the exclusion of one ineligible patient, 64 pts were included in the primary analysis of efficacy and safety. The complete clinical remission rate was 73.4% (47/64, 95% CI, 60.9%-83.7%), with 15 showing CR and 32 showing good PR. The median follow-up was 4.5 years in all eligible pts and the primary endpoint for 3yr-OS was 82.8% (90% CI, 73.4%-89.2%), which met the prespecified hypothesis. The 3yr event-free survival and 3yr local event-free survival were 60.9% (95% CI, 47.9%-71.7%) and 65.6 % (95% CI, 52.5%-75.8%), respectively. With regard to acute adverse events, neutropenia (≥ grade 3), increased creatinine (≥ grade 2), hearing impairment (≥ grade 2), mucositis (≥ grade 3), and stroke (≥ grade 2) were observed in 14.1%, 3.1%, 3.1%, 20.3%, and 1.6% of pts, respectively. One treatment-related death due to a thromboembolic event was reported. Conclusions: RADPLAT is non-inferior to surgery and superior to IV-CRT for pts with T4aN0M0 MS-SCCs and showed manageable toxicities. Therefore, RADPLAT, as well as surgery, should be considered the new standard treatment option for these pts. Clinical trial information: UMIN000013706.