PD-1 blockade with sintilimab plus induction chemotherapy and concurrent chemoradiotherapy (IC-CCRT) versus IC-CCRT in locoregionally-advanced nasopharyngeal carcinoma (LANPC): A multicenter, phase 3, randomized controlled trial (CONTINUUM).

Background: Despite the success of PD-1 blockade plus chemotherapy in recurrent/metastatic NPC, its role in LANPC is unproven. This trial evaluated the efficacy and safety of adding sintilimab (a PD-1 inhibitor) to IC-CCRT in LANPC. Methods: Patients with non-metastatic high-risk LANPC (stage III-IVA, excluding T3-4N0/T3N1) were enrolled at 9 centers in China, and randomized (1:1; stratified by center and stage with block size four) to the Standard Arm (gemcitabine and cisplatin IC plus cisplatin CCRT), or Sintilimab Arm (sintilimab plus IC-CCRT). Sintilimab 200mg was given intravenously once every 3 weeks for up to 12 cycles (3 induction, 3 concurrent, and 6 adjuvant). The primary endpoint was event-free survival (EFS) (i.e. freedom from local/regional/distant failure or death). It is estimated that approximately 417 patients would provide 80% power to detect a hazard ratio (HR) of 0.52 with a two-sided type 1 error of 0.05. Quality of life (QoL) was assessed by EORTC-C30. Biomarkers including tertiary lymphoid structure (TLS), PD-L1, and gene expression were also analyzed. Results: Between December 2018, and March 2020, 425 patients were randomized to the Sintilimab Arm (n = 210) and Standard Arm (n = 215). After a median follow-up of 42 months (94% alive patients >= 36 months), the intention-to-treat analysis showed that 3-year EFS was 86.1% in the Sintilimab Arm and 76.0% in the Standard Arm (stratified HR, 0.59; 95% confidence interval [CI], 0.38-0.92; stratified log-rank p = 0.019). Grade 3-4 adverse events (AEs) occurred in 155 (74.2%) and 140 (65.4%) patients, including immune-related AEs in 20 (9.6%) and 2 (0.9%) patients and grade 5 AEs in 2 (0.95%) and 1 (0.5%) patients in the Sintilimab and Standard Arm, respectively (Table). No minimum clinically important differences in QoL were observed. The benefit of the addition of sintilimab was observed in patients with TLS (HR 0.18; 95% CI, 0.04-0.81; p = 0.011) but not in patients without TLS (HR 0.94; 95% CI, 0.50-1.76; p = 0.85). Conclusions: The addition of sintilimab to standard IC-CCRT results in significant improvement of EFS, manageable safety profile, and comparable QoL in high-risk LANPC. TLS appears to be a predictive biomarker for benefit of sintilimab.