Epstein-Barr virus DNA-guided chemoradiotherapy for patients with low-risk locoregionally advanced nasopharyngeal carcinoma: a secondary 5-year follow-up analysis of an open-label, randomised controlled, phase 2 non-inferiority trial

Abstract

Background: Intensity-modulated radiation therapy (IMRT) with two cycles of concurrent 100 mg/m² cisplatin (DDP) presents a potential alternative for low-risk, locoregionally advanced nasopharyngeal carcinoma (LA-NPC). This study assessed its long-term survival outcomes and late toxicities.

Methods: This is a secondary analysis of an open-label, randomised, controlled, phase 2 non-inferiority trial, which enrolled patients with low-risk LA-NPC (pretreatment plasma Epstein-Barr virus DNA < 4000 copies/mL) at Sun Yat-sen University Cancer Center. Eligible participants were randomly assigned (1:1) to receive either two cycles or three cycles of concurrent cisplatin (100 mg/m² every 3 weeks) with intensity-modulated radiation therapy. The primary endpoint was 5-year progression-free survival (PFS); secondary endpoints were 5-year overall survival (OS), locoregional relapse-free survival (LRRFS), distant metastasis-free survival (DMFS), and late toxic effects. The non-inferiority margin was 10%. This secondary analysis of long-term follow-up was prespecified in the study protocol. Analyses of primary and secondary endpoints included intention-to-treat and per-protocol populations. The trial is registered with ClinicalTrials.gov, NCT02871518.

Findings: Between Sept 28, 2016, to Oct 18, 2018, 332 patients were enrolled and randomly assigned to the two-cycle group (n = 166) or three-cycle group (n = 166). The final follow-up date was Oct 11, 2024. Data were analysed from Oct 11, 2024, to June 5, 2025. At median follow-up of 79.7 months (IQR, 75.4-88.3 months), 5-year PFS rates were 85.0% (95% CI, 79.4-90.4) for the two-cycle group vs. 87.3% (95% CI, 82.2-92.4) for the three-cycle group (difference 2.4%; 95% CI, -5.0%-9.8%; noninferiority P = 0.016). No significant differences were observed in 5-year OS (95.2% [91.9-98.5] vs. 97.6% [95.3-100], HR, 2.02 (95% CI: 0.61-6.72), P _log-rank = 0.240) and the cumulative incidences of locoregional relapse (6.1% [2.4-9.8] vs. 6.0% [2.3-9.7], HR, 1.00 (95% CI: 0.42-2.41), P _Fine-Gray = 0.993) and distant metastasis (6.8% [2.9-10.7] vs. 7.3% [3.4-11.2], HR, 1.08 (95% CI: 0.48-2.44), P _Fine-Gray = 0.855). The three-cycle regimen demonstrated higher incidences of late toxicities: trismus (22.4% [37/165] vs. 12.7% [21/166], P = 0.028), xerostomia (83.0% [137/165] vs. 72.9% [121/166], P = 0.036), and hearing impairment/otitis (55.8% [92/165] vs. 44.0% [73/166], P = 0.042).

Interpretation: Five-year outcomes support the viability of two cycles of concurrent DDP with IMRT as an alternative to the standard three-cycle regimen, offering comparable survival outcomes with fewer late toxicities in patients with low-risk LA-NPC. However, as the study was conducted at a single center, the generalisability of these findings to non-endemic regions warrants further validation.

Funding: National Key Research and Development Program of China, Noncommunicable Chronic Diseases-National Science and Technology Major Project, National Natural Science Foundation of China, Guangdong Basic and Applied Basic Research Foundation, Science and Technology Program of Guangzhou, Sun Yat-sen University Clinical Research 5010 Program, and China Postdoctoral Science Foundation.

Keywords: De-intensified chemoradiotherapy; Long-term outcomes; Nasopharyngeal carcinoma; Randomised trial.

Fig. 1

CONSORT diagram of patients included and excluded in this study. DDP, cisplatin; ITT, intention-to-treat; pEBV, plasma Epstein–Barr virus; RT, radiotherapy.

Fig. 2

Kaplan–Meier and Nelson–Aalen curves for ITT Population. Kaplan–Meier curves for (A) progression-free survival and (B) overall survival. Nelson–Aalen cumulative risk curves for (C) locoregional relapse and (D) distant metastasis in ITT population.

Fig. 3

Treatment effects on progression-free survival subgroup analysis of progression-free survival in the ITT population. Note. The number of events and patients is shown in the study arm. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated in a univariate Cox proportional hazards regression model; interaction and stratified analyses were conducted according to sex, age, overall stage, pretreatment plasma EBV DNA levels, plasma EBV DNA clearance time and post-treatment plasma EBV DNA status. pEBV, plasma Epstein–Barr virus.