Pre-Treatment EBV-DNA/TLG Risk Stratification Predicts Progression-Free Survival and Overall Survival After Definitive Radiotherapy in Non-Metastatic Nasopharyngeal Carcinoma

---

Duymaz M. Ç., Özkaya Duymaz E., Ünsal Akın N., Bekiş R., Tozduman B., Appak Ö., ...Daha Fazla

ESTRO 2026, Stockholm, İsveç, 15 - 19 Mayıs 2026, ss.1, (Özet Bildiri)

• Yayın Türü: Bildiri / Özet Bildiri
• Basıldığı Şehir: Stockholm
• Basıldığı Ülke: İsveç
• Sayfa Sayıları: ss.1
• Dokuz Eylül Üniversitesi Adresli: Evet

Özet

Background: To determine whether combining pre-treatment Epstein–Barr virus (EBV) DNA and PET-derived total lesion glycolysis (TLG) improves prognostic stratification for progression-free survival (PFS) and overall survival (OS) in non-metastatic nasopharyngeal carcinoma (NPC).

Methods: We retrospectively analyzed 86 consecutive, non-metastatic NPC patients treated between 2010 and 2024 with curative-intent radiotherapy (RT) and/or systemic therapy. Pre-treatment ¹⁸F-FDG PET/CT was analyzed in LIFEx; primary and nodal lesions were delineated using a 40% SUVmax relative threshold, and whole-body tumor burden metrics were computed (TLG = MTV × SUVmean). Plasma EBV-DNA (copies/mL) was quantified by qPCR. Staging was performed according to the AJCC/UICC 8th edition. RT technique was 3D-conformal in 16 patients and IMRT/VMAT in 70 patients; all received EQD2₁₀ ≥69 Gy. Neoadjuvant chemotherapy and concurrent chemoradiotherapy were administered in 30 and 75 patients, respectively, while 27 patients received both. The EBV-DNA threshold (3,500 copies/mL) was prespecified based on the literature, whereas the TLG threshold (200) was derived through ROC analysis. Four subgroups were constructed: Low–Low (EBV-DNA_low/TLG_low), High–Low (EBV-DNA_high/TLG_low), Low–High (EBV-DNA_low/TLG_high), and High–High (EBV-DNA_high/TLG_high). The primary endpoint was PFS; OS was a secondary endpoint.

Results: The cohort was predominantly male (67/86, 77.9%), with a median age of 49 years (12–76); 80.2% had stage III–IV disease. Median follow-up was 75.5 months (reverse Kaplan–Meier). The 3-year and 5-year PFS rates were 82.7% and 81.1%, respectively, and the 3-year and 5-year OS rates were 92.3% and 85.9%, respectively. Twenty PFS events (23.3%) occurred, comprising 15 disease progressions and 5 deaths without documented prior progression. In univariate analysis (UVA) for PFS in the complete-case cohort (n = 59), TLG ≥200 (HR = 3.29, p = 0.038), T3–4 classification (HR = 3.21, p = 0.041), and EBV-DNA_high/TLG_high group (HR = 4.01, p = 0.013) were significant predictors; EBV-DNA ≥3,500 copies/mL showed a trend (HR = 3.14, p = 0.058). In UVA for OS, TLG ≥200 (HR = 6.11, p = 0.025) and EBV-DNA_high/TLG_high group (HR = 4.13, p = 0.036) were significant. In multivariate analysis (MVA) for PFS (n = 59), EBV-DNA_high/TLG_high group remained an independent predictor (adjusted HR = 3.97; 95% CI, 1.32–11.93; p = 0.014), along with T3–4 classification (adjusted HR = 3.18; 95% CI, 1.03–9.84; p = 0.045). In MVA for OS, the EBV-DNA_high/TLG_high group remained significant (adjusted HR = 4.13; 95% CI, 1.10–15.52; p = 0.036).

Conclusion: A combined pre-treatment stratification using TLG ≥200 and EBV-DNA ≥3,500 copies/mL identifies the EBV-DNA_high/TLG_high subgroup as having significantly inferior PFS and OS, supporting a simple and potentially clinically applicable risk-stratification approach; however, external multicenter validation is warranted.