Abstract 405O
Background
Plasma EBV DNA is an archetypal tumour marker for endemic NPC and has been employed for disease surveillance, and early detection of occult metastases in the post-radiotherapy (RT) setting. However, the accuracy of this test for this indication is uncertain. Here, we utilised a single-institution RWD to interrogate the accuracy of a harmonised plasma EBV DNA test for early detection of recurrence.
Methods
NCT04340024 was a prospective RT registry from a single institution that included patients with biopsy-proven NPC. Patients who underwent definitive RT ± chemotherapy, and had ≥1 post-treatment EBV DNA test were included. To evaluate its false positive (FP) and negative (FN) rates, we determined the binary presence or absence of relapse within a defined time-period following an EBV DNA test, and performed sensitivity analyses based on 3, 6, 9, and 12-mo window. Relapse was determined by radiological imaging.
Results
Between Jan 2019 to Feb 2023, 1039 tests were performed in 259 patients. Of these, 802 (77.2%), 181 (17.4%) and 56 (5.4%) were recorded as undetectable (U), 1-500 copies/mL, and >500 copies/mL, respectively. For the 802 U readings, FN rates were 6.1% (49/802), 0.1% (1/802), and 0.4% (3/802) within a 12-mo window at 1, 2, and 3 y post-RT, respectively. Of the 237 positive EBV DNA results, FP rates were 57.5%, 58.3%, and 45.5% for 1-500 copies/mL, and 21.2%, 0%, and 0% for >500 copies/mL within a 12-mo window at 1, 2, 3 y post-RT, respectively. The results were consistent on sensitivity analyses adjusting for post-test window duration. To reduce our FP rates for EBV DNA 1-500 reading, we observed that consecutive testing lowered FP rates from 56.9% (103/181) for single EBV DNA readings of 1-500 to 37.1% (23/62) for 1-500 to 1-500/>500, while the proportion of patients with relapse reduced to 17.3% (9/52) for 1-500 to U.
Conclusions
Herein, we showed that plasma EBV DNA has a high negative predictive value for non-recurrence post-RT, although we observed a 6.1% FN rate at 1 y post-RT. However, the FP rates were substantial especially for the 1-500 range, and consecutive testing should be encouraged.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
402O - Preliminary results of phase II study to evaluate safety and efficacy of combination pucotenlimab with epidermal growth factor receptor-ADC (EGFR-ADC) MRG003 in recurrent or metastatic nasopharyngeal carcinoma (R/M-NPC)
Presenter: Danyun Ruan
Session: Proffered Paper session: Head and neck cancers
Resources:
Abstract
403O - Tislelizumab (TIS) plus chemotherapy (CT) vs placebo (PBO) plus CT as first-line (1L) treatment for recurrent or metastatic nasopharyngeal cancer (NPC): 3-year follow-up from the RATIONALE-309 study
Presenter: Wen Feng Fang
Session: Proffered Paper session: Head and neck cancers
Resources:
Abstract
404O - A phase III prospective randomised trial of two- weekly TPF(Taxane/Platinum/5FU) vs three weekly TPF as induction chemotherapy in locally advanced squamous cell carcinoma of head and neck: A preliminary analysis
Presenter: Animesh Gupta
Session: Proffered Paper session: Head and neck cancers
Resources:
Abstract