A multicenter trial in the detection of locally recurrent nasopharyngeal carcinoma with Epstein Barr Viral DNA

Presentation: D218
Topic: Skull Base
Type: Poster
Date: Wednesday, May 1, 2019 (1:00 PM - 7:00 PM) | Thursday, May 2, 2019 (9:00 AM - 7:00 PM)
Session: Wednesday, May 1, 2019 (1:00 PM - 7:00 PM) | Thursday, May 2, 2019 (9:00 AM - 7:00 PM)
Authors: Jason Chan1, Ronald Lai1, Allen Chan1, Jacky Lam1, Joseph Chung2, Raymond Tsang2
Institution(s): 1The Chinese University of Hong Kong, 2The University of Hong Kong

Objectives: Both brush biopsies of the nasopharynx and plasma for EBV DNA have been used in the detection of primary nasopharyngeal carcinoma (NPC) but neither has been thoroughly evaluated in the detection of locally recurrent or persistent NPC. Here we sought to evaluate the sensitivity and specificity of a combination of nasopharyngeal brush biopsy and plasma EBV DNA in the detection of locally recurrent/ persistent NPC.

Methods: Prospective case-control study evaluating patients with biopsy proven NPC and controls with a history of NPC, now proven disease free with clinical examination and MRI The NPscreen was used for the nasopharyngeal brush with a cutoff of <1.7 EBV detection level as being normal. Plasma EBV DNA was performed as described by Lo et al. 1999 with a value of 0 being normal. In the combination analysis, if any one test was positive the result was considered positive.

Results: Twenty males and 8 females with an average age of 57.4 years were recruited.  Among the locally recurrent group, 5 were stage I disease; 1 was stage II; 4 were stage III; one’s staging could not be traced and one had no malignancy found after nasopharyngectomy despite biopsy showing recurrence.  The optimal EDLs were 0.79 (Youden index = 0.845) for NP brush and 2.0 (Youden index = 0.769) for plasma EBV DNA respectively.  The sensitivity and specificity of the NP brush were 91.7% (76%-100%, 95% CI) and 92.9% (79.4%-100%, 95% CI) respectively, while the sensitivity and specificity of plasma EBV DNA were 76.9% (54%-99.8%, 95% CI) and 100% respectively.  When the above EDLs were implemented with a combination of the NP brush and plasma EBV DNA, the sensitivity and specificity in combination were 100% and 92.9% (79.4-100%, 95% CI) respectively, with positive predictive value (PPV) of 92.9% (79.4%-100%, 95% CI) and negative predictive (NNV) value of 100%. Finally, there was a correlation with recurrent T stage with plasma but not nasopharyngeal brush EBV DNA.

Conclusion: The high sensitivity of NP brush biopsy and high specificity of EBV DNA in detecting locally recurrent NPC may provide additional surveillance modalities besides current practice when used in combination for the detection of locally recurrent nasopharyngeal carcinoma.  Further study with a larger sample size would be required to validate the cut-off values.