Prognostic Utility of the Imaging and Biological Signatures in Nasopharyngeal Carcinoma

1. Background: Nasopharyngeal carcinoma (NPC) differs from other head and neck malignancies terms of its epidemiology, pathology, and treatment outcome. NPC has a good local tumor control rate but a higher incidence of distant metastasis compared with squamous cell carcinomas of other regions of the head and neck. Thus, it's usually considered a distinct study group. A good prognostic model for the development of distant failure for NPC patients is still lacking. It is well known that Epstein-Barr virus (EBV) genomes are present in almost every NPC tumor cell, irrespective of histologic differentiation and geographic distribution. Various EBV-derived/related factors, therefore, have been used as NPC tumor markers, including EBV-specific viral capsid antigen (VCA) IgA and cell-free EBV DNA. The measurement of cell-free EBV DNA load has already been shown to be a good pretreatment prognostic indicator. Another important feature of NPC is that the tumors are characterized by heavy infiltration of nonmalignant lymphocytes and might play an important role in tumorigenesis; and cytokines and chemokines may represent the key mediators of this interaction. Compelling evidence has also emerged in recent years suggesting that cytokines and chemokines play important roles in regulating processes critical to tumor progression, such as proliferation and metastasis. Beyond blood biomarkers, standard uptake value (SUV) from FDG PET could predict treatment failure in NPC patients. On the other hand, apparent diffusion coefficient (ADC) from diffusion-weighted MRI (DWI) has been shown to be an important prognostic marker in cancer patients, including head and neck cancer.
2. Aim: To investigate the roles of imaging and blood biomarkers in predicting distant failures, recurrences, and survival in NPC patients.
3. Study design:

Study participants:

Patients with a histological diagnosis of primary M0 NPC were deemed eligible. All study participants received both 18F-FDG PET/CT and head-neck MRI imaging before treatment, during radiotherapy, and 3 months after the definitive treatment. The blood sample was collected on the same day of the PET/CT scan.

18F-FDG PET/CT: Study participants were instructed to fast for six hours before the PET study. FDG (370-444 MBq) was administered intravenously. No intravenous contrast material was administered for CT scans. The participants underwent head to mid-thigh scans. PET and CT images were acquired consecutively 50 to 70 minutes after the injection of 18F-FDG. Before PET acquisition, a standard helical CT scan was acquired from the head to the proximal thighs using the manufacturer's dose reduction software. The PET images were reconstructed using the CT data for attenuation correction and an ordered subsets expectation maximization iterative reconstruction algorithm.

MRI :

MRI was performed at 3 Tesla (MAGNETOM® Trio with Tim; Siemens Medical Solutions, Bonn, Germany). Dedicated MRI of the head and neck region was obtained in the axial projection with T1-weighted turbo spin echo (TSE) sequence and subsequently with T2-weighted TSE sequences with fat saturation. After intravenous administration of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) at 3 mL/sec with a dose of 0.1 mmol/kg, MRI of the head and neck region was performed using a fat-saturated T1-weighted TSE sequence in the axial and coronal planes. DWI was acquired using single shot spin-echo echo-planar imaging with a modified Stejskal-Tanner diffusion gradient pulsing scheme.

Collection and storage of blood plasma:

Ten-milliliter samples of peripheral blood were drawn from subjects using standardized phlebotomy procedures and collected into an EDTA tube for isolation of plasma. Blood samples were centrifuged at 2000×g. Plasma was immediately aliquoted, transferred into plain polypropylene tubes, and was stored in a dedicated freezer at -80°C. No more than one freeze-thaw cycle was allowed for each plasma sample.

Statistical analysis:

Recurrence-free survival (RFS) and overall survival (OS) served as main outcome measures. RFS was defined as the time between the end of treatment and the date of recurrence (tumor relapse or death) or censored at the date of the last follow-up. Distant failure-free survival was also calculated from the date of diagnosis to the date of documented distant recurrences, or censored at the last follow-up date. OS was calculated from the date of diagnosis to the date of death or censor at the date of the last follow-up for surviving study participants. The cutoff values for the clinical variables, blood biomarkers, and imaging parameters in survival analysis were determined using the log-rank test based on the RFS and OS rates observed in the entire study cohort. Survival curves were plotted using the Kaplan-Meier method. The effect of each individual variable was initially evaluated using univariate analysis. Cox regression models were used to identify the predictors of survival. Two-tailed P values < 0.05 were considered statistically significant.