ESTRO 2024 - Abstract Book

S1218

Clinical - Head & neck

ESTRO 2024

There is an increasing interest in the identification of prognostic and predictive patient characteristics in head and neck squamous cell carcinoma. Human papillomavirus (HPV)-positive status is a well-known characteristic that has been linked to a favorable outcome after treatment with (chemo)RT for oropharyngeal squamous cell carcinoma (OPC). The apparent diffusion coefficient (ADC), a quantitative variable derived from diffusion-weighted (DW) MRI, seems to be a less investigated, but promising parameter for pre-treatment identification of (non-)responders to radiotherapy (RT). This study aimed to 1) confirm the predictive value of ADC in response to RT on a large prospective dataset and 2) investigate the differences in quantitative diffusion-weighted (DW) MRI parameters according to HPV status in OPC.

Material/Methods:

We prospectively analyzed all patients undergoing either RT or chemoRT for histologically proven OPC between 2005 and 2018. OPC were considered HPV+ if more than 70% diffuse nuclear and cytoplasmic p16 immunohistochemistry staining was present. MRI with echo-planar DW sequences at 6 b-values (0-1000 s/mm²) were acquired before (chemo)RT treatment and during the fourth week of (chemo)RT. For each scan, an experienced head and neck radiologist manually delineated the entire primary tumor volume on the apparent diffusion coefficient (ADC) map, and several first-order histogram parameters (ADC mean, ADC median, ADC 10th percentile, ADC 90th percentile, ADC minimum and ADC maximum) were extracted. Delta (Δ) ADC values, the percentage of ADC changes between baseline and during RT at week 4, were calculated using the formula ΔADC = [(ADCduring – ADCpre)/ADCpre]*100. All ADC parameters were compared between HPV+ and HPV- OPC using the Mann-Whitney-U test, while oncological outcomes were compared using the log rank test. The predictive value of tumor ADC and ΔADC for local control (LC), loco -regional control (LRC), distant metastases-free survival (DMFS) and overall survival (OS) was examined using Cox regression models, applying logarithmic transformation upon continuous predictors. The significance threshold was set at a p-value of <0.05. One hundred seventy-eight patients with OPC were included. HPV status was known for 158 patients, of which 61 (39%) were considered HPV positive. Median follow-up time was 5.1 years. The estimated 2- and 5-year LRC was 80.1 % (92.6% for HPV+ vs. 71.2% for HPV-, p<0.01) and 75.2 % (89.3 % for HPV+ vs. 63.7% for HPV-, p<0.01), respectively. The 5-year OS was 54.9% (75.2% for HPV+ vs. 46.2% for HPV-, p<0.01). Kaplan-Meier curves are presented in Figure 1. The pretreatment ADC values were significantly higher in patients who developed a locoregional recurrence after RT, with a hazard ratio (HR) of 1.85 for the ADC mean (p=0.04). High tumor ADC values during RT were significantly associated with favorable LC (HR 0.4, p<0.01), LRC (HR 0.4, p<0.01), OS (HR 0.4, p<0.01) and DMFS (HR 0.5, p<0.01). Lower ΔADC was predictive for poorer LC (HR 0.8, p<0.01), LRC (HR 0.8, p<0.01), DMFS (HR 0.9, p<0.01) and OS (HR 0.9, p<0.01). These observations were valid for the entire population and for the HPV negative subgroup only. Boxplots of the mean ADC values pretreatment, during treatment and ΔADC mean according to HPV status are presented in Figure 2. Between HPV+ and HPV- OPC, there was a significant difference in pretreatment first-order ADC 10th percentile (mean value of 82.4 *10^-5 mm2/s for HPV+ vs. 90.3 for HPV-, p=0.03) and a trend towards a lower pretreatment mean ADC in the HPV+ group (mean value of 111.4 *10^-5 mm2/s for HPV+ and 125.2*10^-5 mm2/s for HPV-, p=0.07). During the 4th week of RT, the ADC mean, median, minimum, 10th percentile and 90th percentile ADC values were all significantly higher in HPV+ OPC compared to HPV- OPC. The ΔADC mean value was significantly higher in the HPV+ OPC group (mean value of 95.0% for HPV+ vs. 55.3% for HPV-, p<0.01). Results:

Figure 1: LRC (a) and OS (b) according to HPV status

Made with FlippingBook - Online Brochure Maker