Research Article: FDG PET based prediction of response in head and neck cancer treatment: Assessment of new quantitative imaging features

Date Published: April 19, 2019

Publisher: Public Library of Science

Author(s): Reinhard R. Beichel, Ethan J. Ulrich, Brian J. Smith, Christian Bauer, Bartley Brown, Thomas Casavant, John J. Sunderland, Michael M. Graham, John M. Buatti, Qinghui Zhang.


18 F-fluorodeoxyglucose (FDG) positron emission tomography (PET) is now a standard diagnostic imaging test performed in patients with head and neck cancer for staging, re-staging, radiotherapy planning, and outcome assessment. Currently, quantitative analysis of FDG PET scans is limited to simple metrics like maximum standardized uptake value, metabolic tumor volume, or total lesion glycolysis, which have limited predictive value. The goal of this work was to assess the predictive potential of new (i.e., nonstandard) quantitative imaging features on head and neck cancer outcome.

This retrospective study analyzed fifty-eight pre- and post-treatment FDG PET scans of patients with head and neck squamous cell cancer to calculate five standard and seventeen new features at baseline and post-treatment. Cox survival regression was used to assess the predictive potential of each quantitative imaging feature on disease-free survival.

Analysis showed that the post-treatment change of the average tracer uptake in the rim background region immediately adjacent to the tumor normalized by uptake in the liver represents a novel PET feature that is associated with disease-free survival (HR 1.95; 95% CI 1.27, 2.99) and has good discriminative performance (c index 0.791).

The reported findings define a promising new direction for quantitative imaging biomarker research in head and neck squamous cell cancer and highlight the potential role of new radiomics features in oncology decision making as part of precision medicine.

Partial Text

Quantitative imaging with 18 F-fluorodeoxyglucose (FDG) PET is routinely performed for head and neck squamous cell cancer (HNSCC) patients, yet the information used from these scans is often limited to qualitative analysis determining the presence or absence of disease in an anatomically defined area along with a report of the maximum standardized uptake value (SUVmax). While visual analysis is sufficient for diagnosis and staging, a more quantitative approach to FDG PET/CT analysis holds promise as a predictive tool. For example, a recent review paper by Castelli et al. [1] summarized the results of 45 studies (overall 2928 patients) regarding the predictive value of FDG PET with respect to clinical outcome in head and neck cancer treatment with chemoradiotherapy (CRT). The vast majority of the investigated studies were focused on simple, standard quantitative indices like SUVmax, peak uptake value (SUVpeak) [2], metabolic tumor volume (MTV), and Total Lesion Glycolysis (TLG); only three studies performed texture or shape analysis. The study concluded that MTV and TLG in pre-treatment PET scans showed good correlation with disease free survival (DFS) or overall survival (OS), while simple indices like SUVmax and SUVpeak showed less promise [1].

Patients were followed for a median of 48.8 months (range 5.4–124.3 months). DFS events (recurrence or death) were observed for 25 of the 58 patients with baseline features and 13 of the 25 with post-treatment change features. Hence, 13 of 25 positive scans at 8–12 week follow-up had a true recurrence or died and 12 of 25 had false positive initial follow-up scans and did not have evidence of recurrence with follow-up.

Development of image-based biomarkers is a multi-step process that begins at discovery and migrates to validation and ultimately to regulatory approval with many steps in between [21]. While offering great potential for clinical use, the development of biomarkers is resource intensive and requires development of practical tools for consistent definitions of regions of interest, normalization, and feature calculation, similar to the requirement for development of rapid sequencing technologies for molecularly-based precision techniques. Thus, the goal of this work was to identify new FDG PET based features that are promising for outcome prediction in HNSCC, and therefore, should be further investigated by the community.

The promise of precision medicine is more than the potential to use genetic information, but also includes identification of image-based biomarkers that define useful characteristics of tumors before, during, and after treatment. The presented work compares standard and new features and assesses their suitability for FDG PET based prediction of response in head and neck cancer treatment. The reported findings should help define potential new directions for biomarker research in HNSCC.