Test-retest repeatability of quantitative radiomics features of FDG PET/CT studies in patients with diffuse large B cell lymphoma

Diffuse large B-cell Lymphoma (DLBCL) is an agressive form of lymph node cancer
with a variable prognosis. Up to 40% of the patients do not respond well to
first-line R-CHOP treatment. Current prognostic measures do not accurately
predict therapy failure whereas the need for such a prognostic marker is
increasing due to new promising treatment options. Good prognostic markers
could help identify the patiënt with a high risk of failure after R-CHOP
treatment. Pheno-and genotypic heterogeneity likely explains why pagtients and
even lesions within the same patient, respond differently to therapy.
Quantitative image features from whole body FDG PET-CT capture metabolic
heterogeneity at a lesional and patient basis. The spectrum of such features
(many different parameters covering e.g. shape, metabolice active volume,
heterogeneity) is called radiomics. Recently,studies examaning radiomics of
PET-based tumor phenotyping have shown that radoimics features have prognostic
potential.

At the moment it is unknown which radiomics features are robust against
variations in image characteristics. The objective of this study is to check
the robustness of radiomics features, and to investigate wheter specific pre-
and postprocessing routines should be performed to improve the robustness. We
will check this by scanning 20 patients on two seperate occasions within one
week.

In this monocenter observational study with invasive measurements patients with
diffuse large B-cel lymphoma will be included. All patients will undergo the
current standard diagnostic procedure including an [18F] FDG PET scan. The
second [18F]FDG PET scan will be perfomed for this study within one week after
the first scan. There is no further follow-up period. We expect to include the
total number of 20 patients in 12 months.

During this study, all patients will receive a venous cannula for tracer
injection. This might cause transient intravenous site discomfort. Patients
will undergo PET imaging twice. A PET scan is a regular diagnostic imaging
technique. Each study will be conducted in compliance with the radiation safety
guidelines of the department. However, this imaging technique has its risks as
patients are exposed to radiation. The first [18F]FDG PET/CT scan is standard
of care for these patients and is responsible for 3.5 mSv from the administered
tracer [18F]FDG and 3 mSv from the low-dose CT scan. The radiation burdon for
the retest [18F]FDG PET/CT is 6.5 mSv. The total sum of radiation burdon per
patient for a complete set of test-retest studies will be approximately 13 mSv.
Participating in this study will result in an extra radiation burdon of 6.5 mSv
compared to the standard of care. This is to the same order of magnitude as the
annual natural background radiation in various parts of the world(Netherlands
Commission on Radiation Dosimetry, 2016)

In a general population of age 65, a dose of 6.5 mSv is estimated to increase
the lifetime change of developing fatal cancer with 0.02%, in addition to a
baseline change of 44.9% for males and 37.5% for females (XrayRisk, 2018). The
study population has a decreased survival (Sehn et al., 2015), while the
possible hazard of cancer induction by radiation is only to be expected after
decade(s) post exposure. Consequently, the actual increase in chance of
developing a fatal cancer for these patients is thus only a fraction of the
percentage mentioned above. In the light of this scientific merit the
investigators therefore consider the additional radiation burden acceptable,
balancing ethical considerations with the minimum number required to obtain
reliable information.