Beskrivelse
Background: 2-[18F]Fluoro-2-deoxy-d-glucose ([18F]FDG) positron emission tomography/computed tomography (PET/CT) is a widely used non-invasive imaging modality for various
applications including oncology. Because [18F]FDG is not a specific radiotracer which reflects
glycolytic activity in tumour cells, inflammatory processes and other tissues with high metabolic
activity take up the tracer. Thus, evaluating healthy tissue metabolism (HTM) based on [18F]FDG in
cancer patients receiving cytotoxic anti-cancer treatment may provide prognostic information which
could potentially assist in identifying patients at high risk of developing treatment-adverse events
(AEs) and those who may have poor outcome. Several studies have reported the association of
[18F]FDG uptake in HT with treatment-related AEs and different prognostic markers; however,
unlike cancer imaging HT assessment with [18F]FDG is lacking standardization in the research
setting.
Purpose: The main aim of the thesis project was to identify the prognostic value of HTM based on
[18F]FDG PET/CT in oncologic patients which was investigated through the following objectives:
1) to review the literature for methods used to analyse [18F]FDG uptake in HT for the assessment of
treatment-related AEs in cancer patients.
2) to conduct a preliminary evaluation of commonly used methods for assessment of HTM and
compare selected methods of assessment used for liver and spleen analysis.
3) to assess the test-retest repeatability and interobserver variation in HT using [18F]FDG PET/CT
scans of the thorax in lung cancer patients.
4) to evaluate changes of [18F]FDG uptake in HT during chemotherapy of patients with Hodgkin
lymphoma (HL) and ascertain whether HTM is associated with metabolic tumour volume (MTV),
haematological parameters, haematological AEs, infection, early response to treatment and
progression-free survival (PFS).
Methods:
1) A comprehensive literature search was conducted in PubMed/Medline, Embase (Ovid) and Web
of Science databases to identify published studies from 2000 – 2022 that used [18F]FDG uptake
in HT for assessment of different anti-cancer treatment-related AEs. Reporting AEs had to be
confirmed by the means of clinical symptoms, laboratory or imaging findings. Healthy tissues
included in the search were: heart, thyroid, liver, lungs, spleen, bone marrow (BM), skeletal
muscles, white adipose tissues and organs of the digestive tract.
2) A total of 50 patients were retrospectively identified from the RATHL trial dataset. Three cm
and 6cm diameter volumes of interest (VOI) were placed on the right lobe of the liver and a 3cm
10
diameter VOI was used for spleen which were compared to PET manual segmentation of the
whole liver and spleen. Simple VOIs were also used on ten patients to analyse additional HT. In
addition to VOIs, other methods were also applied in these HT including manual contour for
mediastinal blood pool (MBP), volumetric threshold for myocardium, multiple VOIs for
vertebrae, sectional segmentations for lungs and skeletal muscles, visual analysis of the thyroid
and bowel. For quantifying tissue uptake, the maximum and mean standardized uptake value
(SUVmax and SUVmean) and peak (SUVpeak) if applicable were calculated.
3) A retrospective analysis was performed on 22 patients with non-small cell lung cancer who
underwent [18F]FDG PET/CT of the thorax 2 days apart without receiving treatment. The testretest
repeatability of MBP, left ventricle (LV), BM, lungs and skeletal muscle uptake was
assessed by measuring SUVmax, SUVmean and SUVpeak. Interobserver variation for each
tissue was also assessed by two observers. SUV measurements were analysed using Bland-
Altman plots and variability was expressed using the within-subjects coefficients of variation
(wCV%).
4) Two hundred patients with advanced stage HL were retrospectively included from the RATHL
trial. [18F]FDG PET/CT was performed at baseline and after 2 cycles of chemotherapy for all
patients. Images were analysed by applying VOIs on BM (L3 and T8), spleen, right lobe of the
liver and MBP. Regression models were used for statistical analysis.
Results:
1) A total of 10,626 patients from 85 studies were reviewed with 80(94%) reporting an
association between [18F]FDG uptake and treatment-related AEs. Most studies were
investigating lung-related AEs i.e. radiation pneumonitis, followed by immunotherapyrelated
AEs in multiple organs and lymphoid organ, myocardium, gastrointestinal and
thyroid related AEs. Semi-quantitative analysis was mainly used in the studies, especially
the lung and BM related AEs by applying SUVmean. Assessment of AEs was performed at
baseline or post-therapy in some studies but studies mostly often evaluated changes in HT
uptake from baseline to interim and/or post-therapy scans.
2) In general, SUVmax measurements were higher using larger VOI or contours whereas
SUVmean was lower, however, there was greater variability in SUVmax amongst the
methods than when using SUVmean. SUVpeak could not be assessed with small VOI sizes
or in single slice segmentation. Suggested methods from the literature for myocardial
segmentation were not effective in our cohort. The applied VOIs on different vertebral
levels were comparable and sectional segmentation of skeletal muscle and lungs were
feasible but required a few adjustments.
11
3) The repeatability of SUVmax and SUVpeak were lower than SUVmean in all HT. The wCV
of SUVmean of all HT remained within 20% while it was over 60% for all SUV metrics in
the LV. High interobserver agreement was seen for SUVmean (≈10%) which was better
than SUVmax and SUVpeak in all tissues but with higher variations in LV SUVmean
(≈20%).
4) In HL patients (mean age 37, 106 male), BM and spleen uptake decreased during treatment,
while liver and MBP uptake increased from baseline to PET2. No association was found
between HTM and MTV. High baseline BM uptake was negatively associated with baseline
haemoglobin but positively associated with platelets, neutrophils and white blood cells
(WBC). High liver and MBP uptake was positively associated with haemoglobin. Baseline
BM, liver and spleen uptake was associated with neutropenia at cycles 1-2, but increased
BM uptake at PET2 was associated with lower risk of leukopenia post-cycle 3. Baseline BM
uptake was also associated with failure to achieve an early response to treatment at PET2.
High BM uptake at PET2 in non-responding patients was associated with inferior PFS.
Conclusion:
1) There was an association between [18F]FDG uptake in HT and treatment-related AEs in 94%
of studies included in the review of patients with different cancer types. Lung related AEs
following radiotherapy were widely studied. Most studies relied on semi-quantitative
approaches using SUVmean to correlate changes in HTM with treatment-related AEs.
2) Using different methods for assessment of HTM may result in variability of the SUV
measurements, variation was reduced using SUVmean rather than SUVmax. The application
of simple fixed VOI may be comparable to more sophisticated approaches.
3) HTM is stable in a repeatability setting when no treatment was given and showed good
interobserver agreement. SUVmean measurements were more stable with higher agreement
between observers than SUVmax and SUVpeak. All SUV parameters in LV showed wide
variation between test-retest scans.
4) The clinical study demonstrated significant changes in HTM which was not related to MTV
suggesting a treatment effect induced by chemotherapy rather than a ‘sink effect’
hypothesised by others. The uptake in the BM, spleen and liver was associated with early
haematological AEs. BM uptake in HL could serve as a potential marker of anaemia,
myelosuppression, early AEs and treatment failure and poor outcome in non-responding
patients at PET2.
| Periode | 2023 |
|---|---|
| Begivenhedstitel | Supervisor ved Afnan Malaihs Ph.D-afhandling |
| Begivenhedstype | Andet |