Introduction
The estrogen receptor (ER) is an important target for endocrine treatment in breast cancer patients, and ER expression of the tumor is the main indication to start antihormonal treatment, as success rates heavily rely on ER status [
1,
2]. Although specificity and sensitivity of immunohistochemistry to assess ER expression are high, it is not always feasible to obtain a suitable biopsy. Moreover, ER expression can change over time in the metastatic setting and vary between the primary tumor and its metastases and between metastases within a single patient [
3].
Non-invasive molecular imaging of the ER using positron emission tomography (PET) with 16α-[
18F]fluoro-17β-estradiol (FES) has been found useful to detect the estrogen receptor status of the primary tumor and its metastases. FES-PET has been used in several imaging studies in breast cancer patients to visualize all metastases in a patient to assess tumor heterogeneity [
4‐
13]. FES-PET has a high predictive value with a sensitivity of 85% and a specificity of 98% [
14]. The uptake of FES differs per tissue type and anatomic site and can be influenced by intrinsic (i.e., menopausal status) and extrinsic factors (i.e., hormone therapy) [
6,
9]. Recently, recommendations for the use of FES-PET, including the indications, correct patient preparation, scan acquisition, and analysis of the scans, were published [
15].
The use of FES-PET in daily clinical practice, in patients with clinical dilemmas and for the detection of lesions expressing ER as input for treatment decisions, is growing. Therefore, it is important to gain more insight in the potential pitfalls that are associated with this imaging technique. In the University Medical Center Groningen, extensive experience is available with FES-PET scans in both a research and clinical setting [
10,
11,
14,
15]. In a considerable number of FES-PET scans, heterogeneous uptake in the lungs was noticed, without the presence of an oncologic substrate on the accompanying (contrast-enhanced) CT scan. As this enhanced uptake was seen in the lungs and most patients were irradiated in the thoracic area, we hypothesized that pulmonary fibrosis as a result of earlier radiation therapy might be the cause of this FES uptake.
Since radiation therapy is one of the most frequently administered treatments in patients with breast cancer, and FES-PET is performed more and more in daily clinical practice, it is important for the interpretation of the scans to assess whether radiotherapy leads to enhanced FES uptake in the lungs.
Therefore, the aim of this descriptive study was to evaluate whether radiation therapy in the thoracic area is possibly related to enhanced pulmonary, non-tumor FES uptake.
Methods
In this descriptive, single-center study, we retrospectively analyzed all FES-PET/CT scans that were performed for clinical purposes in our institution from 2008 to 2017. Information on irradiation was retrieved from the patient charts. Scans from patients who had received irradiation in the thoracic area prior to the scan were compared to scans of patients who had never received irradiation in the thoracic area. The medical history and radiation therapy schemes were retrieved from the electronic patient files. Given the retrospective descriptive nature of this study, national legislation does not require medical ethical approval; however, the local database registering patient objection has been checked for patients objecting against using their material.
FES-PET
FES was produced in the University Medical Center Groningen by a two-step method that was extensively described previously [
13]. In short,
18F-fluoride is prepared with a cyclotron by irradiation of
18O-water according to the nuclear reaction
18O(p,n)
18F. The cyclotron-produced
18F-fluoride is allowed to react with 3-O-methoxymethyl-16,17-O-sulfuryl-16-epiestriol (ABX, Germany), followed by removal of the MOM protecting group and the sulfate group by hydrolysis with hydrochloric acid. After HPLC purification, the product is formulated in 10% ethanol in saline and sterilized by filtration [
13]. FES with > 99% radiochemical purity was obtained in a practical yield of 3.4 ± 1.5 GBq. FES had a specific activity of 325 ± 274 GBq/μmol. Approximately 200 MBq of FES was injected intravenously. Whole body emission scans were performed approximately 60 min after tracer injection. PET images were obtained from skull base to mid-thigh with a Siemens 40 or 64 slice mCT (PET/CT) Biograph camera system (Siemens Medical Systems, Knoxville, TN, USA). A low-dose CT scan was performed in all patients for attenuation correction. Attenuation-corrected images were visually analyzed for enhanced non-tumor uptake. To calculate the uptake, a volume of interest (VOI) was drawn over the area of enhanced non-tumor uptake and the maximum standardized uptake value (SUV
max) and the average SUV (SUV
mean) using a 50% isocontour of the hottest pixel were measured using syngo.via software. In patients without visual enhanced non-tumor uptake, a VOI was drawn centrally in the lung (including the basal parts) for the same measurements. All scan acquisitions and calculations were performed according to EANM/EARL guidelines for
18F imaging [
16].
CT scan
All patients had a low-dose CT scan for attenuation correction at the time of FES PET. Part of the patients also had a contrast-enhanced CT scan within 6 weeks of the FES PET of the thorax available when this was clinically indicated. All CT scans were evaluated for fibrosis, or oncological substrates, by an experienced radiologist (MD). There are many features that may imply pulmonary fibrosis, such as honeycombing, traction bronchiectasis, lung architectural distortion, and reticulation. In case of radiation-induced pulmonary fibrosis, also other features may occur such as volume loss, linear scarring, chronic consolidation, mediastinal shift, and pleural thickening. All scans were checked for these features.
Radiation schedules and dose
The patients who were irradiated received variable radiation schedules, depending on the indication and available techniques at the time of radiation therapy. To analyze the effect of different radiation doses and schemes on enhanced non-tumor FES uptake, FES-PET scans were fused with the original radiation therapy planning CT scans, including the radiation fields and doses, using Raystation software. Radiation doses were determined by drawing a VOI in the radiation field in the area with enhanced FES uptake. In patients that were irradiated but did not show enhanced uptake in the lungs, a VOI was drawn in the lungs, in the same region as for the SUV calculation.
Statistics
The main outcome was the presence of enhanced non-tumor FES uptake, defined as visually increased FES uptake above background in the absence of an oncologic substrate on the concordant (low-dose or contrast-enhanced) CT scan. Correlations between enhanced FES uptake and radiation dose and between interval time between radiation therapy and FES PET scan were calculated using a Pearson correlation test. One-way ANOVA was used to analyze the statistical significance between group differences. A probability (p) value < 0.05 was considered statistically significant.
Discussion
In the present study, we found enhanced non-tumor pulmonary FES uptake in a subset of patients, most frequently after radiation therapy in the thoracic area. Uptake of FES is considered to be ER specific, and the cause of this non-tumor uptake is not fully elucidated yet. However, this study supports a possible fibrosis-related origin. This aspect of non-tumor FES uptake on FES-PET has not been described before, and this is the largest series so far to allow hypothesis generation with regard to this aspect.
One possible cause of the enhanced tracer uptake is that the tracer binds to inflammation-related ERβ expression. Two isoforms of ER exist, α and β, and despite the fact that
18F-FES has a 6.3 times higher affinity for ERα compared to ERβ [
12], uptake can be seen in ERβ-driven pathology [
17]. Under normal conditions, low levels of ERβ are present in ovaries, the kidney, the brain, bone, the heart, the lungs, intestinal mucosa, the prostate, the immune system, and endothelial cells [
18]. Also, in patients with interstitial pneumonia and cystic fibrosis, ERβ expression is higher than in healthy lung tissue [
19,
20]. Both cystic fibrosis and interstitial pneumonia are marked by lung fibrosis and inflammation.
Both ERβ and ERα play a role in inflammation and fibrosis. Estrogen-dependent ERα activation is required for normal development of the dendritic cells [
21] and high levels of dendritic cells are present in patients with lung fibrosis [
22]. During inflammation, dendritic cells are activated to initiate and coordinate immune responses. We observed fibrosis or post-radiation inflammation in most patients with enhanced non-tumor FES uptake, but not in all. This could be explained by the timing of the CT scans. Fibrosis may not yet be detectable on a CT scan in an early stage of the formation of fibrosis. Exposure to radiation therapy could lead to side effects, largely depending on the anatomic site and dose received [
23].
The pathogenesis of radiation-induced side effects is not fully understood but seems mostly related to extended inflammatory effects. As part of the inflammatory process, fibrosis may occur several weeks after radiation therapy [
24]. The late phase typically occurs between 6 and 12 months and can continue to progress up to several years [
25]. In 23 out of the 48 patients, enhanced uptake was seen bilaterally, which was beyond the boundaries of the radiation field. It has been reported, both preclinically and clinically, that bilateral radiation therapy toxicity may occur [
26‐
29]. This suggests that enhanced FES uptake may be associated with a (late) inflammatory event caused by irradiation, also outside the irradiation field. Not in all patients, a contrast-enhanced CT scan was available, and due to the lower image quality of the low-dose CT, small areas of fibrosis could be missed.
Not all fibrosis in patients is related to radiation therapy. Extensive literature exists on lung toxicity due to several systemic treatments. With the wide time interval between irradiation and FES-PET treatment types, as well as treatment regimens and doses have changed over the years. With the retrospective design of the current study, we were unable to establish other correlations between fibrosis and FES uptake.
Another explanation for enhanced uptake in irradiated lungs is that radiation results into leakage of the blood vessels, possibly leading to extravasation of FES. In a preclinical rat model, radiation of the lungs showed vascular damage early after irradiation and remodeling leading to increased permeability, perivascular edema, and vascular remodeling [
29,
30]. As a compensatory effect, the blood pressure, blood flow, and thereby shear stress may increase in the vasculature in the non-irradiated part of the lungs. This increase of shear stress may then lead to damage to the non-irradiated vasculature [
30] and potentially explain leakage of the tracer in surrounding tissue. Though unbound FES can readily permeate the endothelium, most FES is bound to the sex hormone-binding globulin (SHBG) which, in case of leaky vessels, may also leak out.
FES-PET scans are increasingly used, both in a research and a clinical setting. The scans are often qualitatively assessed and lesions are identified as ER-positive if the tracer uptake is above the background signal. Therefore, it is important for the analysis of the scans to know that non-tumor uptake in the lungs may occur and that this finding should not be interpreted as pathological uptake. Also, existing lesions in the radiation field may potentially be non-evaluable in cases where the background signal is increased due to the uptake after radiation treatment. Furthermore, to facilitate the interpretation of FES-PET scans, semi-quantitative analysis can be performed and correction for physiologic background uptake is often applied when calculating SUV using the unaffected contralateral site or surrounding tissue of the same origin. In such cases, one should keep in mind that background activity in the reference region can be influenced by radiation therapy and consequently background correction may cause an underestimation of the tracer uptake in the lesion.
Despite the limitations of being a retrospective study over a long period of time, this is the most comprehensive series of patients receiving FES PET scans after radiation therapy described so far. The clinical significance of these findings has to be further investigated, e.g., the relation between the lung function of the patients and enhanced uptake. These data were not available in our patient charts. As such, the findings described here should be regarded as hypothesis generating and should preferably be confirmed in larger, prospective studies.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.