The Utility of 18F-Fluorocholine PET/CT in The Imaging of Parathyroid Adenomas

Aim: to estimate sensitivity of 18F-FCH PET/CT in preoperative localisation of hyperfunctioning parathyroid glands in patients with primary hyperparathyroidism (PHPT). Methods: 65 consecutive patients with PHPT who underwent neck ultrasound (US) and 99mTc/99mTc-MIBI dual-phase parathyroid scintigraphy were prospectively enrolled. Twenty-two patients had unsuccessful parathyroid surgery prior to the study. PET/CT scans were performed 65.0 ± 13.3 min after injection of 218.5 ± 31.9 MBq of 18F-uorocholine (FCH). Three experienced nuclear medicine physicians assessed the detection rate of hyperfunctioning parathyroid tissue. Response to parathyroidectomy and clinical follow-up served as a reference test. Per-patient sensitivity and positive predictive value (PPV) were calculated for patients who underwent surgery. Results: 18F-FCH PET/CT was positive in 61 patients, and negative in four. US and parathyroid scintigraphy showed positive and negative results in 20 , 45 and 17, 48, respectively. US showed nodular goiter in 31 patients and chronic thyroiditis in 9 patients. Parathyroid surgery was performed in 43 (66%) patients. 18F-FCH PET/CT yielded sensitivity of 100% (95% CI [87.99-100]) and PPV of 85.7% (95% CI [70.77-94.06]). Similar values were observed in patients with chronic thyroiditis, nodular goiter, and patients after an unsuccessful parathyroid surgery. PET/CT identied hyperparathyroidism complications (kidney stones, osteoporotic bone fractures and brown tumours) in 11 patients. Conclusions: 18F-FCH PET/CT effectively detected hyperfunctioning parathyroid tissue and its complications. The method showed excellent sensitivity and positive predictive value, including patients with nodular goiter, chronic thyroiditis and prior unsuccessful parathyroidectomy. PET/CT performance was superior to neck ultrasound and parathyroid scintigraphy.


Introduction
Primary hyperparathyroidism (PHPT) is an endocrine disease where excessive amounts of parathormone (PTH) are produced and secreted autonomously by parathyroid glands. It is most commonly caused by a solitary parathyroid adenoma (80-85%), followed by glandular hyperplasia (10-15%) and double adenoma (4%) [1][2][3][4]. Surgical treatment is the only curative therapy. Image-guided minimally invasive parathyroidectomy (MIP) should be the surgical technique of choice as it offers several advantages over traditional bilateral neck explorationapproach. MIP is associated with shorter operation time (local anaesthesia might be used), fewer complications, faster recovery, better cosmetic outcome and lower incidence and severity of symptomatic and biochemical hypocalcaemia. At the same time, it shows similar e cacy to the classic surgery [2,5]. However, for minimally invasive approach a precise localisation is required.
Several imaging modalities have been proposed for preoperative detection and localisation of hyperfunctioning parathyroid glands. Neck ultrasound (US) and 99mTc-methoxy-isobutyl-isonitrile (99mTc-MIBI) scintigraphy are currently the most frequently used methods [6]. US, the most readily available and least expensive one, shows rather limited sensitivity varying from 57% to 76%. It also fails to detect ectopic parathyroid adenomas [7]. Ectopic parathyroid tissue can be, on the other hand, detected in scintigraphy, which is reported to have sensitivity of about 60%. It might be increased to 92% when single photon emission computed tomography/computed tomography (SPECT/CT) is performed [8][9][10].
Currently, it is considered the method of choice for detection of parathyroid adenomas [2,4].
Contemporary research suggests that 18F-uorocholine (FCH) positron emission tomography/computed tomography (PET/CT) might be a feasible tool in this clinical setting [6]. Fluorocholine is a phospholipid analogue that is integrated into the newly synthesized membranes of proliferating cells. Choline uptake increases with the upregulation of choline kinase. Furthermore, its upregulation is linked to PTH secretion [11]. 18F-FCH PET/CT, like scintigraphy, allows for detection of ectopic parathyroid adenomas but its superior spatial resolution is an additional advantage, especially in small lesions. It also exerts lower radiation than scintigraphy and has shorter scanning time (20 min vs 120 min). [6,12].
The aim of the study was to assess the feasibility of 18F-FCH PET/CT in preoperative localisation of hyperfunctioning parathyroid tissue in patients with PHPT.

Patients
Between January 2018 and November 2019 sixty-ve consecutive patients with primary hyperparathyroidism referred to the A dea Mazovian PET/CT Centre in Warsaw, Poland for 18F-FCH PET/CT were prospectively enrolled in the study ( Table 1). The patients were referred by a single centrethe Department of Endocrinology and Radioisotope Therapy in Military Institute of Medicine in Warsaw. The eligibility criteria were: (1) PHPT and (2) neck US and the dual-phase and subtraction 99mTc/99mTc-MIBI SPECT/CT scintigraphy performed prior to the PET/CT scan (both were done in accordance with the recommendations of the European Association of Nuclear Medicine) [9].
All patients signed a written informed consent form. The study was approved by the Military Institute of Medicine Ethical Committee (44/WIM/2016). The study was carried out in accordance with the Declaration of Helsinki and its later amendments. Registered at ClinicalTrials.gov website (NCT04570033; 30/09/2020) Imaging protocol PET/CT imaging was performed using a Discovery 710 PET/CT system (GE Healthcare, Chicago, Illinois, US). A scout view and a non-contrast-enhanced low-dose spiral 64-slice CT scan was performed prior to PET scan for attenuation correction and anatomic localization.
The CT scan was performed with a tube voltage of 140 kV in the helical mode with a Smart/Auto mA (range: 40-120 mA); pitch and table speed were as follows: 0.984:1 and 39.37 mm/rot; the helical thickness was 3.75 mm; X-ray tube rotation time was 0.6 s; the slice thickness was 1.25 mm. The GE Adaptive Statistical Iterative Reconstruction with the level of 20% was used to reduce patient radiation dose from CT scans.
Immediately after CT scanning mid-skull to mid-thigh three-dimensional PET was acquired. The acquisition time was 1 min 45 sec for each bed position (the axial eld of view was 15.7 cm with 23% bed overlap). The emission data was corrected for geometrical response and detector e ciency (normalisation) as well as for system dead time, random coincidences, scatter and attenuation. For attenuation corrected images reconstruction was conducted with 3D iterative algorithm with time of ight PET reconstruction and resolution recovery algorithm with 3 iterations/18 subsets and a lter cut-off of 5.5 mm. The matrix size was 256x256.
Image analysis PET/CT images were analysed with GE Healthcare Advantage Workstation 8.0 (Chicago, Illinois, United States) by three experienced nuclear medicine physicians. The reviewers did not know the results of neck US and scintigraphy the patients had performed prior to PET. We reported PET/CT results as positive and negative (US and SPECT/CT results were divided alike). 18F-FCH PET/CT scans were considered positive for hyperfunctioning parathyroid tissue if at least one tracer-avid focus was identi ed with a corresponding lesion in CT. All disagreements were resolved by consensus. We also recorded the number of discordant results.
Surgery was performed on the basis of PET/CT results. PET/CT ndings were correlated with parathormone levels during and after the operation. Surgery was regarded successful if intraoperative parathormone decreased ≥50% and, within a 6-month follow-up, PTH and calcium levels were normal.
Statistical analysis SPSS Statistics 25 software (Armonk, New York, United States) was used for statistical analysis. A p value <0.05 was considered signi cant. Descriptive analysis was performed by calculating mean, median, standard deviation and range. The data was tested for normality using the Shapiro-Wilk test. Chi 2 test was used to determine if the difference between positive and negative results of PET/CT, US and 99mTc-MIBI SPECT/CT was signi cant. Diagnostic e cacy of 18F-FCH PET/CT was measured for the general studied population as well as for the subgroups of patients with nodular goiter, chronic thyroiditis and patients with a history of parathyroid surgery prior to the study.

Results
No adverse effects were observed after the injections of 18F-FCH. Patients did not report any alarming symptoms.
During the follow-up period 43 (66%) patients had parathyroid surgery (all operations were minimally invasive). All but one operated patients were PET-positive). The patient who was operated despite negative PET/CT also had a negative 99mTc-MIBI SPECT/CT, equivocal neck US and nodular goiter. We observed 6 more cases of unsuccessful surgery in PET-positive patients (FP). PET/CT was TP in 36 of patients who underwent surgery. We did not observe FN results. Sensitivity and positive predictive value (PPV) were 100% and 85.7%, respectively.
We also calculated sensitivity and PPV in subgroups with chronic thyroiditis, nodular goiter and a history of unsuccessful parathyroid surgery ( Table 2).
18F-FCH PET/CT detected 70 choline-avid lesions; seven (10%) of them were equivocal and the decision whether they could be reported as parathyroid adenomas was reached via consensus between the readers. Neck ultrasound showed 22 lesions, including 11 (50%) equivocal ndings (resolved via consensus). In 99mTc-MIBI SPECT/CT 17 lesions were identi ed and 6 (35%) of them were equivocal (resolved via consensus).
We have also veri ed if the lesions visualised in PET/CT, US and SPECT/CT were concordant. From 17 lesions reported in SPECT/CT 14 were visible in PET/CT which yielded 82% accordance. From 22 lesions reported in US PET/CT showed 16 which yielded 75% accordance.
On a per-patient basis, the three methods were concordant (triple positive or triple negative) in 9% and 3% of patients. Conventional imaging was either discordant or negative in 91% of patients (n=59). 18F-FCH PET/CT was positive in 93% of them (Table 3).
PET/CT also identi ed hyperparathyroidism complications (kidney stones, osteoporotic bone fractures and brown tumours) in 17% of patients (Figure 2).
Almost a third of subjects had unsuccessful parathyroidectomy prior to the study. After 18F-FCH PET/CT they were reoperated and surgery resulted in a positive outcome in 85% of them (Figure 3).

Discussion
In our study we performed 18F-FCH PET/CT in patients with primary hyperparathyroidism who had already had US and scintigraphy. The following surgical treatment was based on PET/CT resultscholine-avid lesions were excised.
In our paper we showed that 18F-FCH PET/CT is a feasible method for localisation of parathyroid adenomas. It presents excellent sensitivity and positive predictive value, even in patients with chronic thyroiditis, nodular goiter or with a history of unsuccessful parathyroid surgery. PET/CT also showed superiority over conventional imaging (neck ultrasound and 99mTc-MIBI SPECT/CT).
These results go in line with contemporary publications. Systematic review and meta-analysis recently published by Treglia et al. con rms 18F-FCH PET/CT as an outstanding diagnostic tool for the detection of hyperfunctioning parathyroid tissue. In the analysis of 14 articles (517 patients) the authors report PET/CT to have per-patient sensitivity of 95% (ranging from 85% to 100%), PPV of 97% (86% to 100%) and detection rate of 91% (76% to 100%). A subgroup analysis including only research with PHPT yielded pooled sensitivity, PPV and detection rate of 96%, 96% and 92%, respectively [13]. It has been proved (both in our study and in the recently published research) that 18F-FCH PET/CT is similarly effective in localising hyperfunctioning parathyroid tissue in patients with nodular goiter or with prior unsuccessful surgery [9,13,14]. In our paper we report that PET/CT also maintains its excellent performance in patients with chronic thyroiditis.
18F-FCH PET/CT is useful in detection and localisation of hyperfunctioning parathyroid tissue in patients with negative, equivocal or discordant US and scintigraphy ndings. The APACH1 study (including 25 patients with PHPT and negative or inconclusive conventional imaging) yielded sensitivity of 90.5% and PPV of 86.4% [6]. Fischli et al. and Grimaldi et al. reported that PET/CT detection rate in this clinical setting reaches 91% and 89%, respectively [15,16]. Treglia et al. have concluded that 18F-FCH PET/CT was able to detect hyperfunctioning parathyroid tissue in a high percentage (72-91%) of patients with negative or inconclusive 99mTc-MIBI SPECT/CT [13]. In our study 91% of patients were either negative both on US and 99mTc-MIBI SPECT/CT or had discordant results of conventional imaging. PET/CT detection rate in this group was 93%.
It is important to detect complications of hyperparathyroidism because it often quickens the decision of surgery. Current PET/CT protocols may miss them as they cover only neck and upper chest whilst the most common complications-kidney stones, lumbar spine fractures and pancreatic calci cationslocalise in the abdominal area [13]. Hence, in our study we decided to perform mid-skull to mid-thigh scans. We believe this extended protocol should be the standard procedure in the parathyroid PET/CT imaging.
The lack of histopathological validation of the PET/CT results may be a limitation in this study. However, we believe that biochemical criteria-normalisation of calcium and parathormone levels-are a more reliable reference test. Some factors such as surgeon's experience and distorted neck anatomy (in patients after parathyroidectomy) may still affect it-choline-avid lesions may be overlooked during surgery because of scar tissue (and PET/CT result would be regarded as false positive). Another potential limitation is that only 2/3 of PET-positive patients had surgery prior to the statistical analysis meaning only 2/3 of PET/CT results were veri ed. The remaining 1/3 of patients was awaiting surgical treatment at the time of study closure (waiting time for parathyroid surgery in Poland is currently up to two years).  Table 2. Sensitivity, speci city, positive predictive value of 18F-FCH in different subgroups of patients. 95% con dence intervals are given in square brackets.    PET/CT and surgery results of a subgroup of patients who had unsuccessful parathyroidectomy prior to our study.