A Novel Melanin-Targeted ¹⁸F-PFPN Positron Emission Tomography Imaging for Diagnosing Ocular and Orbital Melanoma

Abstract

Objective

¹⁸F-N-(2-(Diethylamino)ethyl)-5-(2-(2-(2-fluoroethoxy)ethoxy)ethoxy) picolinamide (¹⁸F-PFPN) is a novel positron emission tomography (PET) probe designed to specifically targets melanin. This study aimed to evaluate the diagnostic feasibility of ¹⁸F-PFPN in patients with ocular or orbital melanoma.

Materials and Methods

Three patients with pathologically confirmed ocular or orbital melanoma (one male, two females; age 41–59 years) were retrospectively reviewed. Each patient underwent comprehensive ¹⁸F-PFPN and ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) PET scans. The maximum standardized uptake value (SUV_max) of the lesion and the interference caused by background tissue were compared between ¹⁸F-PFPN and ¹⁸F-FDG PET imaging. In addition, the effect of intrinsic pigments in the uvea and retina on the interpretation of the results was examined. The contralateral non-tumorous eye of each patient served as a control.

Results

All primary tumors (3/3) were detected using ¹⁸F-PFPN PET, while only two primary tumors were detected using ¹⁸F-FDG PET. Within each lesion, the SUV_max of ¹⁸F-PFPN was 2.6 to 8.3 times higher than that of ¹⁸F-FDG. Regarding the quality of PET imaging, the physiological uptake of ¹⁸F-FDG PET in the brain and periocular tissues limited the imaging of tumors. However, ¹⁸F-PFPN PET minimized this interference. Notably, intrinsic pigments in the uvea and retina did not cause abnormal concentrations of ¹⁸F-PFPN, as no anomalous uptake of ¹⁸F-PFPN was detected in the healthy contralateral eyes.

Conclusion

Compared to ¹⁸F-FDG, ¹⁸F-PFPN demonstrated higher detection rates for ocular and orbital melanomas with minimal interference from surrounding tissues. This suggests that ¹⁸F-PFPN could be a promising clinical diagnostic tool for distinguishing malignant melanoma from benign pigmentation in ocular and orbital melanomas.

INTRODUCTION

Melanoma is a malignant neoplasm originating from melanocytes and has a high mortality rate [1]. Ocular and orbital melanomas, the second most common type of melanoma following cutaneous melanomas [2], can be less intuitive to identify because they are often located within or behind the eyeball. Traditional imaging modalities, including computed tomography (CT), magnetic resonance imaging (MRI), ultrasound (US), and certain ocular examinations, such as fundus photography, fundus angiography, and optical coherence tomography, have been utilized to detect these tumors [3,4]. However, these methods have limitations. Not all tumors are easily visible in these images, as some small ocular melanomas and ocular nevi often appear similarly [5]. Furthermore, site-specific imaging does not provide comprehensive views of the entire body.

Positron emission tomography (PET) using ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) is the predominant whole-body imaging modality [6]. In clinical practice, ¹⁸F-FDG PET has proven useful for identifying [7,8], staging [9], and predicting the prognosis [10] of patients with advanced (stage III-IV) or medium-to-large uveal melanomas (UMs). However, it has insufficient sensitivity for detecting stages I–II and small UMs [8]. Additionally, ¹⁸F-FDG PET imaging may be limited by uptake in periocular tissues, such as the brain [11], which can affect the detection of ocular and orbital tumors.

¹⁸F-N-(2-(Diethylamino)ethyl)-5-(2-(2-(2-fluoroethoxy)ethoxy)ethoxy) picolinamide (¹⁸F-PFPN) is an emerging melanin-specific PET tracer targets melanin in tumors [12]. Given that over 90% of melanomas contain melanin [13], the potential applications of ¹⁸F-PFPN are substantial. Recent studies have highlighted the safety and tolerability of ¹⁸F-PFPN in humans [14]. Compared to ¹⁸F-FDG, ¹⁸F-PFPN demonstrated higher detection rates in skin and rectal melanomas and could pinpoint metastatic lesions smaller than 2 mm [14]. To date, there have been no systematic studies on ¹⁸F-PFPN in ocular and orbital melanomas. This study aimed to investigate the diagnostic feasibility of ¹⁸F-PFPN in patients with ocular or orbital melanoma and to examine the effect of intrinsic pigments in the uvea and retina on ¹⁸F-PFPN imaging in the contralateral healthy eyes of these patients.

MATERIALS AND METHODS

Patients

This retrospective analysis included selected patients from a trial (ClinicalTrials.gov number: NCT04747561) approved by the local Institutional Ethics Committee (IRB No. 2021-0016). All the participants provided written informed consent. From August 2022 to September 2023, three patients (one male and two females; age 41–59 years) with pathologically confirmed ocular or orbital melanoma at our hospital were retrospectively analyzed. Demographic data, clinical presentation, tumor location, and size, ¹⁸F-PFPN and ¹⁸F-FDG PET results, and other imaging findings, including CT, MRI, US, fundus photography, and fundus fluorescein angiography, were collected.

Image Acquisition

The patients underwent both ¹⁸F-PFPN PET/MRI and ¹⁸F-FDG PET/CT imaging sessions within a brief interval (no more than one week) of each other. All scans covered the entire body, from the brain to the foot.

¹⁸F-PFPN PET/MR Imaging

¹⁸F-PFPN, a molecular probe that specifically binds to melanin, was synthesized as described previously [12]. ¹⁸F-PFPN PET/MRI was performed using a Signa PET/MR scanner (SIGNA™ PET/MR; GE Healthcare, Chicago, IL, USA). Each patient received an intravenous injection of ¹⁸F-PFPN at a dose of 3.0–5.4 MBq/kg, followed by whole-body PET/MR imaging after a two-hour rest period. Axial T1-weighted MR was performed using the LAVA-FLEX sequence, while coronal T2-weighted MR was performed using the fast-recovery fast spin-echo sequence. PET image reconstruction utilizes the ordered subset maximization iterative method with a combined time-of-flight technique and a point diffusion function comprising two iterations and 28 subsets, followed by Gaussian filtering (3 mm) for image processing.

¹⁸F-FDG PET/CT Imaging

PET/CT imaging was conducted using a PET/CT scanner (Discovery VCT^®; GE Healthcare, Milwaukee, WI, USA). Patients underwent a minimum six-hour fasting period, and ¹⁸F-FDG was administered intravenously at a dose of 3.7–5.4 MBq/kg after ensuring that their blood glucose was below 11.1 mmol/L. Subsequently, PET/CT imaging was performed after sixty minutes of calm rest. CT images were obtained with parameters set at 120 kV, 80 mA, and a thickness layer of 3.75 mm. PET image reconstruction was performed using the ordered subset expectation maximization iterative method with a reconstruction matrix of 512 × 512 and a thickness of 3 mm.

Image Interpretation

The lesion volumes of interest were mapped on PET/MRI and PET/CT fusion images using the three-dimensional outlining method on the Advantage Workstation version AW4.6 (GE Healthcare, Milwaukee, WI, USA). The system automatically calculated the maximum standardized uptake value (SUV_max) of the lesions. The raw SUV_max was normalized using the method described by Zhang et al. [14].

Normalized SUV_max=raw SUV_max/SUV_bkgd

Where SUV_bkgd refers to the mean SUV_max of the descending aorta.

The SUV_max of ¹⁸F-PFPN and ¹⁸F-FDG within lesions was quantitatively compared, using pathological findings as the basis for the final diagnosis. Additionally, the effect of intrinsic pigments in the uvea and retina on the interpretation of the results was examined, with the contralateral non-tumorous eye of each patient serving as a control.

RESULTS

This study evaluated three patients who underwent multimodal imaging. The clinical manifestations and imaging findings of each patient are shown in Table 1. The participants included three types of patients with melanomas occurring in different anatomical locations of the eye. Patient #1 was a 59-year-old female with melanoma of the eyelid and conjunctiva (Supplementary Fig. 1); patient #2 was a 49-year-old female with orbital melanoma (Supplementary Fig. 2); and patient #3 was a 41-year-old male with choroidal melanoma (Supplementary Fig. 3).

Table 1. Clinical presentations and imaging findings of patients with ocular and orbital melanoma.

		Patient #1	Patient #2	Patient #3
Sex		Female	Female	Male
Age, yr		59	49	41
Clinical features		Pigmented mass in the right eye	Swelling in the right eye	A dark shadow covered the right eye
Tumor
	Location	Eyelid and conjunctiva	Orbit	Choroid
	Size, mm	13 × 5 × 5	33 × 27	11 × 6
SUVmax of the primary tumor
	18F-PFPN	16.6	36.2	10.6
	18F-FDG	2.0	9.2	4.1
	18F-PFPN:18F-FDG	8.3:1.0	3.9:1.0	2.6:1.0
Other imaging methods and findings
	MR*	High T1WI and high T2WI	High T1WI and low T2WI	High T1WI and low T2WI
	CT	Nodular enhancement signal	Nodular mixed signal	High-density signal with significant enhancement
	US	-	-	Dome-shaped mass
	FP	-	-	Caesious local solid bulge
	FFA	-	-	Punctate hyper fluorescence in filling phases

^*MR protocols: axial T1-weighted MR was conducted using the LAVA-FLEX sequence, and coronal T2-weighted MR was performed using the fast-recovery fast spin-echo sequence.

MR = magnetic resonance, SUV_max = maximum standardized uptake value, ¹⁸F-PFPN = ¹⁸F-N-(2-(Diethylamino)ethyl)-5-(2-(2-(2-fluoroethoxy)ethoxy)ethoxy)picolinamide, ¹⁸F-FDG = ¹⁸F-fluorodeoxyglucose, T1WI = T1-weighted image, T2WI = T2-weighted image, CT = computed tomography, US = ultrasound, FP = fundus photography, FFA = fundus fluorescein angiography

Our results showed that ¹⁸F-PFPN had a higher detection rate than ¹⁸F-FDG in these melanomas. This enhanced performance is evident from two primary observations. Firstly, abnormal accumulation of ¹⁸F-PFPN was evident in the primary tumors of all participants (3/3). In contrast, only two participants (2/3) showed high uptake levels using ¹⁸F-FDG (patient #1: Fig. 1, patient #2: Fig. 2, patient #3: Fig. 3). One tumor lesion was missed when using ¹⁸F-FDG, likely because of the small size of the tumor. Secondly, the SUV_max of ¹⁸F-PFPN consistently exceeded that of ¹⁸F-FDG within the same patient, ranging from 2.6 to 8.3 times greater.

Fig. 1. A 59-year-old female with melanoma of the eyelid and conjunctiva. A: In ¹⁸F-PFPN PET images, the tumor lesion (red box and arrows) can be clearly visualized. The SUV_max of ¹⁸F-PFPN at the lesion was 16.6. MR imaging demonstrated an abnormal signal shadow in the lower lid of the right eye. B: In ¹⁸F-FDG PET images, the brain and periocular tissues interfered with the visualization of the tumor, which made it difficult to identify the tumor (green box and arrows). The SUV_max of ¹⁸F-FDG at the lesion was 2.0. A thickened lesion site could be seen in the CT image. ¹⁸F-PFPN = ¹⁸F-N-(2-(Diethylamino)ethyl)-5-(2-(2-(2-fluoroethoxy)ethoxy)ethoxy) picolinamide, PET = positron emission tomography, SUV_max = maximum standardized uptake value, MR = magnetic resonance, ¹⁸F-FDG = ¹⁸F-fluorodeoxyglucose, CT = computed tomography, FLAIR = fluid-attenuated inversion recovery.

Fig. 2. A 49-year-old female with orbital melanoma. A: In ¹⁸F-PFPN PET images, showing the upper portion of the lesion and its relationship to the ocular structures (the entire extent of the lesion is presented in Supplementary Fig. 2), the tumor lesion (red box and arrows) can be clearly visualized. The SUV_max of ¹⁸F-PFPN at the lesion was up to 36.2. MR imaging demonstrated an occupation within the right orbit with a high signal on T1-weighted images and a low signal on T2-weighted images. The occupation pushed against the rectus externus, rectus internus, and optic nerve, and breached the adjacent great wing of the sphenoid bone. B: In ¹⁸F-FDG PET images, the brain and periocular tissues interfered with the visualization of the tumor, which made it difficult to identify the tumor (green box and arrows). The SUV_max of ¹⁸F-FDG at the lesion was 9.2. CT demonstrated an abnormal soft tissue density shadow within the right orbit. ¹⁸F-PFPN = ¹⁸F-N-(2-(Diethylamino)ethyl)-5-(2-(2-(2-fluoroethoxy)ethoxy)ethoxy) picolinamide, PET = positron emission tomography, SUV_max = maximum standardized uptake value, MR = magnetic resonance, ¹⁸F-FDG = ¹⁸F-fluorodeoxyglucose, CT = computed tomography.

Fig. 3. A 41-year-old male with choroidal melanoma. A: In ¹⁸F-PFPN PET images, the tumor lesion (red box and arrows) can be clearly visualized. The SUV_max of ¹⁸F-PFPN at the lesion was 10.6. MR imaging demonstrated an abnormal signal shadow in the lateral wall within the right eyeball, with a high signal on T1-weighted images and a low signal on T2-weighted images. B: In ¹⁸F-FDG PET images, the brain and periocular tissues interfered with the visualization of the tumor, which made it difficult to identify the tumor (green box and arrows). The SUV_max of ¹⁸F-FDG at the lesion was 4.1. CT demonstrated an abnormal density shadow in the lateral wall within the right eyeball. ¹⁸F-PFPN = ¹⁸F-N-(2-(Diethylamino)ethyl)-5-(2-(2-(2-fluoroethoxy)ethoxy)ethoxy) picolinamide, PET = positron emission tomography, SUV_max = maximum standardized uptake value, MR = magnetic resonance, ¹⁸F-FDG = ¹⁸F-fluorodeoxyglucose, CT = computed tomography.

Subsequently, we compared the imaging quality of ¹⁸F-PFPN and ¹⁸F-FDG PET scans. The results showed that ¹⁸F-PFPN PET provided a clearer visualization of lesions owing to the low background uptake. In contrast, ¹⁸F-FDG PET images showed interference in the visualization of ocular and orbital melanomas because of physiological uptake in the brain and periocular tissues (Figs. 1B, 2B, 3B: green boxes). This interference was reduced in ¹⁸F-PFPN PET images (Figs. 1A, 2A, 3A: red boxes). Moreover, neither ¹⁸F-PFPN nor ¹⁸F-FDG scans showed abnormal concentrations in the contralateral non-tumor eye of any participant, which is in line with the self-referential comparison.

No metastases were detected in any patient. Patient #2 underwent excision of an intraorbital mass; however, follow-up scans using both ¹⁸F-FDG and ¹⁸F-PFPN PET revealed tumor residue in the orbital cavity.

DISCUSSION

This study describes the clinical presentations and imaging characteristics of a cohort of patients diagnosed with ocular or orbital melanomas. Furthermore, we compared the diagnostic efficacy of ¹⁸F-PFPN with ¹⁸F-FDG in these tumors. We preferred to combine MR with PET, especially when using ¹⁸F-PFPN as a tracer because the inherent paramagnetic properties of pigmented lesions can be observed on MR. These lesions often exhibit characteristic high signals on T1-weighted images and low signals on T2-weighted images, which can enhance the accuracy of image analysis.

This study found that ¹⁸F-PFPN showed higher detection rates for ocular and orbital melanomas than ¹⁸F-FDG, and had less interference from background tissues. This finding is similar to that of Zhang et al. [14], who reported that ¹⁸F-PFPN could detect T2b stage melanoma and that the SUV_max of ¹⁸F-PFPN at the primary site was 3.92 times higher than that of ¹⁸F-FDG. Although none of the patients in this study had metastases, Zhang et al. [14] showed that ¹⁸F-PFPN could detect more and smaller melanoma metastases, identifying 365 metastases that were missed by ¹⁸F-FDG, including one smaller than 2 mm.

Although ¹⁸F-PFPN targets melanin, the normal presence of pigments in the retinal and uveal tissues does not appear to interfere with the results of ¹⁸F-PFPN. In this study, no abnormal concentrations of ¹⁸F-PFPN were detected in the contralateral normal eyes of any patient. In a study of a healthy population, the mean SUV in eye tissue increased from 0.79 ± 0.07 at 30 minutes after injecting ¹⁸F-PFPN to 1.32 ± 0.30 at 240 minutes, but these levels did not meet the criteria for pathological evaluation [14]. This finding suggests that ¹⁸F-PFPN has the potential to distinguish malignant melanomas from benign pigmentation.

Although ¹⁸F-PFPN has shown many advantages in detecting melanoma, it may not be as sensitive as ¹⁸F-FDG for identifying amelanotic melanoma [12,14], which accounts for approximately 2%–8% of melanomas [15]. This study presents a novel diagnostic challenge. This study has some limitations. First, the relative scarcity of these cases led to a limited sample size, preventing a more detailed statistical analysis of the data. Second, patients with nevi or other pigmented diseases requiring differentiation from malignant melanoma were not included. This will be the focus of future studies.

To the best of our knowledge, this is the first study to evaluate the ability of melanin-targeted ¹⁸F-PFPN to detect ocular and orbital melanomas. Compared to ¹⁸F-FDG, ¹⁸F-PFPN demonstrated higher detection rates for ocular and orbital melanomas and experienced minimal interference from the surrounding tissues. This suggests that ¹⁸F-PFPN may be a promising clinical diagnostic tool for ocular and orbital melanomas, with the potential to distinguish malignant melanomas from benign pigmentation. Future research should aim to verify these results with a larger sample size and explore the potential of ¹⁸F-PFPN to differentiate other pigmentary lesions and their use in adjuvant therapy options.

Availability of Data and Material

The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.

Supplement

The Supplement is available with this article at https://doi.org/10.3348/kjr.2024.0243.