18F-DCFPyL PSMA PET/CT Tracheobronchial Uptake in Patients with Prostate Cancer: Incidence and Etiology

Visual Abstract


Patient Population
The study included 100 consecutive PCa patients referred for 18 F-DCFPyL PET/CT.This retrospective study was approved by the Institutional Human Research Ethics Committee with a waiver of informed consent from patients who had been scanned for clinical indications.

F-DCFPyL PET/CT Image Acquisition and Protocol
The 18 F-DCFPyL PET/CT image acquisition and protocol are described in the supplemental materials (available at http://jnm.snmjournals.org).

Data Analysis
For each of the 100 PCa patients who had 18 F-DCFPyL scans, we collected data on age, prostate-specific antigen (PSA) level, injected activity, and length of uptake phase.For each scan, imaging data were analyzed using MIM Encore (MIM Software Inc.).Maximumintensity projection images were visually assessed for the presence or absence of PSMA uptake above the blood pool level in the trachea and main bronchi and semiquantitatively by SUV max .Also, SUV max measurements were obtained for normal physiologic uptake in the lacrimal gland, parotid gland, submandibular gland, liver, spleen, kidney, small bowel, left and right stellate and celiac ganglia, blood pool, and gluteal region.Normal-organ uptake was compared in patients with and without tracheal uptake.The collected data for measured parameters were analyzed for statistical significance using a paired t test.To illustrate and compare tracheal 68 Ga-PSMA-11 and 18 F-DCFPyL uptake in the same patient, PET/CT images using both tracers were evaluated in only one patient.

Immunohistochemistry
To confirm tracheobronchial PSMA expression, immunohistochemistry was performed on tracheobronchial samples taken from 2 men who had surgical resection of lung cancer; the supplemental materials provide immunohistochemistry details.

RESULTS
Of the 100 patients (mean age, 67.4 y [SD, 6.68 y]; range, 50-85 y), tracheobronchial uptake was present in 31 (31%) (Fig. 1).Paired t testing showed that when tracheal uptake was present, the SUV max was significantly higher in the left main bronchus (mean, 2.7) than in the right (mean, 2.3) (P , 0.001) (Fig. 2).Also, patients with tracheobronchial uptake showed significantly higher PSMA uptake in the lacrimal and submandibular glands but lower PSA and uptake in the blood pool (Table 1).
However, there were no significant differences in age, injected PSMA activity, duration of uptake phase, or other measured areas of uptake, including the parotid gland, liver, spleen, kidney, small bowel, ganglia (stellate or celiac), and gluteal muscle. 68Ga-PSMA-11 and 18 F-DCFPyL in the same patient revealed more prominent tracheal uptake of 18 F-DCFPyL (Fig. 3).Immunohistochemistry performed on tracheobronchial samples taken from 2 lung cancer patients demonstrated weak-to-moderate PSMA expression in bronchial submucosal glands in both cases (Fig. 4).

DISCUSSION
PSMA PET/CT is now considered the new standard of care in the management of PCa, with inclusion of PSMA PET/CT in the National Comprehensive Cancer Network guidelines (9).Despite the high diagnostic accuracy of PSMA-based agents in various PCa scenarios, PSMA is not specific to PCa or even to the prostatic gland.Several studies have shown a wide spectrum of nonprostatic benign and malignant PSMA uptake on PET using virtually all PSMA-based agents.We noted an unusual pattern in which physiologic uptake on PSMA PET was frequently seen in the trachea and proximal bronchi, on the left more than the right.Published examples of this pattern of PSMA tracheal uptake are limited to case reports, primarily attributed to underlying lung pathology, and do not localize the site of or determine the mechanism of PSMA expression (4)(5)(6)(7)(8).
In the present study, tracheal uptake was noted in 31% of patients undergoing 18 F-DCFPyL PET/CT (Fig. 1).No relationship was found between tracheobronchial uptake and age, injected activity of 18 F-DCFPyL, or uptake in the parotid gland, liver, spleen, kidney, small bowel, stellate or celiac ganglia, or gluteal muscle (Table 1).However, those with tracheobronchial uptake had significantly higher uptake in the lacrimal and submandibular glands but lower uptake in the blood pool and a lower PSA level.The higher uptake in the salivary glands and trachea might be related to their being secretory glands with a similar origin.Although xerostomia is one of the most common side effects of PSMA radioligand therapy, we could find no report that PSMA therapy can cause adverse airway events or respiratory symptoms.In our study, a lower PSA level was associated with higher tracheobronchial uptake.This might reflect a lower sink effect, with a low PSA level generally indicating a lower disease burden.However, lower blood pool uptake may contribute to a higher physiologic-to-background ratio, thus resulting in more noticeable tracheal uptake.Higher tracheobronchial uptake was also associated with lower uptake in the blood pool.This may reflect uptake time, as blood pool clearance and accumulation of tracer in PSMA-expressing tissues increase with time.The trachea receives its nerve supply from the pulmonary plexus, which is derived from both the sympathetic and the parasympathetic nervous systems.We hypothesized that in patients with higher autonomic tone, higher tracheal uptake may be noted more frequently in those with higher ganglia uptake, particularly stellate, since it has the same nerve supply as the trachea.In our study, tracheal uptake was associated with higher uptake in the left stellate and left celiac ganglia than in the right stellate and right celiac ganglia.Also, tracheal uptake was associated with higher lacrimal and submandibular uptake.Such a pattern might be suggestive of an underlying higher autonomic tone in the trachea, particularly on the left side.Paired t testing showed that when tracheal uptake was present, the SUV max was significantly higher in the left main bronchus  (mean, 2.7) than in the right (mean, 2.3) (P , 0.001).Anatomic differences between the narrower, longer, and more horizontal left bronchus and the wider, shorter, and more vertical right bronchus might also contribute to more PSMA PET visualization on the left side than on the right (Fig. 2).
Several studies showed differences in biodistribution between currently Food and Drug Administration-approved PSMA agents (10)(11)(12).Our data show that a relatively higher injected activity and longer uptake duration may lead to higher contrast in the image as well.In our experience, tracheobronchial uptake is seen relatively more often on 18 F-DCFPyL PET/CT than on 68 Ga-PSMA-11 PET/CT (Fig. 3), but this may be because of higher contrast related to the higher positron yield and lower positron energy of 18 F or because we tend to image later (90 min after injection) and inject higher activity with 18 F-PSMA agents.More recently, we changed our 18 F-DCFPyL imaging protocol from 90 to 60 min, which may lead to less detection of the tracheobronchial uptake.We made this change to align with Food and Drug Administration prescribing information.
Recent studies reported PSMA expression in small salivary glands in the nasopharynx (13).To localize tracheobronchial PSMA expression, immunohistochemistry was performed on tracheobronchial samples taken from 2 men who had surgical resection of lung cancer.Figure 4 shows PSMA expression in bronchial submucosal glands, suggesting that this uptake is related to tracheoglandular function.Submucosal Values are reported as mean followed by SD in parentheses for each variable within tracheal uptake group.P value is based on independent-samples t test.

FIGURE 2 .
FIGURE 2. 18 F-DCFPyL PET maximum-intensity projection showing tracheal uptake to be significantly higher in left main bronchus.