Cerenkov Luminescence Imaging in Prostate Cancer: Not the Only Light That Shines

Abstract

Cerenkov luminescence imaging (CLI) is a novel imaging technology that might have the ability to assess surgical margins intraoperatively during prostatectomy using ⁶⁸Ga-prostate-specific membrane antigen (⁶⁸Ga-PSMA-11). This study evaluated the accuracy of CLI compared with histopathology and, as an exploratory objective, investigated the characteristics of the identified chemiluminescence signal. Methods: After intravenous injection of a mean ⁶⁸Ga-PSMA-11 activity of 69 MBq intraoperatively, all excised specimens were imaged with CLI. Areas of increased signal were marked for histopathologic comparison and scored for the likelihood of being a positive surgical margin (PSM) using a 5-point Likert scale. In addition, the chemiluminescence signal was investigated in 3 radioactive and 3 nonradioactive specimens using CLI. Results: In 15 patients, the agreement between CLI and histopathology was 60%; this improved to 83% when including close surgical margins (≤1 mm). In 6 hot spots, CLI correctly identified PSMs on histopathology, located at the apex and mid prostate. In all 15 patients, an increased signal at the prostate base was observed, without the presence of the primary tumor in this area in 8 patients. This chemiluminescence signal was also observed in nonradioactive prostate specimens, with a half-life of 48 ± 11 min. The chemiluminescence hampered the visual interpretation of 4 PSMs at the base. Conclusion: CLI was able to correctly identify margin status, including close margins, in 83% of the cases. The presence of a diathermy-induced chemiluminescent signal hampered image interpretation, especially at the base of the prostate. In the current form, CLI is most applicable to detect PSMs and close margins in the apex and mid prostate.

Keywords: 68Ga-PSMA; Cerenkov luminescence imaging; intraoperative assessment; prostate cancer; surgical margins.

Graphical abstract

FIGURE 1.

Examples of PET/CT and CLI images from 4 patients: maximum-intensity projections (MIPs) of preoperative PSMA PET, transversal PET/CT images at height of primary tumor, and CLI images of excised prostate specimens without optical filter. Arrows locate hot-spot areas; green, PSM; blue, NSM; and pink, CSM with tumor ≤ 1 mm from margin, according to pathologist. Corresponding tumor–to–specimen edge distances on histopathology are noted below images.

FIGURE 2.

Agreement between CLI and histopathology in all patients divided into 3 regions of prostate. (A) Agreement between CLI hot spot (yes or no) and histopathology (PSM or NSM), excluding CSM. (B) Agreement including CSM (≤1 mm). (C) Agreement when adding LS to CLI hot spot. (D) Agreement with LS, including CSM. Overall agreement is noted below circles. Excl. = excluding; incl. = including.

FIGURE 3.

CLI images from prostate base in ⁶⁸Ga-PSMA-11 patients and bar chart showing SBR of CLI and chemiluminescence. (A and B) Unfiltered images of chemiluminescence at base. (D and E) Corresponding 550-nm short-pass filtered images. (C and F) Base of patient 8, who had multiple PSMs at base (arrows). These images show that visual distinction between chemiluminescence and actual PSMs is difficult. (G) Bar chart displaying SBR of chemiluminescence in nonradioactive specimens and that of PSM, NSM, and CSM in patient nonfiltered and filtered images. Average SBR is derived from all patient data, in which lesions on all sides of prostate were included. Note difference in scaling.

FIGURE 4.

Sequential imaging of nonradioactive prostate specimens to determine effect of time on intensity of chemiluminescence signal. Same scaling is used in all images. Graph displays half-life of chemiluminescence in 3 nonradioactive specimens on filtered and nonfiltered images and of ⁶⁸Ga.

FIGURE 5.

FAR images of ⁶⁸Ga-PSMA-11–containing prostate specimens and nonradioactive prostate specimens to investigate effect of FAR on tumor and chemiluminescent signal levels. FAR reduced chemiluminescence in both nonradioactive (60%) and radioactive specimen (70%) (top and bottom rows). FAR amplified signal originating from tumor (center row). Arrow shows contamination of scintillator. Same scaling is used for all images.