Portable PET probes are a novel tool for intraoperative localization of tumor deposits

Abstract

Background: Positron emission tomography (PET) identifies cancer deposits by detecting sites of gamma emissions that are released from radioactively labeled molecules targeting tumor to formulate a PET image. Correlating preoperative PET scans with intraoperative findings remains a challenge. We investigated whether high-energy gamma emissions detected by a novel hand-held PET probe would detect tumors and offer a real-time method to localize tumor intraoperatively. Furthermore, we investigated the novel beta probe, which detects emissions at a shorter range than gamma emissions, making them undetectable by PET scanners, but potentially valuable for close range intraoperative detection of tumor deposits.

Methods: Six-to-eight-week-old athymic mice were injected with one of four possible tumor cell lines: gastric, pancreas, squamous cell and breast cancer. After tumors reached at least 1 cm in size, they were euthanized and imaged with a micro-PET imager. Hand-held gamma and beta probes were then used in vivo and ex vivo to measure high-energy gamma and beta emissions.

Results: The portable PET probes detected high-energy gamma and beta emissions from all tumors evaluated. These emissions were reproducible and we established that beta emissions correlate with high-energy gamma emissions and conventional PET scans. There was a strong positive correlation (R = 0.8) between gamma and beta counts. Beta emission showed a stronger correlation than gamma emission with overall tissue radioactivity.

Conclusion: This study is the first to demonstrate that gamma emission detected by conventional PET imaging correlates with beta emissions. This study shows that compared to detection of gamma emissions, beta counts may offer superior real-time localization of tumor deposits. Intraoperative portable PET probe may become a useful way to exploit tumor biology and PET technology to guide real-time tissue characterization during surgery.

Figure 1

Conventional PET imaging detects gamma waves emitted from tumors in a murine model. Flank tumors of multiple tumor types (breast, pancreatic, gastric, and squamous cell carcinoma) were grown in nude mice. Mice were injected with radiolabeled ¹⁸F and imaged 1-hour later by a microPET scanner. Animals were then sacrificed and tumors were sectioned for analysis by autoradiography and hematoxylin and eosin (H&E) staining. Shown is a representative animal PET scan with tumor highlighted by a red arrow (A), autoradiograph of tumor section (B), and H&E of tumor section (C).

Figure 2

Portable PET probes detect gamma emission and beta emission from both low and high tested doses of radiation in vitro. Multiple samples of radiolabeled ¹⁸F were obtained at two doses (500 μCi and 5000 μCi), Measurements were taken using the portable probes for both gamma and beta emission at the source and at distances up to 5 cm from the source.

Figure 3

Beta emission directly correlates with gamma emission in all tested tumor types in vivo (R = 0.8). Four tumor types were tested for gamma and beta emission following animal injection with radiolabeled ¹⁸F (n = 10 animals). Pancreatic, gastric, squamous cell and breast cancer cell lines injected subcutaneously into the murine model. Emissions were measured with portable probes on the tumor in vivo as well as on excised tumor ex vivo once tumors reached 1 cm in size. Beta and gamma emission directly correlated as determined by the Pearson correlation calculation of the R value.

Figure 4

Beta detection better pinpoints the source of radiation than gamma detection. Beta and gamma measurements were taken at incremental distances from the tumors 1-hour following intravenous injection of radiolabeled ¹⁸F. Beta detection was nearly zero at distances > 5 cm from the source, while gamma emission was still detectable at 15 cm. Data from a representative animal is shown to demonstrate that at the spot over tumor, the beta counts are high, but after the probe is moved ~2 cm away from the tumor, counts are nearing zero. In contrast, the gamma probe detects emissions up to 10 cm away from the site of tumor.

Figure 5

Ratio of gamma and beta counts to background counts in vivo in mice with pancreatic, gastric, breast, and squamous cell tumors.