Lawrence Berkeley Laboratory

10/01/2026 | Press release | Distributed by Public on 10/01/2026 09:26

Targeted Cancer Therapy Gets a Sharper Focus

SPECT scanners are commonly used in hospitals, but they lack the required sensitivity to accurately generate images from the detected gamma rays in targeted alpha therapy. SPECT uses thick, heavy metal shields (collimators) to detect single gamma rays - but this only works well when the flux of gamma rays is large. And because targeted alpha therapy is so powerful, only small doses of the radioisotope - about 100 times less than a conventional radiotracer - is administered to patients. Altogether, these constraints add up to low gamma-ray sensitivity and noisy images.

On the other hand, PET promises higher gamma-ray sensitivity and image quality because it doesn't use collimators. PET works by picking up "positron annihilation" events - When a positron collides with an electron, it emits two gamma-ray photons back-to-back. The PET algorithm quickly combines tens of thousands of these data points to generate an image of the radioisotope accumulation in the body, such as in a tumor.

Researchers have long assumed that PET scanners wouldn't work with actinium-225 because the radioisotope emits alpha particles, not positrons. "But one day we wondered, what if we could exploit actinium-225's gamma-ray pairs to demonstrate targeted alpha therapy in a PET scanner?"

So Caravaca wrote a new algorithm from scratch - TOF-CGI is an image reconstruction algorithm that uses the detected gamma ray pairs to pinpoint where actinium-225 decayed, and reconstruct that data into a 3D image of actinium-225 in the body. TAT emits thousands of signals as the radionuclide decays, so leveraging the GPU computation capabilities at the National Energy Research Scientific Computing Center (NERSC) was essential to significantly accelerate the image-generation process during the development and testing of this technique.

Caravaca and co-authors at UC San Francisco then conducted a pilot study with a patient who was previously diagnosed with prostate cancer through standard PET scans. A validation test confirmed that TOF-CGI in a PET scanner identified the same prostate tumors that were detected by the initial PET diagnostic scans. Subsequent experiments showed that TOF-CGI in PET detected more actinium-225 decays in the patient than SPECT, resulting in clearer images. The results represent the first TOF-CGI imaging of a human subject.

"This is an exciting first step. Clinicians want to know whether a radioisotope is doing its job. Did it stop the cancer cells from spreading? Or did new tumors emerge? Targeted alpha therapy imaging could help us answer those questions and more, but we still need an imaging modality to match. We think that our TOF-CGI technique could one day help us get there," Caravaca said.

In future studies, Caravaca and team hope to expand the new technique to other radioisotopes and test its efficiency with a full-body PET scanner.

Lawrence Berkeley Laboratory published this content on October 01, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on October 01, 2026 at 15:26 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]