Molecular Imaging, Radiobiology Modeling and Biologically Conformal Radiation Therapy

Anatomic imaging modalities such as CT and MRI do not always provide an accurate picture of the tumor extent, especially in the zone of infiltration that may be the limiting factor in an attempt to use the radical treatment approach. Many efforts have been undertaken to apply various metabolic imaging techniques, including MR spectroscopic imaging (MRSI), SPECT and PET, to attack the problem by adding the metabolic and cellular picture to the anatomic information from conventional imaging techniques. In line with the mission of the Molecular Imaging Program at Stanford (MIPS), we are embarking on a number of efforts on applications of molecular imaging in radiation oncology. We are working to improve the current radiobiology models with consideration of  the most recent biological and molecular imaging findings. We hope to bridge the gaps between modern biology and computer modeling and clinical decision-making so that truly optimal therapeutic treatment plans can be obtained. In collaboration with Dr. D. Spielman's group, we are working on incorporating MRSI into therapeutic treatment. We are also working on a method to quantitate the MRSI abnormality data and integration of hyperpolarized 13C MSRI into radiation oncology research.

In addition to better delineating the tumor volume, biological imaging also promises to afford the spatial distribution of various biological parameters. We are establishing a new paradigm of biologically conformal radiation therapy (BCRT), which takes into account both the new tumor boundaries and the spatial biology distributions mapped by the new imaging modalities. The goal here is to customize the dose distribution according to the spatially heterogeneous tumor burden requirements to truly individualize the RT - as opposed to a spatially uniform dose distribution used in today's clinical practice. This will lay the ground for the next generation of radiation therapy, allowing us to escalate tumor dose more intelligently and effectively.


Radiation Energy Transfer by Gold Nanoclusters (AuNCs)

Few atom gold nanoclusters conjugated with blood serum proteins (AuNCs) were shown to be a highly effective ionizing energy transfer mediators suitable for optical in-vivo radioisotope imaging. Clinically used beta-emitting radioisotopes Fluorine-18 and Yttrium-90, known to produce Cerenkov signal in living tissue, and Technetium-99m, a pure gamma emitter were tested. Results suggest that AuNCs can be excited by radioisotopes via direct and indirect beta particle interactions.

Radioisotope – AuNC Energy Transfer (RET)

AuNCs excited by direct and indirect β particle interactions are highly effective energy transfer mediators, suitable for in vivo optical imaging of radioisotopes


(Small (2015) DOI: 10.1002/smll.201500907)


  • Biodistribution of the beta-emitting radioimmunoconjugates (Bexxar and Zevalin
  • Hybrid optical imaging of tumors small lesions


In vivo and Ex vivo RET studies. (a) Mice bearing subcutaneous breast carcinomas were injected with 750μCi 18F-FDG and 1xPBS (left, n=4) and 750μCi 18F-FDG and BSA-conjugated AuNCs (right, n=7) Optical readings were acquired with Cy5.5 bandpass filter for 5 min. (b) In vivo quantitative analysis of the signal in the tumor before and after AuNC injection (n=7). The ROI (region of interest) was selected from the readings with open filter and used for data analysis at the selected wavelengths. The signal was compensated for radioactive decay and Cerenkov background by comparison to control mice. (c) Ex vivo quantitative analysis of the signal in the tumor. Optical readings were acquired in 695-770 nm range for 5 minutes. The measured activity per tumor was ~2 μCi. Average tumor weight was ~0.15 mg. Tumors harvested from control mice (n=4) were compared to the tumors harvested from mice injected with AuNCs (n=7). (d) RET signal correlation with tumor activity up-take.