# Novel four dimensional magnetic resonance imaging to monitor pancreatic tumor infiltrating blood vessels and tumor response to chemoradiation therapy

> **NIH NIH R21** · UNIVERSITY OF SOUTHERN CALIFORNIA · 2020 · $216,281

## Abstract

Background: Pancreas cancer patients have a dismal prognosis, with the cumulative 5-year
survival of 7%. Most patients present with locally advanced (LAPC) or borderline resectable
(BRPC) pancreatic cancer. The overall survival rates in the patients undergoing margin-negative
resection after chemoradiation therapies are 2-3 times of the unresected group. The superior
outcome of surgery calls for novel local treatment strategies aiming at downstaging patients and
improving resection rate. Targeted therapy such as stereotactic body radiation therapy (SBRT)
with simultaneous integrated boost (SIB) to tumor infiltrating blood vessels has the potential to
sterilize cancerous tissue around the vessels that limit surgery. However, the SIB treatment is not
commonly applied due to challenges in managing internal organ motion, visualization and
segmentation of boost volumes, and tracking the tumor response to the therapy with current CT
based planning and response assessment. Magnetic resonance imaging (MRI) is intrinsically
superior to CT in soft tissue contrast, but until recently, few sequences were available to address
the specific needs in pancreas radiotherapy including the need to resolve motion and differentiate
tumor infiltrated vessels. We pioneered a 3D k-space resorting based four-dimensional MRI (4D-
MRI) technique for the abdomen. Despite its high resolution, the current 4D-MRI images have
insufficient vessel contrast for segmenting and targeting.
Goals: The first goal of this study is to optimize the 4D-MRI technique so it generates a
sufficiently high blood vessel contrast for SBRT-SIB target localization. The second goal is to use
this novel 4D-MRI with vessel highlight to identify imaging makers that predict tumor
resectability after chemoradiation therapy intervention.
Methods: We will optimize the novel sequence by incorporating a new slab-selective function
which highlights tumor encasing blood vessels and quantify the enhancement of contrast-to-noise
(CNR) ratio as well as morphological changes for the vessels within the pancreatic tumor. A flow
phantom and healthy volunteers will be used to optimize 4D-MRI parameters such as the number
of radial lines, resolution and slab-selective volume. The CNR will be correlated to flow rates in
the phantom. To validate the method, we will also recruit 30 patients diagnosed with LAPC or
BRPC and perform pre- and post- chemoradiation 4D-MRI studies. The changes of the vessel
CNR, geometry and morphology will be correlated to standard clinical outcomes.
Impact: Success of this study will provide a non-invasive imaging method capable of localizing
tumor involving blood vessels while overcoming respiratory motion blur. The information will be
available to guide targeted radiation dose escalation to the boost volume defined by the involved
vessels, which initially preclude resection. Therefore, the project answers an unmet need in
current pancreatic cancer management since more than 85% of the patient...

## Key facts

- **NIH application ID:** 9979801
- **Project number:** 5R21CA234637-02
- **Recipient organization:** UNIVERSITY OF SOUTHERN CALIFORNIA
- **Principal Investigator:** Zhaoyang Fan
- **Activity code:** R21 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $216,281
- **Award type:** 5
- **Project period:** 2019-07-17 → 2021-06-30

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9979801

## Citation

> US National Institutes of Health, RePORTER application 9979801, Novel four dimensional magnetic resonance imaging to monitor pancreatic tumor infiltrating blood vessels and tumor response to chemoradiation therapy (5R21CA234637-02). Retrieved via AI Analytics 2026-05-27 from https://api.ai-analytics.org/grant/nih/9979801. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*