Background:
There is an urgent need for non‑invasive biomarkers capable of identifying high‑risk premalignant pancreatic cancer lesions and monitoring their progression and response to emerging interventions. Pancreatic ductal adenocarcinoma (PDAC) carries a poor prognosis because it is typically diagnosed at advanced stages, owing to the absence of early symptoms and the lack of reliable early‑detection tools. Approximately 15% of PDAC cases arise from precursor cystic lesions, particularly intraductal papillary mucinous neoplasms (IPMNs), placing these individuals in a clearly defined high‑risk group1. Once malignant transformation occurs, survival declines sharply, underscoring the importance of intervening during the premalignant window. Currently, surgical resection is the only intervention for high‑risk IPMNs, but it carries substantial morbidity and is not suitable for all patients, highlighting the need for non‑surgical prevention strategies. KRAS mutations, present in more than 80% of IPMNs, are the predominant oncogenic drivers of pancreatic tumorigenesis. Recent advances in targeted therapy have produced KRAS inhibitors (iKRAS), including MRTX1133, a selective KRASG12D inhibitor that demonstrates potent antitumor activity in murine PDAC models2. These promising preclinical results suggest that iKRAS agents may offer a transformative approach for intercepting premalignant lesions and preventing progression to invasive pancreatic cancer.

Hypothesis/Objective:
Metabolism‑based Hyperpolarized Magnetic Resonance (HP‑MR) enables real‑time, in vivo tracking of metabolic alterations in premalignant pancreatic lesions—an imaging capability not achievable with any existing clinical modality. HP‑MR offers the unique ability to non‑invasively assess therapeutic efficacy by monitoring dynamic metabolic shifts in response to treatment. Prior work from our laboratory has shown that HP‑MR provides sensitive and specific detection of premalignant pancreatic lesions capable of progressing to cancer across multiple preclinical models, including an IPMN model driven by KRAS and GNAS mutations4. Building on this foundation, we conducted a proof‑of‑concept HP‑MR study to evaluate the metabolic response and therapeutic efficacy of the KRASG12D inhibitor MRTX1133 in an IPMN mouse model.

Methods:
HP[1-13C]Pyruvate MR was employed to study the metabolic processes in SMAD4/KC mouse model (P48Cre; LSL-KrasG12D/+; Smad4fl/fl), which is known to develop IPMN lesions and progress quickly to PDAC. The dissolution DNP (SpinAligner Polarize, Denmark) operating at 3T was utilized to hyperpolarize 1-13C pyruvate. The 13C magnetic resonance spectra were acquired at 7T Bruker MRI scanner. Mice were imaged before treatment at 15-16 weeks of age and then treated with either MRTX1133 or vehicle control for 5 weeks and imaged again. Pancreatic tissue were then collected for ex vivo experiments, including NMR metabolomics and histological analysis.

Results:
After the 5‑week treatment regimen, anatomical MR imaging revealed a marked reduction in pancreatic tissue volume and lesion burden in MRTX1133‑treated mice compared with vehicle controls. Quantitative HP‑MR analysis demonstrated a significant decrease in the lactate‑to‑pyruvate ratio in the MRTX1133 group (0.18 ± 0.02) relative to vehicle‑treated mice (0.54 ± 0.03), consistent with reduced glycolytic flux following KRASG12D inhibition (Figure 1a), and the anatomical observations. NMR‑based metabolomics further supported these observations, showing a reduced lactate pool in MRTX1133‑treated pancreas compared with vehicle controls (Figure 1b). Ex vivo histological analysis confirmed a striking reversal of disease pathology: pancreatic tissue from treated mice closely resembled normal architecture, whereas vehicle‑treated tissue displayed extensive dysplasia and numerous IPMN lesions.

Conclusion:
In summary, this proof‑of‑concept study demonstrates the potential of HP-MR to non‑invasively assess the therapeutic efficacy of KRAS inhibitors in a pancreatic cancer prevention setting. These findings establish a foundation for more rigorous evaluation of preventative strategies using HP‑MR, including studies incorporating daraxonrasib (RMC‑6236), a next‑generation KRAS inhibitor currently under FDA review and being tested in an ongoing phase I/Ib clinical trial (NCT05379985)3. In parallel, our first‑in‑human HP metabolic imaging trial in high‑risk patients with pancreatic premalignant cystic lesions (NCT05873699) is actively underway, providing a critical translational bridge. If successful, KRAS‑targeted agents could be deployed as early interception therapies, with HP‑MR serving as a powerful tool to monitor therapeutic response and guide personalized prevention strategies.

Image/Figure:
Click to view full size
Click to view full size

Image/Figure Caption:
Figure 1: A) In vivo hyperpolarized MRI data from Smad4 mice treated with vehicle or MRTX1133 for 5 weeks. B) Ex vivo lactate NMR measurements from extracted pancreatic tissue of Smad4 mice either untreated or treated with MRTX1133.

Author
Julia Rose Zickus
The University of Texas MD Anderson Cancer Center