Peder E. Z. Larson, Jenna M. L. Bernard, James A. Bankson, Nikolaj Bøgh, Robert A. Bok, Albert P. Chen, Charles H. Cunningham, Jeremy Gordon, Jan-Bernd Hövener, Christoffer Laustsen, Dirk Mayer, Mary A. McLean, Franz Schilling, James Slater, Jean-Luc Vanderheyden, Cornelius von Morze, Daniel B. Vigneron, Duan Xu
6 min
Hyperpolarized (HP) [1-13C]pyruvate MRI is a powerful, non-invasive imaging modality that enables the real-time visualization of metabolic processes in humans. By significantly increasing the signal-to-noise ratio of 13C-labeled compounds, this technique allows researchers to track the conversion of pyruvate into lactate, alanine, and bicarbonate, which are key indicators of metabolic shifts in diseases such as cancer, heart disease, and neurological disorders. As the field transitions from early feasibility studies to multi-site clinical trials, establishing standardized practices is critical for ensuring data comparability and regulatory compliance.
This paper, developed by the HP 13C MRI Consensus Group, synthesizes current successful practices across four main areas:
As HP 13C MRI continues to expand, the lack of standardized protocols poses a barrier to large-scale clinical adoption. By documenting the current state of the field and identifying evidence gaps, this paper serves as a foundational reference for future consensus-building. It emphasizes that while current methods are safe and effective, the community must move toward unified reporting and analysis standards to maximize the clinical impact of metabolic imaging.
MRI with hyperpolarized (HP) 13C agents, also known as HP 13C MRI, can measure processes such as localized metabolism that is altered in numerous cancers, liver, heart, kidney diseases, and more. It has been translated into human studies during the past 10 years, with recent rapid growth in studies largely based on increasing availability of hyperpolarized agent preparation methods suitable for use in humans. This paper aims to capture the current successful practices for HP MRI human studies with [1-13C]pyruvate - by far the most commonly used agent, which sits at a key metabolic junction in glycolysis. The paper is divided into four major topic areas: (1) HP 13C-pyruvate preparation, (2) MRI system setup and calibrations, (3) data acquisition and image reconstruction, and (4) data analysis and quantification. In each area, we identified the key components for a successful study, summarized both published studies and current practices, and discuss evidence gaps, strengths, and limitations. This paper is the output of the HP 13C MRI Consensus Group as well as the ISMRM Hyperpolarized Media MR and Hyperpolarized Methods & Equipment study groups. It further aims to provide a comprehensive reference for future consensus building as the field continues to advance human studies with this metabolic imaging modality.
Sam: So what's the regulatory dimension? The sterilization pathway must matter enormously for a GMP-grade injectable.
Alex: It does, and this is one of the sharper tensions the paper identifies. There are two production paradigms in current use: terminal sterilization, where the final product is sterilized after formulation, and sterile preparation, where everything is handled aseptically from the start. These aren't just different workflows—they carry different regulatory burdens, different failure modes, and different implications for what you can do with the agent downstream. A site that has built its entire GMP infrastructure around one approach can't trivially switch, and the paper argues the field hasn't converged on which should be standard.
Sam: Which means even if two sites agree on acquisition parameters and analysis pipelines, they may be starting from chemically and regulatorily distinct agents.
Alex: That's the core problem. And it's compounded by the calibration layer. Before you can interpret the metabolic signal quantitatively, you need to know the actual polarization level of what you injected—not just what the polarizer reports, but what survived the fluid path. The paper discusses phantom-based calibration approaches, but acknowledges there's no consensus on what the calibration phantom should look like, what concentration to use, or how to account for coil geometry differences across sites.
Sam: So the quantification is uncertain at both ends—uncertain input, uncertain reference.
Alex: Which is why the authors frame this as a roadmap rather than a solved problem. They're not claiming the field has a working standard. They're arguing the field needs to agree on where the critical control points are before it can build one. The taxonomy is the contribution—not a protocol, but a structured way of thinking about where variability matters and where it doesn't.
Sam: Is there a sense of which of the four modules is closest to consensus?
Alex: Acquisition is probably the most mature. Pulse sequence design and timing constraints are reasonably well understood—the T1 decay sets hard limits on how long you can spend acquiring data, and most sites have converged on similar flip angle strategies for that reason. Analysis is more contested, particularly around partial volume effects and motion correction, but there's at least an active methodological literature to draw from. Preparation and calibration are where the paper sees the most urgent need for coordination.
Sam: And the limitation that most constrains the whole enterprise is that this is a consensus document, not an interventional study. There's no experimental comparison of protocols.
Alex: That's the honest read. The paper maps the landscape carefully, but it can't tell you empirically how much preparation variability actually inflates variance in downstream metabolic readouts across sites—that experiment hasn't been done at scale. What the authors are doing is making the case that it needs to be done, and that you can't design it without first agreeing on what you're controlling.
Sam: So the field is at the stage of writing the protocol for the protocol.
Alex: That's a fair summary. The technology is clinically viable, the biological signal is interpretable, and what's missing is the infrastructure layer—the shared standards that would let a trial coordinator at one institution trust that a number coming out of a scanner at another institution means the same thing. This paper is an attempt to define what that infrastructure needs to cover. Whether the community actually converges around it is a different question, and one the paper can't answer.
Sam: A useful map, even if the territory is still being negotiated.
Alex: Exactly. Thanks for listening to ResearchPod.