Lanthanides go bio: the long road to the clinic

Key points

• Lanthanides offer unique optical and magnetic properties that can be harnessed to create superior probes for biomedical imaging, sensing, and therapy.
• Feasibility studies demonstrate multimodal in vivo imaging, sensing, and theranostic capabilities but rarely benchmark against existing probes or start from an unmet clinical need.
• Closing the gap between laboratory demonstration and clinical use requires closer collaboration with biomedical researchers, standardized evaluation under clinically relevant conditions, and long-term safety testing.

Introduction

Lanthanides form a special group of elements positioned at the bottom of the periodic table, in the row above the radioactive actinides. Together with yttrium and scandium, they form the so-called rare earths—a misleading name, because these elements are quite abundant. Their unique optical and magnetic properties, which arise from the partially filled inner 4f shell, are the basis for many present and future applications (1). Lanthanide research has traditionally been the domain of inorganic/physical chemists and physicists working toward more efficient artificial light sources and stronger permanent magnets for electronics, cars, and wind turbines: europium, cerium, and terbium light up almost all lamps and displays, and the world’s strongest magnets rely on neodymium or dysprosium.

More recently, the development of lanthanide-doped nanostructures has opened up imaging and sensing applications in biological systems, medical diagnostics, and even theranostics, with particularly rapid progress over the past decade. In their Frontiers in Science lead article, Liu et al. outline recent breakthroughs in lanthanide-based upconversion nanoparticles, antenna complexes, and hybrid metal–organic frameworks (2). These nanoprobes can deliver high-resolution imaging; sense pressure, temperature, and water content; provide magnetic resonance imaging (MRI) contrast; and combine several of these functions in a single multimodal probe with therapeutic capabilities. Yet commercial clinical application lags far behind. The field tends to focus on demonstrating what lanthanide probes can do but rarely asks what the clinical needs are and whether these probes meet those needs. Until these questions receive more attention, impressive results will keep accumulating without changing clinical practice.

Spectacular capabilities, demonstrated in the laboratory

Lanthanide-doped upconversion nanocrystals underpin many of the new biomedical imaging and sensing developments discussed by Liu et al. (2). In upconversion, two lower-energy photons are combined via two-step energy transfer into a single higher-energy photon. Lanthanide-doped materials perform this process orders of magnitude more efficiently than other materials, including biological tissue, and infrared-excited upconversion emission therefore produces an optical signal with essentially no competing background signal from the surrounding tissue. Upconversion was discovered 60 years ago in ytterbium–erbium-doped bulk crystals and glasses (3), but it is the recent translation to nanocrystals that has enabled bio-applications. Their extremely high photostability is a further advantage over traditional probes such as fluorescent molecules or quantum dots. A recent study (4) illustrates the possibilities: using photostable ytterbium- and erbium-doped upconversion nanoparticles as labels, Peng et al. tracked single nanocrystals for more than an hour as they carried cargo along neuronal axons over millimeter distances. This work provided new insight into the dynein-mediated transport mechanism in living cells.

The same workhorse ytterbium- and erbium-doped nanocrystals are widely used to combine imaging with sensing of temperature—and more recently pressure—in biological systems including living organisms. Pairing temperature sensing with heating is a promising strategy for theranostics: nanoparticles coated with antibodies that target tumor cells are heated by infrared radiation or an external field, while their luminescence spectrum monitors temperature so it can be controlled at the position of the nanocrystal. A related approach uses infrared upconversion for both optical imaging and photoactivation of singlet oxygen to kill tumor cells. Upconversion nanocrystals embedded in micrometer-sized plastic beads have even been fed to C. elegans worms to record the grinding forces inside the pharynx (5). Photon avalanching, a variant of upconversion, was discovered only recently in nanocrystals and has enabled super-resolution imaging well below the diffraction limit along with further gains in temperature and force sensing (6). New capabilities of this kind continue to be reported, achieving spatial resolution on (or even below) the micrometer scale and penetration depths of centimeters.

The magnetic properties of lanthanides, especially gadolinium, are exploited in MRI. Coupling to gadolinium’s large magnetic moment shortens the T₁ relaxation time of nearby hydrogen nuclei, thereby generating contrast. Incorporating gadolinium into luminescent lanthanide-based nanoprobes therefore offers a route to multimodal probes that combine optical imaging and MRI in a single agent.

These studies, and many more featured in Liu et al. (2), reflect the remarkable capabilities of lanthanide probes for biomedical imaging and sensing. Still, few were designed with clinical implementation in mind. Optimizing performance parameters and demonstrating feasibility are scientifically rewarding, but reaching the clinic requires a different approach.

Why the clinic remains out of reach

Molecular gadolinium complexes for MRI contrast enhancement are one of the few lanthanide imaging probes to have reached the clinic (7). Years after market introduction and widespread clinical use, retention and safety concerns emerged, leading to regulatory restrictions on several agents. The field’s one clear translational success therefore also stands as its clearest warning: problems that laboratory testing cannot anticipate may only surface once a probe is in routine use over timescales not covered by current feasibility studies.

Upconversion nanoparticles in particular are prominently featured in the literature, but real-world implementation faces challenges that are well-known for any biomedical application. Too often, studies lack a realistic connection to a medical challenge that would benefit from the new capability, whether deeper penetration, higher spatial resolution, greater sensitivity, improved targeting, or optical imaging and MRI combined in a single probe. Instead, many focus on a single parameter—relative sensitivity for luminescence thermometry, say—or on a proof-of-principle established under ideal laboratory conditions, giving less attention to the brightness, reproducibility, reliability, and stability that clinical use demands. Safety and toxicity studies are often limited to cell viability tests that show no immediate toxicity, while data on long-term effects remain scarce, as these require complex and lengthy studies monitoring accumulation and release. The surface layer of a nanoparticle, for instance, is typically functionalized for biocompatibility and targeting; ligand exchange with biomolecules in the bloodstream is known to compromise that layer and deserves more attention if performance is to be reproducible and stable in living organisms. Without long-term safety data of this kind, new lanthanide-based probes will not reach the clinic.

What needs to change?

The gap between laboratory demonstration and clinical use may reflect the physics and chemistry background of most researchers in the field, and this calls for closer collaboration with biomedical colleagues. Such collaboration would help identify where the functionalities of lanthanide probes can address unmet clinical needs and which other properties matter when selecting the most promising candidates for joint development. Understanding what state-of-the-art imaging infrastructure the clinic already has would show researchers areas where existing devices need adaptation or where completely new devices are required. Biomedical expertise is equally necessary for benchmarking against existing probes, testing performance under clinically relevant conditions, and ensuring probes are biocompatible, biofunctionalized, and biostable.

Standardization and best-practice protocols are needed to better evaluate and compare reported results—such as for luminescence nanothermometers, where claims about temperature sensitivity and accuracy are too often unrealistic. A recent review (8) addresses some of these issues relating to the accuracy, precision, and reliability of temperature sensing. Experience in image processing shows both the value of such protocols in gaining acceptance for a new technology and the difficulty of establishing them (9). Standardized evaluation should also assess a probe’s complete performance under biologically and clinically relevant conditions, rather than a single optimized metric measured under ideal ones. The most promising probes for the clinic will not be those that excel at one headline parameter but those that combine imaging performance with a wider set of parameters best evaluated together with biomedical researchers.

Conclusion

Lanthanide-based probes have unique and superior capabilities that could redefine the field of biomedical imaging and may also add sensing and therapeutic functions. The road to the clinic is still long. It requires a sharper focus on identifying applications where improved performance genuinely justifies replacing an existing probe and on finding new applications where the added functionality answers a real clinical need. Performance under clinically relevant conditions, biocompatibility, reliability, safety, and standardized protocols must all be addressed. To reach the end of that road, it is crucial to walk it alongside researchers from the biomedical field.

Generative AI statement

The author declared that generative AI was not used in the creation of this manuscript.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Frontiers in Science Editorial Office, Frontiers Media SA, Switzerland