In 68 patients, a tracer aimed at an enzyme that clear cell kidney tumours produce in abundance outperformed standard FDG PET at nearly every site in the body, and changed management in 31% of cases.

Kidney cancer has an awkward relationship with the standard tool of cancer imaging. PET scans usually rely on FDG, a radioactive form of glucose that accumulates in tumours because cancer cells consume sugar rapidly. Renal cell carcinoma often does not take up much of it, and the kidneys themselves excrete the tracer, which clutters the picture in exactly the region of interest.

Researchers presenting at the 39th Annual Congress of the European Association of Nuclear Medicine report a tracer that avoids the problem by aiming at something else entirely. In 68 patients, it detected 90.9% of 492 confirmed lesions, against 56.7% for FDG. In the clear cell subtype, the commonest form of the disease, the gap was wider still: 97.9% versus 47.1%.

An enzyme the tumour makes and the kidney does not

The target is carbonic anhydrase IX, or CAIX, an enzyme that cells produce when starved of oxygen. In clear cell renal cell carcinoma, the gene that controls that response is typically disrupted, so the cells behave as though they are short of oxygen even when they are not, and manufacture CAIX continuously.

“CAIX gives us a highly selective biological target,” said Professor Esra Arslan, head of the Nuclear Medicine Clinic at Istanbul Training and Research Hospital, University of Health Sciences, and the study’s lead author. “It is expressed at a very high rate in ccRCC, while expression in normal renal tissue remains minimal. This gives us the potential to distinguish tumour from surrounding healthy tissue much more clearly.”

The tracer, [⁶⁸Ga]Ga-DPI-4452, binds CAIX and carries gallium-68, a radioactive isotope that a PET scanner detects. Because healthy kidney tissue makes little CAIX, the tumour stands out rather than hiding in a bright background.

Five times the signal, ten times the contrast

The numbers behind the detection rates describe why. Tumour uptake was roughly five times higher with the new tracer than with FDG, and the tumour-to-background ratio, the measure that determines whether a radiologist can actually see a lesion, was about ten times greater.

Contrast matters more than raw signal. A lesion that glows brightly in an equally bright field is invisible; one that glows modestly against a dark field is obvious.

Renal cell carcinoma accounts for the large majority of kidney cancers, and clear cell is its commonest form. The disease is often found incidentally on scans done for other reasons, and a substantial share of patients already have spread elsewhere in the body by the time it is identified, which is why accurate staging bears directly on treatment.

The difference was most extreme in the brain. For brain metastases, the new tracer detected every lesion identified, with a median maximum tumour-to-background ratio of 281.3. The figure for FDG was 0.7, meaning the lesions were no brighter than the surrounding tissue, and in practice invisible. Brain tissue uses glucose heavily, which is precisely why an FDG scan struggles there.

In the lungs, DPI-4452 found all identified lesions while FDG found roughly half. In lymph nodes the rates were 95.0% and 68.9%.

Where FDG still won

One site went the other way. Across the whole cohort, FDG performed better in the liver, which the researchers attribute to lower and more variable CAIX expression in some non-clear-cell subtypes of the disease. Within the clear cell subgroup, the new tracer was better there too.

That exception is worth noting rather than burying. Renal cell carcinoma is a family of diseases, and the biology the tracer exploits is characteristic of one member of it. Papillary and chromophobe tumours, among others, do not reliably carry the same disruption, and a tracer that depends on it will perform unevenly across them.

Patient-level diagnostic accuracy in clear cell disease was 98.2% with DPI-4452 and 77.2% with FDG.

What changed for patients

A better image is only useful if it alters a decision. In this study, clinical management or disease stage changed for 30.9% of patients after the CAIX scan, compared with 5.9% after FDG.

The changes ran in both directions. In three patients, findings that appeared positive on FDG were ruled out by the CAIX scan, which the researchers said avoided unnecessary interventions. Avoiding an operation on a lesion that is not cancer counts as much as finding one that is.

“[⁶⁸Ga]Ga-DPI-4452 has the potential to become a valuable tool across several stages of RCC management,” Arslan said. The roles she identifies are initial staging before removal of a kidney or a metastasis, restaging when recurrence is suspected, and eventual theranostic use.

The same molecule, as treatment

Theranostics is the pairing of diagnosis and therapy using the same targeting molecule. If a compound reliably finds tumour cells while carrying a diagnostic isotope, the diagnostic isotope can be swapped for a therapeutic one, delivering radiation to the cells the scan lit up.

The therapeutic counterpart here would be [¹⁷⁷Lu]Lu-DPI-4452, using lutetium-177, which emits radiation that damages nearby cells over a short range. The approach is established in prostate cancer and in some neuroendocrine tumours, where it has changed treatment for patients with advanced disease.

For kidney cancer, which responds poorly to conventional chemotherapy, a route of that kind would be significant. The scan, in that framing, is also a test of whether a patient’s tumour carries the target the treatment needs.

What has not been shown

Arslan was explicit about the limits, in terms unusual for a conference presentation.

“However, it is premature to claim that this agent represents a ‘new molecular imaging standard’ without validation in larger, multicentre studies stratified by subtype,” she said.

Stratification by subtype is the crux. The headline figures for clear cell disease are striking, and the overall cohort figures are diluted by subtypes where the biology does not apply. Establishing where the tracer helps and where it does not requires enough patients in each category to tell.

The findings were presented as abstract OP-624 at the congress, and a related prospective multicentre comparison by the same group appeared ahead of print in the Journal of Nuclear Medicine in September. Professor Halil Kömek of Gazi Yaşargil Training and Research Hospital in Diyarbakır contributed equally to the work.

---

Notes

1. Arslan, E., Kömek, H., et al. “Beyond FDG: CAIX-targeted PET achieves near-complete tumor detection in renal cell carcinoma.” Abstract OP-624, 39th Annual Congress of the European Association of Nuclear Medicine, 2026.
2. Kömek, H., Arslan, E., et al. “From metabolism to molecular phenotype: carbonic anhydrase IX-targeted [68Ga]Ga-DPI-4452 PET/CT in renal cell carcinoma: a prospective multicenter comparison with [18F]FDG PET/CT.” Journal of Nuclear Medicine, Epub ahead of print, 17 September 2026. DOI: 10.2967/jnumed.126.273011.
3. European Association of Nuclear Medicine. “New PET imaging approach detects almost all lesions in common form of kidney cancer.” EurekAlert!, 20 October 2026. https://www.eurekalert.org/news-releases/1147390.