The tracer, named β-Me-2-[18F]FDG, is described in a study published in Nature Biomedical Engineering. It is derived from [18F]FDG, the glucose analogue that has anchored clinical PET for decades, but with a crucial chemical modification: a methyl group attached at the anomeric C-1 position through a β-glycosidic linkage. That small structural change transforms the molecule from a probe of host metabolism into a probe of bacterial metabolism. Human cells import glucose through GLUT and SGLT transporters, which depend on hydroxyl groups at the C-1 and C-2 positions of the sugar. By replacing the C-1 hydroxyl with a methyl group and lacking the C-2 hydroxyl entirely, β-Me-2-[18F]FDG is no longer recognized by mammalian transporters. Bacteria, however, possess an entirely different carbohydrate import machinery, the phosphoenolpyruvate-dependent phosphotransferase system, or PTS, which simultaneously transports and phosphorylates sugars using phosphoenolpyruvate as the phosphoryl donor. Because this system exists only in prokaryotes, it offers a metabolic doorway that is invisible to human tissue.

The discovery began with a remarkably simple synthetic strategy. Rather than building each candidate tracer from scratch, the researchers took clinical-grade [18F]FDG, dried it by azeotropic distillation, and subjected it to Fischer glycosidation with a series of alcohols under acidic conditions. In a single step, and with isolated radiochemical yields ranging from roughly 33 to 92 percent, they generated a library of nine 18F-labelled alkyl glucosides. Screening these compounds against Staphylococcus aureus in vitro revealed that the methyl glucoside, an anomeric mixture of α and β forms, accumulated in the bacteria far more readily than any of its longer-chain counterparts. Because glucosides do not undergo mutarotation under physiological conditions, the team could separate the two anomers by high-performance liquid chromatography and test them independently. The result was unambiguous: the β-anomer drove essentially all of the bacterial uptake, while the α-anomer was nearly inert.

Crucially, the active β-anomer can be manufactured without any new equipment. The researchers synthesized a β-methyl mannose triflate precursor and showed that it could be radiofluorinated and hydrolyzed using the same workflow, and even the same commercially available cassettes, used for routine [18F]FDG production. Manual synthesis delivered the tracer in 74.6 percent radiochemical yield with purity above 99 percent within 40 minutes, and an automated run on a GE FASTlab module produced it in 31.3 percent yield starting from 37 GBq of fluoride in about 34 minutes. In an era when most experimental PET tracers require bespoke radiochemistry, a compound that can be swapped into existing FDG infrastructure is a significant translational advantage. The team also demonstrated that direct glycosidation of clinical [18F]FDG could serve as an alternative route for imaging centers that receive FDG from commercial vendors but lack their own cyclotrons.

In laboratory assays, β-Me-2-[18F]FDG was taken up robustly and consistently by every clinical isolate of methicillin-resistant and methicillin-susceptible S. aureus tested, spanning the USA100 through USA800 pulsed-field types, as well as by methicillin-susceptible and vancomycin-intermediate Staphylococcus epidermidis. Streptococcus agalactiae and Streptococcus pyogenes also accumulated the tracer, though with somewhat more variability. By contrast, uptake in Escherichia coli, Mycobacterium smegmatis and mammalian cell lines was negligible. Heat-killed bacteria did not take up the compound, and uptake was blocked by excess non-radioactive competitor, confirming that accumulation requires live, metabolically active organisms. The tracer was also stable in both mouse and human serum for at least four hours at 37 degrees Celsius, a prerequisite for clinical use.

Genetic experiments pinpointed the transporter responsible. S. aureus encodes four glucose uptake systems: the PTS components GlcA, GlcB and GlcC, and a non-PTS transporter called GlcU. Using a panel of transposon mutants, the researchers showed that deletion of glcA, which encodes the glucose-specific PTS component EIICBAGlc1, dramatically reduced tracer uptake, whereas deletion of glcB, glcC, glcU or manA had no effect. HPLC analysis of bacterial lysates revealed that most intracellular radioactivity corresponded to a phosphorylated metabolite, consistent with the PTS mechanism of import-coupled phosphorylation, which traps the sugar inside the cell and prevents efflux. Notably, β-Me-2-[18F]FDG was not phosphorylated by mammalian hexokinase, reinforcing its selectivity for bacterial biochemistry. Protein sequence analysis showed that EIICBAGlc1 is highly conserved across Staphylococcus species, explaining why the tracer performed well against multiple staphylococcal pathogens.

In healthy mice, the tracer behaved exactly as a bacteria-specific agent should. Dynamic PET/CT imaging showed rapid distribution to lung, liver and kidneys followed by predominantly renal excretion into the bladder, with low retention in brain, heart and other background tissues. Pharmacokinetic modeling yielded a distribution half-life of 0.48 minutes and an elimination half-life of 44.34 minutes, and extrapolated human dosimetry estimated effective doses of 0.020 to 0.041 mSv/MBq depending on sex and the ICRP model applied, with the urinary bladder receiving the highest absorbed dose. Urine analysis confirmed high in vivo metabolic stability, with no detectable degradation to [18F]FDG, an advantage attributed to the stability of β-glycosidic linkages over their α counterparts.

The imaging performance in disease models was striking. In a murine myositis model, sites inoculated with live S. aureus showed a target-to-non-target ratio of 6.1, rising to 14.6 when imaging was delayed to four hours after injection, while sites inoculated with heat-killed bacteria and tissues of mice with lipopolysaccharide-induced sterile inflammation showed no significant uptake. Conventional [18F]FDG, by comparison, lit up brain, spinal cord and brown adipose tissue in inflamed animals, illustrating precisely the background problem the new tracer is designed to solve. In a wound infection model, signal at infected tissue was nearly seventeenfold higher than in uninfected tissue, and in a pulmonary model, infected lungs accumulated 11.3 percent of the injected dose per cubic centimeter versus 0.6 in uninfected lungs. In a rat model of vertebral discitis-osteomyelitis, tracer signal at the inoculated spinal level rose progressively over six days, reaching sixteenfold above baseline, and expanded to adjacent vertebrae in a pattern that mirrors multifocal spondylodiscitis in patients.

Perhaps the most clinically consequential experiments involved treatment monitoring and disseminated infection. In mice dually infected with oxacillin-susceptible and oxacillin-resistant S. aureus, serial PET scans before and after a three-day course of oxacillin showed a significant drop in signal at the susceptible site, from 5.1 to 2.9 percent injected dose per cubic centimeter, while the resistant site showed persistent uptake and visible spread of infection. Ex vivo bacterial counts correlated with the imaging findings. In a bacteremia model, the tracer revealed secondary foci of infection in the brain, heart, liver, shoulder joint and ilium by day three, sites that the gram-negative-targeting comparator tracer [18F]FDS failed to highlight. The researchers note that interpreting PET signals during bloodstream infection requires caution, since circulating bacteria and sepsis-altered tracer kinetics can confound quantification, but the pattern of focal uptake still mapped plausibly onto hematogenous dissemination.

The team, led by David M. Wilson and Sang Hee Lee at UCSF, is now preparing an investigational new drug application, including toxicity testing of the non-radioactive 19F standard, to support first-in-human studies. Because β-Me-2-[18F]FDG targets gram-positive organisms, it is envisioned as a clinical complement to [18F]FDS, which images gram-negative bacteria, allowing tracer selection based on the likely pathogen. If human trials confirm the preclinical profile, the tracer could enable physicians to distinguish infection from sterile inflammation, localize disseminated disease, and judge antibiotic success within days rather than weeks, potentially shortening therapy, sparing patients unnecessary drugs, and offering a new tool in the fight against resistant pathogens such as MRSA.

Subject of Research: Development of a bacteria-specific PET radiotracer targeting the glucose phosphotransferase system for imaging gram-positive infections

Article Title: Selective PET imaging of bacterial infection using a glycosylated 18F-fluorodeoxyglucose-derived tracer

Article References: Lee, S. H., Kim, J. M., López-Álvarez, M., Wadhwa, A., Bidkar, A. P., Ur Rahim, J., Blecha, J., Flavell, R. R., Ordonez, A. A., Seo, Y., Engel, J., Ohliger, M. A., & Wilson, D. M. (2026). Selective PET imaging of bacterial infection using a glycosylated 18F-fluorodeoxyglucose-derived tracer. Nature Biomedical Engineering. https://doi.org/10.1038/s41551-026-01798-1

Image Credits: AI Generated

DOI: 10.1038/s41551-026-01798-1

Keywords: PET imaging, bacterial infection, Staphylococcus aureus, radiotracer, phosphotransferase system, 18F-FDG, MRSA, antimicrobial resistance, molecular imaging, gram-positive bacteria, antibiotic response monitoring, infection diagnostics

The research, conducted by Berkay Çağdaş of the Department of Nuclear Medicine at Afyonkarahisar State Hospital and Ege University’s Institute of Nuclear Sciences, and Elif Kardelen Çağdaş of the hospital’s Department of Pathology and Ankara University’s Graduate School of Biotechnology, set out to determine whether regional differences in immune and stromal composition inside breast tumors correspond to measurable differences in how the tumor consumes glucose on PET imaging. The premise is biologically plausible: tumors infiltrated by lymphocytes often behave differently from immune-cold tumors, and fibroblast-like stromal cells can alter metabolism, perfusion, and tissue architecture in ways that might plausibly change fluorodeoxyglucose uptake.

The study enrolled 61 women with invasive breast carcinoma who underwent baseline 18F-FDG PET/CT before treatment. Rather than treating each tumor as a single homogeneous blob, the team took a deliberately regional approach. A pathologist, blinded to all PET findings, annotated ten separate intratumoral regions on whole-slide histopathology images, each at least one square millimeter in area. QuPath, an open-access machine-learning platform for digital pathology, then quantified two key features within each region: the density of stromal tumor-infiltrating lymphocytes, the immune cells that gather in the connective tissue surrounding cancer cells, and the density of fibroblast-like stromal cells, spindle-shaped cells identified morphologically on routine hematoxylin and eosin stains.

From those ten regions, the investigators retained the ones with the numerically highest and lowest lymphocyte density for paired analysis, an approach designed to capture the observed range of immune-stromal composition within each tumor. The authors are careful to note that these paired regions represent the extremes of what was measured, not a comprehensive spatial map of every tumor. On the imaging side, the team extracted a rich panel of 30 PET features from each patient’s scan. These included the familiar intensity measures such as maximum, mean, and peak standardized uptake values, volumetric measures including metabolic tumor volume and total lesion glycolysis, texture descriptors derived from gray-level co-occurrence, run-length, and size-zone matrices, and dedicated heterogeneity indices, including the HI2 and HI3 indices and a slope calculated across metabolic volume thresholds from 30 to 50 percent of maximum uptake.

The statistical framework was deliberately conservative. The team ran Spearman correlations between nine immune-stromal endpoints and all 30 PET features, generating 270 individual tests, and then applied Benjamini-Hochberg correction to control the false discovery rate, a standard safeguard against being fooled by chance when many comparisons are made simultaneously. The results were sobering but informative. Thirty-eight of the 270 correlations reached nominal significance at the conventional p-value threshold of 0.05, yet not a single one survived the false discovery rate correction, whether the correction was applied within each endpoint family or across the entire study. The smallest corrected q-value within an endpoint was 0.077, and study-wide it was 0.33, both well above any threshold that would conventionally be called significant.

The strongest raw association was, by the authors’ own framing, weak. The density of stromal tumor-infiltrating lymphocytes in the TIL-rich region correlated with the HI2 heterogeneity index at a Spearman coefficient of 0.379 in absolute value, with a 95 percent confidence interval spanning 0.126 to 0.586 and a nominal p-value of 0.0026. Beyond that single headline number, a coherent pattern emerged in the direction of the nominal findings. Lymphocyte density correlated inversely with metabolic tumor volume, with coefficients ranging from negative 0.29 to negative 0.35, and positively with the HI2 heterogeneity measure. The ratio of lymphocytes to fibroblast-like stromal cells correlated with SUV-based parameters, the largest being a coefficient of 0.299 with mean standardized uptake value. Fibroblast density on its own, and the metrics describing regional differences between the paired areas, produced weaker and less consistent signals.

One of the study’s most technically rigorous contributions is its validation of the pathology measurement itself. Because fibroblast-like stromal cells were defined morphologically on routine H&E stains, the authors cross-checked their machine-learning quantification against alpha-smooth muscle actin immunohistochemistry, a stain that highlights activated fibroblasts and myofibroblasts, in the same 122 paired regions. The agreement was remarkably tight: a correlation of 0.992 and a Lin concordance correlation coefficient of 0.996. When the entire correlation analysis was repeated using the immunohistochemistry-based stromal density instead of the H&E-based one, the profile of results was reproduced almost exactly, with correlations between the two analyses ranging from 0.94 to 0.996. This means the null finding is not easily explained by measurement error in the pathology pipeline.

The team also stress-tested the associations that did reach nominal significance. When extreme lymphocyte densities or the smallest annotated regions were excluded from the analysis, between 9 and 24 of the original 38 nominal associations persisted, indicating that some of the signals were fragile and dependent on particular data points. After statistical adjustment for clinical covariates including T stage, Nottingham histologic score, Ki-67 proliferation index, and molecular subtype, only 13 of the 38 associations remained nominally significant. The picture that emerges is one of faint, directionally interesting signals that dissolve under the weight of proper multiple-comparison correction, exactly the pattern one expects when exploring a genuinely uncertain hypothesis with a modest sample size.

Why does this matter? PET radiomics has generated enormous enthusiasm in oncology, with hundreds of studies proposing imaging texture features as noninvasive surrogates for everything from gene expression to immune infiltration. The promise is seductive: if a scan could reveal the immune landscape of a tumor without a biopsy, clinicians could select patients for immunotherapy, track treatment response, and monitor clonal evolution noninvasively. But the field has been repeatedly criticized for underpowered studies, lax statistical correction, and features that fail to replicate. This study is a refreshing counterexample. By pre-specifying a conservative statistical framework, validating its pathology measurements against an independent stain, and openly reporting that nothing survived correction, the authors have modeled the kind of intellectual honesty that imaging science needs more of.

The authors themselves are unambiguous about the limits of their work. This is a hypothesis-generating, exploratory, single-institution study of 61 patients, and the paired-region design captures only the extremes of immune-stromal composition rather than a full spatial map. They state plainly that the findings require independent validation in larger, multi-institutional cohorts before any biological or clinical interpretation can be drawn. That caution is warranted, but so is the value of the attempt. The study demonstrates a feasible, reproducible workflow for integrating region-level digital pathology with PET radiomics in breast cancer, and it establishes a realistic benchmark for how weak the true links between tissue microenvironment and metabolic imaging may actually be. If future larger studies confirm even the modest directional trends seen here, immune-rich tumors may one day be recognizable on PET as smaller, more heterogeneous glucose-avid lesions. For now, the honest answer to the question of whether a PET scan can see a tumor’s immune landscape is: perhaps faintly, and not yet reliably.

Subject of Research: Associations between regional immune-stromal tumor features and 18F-FDG PET metabolic heterogeneity in breast cancer

Article Title: Exploratory associations between regional immune-stromal features and 18F-FDG PET-derived metabolic heterogeneity in breast cancer

Article References: Çağdaş, B., & Çağdaş, E. K. (2026). Exploratory associations between regional immune-stromal features and 18F-FDG PET-derived metabolic heterogeneity in breast cancer. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02789-z

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02789-z

Keywords: breast cancer, PET imaging, radiomics, tumor microenvironment, tumor-infiltrating lymphocytes, digital pathology, metabolic heterogeneity, 18F-FDG, QuPath, fibroblasts, false discovery rate, machine learning

The stakes are high. The liver is among the most frequent sites of distant spread in malignant tumors, and the number of patients with liver metastases substantially exceeds the number diagnosed with primary liver cancer. Prognosis is poor: one-year overall survival for patients with liver metastases has been reported at roughly 15 percent, compared with 24 percent for patients without such spread. Yet early and precise detection matters enormously, because options ranging from surgical resection and ablation to stereotactic radiotherapy and transarterial therapies can be curative or life-extending in selected patients. In colorectal cancer, for example, resection of limited liver metastases can push five-year survival to between 47 and 60 percent, and some patients with non-colorectal liver metastases also benefit from surgery. Accurate imaging is the gateway to those decisions.

The study, conducted at Fudan University Shanghai Cancer Center and published in Holistic Integrative Oncology, enrolled 76 patients between May 2020 and April 2023 who underwent both scans within one week. The cohort spanned 19 different malignant tumor types, comprising 65 patients with carcinomas and 11 with sarcomas, and a total of 189 liver lesions, of which 173 were ultimately confirmed as metastases through at least three months of clinical and imaging follow-up. Two experienced nuclear medicine physicians, blinded to clinical data and follow-up results, independently evaluated the images using a standardized visual scoring system supplemented by quantitative measurements.

The technical logic behind the comparison hinges on the biology of each tracer. Fluorodeoxyglucose, or FDG, is a glucose analogue that accumulates in cells with high glycolytic activity, the metabolic hallmark of many cancers. But FDG is relatively nonspecific: all living cells consume glucose, and the liver’s own background uptake is brisk, which can mask malignant lesions and inflate the false-negative rate. FAPI-04, by contrast, targets fibroblast activation protein, or FAP, a marker abundantly expressed by cancer-associated fibroblasts in the tumor stroma of a broad spectrum of malignancies. Crucially for liver imaging, FAP expression in normal hepatic parenchyma is low, producing a dark, quiet background against which bright metastatic lesions stand out with high contrast.

The results were emphatic. Across all 76 patients, the sensitivity of gallium-68 FAPI-04 PET/CT for detecting liver metastases was 94.80 percent, compared with 69.94 percent for FDG PET/CT, a difference that was highly statistically significant. Accuracy followed the same pattern, at 91.53 percent versus 70.37 percent. In the carcinoma subgroup the gap widened further: sensitivity of 97.87 percent versus 73.05 percent, and accuracy of 96.05 percent versus 74.34 percent. Even in the smaller sarcoma group, FAPI-04 achieved significantly higher sensitivity, 81.25 percent versus 56.25 percent, although accuracy did not reach statistical significance in that subset. Specificity did not differ significantly between the two tracers in any group.

Quantitative uptake measurements reinforced the visual findings. The researchers compared the maximum standardized uptake value, or SUVmax, of each lesion with the mean uptake of normal liver tissue to derive the tumor-to-background ratio, or TBR. Normal liver background activity was markedly lower with FAPI-04 than with FDG, with a mean SUV of 1.10 versus 2.61. Although overall SUVmax of metastases did not differ significantly across the whole cohort, the median TBR for FAPI-04 was more than double that of FDG, 4.60 versus 1.67, and in carcinomas both SUVmax and TBR were significantly higher for FAPI-04. In essence, even when lesion signal was similar, the quieter liver background made FAPI-04 lesions far easier to see.

The clinical consequences of that contrast were tangible. In ten patients, liver metastases were clearly positive on FAPI-04 PET/CT yet invisible to FDG PET/CT. Because distant spread determines the M stage of the TNM classification, detecting these lesions changed the staging of those patients and could alter treatment planning, from surgical candidacy and radiotherapy target delineation to systemic therapy decisions. For patients with limited, oligometastatic liver disease, more aggressive local interventions such as resection, stereotactic ablative radiotherapy, or microwave ablation may be appropriate, but only if all sites of disease are reliably identified first.

The study’s authors contextualized their findings against existing imaging standards. Magnetic resonance imaging with diffusion-weighted sequences and gadoxetic acid contrast remains the reference method for characterizing liver lesions, with pooled sensitivity of about 95 percent and specificity of about 82 percent in meta-analysis. Notably, the carcinoma subgroup performance of FAPI-04 PET/CT in this study approached those figures, making it a credible alternative for patients who cannot receive gadolinium contrast because of renal impairment or allergy. Prior smaller studies in gastrointestinal cancers and mixed tumor populations had already hinted at FAPI’s advantage, with sensitivities of 96.6 to 98.2 percent, but limited sample sizes and narrow tumor spectra left the picture incomplete. The Shanghai team’s broader cohort and explicit carcinoma-versus-sarcoma stratification add statistical weight and biological nuance.

Why did FAPI-04 perform less decisively in sarcomas? The authors point to fundamental differences in tumor origin. Carcinomas arise from epithelial tissue and typically provoke a robust stromal reaction rich in FAP-expressing fibroblasts, whereas sarcomas of mesenchymal origin display heterogeneous and often lower stromal FAP expression. That heterogeneity likely explains both the lower accuracy of 72.97 percent in sarcoma and the absence of a significant SUVmax advantage, even though the tumor-to-background ratio still favored FAPI-04 significantly. The modest specificity observed for both tracers also has recognized culprits: benign lesions such as angiomyolipoma and focal nodular hyperplasia can take up FAPI, inflammatory focal liver lesions can be FAPI-avid, and fibrotic nodules from chronic hepatitis, alcoholic liver disease, or fatty liver disease may trap the tracer, since fibroblast activation protein is a hallmark of activated hepatic stellate cells in fibrosis.

The researchers are careful to frame their conclusions as preliminary. The analysis was retrospective, most metastases were confirmed by follow-up rather than biopsy, the cohort came from a single center, and the sarcoma subgroup was small enough to invite statistical bias. Only the lesion most suspicious for malignancy per patient was generally included, which may have flattered both scanners’ performance. Still, the magnitude and consistency of FAPI-04’s advantage in carcinoma metastases, the modality where detection most often determines resectability, make a compelling case for larger prospective trials, including direct comparisons with liver MRI. If validated, a routine switch in tracer for patients at risk of hepatic spread would be a rare and consequential upgrade in an imaging workhorse that has remained largely unchanged for decades.

Subject of Research: Comparative diagnostic performance of 68Ga-FAPI-04 and 18F-FDG PET/CT imaging for detecting liver metastases from malignant tumors.

Article Title: Head-to-head comparison of the detection performance and tumor uptake of 68Ga-FAPI-04 and 18F-FDG PET/CT in liver metastases from malignant tumors of different types

Article References: Ma, G., Liu, C., Qi, M., Li, J., & Song, S. (2026). Head-to-head comparison of the detection performance and tumor uptake of 68Ga-FAPI-04 and 18F-FDG PET/CT in liver metastases from malignant tumors of different types. Holistic Integrative Oncology, 5(1), Article 74. https://doi.org/10.1007/s44178-026-00295-4

Image Credits: AI Generated

DOI: 10.1007/s44178-026-00295-4

Keywords: 68Ga-FAPI-04, 18F-FDG, PET/CT, liver metastases, molecular imaging, fibroblast activation protein, carcinoma, sarcoma, SUVmax, tumor-to-background ratio, cancer staging, nuclear medicine