Yeon-koo Kang’s research while affiliated with Seoul National University Hospital and other places

Background Fibroblast activation protein (FAP) is prominently involved in the tumour microenvironment and tissue remodelling processes in most cancers, and its expression is also noted in normal skeletal muscle. This study aims to explore the relationship between FAP expression and age‐related muscle characteristics through FAP inhibitor (FAPI) PET/CT imaging. Methods This retrospective analysis studied 54 patients with lung cancer (n = 27) and pancreatic cancer (n = 27) using FAPI PET/CT. Imaging‐based muscle features including the mean standardised uptake value (SUVmean), skeletal muscle index (SMI) and Hounsfield units (HU) were evaluated. Age‐related FAP expression in skeletal muscles was also evaluated using the Genotype‐Tissue Expression (GTEx) dataset. Statistical analyses included Spearman's rank correlation and Kruskal–Wallis test, with a p‐value of less than 0.05 considered significant. Results Analysis revealed a moderate to strong positive correlation between FAPI SUVmean and age (ρ = 0.368, p = 0.006), with older age groups showing higher muscle uptake. Within specific cohorts, the FAPI‐74 group demonstrated a stronger correlation (ρ = 0.500, p = 0.008) compared to the FAPI‐46 group (ρ = 0.319, p = 0.105). SUVmean also correlated negatively with muscle density (HU) (ρ = −0.298, p = 0.029), suggesting an association with higher fat infiltration. GTEx data supported these findings, showing a significant increase in FAP expression across age groups (p < 0.001), with the highest median FAP in the 70–79 age group. Conclusions This study demonstrates an age‐related increase in FAPI uptake in skeletal muscle, correlated with changes in muscle density and fat infiltration. The role of FAP extends beyond pathology to normal muscle, indicating broader biological functions. Accordingly, FAPI PET shows promise for assessing age‐related muscle health and quality.

Purpose The potential of Large Language Models (LLMs) in enhancing a variety of natural language tasks in clinical fields includes medical imaging reporting. This pilot study examines the efficacy of a retrieval-augmented generation (RAG) LLM system considering zero-shot learning capability of LLMs, integrated with a comprehensive database of PET reading reports, in improving reference to prior reports and decision making. Methods We developed a custom LLM framework with retrieval capabilities, leveraging a database of over 10 years of PET imaging reports from a single center. The system uses vector space embedding to facilitate similarity-based retrieval. Queries prompt the system to generate context-based answers and identify similar cases or differential diagnoses. From routine clinical PET readings, experienced nuclear medicine physicians evaluated the performance of system in terms of the relevance of queried similar cases and the appropriateness score of suggested potential diagnoses. Results The system efficiently organized embedded vectors from PET reports, showing that imaging reports were accurately clustered within the embedded vector space according to the diagnosis or PET study type. Based on this system, a proof-of-concept chatbot was developed and showed the framework’s potential in referencing reports of previous similar cases and identifying exemplary cases for various purposes. From routine clinical PET readings, 84.1% of the cases retrieved relevant similar cases, as agreed upon by all three readers. Using the RAG system, the appropriateness score of the suggested potential diagnoses was significantly better than that of the LLM without RAG. Additionally, it demonstrated the capability to offer differential diagnoses, leveraging the vast database to enhance the completeness and precision of generated reports. Conclusion The integration of RAG LLM with a large database of PET imaging reports suggests the potential to support clinical practice of nuclear medicine imaging reading by various tasks of AI including finding similar cases and deriving potential diagnoses from them. This study underscores the potential of advanced AI tools in transforming medical imaging reporting practices.

Pancreatic ductal adenocarcinoma (PDAC) is associated with poor prognosis even without distant metastases, necessitating in-depth characterization of primary tumors for survival prediction. We assessed the feasibility of using FDG-PET/CT radiomics to predict overall survival (OS) in PDAC. This retrospective study included PDAC patients without distant metastasis who underwent FDG-PET/CT for initial staging. Primary tumors were segmented from FDG-PET/CT images, extracting 222 radiomics features. A radiomics-based risk score (Rad-score) was developed using Cox proportional hazards regression with LASSO to predict OS. The prognostic performance of the Rad-score was compared with a clinical model (demographics, disease stage, laboratory results) using Harrell’s concordance index (C-index) and bootstrapping. 140 patients were included, with a mortality rate was 72.9% during follow-up (total population, 19.5 ± 19.2 months; survivors, 34.4 ± 28.8 months). Eleven radiomics features were significant for survival prediction. The Rad-score predicted OS with a C-index of 0.681 [95% CI, 0.632–0.731]. A model integrating clinical parameters and Rad-score outperformed the clinical-only model in predicting OS (C-index 0.740 [0.715–0.816] vs. 0.673 [0.650–0.766]; C-index difference 0.067 [0.014–0.113]; P < 0.001). These findings suggest that incorporating FDG-PET/CT radiomics into preexisting prognotic stratification paradiagm may enhance survival prediction in PDAC, warranting large-scale studies to confirm its applicability in clinical practice. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-024-77022-4.

Pancreatic ductal adenocarcinoma (PDAC) is associated with poor prognosis even without distant metastases, necessitating in-depth characterization of primary tumours for survival prediction. We assessed the feasibility of using FDG-PET/CT radiomics to predict overall survival (OS) in PDAC. This retrospective study included PDAC patients without distant metastasis who underwent FDG-PET/CT for initial staging. Primary tumours were segmented from FDG-PET/CT images, extracting 222 radiomics features. A radiomics-based risk score (Rad-score) was developed using Cox proportional hazards regression with LASSO to predict OS. The prognostic performance of the Rad-score was compared with a clinical model (demographics, disease stage, laboratory results) using Harrell's concordance index (C-index) and bootstrapping. 140 patients were included, with a mortality rate was 72.9% during follow-up (total population, 19.5 ± 19.2 months; survivors, 34.4 ± 28.8 months). Eleven radiomics features were significant for survival prediction. The Rad-score predicted OS with a C-index of 0.681 [95% CI, 0.632–0.731]. A model integrating clinical parameters and Rad-score outperformed the clinical-only model in predicting OS (C-index 0.736 [0.727–0.812] vs. 0.667 [0.648–0.750]; C-index difference 0.069 [0.028–0.117]; P < 0.001). These findings suggest that incorporating FDG-PET/CT radiomics into preexisting prognotic stratification paradiagm may enhance survival prediction in PDAC, warranting large-scale studies to confirm its applicability in clinical practice.

Introduction: The potential of Large Language Models (LLMs) in enhancing a variety of natural language tasks in clinical fields includes medical imaging reporting. This pilot study examines the efficacy of a retrieval-augmented LLM system considering zero-shot learning capability of LLMs, integrated with a comprehensive PET reading reports database, in improving referring previous reports and decision-making. Methods: We developed a custom LLM framework enhanced with retrieval capabilities, leveraging a database encompassing nine years of PET imaging reports from a single center. The system employs vector space embedding of the reports database to facilitate retrieval based on similarity metrics. Queries prompt the system to retrieve embedded vectors, generating context-based answers and identifying similar cases or differential diagnoses from the historical reports database. Results: The system efficiently organized embedded vectors from PET reading reports, showing that imaging reports were accurately clustered within the embedded vector space according to the diagnosis or PET study type. Based on this system, a proof-of-concept chatbot was developed and showed the framework's potential in referencing reports of previous similar cases and identifying exemplary cases for various purposes. Additionally, it demonstrated the capability to offer differential diagnoses, leveraging the vast database to enhance the completeness and precision of generated reports. Conclusions: The integration of a retrieval-augmented LLM with a large database of PET imaging reports represents an advancement in medical reporting within nuclear medicine. By providing tailored, data-driven insights, the system not only improves the relevance of PET report generation but also supports enhanced decision-making and educational opportunities. This study underscores the potential of advanced AI tools in transforming medical imaging reporting practices.

Purpose Mediastinal nodal staging is crucial for surgical candidate selection in non-small cell lung cancer (NSCLC), but conventional imaging has limitations often necessitating invasive staging. We investigated the additive clinical value of fibroblast activation protein inhibitor (FAPI) PET/CT, an imaging technique targeting fibroblast activation protein, for mediastinal nodal staging of NSCLC. Methods In this prospective pilot study, we enrolled patients scheduled for surgical resection of NSCLC based on specific criteria designed to align with indications for invasive staging procedures. Patients were included when meeting at least one of the following: (1) presence of FDG-positive N2 lymph nodes, (2) clinical N1 stage, (3) central tumor location or tumor diameter of ≥ 3 cm, and (4) adenocarcinoma exhibiting high FDG uptake. [⁶⁸Ga]FAPI-46 PET/CT was performed before surgery after a staging workup including [¹⁸F]FDG PET/CT. The diagnostic accuracy of [⁶⁸Ga]FAPI-46 PET/CT for “N2” nodes was assessed through per-patient visual assessment and per-station quantitative analysis using histopathologic results as reference standards. Results Twenty-three patients with 75 nodal stations were analyzed. Histopathologic examination confirmed that nine patients (39.1%) were N2-positive. In per-patient assessment, [⁶⁸Ga]FAPI-46 PET/CT successfully identified metastasis in eight patients (sensitivity 0.89 (0.52–1.00)), upstaging three patients compared to [¹⁸F]FDG PET/CT. [¹⁸F]FDG PET/CT detected FDG-avid nodes in six (42.8%) of 14 N2-negative patients. Among them, five were considered non-metastatic based on calcification and distribution pattern, and one was considered metastatic. In contrast, [⁶⁸Ga]FAPI-46 PET/CT correctly identified all non-metastatic patients solely based on tracer avidity. In per-station analysis, [⁶⁸Ga]FAPI-46 PET/CT discriminated metastasis more effectively compared to [¹⁸F]FDG PET/CT-based (AUC of ROC curve 0.96 (0.88–0.99) vs. 0.68 (0.56–0.78), P < 0.001). Conclusion [⁶⁸Ga]FAPI-46 PET/CT holds promise as an imaging tool for preoperative mediastinal nodal staging in NSCLC, with improved sensitivity and the potential to reduce false-positive results, optimizing the need for invasive staging procedures.

Objective: 18F-FP-CIT positron emission tomography (PET) is known for its high sensitivity and specificity for evaluating striatal dopamine transporter (DAT) binding. Recently, for the early diagnose of Parkinson's disease, many researchers focused on the diagnosis of synucleinopathy in organs involved in non-motor symptoms of Parkinson's disease. We investigated the feasibility of salivary gland uptake on 18F-FP-CIT PET as a new biomarker in patients with parkinsonism. Materials and methods: A total of 219 participants with confirmed or presumed parkinsonism, including 54 clinically diagnosed idiopathic Parkinson's disease (IPD), 59 suspected and yet undiagnosed, and 106 with secondary parkinsonism, were enrolled. The standardized uptake value ratio (SUVR) of the salivary glands was measured on both early and delayed 18F-FP-CIT PET scans using the cerebellum as the reference region. Additionally, the delayed-to-early ratio (DE_ratio) of salivary gland was obtained. The results were compared between patients with different PET patterns. Results: The SUVR in early 18F-FP-CIT PET scan was significantly higher in patients with IPD pattern compared that in the non-dopaminergic degradation group (0.5 ± 0.19 vs. 0.6 ± 0.21, P < 0.001). Compared with the non-dopaminergic degradation group, the DE_ratio was significantly lower in patients with IPD (5.05 ± 1.7 vs. 4.0 ± 1.31, P < 0.001) or atypical parkinsonism patterns (5.05 ± 1.7 vs. 3.76 ± 0.96, P < 0.05). The DE_ratio was moderately and positively correlated with striatal DAT availability in both the whole striatum (r = 0.37, P < 0.001) and posterior putamen (r = 0.36, P < 0.001). Conclusion: Parkinsonism patients with an IPD pattern exhibited a significant increase in uptake on early 18F-FP-CIT PET and a decrease in the DE_ratio in the salivary gland. Our findings suggest that salivary gland uptake of dual-phase 18F-FP-CIT PET can provide diagnostic information on DAT availability in patients with Parkinson's disease.

Although aptamers have shown excellent target specificity in preclinical and clinical studies either by themselves or as aptamer-drug conjugates, their in vivo tissue pharmacokinetic (PK) analysis is still problematic. We aimed to examine the utility of image based positron emission tomography (PET) to evaluate in vivo tissue PK, target specificity, and applicability of oligonucleotides. For this, fluorine-18-labeled aptamers with erb-b2 receptor tyrosine kinase 2 (ERBB2)-specific binding were synthesized by base-pair hybridization using a complementary oligonucleotide platform. To investigate the PKs and properties in vivo tissue, usefulness of in vivo PET imaging in the development of oligonucleotide based drug as an assessment tool was evaluated in normal and tumor xenografted mice. ERBB2-cODN-idT-APs-[18 F]F ([18 F]1), injected intravenously showed significant and rapid uptake in most tissues except for the initial brain and muscle; the uptake was highest in the heart, followed by kidneys, liver, lungs, gall bladder, spleen, and stomach. The main route of excretion was through the renal approximately 77.8% while about 8.3% through the biliary tract of the total dose. The estimated effective dose for an adult woman was 0.00189 mGy/MBq, which might be safe. ERBB2-positive tumor could be well visualized in KPL4 xenograft animal model by in vivo PET imaging. Consequently, the distribution in each organ including ERBB2 expression could be well determined and quantified by PET with fluorine-18-labeled aptamers. In vivo PK parameters such as T1/2 , Tmax , AUC, and Cmax , were also successfully estimated. These results suggest that image based PET with radioisotope-labeled aptamers could be provide valuable information on properties of oligonucleotide based drugs in drug discovery of targeted therapeutics against various diseases.