Moritz B. Bastian

Large language models (LLMs) are increasingly explored for clinical use. However, the extent to which such models can reliably support physicians in reporting, staging, and the assessment of classification remains an active area of research. This study aimed to evaluate and compare multiple LLMs for automated PROMISE V2 classification for prostate cancer. A total of 126 unambiguous German-language PSMA PET/CT text reports were retrospectively analyzed, with reference standards established by expert consensus based on image interpretation and the original report text. Five LLMs (GPT-5.4, DeepSeek-V3.2, Claude Sonnet 4.6, Gemini 3 Flash and Grok 4) were assessed using two English-language prompting strategies of varying complexity. Agreement with the reference standard served as the primary endpoint. Performance varied in the short-prompt setting (36.5–79.4%) but improved consistently with the long prompt (74.6–86.5%), with Gemini 3 Flash achieving the highest agreement. Across PROMISE V2 subcategories, agreement rates were high (miT: 81.0–92.1%, miN: 92.9–96.0%, miM: 92.9–95.2%), despite inter-model differences. In conclusion, contemporary LLMs demonstrate promising performance in deriving PROMISE V2 scores from unambiguous original report texts, particularly when guided by detailed prompts.

Introduction: To enhance the effectiveness of radioligand therapy (RLT), new prostate-specific membrane antigen (PSMA) targeting radiopharmaceuticals utilizing alternative radionuclides are under active investigation. One promising alternative to the established ¹⁷⁷ Lu is the use of ¹⁶¹ Tb. A key advantage of ¹⁶¹ Tb is its emission of a higher proportion of low-energy conversion and Auger electrons, which may contribute to enhanced therapeutic effectiveness. This study provides a first exploratory evaluation of the efficacy and safety of [ ¹⁶¹ Tb]Tb-PSMA-617 in patients with progression after [ ¹⁷⁷ Lu]Lu-PSMA-617, aiming to generate early clinical insights. Methods: The study included 15 patients with mCRPC, who were enrolled in the “prospective registry to assess outcome and toxicity of targeted radionuclide therapy in patients with mCRPC in clinical routine“ (REALITY Study; NCT04833517). All patients had received conventional PSMA RLT using [ ¹⁷⁷ Lu]Lu-PSMA-617, which resulted in initial biochemical response (PSA decline ≥50%), followed by a biochemical relapse, prompting the re-initiation of PSMA RLT using [ ¹⁶¹ Tb]Tb-PSMA-617. Patients received a median of 3 (range: 2–7) cycles with mean administered activity of 5.4 ± 1.1 GBq and mean cumulative activity of 19.2 ± 6.4 GBq [ ¹⁶¹ Tb]Tb-PSMA-617. Treatment response was assessed both biochemically by serum PSA levels and through molecular imaging by total lesion PSMA (TLP) on [ ⁶⁸ Ga]Ga-PSMA-11 PET/CT scans. Adverse events were assessed at baseline and follow-up using the “Common Terminology Criteria for Adverse Events“ (Version 5.0). Results: [ ¹⁶¹ Tb]Tb-PSMA-617 RLT showed response rates of 66.7% (10/15 patients) based on biochemical assessment and 86.7% (13/15 patients) based on molecular imaging assessment. Beginning with the initiation of [ ¹⁶¹ Tb]Tb-PSMA-617 RLT, the median progression-free survival (PFS) was 6.4 months, and the median overall survival (OS) was 15.5 months. In total, 6 CTCAE grade deteriorations from grade 2 to grade 3 or from grade 3 to grade 4 were observed. No discontinuation of PSMA RLT due to adverse events was reported. Conclusions: [ ¹⁶¹ Tb]Tb-PSMA-617 RLT emerges as a promising treatment option, demonstrating encouraging response rates, preliminary clinical outcomes, and a favorable safety profile as second-line RLT in progressing patients who previously benefited from [ ¹⁷⁷ Lu]Lu-PSMA-617 RLT.

Article: FDG PET/CT imaging in recurrent papillary thyroid cancer: does BRAF mutation status matter? - The Quarterly Journal of Nuclear Medicine and Molecular Imaging 2026 Mar 30 - Minerva Medica - Journals

Background: Aim of this study was to analyze the safety of prostate-specific membrane antigen radioligand therapy (PSMA-RLT) in patients with metastatic castration-resistant prostate cancer (mCRPC) with preexisting mild to moderate leukopenia (CTCAE ≥ 1). Results: Thirty-seven mCRPC patients with preexisting leukopenia (leukocyte count < 3.8 × 10⁹/L) were included in this study. Patients received a median of 3 cycles of [¹⁷⁷Lu]Lu-PSMA-617 (range 1–9). No significant difference in leukocyte counts was observed between baseline and follow-up after each PSMA-RLT cycle: first cycle (3.0 ± 0.5 at baseline vs. 3.4 ± 1.4 at follow up [in × 10⁹/L], p = 0.0921), second cycle (3.1 ± 0.4 vs. 3.8 ± 1.7, p = 0. 0.0509), and third cycle (3.1 ± 0.4 vs. 3.2 ± 2.0, p = 0.2929), respectively. Similarly, baseline and end of treatment values, irrespective of the number of administered cycles, did not reveal a significant difference (3.0 ± 0.5 vs. 3.5 ± 1.4, p = 0.0684). After the end of therapy, irrespective of the number of administered cycles, 27% patients remained stable in terms of CTCAE scoring, 46% changed to a higher score and 27% improved to a lower score. Conclusion: Although marked preexisting leukopenia is often considered a relative contraindication for PSMA-RLT, our findings indicate that PSMA-RLT is feasible in patients with leukopenia of CTCAE grade ≥ 1. In our cohort, leukocyte counts remained stable without significant RLT-induced deterioration. Therefore, patients with leukopenia should not be categorically excluded from receiving PSMA-RLT. Trial registration: Clinicaltrials.gov identifier: NCT04833517, registered 01.01.2016.

Purpose: Given the increasing inclusion of ECOG 3 patients in oncology practice, data on this subgroup in the context of prostate-specific membrane antigen (PSMA)-targeted radioligand therapy (RLT) remain limited. This study evaluates the safety and outcome of PSMA-RLT in metastatic castration-resistant prostate cancer (mCRPC) patients with ECOG performance status 3. Methods: In this analysis, a cohort of 18 mCRPC patients with ECOG performance status 3 who received PSMA-RLT was examined. The median number of treatment cycles was 2 (range: 1–10), with a mean cumulative administered activity of 21.5 ± 15.0 GBq (range: 2.7–62.6 GBq) of [ ¹⁷⁷ Lu]Lu-PSMA-617. Outcome and adverse events including hematologic and renal toxicities, fatigue, and xerostomia were analyzed. Results: 50% of patients achieved either stable disease or a partial biochemical response. Median progression-free survival and overall survival were 1.3 months and 2.8 months, respectively. Severe adverse events were uncommon, occurring in three patients: one developed grade 3 leukopenia, another experienced grade 3 thrombocytopenia, and one patient had pancytopenia of grade 3/4. No higher RLT-induced grade of renal toxicity and xerostomia were observed, whilst symptoms of fatigue improved in the cohort. Conclusion: This study indicates that PSMA-RLT is a feasible and overall well-tolerated treatment for mCRPC ECOG 3 patients with manageable toxicity profile. Despite limited survival outcomes, ECOG 3 status may be considered not to be a categorical exclusion criterion for RLT. Future prospective studies should further investigate the role of PSMA-RLT in this challenging subgroup.

Purpose: With increasing life expectancy, the number of older patients with metastatic castration-resistant prostate cancer (mCRPC) continues to rise, but this group is currently underrepresented in clinical trials. This study aims to assess the efficacy and safety of prostate-specific membrane antigen (PSMA) targeted radioligand therapy (RLT) in patients over 85 years old with mCRPC. Materials and Methods: A retrospective analysis was conducted on 21 mCRPC patients aged 85 years or older receiving PSMA-RLT (range: 85–96 y). The median number of treatment cycles was 4 (range: 1–13 cycles), with a mean cumulative administered activity of 26.9±17.3 GBq (range: 8.0–87.2 GBq) of [¹⁷⁷Lu]Lu-PSMA-617. Adverse events, including anemia, thrombocytopenia, leukopenia, renal impairment, fatigue, and xerostomia were classified according to CTCAE version 5.0. Results: The mean best PSA response across the cohort was –37.6±73.3%, with 62% (13 of 21 patients) showing partial remission. Median progression-free survival and overall survival were 4.6 and 14.7 months, respectively. Severe adverse events were rare, occurring in 2 patients: 1 case of grade 3 anemia and 1 of grade 3 thrombocytopenia. Conclusions: The results demonstrate that PSMA-RLT is an effective and well-tolerated treatment for mCRPC patients over 85 years, with clinical outcomes and safety comparable to those observed in studies involving younger populations. This highlights its potential as a valuable therapeutic option for this aging patient group.

We present an ¹⁸F-fluorodeoxyglucose ([¹⁸F]FDG) positron emission tomography/computed tomography (PET/CT) scan of a 27 y/o patient with long-standing significant B symptoms, diffuse bone pain, increased inflammation parameters, and polydipsia revealing multiple FDG-avid osteolytic lesions of the axial skeleton including a vertebra plana of T7 and paraosseous soft tissue lesions. A CT-guided biopsy confirmed the diagnosis of Langerhans cell histiocytosis (LCH). This case highlights the importance of considering LCH in young patients with vertebral collapse and underscores the role of PET/CT imaging in establishing an accurate diagnosis.

Background: Hypoxic-ischemic brain injury (HIBI) is a feared complication post-cardiac arrest (CA). The timing of brain imaging remains a topic of ongoing debate. Early computed tomography (CT) scans can reveal acute intracranial pathologies but may have limited predictive value due to delayed manifestation of HIBI-related changes. Radiomics analyses present a promising approach to identifying subtle imaging markers, potentially aiding early HIBI detection. Methods: This study retrospectively assessed post-CA patients between 2016 and 2023 who received immediate brain CTs. Patients without acute intracranial pathology on initial scans and who underwent follow-up brain CTs within 14 days post-return of spontaneous circulation (ROSC) were included. Image segmentation involved manual basalganglia segmentation and automated whole-brain segmentation. Radiomics features were calculated using Pyradiomics (v3.0.1) in 3DSlicer (v5.2.2). Feature selection involved reproducibility analysis via ICC and LASSO regression, retaining five features per segmentation method. A logistic regression model for each segmentation method underwent 5-fold cross-validation. Results were summarized with ROC analyses and average sensitivity and specificity. Results: A total of 83 patients (average age: 65 ± 13.3 years, 19 women) with CA and ROSC were included. Follow-up CT scans after 5.2 ± 2.9 days revealed brain edema in 47 patients. The model using manual segmentation achieved an average AUC of 0.76, sensitivity of 0.59, and specificity of 0.78. The automated segmentation model showed an average AUC of 0.66, sensitivity of 0.49, and specificity of 0.68. Conclusions: Radiomics, particularly focused on the basalganglia area in normal-appearing brain CTs after CA and ROSC, may enhance predictive insights for HIBI and the development of brain edema.

Background: Even though the introduction of ¹⁷⁷Lu-PSMA-617 RLT represents a major milestone in the treatment of mCRPC, there are still patients who do not respond adequately to this therapy and for whom there are only limited options left. Augmenting ¹⁷⁷Lu-PSMA-617 RLT with the alpha-emitter ²²⁵Ac-PSMA-617 may present an escalating treatment option to increase efficacy. In this study, we aim to evaluate outcome and safety of ²²⁵Ac-PSMA-617 augmentation to ¹⁷⁷Lu-PSMA-617 RLT in patients who present insufficient response to monotherapy with ¹⁷⁷Lu-PSMA-617 RLT. Patients and Methods: The study included n = 51 mCRPC patients enrolled in a prospective registry receiving ¹⁷⁷Lu-PSMA-617 monotherapy and showing insufficient response, followed by initiation of ²²⁵Ac-PSMA-617 augmentation, with adjusted activities depending on individual patient characteristics. Biochemical response, progression-free survival, and overall survival were assessed. Adverse events were evaluated according to the Common Terminology Criteria for Adverse Events (CTCAE v.5.0). Results: After initiation of ²²⁵Ac-PSMA-617 augmentation to ¹⁷⁷Lu-PSMA-617 RLT, 24/51 patients (47.1%) exhibited partial remission, 16/51 (31.4%) stable disease, and 11/51 (21.6%) progressive disease. The median progression-free survival and overall survival rates were 6.3 months and 9.1 months, respectively. The majority of CTCAE gradings remained stable after initiating augmentation. Severe adverse events were rare, and no treatment termination due to side effects was recorded. Conclusions: ²²⁵Ac-PSMA-617 augmented ¹⁷⁷Lu-PSMA-617 radioligand therapy seems to be an effective escalating treatment option in patients after failure of conventional ¹⁷⁷Lu-PSMA-617 RLT and represents a promising approach balancing antitumor effect and tolerable side effects.

Purpose: This study aims to evaluate the outcome and renal safety of prostate-specific membrane antigen (PSMA)–radioligand therapy (RLT) in patients with metastatic castration-resistant prostate carcinoma (mCRPC) and preexisting renal impairment. Methods: Ninety-four patients with preexisting renal impairment were included in this retrospective analysis. Inclusion criterion was a glomerular filtration rate (GFR) of ≤60 mL/min (equivalent to Common Terminology Criteria of Adverse Events [CTCAE] ≥2). Patients underwent either [¹⁷⁷Lu]Lu-PSMA-617 RLT exclusively (n = 63) or additionally in augmented manner with [²²⁵Ac]Ac-PSMA-617 (n = 31). The median number of administered cycles was 4 (range, 1–16 cycles) with a mean cumulative activity of 29.9 ± 16.3 GBq (range, 6.9–87.2 GBq) [¹⁷⁷Lu]Lu-PSMA-617. Main blood parameters of interest were creatinine, cystatin C, and the respective GFR values. Changes in GFR were categorized according to CTCAE v5.0. Results: In the entire cohort, mean best PSA response was −56.73% ± 45.71%, with 63 of 94 patients (67%) experiencing partial remission. The median progression-free survival and overall survival were 6.7 and 14.1 months, respectively. Under PSMA-RLT, 5 of 94 patients (5.3%) improved to CTCAE grade 0, and 23 of 94 (24.5%) improved to CTCAE grade 1. Three of 94 patients (3.2%) improved from CTCAE grade 3 to grade 2, and only 5 of 94 (5.3%) decreased. The majority (58/94 [61.7%]) of patients stayed stable in terms of CTCAE grading. Conclusion: PSMA-RLT is an effective and safe treatment in mCRPC patients with preexisting impaired renal function (CTCAE ≥2). In daily clinical practice, patients should not be categorically excluded from enrolment to PSMA-RLT due to renal impairment.