Cristiano Pini

The capability of Large Language Models (LLMs) to support and facilitate research activities has sparked growing interest in their integration into scientific workflows. This paper aims to evaluate and compare against human researchers the performance of 6 different LLMs in conducting the various tasks necessary to produce a systematic literature review. The evaluation of the 6 LLMs was split into 3 tasks: literature search, article screening and selection (task 1); data extraction and analysis (task 2); final paper drafting (task 3). Their results were compared with a human-produced systematic review on the same topic, serving as reference standard. The evaluation was repeated on two rounds to evaluate between-version changes and improvements of LLMs over time. Out of the 18 scientific articles to be extracted from the literature for task 1, the best LLM managed to identify 13. Data extraction and analysis for task 2 was only partially accurate and cumbersome. The full papers generated by LLMs for task 3 were short and uninspiring, often not fully adhering to the standard PRISMA 2020 template for a systematic review. Currently, LLMs are not capable of conducting a scientific systematic review in the medical domain without prompt-engineering strategies. However, their capabilities are advancing rapidly, and, with an appropriate supervision they can provide valuable support throughout the review process.

Rationale: Metastatic prostate cancer in the castration-resistant (mCRPC) setting remains challenging to treat. Prostate-specific membrane antigen (PSMA)-targeted alpha therapy (TAT) is emerging as a promising option. We aimed to systematically review the efficacy and safety of PSMA-TAT in patients with prostate cancer.

Background: Although many well-known factors affect the maximum standardized uptake value (SUVmax), it remains the most requested and used parameter, especially among clinicians, despite other parameters, such as the standardized uptake value corrected for lean body mass and the metabolic tumor volume, being proven to be less sensitive to the same factors, more robust, and eventually more informative. This study intends to provide robust evidence regarding the diagnostic and prognostic value of SUVmax in a large cohort of subjects with suspected malignant lung nodules imaged by [¹⁸F]FDG PET/CT. Materials and methods: We performed a retrospective analysis of patients with suspected/confirmed primary lung tumours undergoing [18F]FDG PET/CT. The sample size was 567 patients. Demographics, imaging, surgical, histological, and follow-up data were collected. SUVmax was analysed according to histology, stage, scanner, and outcome. The impact on measured values of different reconstruction protocols was assessed. All potential predictors of patients’ outcome were assessed. Results: 91% cases were primary lung tumours. Lung benign nodules or metastases accounted for 5% and 4% of cases. Most patients presented with adenocarcinoma (70%) and stage I disease (51%); 144 patients relapsed and 55 died. SUVmax failed to effectively differentiate benign lesions from primary tumours or metastases. Stage I patients presented lower SUVmax. SUVmax significantly correlated with patient weight, injected [¹⁸F]FDG activity, and lesion size and differed between reconstructions’ protocols. Survival analyses revealed no independent prognostic significance for SUVmax in progression-free after adjusting for other variables. SUVmax correlated with overall survival, disease stage and tumour histotype. Conclusion: Our study confirms that SUVmax, though widely employed, present relevant limitations in discriminating between benign lesion and lung cancer, in classifying cancer histotypes, and in predicting patient outcomes independently. Known influencing factors significantly impact on numerical values, thus SUV values should be regarded with caution in clinical practice.

Purpose: For several years, oncological positron emission tomography (PET) has developed beyond 2-deoxy-2-[¹⁸F]fluoro-D-glucose ([¹⁸F]FDG). This umbrella review of meta-analyses aims to provide up-to-date, comprehensive, high-level evidence to support appropriate referral for a specific radiopharmaceutical PET/computed tomography (CT) or PET/magnetic resonance (MR) in the diagnosis and staging of solid cancers other than brain malignancies. Methods: We performed a systematic literature search on the PubMed/MEDLINE and EMBASE databases for meta-analyses assessing the accuracy of PET/CT and/or PET/MRI with [¹⁸F]FDG, somatostatin- receptor-targeting ⁶⁸Ga-DOTA-peptides, ¹⁸F-labelled dihydroxyphenylalanine ([¹⁸F]DOPA), prostate-specific membrane antigen (PSMA)-targeted radioligands, and fibroblast activation protein inhibitors (FAPI) in the diagnosis/disease characterisation and staging of solid cancers other than brain tumours. Results: The literature search yielded 449 scientific articles. After screening titles and abstracts and applying inclusion and exclusion criteria, we selected 173 meta-analyses to assess the strength of evidence. One article was selected from references. Sixty-four meta-analyses were finally considered. The current evidence corroborates the role of [¹⁸F]FDG as the main player in molecular imaging; PSMA tracers are useful in staging and re-staging prostate cancer; somatostatin-targeting peptides (e.g. [⁶⁸Ga]Ga- DOTA-TOC and -TATE) or [¹⁸F]DOPA are valuable in neuroendocrine tumours (NETs). FAPI has emerged in gastric cancer assessment. According to search and selection criteria, no satisfactory meta-analysis was selected for the diagnosis/detection of oesophageal cancer, the diagnosis/detection and N staging of small cell lung cancer and hepatic cell carcinoma, the diagnosis/detection and M staging of melanoma and Merkel cell carcinoma, cervical, vulvar and penis cancers, the N and M staging of lung and gastroenteropancreatic NET, testicular cancer, and chondrosarcoma, and the M staging of differentiated thyroid, bladder and anal cancers. Conclusion: The comprehensive high-level evidence synthesised in the present umbrella review serves as a guiding compass for clinicians and imagers, aiding them in navigating the increasingly intricate seascape of PET examinations.

Background/Objectives: Disease recurrence and resistance to radioiodine (RAI) therapy are major challenges in the management of differentiated thyroid cancer (DTC). In particular, the TENIS (Thyroglobulin Elevated Negative Iodine Scintigraphy) syndrome, characterised by elevated thyroglobulin (Tg) serum levels in addition to a negative radioiodine whole body scan (WBS), complicates disease monitoring and treatment decisions. Conventional imaging techniques often fail to detect disease in WBS-negative patients with rising Tg levels, leading to limitations in therapeutic intervention. This systematic review and meta-analysis aims to evaluate the diagnostic accuracy of nuclear imaging modalities in detecting disease recurrence in patients with the TENIS syndrome and to provide insights to guide therapeutic approaches in this complex clinical scenario. Methods: A comprehensive search of PubMed/MEDLINE and EMBASE databases up to March 2024 was performed according to PRISMA guidelines. Eligible studies were selected, and quality assessment was performed with the QUADAS-2 tool. For each study, relevant data were extracted and synthesised. A meta-analysis of the diagnostic accuracy of [¹⁸F]FDG PET/CT was performed, and patient-based pooled sensitivity and specificity were calculated using a random-effects model. Statistical heterogeneity between studies was assessed using the I2 statistic. Results: Of the 538 studies initially identified, 22 were included in the systematic review, of which 18 were eligible for meta-analysis. The eligible studies, mainly focused on [¹⁸F]FDG PET/CT, showed variable sensitivity and specificity for the detection of RAI-refractory thyroid cancer lesions. For [¹⁸F]FDG PET/CT, pooled estimates displayed a sensitivity of 0.87 (95% CI: 0.82–0.90) and a specificity of 0.76 (95% CI: 0.61–0.86), with moderate heterogeneity between studies. Conclusions: [¹⁸F]FDG PET/CT remains central in the detection of disease recurrence in patients with the TENIS syndrome. The emergence of novel radiopharmaceuticals with specific molecular targets is a promising way to overcome the limitations of [¹⁸F]FDG in these patients and to open new theranostics perspectives. This review highlights the great potential of nuclear medicine in guiding therapeutic strategies for RAI-refractory thyroid cancer.

We investigated the therapeutic effect of head down positioning at −15° (head down tilt; HDT15) on cerebral collateral flow and infarct growth in a rat model of large vessel occlusion (LVO) stroke, using multi-modal MRI. Twenty-eight Wistar rats were randomly assigned to HDT15 or flat position for 60 minutes, starting 30 minutes after occlusion of the middle cerebral artery, followed by reperfusion. The perfusion shift analysis, comparing post- versus pre-treatment voxel-level changes in time-to-peak perfusion maps, showed a significant increase in cerebral perfusion in the HDT15 group (common odds ratio 1.50; 95% CI 1.41-1.60; p < 0.0001), but not in the flat group (common odds ratio 0.97; 95% CI 0.92-1.03; p = 0.3503). Infarct growth at 24 hours was + 31.4% in the flat group (343 versus 250 mm ³ ; 95% CI 2.4 to 165.1; p = 0.0447) and + 15.4% in the HDT15 group (224 versus 192 mm ³ ; 95% CI -26.9 to 85.9; p = 0.2272). Our findings indicate that HDT15 acutely increases cerebral perfusion in LVO acute ischemic stroke and provides a tissue-saving effect before recanalization. Further research is needed to develop HDT15 as an emergency therapy to acutely increase collateral flow in ischemic stroke prior to recanalization therapy.