[Large Language Models for Rapid Simplification of Quality Assurance Data Input: Field Trial with Real Data in the Context of Tumour Documentation in Urology]. / Large Language Modelle zur schnellen Vereinfachung der Eingabe von Qualitätssicherungsdaten: Performance-Test mit Echtdaten am Beispiel der Tumordokumentation in der Urologie.

INTRODUCTION: Large Language Models (LLMs) such as ChatGPT have rapidly brought the application of artificial intelligence into widespread use. Among many different use cases for text generation and processing, one application is the extraction of data from existing documents and conversations for simplified and automated form-filling. OBJECTIVE: In the field of quality assurance and documentation of cancer diseases, there is currently a significant workload involved in transferring data under various aspects into slightly varying formats and applying interpretations such as the TNM classification of tumours. However, there is a lack of trials with real data to assess the applicability of LLM-supported processes in this area, which would enable an evaluation of efficiency and practicality. This study aims to implement and assess such a trial. METHODOLOGY: A trial was conducted with N=153 privacy-compliant and ethics committee-cleared medical reports from 25 patients. Using the publicly available version of ChatGPT 4.0, an automated script was used to extract the date of initial diagnosis and common tumor classifications. The results were then individually checked for accuracy. Based on this, the utility of a simple system for guided support in tasks related to tumour documentation was assessed. Additionally, the approach was evaluated in terms of operational costs for the model and its applicability. RESULTS: In summary, the study concludes that the use of generative AI in this field is promising and suitable as a tool even in an untrained state. In a simplified calculation, costs of 35 cents are offset by a value creation of 61,54 euros. However, it also becomes clear that AI can only act in a supportive role, and the correct integration with pre-made specific prompts and tools into the workflow is crucial for a relevant performance. CONCLUSION: The use of generative AI in the context of search, transfer, and interpretation tasks in the creation of tumor documentation is a promising approach. However, its implementation in practical applications must be closely monitored, and the optimal interaction between man and machine should continue to be evaluated and must be accompanied by tools and task-specific prompts.

Real-world Outcome of Patients with Advanced Renal Cell Carcinoma and Intermediate- or Poor-risk International Metastatic Renal Cell Carcinoma Database Consortium Criteria Treated by Immune-oncology Combinations: Differential Effectiveness by Risk Group?

BACKGROUND: Renal c carcinoma (RCC) is one of the most common urinary cancers worldwide, with a predicted increase in incidence in the coming years. Immunotherapy, as a single agent, in doublets, or in combination with anti-vascular endothelial growth factor receptor tyrosine kinase inhibitors (TKIs), has rapidly become a cornerstone of the RCC therapeutic scenario, but no head-to-head comparisons have been made. In this setting, real-world evidence emerges as a cornerstone to guide clinical decisions. OBJECTIVE: The objective of this retrospective study was to assess the outcome of patients treated with first-line immune combinations or immune oncology (IO)-TKIs for advanced RCC. DESIGN, SETTING, AND PARTICIPANTS: Data from 930 patients, 654 intermediate risk and 276 poor risk, were collected retrospectively from 58 centers in 20 countries. Special data such as sarcomatoid differentiation, body mass index, prior nephrectomy, and metastatic localization, in addition to biochemical data such as hemoglobin, platelets, calcium, lactate dehydrogenase, neutrophils, and radiological response by investigator's criteria, were collected. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Overall survival (OS) and progression-free survival (PFS) were estimated using the Kaplan-Meier method. The median follow-up was calculated by the inverse Kaplan-Meier method. RESULTS AND LIMITATIONS: The median follow-up time was 18.7 mo. In the 654 intermediate-risk patients, the median OS and PFS were significantly longer in patients with the intermediate than in those with the poor International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) criteria (38.9 vs 17.3 mo, 95% confidence interval [CI] p < 0.001, and 17.3 vs 11.6 mo, 95% CI p < 0.001, respectively). In the intermediate-risk subgroup, the OS was 55.7 mo (95% CI 31.4-55.7) and 40.2 mo (95% CI 29.6-51.6) in patients treated with IO + TKI and IO + IO combinations, respectively (p = 0.047). PFS was 30.7 mo (95% CI 16.5-55.7) and 13.2 mo (95% CI 29.6-51.6) in intermediate-risk patients treated with IO + TKI and IO + IO combinations, respectively (p < 0.001). In the poor-risk subgroup, the median OS and PFS did not show a statistically significant difference between IO + IO and IO + TKI. Our study presents several limitations, mainly due to its retrospective nature. CONCLUSIONS: Our results showed differences between the IO + TKI and IO + IO combinations in intermediate-risk patients. A clear association with longer PFS and OS in favor of patients who received the IO + TKI combinations compared with the IO-IO combination was observed. Instead, in the poor-risk group, we observed no significant difference in PFS or OS between patients who received different combinations. PATIENT SUMMARY: Renal cancer is one of the most frequent genitourinary tumors. Treatment is currently based on immunotherapy combinations or immunotherapy with tyrosine kinase inhibitors, but there are no comparisons between these.In this study, we have analyzed the clinical course of 930 patients from 58 centers in 20 countries around the world. We aimed to analyze the differences between the two main treatment strategies, combination of two immunotherapies versus immunotherapy + antiangiogenic therapy, and found in real-life data that intermediate-risk patients (approximately 60% of patients with metastatic renal cancer) seem to benefit more from the combination of immunotherapy + antiangiogenic therapy than from double immunotherapy. No such differences were found in poor-risk patients. This may have important implications in daily practice decision-making for these patients.

Bone targeting agents, but not radiation therapy, improves survival in patients with bone metastases from advanced urothelial carcinoma receiving pembrolizumab: results from the ARON-2 study.

The ARON-2 study (NCT05290038) aimed to assess the real-world efficacy of pembrolizumab in patients recurred or progressed after platinum-based chemotherapy. This retrospective analysis reports the outcomes of urothelial carcinoma (UC) patients with bone metastases (BM). Medical records of patients with documented metastatic UC treated by pembrolizumab as second-line therapy were reviewed from60 institutions in 20 countries. Patients were assessed for Overall Response Rate (ORR), Progression-Free Survival (PFS), and Overall Survival (OS). Univariate and multivariate analyses were used to explore the association of variables of interest with OS and PFS. 881 patients were included; of them, 263 (30%) presented BM. Median follow-up time was 22.7 months. Patients with BM showed both shorter median OS (5.9 months vs 13.1 months, p < 0.001) and PFS (3.5 months, vs 7.3 months, p < 0.001) compared to patients without BM. Patients who received bone targeted agents (BTAs) showed a significantly longer median OS (8.5 months vs 4.6 months, p = 0.003) and PFS (6.1 months vs 3.2 months, p = 0.003), while no survival benefits were observed among patients who received radiation therapy for BM during pembrolizumab treatment compared to those who did not. In multivariate analysis, performance status, concomitant liver metastases, and the lack of use of BTAs were significantly associated with worse OS and PFS. Bone involvement in UC patients treated with pembrolizumab predicts inferior survival. Poor performance status and liver metastases may further worsen outcomes, while the use of BTAs is associated with improved outcomes.

Geographical differences in the management of metastatic de novo renal cell carcinoma in the era of immune-combinations.

BACKGROUND: The upfront treatment of metastatic renal cell carcinoma (mRCC) has been revolutionized by the introduction of immune-based combinations. The role of cytoreductive nephrectomy (CN) in these patients is still debated. The ARON-1 study (NCT05287464) was designed to globally analyze real-world data of mRCC patients receiving first-line immuno-oncology combinations. This sub-analysis is focused on the role of upfront or delayed partial or radical CN in three geographical areas (Western Europe, Eastern Europe, America/Asia). METHODS: We conducted a multicenter retrospective observational study in mRCC patients treated with first-line immune combinations from 55 centers in 19 countries. From 1152 patients in the ARON-1 dataset, we selected 651 patients with de novo mRCC. 255 patients (39%) had undergone CN, partial in 14% and radical in 86% of cases; 396 patients (61%) received first-line immune-combinations without previous nephrectomy. RESULTS: Median overall survival (OS) from the diagnosis of de novo mRCC was 41.6 months and not reached (NR) in the CN subgroup and 24.0 months in the no CN subgroup, respectively (P<0.001). Median OS from the start of first-line therapy was NR in patients who underwent CN and 22.4 months in the no CN subgroup (P<0.001). Patients who underwent CN reported longer OS compared to no CN in all the three geographical areas. CONCLUSIONS: No significant differences in terms of patients' outcome seem to clearly emerge, even if the rate CN and the choice of the type of first-line immune-based combination varies across the different Cancer Centers participating in the ARON-1 project.

Real-world effectiveness of pembrolizumab as first-line therapy for cisplatin-ineligible patients with advanced urothelial carcinoma: the ARON-2 study.

BACKGROUND: The advent of immune-checkpoint inhibitors has challenged previous treatment paradigms for advanced urothelial carcinoma (UC) in the post-platinum setting as well as in the first-line setting for cisplatin-ineligible patients. In this study, we investigated the effectiveness of pembrolizumab as first-line treatment for cisplatin-ineligible UC. METHODS: Data from patients aged ≥ 18 years with cisplatin-ineligible UC and receiving first-line pembrolizumab from January 1st 2017 to September 1st 2022 were collected. Cisplatin ineligibility was defined according to the Galsky criteria. Thirty-three Institutions from 18 countries were involved in the ARON-2 study. RESULTS: Our analysis included 162 patients. The median follow-up time was 18.9 months (95%CI 15.3-76.9). In the overall study population, the median OS was 15.8 months (95%CI 11.3-32.4). The median OS was significantly longer in males versus females while no statistically significant differences were observed between patients aged < 65y versus ≥ 65y and between smokers and non-smokers. According to Recist 1.1 criteria, 26 patients (16%) experienced CR, 32 (20%) PR, 39 (24%) SD and 55 (34%) PD. CONCLUSIONS: Our data confirm the role of pembrolizumab as first-line therapy for cisplatin-unfit patients. Further studies investigating the biological and immunological characteristics of UC patients are warranted in order to optimize the outcome of patients receiving immunotherapy in this setting.

How handedness influences perceptual and attentional processes during rapid serial visual presentation.

Stimulus-driven orienting of visual attention is lateralized to the right hemisphere (RH). This lateralization has been studied in the dual-stream rapid serial visual presentation task (dual RSVP). In this task a second target (T2), hard to discern by being embedded in one of two lateral streams of rapidly changing distractors, is better identified on the left than on the right. This phenomenon is called the left visual field advantage (LVFA). Furthermore, in terms of event related potentials (ERPs), the N2pc component and visual evoked potentials (VEPs) are evoked earlier at the RH than at the left hemisphere (LH). All previous dual RSVP experiments were performed on right-handers. In the present study it was investigated how the LVFA and its neural correlates are modulated by handedness. To that end, the size of the LVFA and ERPs (VEPs and N2pc) were compared between right- and left-handers. VEPs were evoked earlier at the RH than the LH in right-handers but not in left-handers. Besides this effect, handedness modulated neither the size of the LVFA nor T2-evoked N2pc. Thus, the LVFA seems to be independent of handedness. Rather than for lateralization of attention, handedness might be relevant for lateralization of early visual perception processes.