Search

Introduction: Large Language Models (LLMs), such as the GPT model family from OpenAI, have demonstrated transformative potential across various fields, especially in medicine. These models can understand and generate contextual text, adapting to new tasks without specific training. This versatility can revolutionize clinical practices by enhancing documentation, patient interaction, and decision-making processes. In oncology, LLMs offer the potential to significantly improve patient care through the continuous monitoring of chemotherapy-induced toxicities, which is a task that is often unmanageable for human resources alone. However, existing research has not sufficiently explored the accuracy of LLMs in identifying and assessing subjective toxicities based on patient descriptions. This study aims to fill this gap by evaluating the ability of LLMs to accurately classify these toxicities, facilitating personalized and continuous patient care. Methods: This comparative pilot study assessed the ability of an LLM to classify subjective toxicities from chemotherapy. Thirteen oncologists evaluated 30 fictitious cases created using expert knowledge and OpenAI’s GPT-4. These evaluations, based on the CTCAE v.5 criteria, were compared to those of a contextualized LLM model. Metrics such as mode and mean of responses were used to gauge consensus. The accuracy of the LLM was analyzed in both general and specific toxicity categories, considering types of errors and false alarms. The study’s results are intended to justify further research involving real patients. Results: The study revealed significant variability in oncologists’ evaluations due to the lack of interaction with fictitious patients. The LLM model achieved an accuracy of 85.7% in general categories and 64.6% in specific categories using mean evaluations with mild errors at 96.4% and severe errors at 3.6%. False alarms occurred in 3% of cases. When comparing the LLM’s performance to that of expert oncologists, individual accuracy ranged from 66.7% to 89.2% for general categories and 57.0% to 76.0% for specific categories. The 95% confidence intervals for the median accuracy of oncologists were 81.9% to 86.9% for general categories and 67.6% to 75.6% for specific categories. These benchmarks highlight the LLM’s potential to achieve expert-level performance in classifying chemotherapy-induced toxicities. Discussion: The findings demonstrate that LLMs can classify subjective toxicities from chemotherapy with accuracy comparable to expert oncologists. The LLM achieved 85.7% accuracy in general categories and 64.6% in specific categories. While the model’s general category performance falls within expert ranges, specific category accuracy requires improvement. The study’s limitations include the use of fictitious cases, lack of patient interaction, and reliance on audio transcriptions. Nevertheless, LLMs show significant potential for enhancing patient monitoring and reducing oncologists’ workload. Future research should focus on the specific training of LLMs for medical tasks, conducting studies with real patients, implementing interactive evaluations, expanding sample sizes, and ensuring robustness and generalization in diverse clinical settings. Conclusions: This study concludes that LLMs can classify subjective toxicities from chemotherapy with accuracy comparable to expert oncologists. The LLM’s performance in general toxicity categories is within the expert range, but there is room for improvement in specific categories. LLMs have the potential to enhance patient monitoring, enable early interventions, and reduce severe complications, improving care quality and efficiency. Future research should involve specific training of LLMs, validation with real patients, and the incorporation of interactive capabilities for real-time patient interactions. Ethical considerations, including data accuracy, transparency, and privacy, are crucial for the safe integration of LLMs into clinical practice.

Objectives The aim of the study was to ascertain the percentage of Spanish lung cancer cases that would fulfil the lung cancer screening inclusion criteria recommended by the United States Preventive Service Task Force (USPSTF) in 2013 and 2021. Methods A cross-sectional study was carried out. All lung cancer cases registered in the Thoracic Tumor Registry with data on date of birth, date of diagnosis, smoking habit, number of pack-years and time elapsed since smoking cessation were included. Results The study included 15 006 patients diagnosed with lung cancer in Spain between 2016 and 2022. Eligibility to participate in screening increased from 53.7% to 63.5% (an increase of 9.8%) according to the 2013 and 2021 recommendations, respectively. The percentage of eligible men rose by 9.2 percentage points with the 2021 versus 2013 recommendations, whereas this rise was 11.5 percentage points in women. Under the 2021 recommendations, 36.6% of women and 5.3% of men would not have fulfilled the screening inclusion criteria due to being never-smokers; 14.9% of women and 11.0% of men would not have fulfilled the age criterion; and 27.0% of ex-smokers among women compared to 35.6% among men would not have been eligible due to >15 years having elapsed since smoking cessation. Conclusions In Spain, over one-third of lung cancer cases could not be detected through screening, by virtue of not meeting the most recent inclusion criteria stated by the USPSTF. The degree of fulfilment in a potential nationwide screening programme should be analysed, with the aim of establishing inclusion criteria in line with each country's context.

Median survival is 10 months and 2-year survival is 20% in metastatic non-small-cell lung cancer (NSCLC) treated with platinum-based chemotherapy. A small fraction of non-squamous cell lung cancers harbor EGFR mutations, with improved outcome to gefitinib and erlotinib. Experimental evidence suggests that BRCA1 overexpression enhances sensitivity to docetaxel and resistance to cisplatin. RAP80 and Abraxas are interacting proteins that form complexes with BRCA1 and could modulate the effect of BRCA1. In order to further examine the effect of EGFR mutations and BRCA1 mRNA levels on outcome in advanced NSCLC, we performed a prospective non-randomized phase II clinical trial, testing the hypothesis that customized therapy would confer improved outcome over non-customized therapy. In an exploratory analysis, we also examined the effect of RAP80 and Abraxas mRNA levels.We treated 123 metastatic non-squamous cell lung carcinoma patients using a customized approach. RNA and DNA were isolated from microdissected specimens from paraffin-embedded tumor tissue. Patients with EGFR mutations received erlotinib, and those without EGFR mutations received chemotherapy with or without cisplatin based on their BRCA1 mRNA levels: low, cisplatin plus gemcitabine; intermediate, cisplatin plus docetaxel; high, docetaxel alone. An exploratory analysis examined RAP80 and Abraxas expression. Median survival exceeded 28 months for 12 patients with EGFR mutations, and was 11 months for 38 patients with low BRCA1, 9 months for 40 patients with intermediate BRCA1, and 11 months for 33 patients with high BRCA1. Two-year survival was 73.3%, 41.2%, 15.6% and 0%, respectively. Median survival was influenced by RAP80 expression in the three BRCA1 groups. For example, for patients with both low BRCA1 and low RAP80, median survival exceeded 26 months. RAP80 was a significant factor for survival in patients treated according to BRCA1 levels (hazard ratio, 1.3 [95% CI, 1-1.7]; P = 0.05).Chemotherapy customized according to BRCA1 expression levels is associated with excellent median and 2-year survival for some subsets of NSCLC patients , and RAP80 could play a crucial modulating effect on this model of customized chemotherapy.(ClinicalTrials.gov) NCT00883480.