Impact of body surface area on efficacy and safety in patients with EGFR-mutant non-small cell lung cancer treated with osimertinib as a first-line treatment.

BACKGROUND: The most recommended treatment for stage IV EGFR-positive lung cancer is osimertinib monotherapy. The dosage of osimertinib is fixed at 80 mg/day regardless of body surface area (BSA), however some patients withdraw or reduce the dosage due to adverse events (AEs). METHODS: We performed a retrospective cohort study of 98 patients with EGFR mutation-positive non-small cell lung cancer (NSCLC), who received 80 mg osimertinib as the initial treatment. We investigated the impact of BSA on efficacy and safety of osimertinib. RESULTS: The cut-off value of BSA was estimated using the receiver operating characteristics curve, and was determined to be 1.5 m2. There were 44 patients in the BSA < 1.5 group and 54 patients in the BSA ≥ 1.5 group. There was no significant difference in the incidence of AEs (hematologic toxicity of ≥grade 3 or higher, and non-hematologic toxicity of ≥grade 3) between the two groups. However, the incidence of dose reduction due to AEs was significantly higher in the BSA < 1.5 group compared with the BSA ≥ 1.5 group (16 patients vs 5 patients, p = 0.003). The main reasons were fatigue, anorexia, diarrhea, and liver disfunction. Median progression-free survival (PFS) was not significantly different (16.9 months in the BSA < 1.5 group vs 18.1 months in the BSA ≥ 1.5 group, p = 0.869). CONCLUSION: Differences in BSA affected the optimal dose of osimertinib. However, the PFS with osimertinib treatment was not affected by BSA. Therefore, when using osimertinib as an initial treatment for patients with EGFR-mutant NSCLC, dose reduction to control AEs should be considered, especially in the BSA<1.5 group.

Assessing domain adaptation in adverse drug event extraction on real-world breast cancer records.

BACKGROUND: Adverse Drug Events (ADE) are key information present in unstructured portions of Electronic Health Records. These pose a significant challenge in healthcare, ranging from mild discomfort to severe complications, and can impact patient safety and treatment outcomes. METHODS: We explore the influence of domain shift between a set of dummy clinical notes and a real-world hospital corpus of Japanese clinical notes of breast cancer treatment when extracting ADEs from free text. We annotated a subset of the hospital dataset and used it to fine-tune a Named Entity Recognition (NER) model, initially trained with the set of dummy documents. We used increasing amounts of the annotated data and evaluated the impact on the model's performance. Additionally, we examined the extracted information to identify combinations of drugs that are likely to cause ADEs. RESULTS: We show that domain adaptation can significantly improve model performance in the new domain, as by feeding a small subset of 100 documents for the fine-tuning process we saw a 40% improvement in model performance. However, we also noticed diminishing returns when fine-tuning the model with a larger dataset. For instance, by feeding eight times more data, we only saw further 18% improvement in extraction performance. CONCLUSION: While variations in writing style and vocabulary in clinical corpora can significantly impact the quality of NER results. We show that domain adaptation can be of great aid in mitigating these discrepancies and achieving better performance. Yet, while providing in-domain data to a model helps, there are diminishing returns when fine-tuning with large amounts of data.