Publications by authors named "I Locke"
Background: Pneumonitis is a well-described, potentially disabling, or fatal adverse effect associated with both immune checkpoint inhibitors (ICI) and thoracic radiotherapy. Accurate differentiation between checkpoint inhibitor pneumonitis (CIP) radiation pneumonitis (RP), and infective pneumonitis (IP) is crucial for swift, appropriate, and tailored management to achieve optimal patient outcomes. However, correct diagnosis is often challenging, owing to overlapping clinical presentations and radiological patterns.
View Article and Find Full Text PDFLocoregional control and toxicity following high-dose hypofractionated and accelerated palliative radiotherapy regimens in breast cancer.
Clin Oncol (R Coll Radiol)
September 2023
Aims: For patients with locally advanced primary/recurrent breast cancer, radiotherapy is an effective treatment for locoregional control. 36 Gy in 6 Gy once-weekly fractions is a commonly used schedule, but there are no data comparing local control and toxicity between 36 Gy delivered once-weekly versus accelerated schedules of multiple 6 Gy fractions per week. This retrospective study compared local control rates and acute and late toxicity in patients undergoing 30-36 Gy in 6 Gy fractions over 6 weeks versus more accelerated schedules over 2-3 weeks for an unresected breast cancer.
View Article and Find Full Text PDFArticle Synopsis
- Recurrence of lung cancer after radiotherapy occurs in up to 36% of patients, highlighting the need for better prediction of who is at higher risk.
- Researchers developed radiomic classification models using CT scans from over 900 patients with NSCLC to predict overall survival (OS), recurrence-free survival (RFS), and recurrence rates two years post-treatment.
- The models showed promising results in predicting outcomes and could be used to create personalized surveillance strategies, potentially leading to improved patient care in future clinical trials.
Article Synopsis
- A study focused on non-small cell lung cancer (NSCLC) patients, aimed to develop and validate machine learning models using patient, tumor, and treatment data for predicting recurrence, recurrence-free survival (RFS), and overall survival (OS) following radiotherapy.
- The research included 657 patients from 5 hospitals and involved various data pre-processing and machine learning techniques to create risk-stratification models, assessed through cross-validation and external testing.
- Findings indicated that the machine learning models outperformed traditional TNM stage and performance status assessments in predicting recurrence and overall survival, with promising AUC scores demonstrating their effectiveness.