Tolerability and outcomes of curative radiotherapy in patients aged 85 or more years.

OBJECTIVES: To assess the tolerability and survival outcome of curative radiotherapy in patients over the age of 85 years. DESIGN, SETTING, AND PARTICIPANTS: Retrospective analysis of all patients aged over 85 years who received radiotherapy as part of curative treatment for any cancer (excluding insignificant skin cancers) at the Peter MacCallum Cancer Centre between 1 January 2000 and 1 January 2010. MAIN OUTCOME MEASURES: Poor treatment tolerability (defined as hospital admission during radiotherapy, treatment break, or early treatment cessation); predictors for poor treatment tolerability, overall survival and cancer-specific survival. RESULTS: 327 treatment courses met eligibility criteria. The median age of patients was 87 years. The most common treatment sites were pelvis (30%), head and neck (25%), and breast (18%). The Eastern Cooperative Oncology Group performance status (ECOG PS) score was 0 or 1 for 70% of patients. Overall, 79% of patients completed the prescribed treatment without poor treatment tolerability, and 95% of patients completed all treatment. Only unfavourable ECOG PS score (odds ratio [OR], 1.80; P = 0.005) and increasing age (OR, 1.18; P = 0.018) predicted poor treatment tolerability. ECOG PS score predicted overall survival (hazard ratio, 1.53; P = 0.001). CONCLUSION: Age should not be the sole discriminator in decisions to prescribe aggressive loco-regional radiotherapy. ECOG PS score predicts for treatment tolerability, and also overall survival. The risk of cancer death was higher than non-cancer death for more than 5 years after treatment.

Practical implementation of an existing smoking detection pipeline and reduced support vector machine training corpus requirements.

This study aimed to reduce reliance on large training datasets in support vector machine (SVM)-based clinical text analysis by categorizing keyword features. An enhanced Mayo smoking status detection pipeline was deployed. We used a corpus of 709 annotated patient narratives. The pipeline was optimized for local data entry practice and lexicon. SVM classifier retraining used a grouped keyword approach for better efficiency. Accuracy, precision, and F-measure of the unaltered and optimized pipelines were evaluated using k-fold cross-validation. Initial accuracy of the clinical Text Analysis and Knowledge Extraction System (cTAKES) package was 0.69. Localization and keyword grouping improved system accuracy to 0.9 and 0.92, respectively. F-measures for current and past smoker classes improved from 0.43 to 0.81 and 0.71 to 0.91, respectively. Non-smoker and unknown-class F-measures were 0.96 and 0.98, respectively. Keyword grouping had no negative effect on performance, and decreased training time. Grouping keywords is a practical method to reduce training corpus size.