Dataset Information

A prediction model for early death in non-small cell lung cancer patients following curative-intent chemoradiotherapy.

ABSTRACT: BACKGROUND:Early death after a treatment can be seen as a therapeutic failure. Accurate prediction of patients at risk for early mortality is crucial to avoid unnecessary harm and reducing costs. The goal of our work is two-fold: first, to evaluate the performance of a previously published model for early death in our cohorts. Second, to develop a prognostic model for early death prediction following radiotherapy. MATERIAL AND METHODS:Patients with NSCLC treated with chemoradiotherapy or radiotherapy alone were included in this study. Four different cohorts from different countries were available for this work (N?=?1540). The previous model used age, gender, performance status, tumor stage, income deprivation, no previous treatment given (yes/no) and body mass index to make predictions. A random forest model was developed by learning on the Maastro cohort (N?=?698). The new model used performance status, age, gender, T and N stage, total tumor volume (cc), total tumor dose (Gy) and chemotherapy timing (none, sequential, concurrent) to make predictions. Death within 4 months of receiving the first radiotherapy fraction was used as the outcome. RESULTS:Early death rates ranged from 6 to 11% within the four cohorts. The previous model performed with AUC values ranging from 0.54 to 0.64 on the validation cohorts. Our newly developed model had improved AUC values ranging from 0.62 to 0.71 on the validation cohorts. CONCLUSIONS:Using advanced machine learning methods and informative variables, prognostic models for early mortality can be developed. Development of accurate prognostic tools for early mortality is important to inform patients about treatment options and optimize care.

SUBMITTER: Jochems A

PROVIDER: S-EPMC6108087 | biostudies-literature | 2018 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A prediction model for early death in non-small cell lung cancer patients following curative-intent chemoradiotherapy.

Jochems Arthur A El-Naqa Issam I Kessler Marc M Mayo Charles S CS Jolly Shruti S Matuszak Martha M Faivre-Finn Corinne C Price Gareth G Holloway Lois L Vinod Shalini S Field Matthew M Barakat Mohamed Samir MS Thwaites David D de Ruysscher Dirk D Dekker Andre A Lambin Philippe P

Acta oncologica (Stockholm, Sweden) 20171014 2

<h4>Background</h4>Early death after a treatment can be seen as a therapeutic failure. Accurate prediction of patients at risk for early mortality is crucial to avoid unnecessary harm and reducing costs. The goal of our work is two-fold: first, to evaluate the performance of a previously published model for early death in our cohorts. Second, to develop a prognostic model for early death prediction following radiotherapy.<h4>Material and methods</h4>Patients with NSCLC treated with chemoradiothe ...[more]

PMID: 29034756

Similar Datasets

Project description:BackgroundSurveillance is universally recommended for non-small cell lung cancer (NSCLC) patients treated with curative-intent radiotherapy. High-quality evidence to inform optimal surveillance strategies is lacking. Machine learning demonstrates promise in accurate outcome prediction for a variety of health conditions. The purpose of this study was to utilise readily available patient, tumour, and treatment data to develop, validate and externally test machine learning models for predicting recurrence, recurrence-free survival (RFS) and overall survival (OS) at 2 years from treatment.MethodsA retrospective, multicentre study of patients receiving curative-intent radiotherapy for NSCLC was undertaken. A total of 657 patients from 5 hospitals were eligible for inclusion. Data pre-processing derived 34 features for predictive modelling. Combinations of 8 feature reduction methods and 10 machine learning classification algorithms were compared, producing risk-stratification models for predicting recurrence, RFS and OS. Models were compared with 10-fold cross validation and an external test set and benchmarked against TNM-stage and performance status. Youden Index was derived from validation set ROC curves to distinguish high and low risk groups and Kaplan-Meier analyses performed.FindingsMedian follow-up time was 852 days. Parameters were well matched across training-validation and external test sets: Mean age was 73 and 71 respectively, and recurrence, RFS and OS rates at 2 years were 43% vs 34%, 54% vs 47% and 54% vs 47% respectively. The respective validation and test set AUCs were as follows: 1) RFS: 0·682 (0·575-0·788) and 0·681 (0·597-0·766), 2) Recurrence: 0·687 (0·582-0·793) and 0·722 (0·635-0·81), and 3) OS: 0·759 (0·663-0·855) and 0·717 (0·634-0·8). Our models were superior to TNM stage and performance status in predicting recurrence and OS.InterpretationThis robust and ready to use machine learning method, validated and externally tested, sets the stage for future clinical trials entailing quantitative personalised risk-stratification and surveillance following curative-intent radiotherapy for NSCLC.FundingA full list of funding bodies that contributed to this study can be found in the Acknowledgements section.

Project description:ObjectivesTo inform personalised home-based rehabilitation interventions, we sought to gain in-depth understanding of lung cancer survivors' (1) attitudes and perceived self-efficacy towards telemedicine; (2) knowledge of the benefits of rehabilitation and exercise training; (3) perceived facilitators and preferences for telerehabilitation; and (4) health goals following curative intent therapy.DesignWe conducted semi-structured interviews guided by Bandura's Social Cognitive Theory and used directed content analysis to identify salient themes.SettingOne USA Veterans Affairs Medical Center.ParticipantsWe enrolled 20 stage I-IIIA lung cancer survivors who completed curative intent therapy in the prior 1-6 months. Eighty-five percent of participants had prior experience with telemedicine, but none with telerehabilitation or rehabilitation for lung cancer.ResultsParticipants viewed telemedicine as convenient, however impersonal and technologically challenging, with most reporting low self-efficacy in their ability to use technology. Most reported little to no knowledge of the potential benefits of specific exercise training regimens, including those directed towards reducing dyspnoea, fatigue or falls. If they were to design their own telerehabilitation programme, participants had a predominant preference for live and one-on-one interaction with a therapist, to enhance therapeutic relationship and ensure correct learning of the training techniques. Most participants had trouble stating their explicit health goals, with many having questions or concerns about their lung cancer status. Some wanted better control of symptoms and functional challenges or engage in healthful behaviours.ConclusionsFeatures of telerehabilitation interventions for lung cancer survivors following curative intent therapy may need to include strategies to improve self-efficacy and skills with telemedicine. Education to improve knowledge of the benefits of rehabilitation and exercise training, with alignment to patient-formulated goals, may increase uptake. Exercise training with live and one-on-one therapist interaction may enhance learning, adherence, and completion. Future work should determine how to incorporate these features into telerehabilitation.

Project description:PurposeThe optimum dose and fractionation in radiation therapy of curative intent for non-small cell lung cancer remains uncertain. We undertook a published data meta-analysis of randomized trials to examine whether radiation therapy regimens with higher time-corrected biologically equivalent doses resulted in longer survival, either when given alone or when given with chemotherapy.Methods and materialsEligible studies were randomized comparisons of 2 or more radiation therapy regimens, with other treatments identical. Median survival ratios were calculated for each comparison and pooled.Results3795 patients in 25 randomized comparisons of radiation therapy dose were studied. The median survival ratio, higher versus lower corrected dose, was 1.13 (95% confidence interval [CI] 1.04-1.22) when radiation therapy was given alone and 0.83 (95% CI 0.71-0.97) when it was given with concurrent chemotherapy (P for difference=.001). In comparisons of radiation therapy given alone, the survival benefit increased with increasing dose difference between randomized treatment arms (P for trend=.004). The benefit increased with increasing dose in the lower-dose arm (P for trend=.01) without reaching a level beyond which no further survival benefit was achieved. The survival benefit did not differ significantly between randomized comparisons where the higher-dose arm was hyperfractionated and those where it was not. There was heterogeneity in the median survival ratio by geographic region (P<.001), average age at randomization (P<.001), and year trial started (P for trend=.004), but not for proportion of patients with squamous cell carcinoma (P=.2).ConclusionsIn trials with concurrent chemotherapy, higher radiation therapy doses resulted in poorer survival, possibly caused, at least in part, by high levels of toxicity. Where radiation therapy was given without chemotherapy, progressively higher radiation therapy doses resulted in progressively longer survival, and no upper dose level was found above which there was no further benefit. These findings support the consideration of further radiation therapy dose escalation trials, making use of modern treatment methods to reduce toxicity.

Dataset Information

A prediction model for early death in non-small cell lung cancer patients following curative-intent chemoradiotherapy.

Publications

A prediction model for early death in non-small cell lung cancer patients following curative-intent chemoradiotherapy.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets