Dataset Information

Determining the impact of an artificial intelligence tool on the management of pulmonary nodules detected incidentally on CT (DOLCE) study protocol: a prospective, non-interventional multicentre UK study.

ABSTRACT:

Introduction

In a small percentage of patients, pulmonary nodules found on CT scans are early lung cancers. Lung cancer detected at an early stage has a much better prognosis. The British Thoracic Society guideline on managing pulmonary nodules recommends using multivariable malignancy risk prediction models to assist in management. While these guidelines seem to be effective in clinical practice, recent data suggest that artificial intelligence (AI)-based malignant-nodule prediction solutions might outperform existing models.

Methods and analysis

This study is a prospective, observational multicentre study to assess the clinical utility of an AI-assisted CT-based lung cancer prediction tool (LCP) for managing incidental solid and part solid pulmonary nodule patients vs standard care. Two thousand patients will be recruited from 12 different UK hospitals. The primary outcome is the difference between standard care and LCP-guided care in terms of the rate of benign nodules and patients with cancer discharged straight after the assessment of the baseline CT scan. Secondary outcomes investigate adherence to clinical guidelines, other measures of changes to clinical management, patient outcomes and cost-effectiveness.

Ethics and dissemination

This study has been reviewed and given a favourable opinion by the South Central-Oxford C Research Ethics Committee in UK (REC reference number: 22/SC/0142).Study results will be available publicly following peer-reviewed publication in open-access journals. A patient and public involvement group workshop is planned before the study results are available to discuss best methods to disseminate the results. Study results will also be fed back to participating organisations to inform training and procurement activities.

Trial registration number

NCT05389774.

SUBMITTER: O'Dowd E

PROVIDER: S-EPMC10773382 | biostudies-literature | 2024 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Determining the impact of an artificial intelligence tool on the management of pulmonary nodules detected incidentally on CT (DOLCE) study protocol: a prospective, non-interventional multicentre UK study.

O'Dowd Emma E Berovic Marko M Callister Matthew M Chalitsios Christos V CV Chopra Disha D Das Indrajeet I Draper Adrian A Garner Justin L JL Gleeson Fergus F Janes Sam S Kennedy Martyn M Lee Richard R Mauri Fabrizio F McKeever Tricia M TM McNulty William W Murray James J Nair Arjun A Park John J Rawlinson Janette J Sagoo Gurdeep Singh GS Scarsbrook Andrew A Shah Pallav P Sudhir Rajini R Talwar Ambika A Thakrar Ricky R Watkins Johnathan J Baldwin David R DR

BMJ open 20240104 1

<h4>Introduction</h4>In a small percentage of patients, pulmonary nodules found on CT scans are early lung cancers. Lung cancer detected at an early stage has a much better prognosis. The British Thoracic Society guideline on managing pulmonary nodules recommends using multivariable malignancy risk prediction models to assist in management. While these guidelines seem to be effective in clinical practice, recent data suggest that artificial intelligence (AI)-based malignant-nodule prediction sol ...[more]

PMID: 38176863

Similar Datasets

Project description:ImportanceWhether guideline-concordant lung nodule evaluations lead to better outcomes remains unknown.ObjectiveTo examine the association between the intensity of lung nodule diagnostic evaluations and outcomes, safety, and health expenditures.Design, setting, and participantsThis comparative effectiveness research study analyzed health plan enrollees at Kaiser Permanente Washington in Seattle, Washington, and Marshfield Clinic in Marshfield, Wisconsin, with an incidental lung nodule detected between January 1, 2005, and December 31, 2015. Included patients were 35 years or older, had no high suspicion of infection, had no history of malignant neoplasm, and had no evidence of advanced lung cancer on nodule detection. Data analysis was conducted from January 7 to August 19, 2020.ExposuresWith the 2005 Fleischner Society guidelines (selected for their applicability to the time frame under investigation) as the comparator, 2 other intensities of lung nodule evaluation were defined. Guideline-concordant evaluation followed the guidelines. Less intensive evaluation was the absence of recommended testing, longer-than-recommended surveillance intervals, or less invasive testing than recommended. More intensive evaluation consisted of testing when the guidelines recommended no further testing, shorter-than-recommended surveillance intervals, or more invasive testing than recommended.Main outcomes and measuresThe main outcome was the proportion of patients with lung cancer who had stage III or IV disease, radiation exposure, procedure-related adverse events, and health expenditures 2 years after nodule detection.ResultsAmong the 5057 individuals included in this comparative effectiveness research study, 1925 (38%) received guideline-concordant, 1863 (37%) less intensive, and 1269 (25%) more intensive diagnostic evaluations. The entire cohort comprised 2786 female patients (55%), and the mean (SD) age was 67 (13) years. Adjusted analyses showed that compared with guideline-concordant evaluations, less intensive evaluations were associated with fewer procedure-related adverse events (risk difference [RD], -5.9%; 95% CI, -7.2% to -4.6%), lower mean radiation exposure (-9.5 milliSieverts [mSv]; 95% CI, -10.3 mSv to -8.7 mSv), and lower mean health expenditures (-$10 916; 95% CI, -$16 112 to -$5719); no difference in stage III or IV disease was found among patients diagnosed with lung cancer (RD, 4.6%; 95% CI, -22% to +31%). More intensive evaluations were associated with more procedure-related adverse events (RD, +8.1%; 95% CI, +5.6% to +11%), higher mean radiation exposure (+6.8 mSv; 95% CI, +5.8 mSv to +7.8 mSv), and higher mean health expenditures ($20 132; 95% CI, +$14 398 to +$25 868); no difference in stage III or IV disease was observed (RD, -0.5%; 95% CI, -28% to +27%).Conclusions and relevanceThis study found inconclusive evidence of an association between less intensive diagnostic evaluations and more advanced stage at lung cancer diagnosis compared with guideline-concordant care; higher intensities of diagnostic evaluations were associated with greater procedural complications, radiation exposure, and expenditures. These findings underscore the need for more evidence on better ways to evaluate lung nodules and to avoid unnecessarily intensive diagnostic evaluations of lung nodules.

Project description:BackgroundPulmonary nodules are a common incidental finding on CT imaging. Few studies have described patient and nodule characteristics associated with a lung cancer diagnosis using a population-based cohort.Research questionDoes a relationship exist between patient and nodule characteristics and lung cancer among individuals with incidentally detected pulmonary nodules, and can this information be used to create exploratory lung cancer prediction models with reasonable performance characteristics?Study design and methodsWe conducted a retrospective cohort study of adults older than 18 years with lung nodules of any size incidentally detected by chest CT imaging between 2005 and 2015. All patients had at least 2 years of complete follow-up. To evaluate the relationship between patient and nodule characteristics and lung cancer, we used binomial regression. We used logistic regression to create prediction models, and we internally validated model performance using bootstrap optimism correction.ResultsAmong 7,240 patients with a median age of 67 years, 56% of whom were women, with a median BMI of 28 kg/m2, 56% of whom were ever smokers, 31% of whom had prior nonlung malignancy, with a median nodule size 5.6 mm, 57% of whom had multiple nodules, and 40% of whom had an upper lobe nodule, 265 patients (3.7%; 95% CI, 3.2%-4.1%) had a diagnosis of lung cancer. In a multivariate analysis, age, sex, BMI, smoking history, and nodule size and location were associated with a lung cancer diagnosis, whereas prior malignancy and nodule number and laterality were not. We were able to construct two prediction models with an area under the curve value of 0.75 (95% CI, 0.72-0.80) and reasonable calibration.InterpretationLung cancer is uncommon among individuals with incidentally detected lung nodules. Some, but not all, previously identified factors associated with lung cancer also were associated with this outcome in this sample. These findings may have implications for clinical practice, future practice guidelines, and the development of novel lung cancer prediction models for individuals with incidentally detected lung nodules.

Project description:BackgroundMajor issues in the implementation of screening for lung cancer by means of low-dose computed tomography (CT) are the definition of a positive result and the management of lung nodules detected on the scans. We conducted a population-based prospective study to determine factors predicting the probability that lung nodules detected on the first screening low-dose CT scans are malignant or will be found to be malignant on follow-up.MethodsWe analyzed data from two cohorts of participants undergoing low-dose CT screening. The development data set included participants in the Pan-Canadian Early Detection of Lung Cancer Study (PanCan). The validation data set included participants involved in chemoprevention trials at the British Columbia Cancer Agency (BCCA), sponsored by the U.S. National Cancer Institute. The final outcomes of all nodules of any size that were detected on baseline low-dose CT scans were tracked. Parsimonious and fuller multivariable logistic-regression models were prepared to estimate the probability of lung cancer.ResultsIn the PanCan data set, 1871 persons had 7008 nodules, of which 102 were malignant, and in the BCCA data set, 1090 persons had 5021 nodules, of which 42 were malignant. Among persons with nodules, the rates of cancer in the two data sets were 5.5% and 3.7%, respectively. Predictors of cancer in the model included older age, female sex, family history of lung cancer, emphysema, larger nodule size, location of the nodule in the upper lobe, part-solid nodule type, lower nodule count, and spiculation. Our final parsimonious and full models showed excellent discrimination and calibration, with areas under the receiver-operating-characteristic curve of more than 0.90, even for nodules that were 10 mm or smaller in the validation set.ConclusionsPredictive tools based on patient and nodule characteristics can be used to accurately estimate the probability that lung nodules detected on baseline screening low-dose CT scans are malignant. (Funded by the Terry Fox Research Institute and others; ClinicalTrials.gov number, NCT00751660.).

Project description:ObjectivesTo validate the performances of two prediction models (Brock and Lee models) for the differentiation of minimally invasive adenocarcinoma (MIA) and invasive pulmonary adenocarcinoma (IPA) from preinvasive lesions among subsolid nodules (SSNs).DesignA retrospective cohort study.SettingA tertiary university hospital in South Korea.Participants410 patients with 410 incidentally detected SSNs who underwent surgical resection for the pulmonary adenocarcinoma spectrum between 2011 and 2015.Primary and secondary outcome measuresUsing clinical and radiological variables, the predicted probability of MIA/IPA was calculated from pre-existing logistic models (Brock and Lee models). Areas under the receiver operating characteristic curve (AUCs) were calculated and compared between models. Performance metrics including sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV) were also obtained.ResultsFor pure ground-glass nodules (n=101), the AUC of the Brock model in differentiating MIA/IPA (59/101) from preinvasive lesions (42/101) was 0.671. Sensitivity, specificity, accuracy, PPV and NPV based on the optimal cut-off value were 64.4%, 64.3%, 64.4%, 71.7% and 56.3%, respectively. Sensitivity, specificity, accuracy, PPV and NPV according to the Lee criteria were 76.3%, 42.9%, 62.4%, 65.2% and 56.3%, respectively. AUC was not obtained for the Lee model as a single cut-off of nodule size (≥10 mm) was suggested by this model for the assessment of pure ground-glass nodules. For part-solid nodules (n=309; 26 preinvasive lesions and 283 MIA/IPAs), the AUC was 0.746 for the Brock model and 0.771 for the Lee model (p=0.574). Sensitivity, specificity, accuracy, PPV and NPV were 82.3%, 53.8%, 79.9%, 95.1% and 21.9%, respectively, for the Brock model and 77.0%, 69.2%, 76.4%, 96.5% and 21.7%, respectively, for the Lee model.ConclusionsThe performance of prediction models for the incidentally detected SSNs in differentiating MIA/IPA from preinvasive lesions might be suboptimal. Thus, an alternative risk calculation model is required for the incidentally detected SSNs.

Project description:Background Incidental detection of pancreatic cystic neoplasm (PCN) has increased. Since a small percentage of PCNs possess malignant potential, management is challenging. The recently revised American College of Radiology (ACR) recommendations define PCN measurement and growth for different categories based on baseline cyst size. However, no data are available regarding PCN growth rate under the ACR-defined size categories. Purpose To assess growth of incidentally detected PCNs on long-term imaging follow-up using revised ACR recommendations and to evaluate the association between baseline imaging features and growth. Materials and Methods This retrospective study included PCNs with baseline imaging performed between January 2002 and May 2017, with two or more cross-sectional imaging studies performed at least 12 months apart. PCN assessment was based on ACR 2017 recommendations. Cyst features, including location, septations, and mural nodules and multiplicity, were noted. Time to cyst progression (growth by ACR criteria) was examined by using baseline PCN size, among other factors. Results A total of 646 cysts in 390 patients were followed up for a median of 50 months (range, 12-186 months). A total of 184 (28.5%) cysts increased in size, 52 (8.1%) decreased in size, and 410 (63.4%) remained stable. For groups in which baseline PCN size was smaller than 5 mm, 5-14 mm, 15-25 mm, and larger than 25 mm, growth was noted in seven (13.2%), 106 (28.9%), 49 (32.2%), and 22 (29.7%) cysts, respectively. ACR baseline size categories (subhazard ratio: 2.8 [5-14-mm PCN group], 3.4 [15-25-mm PCN group], and 2.7 [>25 mm group], as compared with the <5 mm PCN group; P < .05 for each) demonstrated association with growth. Presence of mural nodules, septations, or lesion multiplicity failed to demonstrate association with growth. Among PCNs smaller than 5 mm at baseline, 100% of PCNs at 3-year follow-up and 94.2% of PCNs at 5-year follow-up were likely to remain stable. Conclusion American College of Radiology baseline size category of 15-25-mm pancreatic cystic neoplasms (PCNs) demonstrated the highest (3.1 times) likelihood of growth, as compared with the category of PCNs smaller than 5 mm. PCNs smaller than 5 mm at baseline did not demonstrate growth at 3-year imaging follow-up. © RSNA, 2019 Online supplemental material is available for this article.

Dataset Information

Determining the impact of an artificial intelligence tool on the management of pulmonary nodules detected incidentally on CT (DOLCE) study protocol: a prospective, non-interventional multicentre UK study.

Introduction

Methods and analysis

Ethics and dissemination

Trial registration number

Publications

Determining the impact of an artificial intelligence tool on the management of pulmonary nodules detected incidentally on CT (DOLCE) study protocol: a prospective, non-interventional multicentre UK study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets