Cancer Risk Prediction Research Articles

Validity of nomograms for predicting cancer risk in oral leukoplakia and oral lichen planus.

This study assessed the validity of nomograms for predicting malignant transformation (MT) among patients with oral leukoplakia (OL) and oral lichen planus (OLP). Two nomograms were identified following a systematic search. Variables to interrogate both nomograms were obtained for a retrospective OL/OLP cohort. Then, the nomograms were applied to estimate MT probabilities twice and their average was used to calculate the discriminatory performance, calibration, and potential net benefit of the models. Subgroup analyses were performed for patients with OL, OLP, and oral epithelial dysplasia. Predicted probabilities were mostly significantly higher among OL/OLP patients who developed MT compared to those who did not (p = <0.001-0.034). AUC values and Brier scores of the nomograms were 0.644-0.844 and 0.040-0.088 among OL patients and 0.580-0.743 and 0.008-0.072 among OLP patients. Decision curve analysis suggested that the nomograms had some net benefit for risk stratification. However, the models did not best binary dysplasia grading in discriminatory validity and net benefit among patients with OL and oral epithelial dysplasia. Nomograms for predicting MT may have satisfactory validity among patients with OL than OLP, but they do not outperform binary dysplasia grading in risk stratification of OL.

A pancreatic cancer risk prediction model (Prism) developed and validated on large-scale US clinical data

Pancreatic Duct Adenocarcinoma (PDAC) screening can enable early-stage disease detection and long-term survival. Current guidelines use inherited predisposition, with about 10% of PDAC cases eligible for screening. Using Electronic Health Record (EHR) data from a multi-institutional federated network, we developed and validated a PDAC RISk Model (Prism) for the general US population to extend early PDAC detection. Neural Network (PrismNN) and Logistic Regression (PrismLR) were developed using EHR data from 55 US Health Care Organisations (HCOs) to predict PDAC risk 6-18 months before diagnosis for patients 40 years or older. Model performance was assessed using Area Under the Curve (AUC) and calibration plots. Models were internal-externally validated by geographic location, race, and time. Simulated model deployment evaluated Standardised Incidence Ratio (SIR) and other metrics. With 35,387 PDAC cases, 1,500,081 controls, and 87 features per patient, PrismNN obtained a test AUC of 0.826 (95% CI: 0.824-0.828) (PrismLR: 0.800 (95% CI: 0.798-0.802)). PrismNN's average internal-external validation AUCs were 0.740 for locations, 0.828 for races, and 0.789 (95% CI: 0.762-0.816) for time. At SIR=5.10 (exceeding the current screening inclusion threshold) in simulated model deployment, PrismNN sensitivity was 35.9% (specificity 95.3%). Prism models demonstrated good accuracy and generalizability across diverse populations. PrismNN could find 3.5 times more cases at comparable risk than current screening guidelines. The small number of features provided a basis for model interpretation. Integration with the federated network provided data from a large, heterogeneous patient population and a pathway to future clinical deployment. Prevent Cancer Foundation, TriNetX, Boeing, DARPA, NSF, and Aarno Labs.

Artificial Intelligence for Image-Based Breast Cancer Risk Prediction Using Attention.

Accurate prediction of individual breast cancer risk paves the way for personalised prevention and early detection. The incorporation of genetic information and breast density has been shown to improve predictions for existing models, but detailed image-based features are yet to be included despite correlating with risk. Complex information can be extracted from mammograms using deep-learning algorithms, however, this is a challenging area of research, partly due to the lack of data within the field, and partly due to the computational burden. We propose an attention-based Multiple Instance Learning (MIL) model that can make accurate, short-term risk predictions from mammograms taken prior to the detection of cancer at full resolution. Current screen-detected cancers are mixed in with priors during model development to promote the detection of features associated with risk specifically and features associated with cancer formation, in addition to alleviating data scarcity issues. MAI-risk achieves an AUC of 0.747 [0.711, 0.783] in cancer-free screening mammograms of women who went on to develop a screen-detected or interval cancer between 5 and 55 months, outperforming both IBIS (AUC 0.594 [0.557, 0.633]) and VAS (AUC 0.649 [0.614, 0.683]) alone when accounting for established clinical risk factors.

Cross-population evaluation of cervical cancer risk prediction algorithms

BackgroundCervical cancer is a preventable disease, despite being one of the most common types of female cancers worldwide. Integrating existing programs for cervical cancer screening with personalized risk prediction algorithms can improve population-level cancer prevention by enabling more targeted screening and contrive preventive healthcare innovations. While algorithms developed for cervical cancer risk prediction have shown promising performance in internal validation on more homogeneous populations, their ability to generalize to external populations remains to be assessed. MethodsTo address this gap, we perform a cross-population comparative study of personalized prediction algorithms for more personalized cervical cancer screening. Using data from the Norwegian and Estonian populations, the algorithms are validated on internal and external datasets to study their potential biases and limitations when applied to different populations. We evaluate the algorithms in predicting progression from low-grade precancerous cervical lesions, simulating a clinically relevant application of more personalized risk stratification. ResultsAs expected, our numerical experiments show that algorithm performance varies depending on the population. However, some algorithms show strong generalization capacity across different data sources. Using Kaplan-Meier estimates, we demonstrate the strengths and limitations of the algorithms in detecting cancer progression over time by comparing to the trends observed from data. We assess their overall discrimination performance in personalized risk predictions by analyzing the accuracy and confidence in individual risk estimates. Discussion and ConclusionThis study examines the effectiveness of personalized prediction algorithms across different populations. Our results demonstrate the potential for generalizing risk prediction algorithms to external populations. These findings highlight the importance of considering population diversity when developing risk prediction algorithms.

Risk Prediction of Metachronous Colorectal Cancer from Molecular Features of Adenomas: A Nested Case-Control Study.

Identifying new biomarkers may improve prediction of me-CRC for individuals with adenomas and optimize surveillance intervals to reduce risk of colorectal cancer and reduce oversurveillance of patients with low risk of colorectal cancer. Use of DNA CNAs alone does not improve prediction of me-CRC. Further research to improve risk classification is required.

A Systematic Review and Critical Assessment of Breast Cancer Risk Prediction Tools Incorporating a Polygenic Risk Score for the General Population.

Single nucleotide polymorphisms (SNPs) in the form of a polygenic risk score (PRS) have emerged as a promising factor that could improve the predictive performance of breast cancer (BC) risk prediction tools. This study aims to appraise and critically assess the current evidence on these tools. Studies were identified using Medline, EMBASE and the Cochrane Library up to November 2022 and were included if they described the development and/ or validation of a BC risk prediction model using a PRS for women of the general population and if they reported a measure of predictive performance. We identified 37 articles, of which 29 combined genetic and non-genetic risk factors using seven different risk prediction tools. Most models (55.0%) were developed on populations from European ancestry and performed better than those developed on populations from other ancestry groups. Regardless of the number of SNPs in each PRS, models combining a PRS with genetic and non-genetic risk factors generally had better discriminatory accuracy (AUC from 0.52 to 0.77) than those using a PRS alone (AUC from 0.48 to 0.68). The overall risk of bias was considered low in most studies. BC risk prediction tools combining a PRS with genetic and non-genetic risk factors provided better discriminative accuracy than either used alone. Further studies are needed to cross-compare their clinical utility and readiness for implementation in public health practices.

Artificial Intelligence-Driven Mammography-Based Future Breast Cancer Risk Prediction: A Systematic Review

PurposeTo summarize the literature regarding the performance of mammography-image based artificial intelligence (AI) algorithms, with and without additional clinical data, for future breast cancer risk prediction. Materials and methodsA systematic literature review was performed using six databases (medRixiv, bioRxiv, Embase, Engineer Village, IEEE Xplore, and PubMed) from 2012 through September 30, 2022. Studies were included if they used real-world screening mammography examinations to validate AI algorithms for future risk prediction based on images alone or in combination with clinical risk factors. The quality of studies was assessed, and predictive accuracy was recorded as the area under the receiver operating characteristic curve (AUC). ResultsSixteen studies met inclusion and exclusion criteria, of which 14 studies provided AUC values. The median AUC performance of AI image-only models was 0.72 (range 0.62-0.90) compared with 0.61 for breast density or clinical risk factor–based tools (range 0.54-0.69). Of the seven studies that compared AI image-only performance directly to combined image + clinical risk factor performance, six demonstrated no significant improvement, and one study demonstrated increased improvement. ConclusionsEarly efforts for predicting future breast cancer risk based on mammography images alone demonstrate comparable or better accuracy to traditional risk tools with little or no improvement when adding clinical risk factor data. Transitioning from clinical risk factor–based to AI image-based risk models may lead to more accurate, personalized risk-based screening approaches.

The local environment and germline genetic variation predict cancer risk in the UK Biobank prospective cohort

BackgroundThere is a growing body of evidence on the effect of the local environment exposure on cancer susceptibility. Nonetheless, several of the associations remain controversial. Moreover, our understanding of the possible interaction between the local environment and the genetic variability is still very limited. ObjectiveThe aim of this study was to clarify the role of the local environment and its possible interplay with genetics on common cancers development. MethodsUsing the UK Biobank (UKBB) prospective cohort, we selected 12 local environment exposures: nitrogen oxides, nitrogen dioxides, particulate matter (10 and 2.5 μm), noise pollution, urban traffic, living distance from the coast, percentage of greenspace, natural environment, water, and domestic garden within 1000 m from the residential coordinates of each participant. All these exposures were tested for association with 17 different types of cancer for a total of 53,270 cases and 302,645 controls. Additionally, a polygenic score (PGS) was computed for each cancer, to test possible gene-environment interactions. Finally, mediation analyses were carried out. ResultsThirty-six statistically significant associations considering multiple testing (p < 2.19 × 10−4) were observed. Among the novel associations we observed that individuals living farther from the coast had a higher risk of developing prostate cancer (OR = 1.13, CI95% = 1.06–1.20, P = 1.98 × 10−4). This association was partially mediated by physical activity (indirect effect (IE) = -8.48 × 10−7) and the time spent outdoor (IE = 9.07 × 10−6). All PGSs showed statistically significant associations. Finally, genome-environment interaction analysis showed that local environment and genetic variability affect cancer risk independently. DiscussionLiving close to the coast and air pollution were associated with a decreased risk of prostate cancer and skin melanoma, respectively. These findings from the UKBB support the role of the local environment on cancer development, which is independent from genetics and may be mediated by several lifestyle factors.

Lung Cancer Risk Prediction Models for Asian Ever-Smokers

IntroductionAlthough lung cancer prediction models are widely used to support risk-based screening, their performance outside Western populations remains uncertain. This study aims to evaluate the performance of 11 existing risk prediction models in multiple Asian populations and to refit prediction models for Asians. MethodsIn a pooled analysis of 186,458 Asian ever-smokers from 19 prospective cohorts, we assessed calibration (expected-to-observed ratio) and discrimination (area under the receiver operating characteristic curve [AUC]) for each model. In addition, we developed the “Shanghai models” to better refine risk models for Asians on the basis of two well-characterized population-based prospective cohorts and externally validated them in other Asian cohorts. ResultsAmong the 11 models, the Lung Cancer Death Risk Assessment Tool yielded the highest AUC (AUC [95% confidence interval (CI)] = 0.71 [0.67–0.74] for lung cancer death and 0.69 [0.67–0.72] for lung cancer incidence) and the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Model had good calibration overall (expected-to-observed ratio [95% CI] = 1.06 [0.90–1.25]). Nevertheless, these models substantially underestimated lung cancer risk among Asians who reported less than 10 smoking pack-years or stopped smoking more than or equal to 20 years ago. The Shanghai models were found to have marginal improvement overall in discrimination (AUC [95% CI] = 0.72 [0.69–0.74] for lung cancer death and 0.70 [0.67–0.72] for lung cancer incidence) but consistently outperformed the selected Western models among low-intensity smokers and long-term quitters. ConclusionsThe Shanghai models had comparable performance overall to the best existing models, but they improved much in predicting the lung cancer risk of low-intensity smokers and long-term quitters in Asia.

Application of radiomics in diagnosis and treatment of lung cancer.

Radiomics has become a research field that involves the process of converting standard nursing images into quantitative image data, which can be combined with other data sources and subsequently analyzed using traditional biostatistics or artificial intelligence (Al) methods. Due to the capture of biological and pathophysiological information by radiomics features, these quantitative radiomics features have been proven to provide fast and accurate non-invasive biomarkers for lung cancer risk prediction, diagnosis, prognosis, treatment response monitoring, and tumor biology. In this review, radiomics has been emphasized and discussed in lung cancer research, including advantages, challenges, and drawbacks.

43P Performance evaluation of a combined risk model for breast cancer risk prediction in Indonesian population (TRIP Study)

43P Performance evaluation of a combined risk model for breast cancer risk prediction in Indonesian population (TRIP Study)

LncRNA polymorphisms and lung cancer risk.

Lung cancer (LC) imposes a significant burden, and is associated with high mortality and morbidity among malignant tumors. Aberrant expression of particular lncRNAs is closely linked to LC. LncRNA polymorphisms cause abnormal expression levels and/or structural dysfunction. They can affect the progression of cancer, survival, response to chemotherapyand recurrence rates in cancer patients. The present article provides a comprehensive overview of the effect of lncRNAgenetic polymorphisms on LC. It is proposed that lncRNA-related variants can be used to predict cancer risk and therapeutic outcomes. More large-scale trials on diverse ethnic groups are required to validate the results, thus personalizing LC therapy based on lncRNA genotypes.

MA19.03 Sex Differences in the Prediction of Future Lung Cancer Risk Based on a Single Low-Dose Chest Computed Tomography Scan

MA19.03 Sex Differences in the Prediction of Future Lung Cancer Risk Based on a Single Low-Dose Chest Computed Tomography Scan

Breast density analysis of digital breast tomosynthesis

Mammography shifted to digital breast tomosynthesis (DBT) in the US. An automated percentage of breast density (PD) technique designed for two-dimensional (2D) applications was evaluated with DBT using several breast cancer risk prediction measures: normalized-volumetric; dense volume; applied to the volume slices and averaged (slice-mean); and applied to synthetic 2D images. Volumetric measures were derived theoretically. PD was modeled as a function of compressed breast thickness (CBT). The mean and standard deviation of the pixel values were investigated. A matched case–control (CC) study (n = 426 pairs) was evaluated. Odd ratios (ORs) were estimated with 95% confidence intervals. ORs were significant for PD: identical for volumetric and slice-mean measures [OR = 1.43 (1.18, 1.72)] and [OR = 1.44 (1.18, 1.75)] for synthetic images. A 2nd degree polynomial (concave-down) was used to model PD as a function of CBT: location of the maximum PD value was similar across CCs, occurring at 0.41 × CBT, and PD was significant [OR = 1.47 (1.21, 1.78)]. The means from the volume and synthetic images were also significant [ORs ~ 1.31 (1.09, 1.57)]. An alternative standardized 2D synthetic image was constructed, where each pixel value represents the percentage of breast density above its location. Several measures were significant and an alternative method for constructing a standardized 2D synthetic image was produced.

P1.14-02 Validation of Lung Cancer Risk Prediction Models in an Asbestos Exposed Population

P1.14-02 Validation of Lung Cancer Risk Prediction Models in an Asbestos Exposed Population

OA07.06 Development and Validation of a Protein-Based Lung Cancer Risk Prediction Model - Initial Results from the Lung Cancer Cohort Consortium (LC3)

OA07.06 Development and Validation of a Protein-Based Lung Cancer Risk Prediction Model - Initial Results from the Lung Cancer Cohort Consortium (LC3)

Ethical and legal implications of implementing risk algorithms for early detection and screening for oesophageal cancer, now and in the future.

Oesophageal cancer has significant morbidity and mortality but late diagnosis is common since early signs of disease are frequently misinterpreted. Project DELTA aims to enable earlier detection and treatment through targeted screening using a novel risk prediction algorithm for oesophageal cancer (incorporating risk factors of Barrett's oesophagus including prescriptions for acid-reducing medications (CanPredict)), together with a non-invasive, low-cost sampling device (CytospongeTM). However, there are many barriers to implementation, and this paper identifies key ethical and legal challenges to implementing these personalised prevention strategies for Barrett's oesophagus/oesophageal cancer. To identify ethical and legal issues relevant to the deployment of a risk prediction tool for oesophageal cancer into primary care, we adopted an interdisciplinary approach, incorporating targeted informal literature reviews, interviews with expert collaborators, a multidisciplinary workshop and ethical and legal analysis. Successful implementation raises many issues including ensuring transparency and effective risk communication; addressing bias and inequity; managing resources appropriately and avoiding exceptionalism. Clinicians will need support and training to use cancer risk prediction algorithms, ensuring that they understand how risk algorithms supplement rather than replace medical decision-making. Workshop participants had concerns about liability for harms arising from risk algorithms, including from potential bias and inequitable implementation. Determining strategies for risk communication enabling transparency but avoiding exceptionalist approaches are a significant challenge. Future challenges include using artificial intelligence to bolster risk assessment, incorporating genomics into risk tools, and deployment by non-health professional users. However, these strategies could improve detection and outcomes. Novel pathways incorporating risk prediction algorithms hold considerable promise, especially when combined with low-cost sampling. However immediate priorities should be to develop risk communication strategies that take account of using validated risk algorithms, and to ensure equitable implementation. Resolving questions about liability for harms arising should be a longer-term objective.

Polygenic Risk Scores for cancer mortality prediction

Abstract Background Polygenic risk scores (PRSs) have shown great accuracy in predicting cancer risk. In screening-based cancer prevention, it is important to be able to predict lethal cancer risk and PRSs have the potential to improve such prediction. This preliminary analysis, conducted on the UK BioBank, investigates the ability of 5 cancer PRSs to predict cancer incidence and mortality. Methods Five cohorts were derived from the UK BioBank for 5 common types of cancer (breast, colorectal, lung, pancreas and prostate). PRSs for cancer were constructed for each participant. The PRS hazard ratio per standard deviation (HR×SD) for cancer incidence and mortality was estimated using two proportional hazards Cox regression models for each type of cancer controlling for genotyping array, the first ten principal components of ancestry, family history, sex (where applicable), and age at enrolment. Results In the breast cancer cohort, PRS showed a HR×SD of 1.74 (95%CI:1.70-1.78) for cancer incidence and 1.66 (95%CI:1.52-1.81) for cancer mortality. In the prostate cancer cohort the PRS had a HRxSD of 1.78 (95%CI:1.74-1.81) for incidence and 1.73 (95%CI:1.61-1.86) for mortality. Similar results were found in the colorectal cancer cohort with a 1.27 (95%CI:1.24-1.31) PRS HR×SD for incidence and 1.30 (95%CI:1.24-1.37) for mortality, as well as in the pancreas cancer cohort with a 1.44 (95%CI:1.38-1.51) PRS HR×SD for incidence and 1.48 (95%CI:1.40-1.57) for mortality. In the lung cancer cohort a marginal PRS HR×SD was found for incidence and for mortality [1.07 (95%CI:1.04-1.10) and 1.07 (95%CI:1.03-1.12), respectively]. Conclusions These results suggest that, for most types of cancer, PRSs can predict cancer mortality, as well as incidence, indicating their potential use as a tool to identify people at risk of lethal disease. Key messages • These results may contribute to outline clinical utility of PRSs in cancer prevention. • PRSs may be a useful addition to screening-based cancer prevention, in order to prioritise patients at risk of lethal cancer.

Machine learning endometrial cancer risk prediction model: integrating guidelines of European Society for Medical Oncology with the tumor immune framework

ObjectiveCurrent prognostic factors for endometrial cancer are not sufficient to predict recurrence in early stages. Treatment choices are based on the prognostic factors included in the risk classes defined by...

Statistical analysis and risk prediction of lung cancer exploiting logistic regression

Cancer is a disease with a high mortality rate. Early prediction of cancer is an important means to completely cure the disease. Therefore, there is an ever-increasing demand for technologies to detect early cancer nodules. As an easily misdiagnosed disease, the mortality rate of lung cancer has reached the highest level in recent years. Early diagnosis of lung cancer can save many lives. The study used logistic regression to predict cancer risk by collecting and analyzing patient data. Logistic regression, referred to as logistic regression analysis, is a general linear regression analysis model that relates to supervised learning in machine learning. First, n sets of data are given to the model as a training set for training, and then the model could be leveraged to classify other data (test sets), and finally the classification results could be achieved. P indicators make up each batch of data. This study used a dataset from Kaggle, which contains information from one thousand patients with lung cancer, and it has 23 eigenvalues. This dataset classifies the predicted values as high, medium, and low. The regression results show that the precision value of the low prediction value is 84%, the precision value of the middle prediction value is 88%, and the precision value of the low prediction value is 93%. The result is that as the number of samples with different predicted values increases, the precision values increase.