Quantitative models of lung cancer mortality. II. Predicting lung cancer mortalities for a population depending on the level of smoking.
Explore the source record for details and available documents.
Adolescent↗
SEARCH · Search PubMed
Results for “Cancer prediction”
Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
SurvGRN: a multi-feature fusion framework for bladder cancer survival prediction.
Bladder cancer survival outcomes exhibit significant heterogeneity, influenced by multifaceted factors. While digital pathology-based survival models leveraging artificial intelligence show promise, they often overlook complementary data sources. Conversely, imaging lacks cellular detail, and genomics/proteomics entail complexity and cost. To integrate multidimensional data for enhanced survival prediction, we propose SurvGRN, a multi-feature fusion framework. SurvGRN synergistically combines clinical variables, transcriptomics, and digital pathology slides using a gated residual network architecture. Pathological features are extracted via multiple instance learning, while clinical and transcriptomic data are processed as static inputs. These features are dynamically fused using a long short-term memory (LSTM) network for comprehensive survival risk assessment. Evaluated on 400 bladder cancer patients, SurvGRN significantly outperformed existing methods: improving the C-index by 12.6% over DeepMISL; 20.6% and 7.1% over graph-based models (DeepGraphConv and Patch-GCN); and 5.4% and 4.0% over attention-based approaches (Surformer and HVTSurv). Ablation studies confirmed the contributions of pathology features (extracted via ResNet-50 pre-trained on bladder tissue), clinical/transcriptomic data, and the LSTM fusion. SurvGRN also enabled significant stratification of patients into distinct risk cohorts. This work demonstrates that holistic integration of multi-source data through tailored fusion architectures substantially improves bladder cancer survival prediction.
bladder cancer↗
RCoxNet: A Deep Learning Framework Integrating Random Walk with Restart, Mutation, and Clinical Data for Cancer Survival Prediction.
Accurate survival prediction in cancer remains challenging due to the sparsity of somatic mutation profiles and the failure of existing models to capture higher-order gene-gene dependencies. Network diffusion methods such as Random Walk with Restart (RWR) can propagate mutation signals across protein-protein interaction (PPI) networks to address sparsity, yet their integration within a deep learning Cox survival framework has not been comprehensively benchmarked across multiple cancer cohorts. We present RCoxNet, a deep learning framework that maps somatic mutation profiles onto a ConsensusPathDB-derived PPI network via RWR, selects prognostic genes by log-rank filtering, and processes network-informed mutation scores through three fully connected hidden layers feeding into a Cox proportional hazards output. RCoxNet was evaluated on The Cancer Genome Atlas (TCGA) cohorts for four cancer types (breast invasive carcinoma [BRCA], lung adenocarcinoma [LUNG], glioblastoma multiforme [GBM], and ovarian serous cystadenocarcinoma [OV]) using 20 independent random splits. The model achieved mean C-index values of 0.807 ± 0.044 (BRCA), 0.750 ± 0.039 (LUNG), 0.704 ± 0.041 (GBM), and 0.668 ± 0.036 (OV), consistently outperforming DeepSurv, Cox-nnet, SurvivalNet, Cox Elastic-Net (Cox-EN), and DeepHit, with statistically significant gains over Cox-EN, Cox-nnet, SurvivalNet, and DeepHit across the majority of cohorts. RCoxNet demonstrates that embedding sparse mutation profiles into a PPI network context substantially improves cancer survival prediction and yields biologically interpretable prognostic features relevant to precision oncology.
cancer survival prediction↗
Ethical implications of predictive DNA testing for hereditary breast cancer.
Predictive medicine offers the possibility of detecting many common diseases that have a genetic basis, such as cancer; however, a genetic alteration might only indicate susceptibility to, not certainty of, disease. Whereas means for identifying a greater susceptibility to disease have been developed, effective interventions have progressed much more slowly. Awareness of one's susceptibility to disease without an actual possibility of intervention can lead to an unacceptable use of such information, or have a dramatic psychological impact on the person involved. Are the risks connected with the knowledge of susceptibility to genetic disease proportional to the benefits that such knowledge may provide? Does the knowledge of one's genetic condition constitute a service to the individual and society, or is this predominantly harmful for the person involved? The problem is vast, and involves medical, psychological, social, political and ethical dilemmas. These dilemmas, common to all predictive medicine, are most evident in predictive DNA testing for hereditary breast cancer. In our analysis, we will first examine the ethical values involved in genetic testing, highlighting the special ethical issues raised by predictive DNA testing for hereditary breast cancer. Next we will deal with genetic counseling, which, in our opinion, is the 'ethos' for ethically justifying predictive DNA testing.
Breast Neoplasms↗
Estrogen receptor status of primary breast cancer is predictive of estrogen receptor status of contralateral breast cancer.
BACKGROUND: Tamoxifen reduces the risk for contralateral breast cancer by approximately 30%-50%, with benefits probably limited to women with estrogen receptor (ER)-positive primary disease. In a retrospective analysis of data from National Surgical and Adjuvant Breast and Bowel Project trials B-18, B-22, and B-25, we determined whether the ER status of primary breast cancer predicts the ER status of a subsequent contralateral breast cancer and whether tamoxifen treatment affects this relationship. In these trials, tamoxifen at 20 mg/day had been administered only to women aged 50 years or older, rather than to those determined by the ER status of their primary tumor, allowing an assessment of the treatment's effects in ER-negative disease. METHODS: Among the 5513 eligible patients, 176 patients developed a contralateral breast cancer. The ER status of the primary and contralateral tumor was determined and cross-classified for women who did not receive tamoxifen (i.e., those aged 49 years or younger) and for women who did (i.e., those aged 50 years or older). ER data were available for 110 evaluable invasive contralateral breast cancers. RESULTS: Among patients who did not receive tamoxifen (n = 62), 89% with an ER-positive primary cancer had an ER-positive contralateral breast cancer and 70% with an ER-negative primary breast cancer had an ER-negative contralateral breast cancer (odds ratio for the association between primary and contralateral ER status = 14.8, 95% confidence interval = 3.8 to 74.3; P<.001). Among patients who received tamoxifen (n = 48), 56% with an ER-positive primary cancer had an ER-positive contralateral breast cancer and 78% with an ER-negative primary cancer had an ER-negative contralateral breast cancer (odds ratio = 3.4, 95% confidence interval = 0.53 to 39.2; P =.25). CONCLUSION: The ER status of the primary breast cancer was associated with that of the contralateral breast for patients not receiving tamoxifen. Patients with an ER-positive primary cancer who received tamoxifen had a lower concordance rate with fewer ER-positive contralateral breast cancers, which may be a result of tamoxifen treatment.
Age Factors↗
Prediction of clinical behaviour and treatment for cancers.
Prediction of clinical behaviour and treatment for cancers is based on the integration of clinical and pathological parameters. Recent reports have demonstrated that gene expression profiling provides a powerful new approach for determining disease outcome. If clinical and microarray data each contain independent information then it should be possible to combine these datasets to gain more accurate prognostic information. Here, we have used existing clinical information and microarray data to generate a combined prognostic model for outcome prediction for diffuse large B-cell lymphoma (DLBCL). A prediction accuracy of 87.5% was achieved. This constitutes a significant improvement compared to the previously most accurate prognostic model with an accuracy of 77.6%. The model introduced here may be generally applicable to the combination of various types of molecular and clinical data for improving medical decision support systems and individualising patient care.
Algorithms↗
Integrating genetic predictors into subsequent breast cancer risk prediction in survivors of childhood cancer.
PURPOSE: Female survivors of childhood cancer are at high risk for developing breast cancer. The contributions of most general population primary breast cancer genetic predictors to this risk have not been explored. METHODS: Analyses included females who survived ≥5 years after their childhood cancer diagnosis with available array (N = 2096, subsequent breast cancer [SBC]=218) or whole-genome sequencing (WGS; N = 3292, SBC=101) data from the Childhood Cancer Survivor Study and St. Jude Lifetime Cohort. We computed 99 externally-validated primary breast cancer polygenic risk scores (PRS). Using deep-coverage WGS, ClinVar-annotated pathogenic/likely pathogenic (P/LP) variants in breast cancer susceptibility genes were identified. Cox proportional hazards models assessed associations with SBC risk, adjusting for treatments and genetic ancestry. RESULTS: Among 5388 female survivors (genetic ancestry, European: N = 4,752; African: N = 444; East Asian: N = 192), 319 developed SBC. Most (90.9%) PRSs were nominally associated with SBC risk (P < 0.05), but effect sizes varied substantially. PRSs with superior discriminatory ability had greater genome-wide coverage (e.g., 6.4 million-variant PRS, HR per SD = 1.71, 95% CI = 1.43 to 2.05; P = 4.2x10-9) and 7.7-fold higher odds (P = 7.0x10-4) of including variants in multiple DNA damage repair pathways compared with PRSs with weaker risk associations. Among survivors with WGS, 1.6% carried P/LP variants in clinical testing panel genes, which was associated with a 7.4-fold greater risk (95% CI = 3.16 to 17.19). Including genetic factors improved SBC risk prediction by age 40 (P < 0.001) compared to treatment exposures alone. CONCLUSIONS: Externally-validated primary breast cancer genetic susceptibility predictors are relevant for SBC risk prediction and should be prioritized for risk stratification in survivors.
Journal Article↗
Lung cancer rate predictions using generalized additive models.
Predictions of lung cancer incidence and mortality are necessary for planning public health programs and clinical services. It is proposed that generalized additive models (GAMs) are practical for cancer rate prediction. Smooth equivalents for classical age-period, age-cohort, and age-period-cohort models are available using one-dimensional smoothing splines. We also propose using two-dimensional smoothing splines for age and period. Variance estimation can be based on the bootstrap. To assess predictive performance, we compared the models with a Bayesian age-period-cohort model. Model comparison used cross-validation and measures of predictive performance for recent predictions. The models were applied to data from the World Health Organization Mortality Database for females in five countries. Model choice between the age-period-cohort models and the two-dimensional models was equivocal with respect to cross-validation, while the two-dimensional GAMs had very good predictive performance. The Bayesian model performed poorly due to imprecise predictions and the assumption of linearity outside of observed data. In summary, the two-dimensional GAM performed well. The GAMs make the important prediction that female lung cancer rates in these countries will be stable or begin to decline in the future.
Adult↗
Protein expression profiling identifies subclasses of breast cancer and predicts prognosis.
Breast cancer is a heterogeneous disease whose evolution is difficult to predict by using classic histoclinical prognostic factors. Prognostic classification can benefit from molecular analyses such as large-scale expression profiling. Using immunohistochemistry on tissue microarrays, we have monitored the expression of 26 selected proteins in more than 1,600 cancer samples from 552 consecutive patients with early breast cancer. Both an unsupervised approach and a new supervised method were used to analyze these profiles. Hierarchical clustering identified relevant clusters of coexpressed proteins and clusters of tumors. We delineated protein clusters associated with the estrogen receptor and with proliferation. Tumor clusters correlated with several histoclinical features of samples, including 5-year metastasis-free survival (MFS), and with the recently proposed pathophysiologic taxonomy of disease. The supervised method identified a set of 21 proteins whose combined expression significantly correlated to MFS in a learning set of 368 patients (P < 0.0001) and in a validation set of 184 patients (P < 0.0001). Among the 552 patients, the 5-year MFS was 90% for patients classified in the "good-prognosis class" and 61% for those classified in the "poor-prognosis class" (P < 0.0001). This difference remained significant when the molecular grouping was applied according to lymph node or estrogen receptor status, as well as the type of adjuvant systemic therapy. In multivariate analysis, the 21-protein set was the strongest independent predictor of clinical outcome. These results show that protein expression profiling may be a clinically useful approach to assess breast cancer heterogeneity and prognosis in stage I, II, or III disease.
Adult↗
Cancer disease predictive diagnosis: BAT/CD3-positive lymphocytes in cancer patients.
BAT is an immune-activating monoclonal antibody produced against Daudi cell membranes and selected for stimulating lymphocyte proliferation. The anti-tumor activity of BAT is related to its immunostimulatory properties. Both T and NK cells mediate the anti-tumor activity of BAT. CD4-positive T cells respond to BAT activation by proliferation and INF-gamma production. The aim of the study was to assess the probability that the BAT monoclonal antibody binding capacity to T cells is a marker for different cancers. Human peripheral blood T cells from colon, breast and prostate cancer patients, as well as healthy volunteer donors, were tested for the percentage of binding to BAT mAb (BAT/CD3 cells) by FACS analysis. All patients were tested before undergoing surgery or treatment, and their diagnosis was confirmed by histology. The results showed that the percentage of BAT monoclonal antibody binding to CD3-positive T cells in the peripheral blood was different in cancer patients with diverse tumor types. We found that lymphocytes from the blood of healthy donors contained 25% BAT/CD3 cells. In colon and breast cancer patients, a significant decrease to 13 and 11% of BAT/CD3 cells was found. In contrast, these cells increased ><50% in patients with prostate cancer. These findings may have a potential diagnostic significance and also assist in the evaluation of strategies for the therapeutic use of BAT for different cancer patients.
Adult↗
CA 125 reliability in predicting ovarian cancer recurrence.
The predictive value of CA 125 assay for recurrence in ovarian cancer patients in follow-up was analyzed in a study from April 1984 through June 1987. Forty-two patients with no evidence of disease (NED), with positive antigen levels at diagnosis and negative at the end of active treatment, were considered eligible for the analysis. Median follow-up time was 16 months (range, 5-34). Outcome analysis revealed 19 cases still NED: 16 had normal CA 125 levels (less than 35 U/ml). The 3 patients with positive antigen titers were intensively investigated with no evidence of recurrence. Twenty-three cases had disease recurrence: 13 of them had elevated marker levels prior to relapse diagnosis, with a median lead time of 5 months (range, 2-13). In contrast, 10 patients had positive titers at or soon after the recurrence. Test sensitivity was therefore 56% and specificity 84%. Predictive value for recurrence of elevated CA 125 levels was 0.81.
Adult↗
14-3-3sigma methylation in pretreatment serum circulating DNA of cisplatin-plus-gemcitabine-treated advanced non-small-cell lung cancer patients predicts survival: The Spanish Lung Cancer Group.
PURPOSE: Survival in patients with advanced non-small-cell lung cancer (NSCLC) who are treated with platinum-based chemotherapy is rather variable. Methylation-dependent transcriptional silencing of 14-3-3sigma, a major G2-M checkpoint control gene, could be a predictor of longer survival. PATIENTS AND METHODS: A sensitive methylation-specific polymerase chain reaction assay was used to evaluate 14-3-3sigma methylation status in pretreatment serum DNA obtained from 115 cisplatin-plus-gemcitabine-treated advanced NSCLC patients. RESULTS: 14-3-3sigma methylation was observed in all histologic types of 39 patients (34%). After a median follow-up of 9.8 months, median survival was significantly longer in the methylation-positive group (15.1 v 9.8 months; P = .004). Median time to progression was 8 months in the methylation-positive group and 6.3 months in the methylation-negative group (log-rank test, P = .027). A multivariate Cox regression model identified only 14-3-3sigma methylation status and Eastern Cooperative Oncology Group performance status as independent prognostic factors for survival. In an exploratory analysis, median survival for 22 methylation-positive responders has not been reached, whereas survival was 11.3 months for 29 methylation-negative responders (P = .001). CONCLUSION Methylation of 14-3-3sigma is a new independent prognostic factor for survival in NSCLC patients receiving platinum-based chemotherapy. It can be reliably and conveniently detected in the serum, thus obviating the need for tumor tissue analysis.
14-3-3 Proteins↗
Method of test administration as a factor in test validity: the use of a personality questionnaire in the prediction of cancer and coronary heart disease.
This study examines the predictive accuracy of four different methods of administration of a questionnaire designed to predict cancer and coronary heart disease (CHD) in healthy probands. The method of administration uses the establishment of trust and the explanation of questions as variables in all four possible combinations, i.e. trust and explanation, trust only, explanation only, and neither, the prediction being that the combination of trust and explanation would produce the most accurate prediction, the treatment using neither the worst prediction, with methods using either trust alone or explanation alone intermediate. The criterion was the successful prediction of cancer and CHD. A total population of 3563 men and women was used, and followed up over 15 years, death certificates being used to establish cause of death. As predicted, the combination of trust and explanation did best, use of neither worst. Explanation seemed more important than trust, and the combination seemed to have a synergistic effect. It is apparent that method of administration had an important effect on the outcome of the experiment.
Adult↗
Prediction of response to hormonal treatment in metastatic breast cancer.
Prediction of outcome and individualization of therapeutic strategies are challenging problems in oncology. Predictive parameters for response to hormonal treatment include the expression of hormone receptor, the extent and location of metastatic spread, disease-free interval, patient age, response to prior hormonal therapy, grading, and more recently, some molecular markers like the expression of HER-2/neu. The use of conventional statistics for prediction of response to hormonal treatment is limited by non-linearities and complex interactions between predictive factors. Modern computational mathematical models like artificial neural networks, entropy-based inductive algorithms or chi(2) interaction detection algorithms can describe these interactions and generate classification models and decision structures. They can be used to predict the clinical outcome for individual patients. In contrast to conventional methods, the level of confidence for the predictions can reach 90% and more. This might be an important step towards further individualization of therapeutic strategies.
Antineoplastic Agents, Hormonal↗
Predictive factors for response to endocrine therapy in patients with recurrent breast cancer.
Predictive markers and variables for response to anticancer therapy provide cancer patients with refinement of therapeutic options and a decreased likelihood of receiving an ineffective therapy. The best-established predictive marker for response to endocrine therapy for breast cancer is the status of estrogen receptors (ER) in the primary breast tumor. However, although patients with ER-positive tumors have a greater than 50% objective response rate to endocrine therapy, other patients can not obtain an objective response. Therefore additional markers, such as better molecular biologic markers, are needed. Our previous study using multivariate analysis revealed that the ER status of primary tumors and the dominant site of metastasis are independent predictors for response to first-line endocrine therapy and that a response to first-line endocrine therapy is only an independent predictor for response to second-line endocrine therapy. However, all these factors are already well-established predictive markers for response to endocrine therapy. Recently, a number of new hormonal agents, such as more selective aromatase inhibitors and specific antiestrogens, have been developed and introduced. However, several questions, such as the best sequences when using hormonal agents, remain to be elucidated. On the other hand, several molecular biologic markers predicting response to endocrine therapy, such as the expression of the HER family of tyrosine kinase receptors, pS2, Bcl-2, and vascular endothelial growth factor, have been reported. To elucidate the most effective use of endocrine therapy for recurrent breast cancer, classical and new predictive factors for response to endocrine therapy are reviewed, and the clinical implications of these factors are discussed.
Antineoplastic Agents, Hormonal↗
Will the Scottish Cancer Target for the year 2000 be met? The use of cancer registration and death records to predict future cancer incidence and mortality in Scotland.
Cancer mortality data reflect disease incidence and the effectiveness of treatment. Incidence data, however, reflect the burden of disease in the population and indicate the need for prevention measures, diagnostic services and cancer treatment facilities. Monitoring of targets mandates that both be considered. The Scottish Cancer Target, established in 1991, proposed that a reduction of 15% in mortality from cancer in the under-65s should be achieved between 1986 and 2000. Each year in Scotland approximately 8300 persons under 65 are diagnosed with cancer and 4500 die from the disease. The most common malignancies, in terms of both incident cases and deaths, in the under-65s, are lung and large bowel cancer in males, and breast, large bowel and lung cancer in females. A decrease of 6% in the number of cancer cases diagnosed in males under 65 is predicted between 1986 and 2000, whereas the number of cases in females in the year 2000 is expected to remain at the 1986 level. In contrast, substantial reductions in mortality are expected for both sexes: 17% and 25% in males and females respectively. Demographic changes will influence the numbers of cancer cases and deaths in the Scottish population in the year 2000. However, long-term trends in the major risk factors, such as smoking, are likely to be the most important determinants of the future cancer burden.
Aged↗
Use of nomograms as predictive tools in bladder cancer.
Bladder cancer is a common genitourinary malignancy that demonstrates a great variation in risk of tumor recurrence and progression following treatment. The dramatic differences in clinical behavior dictate vastly differing treatments, which may range from simple surveillance to combination radical surgery with systemic chemotherapy. For non-muscle invasive bladder cancer prediction of the risk of recurrence and progression is necessary to assess the need for intravesical therapy and possible early cystectomy. In contrast, prediction of advanced disease response to primary treatment such as cystectomy and the response to systemic chemotherapy plays an important role in treatment assignment for patients with muscle invasive disease. To estimate these risk traditional risk grouping schemes such as the present TNM staging system has been used to guide patient treatment. More recently, improved prognostic tools such as nomograms have been developed to provide a more accurate assessment of outcomes. Clinicians are enthusiastically working to utilize these statistical methods in bladder cancer. We summarize the current status of outcome predictive models for bladder cancer; and focus particularly on the ability of nomograms to predict disease recurrence, progression, and patient survival.
Algorithms↗
High-grade prostatic intraepithelial neoplasia with adjacent small atypical glands on prostate biopsy.
With high-grade prostatic intraepithelial neoplasia with adjacent small atypical glands (PINATYP), the issue is whether the small glands represent budding or tangentially sectioned glands off of adjacent high-grade prostatic intraepithelial neoplasia (PIN) or invasive cancer next to high-grade PIN. The histology and significance of PINATYP on biopsy have not been described. Among 574 cases of high-grade PIN lesions on needle biopsy, we identified 71 cases of PINATYP. Most cases were consultations, and 51 cases were available for histologic review. At least 1 follow-up prostate biopsy was performed in each of 55 cases. Immunohistochemistry for high-molecular-weight cytokeratin (HMWCK) was performed on cases in which material was available. The average patient age at diagnosis was 65.5 years (range, 48 to 103 years). The initial digital rectal examination, transrectal ultrasound, serum prostate-specific antigen (PSA) level, PSA velocity, and family history of prostate cancer did not predict cancer on repeat biopsy. In 39% of cases, high-grade PIN had a predominantly flat pattern, and remaining cases showed a predominance of other patterns (tufting, micropapillary, cribriform). The average number of high-grade PIN glands and adjacent small atypical glands were 11.5 (1 to 60) and 5.3 (1 to 21), respectively. The farthest adjacent small atypical gland averaged 0.12 mm from the high-grade PIN (0.01 mm to 0.4 mm), as measured with an ocular micrometer. The following histologic features did not predict cancer on repeat biopsy: more than 1 core involved by the high-grade PIN; number of high-grade PIN glands; number of small atypical glands; distance of small atypical glands from the high-grade PIN; size and percentage of nucleoli; marked nuclear pleomorphism; and mitoses. Overall, the risk of cancer on repeat biopsy was 46%. Two findings predicted a lower risk of cancer on repeat biopsy: younger age (62.2 years benign v 68.3 years cancer; P =.004) and predominantly flat high-grade PIN (P =.007). In our material, PINATYP appears to be a greater risk factor than high-grade PIN alone in predicting cancer on rebiopsy. Although age and predominant pattern of associated high-grade PIN may be helpful in predicting which men with this lesion will have cancer on rebiopsy, they cannot be used reliably; therefore, all men with PINATYP should undergo repeat biopsy. HUM PATHOL 32:389-395.
Aged↗