Search PubMed⌕ Search

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.

At least 235 records · Page 13Linked to original sources
PubMed2000

Chromosomal aberrations in lymphocytes predict human cancer independently of exposure to carcinogens. European Study Group on Cytogenetic Biomarkers and Health.

An increased risk of cancer in healthy individuals with high levels of chromosomal aberrations (CAs) in peripheral blood lymphocytes has been described in recent epidemiological studies. This association did not appear to be modified by sex, age, country, or time since CA test, whereas the role played by exposure to carcinogens is still uncertain because of the requisite information concerning occupation and lifestyle was lacking. We evaluated in the present study whether CAs predicted cancer because they were the result of past exposure to carcinogens or because they were an intermediate end point in the pathway leading to disease. A nested case-control study was performed on 93 incident cancer cases and 62 deceased cancer cases coming from two prospective cohort studies performed in Nordic countries (Denmark, Finland, Norway, and Sweden) and Italy. For each case, four controls matched by country, sex, year of birth, and year of CA test were randomly selected. Occupational exposure and smoking habit were assessed by a collaborative group of occupational hygienists. Logistic regression models indicated a statistically significant increase in risk for subjects with a high level of CAs compared to those with a low level in the Nordic cohort (odds ratio, 2.35; 95% confidence interval, 1.31-4.23) and in the Italian cohort (odds ratio, 2.66; 95% confidence interval, 1.26-5.62). These estimates were not affected by the inclusion of occupational exposure level and smoking habit in the regression model. The risk for high versus low levels of CAs was similar in subjects heavily exposed to carcinogens and in those who had never, to their knowledge, been exposed to any major carcinogenic agent during their lifetime, supporting the idea that chromosome damage itself is involved in the pathway to cancer. The results have important ramifications for the understanding of the role played by sporadic chromosome damage for the origin of neoplasia-associated CAs.

Adult↗
PubMed2005

Predicting breast cancer survivability: a comparison of three data mining methods.

OBJECTIVE: The prediction of breast cancer survivability has been a challenging research problem for many researchers. Since the early dates of the related research, much advancement has been recorded in several related fields. For instance, thanks to innovative biomedical technologies, better explanatory prognostic factors are being measured and recorded; thanks to low cost computer hardware and software technologies, high volume better quality data is being collected and stored automatically; and finally thanks to better analytical methods, those voluminous data is being processed effectively and efficiently. Therefore, the main objective of this manuscript is to report on a research project where we took advantage of those available technological advancements to develop prediction models for breast cancer survivability. METHODS AND MATERIAL: We used two popular data mining algorithms (artificial neural networks and decision trees) along with a most commonly used statistical method (logistic regression) to develop the prediction models using a large dataset (more than 200,000 cases). We also used 10-fold cross-validation methods to measure the unbiased estimate of the three prediction models for performance comparison purposes. RESULTS: The results indicated that the decision tree (C5) is the best predictor with 93.6% accuracy on the holdout sample (this prediction accuracy is better than any reported in the literature), artificial neural networks came out to be the second with 91.2% accuracy and the logistic regression models came out to be the worst of the three with 89.2% accuracy. CONCLUSION: The comparative study of multiple prediction models for breast cancer survivability using a large dataset along with a 10-fold cross-validation provided us with an insight into the relative prediction ability of different data mining methods. Using sensitivity analysis on neural network models provided us with the prioritized importance of the prognostic factors used in the study.

Breast Neoplasms↗
PubMed2005

Utility of saturation biopsy to predict insignificant cancer at radical prostatectomy.

OBJECTIVES: To determine whether potential candidates for watchful waiting have undersampling of more substantial cancer. METHODS: A total of 103 men were studied, who were predicted to have insignificant cancer in their radical prostatectomy (RP) specimen. All had limited cancer on routine needle biopsy (no core with more than 50% involvement; Gleason score less than 7, and fewer than 3 cores involved) with a serum prostate-specific antigen density of 0.15 or less. Insignificant tumor at RP was considered organ-confined tumor, no Gleason pattern 4 or 5, and a tumor volume of less than 0.5 cm3. Saturation biopsy (average 44 cores) and an alternate biopsy saturation scheme with one half the number of cores using an 18-gauge Biopty gun was performed in the pathology laboratory on totally embedded and serially sectioned RP specimens. RESULTS: Of the tumors, 97% were organ confined. The RP Gleason score was less than 7 in 84% of the cases. The RP tumor volume was 0.01 to 2.39 cm3 (median 0.14). Of the cancer specimens, 71% were insignificant and 29% had been incorrectly classified before surgery using standard biopsy schemes. Using the full saturation biopsy scheme, if we predicted significant cancer, the probability of having insignificant cancer was only 11.5% (false-positive rate). If the model predicted insignificant cancer, the probability of significant cancer was also only 11.5% (false-negative rate; sensitivity 71.9% and specificity 95.8%). Using the alternate biopsy sampling scheme, the false-positive rate was 8% and the false-negative rate was 11.4% (sensitivity 71.9% and specificity 97.1%). CONCLUSIONS: Saturation biopsy provides accurate predictability of prostate tumor volume and grade to select suitable candidates for watchful waiting therapy.

Adult↗
PubMed1994

Use of mechanistic models to estimate low-dose cancer risks.

The utility of mechanistic models of cancer for predicting cancer risks at low doses is examined. Based upon a general approximation to the dose-response that is valid at low doses, it is shown that at low doses the dose-response predicted by a mechanistic model is a linear combination of the dose-responses for each of the physiological parameters in the model that are affected by exposure. This demonstrates that, unless the mechanistic model provides a theoretical basis for determining the dose-responses for these parameters, the extrapolation of risks to low doses using a mechanistic model is basically "curve fitting," just as is the case when extrapolating using statistical models. This suggests that experiments to generate data for use in mechanistic models should emphasize measuring the dose-response for dose-related parameters as accurately as possible and at the lowest feasible doses.

Carcinogens↗
PubMed1996

Serial cytological assay of micronucleus induction: a new tool to predict human cancer radiosensitivity.

BACKGROUND AND PURPOSE: The micronucleus test, generally done in cultured tumour cells irradiated in vitro, has not gained wide acceptance in predicting human cancer radiosensitivity. The purpose of this study was to see if micronucleus assay by serial scrape smear cytology can predict oral cancer radiosensitivity. MATERIALS AND METHODS: Forty nine oral cancer patients given radiotherapy (60 Gy/25 fractions/5 weeks) form the study population. Serial scrape smears were taken from their tumours before treatment and after delivery of 2, 5, 8 and 12 fractions, stained by Giemsa and the number of micronucleated cells (MNC) noted. The patients were grouped to those who developed tumour recurrence ('Resistant') and those who did not ('Sensitive'), and the pattern of micronucleus induction compared. RESULTS: Both groups of tumours had MNC even before treatment, with statistically significant dose-related increase with radiotherapy. The sensitive group had a higher mean increase in MNC count than the resistant group (6.1 times and 3.6 times the pre-treatment value, respectively) and better correlation with dose (r = 0.54 vs. 0.43). The increase in MNC count occurred earlier in the resistant group than in the sensitive, the TMNC (time for the pre-treatment value to double) being 3.3 days and 7.6 days, respectively. Also, the resistant group showed a plateauing of the MNC count which the sensitive group lacked. CONCLUSIONS: The higher MNC induction in the sensitive tumours suggests the usefulness of the assay as a test of radiosensitivity. The differing patterns of MNC increase suggest that differences in proliferation rate is an important cause of tumour failure. Serial cytological assay of micronucleus induction can identify both radiosensitivity and proliferation characteristics of tumours, and thus may turn out to be a useful test of radiocurability.

Carcinoma, Squamous Cell↗
PubMed1995

Core cancer length in ultrasound-guided systematic sextant biopsies: a preoperative evaluation of prostate cancer volume.

OBJECTIVES: Evidence has been presented that biologic aggressiveness of prostate carcinoma increases with volume and that cancers less than 0.5 cc may be regarded as clinically insignificant for immediate treatment. We have analyzed the ability of core cancer length on needle biopsy to predict cancer volumes near the 0.5 cc threshold for distinction between incidental versus clinical carcinoma. METHODS: Systematic sextant transrectal ultrasound-guided prostate biopsies were performed on 110 men who then underwent radical prostatectomy for adenocarcinoma. The core cancer length within each biopsy was compared with the volumes of clinical and incidental carcinomas in the prostatectomy specimen. RESULTS: Among incidental (nonpalpable) cancers, 14% of those under 0.2 cc were detected, but 44% at 0.2 to 0.5 cc and 92% of those more than 0.5 cc were detected. Among clinically suspected carcinomas, 2 mm or longer core cancer length reliably predicted a cancer of 0.5 cc or larger, but among incidental cancers, it predicted a tumor of 0.2 cc or larger. A 3-mm core cancer length threshold predicted 0.5 cc for both groups. The high frequency of incidental cancers under 0.5 cc impaired the predictive value of multiple positive needle cores. Bilateral positive biopsy results indicated bilateral extension of clinical cancer in only 59% of cases. CONCLUSIONS: A core cancer length of 3 mm or more on one or two needle biopsies reliably predicts cancer of clinically significant volume (0.5 cc or larger). The high detection frequency of smaller incidental carcinomas on biopsy impairs the reliability of volume estimation from multiple positive needle cores and mandates that treatment decisions be made with knowledge of core cancer length.

Adenocarcinoma↗
PubMedDate not supplied

Prognostic significance of routine clinical and laboratory data in advanced head and neck cancers.

Predictive factors for toxicity and response to chemotherapy in patients with advanced head and neck cancer are seldom reported. Therefore, from a short series of patients with a histologically proven cancer, who were treated by a neo-adjuvant protocol with cisplatin and fluorouracil, routine clinical and laboratory data were investigated. ALT (alanine aminotransferase) and Hb (hemoglobin) appeared to be predictive for efficacy. By multivariate analysis (principal component analysis), these laboratory data were involved in two independent axes: one which was considered as "inflammatory" and the other as "hepatic". Initial obesity indices were associated with the former. The predictive variables for toxicity (i.e. age, serum creatinine level, weight loss and plasma cisplatin) were probably biased in this series. Nevertheless cisplatin concentration regularly increased in each cycle. Hence it was dependent on the rank of the course. According to this preliminary study, it would be of interest to conduct future investigations on acquired protein-energy malnutrition, as well as on selected soluble mediators of cellular and humoral immune response.

Adult↗
PubMed2007

Bleomycin-induced DNA damage and repair in Xenopus laevis and Xenopus tropicalis.

Microgel cell electrophoresis has been used with various species to measure breakage of DNA and DNA repair following exposure to the radiomimetic antibiotic, bleomycin. With humans, a high degree of DNA damage is considered to be predictive of cancer susceptibility. Non-isogeneic Xenopus laevis, the South African clawed toad, rarely develop spontaneous or induced cancers. Here, we investigate bleomycin-induced DNA damage and repair in splenic lymphocytes of this species to test consistency with cancer predictability. As X. laevis is pseudotetraploid in nature, while Xenopus tropicalis is diploid, we additionally explore the effect of polyploidy on DNA damage and repair in these vertebrates. The results show that higher doses of bleomycin are required to induce comparable levels of DNA damage in both Xenopus species, than in humans. X. tropicalis, the diploid, is more bleomycin-sensitive than is X. laevis. Additionally, repair rates of damaged DNA of X. laevis lymphocytes are more rapid than those of X. tropicalis, although both are hours slower than human leukocytes. While no data exist on cancer susceptibility in X. tropicalis, the results suggest greater susceptibility to cancer than X. laevis, but less than in humans. Thus, polyploidy serves as a protection against DNA damage and allows more rapid repair.

Animals↗
PubMed2000

Ovarian carcinoma in situ with germline BRCA1 mutation and loss of heterozygosity at BRCA1 and TP53.

BACKGROUND: The two-hit hypothesis for the genesis of cancer predicts that cancer can develop when the wild-type allele of a tumor suppressor gene is lost in an individual with a germline mutation in that gene. Neither loss of heterozygosity (LOH) for BRCA1 nor mutations of the TP53 (also known as p53) gene have been documented prior to invasion in ovarian cancers arising in women with germline BRCA1 mutations. Such documentation is difficult because lesions are rarely identified in ovarian epithelium. We, therefore, looked for LOH at microsatellite polymorphisms linked to the BRCA1 and TP53 tumor suppressor loci in an incidental carcinoma in situ of the ovary removed prophylactically from a woman with a germline BRCA1 mutation. METHODS: By use of laser-capture microdissection, we obtained pure populations of atypical ovarian epithelial cells and normal stromal cells. DNA was extracted, amplified with primers flanking polymorphic microsatellites linked to BRCA1 (D17S855 and D17S579) and TP53 (TP53 and D17S786), and analyzed for LOH at these microsatellites. We also tested for p53 expression in the abnormal epithelium by immunohistochemistry. RESULTS: Both of the markers linked to TP53 showed LOH, as did an intragenic BRCA1-linked marker (D17S855). The other microsatellite marker for BRCA1 was uninformative. Immunohistochemical staining with an antibody to p53 showed strong immunoreactivity confined to the atypical epithelium. CONCLUSIONS: BRCA1, as well as TP53, can undergo LOH prior to stromal invasion in BRCA1-associated ovarian cancer. Strong immunoreactivity for p53 suggests the presence of mutated p53 in these cells as well. These findings suggest that loss of function of these two tumor suppressor genes occurs early in ovarian carcinogenesis in BRCA1 mutation carriers.

Antibodies, Neoplasm↗
PubMed2007

Lung cancer risk prediction: a tool for early detection.

Although 45% of men and 39% of women will be diagnosed with cancer in their lifetime, it is difficult to predict which individuals will be affected. For some cancers, substantial progress in individual risk estimation has already been made. However, relatively few models have been developed to predict lung cancer risk beyond effects of age and smoking. This paper reviews published models for lung cancer risk prediction, discusses their potential contribution to clinical and research settings and suggests improvements to the risk modeling strategy for lung cancer. The sensitivity and specificity of existing cancer risk models is less than optimal. Improvement in individual risk prediction is important for selection of individuals for prevention or early detection interventions. In addition to smoking, factors related to occupational exposure, personal medical history and family history of cancer can add to the predictive power. A good risk prediction model is one that can identify a small fraction of the population in which a large proportion of the disease cases will occur. In the future, genetic and other biological markers are likely to be useful, although they will require rigorous evaluation. Validation is essential to establish the predictive effect and for ongoing monitoring of the model's continued relevance.

Early Diagnosis↗
PubMed2004

Preoperative neural network using combined magnetic resonance imaging variables, prostate specific antigen and Gleason score to predict prostate cancer stage.

PURPOSE: We developed an artificial neural network analysis (ANNA) to predict prostate cancer pathological stage more effectively than logistic regression (LR) based on the combined use of prostate specific antigen (PSA), biopsy Gleason score and pelvic coil magnetic resonance imaging (pMRI) in patients with clinically organ confined disease before radical prostatectomy. MATERIALS AND METHODS: In 201 consecutive patients undergoing radical retropubic prostatectomy with pelvic lymphadenectomy the radiological-pathological correlation was evaluated using pMRI. Predictive variables were clinical TNM classification, preoperative serum PSA, biopsy Gleason score and pMRI findings. The predicted results were organ confined vs nonorgan confined disease and lymphatic vs no lymphatic involvement. The predicted ability of ANNA with several parameters in a set of 160 randomly selected test data was compared with that of LR and the Partin tables by area under the receiver operating characteristic curve analysis. RESULTS: The overall accuracy of ANNA and LR was 88% and 91%, and 77% and 84% for nonorgan confined and lymphatic involvement, respectively. For nonorgan confined disease and lymph node involvement the area under the curve of ANNA (0.895 and 0.899) was significantly larger than that of LR and the Partin tables (0.722 and 0.751, and 0.750 and 0.733, respectively, p <0.05). Gleason score represented the most influential predictor (relative weight 2.05) of nonorgan confined disease, followed by pMRI findings (1.96), PSA (1.73) and clinical stage (0.89). CONCLUSIONS: ANNA is superior to LR for accurately predicting pathological stage. The relative importance of pMRI findings and the usefulness of ANNA for predicting pathological stage in individuals must be confirmed in a prospective trial.

Adult↗
PubMed1997

Value of preoperative CA 125 level in the management of uterine cancer and prediction of clinical outcome.

OBJECTIVE: To enhance cost-effective management of uterine cancer by predicting the likelihood of extrauterine disease and survival on the basis of preoperative parameters. METHODS: A retrospective review of preoperative CA 125 levels from 210 women with endometrial carcinoma was performed. The relationship of preoperative CA 125 levels to various preoperative and postoperative histopathologic factors was investigated. RESULTS: Elevated CA 125 (greater than 35 U/mL) correlated (P < .05) with higher stage, higher grade, increased depth of myometrial invasion, positive cytology, pelvic or para-aortic lymph node metastases, and reduced actuarial survival (P < .001). Multivariate analysis of preoperative factors showed that an elevated CA 125 level was the most important predictor for poor survival (P < .001). Moreover, a preoperative CA 125 level greater than 65 U/mL was the most significant predictor of extrauterine disease and carried a 6.5-fold higher risk (95% confidence interval 2.5, 17.1). A logistic model to predict extrauterine disease was developed. The model has a sensitivity of 62%, specificity of 91%, positive predictive value of 69%, and negative predictive value of 88%. CONCLUSION: A CA 125 level should be included as part of the preoperative workup for all patients with uterine cancer. Patients with a preoperative CA 125 level less than or equal to 20 U/mL should be considered as candidates for vaginal hysterectomy unless unfavorable histology or a high-grade (grade II or III) tumor is present. In our series, this approach would have eliminated 24% of the abdominal staging procedures, with a risk of less than 3% for extrauterine disease, while lowering treatment-related morbidity and cutting costs in the treatment of this common female cancer.

Adult↗
PubMed2003

Artificial intelligence in predicting bladder cancer outcome: a comparison of neuro-fuzzy modeling and artificial neural networks.

PURPOSE: New techniques for the prediction of tumor behavior are needed, because statistical analysis has a poor accuracy and is not applicable to the individual. Artificial intelligence (AI) may provide these suitable methods. Whereas artificial neural networks (ANN), the best-studied form of AI, have been used successfully, its hidden networks remain an obstacle to its acceptance. Neuro-fuzzy modeling (NFM), another AI method, has a transparent functional layer and is without many of the drawbacks of ANN. We have compared the predictive accuracies of NFM, ANN, and traditional statistical methods, for the behavior of bladder cancer. EXPERIMENTAL DESIGN: Experimental molecular biomarkers, including p53 and the mismatch repair proteins, and conventional clinicopathological data were studied in a cohort of 109 patients with bladder cancer. For all three of the methods, models were produced to predict the presence and timing of a tumor relapse. RESULTS: Both methods of AI predicted relapse with an accuracy ranging from 88% to 95%. This was superior to statistical methods (71-77%; P < 0.0006). NFM appeared better than ANN at predicting the timing of relapse (P = 0.073). CONCLUSIONS: The use of AI can accurately predict cancer behavior. NFM has a similar or superior predictive accuracy to ANN. However, unlike the impenetrable "black-box" of a neural network, the rules of NFM are transparent, enabling validation from clinical knowledge and the manipulation of input variables to allow exploratory predictions. This technique could be used widely in a variety of areas of medicine.

Artificial Intelligence↗
PubMed1996

Can insignificant prostate cancer be predicted preoperatively in men with stage T1 disease?

This study reviews the pathological findings in nonpalpable prostate cancer detected on transurethral resection (TUR) (stages T1a and T1b) and on needle biopsy (stage T1c) to determine whether insignificant cancers can be predicted preoperatively. The majority of stage T1a cancers are insignificant and all stage T1b cancers are significant. Preoperative pathological parameters cannot identify the minority of T1a cancers that are more significant and the minority of T1b cancers with relatively little residual cancer. Although most nonpalpable tumors diagnosed by screening techniques are usually significant tumors, which would be treated aggressively if they were palpable, 12% to 26% of cancers removed from nonpalpable disease are relatively insignificant. There is conflicting data as to whether a combination of needle biopsy findings and PSA density measurements can aid in the prediction of insignificant cancers. If conservative management is considered, a patient should undergo repeat biopsies (including transition zone biopsies) to detect those more significant cancers that may have been undersampled. In addition, serial serum PSA measurements should be performed to detect whether serum PSA levels increase to a point where these tumors would be reclassified as significant. When planning therapy for T1 patients, the age, comorbidity, and treatment preferences of the patient are key factors that must also be taken into account.

Biopsy, Needle↗
PubMed2003

Application of breast cancer risk prediction models in clinical practice.

Breast cancer risk assessment provides an estimation of disease risk that can be used to guide management for women at all levels of risk. In addition, the likelihood that breast cancer risk is due to specific genetic susceptibility (such as BRCA1 or BRCA2 mutations) can be determined. Recent developments have reinforced the clinical importance of breast cancer risk assessment. Tamoxifen chemoprevention as well as prevention studies such as the Study of Tamoxifen and Raloxifene are available to women at increased risk of developing breast cancer. In addition, specific management strategies are now defined for BRCA1 and BRCA2 mutation carriers. Risk may be assessed as the likelihood of developing breast cancer (using risk assessment models) or as the likelihood of detecting a BRCA1 or BRCA2 mutation (using prior probability models). Each of the models has advantages and disadvantages, and all need to be interpreted in context. We review available risk assessment tools and discuss their application. As illustrated by clinical examples, optimal counseling may require the use of several models, as well as clinical judgment, to provide the most accurate and useful information to women and their families.

Breast Neoplasms↗
PubMed2003

The natural history of prostate cancer.

Predicting the long-term outcome of patients who choose watchful waiting as initial therapy for prostate cancer is difficult. The wide variation in disease progression, the impact of competing medical hazards, and the potential impact of early hormonal therapy that is characteristic of contemporary patients all conspire to compromise survival estimates dating from the pre-PSA era. The survival analysis figure developed by Albertsen et al (Fig. 1) estimates a 15-year survival rate based on patient age and Gleason score at diagnosis from patients diagnosed in the pre-PSA era. Although no effort was made to adjustfor competing medical hazards, patients and clinicians can adjust a patient's chronological age to match his "physiological" age. The advent of widespread PSA testing appears to have advanced the date of diagnosis by approximately 5 years and the onset of secondary treatment by at least as many years. Therefore, the figure describing the natural history of prostate cancer most likely underestimates rather than overestimates survival among men with newly diagnosed, localized prostate cancer who select watchful waiting as their treatment choice. As contemporary databases of men with localized prostate cancer mature, more data on the natural history of this disease will become available. Only time will tell how the use of PSA has altered the precision of historic case-series data.

Adenocarcinoma↗
PubMed1989

[Future prediction of cancer deaths in Japan].

The number of cancer deaths and age-adjusted cancer death rates up to 2000 in Japan were predicted based on cancer death rates in 1972-1986. A simple linear regression model (y = a + bt, where t is calendar year and y is cancer death rate per 100,000 population) was fitted to the sex-age specific cancer mortality rates from 1972 to 1986 and cancer death rates in 1990, 1995 and 2000 were predicted by extrapolation method. The number of future cancer deaths was estimated after taking into account future population in Japan. The age-adjusted cancer rates up to 2000 were also estimated. The present study revealed that the numbers of deaths from stomach cancer (both sexes), uterine cancer and esophageal cancer (females) would keep declining, while all other cancers would increase in the future. The total number of cancer deaths in 2000 was estimated to be about 310,000 which is 1.62 times that of the total cancer deaths in 1986 (191,654). It was estimated that in 2000, lung cancer would rank top accounting for 22.2% of all cancer deaths, followed by cancers of the large intestine, liver, stomach, pancreas, biliary tract, leukemic, lymphoma and breast.

Adolescent↗