Late rectal bleeding after 3D-CRT for prostate cancer: development of a neural-network-based predictive model.
Summary of "Late rectal bleeding after 3D-CRT for prostate cancer: development of a neural-network-based predictive model."
The aim of this study was to develop a model exploiting artificial neural networks (ANNs) to correlate dosimetric and clinical variables with late rectal bleeding in prostate cancer patients undergoing radical radiotherapy and to compare the ANN results with those of a standard logistic regression (LR) analysis. 718 men included in the AIROPROS 0102 trial were analyzed. This multicenter protocol was characterized by the prospective evaluation of rectal toxicity, with a minimum follow-up of 36 months. Radiotherapy doses were between 70 and 80 Gy. Information was recorded for comorbidity, previous abdominal surgery, use of drugs and hormonal therapy. For each patient, a rectal dose-volume histogram (DVH) of the whole treatment was recorded and the equivalent uniform dose (EUD) evaluated as an effective descriptor of the whole DVH. Late rectal bleeding of grade ≥ 2 was considered to define positive events in this study (52 of 718 patients). The overall population was split into training and verification sets, both of which were involved in model instruction, and a test set, used to evaluate the predictive power of the model with independent data. Fourfold cross-validation was also used to provide realistic results for the full dataset. The LR was performed on the same data. Five variables were selected to predict late rectal bleeding: EUD, abdominal surgery, presence of hemorrhoids, use of anticoagulants and androgen deprivation. Following a receiver operating characteristic analysis of the independent test set, the areas under the curves (AUCs) were 0.704 and 0.655 for ANN and LR, respectively. When evaluated with cross-validation, the AUC was 0.714 for ANN and 0.636 for LR, which differed at a significance level of p = 0.03. When a practical discrimination threshold was selected, ANN could classify data with sensitivity and specificity both equal to 68.0%, whereas these values were 61.5% for LR. These data provide reasonable evidence that results obtained with ANNs are superior to those achieved with LR when predicting late radiotherapy-related rectal bleeding. The future introduction of patient-related personal characteristics, such as gene expression profiles, might improve the predictive power of statistical classifiers. More refined morphological aspects of the dose distribution, such as dose surface mapping, might also enhance the overall performance of ANN-based predictive models.
Department of Medical Physics, Fondazione IRCCS Istituto Nazionale Tumori, via Venezian 1, 20133 Milano, Italy.
This article was published in the following journal.
Name: Physics in medicine and biology
- PubMed Source: http://www.ncbi.nlm.nih.gov/pubmed/22349550
- DOI: http://dx.doi.org/10.1088/0031-9155/57/5/1399
This study was done to assess the impact of clinical factors and in particular the use of drugs for concomitant illnesses on late radiation-induced rectal bleeding in patients with prostate cancer.
Prostate cancer is a very common and highly unpredictable form of cancer. Whereas many prostate cancers are slow growing and could be left without treatment, others are very aggressive. Additionally,...
In spite of increased prostate cancer patients, little is known about the impact of treatments for prostate cancer patients and outcome of different treatments based on nationwide data. In order to ob...
Prostate cancer is the second leading cause of cancer-related death among men in North America. While a majority of prostate cancer cases remain indolent, subsets of patients develop aggressive cancer...
Prostate cancer is the most common cancer in Western society males, with incidence rates predicted to rise with global aging. Etiology of prostate cancer is however poorly understood, while current di...
There is controversy concerning the preferred treatment for men with localized prostate cancer with regard to the relative effectiveness of different treatments in controlling the cancer a...
Primary objective: Estimate the incidence of acute and late side effects of proton radiation therapy for prostate cancer including the development of second malignancies. Secondary...
There are theoretical reasons to believe that radiotherapy (RT) given in larger fractions during a shorter period of time might result in more late side effects than giving a conventional,...
Prostate cancer is the most common malignancy in males, and radiotherapy is a commonly chosen treatment option for patients with localized disease. Technical innovations such as three-dime...
RATIONALE: Vitamin E supplements may stop or delay the development of prostate cancer in patients who are at risk of prostate cancer or who have prostate cancer. It is not yet known which...
Medical and Biotech [MESH] Definitions
A glycoprotein that is a kallikrein-like serine proteinase and an esterase, produced by epithelial cells of both normal and malignant prostate tissue. It is an important marker for the diagnosis of prostate cancer.
A synthetic retinoid that is used orally as a chemopreventive against prostate cancer and in women at risk of developing contralateral breast cancer. It is also effective as an antineoplastic agent.
The two longitudinal ridges along the PRIMITIVE STREAK appearing near the end of GASTRULATION during development of nervous system (NEURULATION). The ridges are formed by folding of NEURAL PLATE. Between the ridges is a neural groove which deepens as the fold become elevated. When the folds meet at midline, the groove becomes a closed tube, the NEURAL TUBE.
Tissue ablation of the PROSTATE performed by ultrasound from a transducer placed in the RECTUM. The procedure is used to treat prostate cancer (PROSTATIC NEOPLASMS) and benign prostatic hypertrophy (PROSTATIC HYPERPLASIA).
Proteins secreted by the prostate gland. The major secretory proteins from the human prostate gland include PROSTATE-SPECIFIC ANTIGEN, prostate-specific acid phosphatase, prostate-specific membrane antigen, and prostate-specific protein-94.