Prostate Cancer

As many of you may already know, carcinoma of the prostate (CaP) is one of the most common types of malignancy in men; in the U.S. alone there will be an estimated 186,000 men diagnosed with it in 2008.  CaP is usually suspected due to an abnormal blood test, the prostate-specific antigen (PSA), a suspicious feeling prostate on physical exam, or both. 

Biopsies of the prostate are taken through the rectal area to make the diagnosis of prostate cancer.  In each man whose biopsies show cancer, he must decide how to arrive at the right approach to managing this disease with the guidance and support of his doctors and loved ones.

Estimating Disease Extent and Making Decisions About Treatment

Until you have a problem, often you don’t know much about how to pick the most probable, effective solution.  With prostate cancer, many newly diagnosed men are bombarded with a tremendous amount of complex information in a very short period of time.  Men with prostate cancer are then given treatment recommendations and are asked to determine what the best approach is for them.

The health care professionals usually best trained in the management of prostate cancer are urologists, radiation oncologists and medical oncologists.   However, doctors in each area may have different expertise, perspectives upon treatment philosophy, and sources for information for making helpful recommendations to patients.  Furthermore, each health care professional may recall specific prostate cancer outcomes from published medical articles.  Whether that information applies to each individual patient may be challenging to determine quickly without reviewing the individual articles, which are not always easily available for review.

Using biostatistics, many prostate cancer researchers have developed predictive models to assist doctors in making evidence-based recommendations to patients.  Many of these predictive models have been published and are comprehensive enough to be applied to many (though not all) men with newly diagnosed prostate cancer.  Historically, predictions of outcome were based more commonly upon patients being placed in a risk group (e.g. low, intermediate or high risk).  However, more sophisticated biostatistical techniques such as nomograms and artificial neural networks have allowed outcomes to be predicted in a more individualized manner.  Although some of these more sophisticated predictive tools are available, they may not be easily available in a timely, user-friendly format.

In newly diagnosed prostate cancer, the three main tumor-related factors guiding treatment decisions since the early 1990s have been AJCC clinical stage, Gleason score, and PSA level.  However, the important features predicting outcomes in one group of prostate cancer patients may not be proven statistically to be important in another group.  As a result, different details (also called ‘variables’) of each individual and his prostate cancer may be considered relevant depending upon which predictive model is used.  In some circumstances, not all the information used to make treatment decisions at an academic research center is widely available elsewhere.  Therefore, applying information from one published study may not be widely applicable outside of that institution.

In addition to possibly having incomplete information at the time of diagnosis, the knowledge base of both patient and physician, their rapport, and time constraints all represent challenges to making informed treatment decisions regarding prostate cancer.

CaP Calculator:  Compare Different Systems Predicting Prostate Cancer Outcomes

Because of these numerous challenges, we decided to develop a single evidence-based resource that uses peer-reviewed published articles to help both doctors and patients have a more meaningful discussion of risk assessment, informed consent and shared decision-making in newly diagnosed prostate cancer.  Instead of focusing upon one article or approach, we developed a simple electronic application that allows comparison of multiple predictive models in prostate cancer simultaneously in an accessible, user-friendly format. 

With CaP, disease spread beyond the prostate and seminal vesicles (lymph node involvement or other parts of the body) at diagnosis are less common that it used to be before the use of PSA screening.  We hope there will be enough interest to expand CaP Calculator into other important areas in prostate cancer in the future, but currently this tool’s potential usefulness is limited to men with clinically T1-3 prostate cancer.

No actual patient information is available through CaP Calculator.  We use the materials presented in peer-reviewed journals rather than the original data.  We do not have access to the original data used to generate any of these predictive tools, only the means to reproduce them for comparison to other research.  We do hope to validate this tool in the near future to confirm its usefulness.  If you would like to help us with this project or develop CaP Calculator further, please contact us.

While it does provide multiple outcomes predictions in a single location, the convenience of CaP Calculator filters out potentially important details published with each study that makes up part of this tool.  Therefore, we have included a bibliography of the articles included in CaP Calculator. 

CaP Calculator was originally designed in Microsoft Excel, but we felt it would be more accessible by placing it on the Internet for use online.  

To learn more about the intended use of CaP Calculator, use the highlighted link.