Cancer Screening Tests

The statistics on cancer are grim. We are moving toward a time when half of those born will be diagnosed with cancer in their lifetime. And while improvement has been made in the treatment of a few types of cancer, overall the effectiveness of conventional cancer treatment has not changed much over the past 30 years.

A great deal of the improved outcomes that have occurred with cancer treatment can be attributed to earlier detection. The more advanced a cancer is at the time it is diagnosed, the less likely a therapy will be beneficial. Unfortunately, the most thorough and accurate means of diagnosing cancers can increase the risk of cancer. CT scans and X-rays both carry the risk of causing the very problem they are trying to diagnose. For this reason they are not used for screening.

The exception to this is the use of mammography, though even recommendations around this procedure are beginning to change. Evidence is showing that the radiation associated with mammography could be contributing to the development of breast cancer in some women.

MRIs do not involve radiation exposure, but are too expensive to be used for screening.

An ideal screening test would be a blood test that identifies something in the blood that is uniquely associated with cancer activity. There are a wide range of “tumor markers” that are often used to monitor various types of cancer after they have been diagnosed, but such markers are not typically used a screening tests due to their poor sensitivity and specificity. These words refer to very important aspects of any test.

Sensitivity is a measure of how often the test will be positive in a person who really has the condition being tested. So, if there are 100 people known to have cancer, and if a certain cancer-screening blood test is positive in all 100 of them, then we can say that the test has a sensitivity of 100%. That is a very high sensitivity.

However, suppose we do the same blood test on 100 people known to *not* have cancer, but who have conditions such as rheumatoid arthritis, lupus, migraines, as well as some who are completely healthy. If that same blood test is positive in all 100 of those people, then we know that the test is not specific for cancer. In fact, it’s specificity for cancer is 0%. That would be a very poor test to use as a screening test, because you would never know if a positive test meant that cancer is present.

Most tumor marker tests currently used today have moderately good sensitivity and specificity. However, if cancer is what’s being screened for, “moderately good” isn’t good enough. A test that incorrectly indicates someone has cancer could lead to extreme stress and thousands of dollars worth of unneeded additional tests and scans. For this reason, use of standard tumor markers is not recommended as screening tools.

There are no tests currently used in the conventional medical world that can safely and accurately screen individuals for the presence of cancer with high sensitivity and specificity. Researchers around the world are working to develop such a test. Next week I will review some of the most promising tests that are currently in use by some physicians (myself included), but that have not been approved by the FDA for widespread use. As such they are considered experimental.