In addition to false-positive tests, harms may also follow from true-positive tests. Not all people with true-positive screening tests benefit from the earlier detection of cancer. One can think of people with true-positive tests as having cancers in one of four categories.
Category 1: People with an aggressive, malignant cancer may not benefit from screening because the cancer has already metastasized before it can be detected. We are learning, in fact, that some cancers may metastasize within the first few cell divisions, too early to be the target of screening. Category 2: Other people with slower-growing cancers may be highly treatable even after clinical detection. Testicu-lar cancer may be such a tumor; our treatments are highly effective without the need of early detection. People with such cancers do not benefit from screening. Category 3: Some people may have pseudodisease, cancers that do not need treatment at all. These people either have intermediate lesions that would not progress but are still considered positive tests (e.g., small colonic adenomas) or have cancer that would not cause clinically important problems for the person in his/her lifetime. These are lesions that appear to be cancer but do not act as we think cancer usually acts. These people cannot benefit from early detection of their "cancer." Category 4: These are people who can benefit from earlier detection. These people have cancers that are potentially lethal but which can be treated more effectively because they were found earlier. In this case, the criterion is met that the treatment must be more effective if applied after screening detection than later, after clinical detection. Usually, this group of true-positive cancers is a minority of all true positives. The randomized controlled trials of breast cancer screening, for example, tell us that less than 20% of potentially lethal breast cancers (categories 1 and 4) belong to group 4.
A problem with this formulation, however, is that many cancers can only be placed in their proper category retrospectively. That is, the people in category 3, who do not need to be detected or treated, are often initially difficult to distinguish from the other groups. Thus, people in this category are still treated. An example is men with prostate cancer detected by screening. The majority of men with screening-detected prostate cancer have tumors that are moderately differentiated. Some cancers of this type are potentially lethal whereas others will never cause clinical problems. Because it is impossible to distinguish these cancers with high confidence at the time of diagnosis, virtually all men with this type of cancer are treated. This constitutes overdiagnosis, as we are diagnosing some men with cancer who do not need to be diagnosed, and overtreatment, as we are treating some men who do not need treatment.
The fact of overtreatment is undeniable and likely occurs with many cancers. The most important question is how often it occurs. Determining the number of people in category 3 (the primary group that is affected by overtreatment), however, is not simple. One can consider the issue in either of two ways: pathologically or epidemiologically. These different approaches explain much of the debate about the frequency of "clinically important" prostate cancers.
The pathologic approach to determining the frequency of cancers that do not need treatment uses grade and other cellular prognostic characteristics to determine prognosis at the time of diagnosis. People who are at risk of overtreatment have cancers with more benign characteristics. The problem with this approach is that none of the known prognostic characteristics is able to separate benign from malignant cancers with a high degree of accuracy. For example, one population-based study found that from 40% to 70% of men (depending upon age) with localized Gleason score 7 prostate cancer died of prostate cancer within 15 years of diagnosis.13 This finding also means that 30% to 60% of men with this type of cancer did not die of prostate cancer in that time. As these men were diagnosed before widespread PSA screening, it is likely that these survival figures would be higher today, independent of any changes in the effectiveness of treatment. Thus, the Gleason score and extent of tumor only give partial information about prognosis, and we are uncertain about whether an individual man will die of prostate cancer.
Another approach is based on the epidemiology of the cancer. This approach examines such issues as the difference between incidence and mortality; trends over time; changes in the effectiveness of treatment; and the lead time produced by the screening test. Using these assumptions with statistical modeling, investigators can calculate an approximation of the proportion of cancers that would not have caused problems during the person's lifetime. The problem with this approach is that it is based on a number of assumptions, at least some of which may be incorrect.
The best way to calculate the percentage of cancers that would never become clinically apparent is an analysis of results from a randomized controlled trial (RCT) of screening, comparing invited and control groups. If the trial screens people in the invited group for several years and then stops screening, the initial increase in incidence usually seen in the invited group compared with the control group should gradually decrease after the end of screening, as the cancers in the control group are detected at a later time. If the cumulative incidence of cancer in the control group never catches up with the invited group, this is evidence of detection by screening (in the invited group) of cancers that would never become clinically apparent. This approach may theoretically underestimate the true frequency of overdiagnosis, however, as it does not count cancers that produce only minimal symptoms (but symptoms sufficient to be diagnosed) in the overdiagnosis category. Although such cancers do cause some symptoms, they may grow so slowly that they would never progress to important clinical problems within the lifetime of the individual. The extent to which such cancers exist is unknown, but they do not need to be diagnosed early.
Overtreatment causes harm in a number of ways. First, the individual has been labeled as a "cancer patient," with likely important consequences for the person's life. Second, most cancer treatments have some side effects, some of which may be long lasting. Thus, in attempting to gain additional life in the future, people must undergo immediate harm from treatment. Finally, the large number of people being treated leads to an exaggerated view by professionals and the public of the true frequency of the cancer and the effectiveness of treatment.14 Further, 5-year survival statistics, which improve as more benign cancers are detected and treated, provide an erroneous overestimate of the efficacy of treatment (15) (see following discussion), and many "cancer survivors" are actually people who had either benign-type cancers (category 2) or pseudodisease (category 3) (see Cancer Survivors, later in this chapter).
Was this article helpful?
Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.