Screening for cancer: Colon, lung, and skin cancers

When it comes to a disease like cancer, early diagnosis is particularly important, even urgent. As Harvard Medical School explains, there are many ways to facilitate early detection.

  • E-Mail
By  Harvard Medical School Published  November 8, 2005

|~||~||~|One of modern medicine’s proudest achievements is its ever-improving ability to detect disease early. The rationale is obvious: Early diagnosis leads to early treatment and a better outcome. When it comes to a disease like cancer, early diagnosis is particularly important, even urgent.

There are many ways to facilitate early detection. People should learn to recognise warning symptoms and call them to a doctor’s attention. Physicians should perform careful exams and follow up on any abnormalities they find. And testing should be used to root out disease even before it shows up on a doctor’s exam or causes any symptoms at all.

Screening tests are procedures performed on a routine basis when people feel well and neither patients nor their doctors have any specific cause for concern. Screening tests have been spectacularly successful in detecting cardiovascular risk factors. Routine blood pressure checks and cholesterol tests are the best examples.

Both can detect abnormalities long before they cause any symptoms, and both can lead to safe and effective lifestyle and medical treatments. Screening for high blood pressure and increased cholesterol levels gets much of the credit for the 56% decrease in deaths from heart disease and the 70% decline in stroke deaths that Americans have enjoyed over the past 50 years.

Cancer provides another instructive example from the United States, where it is the second leading cause of death. It takes more than 570,000 lives each year, making screening crucial In fact, the first medical screening test that produced major benefits was a cancer test: In the 1930s, Dr. George Papanicolaou introduced the simple screening test that bears his name.

As a result of the Pap test, deaths from cervical cancer have declined by more than 85% — and most of the 4100 American women who die from the disease each year could have been saved if they had had their recommended Pap smears.

The Pap test has set a high standard for cancer screening. Unfortunately, no other test can match it. But if screening mammograms are not as effective as Pap tests, most authorities still believe they are very beneficial.

At present, the only cancer screening procedures of proven value for men are the tests for colorectal cancer — and even here, experts debate the best use of the various tests. Still, the choice a man has to make about colon cancer screening seems easy compared with his decision about prostate cancer screening, the most controversial issue of all.

Screening the tests
Before a test becomes part of routine medical care, it should meet several standards. To be useful, a test should have a high sensitivity: It should be able to detect a large percentage of cases, with few false-negative results. It should also have a high specificity: It will not produce many false-positive results, diagnosing disease when none is present.

A good test should be reliable and reproducible, so a patient can expect the same results wherever he is tested. The test should be convenient and comfortable enough to gain wide acceptance and inexpensive enough to be affordable. Needless to say, tests that are performed on people who feel perfectly well should be extremely safe.

Above all, perhaps, good tests should lead to an effective treatment that will improve a person’s outcome. In a word, a test should do more good than harm.

Testing the tests
Early diagnosis is always scored as a success for the scientists who developed the tests, but it’s not necessarily a success for the patient. The measure of a test’s value is not its ability to find a disease but its ability to improve the quality and duration of life.

A randomised clinical trial is the best way to evaluate a test. Volunteers are randomly divided into two groups, one of which gets the test while the other does not. Because the assignment is by lot, the two groups will have an identical mix of risk factors, and the two should receive identical medical care and follow-up treatment during the trial. At the end of the study, the researchers evaluate the groups to see if the test has led to a better outcome.

Randomised clinical trials require large numbers of patients and many years of follow-up. As an example, scientists calculate that a study designed to detect a 25% decline in deaths from testicular cancer would require 6.5 million volunteers and take many years.

Randomised clinical trials are slow and expensive. They are also very complex, which is why there has been some debate about the various mammogram trials that have been completed to date. The only cancer screening test for men that has run the gauntlet of a randomised clinical trial is the fecal occult blood test for colon cancer. It passed — but other procedures, such as colonoscopy and sigmoidoscopy, which have not been evaluated as stringently, are likely to be substantially better.

In the real world, we don’t have the luxury of a full answer to many important medical questions. Lacking proof, we have to make decisions based on the best available evidence.

Should you be tested?
Cancer triggers so much anxiety, and early diagnosis is so logical, that many people choose to be tested even if proof of benefit is lacking. In many cases it’s a reasonable choice, but before a man rolls up his sleeve or pulls down his pants for a test, he should understand the potential drawbacks.

Tests may be frightening, time-consuming, inconvenient, or uncomfortable. These drawbacks are easy to understand. It’s also easy to realise that invasive tests may have side effects that can be serious. Less obvious is the fact that a test that’s safe and simple can lead to treatments that are not.

In some cases, the treatment is worse than the disease, in which case the test that starts the ball rolling ends up doing more harm than good. And testing is expensive, particularly if a false-positive test leads to additional procedures. In the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial, for example, 43% of participants received at least one false-positive result. On average, each person received about US$1100 in extra medical care in the year following the false alarm.

Touting the tests
“Who shall decide,” asked Alexander Pope, “when doctors disagree?” It’s a good question. In the case of cancer screening, smart, well-intentioned scientists can come up with different conclusions and recommendations.

In general, advocacy groups, such as the American Cancer Society, and associations of medical specialists who perform tests and treatments, such as the American Urological Association, are more aggressive about testing than academic panels, such as the U.S. Preventive Services Task Force, and government agencies, such as the National Cancer Institute. And the problem of competing guidelines has become more acute since commercial interests have entered the arena, offering body scans and other tests to the public through direct advertising, often very forcefully.

The patient makes the call
In the last analysis, each man must decide for himself. Doctors should be ready to answer a litany of questions associated with diffferent screening procedures: what a test will cost in term of time and discomfort, as well as money; what side effects might occur; and the patient’s risk for the particular cancer, since, in general, men at high risk tend to benefit from testing much more than men at low risk.

Although it can be hard to think about the consequences of a cancer diagnosis, patients are likely to ask what will happen if they find out they have the disease and if treatment is apt to change that result.

Colorectal cancer
About 29,000 American men die from colorectal cancer each year. That puts the disease far behind lung cancer and slightly behind prostate cancer on the hit list. Even so, screening for colorectal cancer is more important than screening for the other two because it really can save lives.

Testing can detect prostate cancer at an early stage, and new technology may do the same for lung cancer, but doctors don’t yet know if early diagnosis will translate into longer survival. Surprisingly, perhaps, only 48% of American men over 50 have had up-to-date colon cancer screening, whereas about 75% have been screened for prostate cancer at least once and 54% report regular screenings. Colon cancer is an ideal target for screening because it develops slowly in a predictable pattern and because early detection is highly effective but late treatment is not.

In nearly all cases, the first abnormality is a benign adenomatous polyp. About 99% of polyps remain harmless, but some become cancerous. Even then, the disease progresses predictably, spreading from the polyp into the wall of the colon, then into other tissues at a measured pace. If doctors remove polyps while they are benign, they can actually prevent colon cancer. And by removing cancers that are diagnosed early, doctors can cure more than 90% of patients, a figure that drops precipitously if the disease is diagnosed at a later stage.

The simplest is fecal occult blood testing (FOBT), which takes advantage of the fact that polyps and cancers are more likely to bleed than normal tissues. In most cases, the blood is too scant to be visible, but it can be detected by a sensitive chemical test. The theory is simple, but like most easy answers, it has flaws. Many polyps and cancers don’t bleed or bleed only intermittently, so FOBTs are often falsely negative (i.e., they have a low sensitivity).

On the other hand, benign abnormalities, ranging from gastritis to hemorrhoids, can bleed; as a result, FOBTs are often falsely positive (i.e., they have a low specificity). Despite the test’s low cost, wide availability, ease, and safety, the problem of false negatives and positives would seem to rule it out as a screening test but for one fact: It works. Five randomized clinical trials show that people who submit three stool specimens a year (or even every other year) and follow up a positive result with a colonoscopy are 15%–34% less likely to die from colon cancer than people who are not screened at all. That’s a high yield from an easy test. Remember, though, that a digital rectal exam with a single FOBT in a doctor’s office is not a useful test for colon cancer.

Sigmoidoscopy allows doctors to see polyps and cancers in the lower third of the colon. A patient prepares for the test by taking a laxative and/or enema at home and skipping breakfast. The doctor passes a fiberoptic tube through the rectum and advances it about 25 inches upward. If he spots a polyp, he can biopsy or remove it through the scope. The test takes about 15 minutes and is mildly uncomfortable.

A sigmoidoscopy is highly accurate as far as it goes, but it only goes a third of the way up the colon. A colonoscopy goes the distance, making it the best way to detect polyps and early colon cancers. Unfortunately, it will miss 5% of polyps. It’s also the most arduous test. Colonoscopy requires a much more vigorous cleansing preparation than sigmoidoscopy, and it also requires a sedative.

Although the preparation is unpleasant, sedation minimises any discomfort from the test itself. A colonoscopy takes 30–40 minutes, but it takes several hours for the sedation to wear off. For most people, the test will disrupt normal activities for the better part of a day. At present, colonoscopy is the “gold standard” in screening for colon cancer.

A double-contrast barium enema relies on x-rays to detect polyps and cancers. An enema is used to fill the colon partially with barium. Next, the colon is inflated with air and x-rays are obtained. The preparation is similar to that for a sigmoidoscopy, and the test takes about 30 minutes. Barium enemas cannot distinguish between polyps and cancers, so patients with abnormalities require colonoscopies. The test can miss small polyps and cancers.

Men at high risk may be candidates for genetic testing and require frequent colonoscopies, sometimes beginning at a young age. Men at moderate risk should choose a colonoscopy when they turn 50; even if the test is negative, it should be repeated every 5–10 years. Men at average risk can choose a colonoscopy every 10 years, a sigmoidoscopy every five years (preferably with annual FOBT), or annual FOBT; colonoscopy offers the best protection (but is the most arduous). A barium enema every 5–10 years is also an approved option but seems less desirable for the patient.

New tests for colon cancer are being developed. The most promising is a stool test for cancer genes, but it is at least a few years away from clinical use. Virtual colonoscopy, which uses CT scanning to visualise the colon, is also being studied; early results have varied, and most experts agree it’s not ready for prime time. More progress will follow, but it’s a mistake to put off colon cancer screening with the excuse that a better test is on the way.

Lung cancer
It’s a dreadful and highly lethal disease. And the real tragedy of lung cancer is that it is largely preventable. The key is to avoid exposure to tobacco in all its forms, including secondhand smoke; reducing radiation exposure, including residential radon, will also help.

In 1974, the American Cancer Society (ACS) endorsed lung cancer screening with annual chest x-rays. It was a well-intentioned and logical recommendation, and it was widely accepted by the medical community and the general public. But as the results of four large randomised trials were known, it became clear that annual chest x-rays don’t reduce deaths from lung cancer, and the ACS reversed itself in 1980.

Screening chest x-rays don’t save lives because by the time a lung cancer is large enough to show up, it is likely to have already spread, making surgical cure unlikely. But a new radiology procedure, low-dose high-resolution spiral computed tomography (spiral CT) can pick up cancers that are much smaller, when they may still be cured. The problem is that spiral CTs also detect many tiny nodules that are not malignant. As a result, patients are faced with the worry and expense of repeat spiral CTs every three or four months, or with the risks of surgery.

Spiral CTs are being promoted as an effective screening test for lung cancer. They may turn out to meet that goal, but they may not — and because an abnormal spiral CT often leads to lung surgery, they could even do more harm than good. The only way to find out is with randomised clinical trials, which are in progress.

Despite mounting publicity for testing, the ACS and other medical groups do not recommend screening for lung cancer, even in smokers.

The fear of cancer is understandable and legitimate, but turning fear into profit is reprehensible. Because spiral CTs can detect small nodules in the abdomen as well as the chest, whole-body scans to detect cancer are being marketed directly to the public in many places.

As a result, a 2004 survey found that 73% of respondents would choose a whole-body CT over US$1000 in cash. While preliminary evidence raises the possibility that spiral CTs may be able to detect lung cancer in smokers, there are no data to support whole-body scans, which may do more harm than good, both from false-positive results that lead to anxiety and unnecessary treatment and from radiation exposure. Whole-body scans are not recommended.

Skin cancer
Basal cell and squamous cell skin cancers are very common but rarely serious. Malignant melanomas are much less common but much more serious; about 33,500 cases are diagnosed in American men each year, and 4900 will be lethal.

Because melanomas are visible to the naked eye, screening involves little more than a careful look. Systematic screening has not been validated by randomised trials, but it’s logical, easy, safe, and inexpensive. Doctors can even counsel their patients on how to perform this screening test on their own by telling them all about the warning signs and encouraging them to establish a regular routine of self-examination.

This article is provided courtesy of Harvard Medical International.
© 2005 President and Fellows of Harvard College||**||

Add a Comment

Your display name This field is mandatory

Your e-mail address This field is mandatory (Your e-mail address won't be published)

Security code