Some people are concerned that the new recommendations (see sidebars) are designed to ration care and reduce health costs. As a volunteer member of the Canadian panel, who also attended U.S. panel meetings, I can assure you that this is not the case. Panel members are not paid. Their focus is on how these tests can best help people.
So why is there such a split of ideas? One major reason is that people diagnosed and treated early believe that they have been cured. They are grateful to the test and the doctors who treated them. However, the situation is actually more complicated.
To understand this complex question, it helps to understand a little about cancer and how it may grow.
Not all cancers are the same. Any given type of cancer has a range of growth patterns. Some grow very fast, invade rapidly, spread through the body and kill within a short time. Others grow so slowly that they might cause death by the time a person reaches extreme old age (well over 100). Some grow at intermediate speeds.
As well, some cancers can be treated using current methods, while others cannot. Some are treatable even if discovered late. Others will kill even if found really early. Most cancers occur late in life, and many of these grow slowly. However, a few start in young people, and sadly more of these grow rapidly.
For instance, consider cancer of the cervix. This cancer is found in women, usually after age 40. It is highly treatable when caught early. A Pap smear test scrapes cells directly from the top layer of skin on the cervix, where cancer changes occur. Using this test to find really early changes can lower the death rate by 70 per cent or more.
While this cancer is rare among young women, it tends to be worse for them and more difficult to cure. Even though the Pap smear is a good early test, some women still die from the disease. Many early changes (pre-cancers) are found before they develop into cancer. Left alone, these heal themselves in many women. However, all women are treated to make sure that any cancers are prevented from developing. Perhaps three times more women with early pre-cancer changes are treated than would get cancer, to be sure that none do. Some get complications from the treatment, such as infection or an incompetent cervix, which can lead to pregnancy loss later on.
Balancing the harm and benefit of early testing and treatment is always necessary. With cervical cancer, testing is done in the hope that preventing invasive cancer outweighs the risk of harm, such as miscarriage or premature birth. If changes are not found at the pre-cancer stage, only about 1.5 per cent of women would get cancer of the cervix. If we did not use pap smears, maybe one per cent of women would die from it. Pap smears are offered to all women between 21 and 65, even though over 48 out of 50 women will never benefit from the test.
Most other types of cancer are more difficult to detect early and to treat.
An issue to consider is the potential for error. Cancer at an advanced stage is easy to diagnose. With early cancers, in contrast, cells under the microscope or shadows on an X-ray barely differ from normal ones. It is very hard to decide whether the cells are abnormal or whether they are just a variation of normal. There are many more samples with slightly abnormal cells than severely abnormal ones.
Pathologists are doctors who look at body tissues and cells to check for disease. They have very difficult decisions to make about whether to call a specimen cancer or not. They know lives depend on their analysis. Pathologists often call in other doctors to assist with difficult decisions, but even with agreement, they may still be wrong. Radiologists who read X-rays face similar situations.
Interestingly, 'abnormal' may differ from country to country. British radiologists usually call about five per cent of mammograms abnormal, while in North America, up to 10 per cent may be called abnormal. A tiny percentage of the extra women that North American radiologists call abnormal have cancer, and would be 'missed' by British radiologists.
Early detection of breast cancer is difficult. The most effective method is mammography, an X-ray of the breast. This test often shows early changes, such as small clumps of calcium, which can be seen by expert radiologists. A biopsy, usually done by inserting a fine needle into the area of change and sucking out cells, may be the next step. The cells are studied under a microscope. If possible cancer cells are seen, surgery may be needed to take out the lump. Cells from the lump will be checked under the microscope. If the pathologists see what they think are cancer cells, further treatment is done. This could involve major surgery, radiotherapy or chemotherapy.
As X-ray is an indirect method of testing, breast cancer is more difficult to find than cervical cancer. Generally, the cancer must be larger to be seen. Even so, for women between 50 and 70 years, screening by mammography reduces death rates by 15 to 20 per cent, or 33 per cent in the highest estimates. Mammography does not help most women who get breast cancer. Some studies show a smaller benefit among women aged 40 to 49, but this is disputed. It is not clear whether there is any value in screening at this age. Even worse, far more women are diagnosed as having some abnormality than would ever get cancer.
Table 1 shows the results of a study from Australia, comparing women who were screened with those who were not. For each age group, the table shows what happens over ten years when 1000 women are screened. At each age, a large proportion of the women had a positive mammogram. This means that some abnormality was noticed. Further tests were done for those women, and a few were diagnosed as having cancer. With women who were not screened, fewer were diagnosed as having cancer. So, screening created a large amount of over-diagnosis, meaning diagnosis of cancer in women who would never develop invasive cancer if left alone. The trouble is that all these women are given the full treatment – surgery, radiotherapy and chemotherapy. Many women still die in spite of the screening. Still, there are fewer deaths among screened women, so some lives are 'saved' by this process.
Table 1: Outcomes of Mammography from Australian study
What proportion of breast cancer deaths can be reduced by mammogram screening? The best estimate is 30 per cent. This table shows ten-year biennial results per 1000 women.
|Diagnosis of cancer||21||33||38||41|
|Number needed to save one life||2000||500||333||400|
At each age, there is a trade-off between the possible benefit in lives saved, against the harm to those told they have cancer and treated for it unnecessarily. Here is the dilemma — most women who are 'survivors' of breast cancer had that diagnosis because their cells looked like cancer under the microscope. However, they would never have developed disease that would grow, spread and cause death.
Women have to decide whether this trade-off is a good choice for them, and at what age they would like to start mammography screening. This is a difficult decision with no right answer.
The situation with prostate cancer is similar. Although a test called the prostate specific antigen (PSA) is available, it is generally recognized as a poor test. It finds many false positives, but misses many men with prostate cancer. The follow-up test is prostate biopsy. This is a series of tissue samples taken from the prostate gland with a needle. Samples are examined under the microscope. Deciding whether prostate biopsy samples contain cancer is also difficult, since cancers are common. Over 40 per cent of men aged 60 have them, but most are very slow growing, and would never kill the man before he dies of something else.
Draft Recommendation: Do not screen for prostate cancer, with either PSA (Prostate Specific Antigen) testing or rectal examination.
Review of the trials performed to test the value of screening show that PSA screening results in small or no reduction in deaths from prostate cancer, and can cause harm, including post-operative death (0.5 per cent), cardiovascular events (0.6 per cent to 3 per cent) and one in five who have a surgical prostatectomy will have continuing incontinence of urine.
Canadian Cancer Society
Prostate cancer can be detected early using a PSA test and a digital rectal exam. However, research does not clearly show if the benefits of testing for prostate cancer outweigh the harms. We recommend that you talk to your doctor about your risk for prostate cancer and about the benefits and harms of early detection.
Read more: tinyurl.com/7sdebqh
American Cancer Society
Research has not yet proven that the potential benefits of testing outweigh
the harms of testing and treatment.
The American Cancer Society believes that men should not be tested without learning more about what we know, and don't know, about the risks and possible benefits of testing and treatment.
UK: NHS Screening Committee
The National Screening Committee has recommended that a national screening program not be introduced at this time.
The Prostate Cancer risk management program was introduced so that men who ask about a PSA test can make an informed choice, based on quality information about the advantages and disadvantages of testing.
Cancer Council Australia Position Statement (May 2010)
Current evidence indicates that the PSA test is not suitable for population screening, as the harms outweigh the benefits.
Canadian Task Force on Preventive Health Care, CTFPHC Recommendation for Screening for Breast Cancer with Mammography. Retrieved January 24th, 2012. (See www.canadiantaskforce.ca)
A few organizations, mostly in North America, strongly urge men to start screening at age 40. However, the majority of cancer expert organizations around the world remain doubtful.
Several trials have now been done to test this screening process. Each has its strengths and weaknesses. The best one was done at several different centres in Europe (the Economic and Social Research Council or ESRC trial). It showed that for men between 55 and 75 years, doing one test every three to four years reduced prostate cancer deaths by 20 per cent. That sounds good, but 10,000 men screened every three to four years will have 34 fewer deaths after 14 years from prostate cancer, but no difference in total deaths. In other words, they will die of something else.
Table 2 compares what happened among 10,000 men screened in the ERSC trial, compared to the same number who were not. The number of men harmed is shown in the net benefit column as negative numbers. This does not include harm to men treated with either chemotherapy or hormone therapy, which is likely in North America as treatment is more aggressive here than in Europe.
Table 2: Effects of Prostate Screening from Canadian Partnership against Cancer
Estimate of effects experienced by 10,000 men aged 55–69 years, each screened every 4 years, compared with those not screened. All were followed for 9 years.
|Screened||Not Screened||Net Effect||Net Benefit||Net Harm|
|Number invited for screening/not screened||10,000||10,000|
|Number Number of positive PSA results (> 3 ng/ml) 1of positive PSA results (> 3 ng/ml) 1||1,620|
|Number of biopsies 1||1,393|
|Number of cancers detected 1||820||480||+ 340|
|Number of potentially aggressive cancers (Gleason score > 7) 1||228||217||11|
|Number of low-grade cancers
(≤ Gleason score 6) 1
|Number undergoing radical prostatectomy 2||220||100|
|Number undergoing radiotherapy 2||227||123||+ 120|
|Complications of therapy 3||+ 104|
|Number of deaths due to prostate cancer||29||36||7|
Trials that screened more often had no effect. This may be because they cause more harm because of more false positive results, and this outweighs the small benefit. In spite of the benefit being limited to ages 55 to 75, in a trial that only tested every three to four years, enthusiasts for screening recommend starting screening from age 40, and testing annually from age 50!
Unfortunately, much of the publicity about screening comes from the United States. Big business drives many of those recommendations. Many American doctors who specialize in issues of the urinary system make more money from selling drug treatment and radiotherapy than from doing their main job of advising and treating patients. Even the PSA test makes large profits for its maker and for medical laboratories. No wonder it is advertised. Moreover, the makers and marketers of adult diapers likely look forward to more prostate treatment, since that brings more business from men who have trouble holding their urine.
Personally, as a male, I have no intention of letting anyone do prostate screening on me until I have problems that need investigation. Dr. Otis Brawley, Chief Medical Officer of the American Cancer Society, is African American (black) and so has a higher risk than a Caucasian (white) man. He also advises against such testing.
All screening tests provide a balance between a small possibility of benefit, and some chance of harm. When the balance is strongly in favour, it is appropriate for doctors to recommend such tests. When the balance is uncertain, they should be more cautious in their recommendations. But even then, it is ultimately you who must decide what is right for you, with doctors helping you to interpret the information.