First, let me be clear about the definition of cancer screening. Screening for cancer is done for people who feel well and have no symptoms that suggest they might have cancer.The goal is to detect cancer early, before you have symptoms, so that hopefully it is easier to treat and cure.
If you have symptoms of cancer, then tests such as mammograms or colonoscopy are for diagnosis of a disease, not screening. If you are a woman who has a breast lump or abnormal uterine bleeding, or a man with a lump in the testicle, or anyone who has rectal bleeding or difficulty swallowing food, or feeling full after just a little food, you need to see your doctor right away and be tested for cancer. Here is a link to a web page from UCSF that has a more comprehensive look at symptoms of cancer for which you should see your doctor: 17 Cancer Symptoms You Shouldn’t Ignore.
In this post I am going to write about the available screening tests for cancer. For many cancers there are no good screening tests. A good screening test has to meet several criteria. The first is that it detects a kind of cancer that responds better to treatment when detected early. There are other criteria for a good screening test as well. Here is a link to a web page from the American Medical Society Journal of Ethics that talks about all the criteria for a good screening test: What Makes a Screening Exam “Good”?
I will discuss how well each of them work and how well they fit the criteria for a good screening test. I will also write about the risks of getting cancer screening tests (there are some substantial risks).
Cervical Cancer Screening (and Prevention)
Screening for cervical cancer is the poster child for an effective screening test. It is inexpensive, has no significant risk and detects both cervical cancer and pre-cancerous changes. It develops slowly, so it is an ideal cancer for screening. Early detection leads to much more effective treatment even before cancer develops. Since the 1940’s, when the pap smear was invented, death from cervical cancer has decreased by over 70%.
We have learned a lot about cervical cancer since the 1940’s. We now know that cervical cancer is caused by certain strains of the wart virus, otherwise known as HPV (human papilloma virus). The recommendation used to be that women should have a pap smear every year starting after beginning sexual intercourse. We now know that women under 21 most often clear HPV on their own, and there is no reason to do pap smears before age 21. If the pap smear is normal and the HPV test is negative, then for women 30-65 years of age, it is not necessary to do another one for five years. Women over 65 who have had two previous normal pap smears can stop cervical cancer screening.
Not only can the pap smear and HPV test detect cervical cancer early, but they can detect pre-cancerous changes that can be treated before cervical cancer develops. Even better, there is now a vaccine that prevents HPV and therefore cervical cancer altogether. It is called Gardisil, and should be given to adolescent girls and boys before the age they start having sexual intercourse.
Like other screening tests, most women will not benefit from cervical cancer screening. We would have to do pap smear and HPV screening on 1,140 women for ten years to prevent one death from cervical cancer. That means that 1,140 out of every 1,141 women will not benefit from cervical cancer screening,
Breast Cancer Screening
Mammograms are the primary and most studied screening test for breast cancer. Breast cancer is not as ideal for screening as cervical cancer, because some breast cancers develop very rapidly and can spread aggressively between mammography screenings. That means that some women die from breast cancer even if they get regular mammograms. Mammograms also have a high false positive rate, which can lead to more invasive unnecessary tests such as breast biopsies. The false positive rate for any kind of mammogram, including the newer 3D mammograms is 50% over 10 years!
Mammograms do save lives, but a lot fewer than people think. Here is a table from the US Preventive Services Task Force website that shows the benefits and harms of mammograms every two years for every 1000 women screened.
| Variable | Ages 40–74 y | Ages 50–74 y |
|---|---|---|
| Lives Saved | 8 (5–10) | 7 (4–9) |
| False-positive tests | 1529 (1100–1976) | 953 (830–1325) |
| Unnecessary breast biopsies | 213 (153–276) | 146 (121–205) |
| Overdiagnosed breast tumors | 21 (12–38) | 19 (11–34) |
There are a few things to notice about this table. First, almost all the benefit in lives saved by screening is in the 50-74 age group but more false positives, unnecessary biopsies and overdiagnosis occurs in the group that starts mammography at age 40. The other thing to notice is the number of overdiagnosed breast cancers (overdiagnosed means that these tumors would never result in illness or death from the cancer) are much larger than the number of lives saved in each age category.
What this table says is that 125 women need to be screened with mammography to save one life from breast cancer. That is a great number! On the other hand, for every 47 women screened by mammography one woman gets overdiagnosed with breast cancer that would never harm her. That is a terrible number!
Given all this information, what should women do? Here are the US Preventive Services Task Force recommendations as of 2016. These recommendations are currently being updated, but the new recommendations are not available yet. The USPSTF recommends mammograms every other year, rather than every year. Every other year mammograms cut the risk of false positives in half, but result in the same number of lives saved.
| Population | Recommendation | Grade |
|---|---|---|
| Women aged 50 to 74 years | The USPSTF recommends biennial screening mammography for women aged 50 to 74 years. | B |
| Women aged 40 to 49 years | The decision to start screening mammography in women prior to age 50 years should be an individual one. Women who place a higher value on the potential benefit than the potential harms may choose to begin biennial screening between the ages of 40 and 49 years. . For women who are at average risk for breast cancer, most of the benefit of mammography results from biennial screening during ages 50 to 74 years. Of all of the age groups, women aged 60 to 69 years are most likely to avoid breast cancer death through mammography screening. While screening mammography in women aged 40 to 49 years may reduce the risk for breast cancer death, the number of deaths averted is smaller than that in older women and the number of false-positive results and unnecessary biopsies is larger. The balance of benefits and harms is likely to improve as women move from their early to late 40s. . In addition to false-positive results and unnecessary biopsies, all women undergoing regular screening mammography are at risk for the diagnosis and treatment of noninvasive and invasive breast cancer that would otherwise not have become a threat to their health, or even apparent, during their lifetime (known as “overdiagnosis”). Beginning mammography screening at a younger age and screening more frequently may increase the risk for overdiagnosis and subsequent overtreatment. . Women with a parent, sibling, or child with breast cancer are at higher risk for breast cancer and thus may benefit more than average-risk women from beginning screening in their 40s. | C |
Other groups recommend more intensive screening, but screening younger women and screening every year results in less than half a percent more lives saved and markedly increases the harms of screening including false positives and overdiagnosis. My feeling is that the USPTF recommendations make the most sense for women.
Prostate Cancer Screening
Screening for prostate cancer with the PSA (prostate specific antigen) is the most controversial of the cancer screening tests. Overdiagnosis with the PSA test is a huge problem. Autopsy studies done on men who died of other causes find that by age 80 more than half the autopsies show prostate cancer. Many, perhaps most prostate cancers grow so slowly that they will never cause any symptoms. More men die with prostate cancer than from prostate cancer. Overdiagnosis leads to radical prostatectomy for tumors that would never cause symptoms or death. Surgery for overdiagnosed tumors results in permanent urinary incontinence for half of men and sexual dysfunction for most of them. On the other hand, prostate cancer kills one out of every 41 men in the US. A good screening test for aggressive prostate cancer would be a wonderful thing. Unfortunately at this point we do not have such a test.
The US Preventive Services Task Force reviews all the available studies of screening tests and makes recommendations based on that evidence. Here is that review for the PSA test if 1000 men are screened with the PSA test annually for13 years,
| Number of Men Affected | |
|---|---|
| Men invited to screening | 1000 |
| Men who received at least 1 positive PSA test result | 240 |
| Men who have undergone 1 or more transrectal prostate biopsies | 220b |
| Men hospitalized for a biopsy complication | 2 |
| Men diagnosed with prostate cancer | 100 |
| Men who initially received active treatment with radical prostatectomy or radiation therapy | 65 |
| Men who initially received active surveillance | 30 |
| Men who initially received active surveillance who went on to receive active treatment with radical prostatectomy or radiation therapy | 15 |
| Men with sexual dysfunction who received initial or deferred treatment | 50 |
| Men with urinary incontinence who received initial or deferred treatment | 15 |
| Men who avoided metastatic prostate cancer | 3 |
| Men who died of causes other than prostate cancer | 200 |
| Men who died of prostate cancer despite screening, diagnosis, and treatment | 5 |
| Men who avoided dying of prostate cancer | 1.3 |
What this table shows is that of 1000 men screened for 13 years, there will be 100 men diagnosed with prostate cancer, but only 1 life saved as a result of screening and 2 other men who avoid metastatic prostate cancer. The cost of that life saved will be 97 men overdiagnosed with prostate cancer and 220 men that have a prostate biopsy. This is not a pleasant procedure and can sometimes result in complications such as infection, as indicated by the 2 men out of the 220 men that were biopsied that were hospitalized because of the biopsy. Five men out of the 1000 died from prostate cancer anyway despite screening, diagnosis and treatment,
African American men and men with a strong family history of aggressive prostate cancer are at substantially higher risk for aggressive prostate cancer. It is worth considering screening for this group. Men at average risk are more at risk for harm than help from prostate cancer screening, at least with the tests we have available now. The USPTF recommendation is that prostate cancer screening is an individual decision and should be discussed with your doctor.
Colorectal Cancer Screening
Colorectal cancer if caught early is almost 100% curable, but if it is not discovered until it invades the intestinal wall, the cure rate gets progressively worse.
Unlike breast cancer and prostate cancer, which have only one kind of screening test, there are a number of different screening tests for colon cancer. They each have advantages and disadvantages.
FOBT (Fecal Occult Blood Test)
This is the oldest test and also the one that has been studied the most. It is also the least expensive ($5-$10). People who do this test every year, and who have a colonoscopy if they have a positive test have reduced deaths from colorectal cancer by about 27%. That means that out of 1000 people who screen annually with FOBT, there will be 270 fewer deaths from colorectal cancer than people who do not do screening.
The biggest problem with this test is that it has a fairly low sensitivity of 50%, which means it will miss half of early colorectal cancers. It also has a fairly low specificity of 78%, which means that about 1/4 of the tests will be false positives.
FIT (Fecal Immunochemical Test)
This newer test is only positive for blood coming from the colon. That makes both the sensitivity and the specificity higher than FOBT. It is also fairly inexpensive (the home test costs about $25)
The sensitivity of the FIT test is about 75%, which means it will miss about 25% of early colorectal cancers. The specificity of the FIT test is about 90%, which means that the false positive rate is only 10%. Because the FIT test is relatively new, there are no randomized controlled trials of lives saved by FIT tests vs no screening. Estimates based on computer models suggest that annual FIT testing with colonoscopy for positive tests would reduce deaths from colorectal cancer by 74%! That means that of 1000 people who screen with the FIT test and get colonoscopy for positive tests, deaths from colorectal cancer will be reduced from 1782 deaths to 457 deaths.
There are several different kinds of FIT tests. Some require sending a stool sample to a lab, but the home test is just as good and is less expensive. The one that seems to work the best is from Pinnacle Biolabs. Here is a link to their website where you can order a test: Second Generation FIT® 1 Pack.
Cologuard
Cologuard is the only commercially available DNA test. It measures the specific DNA shed by cancer cells. It is actually a combination test and includes a FIT test as well. This increases the sensitivity, but decreases the specificity, which means that the Cologuard test has a higher false positive rate than the FIT test alone. It is also very expensive, about $600 per test. A stool sample has to be collected and sent to the lab. Computer modeling suggests that deaths from colon cancer for 1000 people would be reduced from 1782 deaths to 1143 deaths. Not nearly as good as the FIT test alone.
Colonoscopy
Although some groups consider colonoscopy to be a screening test, many organizations feel that colonoscopy should be used only when another screening test is positive. People who are at high risk of colon cancer should probably have colonoscopy as a screening test. Colonoscopy done every ten years, with follow-up exams every 3 to 5 years when polyps were found, would reduce deaths from 1782 deaths to 624 deaths.
Colonoscopy is the most sensitive and specific test, but also carries much more risk as a screening test. Serious complications (bleeding or perforation) occur in 44 people out of every 10,000. 3 people out of 100,000 die from colonoscopy. The average number of colonoscopies to result in one serious complication is 225. It is also very expensive (average cost $1,700). It may seem counterintuitive that FIT testing saves more lives, but that is because it is easy to do FIT testing every year and screening colonoscopies are done only every 10 years. One advantage of colonoscopy is that it reduces the chance of getting colon cancer by identifying and removing pre-cancerous polyps.
Lung Cancer Screening
For a long time we had no good screening test for lung cancer. Annual chest x-rays were shown to be worthless for lung cancer screening. Now there is a screening test for lung cancer. It is a low radiation dose CT scan. It should be done every year, but only for people aged 50 to 80 years who have a 20 pack-year smoking history and currently smoke or have quit within the past 15 years.
False positives are a huge problem with lung cancer screening. Of people who get a positive result, 97% of those will be a false positive, meaning no cancer will be found on further testing. Further testing may include a needle biopsy of the lung, which can result in a collapsed lung.
The number needed to screen to prevent 1 lung cancer death is about 300. The number needed to harm (false positives or over diagnosis is 19. Although lung cancer screening with low dose CT does save lives, it comes at a substantial cost of false positives and overdiagnosis.
Common Cancers with No Good Screening Test
There are no good screening tests for ovarian cancer, uterine cancer, pancreatic cancer, esophageal cancer, leukemia, or Hodgkin’s disease (cancer of the lymph nodes. Many of these cancers are treatable once they are found, but they are found because of symptoms, not because of screening.
Galleri test
This is an update to this post, which I just published. Galleri is a new blood test that detects DNA shed by cancer cells. The company claims that it will detect 50 different types of cancer. Although this technology is promising, it is not yet ready for prime time. It is very good at detecting advanced cancer, but not very good at detecting stage 1 cancers when they are most treatable. There is also a high false positive rate of about 30%. Hopefully this technology will improve. If that happens, it will be a game changer with regard to cancer screening.
Bottom Line
If you are higher risk for any of the cancers for which we have screening tests, then getting screened is something you should do. If you are at average or low risk, screening tests are unlikely to help you and may harm you. Everyone should be alert for red flag symptoms that mean you might have cancer. The purpose of this post is not to discourage cancer screening, but to give you information that you should have when making a decision about whether and which cancer screening tests you choose.