Medicine For People!

August 2010

Colon Cancer Screening

"It's cancer, her colon, her liver, everywhere. All we can do is relieve the obstruction and close her up." Three a.m. at the Oakland Kaiser Hospital operating room, 1969, and all the surgical equipment in the suite suddenly seems so useless. Our anesthetized patient had noticed a bit of blood in her stool months ago but had attributed it to a hemorroid. In the months that followed, she experienced increasing pain. Finally it had become too much and she'd come to the ER earlier that night. Evaluation indicated a bowel obstruction. We were hoping for a twisted bowel, but this was worse than we'd expected.

It is this type of experience that stimulates our current practice of nationwide screening for colon cancer. Colon cancer kills people and early detection can save some of those lives. This we know to be true. But is mass colonoscopy screening the best way to find the cancers and save the lives?

Conventional Wisdom about Colonoscopy

This month we begin a three-part series questioning the conventional thinking about colonoscopy, the generally accepted idea that, starting at age 50, every five or ten years we should load up on laxatives, write a $1500 to $2000 check, then let someone run a tube where the sun never shines. Certainly we need to perform colonoscopy when symptoms or other indications call for it. But when it comes to screening healthy people for colon cancer, there is a less expensive, safer, similarly effective way. I will discuss the better solution in depth in the third part of this series.

Why Screen for Colon Cancer?

Ask the experts, and you'll hear that colon cancer is the second leading cause of death from cancer in the United States. With care and attention we can detect it early and prevent many of those deaths1. For every 1000 men and women in the middle of life, about five will die of colon cancer in the next ten years.2 Each one who dies will lose, on average, 13 years of life.3 All that is correct. Let's open the window a little wider, however, and get the bigger picture.

Colon Cancer: The Degree of the Threat

The chart4 below shows deaths, per 1000 men who never smoked, over a ten year period beginning at age 50. Death rates for smokers vary by cause but are in the range of two to three times greater. Empty cells indicate less than one death per 1000. There are many other causes of death besides those listed, so "all causes combined" is much greater than the sum of the numbers in each row.

Age

Vascular Disease

Cancer

Infection

Lung disease

Accidents

All causes combined


Heart attack

Stroke

Lung

Colon

Prostate





50

11

1

1

2

1

2


5

49

55

19

3

1

3

2

2

1

5

74

60

32

5

2

5

3

3

1

5

115

65

52

9

4

8

6

3

3

6

176

70

87

18

6

10

12

6

5

7

291

75

137

32

8

13

19

12

6

11

449

Deaths, per 1000 women who never smoked, over a ten year period beginning at age

Age

Vascular Disease

Cancer

Infection

Lung disease

Accidents

All causes combined


Heart attack

Stroke

Lung

Breast

Colon

Ovary

Cervix





50

4

1

1

4

1

1




2

37

55

8

2

2

6

2

2

1

1

1

2

55

60

14

4

3

7

3

3

1


2

2

84

65

25

7

5

8

5

4

1

2

3

3

131

70

46

14

7

9

7

4

1

4

5

4

207

75

86

30

7

10

10

5

1

8

6

7

335

Source. (At this site you can find similar tables for smokers.)

You can see that although colon cancer ranks high among cancer deaths, our greatest threats lay elsewhere. So while the American Cancer Society reports that about 50,000 people died of colon cancer last year, you need a table such as the above to put that number in context.

All About Polyps

Just as our skin can grow barnacles and other nubbins as we age, benign growths, called polyps, occur in the colon (which we might consider to be an interior skin) in over 20 percent of people in the over-fifty age group5 and in about 50 percent of people at age 70.6 Like moles on our exterior skin, some will become cancerous as the years pass but most do not.

Colon Polyps

Source.

Tests on apparently healthy people to look for cancer, such as mammography, are called screening tests. Most screening tests do not also treat or prevent cancer. Colonoscopy, uniquely, not only identifies otherwise hidden early cancer but, by removing the polyps or early tumors, also treats and prevents it. That is why many physicians recommend this test.

Non-Polyp Cancers

A closer look at the natural history of colon polyps, however, casts doubt upon this recommendation. In Taiwan, doctors scoped almost 14,000 people of whom about 2600 had polyps and about 750, colon cancer. They used statistical modeling to estimate that many colon cancers, perhaps a third, arise "de novo," not from a polyp but right from the colonic wall.7 Standard colonoscopy does a poor job of identifying such tumors. They can easily progress from apparently normal colon lining to invasive cancer between the standard ten-year colonoscopy screenings.

Polyps can Disappear or Become Cancer

When we consider cancers that arise from the standard polyp-adenoma-cancer sequence, these progress at a rate generally slow enough that colonoscopy often becomes an unnecessarily invasive method of surveillance. The chart below details the microscopic structure of polyps and the sequence through which they change.

Type of Polyp

Description

Size range

What it can do

Normal colon lining

Healthy tissue

n/a

Stay healthy and function well.

Adenoma

Benign growth of glandular tissue. The first stage of an adenoma is a tubular adenoma, named for the cell's arrangement under the microscope, a normal tubular arrangement.

From barely visible to something under 2/5 of an inch in diameter.

Often tubular adenomas remain stable in size8 or go away on their own9. Repeat colonoscopies on such polyps found that only one of 189 lesions exceeded the 2/5 of an inch threshold after 1 year. In another study, at the 3-year mark most such polyps remained stable or had regressed in size, and there was an overall tendency to net regression among the medium-sized (1/5 to 2/5 of an inch) polyps.10 Sometimes, usually after quite a few years, an adenoma will become a tubulo-villous adenoma. Fewer than 0.1% of adenomas under 2/5 of an inch in diameter contain cancer11.

People who have an adenoma and refuse to have it removed developed colon cancer at a rate of about 4 percent after 5 years and 14 percent after 10 years.12

A tubulo-villous adenoma

While not a cancer, this begins to show disordered growth, arrangements of cells not normally found in the body.

Generally in the range of 2/5 of an inch diameter, give or take, or larger.

These can progress to

villous adenoma. When polyps show up on CT scans, and we can only see the size of the polyp, the cumulative risk of cancer for large polyps at 5 and 10 years is less than 3 percent and 10 percent, respectively.13

Villous adenoma

Have no remaining characteristics of normal colon lining, but not yet a cancer.

Usually over 2/5 of an inch diameter

Can progress to cancer. See next table for rates of advancement.

Colon Cancer

A malignant growth than remains localized for some time, then spreads through the colon wall and then to other organs. A localized cancer can be removed with a high probability of cure. Once it has spread, the chances of death increase.

Usually over 2/5 of an inch diameter

"Although nearly all proven cancers in their study showed a more rapid growth rate compared with benign lesions, the estimated doubling times were still relatively long, often approximately 1,000 days. Even the fastest growing cancer required 100 days to increase its diameter by only 2.5 mm (1/10 inch), whereas the slower growing cancers might have required well over half the maximum human life span to attain a diameter of 6 cm (2.4 inches),"14

How Fast Do Polyps Grow

The Taiwanese group's statistics in the table below show that the number of years required for a typical normal colon lining to progress to a tubular adenoma, a common polyp, was about 450 years. Since none of us live that long, what this means is that in about seven colons, belonging to seven people, one of those colons will develop a tubular adenoma during its lifetime (and usually in the latter third of that lifetime.) This percentage is in general agreement with other studies.15

If we look just at the people with a tubular adenoma, it takes, on the average, about 43 years for one of those to progress to a tubulovillous adenoma. If every single one took 43 years, no worries, we'd die of something else first. Except, some never progress, and some do, and in much less than 43 years. For those that reach the tubulovillous stage, another 20 years on average are required to get to the villous stage, which if they reach, require nine years on average to become an invasive cancer. We might be tempted to add up these intervals and guess that we will all die of something else before the evolution of a colon cancer, remember these are average times. Some progress more quickly, others more slowly or not at all. A Norwegian study16 estimated that tubulovillous adenomas advanced to cancer at a rate of about 17 percent per year, which is consonant with the Taiwanese estimate of nine years progression time on the average.

Years, on average, required for lesion to progress

Average: varies greatly between individuals

By size of lesion

All estimates below in years

Normal-diminutive adenoma

476 years

Diminutive-small adenoma

48

Small-large adenoma

18

Large adenoma-invasive CRC

16



By tissue diagnosis

9

Normal-tubular adenoma

455

Tubular-tubulovillous

43

Tubulovillous-villous

20

Villous adenoma-invasive CRC

9

Data in table from Br J Cancer 2003 Jun 16;88(12):1866-73

As confirmation of this analysis, note that on repeat colonoscopies, often done within 3 to 5 years when adenomas are found, cancer is no more often found than it is in a normal population17.

Polyps and Colon Cancer: Executive Summary

As far as we can tell, colon cancer can arise straight from the colon wall, the so-called de novo cancers, or from a common polyp, through a series of evolutions marked by advance or retreat. In this series of newsletters, I will argue that colonoscopy screening once every ten years fails to find many de novo cancers and is not an optimal strategy for colon cancer arising from polyps.

Coming Next Month

In next month's newsletter I will take a closer look at colonoscopies and examine what they can and cannot do for you.

Endnotes

1 N Engl J Med 2009;361:1179-87.

2 Gilbert Welch, MD Should I be tested for cancer? pg 166

3 Gilbert Welch, MD Should I be tested for cancer? pg 191

4 This table and the following table regarding women has been adapted from "Know your chances: Understanding Health Statistics" by Steven Woloshin, Lisa Schwartz, and Gilbert Welch, MD, page 128-9.

5 n engl j med 362;19 nejm.org may 13, 2010

6 Lecture by OS Lin, MD, of the Virginia-Mason Clinic

7 Br J Cancer 2003 Jun 16;88(12):1866-73 Early detection of flat- or depressed type cancer arising from de novo carcinoma is difficult with endoscopy. While estimates vary widely, a reasonable working fraction is that about 30% of colon cancers arisie without going through the adenomatous phase. While dye techniques can facilitate their identification on colonoscopy, such techniques are not commonly used.

8 Welin S, Youker J, Spratt JS Jr. The rates and patterns of growth of 375 tumors of the large intestine and rectum observed serially by double-contrast enema study (Malmö Technique). Am J Roentgenol Radium Ther Nucl Med 1963; 90:673–687

9 Int. J. Cancer: 111, 633–639 (2004)

NATIONAL POLYP STUDY DATA: EVIDENCE FOR REGRESSION OF ADENOMAS

The data of the National Polyp Study, a large longitudinal study on surveillance of adenoma patients, is used for testing assumptions on the adenoma-carcinoma sequence. The observed adenoma and colorectal cancer incidence in the National Polyp Study were compared with the simulated outcomes of the MISCAN-COLON model of epidemiology and control of colorectal cancer for the U.S. population based on expert opinion. Variants of this model were explored in order to identify assumptions on the adenoma-carcinoma sequence that are consistent with the study observations. The high observed adenoma detection rates at surveillance and low observed colorectal cancer incidence in the National Polyp Study could only be explained by assuming a high incidence rate of adenomas accompanied by regression of adenomas. The National Polyp Study data suggest that adenoma prevalence results from a dynamic process of both formation as well as regression of adenomas. This lowers the expectations for the effects of colorectal cancer screening strategies that focus on adenoma detection.

10 AJR Am J Roentgenol 2009 Jul;193(1):40-6

11 AJR Am J Roentgenol 2009 Jul;193(1):40-6

12 Med Clin N Am 89 (2005) 1–42

13 Colorectal cancer screening with CT colonography: key concepts regarding polyp prevalence, size, histology, morphology, and natural history.

AB - OBJECTIVE: The purpose of this article is to provide a timely update on a variety of key polyp topics to construct a proper framework for physicians who are interested in providing CT colonography screening as a clinical service. CONCLUSION: As the medical community considers the expansion of CT colonography for screening, we believe it is prudent to update and review several key concepts regarding colorectal polyps. In particular, it is important to replace the older literature derived from high-risk and symptomatic cohorts with the wealth of newer and more applicable data from average-risk and asymptomatic screening cohorts. Familiarity with current concepts regarding flat (nonpolypoid) lesions and the natural history of small colorectal polyps is also vital to the effective application of this technique.

AJR Am J Roentgenol 2009 Jul;193(1):40-6

14 AJR Am J Roentgenol 2009 Jul;193(1):40-6

15 This study did not correct for the presence of de novo carcinogenesis, as a result of which their progression estimates are more rapid than they would otherwise be.

Deutsches Ärzteblatt International - Dtsch Arztebl Int 2008; 105(24): 434–40 colonoscopy rate quality.pdf

This 24% differs from 16% (476 years to progress to adenoma divided by lifespan of 76 years) but given the varying nature of human populations and statistical and observational techniques, this variance strikes me as within the range of expectation.

16 If we don't undergo colonoscopy, and those pre-cancerous polyps are present, even the worst type, the tubulovillous adenomas, advance to cancer at a rate of about 17 percent per year

Risk of colorectal cancer in adenoma-bearing individuals within a defined population.

Eide TJ.

Int J Cancer. 1986 Aug 15;38(2):173-6. 

Assuming that all colorectal cancers develop from preexisting adenomas, the annual conversion rate, defined as the number of cancers occurring each year as a percentage of all adenoma-bearing individuals, was determined. The number of adenoma-bearing individuals in the living population of northern Norway was estimated in each cohort for the period 1974-76 by determining the prevalence of colorectal adenomas in an autopsy population of 271 consecutive cases, representative of the population of the area with regard to underlying causes of death. During the 10-year period 1974-1983 a total of 656 colorectal cancers were recorded among an estimated number of 26,419 adenoma-bearing individuals aged over 35 years. The annual conversion rate was found to be 0.25%, indicating that an average adenoma-bearing individual is only at a moderate risk of colorectal cancer. The annual conversion rates for individuals having large adenomas, or adenomas with villous structures, or severe dysplasia were roughly estimated to be 3%, 17% and 37% respectively, assuming that colorectal cancer develops from one of these sub-groups of adenomas only.

Eide TJ: Risk of colorectal cancer in adenoma-bearing individuals within a defined population. Int J Cancer 1986; 38: 173–6.

17 "in patients undergoing adenoma surveillance, there was an increased risk of being diagnosed with an adenoma, but the prevalence of advanced neoplasia was similar to that seen in the screening population.

The annual adenoma-to-carcinoma conversion rate was 3% for large adenomas, 17% for adenomas with villous histology, and 37% for adenomas with high-grade IEN (8).

Deutsches Ärzteblatt International - Dtsch Arztebl Int 2008; 105(24): 434–40 colonoscopy rate quality.pdf

story: 

Medicine for People! is published by Douwe Rienstra, MD at Port Townsend, Washington.