Secondary Logo

Journal Logo

Much Discussion Sparked by Article Questioning Relook at What Is and Is Not Considered ‘Cancer’

Rosenthal, Eric T.

doi: 10.1097/01.COT.0000434535.95963.a5

A much-talked-about Viewpoint article, “Overdiagnosis and Overtreatment in Cancer: An Opportunity for Improvement,” published online ahead of print in the Journal of the American Medical Association (doi:10.1001/jama.2013.108415) has raised the provocative issue of how and when the term “cancer” should be applied.

Coauthors Laura J. Esserman, MD, MBA, Ian M. Thompson Jr., MD, and Brian Reid, MD, PhD, were part of a National Cancer Institute working group convened in March 2012 by Barnett Kramer, MD, MPH, to evaluate the problem of overdiagnosis and how it can lead to overtreatment. The group made the following recommendations to NCI:



  • Physicians, patients, and the general public should be helped to recognize that overdiagnosis—identification of indolent cancers—is common, especially in breast, lung, prostate, and thyroid cancers, and occurs more frequently with cancer screening;
  • Cancer terminology should be changed based on companion diagnostics: (1) The word “cancer” should be reserved for lesions with a reasonable likelihood of lethal progression if untreated; (2) the terms “cancer” and “neoplasia” should not be used for premalignant conditions; and (3) molecular diagnostic tools should be used to identify indolent or low-risk lesions, which should be reclassified as “IDLE”—indolent lesions of epithelial origins;
  • Observational registries should be created for low-malignant-potential lesions;
  • Overdiagnosis should be mitigated by reducing low-yield diagnostic evaluations appropriately, reducing the frequency of screening examinations, and focusing screening on high-risk populations; and
  • The concept of how to approach cancer progression should be expanded, with future research that includes controlling the environment in which precancerous and cancerous conditions arise, as an alternative to surgery.


Back to Top | Article Outline

‘Starting Point for Conversation’

The issues raised offer a starting point for conversation about the terminology for distinguishing between conditions that are life threatening and those that are not likely to progress to problematic stages; about how screening can lead to overdiagnosis and thus unnecessary treatment; and about how physicians talk to their patients.

Those interviewed for this article noted that eliminating the emotionally charged word “cancer” from certain diagnoses can also have several ripple effects—lessening anxiety but also perhaps decreasing vigilance and follow-up, as well as changing the statistics for cancer incidence and survivorship.

Kramer, Director of NCI's Division of Cancer Prevention, said during a telephone interview that the goal of the March 2012 meeting “was to match terminology with biological behavior,” so that lesions that generally don't progress such as ductal carcinoma in situ (DCIS) and some screening-detected slow-growing prostate cancers are not termed as carcinomas, which can frighten patients and lead to very aggressive therapy and potential overtreatment.





Kramer said that part of this discussion stemmed from a 2009 National Institutes of Health State-of the-Science Conference on “Diagnosis and Management of Ductal Carcinoma in Situ” (OT, 10/25/09), and that the working group topic two and a half years later was one of his early initiatives when he returned to NCI to take over his current prevention position (OT, 12/25/11).

Kramer said that some of the recommendations would be incorporated into NCI's research plans, including a collaborative project between his prevention division and the Division of Cancer Biology that would stimulate research on the molecular biology of overdiagnosis and missed diagnoses of interval cancers that come in between screenings and could help provide targets for chemoprevention.

Focus and tissue collection, though, should not be restricted to just premalignant cells but should also include normal surrounding tissue, he said. In addition, some of the other recommendations, such as changing terminology, does not actually fit within the research purview of the NCI and might more properly fit within the scope of the Institute of Medicine.

Back to Top | Article Outline

Time for a Paradigm Shift

Also speaking in a telephone interview several days after the article appeared, Esserman, Director of the Carol Franc Buck Breast Cancer Center at the UCSF Helen Diller Family Comprehensive Cancer Center and Professor of Surgery and Radiology at UCSF, said that most of the feedback she had received so far was positive, in contrast to that for an earlier article, “Rethinking Screening for Breast Cancer and Prostate Cancer,” that she, Yiwey Shieh, and Ian Thompson had written for JAMA in October 2009 (302:1685-1692), which was greeted with a fair amount of hostility.

Noting that more isn't necessarily better, and pointing to use of the Halsted radical mastectomy versus lumpectomy as an example for breast cancer, Esserman said that it was time for a paradigm switch involving changing how cancer is thought about and what it is called.

“Many oncologists don't like it [changing the terminology] because they see a lot of metastatic cancer and many of their patients have progressive cancer,” she explained.

Screening was started, of course, she related, because it was thought that if cancer was caught early it would decrease advanced-stage cancers, but it is now known that assumption is too simplistic because everyone does not need extensive treatment, especially since screening uncovers so many slow-growing low-risk cancers.

She cautioned about rushing to treatment: “Unless it's leukemia, it's not an emergency,” and said she also thought it appropriate for an Institute of Medicine committee to be created to help better define what cancer is called.

Esserman is also the principal investigator for the ATHENA Breast Health Network (—an initiative involving all five University of California medical centers and about 150,000 women throughout California to develop risk-based screening programs, which also includes a DCIS registry.

Back to Top | Article Outline

‘Statement of the Obvious’

Thompson, Director of the Cancer Therapy and Research Center at the University of Texas Health Science Center at San Antonio, said he thought the JAMA Viewpoint article “was a statement of the obvious,” and was surprised at some of the national media attention.

He related seeing a newly diagnosed patient that week with a slow-growing prostate cancer who was “simply distraught by his diagnosis of cancer,” even though it was not life-threatening. Thompson noted that the article was originally a comprehensive paper on the subject but that JAMA thought publishing a commentary would be more appropriate. (The original paper has still not been published.)

He said the visceral response when seeing a headline that says: “Too much cancer is diagnosed,” might be to say, ”Oh my goodness, will the government or medical establishment withhold the opportunity to diagnose cancer that might save a life?”

He added he would like to see an intelligent discussion about individualizing that type of decision and noted you had to be careful that “in the process you don't swing the pendulum too far in the other direction and cause harm by not identifying.”

Thompson said that after the U.S. Preventive Services Task Force recommended last year against PSA-based screening for prostate cancer (, he saw patients with clearly very high risks of aggressive cancers who were told not to worry about it by a physician. “I also saw patients with whom it was necessary to do damage control because they were told that prostate cancer is nothing to worry about and they had been diagnosed with an aggressive prostate cancer that posed them considerable harm,” he said.

He said he was concerned about what he termed a “sound-bite front-page” perspective that prostate cancer is not bad when there is actually risk, and emphasized the need to seek a more precise definition.

He said that as a surgeon he believes in active surveillance and has about 200 to 300 patients on active surveillance for prostate cancer. He recalled the early days when it was difficult to get men to agree to active surveillance and he had to weather the storm of people saying: “How can you do this?”

“Now it's becoming more common and there are surgeons who ask me how to do that [active surveillance]. This is a car that nobody's driven before,” he said, noting that the NCI and the Canary Foundation's active surveillance trial was about to enroll its 1,000th patient.

“My biggest concern is that [others will think] it's too hard to make the changes [in nomenclature]—and that's not a good thing for the practice of medicine.”

Back to Top | Article Outline

Range of Reactions

OT also spoke with experts in various oncology subspecialties about their respective reactions to the JAMA commentary:

Back to Top | Article Outline

30% Overtreated?



Otis W. Brawley, MD, Chief Medical and Scientific Officer and Executive Vice President of the American Cancer Society, said that much of what is believed to be cancer today is based on pathology from 163 years ago using microscopes (i.e., by Rudolf Virchow in 1850).

“Our new 21st century definitions should use modern-technology tumor profiles,” he said. “In prostate cancer, there were about a dozen articles last year that could be summarized by saying that Gleason 6 shouldn't be called prostate cancer.” In Europe, he noted, patients with Gleason 6 typically have watchful waiting, whereas in the U.S., the attitude is that it has to be treated right away: “People hear the word ‘cancer’ and think it has to be treated immediately.”

He pointed out that some pathology textbooks have called DCIS “precancerous” and not cancer, but that others have changed the terminology to “stage 0 cancer.”

“Referring to DCIS as a non-invasive cancer is oxymoronic: there's no such thing as non-invasive cancer,” he said, adding a political observation that including DCIS as a cancer had increased the numbers of breast cancer survivors.

Overdiagnosis is a problem and it's necessary to set a scientific agenda determining which cancers need treating and which need waiting: “The biggest part of the JAMA article was a call to get this on the scientific agenda,” he said.

“Some of us think that about 30 percent of the 11 million cancer survivors got overtreatment.”

Back to Top | Article Outline

‘First Understand the Biology’

Richard L. Schilsky, MD, Chief Medical Officer of the American Society for Clinical Oncology, said that he agreed with much in the JAMA article.

“There is tremendous biological heterogeneity in cancer. Some cancers are indolent and some are aggressive, and screening has the propensity to turn up more indolent cancers that are not that clinically important,” he said.



Screening should be “risk-adapted,” he said, but also expressed concern that it might be premature to change the terminology: “It's not the nomenclatural, it's the biology, and until we understand the biology better we shouldn't change the nomenclature.”

He said that the tools are not yet available to determine which cancers are “good” or bad and an unintended consequence of renaming could be that patients won't adhere to follow-up recommendations from screening tests: “If we tell patients that something really isn't cancer, they may not take it as seriously, and some may become complacent about their diagnosis,” he said, adding that he often struggles with how to discuss pre-invasive lesions with patients.

“Should I say you have cancer, but it's early, highly curable, and won't spread? Or that you have something that may become cancer, but don't worry about it, it isn't cancer? It's subtle, but may affect how some patients continue health care.”

A missing piece of the puzzle, he said, is health behavior and how communicating about a diagnosis can influence subsequent health behaviors: “How does changing the name impact how a patient's health behavior changes?”

Back to Top | Article Outline

‘Mixed Blessing’



Robert J. Mayer, MD, the Stephen B. Kay Family Professor of Medicine at Harvard Medical School; Vice Chair for Academic Affairs, Medical Oncology, at Dana-Farber Cancer Institute; Faculty Associate Dean for Admissions at Harvard Medical School, said that in retrospect, it was fairly primitive how cancer was initially identified, codified, and classified by morphologists looking at the appearance of cells.

“And the appearance does not define biological behavior and does not necessarily define virulence. On the other hand, these were the only tools that were present then. It's only been in the last 35 or 40 years that use of immunological or immunohistochemical staining, genetic subtyping, genomics, and microarrays have given us a great deal more information than ever before.”

Mayer, a member of OT's Editorial Board, pointed out that Barrett's esophagus and ulcerous colitis are premalignant conditions for certain GI cancers and that the recommendation was for brushings or biopsies that didn't necessarily look for cancer but rather for dysplasia that was somewhere between normal and abnormal and akin to DCIS in breast cancer.

“But it's not a quantitative measurement, and to find dysplasia and then to talk about high grade or low grade could [result] in a wide variability among different pathologists,” he said, adding that being saddled with a nomenclature derived from morphology doesn't take into account the biological activity of cells, virility, or life-threatening predictive value or appearance of the cells.

Mayer noted the tremendous anxiety, stress, and fear placed on patients by cancer, especially when one size doesn't fit all in cancer, and that using a single name can be misleading and inaccurate and lead to overtreatment or undertreatment.

Society needs to educate consumers when they are healthy about what the different terms mean, he said. “It's a mixed blessing that mass screenings such as mammography and PSAs have found many of these non-life-threatening conditions through screenings endorsed by the cancer establishment, since it can also lead to an incorrect sense of security that if testing is normal, people are safe and should not be concerned.”

Back to Top | Article Outline

‘Don't Want to Miss Opportunities to Intervene’

Lynn M. Schuchter, MD, the C. Willard Robinson Professor and Chief of Hematology/Oncology and Program Leader for the Melanoma and Cutaneous Malignancies Program at the University of Pennsylvania's Abramson Cancer Center, said that at her center there had been a lot of reaction to the JAMA article and that the article had been forwarded several times to an internist involved in screening.

“It's a provocative piece and a good first step in allowing for a fruitful conversation,” she said, adding that she thought the most important message was in the conclusion that the task force's recommendations are intended to be initial approaches only, and that “physicians and patients should engage in open discussion about these complex issues.”

“From my own perspective,” Schuchter said, “I get a little concerned. I see that in my area of melanoma sometimes when the word ‘atypical’ is used, some take it to mean everything's fine and don't read all the details of a pathology report and don't do the next step that is required.”

She said that many times a condition such as DCIS doesn't require treatment, but sometimes it does, and that language and communication are very important.

“It's valid that we don't want to overdiagnose or overtreat, but we also don't want to miss opportunities to intervene.”

Schuchter said she gets a little concerned that when you take the word “cancer” out, there is a false sense of security that there is no need for follow-up or to act: “The most difficult part is finding the sweet spot in terms of appropriately alerting clinicians and patients that some action should be taken, but without overreacting.”

Back to Top | Article Outline

‘Challenge Is to Identify Biomarkers’

Ramaswamy Govindan, MD, Co-Director of the Section of Medical Oncology and Professor of Medicine at Washington University School of Medicine, Alvin J. Siteman Cancer Center, said the heart of the JAMA article is that only those conditions that have a risk of lethal progression if left untreated should be labeled as cancer. Govindan, OT's Clinical Advisory Editor for Oncology and a lung cancer specialist, said routine CT screening of smokers is likely to identify a lot of premalignant lesions that may never become invasive cancer.

“I think avoiding the term ‘cancer or carcinoma’ is not a bad idea for lesions that have low to no potential for developing cancer, but the challenge really is to identify biomarkers that would clearly identify the lesions at high risk of progression and those that would pursue a more indolent course.”

© 2013 by Lippincott Williams & Wilkins, Inc.
Home  Clinical Resource Center
Current Issue       Search OT
Archives Get OT Enews
Blogs Email us!