Among all the battles we fight with cancer, one has gained the most public attention is breast cancer. Yet, after after years of research and millions of dollars in funding, one in eight women will still be diagnosed with breast cancer in her lifetime (1). The best defense we have so far against this malady is early detection through regular mammogram screenings. However, the treatment of breast tumors may be revolutionized by a recent study published in the Archives of Internal Medicine, where Welch claims that tumors may disappear without treatment (2).
Today, there are about 34 million mammogram procedures performed in U.S. each year (3) and the FDA recommends that all women over the age of 40 get regular screenings once every year (4). Once a woman is suspected of having cancer due to a positive mammogram result, she undergoes various tests, biopsies, and sometimes surgeries to follow up. Five to fifteen percent of mammogram studies require additional mammograms or ultrasound tests because of abnormal results, and if follow-up studies still result in suspicious findings, the next step is to perform a biopsy (5).
Biopsies are usually performed by using a needle to take out a sample of tissue for a pathologist to analyze for malignancy. This is the only sure way to diagnose cancer. Four out of five biopsies are noncancerous (6), but many patients are subjected to much unnecessary stress throughout the testing process. Studies have shown that women experience long-term as well as short-term anxieties after receiving a false-positive diagnosis (7).
A woman receiving a mammogram at a clinic. DMS prof. H. Gilbert Welch argues that mammogram screenings might do less good for women that we believe.
Researchers are trying to find ways to ameliorate and perhaps eventually circumvent this imperfect method of treating breast cancers. As seven to eight percent of women aged 40-49 who have annual mammograms undergo breast biopsies, the majority of which are benign, this recent study could have potential benefits for many women.
Methods
In Welch’s study, entitled “The Natural History of Invasive Breast Cancers Detected by Screening Mammography,” two groups of women aged 55-69 were observed, a screened group and a control group of about 100,000 women each.
The screened group underwent three rounds of mammogram screenings throughout the years 1996-2001. Once an invasive cancer was diagnosed, the patient immediately received treatment and the case was counted towards the total number of cancerous findings in the group. The control group was observed through the years 1992-1997 and was administered a one-time prevalence screen at the end of those six years. The total number of invasive tumors of the screened group collected through the three rounds was compared to the one-time prevalence screen of the control group.
Results
The occurrence of cancer should be statistically equivalent in both groups, so the total diagnoses of cancer were expected to be fairly similar. However, the results showed that the control group had a 20 percent lower incidence of tumor diagnoses than the screened group. This study implies that some tumors that were supposed to have existed during the observation period in the control group, by comparison to the screened group, disappeared by the end of the six years and resulted in significantly fewer cases of cancer in the control group.
To ensure accurate results, the women tested had similar education backgrounds, incomes, and reproductive histories. Although the time of screenings for the two groups differed by six years and may have affected the expected cancer statistics due to lifestyle changes and differences in standards of living, Welch estimated those differences should only account for four percent of the difference at most, which is too far from the observed 20 percent to explain the discrepancy (8).
Also, in response to doubt concerning the sensitivity of mammograms and the imperfections of the technology, a follow-up study was conducted on the two groups. If the prevalence screen had missed some tumors during the first study, then the margin of difference between the results of the control group and the screened group should have narrowed in the follow-up study. However, the follow up study’s results remained relatively consistent with the first study.
Impact
Featured in the New York Times on November 24, 2008, this study has spurred a new way of thinking among groups ranging from cancer researchers to doctors prescribing mammography screenings to patients, although some specialists are hesitant to accept the accuracy of the findings.
Mary Hu, a cancer specialist in New York City who performs thousands of mammograms each year, states, “Based on my ten years’ experience, I find it hard to pile sufficient data and even harder to come to any solid conclusion. […] In general, there is still a long way to go in establishing the definite conditions and circumstances of disappearing tumors” (9).
Regardless of whether or not the medical community accepts these conclusions as definitive, there cannot be any immediate changes to the screening and treatment process yet, as it is not worth risking the dangers of invasive cancer to see if a tumor will spontaneously dissipate. Welch states, “The most important idea […] is to recognize the need for a more nuanced view of cancer. Some cancers do progress maliciously, but others do regress as well. Cancer is much more dynamic than commonly perceived” (8). Welch has also written numerous other publications in the New York Times, Washington Post, Los Angeles Times, and USA Today about the judiciousness of various cancer screenings and potential pitfalls of early diagnoses. However, statistics from the American Cancer Society still favor early detection and treatment of cancer, as it saves thousands of lives each year (10).
There have been rare cases of other cancers such as melanoma, kidney cancer, and neuroblastoma disappearing on their own without treatment as well (11), but most doctors do not place much importance on them, as they are considered rare. However, this study encourages researchers to give more consideration to this behavior of cancer, which is no longer considered such an anomaly. Further investigation could lead to the reduction of painful and emotionally stressful ordeals for many patients diagnosed with breast cancer if progressing cancers could eventually be differentiated from regressing cancers.
References
1. Surveillance Epidemiology and End Results, SEER Stat Fact Sheets (2008). Available at http://seer.cancer.gov/statfacts/html/breast.html (17 January 2009).
2. H. G. Welch, The excessive focus on mammography (Los Angeles Times, November 3, 2008).
3. National Cancer Institute, Digital vs. Film Mammography in the Digital Mammographic Imaging Screening Trial (DMIST): Questions and Answers. Available at http://www.cancer.gov/newscenter/pressreleases/dmistqanda (17 January 2009).
4. US Food and Drug Administration, Mammograms and Breast Cancer. Available at http://www.fda.gov/opacom/lowlit/mammo.html (17 January 2009).
5. Radiological Society of North America, Inc., Mammography (1 September 2008). Available at http://www.radiologyinfo.org/en/info.cfm?PG=mammo (7 April 2009)
6. American Cancer Society, For Women Facing a Breast Biopsy (18 August 2008). Available at http://www.cancer.org/docroot/CRI/content/CRI_2_4_6x_For_Women_Facing_a_Breast_Biopsy.asp (12 April 2009).
7. N.T. Brewer, T. Salz, S. E. Lillie, Annals of Internal Medicine. 146, 502-510 (April 2007)
8. H.G. Welch, Personal interview, 12 January 2009.
9. M.X. Hu, Persional interview, 16 January 2009.
10. American Cancer Society, Breast Cancer: Early Detection (30 September 2008). Available at http://www.cancer.org/docroot/CRI/content/CRI_2_6x_Breast_Cancer_Early_Detection.asp (15 January 2009).
11. G. Kolata, Study Suggests Some Cancers May Go Away (New York Times, November 25, 2008, p. A19 of the New York edition).