Currently, one of the most challenging problems in oncology is to accurately predict whether neoplastic lesions detected by screening tests will progress. The focus on developing ever-more sensitive cancer screening tests has produced the clinical dilemma of overdiagnosis. Overdiagnosis occurs when a cancer case identified by a screening test would have gone undiagnosed otherwise and not caused clinical symptoms or death in the person’s lifetime. It triggers overtreatment, which causes not only overutilization of resources, but also substantial physical, emotional, and financial harm to the patient, and sometimes even treatment-related death.
In a recent study, examination of the Surveillance, Epidemiology, and End Results (SEER) data for trends in breast cancer incidence from 1976 through 2008 among women 40 years of age or older showed that overdiagnosis accounted for an estimated 31% of all breast cancers diagnosed in 2008.1 However, population-based ecologic trends can only provide suggestive evidence of overdiagnosis. Randomized controlled screening trials with long-term follow-up provide more direct evidence.
Recently, as reviewed in The ASCO Post, early release online, the Canadian National Breast Screening Study, with 25-year follow-up data, reported a lack of mortality benefit of annual mammography screening for breast cancer compared with physical examination or usual care in women aged 40 to 59 years.2 This was not a new finding from the trial. The new, headline-grabbing results of the trial were that about 1 in 5 of the screen-detected invasive cancers was overdiagnosed (and if ductal carcinoma in situ had been included, the proportion would be even higher).
While debates on the efficacy of mammography screening continue, we should not overlook strategies to mitigate harms. Specifically, how can we mitigate the harms of overdiagnosis? The problem can be addressed at several levels. Two obvious approaches are to (1) improve patient (and physician) understanding of overdiagnosis, and (2) promote the development of additional complementary tissue- and blood-based biomarkers that can be used to predict behavior of screen-detected lesions.
Understanding and Communication
How best to improve understanding and communication between physicians and patients? Recently there has been a call for changing the terminology for early, nonlethal lesions.3,4 A change in terminology to reflect what we do know about the underlying biology of tumors may diminish the angst associated with the word “cancer,” a term that can drive overtreatment. Taking the lead from the field of cervical cancer screening, some have suggested changing the term “ductal carcinoma in situ’” (DCIS) to “ductal intraepithelial neoplasia” (DIN).
Unfortunately, the dilemma of overtreatment would still not be fully resolved for the patient or physician by simply changing terminology, because the ability to predict lesion progression remains limited by our understanding of the underlying molecular drivers. This suggests new avenues for research.
There are already a few molecular indicators. For example, interval breast cancers (those missed in spite of screening) are 1.8- to 2.6-fold more likely to be estrogen receptor (ER)-negative compared to screen-detected tumors,5,6 and are also more likely to be diagnosed at a more advanced, clinically aggressive stage.7,8 A more refined molecular characterization of screen-detected lesions could improve therapeutic decisions. However, because of the heterogeneous nature of breast cancer and its interaction with the surrounding tissue microenvironment, it is unlikely that single genes or gene products will suffice.
With the advent of the “omics” revolution, there has been an explosion in our knowledge of the molecular basis for breast cancer and the identification of intrinsic subtypes of breast cancer and correlation of these subtypes with disease prognosis.9-11 However, these studies have not been well linked to the mode of detection. Without better annotation on method of detection, sorting out the spectrum of behavior of screen-detected lesions is extremely difficult.9-11
Pilot Program
The National Cancer Institute’s Early Detection Research Network (EDRN) (www.cancer.gov/edrn) investigators have therefore initiated a pilot program to study the biologic behavior of breast cancer that incorporates annotation of method of diagnosis, along with the other well-known risk factors for progression. There is also a separate Funding Opportunity Application (FOA) to study the spectrum of changes associated with tissue microenvironment, genomic and proteomic changes, and genomic heterogeneity both between and within premalignant lesions combined with imaging to study the biology of early breast cancer lesions.
When possible, tissues from cohorts that prospectively follow patients for progression will be used. We also encourage studies of similarities and differences in molecular characteristics between interval-detected cancers and screen-detected cancers or cancers that recur after initial therapy.
These new research directions could help guide clinical decision-making with respect to both extent of treatment and frequency of follow-up for breast cancer survivors. The goal is to develop clinical strategies that may retain any benefits of screening while diminishing the harms inherent in increasingly sensitive screening tests.
Disclosure: Drs. Srivastava and Kramer reported no potential conflicts of interest.
Dr. Srivastava is Chief, Cancer Biomarkers Research Group, Division of Cancer Prevention, National Cancer Institute, National Institutes of Health. Dr. Kramer is Director, Division of Cancer Prevention, National Cancer Institute, National Institutes of Health.
The opinions expressed in this article represent those of the authors, and not official positions of the U.S. Department of Health and Human Services or the National Institutes of Health.
For more discussion of the Canadian National Breast Screening Study results, read the accompanying commentaries by Therese B. Bevers, MD, and Daniel B. Kopans, MD, FACR.
References
1. Bleyer A, Welch HG: Effect of three decades of screening mammography on breast-cancer incidence. N Engl J Med 367:1998-2005, 2012.
2. Miller AB, Wall C, Baines CJ, et al: Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: Randomized screening trial. BMJ 348:g366, 2014.
3. Esserman LJ, Thompson IM, Reid B: Overdiagnosis and overtreatment in cancer. JAMA 310:797-798, 2013.
4. Dunn BK, Srivastava S, Kramer BS: The word “cancer”: How language can corrupt thought. BMJ 347:f5328, 2013.
5. Nelson HD, Tyne K, Naik A, et.al: Screening for breast cancer: An update for the US Preventive Services Task Force. Ann Intern Med 151:727-742, 2009.
6. Collett K, Stefansson IM, Eide J, et al: A basal epithelial phenotype is more frequent in interval breast cancers compared with screen detected tumors. Cancer Epidemiol Biomarkers Prev 14:1108-1112, 2005.
7. Porter PL, El-Bastawissi AY, Mandelson MT, et al: Breast tumor characteristics as predictors of mammographic detection: Comparison of interval- and screen-detected cancers. J Natl Cancer Inst 91:2020-2028, 1999.
8. Otto SJ, Fracheboud J, Verbeek AL, et al: Mammography screening and risk of breast cancer death: A population-based case-control study. Cancer Epidemiol Biomarkers Prev 21:66-73, 2012.
9. Perou CM, Sorlie T, Eisen MB, et al: Molecular portraits of human breast tumours. Nature 406:747-752, 2000.
10. Sorlie T, Tibshirani R, Parker J, et al: Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci USA 100:8418-8423, 2003.
11. Curtis C, Shah SP, Chin SF, et al: The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature 486:346-352, 2012.
This commentary represents the views of the author and may not necessarily reflect the views of ASCO®.
