In reflecting on the points raised by Dr Wilkinson in the May 2024 issue of Canadian Family Physician,1 we would like to offer the following responses.
On earlier detection. On how earlier detection is needlessly identifying cancers that would not impact outcomes, Dr Wilkinson wrote: “Scientific data do not support the spontaneous regression of cancers: in a study of 479 untreated breast cancers followed over 10 years, zero cancers spontaneously regressed or disappeared.”1,2
We are puzzled by this point, as spontaneous regression is tangential to our question: Does earlier detection of cancer result in better health outcomes? On this, the most important question, we provided 3 examples in cancer screening (ie, melanoma, neuroblastoma, and thyroid cancer).3 Scientists know that high-quality evidence is required to be confident that earlier is better. To counter this point in our article, Dr Wilkinson referenced a single cohort study of the persistence of screen-detected breast tumours (ductal carcinoma in situ or invasive cancer).1,2 This reference was to an observational study that did not report health benefits that would truly matter to patients. As we write this letter, the benefit of detecting ductal carcinoma in situ remains uncertain.
Bottom line? Earlier detection is essential for screening to be of benefit. Importantly, early detection is often not beneficial.
On new technology. On newer technology and health benefits for our patients, Dr Wilkinson wrote: “From 1975 to 2019, US breast cancer mortality decreased by 58%, attributable to both screening and treatment.”1
To be clear, we referred to technology in our article in the context of imaging for cancer screening.2 However, Dr Wilkinson is touting the benefits of newer “cancer diagnostics and therapies.”1 Interestingly, as cancer treatment improves, screening to achieve earlier detection becomes less important. In the context of screening, it is challenging to disentangle the fraction of cancer deaths prevented by improved treatment from that attributable to improved imaging. Research shows improvements in treatment were responsible for most of the observed reduction in breast cancer mortality in the United States.4
On screening saving lives. On the point that “Cancers diagnosed through screening are earlier-stage cancers with better survival and decreased mortality, meaning that lives are saved,”1 5-year survival statistics are presented as evidence.
An early article in the Prevention in Practice series made the following key point: In screening for cancer, appropriate outcome measures for determining benefit include overall and disease-specific mortality; inappropriate measures include incidence (new cases) and 5- or 10-year survival.5 The use of a metric such as 5-year survival is highly inappropriate to judge the effect of screening because of the problems of lead-time bias, length-time bias, and overdiagnosis of screening-detected cancers. By definition, an overdiagnosed cancer does not kill. Overdiagnosis, lead-time bias, and length-time bias lead to a mirage of benefit.
To contend that lives are saved at the population level is inaccurate, as explained in our article. This important point received further support from a recent analysis of estimated lifetime gained by cancer screening tests.6
We would all hope to reduce premature mortality from cancer, or indeed from any disease. Our patients need clinicians who relay accurate information they can understand in a calm, nonemotional way. It is profoundly unjustified to suggest that members of the Canadian Task Force on Preventive Health Care are anti-screening when we write about the science and the need to balance potential harms against any benefits of cancer screening.
Our patients would be better served by collective efforts to attack our lack of knowledge about screening, rather than continuing to attack the myths or those who point to them. We suffer from many knowledge gaps with respect to the value of cancer screening interventions, including that of newer screening tests. Randomized controlled trials are sorely needed to address these gaps.
Footnotes
Competing interests
None declared
The opinions expressed in letters are those of the authors. Publication does not imply endorsement by the College of Family Physicians of Canada.
References
- 1.Wilkinson AN. Evidence-based truths about the benefit of cancer screening [Letters]. Can Fam Physician 2024;70:296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kagan Arleo E, Monticciolo DL, Monsees B, McGinty G, Sickles EA.. Persistent untreated screening-detected breast cancer: an argument against delaying screening or increasing the interval between screenings. J Am Coll Radiol 2017;14(7):863-7. Epub 2017 Apr 28. [DOI] [PubMed] [Google Scholar]
- 3.Thériault G, Reynolds DL, Grad R, Dickinson JA, Singh H, Szafran O, et al. Debunking myths about screening. How to screen more judiciously. Can Fam Physician 2023;69:767-71 (Eng), e216-20 (Fr). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Welch HG, Prorok PC, O’Malley AJ, Kramer BS.. Breast-cancer tumor size, overdiagnosis, and mammography screening effectiveness. N Engl J Med 2016;375(15):1438-47. [DOI] [PubMed] [Google Scholar]
- 5.Bell NR, Grad R, Dickinson JA, Singh H, Moore AE, Kasperavicius D, et al. Better decision making in preventive health screening. Balancing benefits and harms. Can Fam Physician 2017;63:521-4 (Eng), 525-8 (Fr). [PMC free article] [PubMed] [Google Scholar]
- 6.Bretthauer M, Wieszczy P, Løberg M, Kaminski MF, Fiskergård Werner T, Helsingen LM, et al. Estimated lifetime gained with cancer screening tests: a meta-analysis of randomized clinical trials. JAMA Intern Med 2023;183(11);1196-203. [DOI] [PMC free article] [PubMed] [Google Scholar]