Does the NCI Have a Problem With Cancer Screening?

Article

Reading the contents of the NCI Cancer Bulletin in recent years, I have become increasingly concerned by what I perceive as an irrational, pervasive, and persistent editorial bias against cancer screening.

Frederic W. Grannis Jr., MD

Frederic W. Grannis Jr., MD

Reading the contents of the NCI Cancer Bulletin in recent years, I have become increasingly concerned by what I perceive as an irrational, pervasive, and persistent editorial bias against cancer screening. This bias is particularly evident in the November 27, 2012 issue as the editors devote the entire issue emphasizing shortcomings and risks of screening. Although the issue is touted as an analysis of evidence, I suggest that precious little new evidence is cited; articles consist primarily of rehashed opinion from National Cancer Institute (NCI) epidemiologists and their collaborators.

Otis Brawley, MD, chief medical and scientific officer of the American Cancer Society, opens by accusing our society of “bombarding” and “luring” Americans into cancer screening programs. He correctly emphasizes that screening is complicated and that patients must be educated about risks and benefits, but glosses over benefits, to stress risk of anxiety and overdiagnosis. In fact, published information suggests that anxiety in screened individuals is relatively minor and well tolerated in comparison to the enormous physical and psychological distress experienced by dying cancer patients, and their caregivers and families.

Virginia Moyer, MD, chair of the US Preventive Services Task Force (USPSTF), describes the USPSTF philosophy and methodology, and suggests that the recommendations of the group with regard to breast cancer screening in younger women were not wrong, just badly communicated. Many USPSTF critics dispute this interpretation. Moyer does not inform NCI Cancer Bulletin readers that the USPSTF is delaying publication of a guideline on lung cancer screening until 2014. Delay in implementation of lung cancer screening will cost many lives.

Crunching Numbers: What Cancer Screening Statistics Really Tell Us” remodels old NCI graphics illustrating lead-time and overdiagnosis biases, using theoretical scenarios in which biases theoretically skew research results, rather than actual clinical data. In the lead-time example posited, a cohort of patients has 100% survival at 5 years, but 0% survival at 10 years. In the 45 years spent reading medical journals, I have never seen anything remotely resembling such data in a clinical trial. Although NCI epidemiologists hypothesized that there was a substantial proportion of overdiagnosed lung cancers in the Mayo Lung Project trial, incidence curves from the NCI's Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial show no excess lung cancers in research subjects screened with chest roentgenograms. National Lung Screening Trial (NLST) incidence curves show some excess lung cancer cases, but when a lead-time in diagnosis of approximately 2 years is factored into the equation, there is no evidence of overdiagnosed lung cancers.

The article entitled “After Landmark Study, Exploring Questions about Lung Cancer Screening” notes that “only a handful of US clinicians and medical centers have adopted the screening method and the patient selection criteria and evaluation criteria used in NLST.” Though strictly true, it omits the fact that more than 125 centers have signed on to National Framework for Excellence in Lung Cancer Screening and Continuum of Care to begin offering lung cancer screening using far more rigorous methods than those used by NLST investigators. In addition, eight national and international organizations have issued guideline advice to clinicians and patients, with three organizations (NCCN, Society of Thoracic Surgeons, and American Association for Thoracic Surgeons) recommending broader patient selection criteria than NLST.

This article also warns of false positive rates of 25% in CT lung cancer screening, failing to cite multiple publications from groups using stronger diagnostic algorithms than NLST that report substantially lower false positive rates.

All That Glitters: A Glimpse into the Future of Cancer Screening” describes encouraging research in molecular screening, but candidly reports “we haven’t had any major breakthroughs in early detection.”

I found one jewel in a linked audio discussion of “absolute” risks and benefits by long-time screening critic Steven Woloshin, MD, who offers an example of 20% reduction in cost applied to both $40,000 and $10 products, concluding that a cost saving of $8,000 is better than $2. This set me to ponder the question-what is the value of my wife or my child?-alternatively, what cost would I incur to prevent their cancer deaths? A no-brainer; the answer is-whatever I own, in addition to everything I can beg, borrow, or steal. I believe that this question is an important consideration in the calculus of risks and “absolute” benefits of screening. Incidentally, my wife is a long-term survivor of breast cancer detected by mammographic screening, and I can assure readers that neither of us loses any sleep brooding whether the cancer was an overdiagnosed “potato chip.”

The issue closes with a report on a prestigious award to Steven Rosenberg, NCI’s chief of surgery at their Center for Cancer Research. The article does not comment that NCI surgical research has virtually ignored surgical research in the nation’s number one cancer killer, lung cancer, and instead narrowly focused on immunological research, which, while producing much important basic science information, has yielded few advances that translate into practical, curative treatment of cancer patients.

Recent Videos
The FirstLook liquid biopsy, when used as an adjunct to low-dose CT, may help to address the unmet need of low lung cancer screening utilization.
An 80% sensitivity for lung cancer was observed with the liquid biopsy, with high sensitivity observed for early-stage disease, as well.
9 Experts are featured in this series.
9 Experts are featured in this series.
Harmonizing protocols across the health care system may bolster the feasibility of giving bispecifics to those with lymphoma in a community setting.
2 experts are featured in this series.
Patients who face smoking stigma, perceive a lack of insurance, or have other low-dose CT related concerns may benefit from blood testing for lung cancer.
9 Experts are featured in this series.
Related Content