Cancer Screening Principles

November 20, 2021  •  Leave a Comment
  • Cancer screening aims to detect cancer early before symptoms develop so that treatment (hopefully a cure) can be initiated, thereby extending the patient's life without causing additional harm(s) from the cancer treatment. To date, the best cancer screening available is for cervical cancer, where ideally, a pre-cancerous cervical lesion is identified by screening and then treated, actually preventing cancer from ever occurring. Colorectal cancer screening is also advantageous because pre-cancerous polyps can be detected by screening and subsequently removed. In these scenarios, cancer screening is also a cancer prevention method, identifying and treating pre-cancerous lesions in conjunction with cancer screening. In addition to cervical and colon cancer screening, effective screening exists for breast cancer. The high false-positive rates associated with prostate cancer screening and lung cancer screening prevent prostate and lung cancer screening from being classified as effective screening programs for the general public.
  • Both sensitivity (not missing cancers) and specificity (low false positives) are essential in cancer screening. The prevalence of a specific cancer is also significant. Prevalence impacts the positive predictive value of a screening test. The positive predictive value represents the ratio of true positives to all positives (true positives plus false positives). Therefore, more true positives will come from a population with a higher disease prevalence. And more true positives will support a higher positive predictive value. Although, a high prevalence of cancer is only a component to boost the positive predictive value. The percentage of false positives is still dependent on the specificity of the screening test. A low specificity will generate more false positives, which negatively affects the positive predictive value.
  • Screening tests are also vulnerable to bias. Cancer screening tests can be culprit to lead-time bias. Early detection for a cancer with no effective treatment will increase survival time. For example, person A is diagnosed with cancer by a screening test ten years before passing away from that cancer versus person B, who is diagnosed with cancer after noticing a lump underneath their skin and then passes away a year later from that cancer. Person A and person B received no treatment for their cancers. However, person A lived for ten years with cancer compared with person B, who only lived a year with cancer. Lead-time bias can be avoided by measuring a screening test's performance with cause-specific mortality instead of survival time following diagnosis. The cause-specific mortality rate is determined by the number of deaths from a specific cancer divided by the population size.
  • The cause-specific mortality rate would be the same for person A and person B in the scenario above. It may seem that cause-specific mortality would only be applicable for cancers that have a cure. It is true, for cancers where a cure is available, cause-specific mortality will most certainly differ from cancers where no treatment is available. However, the cause-specific mortality rate can also increase for cancers where treatment significantly prolongs life but does not cure because an individual's extended life may lead to them dying from something other than cancer, such as heart disease, for example.
  • Screening tests are only performed on asymptomatic individuals. Individuals undergoing cancer screening are healthy. Therefore, the benefit of cancer screening must outweigh the risk of a harmful outcome caused by the screening test. Using prostate and lung cancer screening as examples, false positives from cancer screening are associated with multiple harms. False positives from both prostate and lung cancer still cause patients anxiety. Prostate biopsies following positive screening tests are associated with incontinence and erectile dysfunction. And following a positive lung cancer screening, patients may be subjected to the additional stress associated with waiting six months to a year for their follow-up CT scan as the next step in evaluation. There is also the consideration of overdiagnosis when considering cancer screening. Overdiagnosis in cancer is the diagnosis of very slow-growing or indolent cancer that would not cause symptoms or medical problems. Screening tests may identify lesions classified as cancer based on pathologic classification that clinically does not behave as harmful cancer does. Indolent cancers are found but are not limited to the spectrum of breast cancer, thyroid cancer, prostate cancer, and some lymphomas.
  • References:
  • MacLean C. D. (1996). Principles of cancer screening. The Medical clinics of North America, 80(1), 1–14. https://doi.org/10.1016/s0025-7125(05)70423-5
  • Croswell, J. M., Ransohoff, D. F., & Kramer, B. S. (2010). Principles of cancer screening: lessons from history and study design issues. Seminars in oncology, 37(3), 202–215. https://doi.org/10.1053/j.seminoncol.2010.05.006
  • Roder, D. M., & Buckley, E. (2020). Overdiagnosis of cancer in Australia: the role of screening. The Medical journal of Australia, 212(4), 159–160. https://doi.org/10.5694/mja2.50494

Cancer Screening Principles


Comments

No comments posted.
Loading...

Archive
January February March April May June July August September October November (1) December (3)
January February (1) March (1) April May June (1) July August (2) September (1) October (3) November December
January February March April May June July August September October November (1) December
January February March April May June July August September October November December
January February March (1) April May June July August September October (3) November (7) December (2)
January (3) February March April May (1) June July (1) August September October (2) November (2) December (1)
January February (2) March April (1) May June July August September October November December (1)
January February March April May June July August September October (4) November (44) December (30)
January (15) February (8) March (2) April May June (3) July (1) August September October November December (6)
January (4) February March April May June July August September October November December
January February (1) March April (1) May (17) June (21) July (88) August September October November (3) December