For thyroid cancer , the rate of diagnosis has doubled in the last 30 years, with no change in death rate. The increased new cases are confined to papillary thyroid cancer, which has the most favorable prognosis [23]. It is estimated that overdiagnosis in women accounts for 90% of thyroid cancer cases in South Korea; 70–80% in the USA, Italy, France, and Australia; and 50% in Japan, the Nordic countries, England, and Scotland [24]. In Japan, thyroid cancer incidence among screened children and adolescents was approximately 30 times as high as the national average only a few months after intensive screening programs for these age groups began in response to the 2011 nuclear accident [25]. For melanoma , the diagnosis rate has increased almost threefold, from 7.9 to 21.5 per 100,000 [26]. Most of these are localized, in situ melanomas, and their rate of diagnosis closely mirrors population skin biopsy rates. Kidney cancer rates have doubled from 7.1 to 13.4 per 100,000, reflecting the widespread utilization of ultrasound and CT imaging. A number of recent series have confirmed the indolent behavior of many kidney cancers [27, 28]. A study of the growth rate of 53 solid renal tumors, in which each tumor had at least two CT volumetric measurements 3 months apart before nephrectomy, demonstrated their variable natural history and frequent indolence [29]. Twenty-one (40%) had a volumetric doubling time of more than 2 years and seven (14%) regressed. Furthermore, slow-growing tumors were more common in the elderly. Many renal tumors thus are overdiagnosed either because they do not grow at all or because their growth is too slow for the tumor to cause symptoms before the patient dies of other causes. In the absence of systematic screening for renal cancer, the increased rate of diagnosis is likely due to the increased use of abdominal imaging.

For both breast and prostate cancer , mortality rates have decreased despite the marked increase in diagnosis. Prostate cancer mortality in the USA has fallen by about 40% since 1993, from 38.6 to 24.6 per 100,000. A similar trend has been seen in breast cancer. This decrease has multiple causes. The two most probable are the effects of early detection and improved therapy. Thus, in these two cancers, early detection is likely producing both overdiagnosis and a mortality benefit.

This is a classic benefit-harm conundrum. In prostate cancer , there appears to be an undeniable benefit of early detection, reflected by a substantial and very clinically meaningful fall in mortality. This comes at the cost of many patients being treated for each one who benefits. This overtreatment problem is a major concern.

Overdiagnosis, along with the subsequent unnecessary treatment and associated risks, is a critically important adverse effect of early cancer detection. With false-positive screening test, the adverse effects of anxiety and additional tests are short term, until the absence of cancer is confirmed. In contrast, the impact of overdiagnosis is lifelong. A cancer diagnosis may influence patients’ sense of well-being, their physical and emotional health , their relationship with loved ones, and their ability to purchase health insurance.

Many have written eloquently about the medicalization of the healthy and the use of fear in overdiagnosis and overtreatment. “Today the kingdom of the well is being rapidly absorbed into the kingdom of the sick, as clinicians and health services busy themselves in ushering people across this important border in ever increasing numbers” [30]. The problem of overdiagnosis is a malady of modern medicine, not just oncology. Some argue that this problem is an inevitable but somewhat unforeseen consequence of well-meaning attempts to diagnose serious diseases at a point where they are more amenable to cure; others argue that it reflects vested medical and commercial interests in medicalizing the normal [31].

The risk of overdiagnosis and overtreatment makes informed decision-making more complex. Early treatment may help some but hurts others. This trade-off should be calculated by each individual patient based on a sophisticated understanding of the risks and benefits involved and insight into their own personal values and risk tolerance. The decision involves balancing many factors. This ideal is often not achieved.

Four strategies are warranted to improve this situation: (1) develop clinical and patient tools to support informed decisions about prevention, screening, biopsy, and treatment and offer treatments tailored to tumor biology; (2) focus on development and validation of markers that identify and differentiate significant- and minimal-risk cancers; (3) reduce treatment for minimal-risk disease; and (4) identify the highest-risk patients and target preventive interventions.

Patient education is a key solution to this problem. Patients should be adequately informed of the nature and the magnitude of the trade-offs involved. This kind of discussion is challenging for patients. Scientific illiteracy and lack of numeracy contribute to the challenge [32]. (Indeed, failure of most people to understand the nature and magnitude of risk is a major social issue and results in support for many inappropriate policies.) Patients must clearly understand the nature of the trade-off that although early treatment may offer the opportunity to reduce the risk of cancer death, it also can lead one to be treated for a “cancer” that is not destined to cause problems. These ideas are often foreign and must be presented clearly. The cancer “zeitgeist” referred to earlier in this chapter, i.e., that it is uniformly a lethal and aggressive disease, contributes to the challenge.

Quantifying overdiagnosis is often challenging. There are only a few randomized trials of prostate cancer screening and even fewer provide the needed long-term follow-up data. Nonetheless, “best guess” estimates about the magnitude of overdiagnosis are useful in decision-making. These estimates involve modeling the natural history of the cancer, the impact of early diagnosis, and competing mortality risks. It isn’t clear, for example, how patient preferences are influenced by whether the number needed to treat is 12 (Hugosson Scandinavian screening study) [33] or 48 (ERSPC) [19], for each prostate cancer death avoided. Simple and transparent models with explicit assumptions and input values can be instructive.

Overdiagnosis and overtreatment generate a cycle of positive feedback for more. As the disease is more widely diagnosed, more and more people have a connection to someone, whether a family member, friend, or celebrity, who “owes their life” to early cancer detection and treatment. This is the popularity paradox of screening : The more screening causes overdiagnosis, the more people feel they owe it their life and the more popular screening becomes [34]. The problem is compounded by media reports about the dramatic improvements in survival statistics, which may reflect nothing more than lead- and length-time effects.

Volume criteria can be used to identify candidates for conservative management. This is now widely accepted for small pulmonary nodules [35] and adrenal masses [36] detected incidentally. Identifying growth over time is another parameter that can reduce overtreatment. With lung cancer screening using CT, biopsies of small lesions are now restricted to those that grow over time [37].

Another solution is to relabel the disease with a term that doesn’t include words for cancer. This was done effectively for what was formerly grade 1 papillary transitional cell carcinoma of the bladder [38, 39] and is now termed PUNLMP or papillary urothelial neoplasia of low malignant potential . It has been proposed that small-volume, Gleason 6 prostate cancer be termed “IDLE” tumors (indolent lesions of epithelial origin) [40]. This would go a long way toward reducing the problem convincing patients with a “cancer” diagnosis to remain untreated. IDLE tumors would be managed as ASAP is currently with serial PSA and repeat biopsy. However, most pathologists believe that, since low-grade prostate cancer can demonstrate local invasion, it deserves to be labeled cancer. The new grade grouping of prostate cancer is a step in this direction. Gleason 6/10, implying an intermediate grade, will now be called Group 1, reinforcing the concept of a favorable lesion [41].

The problem of overdiagnosis and overtreatment goes beyond the prostate cancer field. As physicians, we have a responsibility to recognize the phenomenon, protect our patients from it where possible, and minimize the impact in other ways. These include developing a clear definition of where it exists; describing it in simple, easily accessible terms (i.e., “too much medicine”) [42]; recognizing the competing values and risks/benefits involved and developing strategies to account for these; and promoting public debate on the inherent uncertainty and limitations of health care and their implications for overdiagnosis.

Active surveillance, the focus of this book, is a major step forward in addressing this concern, not only in prostate cancer, but in many other human conditions.