Prostate Cancer: People Transforming A Diagnosis,A Diagnosis Transforming People

Abstract

Prostatic specific antigen (PSA) screening has been controversial since its inception. Controversy has persisted despite more and higher quality clinical evidence. Attention to lead and length time biases, overdiagnosis, overtreatment, medicalisation, iatrogenesis and financial conflict of interest has had limited impact. I undertook a social history of the prostate cancer diagnosis to reassess the causes of controversy and suggest different clinical and policy responses.

For much of the twentieth century, clinicians were uninterested in early detection and radical treatment, believing that cancers revealed after obstruction-relieving surgery or autopsy could be ignored. In 1985, the FDA approved PSA diagnostic tests, which rapidly catalysed two self-reinforcing cycles of action and perception. One occurred when the increased diagnoses made the disease more prevalent and feared, and efforts at prevention and treatment seem more efficacious, leading to more screening, and so on. The other cycle occurred among men with screening detected cancer who initially eschewed radical treatments or imagined doing so, whose lives were often consumed with fear and surveillance, increasing demand for radical cures.

This history underscores the need for novel clinical and policy responses to the looping effects—self-reinforcing cycles of action and perception—which can radically transform so much of what we believe and do about disease.

Keywords

Prostate cancer risk PSA cancer screening overdiagnosis looping effects

The Diagnosis Today

Imagine a frightening disease that is making men impotent and incontinent. It has attacked over a hundred thousand, previously healthy, middle aged and older American men yearly since the mid-1980s. It does not appear to spread from person to person and its effects are limited to a few affluent countries. Careful epidemiological study within the USA has revealed that affected men have high levels of prostatic serum antigen (PSA), a blood test ordered by their doctors, and subsequently had prostate biopsies that showed pathological evidence of cancer. Acutely, the disease leads to hospitalisation for intensive and painful urological interventions, which in some cases results in urinary infection and deep venous thrombosis. Nearly every stricken individual has experienced long-term impotence and short-term incontinence. There is a silver lining, however. The disease’s permanent damage to the prostate gland has saved a few men from death by prostate cancer.

This scenario is not imaginary but reverses the way we understand intervention and disease. Looked at this way, American men have been experiencing a problematic epidemic of invasive treatments triggered by prostate cancer screening. It is possible to reverse figure and ground because there is little consensus about whether efforts to prevent and treat prostate cancer have caused more harm than good. Results from clinical trials comparing screened to unscreened populations did not appear until 2009—one showed no net benefit (Andriole et al., 2009) and the other raised serious questions about whether the harm of cancer treatments to many was worth the benefit to the few whose death from cancer was prevented (Schröder et al., 2009). Although the US Preventative Services Task Force, an authoritative body on prevention, subsequently issued recommendations against screening, controversy has continued, and remains unresolved despite longer-term follow-up from these trials and other evidence.

Where one stands on this controversy seems to depend on what one believes about the natural history of prostate cancer—the disease’s clinical course in the absence of intervention. How likely is untreated, screening detected cancer to progress and do harm? Just what prostate cancer ‘is’ and what the diagnosis ‘means’ have for the past century been the central unresolved questions.

Unravelling how these questions have been asked and answered in the USA throughout the past century is the subject of this essay. My central argument is that everything clinicians and patients know, or think they know, about the disease is entangled with our interventions to prevent, diagnose and treat it. Making sense of these entanglements is crucial to resolving current screening and treatment controversies. It also represents an opportunity to understand diagnostic transformations more generally in an increasingly intervened-in world.

I explain the historical transformation of the prostate cancer diagnosis by considering it as a five act drama, starting with Act I, spanning the first half of the twentieth century, a period in which clinicians were pessimistic about and largely uninterested in prostate cancer.^{In Act II,} I give a brief overview of a mid-century failed attempt to jump start a screen and radical treatment paradigm. Act III, roughly spanning from 1960 to 1985, was a period in which a diffuse set of actors, frustrated with lack of progress in prostate cancer but focused on smaller problems, tinkered with different diagnostic and therapeutic technologies. Act IV covers the approval and introduction of the PSA test in 1985: an event test developers and regulators thought of as a minor, incremental change but which ended up catalysing a rapid transformation of the diagnosis and disease.

Act V has two scenes, both of which started with PSA screening and continue through the present. I describe two self-reinforcing cycles of action and perception, what philosopher Ian Hacking (1995) has characterised as the looping effects of human kinds. One scene occurred at the level of American society: the growing numbers of people diagnosed with prostate cancer transformed the meaning of prostate cancer diagnosis, which in turn impacted the beliefs and actions of clinicians, patients and broad swaths of the population, which further increased the numbers of people diagnosed. The other scene occurred in the lives of people with screening detected cancer who initially decided to live with the uncertainty of the diagnosis and among the much larger group of clinicians and patients who imagined this choice. Many of these individuals have experienced a dizzying array of diagnostic tests and routines, which in sum have led to a life at risk, which I argue has created its own demand for radical interventions that promise a cure for the experience of uncertainty and fear (Aronowitz, 2015).

ACT I: Pessimism and Inaction, 1904–1950

The prostate cancer diagnosis in the early twentieth century was attached to three different clinical situations. The ‘prototypical prostate cancer’ diagnosis was given to men presenting with pain and other evidence of severe disease and who were found by X-ray and clinical examination to have prostate cancer metastatic to bone and/or other organs. There were no cures for such tragic patients. Accordingly, urologists devoted little space in clinical journals and textbooks to prostate cancer. They were busy, however, doing prostatectomies (surgical excision: partial or complete), for another reason—to reduce or relieve the obstruction caused by large glands, which later in the century was called ‘benign prostatic hypertrophy,’ emphasising that this enlargement was not due to cancer.

Surgical specimens from such obstruction relieving surgery were sometimes sent to the pathology laboratory. This led to a second, incidentally diagnosed prostate cancer, as pathologists frequently identified clusters of cancer-appearing cells in the apparently benign specimens. Because this type of cancer was only incidentally found and not associated with any other symptoms or signs and occurred much more commonly than prototypical clinical prostate cancer, urologists generally thought this type of cancer was a separate, if not problematic, entity. It was often ignored or not mentioned to patients and typically triggered no clinical response. Such cancer was often called ‘latent cancer’. As it was typically contained within the excised prostatic tissue removed by surgery, it was essentially ‘cured’ by the obstruction relieving surgery through which it had also been discovered. Later in the century, especially after Rich’s (1935) studies at Johns Hopkins University, a similar latent or incidental cancer was noted in a high percentage of men who underwent routine autopsies after death for other reasons besides prostate cancer; as high as 30 per cent in some early series, with rates increasing as men aged.

The third type of prostate cancer—‘curable cancer’—was a virtual diagnosis for most of the first three quarters of the twentieth century. Extrapolating from the high prevalence found incidentally by prostatectomy and autopsy, a small number of pathologists and clinicians wondered about the potential harms of cancer that must exist within many men’s bodies, especially as men were living longer and thus had more time for it to evolve into something dangerous. Both lay and medical people generally believed that the prototypical cancer had to start somewhere. Fuelling disinterest and inaction, however, was the fact that there remained no safe and reliable way to diagnose curable cancer. However, this fact itself calls for explanation, as there was almost no clinical or laboratory research or calls for research into new diagnostic tests to detect cancer early or even into disease pathogenesis.

There was some interest in curative surgery for men, even though urologists almost never encountered cancer diagnosed ‘early enough’ on clinical grounds to be cured. In 1905, Johns Hopkins University’s Hugh Hampton Young published a case series of prostate cancer patients who underwent radical perineal prostatectomies for cure, an operation he developed. Young had founded Hopkins’ storied urology programme, that would soon become the Brady Urological Institute after a generous gift from James Buchanan Brady—a larger than life Gilded age businessman and gambler. Over the next five decades, Young and his associates were the leading proponents of this approach in the USA. Yet only a small number of men, even at the Hopkins—the epicentre of prostate cancer surgical cures—were diagnosed with prostate cancer and treated with radical prostatectomy for cure during this period.

In my own archival work with the extant records of cases in which radical prostatectomy for cure of prostate cancer was attempted (Chesney Archives),¹ Young and his associates could sometime palpate very hard nodules in rectal exams among men who were symptomatic from urinary obstruction and also suspect cancer from physical findings of local extension, for example, swollen testicles or seminal vesicles. Since these men, even though they had no clinical evidence of bone or other distant metastasis, were still believed to likely be harbouring metastatic disease, the operation for cure was infrequently attempted even at Hopkins. For many other leading urologists, there was disbelief that cures could ever be achieved by surgery (Keyes, 1917). Or as one urologist put it,

once the diagnosis of carcinoma of the prostate has been made it is technically impossible to do any sort of surgical procedure that would effect a cure. Once the diagnosis of carcinoma of the prostate has been definitely made, the condition is no longer surgical.

(cited by Smith, 1942, p. 1514)

In comparative cancer terms, especially in relation to female breast and cervical cancer—which were subject to a ‘do not delay campaign’ starting in the early twentieth century (Aronowitz, 2001)—it is noteworthy that there was essentially no research into or promotion of any type prostate cancer early detection until much later in the century.² Young’s radical perineal operation for cure had been developed and carried out with the assistance of William Halsted, who was able to develop and promote an analogous operation for cure of breast cancer, and change the minds and practices of a large number of other physicians and lay people. Why did scepticism prevail in prostate cancer but not in breast (and cervical) cancer?

Male anatomy and the nature of prostate cancer itself played some role. The prostate gland is small and not capable of being self-examined, although induration and hard nodules could be palpated by a physician’s rectal exam. Prostate cancer itself can be multifocal and diffuse, so it was not possible to do less mutilating surgery such as lumpectomy for breast cancer, done much later in the century, or to extirpate it with relatively non-mutilating procedures such as was the case for cervical cancer, where gynaecologists could deploy less-invasive colposcopy or the well accepted hysterectomy.

Besides these purely anatomic and biological details, and technical differences in available tools, male urologists plausibly had a much higher threshold for doing mutilating surgery (reliably leading to incontinence and impotence) of uncertain efficacy on other males. Urologists may also have had less fire in their belly—financial or technical interest—to promote radical prostatectomies for cure as they were busy enough doing similar operations for obstruction. While calls for physician rectal examinations to screen for prostate cancer were sometimes made, purposeful early detection was rare in practice and hardly ever the subject of lay or medical concern until the 1970s and 1980s (Catalona & Scott, 1978).

ACT II: Bowery Interlude, 1951–1966

Elsewhere I have written about the ‘Bowery series’, a set of practices on the border of research and innovative therapeutics, led by urologist Perry Hudson (Aronowitz, 2014). Hudson observed the pessimism and inaction about detecting and curing prostate cancer at Hopkins, where he had done his urological training. Put in charge of urology at Delafield hospital, a newly opened public cancer hospital in New York City affiliated with Columbia University, he decided to try and break the inertia with an aggressive attempt at detecting and treating curable cancer. Starting in 1951 and continuing until 1966, Hudson and his colleagues recruited more than 1,200 homeless, alcoholic men from New York City’s skid row—the Bowery—and brought them to Delafield. These men were subjected to many invasive tests and procedures, culminating in open perineal biopsy of the prostate gland—a non-trivial, impotence-causing surgical procedure in which a 2 × 2 cm piece of prostatic tissue was removed, frozen and sent to the pathology laboratory for examination. If this tissue was positive for cancer, the affected man underwent immediate radical prostatectomy and surgical castration, followed later by a course of hormonal treatment. These poorly informed, vulnerable men participated in these practices because they were often told they had a prostate problem that could be fixed, were promised free medical care, and were offered a few days respite from the ravages of homelessness, hunger and alcoholism with a stay in the hospital. Although some kind of consent may have been obtained, these studies were conducted on poor, helpless men because investigators would and could not do these experiments on people with more autonomy, power and dignity, such as the paying private patients at nearby Columbia Presbyterian Hospital.

Hudson believed that his fellow urologists’ scepticism and inaction towards prostate cancer needed to be overcome by a demonstration that curable cancer could be found early and among asymptomatic men—something akin to the way breast cancer could be detected early by self and/or clinical examination or cervical cancer by colposcopy and the newly introduced PAP smear. He wanted to transform the ignored latent cancer into curable cancer, something diagnosable by intention in living patients and radically treated. Hudson’s research-within-practice, however, was too poorly designed to demonstrate that this proto-screening followed by radical interventions could decrease mortality.

Hudson’s ambitions illustrate the frustrating—for someone determined to make advances against the horrible human toll of suffering caused by prostate cancer—state of affairs at mid-century. While this unethical and problematic research was published in leading medical journals, cited frequently in the medical literature, and was the subject of popular news coverage, they were ultimately forgotten and had minimal direct impact on subsequent clinical developments. Yet the studies’ history is significant because it provides a provocative and illuminating perspective with which to view subsequent events. The ‘Bowery series’ was a prescient attempt to combine a set of existing practices for diagnosing and treating prostate cancer into a new early detection and radical treatment paradigm.

ACT III: Tinkering, 1960s–1985

The history of the Bowery series shows that through the mid-1960s there were substantial obstacles to starting an early detection and radical treatment paradigm for prostate cancer. In the ensuing decades, researchers, device makers and clinicians would tinker in a largely uncoordinated way with different diagnostic and therapeutic practices and devices that in sum led to more favourable conditions for such a paradigm to take off. These actors gradually found more palatable ways to do prostatic biopsies and cancer surgeries, and visualise the prostate gland. While this tinkering was done within the routine care of patients suspected or diagnosed with prostate cancer, and not for screening, it laid the groundwork for the later transformation of the previously ignored latent cancer into something worth pursuing intentionally and for the affectual and attitudinal change Hudson pursued—to replace medical pessimism and inaction with an optimism and enthusiasm for intervening.

One minor front was the promotion of rectal exams for detection of treatable prostate cancer. A 1980 study of different ways of detecting prostate cancer among men with obstructive symptoms, described the subjects as ‘clinical material,’ and like the Bowery studies, subjected them to invasive diagnostics (Guinan et al., 1980). The investigators compared rectal exams, prostatic acid phosphatase (PAP) blood tests (discussed in detail below) and other ways of potentially diagnosing prostate cancer. In order to accurately assess and compare the sensitivity and specificity of these tools, the clinician investigators needed a ‘gold standard’ for comparison. Each of the 300 men, all patients at Chicago’s Cook County’s urological service, underwent prostatic needle biopsy. While significantly less mutilating than the open perineal biopsy Hudson deployed, it is difficult to imagine men other than poor, uninsured patients at a large public hospital agreeing to get a needle biopsy (the study did not detail who these men were nor whether consent was obtained). A relatively high 23 per cent who otherwise had no reason to suspect prostate cancer turned out to have the disease. The study’s major finding was that rectal exams had the best combination of sensitivity and specificity when compared to the ‘gold standard’ needle biopsy.

The early 1980s witnessed a flurry of other research and demonstration projects for mass rectal exams to screen for prostate cancer (e.g., Chodak & Schoenberg, 1984). Unlike the Chicago study’s conclusions, most studies found that screening for prostate cancer with rectal exams did not have enough sensitivity or specificity for diagnosing prostate cancer to be used as a mass screening tool (e.g., Thompson et al., 1984).

There was considerable interest in tweaking the technical characteristics of the PAP test. PAP had been used for diagnosis and to track disease progression starting in the 1930s by some of Hudson’s Columbia colleagues (Gutman et al., 1936). PAP was understood to be a marker of tumour volume and metastatic spread. But the test was not sensitive enough to reliably distinguish men with early cancer from unaffected men (there were many false negatives), nor specific enough to distinguish prostate cancer from benign hypertrophy and a host of other, and often non-specific, conditions which led to high serum levels (there were many false positives).

In the 1980s, PAP was tweaked, rebranded and promoted for diagnosis and screening. Different radio-immunoassays, monoclonal and polyclonal antibody tests, and commercial kits were tested and marketed. Most notably, Hybritech, one of the first and most important biotech companies, which would eventually develop PSA tests, used then new hybridoma technology to make and market a more accurate PAP test. In some consumer advertising, this blood test began to be called the ‘male PAP test’, linking the common abbreviation for the test with the heavily promoted Papanikolaou screening cytological one for cervical cancer.

One critical New England Journal of Medicine (NEJM) commentary (Watson & Tang, 1980) called attention to the evidence-challenged promotion of PAP screening. It critically cited a New York Times advertisement by a local hospital promoting PAP screening, a National Prostate Cancer Project announcement that new PAP assays ‘herald the era of routine prostate cancer screening in the near future’, and a trade magazine’s front page headline that ‘New Prostate Ca Test recommended as Screen.’ The commentators then reviewed data on the new immunoassay’s sensitivity and specificity and computed predictive values of the test with estimates of the disease’s prevalence and found no evidence to support screening either adult males or men presumed at high risk. The authors somewhat cheekily wrote that ‘blind biopsies’ would have superior sensitivity among such men (Watson & Tang, 1980); I return to this ‘modest proposal,’ essentially Hudson’s practice, later.

An important catalyst to increased demand for a screen and early detection paradigm in prostate cancer was the dramatic increase, starting in the early 1970s, in the number of men who underwent transurethral resection of the prostate (TURP)—basically the removal of obstructing prostatic tissue using instruments to visualise and cut tissue through a catheter inserted through the urethra. Although TURPs were commonly done since the 1930s, the development of rod-lens telescopes and improved endoscopic illumination made them much more common in the 1970s. As they required no pelvic or abdominal incision, TURPs were more popular to many patients and physicians than open surgery (Merrill et al., 1999).

The increased number of TURPs meant there were more pathological examinations of removed tissues, which led to much greater numbers of latent cancer diagnoses. The resulting increased incidence led to more awareness of the magnitude of prostate cancer in men as they age, creating some pressure for a prevention paradigm, although the increasing numbers of incidentally diagnosed cancer did not by itself resolve the uncertain meaning of such diagnoses. Moreover, the resulting increased incidence necessarily led to improved positive predictive values for proto-screening tests like the male PAP (and later PSA). This is a Bayesian statistical principle: screening tests have better predictive value at the same sensitivity and specificity as the prevalence of the underlying disease rises. This rising tide would have helped the utility of any diagnostic or screening test.

The plausibility of detecting and curing prostate cancer also increased with the modifications to cancer surgery labelled and promoted as ‘nerve sparing’ by Patrick Walsh and colleagues in the early 1980s (Walsh et al., 1983). One Georgetown urologist was quoted in a 1986 Washington Post report: ‘Before Walsh, men were choosing not to have the prostatectomy because of the impotency problem and the fear of incontinence’ (Thompson, 1987, p. 13). The label ‘nerve sparing’ alone made the operation less frightening to patients and clinicians alike.

New technologies were also developed for visualising the prostate gland. One of them, trans-rectal ultrasonography (TRUS) was used for the diagnosis of prostate cancer abnormalities. There was some tinkering with its possible use for screening, but it did not find its niche until after the introduction of PSA screening, when it was used to better visualise the prostate gland for both abnormalities to biopsy and to systematically sample the gland with multiple biopsies. TRUS was accompanied by needle biopsies, as discussed above, and later by the spring-loaded biopsy ‘gun,’ which allowed easy, fast (and thus reducing pain) and safer tissue biopsies (Applewhite et al., 2001). Also important was the new and more widely accepted staging system for prostate cancer, the Gleason score. This system was developed in the 1960s and 1970s to have standardised classification of prostate cancer in cooperative clinical trials of hormone treatments, but was later assimilated into routine practice (Chen & Zhou, 2016).

Overall, the many actors and groups who did this tinkering by the mid-1980s shared a goal of improving prostate cancer diagnosis and treatment, although there were some fitful attempts to promote screening PAP tests and rectal examinations. The situation would change with the development, diffusion and widespread use of a new way of detecting early disease.

ACT IV: The Non-event of PSA Development and Regulatory Approval

The PSA blood test to screen American men for prostate cancer, followed by aggressive curative treatments, catalysed the transformation of the disease and diagnosis, and the lives of a large fraction of American men. However, PSA’s initial development and regulatory approval were largely under the radar events.

Attributing priority and credit for discovering and developing the PSA test has been controversial. Immunologist Richard Ablin, who has emerged as a fierce critic of PSA screening, believes he and colleagues have priority from their early 1970s research. But in the initial FDA regulatory hearings on PSA, a Hybritech clinical director gave a history in which ‘PSA was first identified in 1979 by Wang and co-workers at Roswell Park Institute,’ ignoring Ablin and co-workers (Immunology Device Panel, 1985, p. 7).

Ablin has acknowledged that it was researchers at the Roswell Park Institute, starting in 1979, who purified and characterised PSA in greater detail, and with the help of PCR (polymerase chain reaction) amplification, were able to develop PSA into a clinically useful blood test (Ablin & Piana, 2014). Roswell investigators received a patent for their PSA immunoassay in 1984. Believing that PSA had superior accuracy to PAP for identifying prostate cancer, this patent attracted the attention of Hybritech, who as noted earlier, was already trying to commercialise PAP tests. Hybritech was mostly interested in developing cancer therapeutics but as it was costly and time consuming to do so, fraught with possibilities of failure and difficult to get through the FDA regulatory process, the company’s initial business strategy was to first use their expertise in monoclonal antibody technology to develop commercially viable diagnostic tests (Jones, 2005).

Hybritech sought approval for their PSA diagnostic kit from the FDA device regulators in 1985 for monitoring patients with known diagnoses of prostate cancer, especially to determine whether the disease had spread. To make the argument for PSA’s clinical utility, urologist Paul Lange testifying for Hybritech before the Immunological Device Panel (1985) of the FDA, argued that clinicians had already used PAP tests to follow patients for worsening disease or reassurance and that PSA was a more accurate way to follow the disease. However, Lange did not have any head-to-head comparative evidence for this claim, as was noted with disapproval by FDA regulators. Instead, he presented separate, retrospective evidence of PSA’s modest superiority over PAP for following the course of prostate cancer. However even these weak data were poorly analysed. When pressed about the sensitivity and specificity of PSA for detecting cancer (a crucial determination, as PSA and PAP are substances present in prostate glands: benign and malignant), Lange conceded that ‘I didn’t show sensitivity. The sensitivity—I didn’t put that slide in because I got all mixed up between sensitivity and specificity, predictability and all this’ (Immunological Device Panel, 1985, p. 64).

Lange anticipated that regulators would be sceptical of approving it. Both PAP (which had diffused in an era where there was essentially no regulation of devices) and PSA gave clinicians a signal of disease progression, but there was no evidence that treating asymptomatic recurrences was superior to waiting until symptoms appeared, and overall there were no effective treatments against metastatic prostate cancer. So, Lange asked rhetorically and defensively, ‘why bother doing any of this?’ Lange answered that in the real, as opposed to the rarefied world of evidence evaluation and policymaking, patients wanted to know the status of their cancer even if it was not strictly necessary for therapeutic decisions. ‘Every urologist who treats cancer,’ Lange noted, ‘gets acid phosphatase levels and uses that information in a variety of ways. And we can debate this all day, but I can just tell you that the patients need to know, particularly the men’ (Immunological Device Panel, 1985, p. 75)

FDA regulators were disappointed that there was no evidence that the PSA test had clinically significant impact such as reducing morbidity, or saving lives or money. When pressed about doing a prospective trial with such ‘hard’ endpoints by Dr Ladoulis, one of the FDA panel members, Lange called this question ‘political’ and pointed out that PAP was not evaluated by such high evidentiary standards and was already in wide use. ‘Well, we cannot do anything about PAP now,’ responded Ladoulis. ‘That is the past’ (Immunological Device Panel, 1985, p. 79).

The potential large market for using PSA as a screening test may very well have been anticipated by Hybritech, especially given the efforts by others to fashion PAP for such purposes, but not mentioned at the hearings because it might have triggered closer FDA scrutiny. While Lange mentioned that clinicians sometimes used PAP to screen for cancer, none of the panel members seemed to worry about this potential for PSA (Immunology Device Panel, 1985). In hindsight, given the almost immediate application of PSA to screening after regulatory approval, this lack of concern is puzzling.

Despite Lange’s fuzzy performance and panellists’ doubts, the FDA approved the new PSA test for monitoring the disease. Soon after, later in 1985, the PSA blood test became commercially available and was immediately used along with the other practices and technologies developed in the preceding two decades (ultrasounds, needle biopsies, and soon biopsy guns and nerve sparing surgery) to make more cancer diagnoses—including doing the test on men without cancer symptoms and signs. ‘I would advocate that once a man hits 50 he not only have a digital rectal exam but that he have a baseline ultrasound and a PSA,’ observed Michigan radiologist Fred Lee in 1987 ‘and that if a biopsy is needed, the new spring-loaded biopsy gun is painless … no anesthesia, no catheter, and far superior to the older biopsy techniques which had significant complication rates and were inaccurate’ (Thompson, 1987, p. 14). Lee was not alone. An increasing number of asymptomatic men had their PSA-detected cancer treated with surgery or radiation, a sharp break from the inaction that had long been the norm for cancers believed to be incidental or latent. Curable cancer was no longer a virtual diagnosis.

ACT V: PSA Ignites Two Looping Cycles

Scene 1: Looping Effects Mediate the Transformation of a Diagnosis

The spread of PSA testing after FDA regulatory approval was quick and massive. Among men between 65 and 74, from 1988 to 1991, the rate of PSA testing went from 1,430 to 18,000 per 100,000 men: from 1986 to 1991, the rate of prostatic biopsies nearly quadrupled (Potosky et al., 1995). According to the Surveillance, Epidemiology and End Results (SEER) databases, the incidence of prostate cancer doubled and the rate of radical prostatectomy among white men quadrupled between 1986 and 1992; changes universally attributed to PSA testing (Stanford, 1999).

These early rapid changes were not driven by top-down public health messages or mass media publicity about PSA testing. A search of the New York Times in ProQuest with ‘PSA’ or ‘prostatic specific antigen’ as the search terms yielded only seven articles for the entire 1970–1990 period, some of which had little to do with the use of the PSA test.

Rather than being a top-down change, there appears to have been a self-reinforcing interaction between clinical decisions made by, at first small numbers of clinicians and patients, and perceptions of the efficacy of screening and radical treatments by an ever growing, larger group. PSA screening and diagnostic use, even on a small scale, led to more prostate cancer diagnoses, radical treatment and an increase in the numbers of men who seemed to have survivable disease. Improved case fatality and the related survival rate necessarily follows from greater numbers of diagnoses because these are ratios of the number of cases to mortality from disease (assuming, for the sake of argument, no direct biological impact from surgery or radiation). Even small increases in disease prevalence and apparent greater survival could plausibly drive more men to screening and to the biopsies and interventions that follow, leading to more awareness of prostate cancer and the apparently new survivability with screening and treatment, leading to more PSA use and even better survival rates, and so on.

This rapid transformation shows the centrality of looping effects in the construction of disease diagnosis, prognosis, treatment and epidemiology. PSA screening catalysed mass behavioural change and led to many men’s reclassification as ‘cancer patients.’ Prostate and other cancers are thus not stable ontologies, but are, as Hacking observes, ‘moving targets because our investigations interact with them, and change them’ (Hacking, 2006, p. 23).

Some later observers of the rapid rise of screening and treatment in prostate cancer labelled this interactive process the ‘popularity paradox,’ and used it to explain individual enthusiasm for screening. ‘The greater the harm through overdiagnosis and overtreatment from screening,’ observed Raffle and Grey (2007, p. 68), ‘the more people there are who believe they owe their health, or even their life, to the programme.’ A clear articulation of this belief system at the level of the individual was offered by ex-NYC mayor Rudy Giuliani (2007), after he was diagnosed and treated for prostate cancer. In a television interview, he expressed extreme satisfaction with the USA health care system compared to socialistic ones, noting that the survival rates for his cancer were so much higher in the USA than in Britain or Cuba. While technically true, better survival in the USA could be explained by more screening alone.

In 1989, my then 70-year-old uncle told me that he had been diagnosed with a low-grade prostate cancer after a high PSA level. When I asked him why he got the PSA test, he told me that he was influenced by the experience of a close friend who had recently been through the screening-biopsy-treatment routine. He believed that the PSA test had uncovered a silent cancer in his friend that would have likely killed him had it not been surgically removed. This led my uncle to ask his internist for a PSA test. My uncle ultimately chose radiation therapy and died decades later of other causes. Whether a good or bad idea, my uncle’s experience became part of the growing cancer prevalence and improved survival story, influencing others.

Other factors were also at work. New York Times medical reporter Gina Kolata, in a 1993 account of the transformations then underway, emphasised the marketing of the PSA test to doctors by test makers. Leading clinicians were quoted observing the role of vested interests in drumming up demand for PSA screening, especially the ‘annual Prostate Cancer Awareness Week, paid for by the makers of drugs to treat the cancer and by makers of the PSA test and the numerous celebrity advertisements paid for by ‘TAP Pharmaceuticals of Deerfield, Ill., a maker of a prostate cancer treatment.’ Kolata also wrote about the role played by the American Cancer Society (ACS), which legitimated screening by issuing guidelines. Defending this legitimation despite the lack of evidence, an ACS spokesman made a realpolitik argument—‘men were having the test anyway’ (Kolata, 1993b).

Later in the 1990s, test makers, professional organisations like the ACS and urologists added ethnic/racial disparities and gender imbalance to the argument for PSA screening and radical treatment. The greater prevalence and mortality among African Americans was used to argue for more research, screening and ‘prostate cancer awareness’ to redress this disparity. The lack of grass roots disease advocacy by men, unlike the situation in breast cancer, was linked to putative male traits of denial and lack of attention to self-care as well as to fears of impotence. Such deficits were harnessed as an argument for greater prostate cancer awareness and screening, leading to the mobilisation of prominent male celebrities, such as General Norman Schwarzkopf (Crawford, 1997).

While it was not unnoticed that the screen and radical treatment paradigm widely diffused without any direct clinical trial evidence comparing screening to no screening, these objections seemed to have little effect. Nor did this evidence-challenged, rapid diffusion evoke much ethical revulsion, as there is, retrospectively at least, for the previously described Bowery practices. Why? One answer is that the tinkering in Act III made the steps leading to radical treatment much more palatable; the new PSA screening programme identified men at high risk of cancer, or at least appeared to, in a rational and intentional manner; and a broad swath of the American male population, informed by their doctors about what was happening and operating with similar knowledge and assumptions, were now involved rather than a small, poorly informed and vulnerable population of homeless alcoholics. The absence of significant ethical qualms was also sustained by there being no single actor or set of actors in control of all the tinkering in diagnostic technology and practices that led up to this development. In addition, PSA seemed to many actors, just an incremental improvement over PAP testing rather than ushering in a brave new world.

American men and their doctors generally shared the belief that those men selected for biopsy because of high PSA levels were facing high risks of invasive cancer and death. So it was only logical to try to detect and treat such curable cancer. Fears of not taking advantage of the chance to prevent cancer trumped contemporary voices who believed there should be evidence of the efficacy of screening before it was put into widespread use. ‘If they could spend an afternoon with me,’ prominent urologist William Catalona told Kolata (1993a) about the paradigm’s critics, ‘having to tell a patient who’s relapsed after hormonal therapy that there’s nothing else we can do because we detected his cancer too late, they would have a different view.’ Catalona argued there was more danger in inaction than screening and treatment during the interval in which evidence about efficacy and safety of the whole paradigm was not yet available.

Let’s say we don’t do PSA screening for 10 or 15 years and let’s say that after 15 years we find it was a good thing to do. What about the 500,000 men who died of prostate cancer in that time? This is my generation of men, too.

Catalona told Kolata he was getting ‘monthly’ PSA tests.

Within a few years after widespread PSA screening was introduced, improvement in prostate cancer mortality were noted and marshalled to support the early detection and radical treatment paradigm. Some mortality improvement, however, was to be expected as a direct result of more prostatectomies: fewer men with prostate glands necessarily meant fewer cancers. Some improvement might also have been an artefact of the greater attribution of deaths due to prostate cancer in the immediate pre-PSA era due to rising prevalence from TURP, TRUS, etc., which established a ‘high’ baseline from which to measure later historical drop in mortality rates. But a crucial issue, even if one believed declining prostate cancer mortality was due partly or largely to screening, was whether the costs and side effects of treatment to many people were worth the much fewer number of lives saved.

Whatever one believes about its efficacy and safety, screening changed the meaning of ‘latent’ cancers, just as Hudson had hoped. Urologists came to believe that the PSA test was a marker for cancer risk, making PSA-detected cancers categorically different from the incidental or latent cancers of the same grade long known to them from pathological examination after open prostatectomies, TURPs, or autopsy. The new PSA-detected cancer was intentionally sought. It was the target of screening. This change from incidental/latent to purposeful had little to do with what PSA was biochemically or functionally, but more from the belief that the PSA test was a rational way to diagnose prostate cancer and its logical role in the screen, biopsy, diagnosis and cure sequence. This sequence, and the belief in its rationality and efficacy, shifted urological pessimism and inaction. Even the historical existence of the older latent or incidental cancer was rarely recalled in an era marked by widespread enthusiasm, rationality, interlocked routines and economic rewards.

Urologists also benefited from being outside the decision-making realm of the most controversial aspect of the new paradigm, initiating screening, which usually occurred in primary care settings. Patients most often sought out urologists once they had a high PSA level. This was a moment, as well as the one after biopsy results were available, for urologists to apply their clinical expertise. And with so many more men appearing with high PSA levels and submitting to biopsies and later treatment, this expertise was growing. PSA, TRUS and spring-loaded biopsies, and the promise of nerve sparing surgery were not only new, attractive tools which rationalised screening and constituted an early detection techno-system that Valier (2016, p. 135) aptly describes as ‘mutually defining, and ultimately mutually vindicating.’ They also gave urologists financially rewarding ways to use their specialised expertise.

Scene 2: Embodied Risk, Fear, Work and Definitive Cures

If men with high PSAs consented to biopsies, urologists typically took multiple ones with the self-loading ‘gun,’ sometimes directed by suspicious findings on ultrasound but often blindly. If cancer was found, it was given a Gleason score, which enabled some standardisation of the often complex and variable findings in prostatic biopsies. Pathologists graded the often multiple and variable areas of cancer within the biopsy specimen on a 1 to 5 scale, from benign to very cancerous. The Gleason score is the sum of the pathological impression of the most common pattern and the second most common one. Scores of 6–7 (e.g., a score of 7 might be the sum of one area of ‘3’ and another of ‘4’) have been the most frequent and are generally believed to signal a good, if uncertain, prognosis.

Such men, numbering over a hundred thousand Americans yearly in the post-PSA era, have generally been told that they could have their prostate glands surgically removed or irradiated, or could be watched for some period. We do not know how such difficult decisions were made, but I want to suggest that these decisions have often been part of another dynamic process—one that parallels at the individual level what was just described as a looping effect at the population level. Many men with PSA-detected cancer have experienced a new embodied risk state, characterised by intense clinical surveillance—both experienced and anticipated—which has triggered fear, uncertainty and hassle, creating its own demand for ‘cure’ by surgery or radiation. This state was also anticipated in decisions to go directly to radical treatments. The increasing numbers of clinicians and patients who made such decisions helped make aggressive treatment for screening detected cancer the new normal. A similar embodied risk state and dynamic, most certainly characterises the many men who NEJM had high PSA levels but initially eschewed biopsy, but as I have less evidence for their decision-making, will not explore it in any detail.

To understand how this dynamic process works, I analysed a online poll of readers’ responses to a clinical vignette of a 63-year-old man with a rising PSA level (Schröder et al., 2008). Subsequent biopsies showed good prognosis prostate cancer (Gleason score 6) in two of twelve samples. Three experts gave their rationale for one of three courses of action—expectant management (EM), surgery, or radiotherapy. Of the 3,720 votes cast, 29 per cent were for EM, 33 per cent for radiotherapy and 39 per cent for radical prostatectomy. The NEJM also posted 195 of the 218 written comments they received explaining respondent reasons for choosing one therapy over another.³ The majority of these comments were from US-based physicians, but a sizeable number were from physicians outside the USA and lay people, many of whom had been treated for prostate cancer.

The vignette curiously did not mention why the patient initially had a PSA test. Was the man screened or did he have some symptom or clinician discovered sign? Presumably the authors believed this detail was irrelevant to the decision or simply assumed that the patient’s first PSA test was for screening. However the prognosis and meaning of a cancer at a certain stage and grade is not independent of the circumstances of its discovery: a phenomenon captured by the notion of length time bias but also by historical analysis of the performative impact of changed definitions of cancer, and norms and means of diagnosis on prognosis, as discussed in scene 1 and elsewhere (Aronowitz, 2007). So this absent detail itself reflects a bias by the vignette framers, in which a ‘cancer is a cancer’ regardless of how it was diagnosed; a belief that favours active treatment, especially since evidence from the pre-PSA era suggested the efficacy of radical treatment for low-grade cancers. The summary document of the NEJM survey similarly revealed the polltakers’ bias towards active treatments. ‘Most voters for expectant management were actually voting for a delay before invasive treatment, not for years of such management,’ they observed. ‘Ultimately, most voters in favour of expectant management would have to choose a treatment’ (Schwartz, 2009, E4). But many respondents explicitly advocated EM as a long-term strategy, not just a pause.

Very few respondents cited evidence from clinical studies to support their choice. Some EM supporters pointed out there was no evidence of the efficacy and safety of surgery and radiotherapy from robust clinical trials, while supporters of these active treatments often argued that a pathological diagnosis of cancer meant inevitable harm and death unless the cancer was taken out prior to the point of metastasis. ‘No point in waiting,’ proclaimed one survey respondent. ‘It will progress. Do the definitive surgery now when chance of cure is highest, and when in good health’ (F85F37).⁴

Respondents often articulated that their own choice of therapy was motivated by their knowledge of the preferences or biases of others in the medical profession. One respondent noted that EM is not ultimately a viable long-term option because ‘urologists are totally uninterested in watchful waiting’ (AF0EB3). For a few respondents, the most important bit of pragmatic knowledge shaping their choice concerned the economic influences on practitioners and others who promote this or that option. From this perspective, the most efficacious option was the one least tainted with conflict of interest. One respondent pointed to the money being made in Da Vinci surgery (robotic surgical removal of the prostate gland). ‘Personal experience with robotic prostatectomy: Incontinence, impotence, other problems much more common than reported in the literature. A big business for the manufacturer of the robot, and for the surgeons’ (657B7E). Some respondents said that they opted for surgery over radiation because the promoter of radiotherapy in the vignette admitted to the most potential conflict of interest. Another respondent observed

that only one of the three options is written by someone with a potential…and acknowledged…conflict of interest. In this case that makes it harder for a family physician to evaluate this potentially important detail. This all taken into consideration I cast a wary vote for option number three. (E859C9)

The central argument in support of EM was that it avoided the serious side effects of either active treatment. A 66-year-old male respondent who found himself in similar circumstances to the man in the vignette

opted to wait. When my children urged me to go ahead and have it ‘fixed’ I laughed and said I was enjoying life too much. If I had surgery now and removed the cancer but had sexual dysfunction and died in a car accident (angry at the world), what a waste of time the surgery (or radiation) would have been. (159A5D)

For this respondent, EM’s benefits accrued from the moment active treatments were eschewed. He and others argued that there existed a reasonable trade-off between protecting the quality of life against the small possibility of future death and disability. Another respondent noted that

benefits of any treatment for PSA-detected cancers are still unproven. The risk of incontinence and of sexual impotence are high and would be devastating for the lifestyle of this 63-year-old active man. Complications of cancer would develop in a minority of cases and decades later, at an age when he’s retired, with a totally different lifestyle. I’d trade some healthy years in relative youth, for less years in an older and less enjoyable life. (B0D225)

Respondents also noted that EM could buy time for a future when effective therapies would be developed, and which might only or better work on un-intervened bodies. ‘Only the watchful waiter can appreciate a breakthrough in cancer treatment that makes prostate removal obsolete’ (159A5D).

Other clues to EM’s meaning are the different names used in this poll and elsewhere. Early on in my clinical practice, ‘watchful waiting’ was in vogue, but this term suggested inaction, which posed a larger problem for the early detection and radical treatment paradigm: why diagnose cancer if not to do something? This point was explicitly made by a French physician respondent, who pointed out that

not to have done a PSA would have been the best option! Now that fear of death has been poured in this poor chap the least aggressive approach is warranted, and maybe the one proposed here is still too much, at least it has not been properly tested. ‘Prevention can be bad for your health.’ (ED66CD)

Sometimes ‘active surveillance’ has been used to capture the increasing amount and intensity of the surveillance work often expected of patients who do not have surgery or radiation. One respondent described the intensive work this way:

To track progression, PSA should be measured every 3 months, trans-rectal ultrasound (TRUS) performed every 6 to 12 months and repeat prostate needle biopsy at 12 to 24 month intervals. Progression defined as PSA velocity greater than 0.75 ng/mL/yr, a rise in Gleason score, or greater than 50 per cent increase in lesion size on TRUS. The urine test for the PCA3 gene, already marketed for use in diagnosing prostate cancer, could also be useful in prognostication. Other tests: TRUSP (trans-rectal ultrasound of the prostate) at intervals perhaps of every year to 2–3 years pending stability. Endorectal MRI with or without spectroscopy if item above not done or if you wish further imaging input. PAP (Prostatic Acid Phosphatase) every 2 years to make sure that a PC confined to the prostate is not approaching a critical PAP threshold of 3.0 or higher which confers a poorer prognosis in the face of RP, or any form of RT. (E84272)

There are obvious downsides of this intensive surveillance. These practices are invasive, costly, spread fear, and produce information that is often difficult to interpret. The resulting incompleteness itself becomes a central argument for more definitive and ‘complete’ interventions that promise to put an end to the incomplete character of the risk experience; in effect one intervention reducing the harms of others. This line of argument is similar to how the efficacy of aggressive breast cancer surgery was rationalised in the early twentieth century. Radical surgery for breast cancer was called the ‘complete operation’ which in one stroke made less extensive surgery, which was often followed by local recurrences and the need for more surgery, incomplete (Aronowitz, 2007). The promise to achieve completeness can substitute for, and obscure, efficacy understood as curing cancer and saving lives.

Many respondents noted that the intensive surveillance routines for EM constituted a severe practical and psychological burden for patients. Active treatments ‘worked’ by avoiding the uncertainties and burdens associated with EM. One respondent wrote that ‘it’s a rare patient indeed, who is willing to follow “expectant management” after being told they have prostate cancer’ (278B3E). In some cases, EM involves so much responsibility and work as well as fear and disturbance to peace of mind, that surgery could be rationalised as psychologically efficacious. According to one respondent, ‘a validated diagnosis of prostate cancer, from a psychological point of view, needs a treatment’ (9FDABD). As Whyte et al. (2002) point out in their ethnographic account of drug efficacy, it is common for the pharmaceutical industry, whose long patent protection and status derives from the hard scientific evidence of drug efficacy, to resort to psychological efficacy when such evidence is absent. ‘Once the biopsy report is discussed with the patient,’ one defender of active treatments argued along these lines, ‘he will go through a period of heightened stress with its attendant complications. Radical prostatectomy with removal of any malignant tissue from the body will improve the quality of life’ (D976F2).

The concern with psychological efficacy is real and deeply felt. There is a heavy burden of responsibility that follows not taking the cancer out of the body, where it sits like a time bomb. Active treatment supporters offered definitive aphoristic statements that all cancers should be excised: ‘the best place for prostate cancer is into a bottle in the pathology department’ (53637A). ‘I was diagnosed very similarly at age 51 and had a radical prostatectomy with nerve sparing 14 months ago’ another respondent explained.

My PSA is now non-traceable. I have no issues at all with incontinence. I have ED, but am going through therapy to hopefully correct it soon. I feel that this is a good trade off to be cancer free (my italics). I do not recommend any other therapy. Get the cancer out of there before it kills! Wait too long and you lose the nerves, wait longer and you won’t be able to contain it. Dead is dead, people, quit messing around with inadequate treatments!’ (C5ABB3)

For many respondents, the burden of responsibility for a cancer diagnosis is only exacerbated by surveillance monitoring. One respondent described a rising PSA level as an ‘anxiety producing experience’ (B5C0B9). Others observed that it takes unusual psychological strengths to live with cancer and cancer risk. One doctor noted that before recommending EM, he would

first determine if the patient’s temperament is suitable. The chances are 25 per cent he will be upstaged at prostatectomy and 50 per cent he will either request therapy or shows signs of progression within a couple of years, so it takes a ‘well-informed’ patient with ‘courage’ to do this. (CC298F)

Another doctor explained he favoured active treatments but recognised that EM might be a possibility because ‘the description of the patient background made me believe that he’s quite smart, and would follow the physician’s instructions strictly’ (51830D).

In sum, many of the clinician respondents who supported active treatments generally believed that EM results in a heavy psychological and moral burden. It leads to a heightened state of risk, a highly incomplete state that demands a more definitive cure. And it is not simply a psychological burden, but the work of surveillance practices. A few respondents noted that such practices were also not feasible in poor parts of the world, so active treatments were preferred. ‘Expectant management of prostate cancer seems to be attractive but impractical in India because of strict protocol and cost, moreover the feasibility of long-term expectant management remains unproved. So I strongly feel that nerve sparing surgery is excellent choice in Indian conditions’ (DA6165).

The efficacy of the two active treatments in relieving the burdens of the at-risk state, experienced or anticipated, is similar to the appeal of other contemporary risk interventions, such as the rising rates of contralateral prophylactic mastectomies (Aronowitz, 2015). The logic of relieving the burden of the at-risk state also suggests a future, dystopic or rational depending on your perspective, in which prostate cancer intervention might be the leading salient. It is possible that a version of Hudson’s programme, biopsying all men to see if they have prostate cancer, might be one consequence of the diffusion of risk practices and cancer fears.

This ‘biopsy all’ vision is not fantasy. The Prostate Cancer Prevention Trial (PCPT) was a large randomised, placebo-controlled clinical trial in which some men were treated with an androgen antagonist (finasteride) to determine if prostate cancer could be prevented. At the end of the trial, men in both arms of the study were offered a prostatic biopsy. Biopsying men at no special risk was and is not an accepted clinical practice and was not part of the required research protocol. For investigators, it represented an opportunity to more objectively determine if finasteride worked to prevent cancer without the intervening noise introduced by PSA determinations (which were frequently done, and in the main study rising levels were the tripwire for biopsy and cancer determinations). Approximately three quarters of the men in both the finasteride and placebo arms agreed at the end of the study to have a biopsy; almost 8,000 men in total (Thompson et al., 2003). I interpret their voluntary consent as an indicator of just how much uncertainty and fear was generated by the surveillance aspects of the study. This mass consent suggests the plausibility of a mass biopsy programme in the future, if the level of screening and surveillance increases for prostate cancer in clinical practice in ways resembling PCPT. While the ‘offer’ to biopsy everyone was not part of an explicit research hypothesis, I consider it the PCPT’s most important question and result, with profound implications for the future.

Implications

Critical perspectives on prostate cancer screening and its sequelae—and other similar controversies, for example, over screening mammography for women under 50—are typically articulated in terms of lead and length time biases, overdiagnosis, overtreatment, medicalisation, iatrogenesis and financial conflict of interest. These perspectives capture elements of what is problematic about our present situation, but fail to capture other, more historically conditioned and contingent, developments. So while financial self-interest and even greed among contemporary clinicians and test makers has clearly played a role in the rapid and widespread diffusion of the screen and radical treatment paradigm, as one observer recently quipped about the role of greed as a cause of the 2007/2008 financial crisis, ‘you can’t explain change with constant’ (WNYC studios, 2018). The strikingly different historical course of the prostate cancer diagnosis relative to breast and cervical cancer highlights the role of gender, male anatomy, and other factors which shaped urological pessimism and inaction throughout most of the century. The seeming lack of awareness of what might be at stake in the FDA’s 1985 regulatory approval of PSA suggests a powerful role of largely unplanned and poorly visible looping effects—mediated by hope, fear, unproven promise, uncertainty and risk—in the rapid transformation of beliefs and practices which followed. These effects happened within a particular socio-historical context, as evidenced by the much greater diffusion in the USA in the late twentieth century as compared to economically advanced Western countries that have different health systems (not based on fee-for-service) and regulatory regimes.

Some of the implications of this history are clinical and policy-oriented. The great potential for some risk reducing practices, like the PSA test, to spark autocatalytic like transformations of beliefs, practices and bodies argues for solid empirical evidence of efficacy and safety before they are permitted to diffuse. This history has emphasised how radical interventions such as prostatectomy for cure have ‘worked’ not only by their direct biological impact on bodies but by the ways they ‘cure’ people of the work, burden of fear and uncertainty of the cancer risk experience. In clinical and policy evaluation of such practices, relying on reductive conceptions of biomedical efficacy will therefore be insufficient. More attention and response to these more psychological efficacies are in order. The ways that biomedical and public health technologies and practices have led to states of ‘embodied risk’ may be just as important as their impact on say, quality adjusted life years.

Policymakers and clinicians might lessen fear by specific actions, such as giving and promoting non-cancerous sounding names for embodied risk states (e.g., Esserman et al., 2014). They could do more ‘upstream’ regulation of knowledge production, and factor into decisions the discontinuities between past and present meanings of diseases and diagnoses, and the contingent nature of almost everything they as well as patients know or think they know about prostate cancer and many other diseases. In our response to prostate cancer controversies today, biomarkers and other technological fixes are often served up as solutions just around the corner and help rationalise staying the present course. The historical narrative I have crafted underscores the hubris inherent in assuming that the natural history of the disease is already pre-determined by genetics or anything else and in ignoring the potential of new biomarkers to further complicate judgement by transforming what we believe and do about prostate cancer.

Some of the older clinical beliefs and practices sketched above might also be profitably recalled and cautiously utilised. The humility and scepticism of early twentieth century clinicians, who acknowledged prostate cancer’s confusing and uncertain natural history and who were aware of the difficulty of reconciling into one coherent image the different views of disease which resulted from the diverse ways it was discovered, might be recalled and harnessed in our responses to the complexities of present challenges. Recognising that our own practices have made prostate and other cancers a problematic nemesis suggests that it is also in our power to lessen fear and respond to present challenges with more sobriety.

Footnotes

Declaration of conflicting interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author received financial support from the Leonard Davis Institute of the University of Pennsylvania for the research of this article.

Notes

References

Ablin

R. J.

, & Piana

(2014). The great prostate hoax: How big medicine hijacked the PSA test and caused a public health disaster. St. Martin’s Pres.

Andriole

G. L.

, Crawford

E. D.

, Grubb

R. L.

III , Buys

S. S.

, Chia

, Church

T. R.

, Fouad

M. N.

, Gelmann

E. P.

, Kvale

P. A.

, Reding

D. J.

, Weissfeld

J. L.

, Yokochi

L. A.

, O’Brien

, Clapp

J. D.

, Rathmell

J. M.

, Riley

T. L.

, Hayes

R. B.

, Kramer

B. S.

, Izmirlian, … PLCO Project Team (2009). Mortality results from a randomised prostate-cancer screening trial. New England Journal of Medicine, 360(13), 1310–1319.

Applewhite

J. C.

, Matlaga

B. R.

, Mccullough

D. L.

, Hall

M. C.

(2001). Transrectal ultrasound and biopsy in the early diagnosis of prostate cancer. Cancer Control, 8(2), 141–150.

Aronowitz

R. A.

(2001). Do not delay: Breast cancer and time, 1900–1970. The Milbank Quarterly, 79(3), 355–386.

Aronowitz

R. A.

(2007). Unnatural history: Breast cancer and American society. Cambridge University Press.

Aronowitz

(2014). From skid row to main street: The Bowery series and the transformation of prostate cancer, 1951–1966. Bulletin of the History of Medicine, 88(2), 287–317.

Aronowitz

(2015). Risky medicine: Our quest to cure fear and uncertainty. University of Chicago Press.

Catalona

W. J.

, & Scott

W. W.

(1978). Carcinoma of the prostate: A review. The Journal of Urology, 119(1), 1–8.

Chen

, & Zhou

(2016). The evolving Gleason grading system. Chinese Journal of Cancer Research, 28(1), 58.

10.

Chodak

G. W.

, & Schoenberg

H. W.

(1984). Early detection of prostate cancer by routine screening. Journal of the American Medical Association, 252(23), 3261–3264.

11.

Crawford

E. D.

(1997). Prostate cancer awareness week: September 22 to 28, 1997. CA: A Cancer Journal for Clinicians, 47(5), 288–296.

12.

Esserman

L. J.

, Thompson

I. M.

, Reid

, Nelson

, Ransohoff

D. F.

, Welch

H. G.

, Hwang

, Berry

D. A.

, Kinzler

K. W.

, Black

W. C.

, Bissell

, Parnes

, & Srivastava

(2014). Addressing overdiagnosis and overtreatment in cancer: A prescription for change. Lancet Oncology, 15(6), e234–e242.

13.

Giuliani

(2007). Rudy on Cavuto. https://www.youtube.com/watch?v=AP-H3JxVyyE

14.

Guinan

, Bush

, Ray

, Vieth

, Rao

& Bhatti

(1980). The accuracy of the rectal examination in the diagnosis of prostate carcinoma. New England Journal of Medicine, 303(9), 499–503.

15.

Gutman

E. B.

, Sproul

E. E.

, & Gutman

A. B.

(1936). Significance of increased phosphatase activity of bone at the site of osteoplastic metastases secondary to carcinoma of the prostate gland. The American Journal of Cancer, 28(3), 485–495.

16.

Hacking

(1995) The looping effect of humankinds. In Sperber

, Premack

, & Premack

A. J.

(Eds.), Causal cognition: A multidisciplinary debate (pp. 351–383). Clarendon Press.

17.

Hacking

(2006). Making up people. London Review of Books, 28(16), 23–26.

18.

Immunological Device Panel, U.S. Food and Drug Administrations (1985). Minutes, 12/9/1985 (obtained by FOIA request, 10/24/2017).

19.

Jones

M. P.

(2005). Biotech’s perfect climate: The hybritech story [PhD Thesis]. University of California, San Diego.

20.

Keyes

E. L.

(1917). Urology: Diseases of the urinary organs, diseases of the male genital organs, the venereal diseases. D. Appleton.

21.

Kolata

(1993a, 23 June. Whether positive or negative, result of prostate cancer test can create maze of questions. New York Times. https://www.nytimes.com/1993/06/23/health/whether-positive-negative-result-prostate-cancer-test-can-create-maze-questions.html

22.

Kolata

(1993b, 29 September. How demand surged for prostate test. New York Times (p. 9).

23.

Merrill

R. M.

, Feuer

E. J.

, Warren

J. L.

, Schussler

, & Stephenson

R. A.

(1999). Role of transurethral resection of the prostate in population-based prostate cancer incidence rates. American Journal of Epidemiology, 150(8), 848–860.

24.

Potosky

A. L.

, Miller

B. A.

, Albertsen

P. C.

, & Kramer

B. S.

(1995). The role of increasing detection in the rising incidence of prostate cancer. Journal of the American Medical Association, 273(7), 548–552.

25.

Raffle

A. E.

, & Muir Gray

J. A.

(2007). Screening: Evidence and practice. Oxford University Press.

26.

Rich

A. R.

(1935). On the frequency of occurrence of occult carcinoma of the prostate. The Journal of Urology, 33(3), 215–223.

27.

Schröder

F. H.

, Hugosson

, Roobol

M. J.

, Tammela

T. L.

, Ciatto

, Nelen

, Kwiatkowski

, Lujan

, Lilja

, Zappa

, Denis

L. J.

, Recker

, Berenguer

, Määttänen

, Bangma

C. H.

, Aus

, Villers

, Rebillard

, van der Kwast

, … Auvinen

(2009). Screening and prostate-cancer mortality in a randomised European study. New England Journal of Medicine, 360(13), 1320–1328.

28.

Schröder

F. H.

, Roach

, & Scardino

(2008). Clinical decisions. Management of prostate cancer. New England Journal of Medicine, 359(24), 2605–2609. https://doi.org/:10.1056/NEJMclde0805491

29.

Schwartz

R. S.

(2009). Management of Prostate Cancer—Polling Results. NEJM, 360, e4–e4.

30.

Smith

G. G.

(1942). Surgery for cancer of the prostate. Journal of the American Medical Association, 118(17), 1514. https://doi.org/10.1001/jama.1942.02830170092023

31.

Stanford

J. L.

, Stephenson

R. A.

, Coyle

L. M.

, Cerhan

, Correa

, Eley

J. W.

, Gilliland

, Hankey

, Kolonel

L. N.

, Kosary

, Ross

, Severson

, & West

(1999). Prostate cancer trends 1973–1995, SEER Programme National Cancer Institute. https://pdfs.semanticscholar.org/b0bd/3d8235ea698c1b70d7301225d40454d23a1b.pdf

32.

Thompson

(1987, 10 November. Prostate cancer: What men don’t know. Washington Post.

33.

Thompson

I. M.

, Ernst

J. J.

, Gangai

M. P.

, & Spence

C. R.

(1984). Adenocarcinoma of the prostate: Results of routine urological screening. The Journal of Urology, 132(4), 690–692.

34.

Thompson

I. M.

, Goodman

P. J.

, Tangen

C. M.

, Lucia

M. S.

, Miller

G. J.

, Ford

L. G.

, Lieber

M. M.

, Cespedes

R. D.

, Atkins

J. N.

, Lippman

S. M.

, Carlin

S. M.

, Ryan

, Szczepanek

C. M.

, Crowley

J. J.

, & Coltman Jr

C. A.

(2003). The influence of finasteride on the development of prostate cancer. New England Journal of Medicine, 349(3), 215–224.

35.

Valier

(2016). A history of prostate cancer: Cancer, men and medicine. Palgrave Macmillan.

36.

Walsh

P. C.

, Lepor

, & Eggleston

J. C.

(1983). Radical prostatectomy with preservation of sexual function: Anatomical and pathological considerations. The Prostate, 4(5), 473–485.

37.

Watson

R. A.

& Tang

D. B.

(1980). The predictive value of prostatic acid phosphatase as a screening test for prostatic cancer. New England Journal of Medicine, 303(9), 497–499.

38.

Whyte

S. R.

, van der Geest

, & Hardon

(2002). Social lives of medicines. Cambridge University Press.

39.

WNYC Studios (2018). One financial crisis, many explanations. On the Media. https://www.wnycstudios.org/story/one-financial-crisis-many-explanations

40.

Young

H. H.

(1905). The early diagnosis and radical cure of carcinoma of the prostate. Being a study of 40 cases and presentation of a radical operation which was carried out in four cases. Bulletin Johns Hopkins Hospital. 16, 315–321.