A 72-Year-Old Man With Localized Prostate Cancer—14 Years Later

In the first of 136 Clinical Crossroads articles published in JAMA since 1995, Peter C. Albertsen, MD, discussed Mr S, a 72-year-old man with localized prostate cancer.¹

In 1992, Mr S was found by Dr K, his primary physician, to have slight induration of his prostate on a routine physical examination. A consulting urologist confirmed the finding and suggested a prostate biopsy after noting an elevated prostate-specific antigen (PSA) value of 4.3 ng/mL. Needle biopsy performed in 1992 was “suspicious for adenocarcinoma, but not diagnostic,” but repeat biopsy 1 year later revealed adenocarcinoma in 2 of 8 cores: 1 with a “minute focus” and 1 with a “small focus.” The areas with carcinoma were read as Gleason 3 + 3, and another area revealed “high-grade intraepithelial neoplasia.” Evaluation for metastases was negative, and prostate volume was measured as 32 cm³. A repeat PSA value was 3.8 ng/mL. In February 1994, Mr S was offered the following options: (1) he could be followed up for onset of symptoms and his PSA could be monitored; (2) he could receive radiation therapy; or (3) he could undergo a radical prostatectomy. Mr S elected to pursue “watchful waiting.”

Mr S smoked heavily until age 40 years and had a long-standing history of hyperlipidemia, treated first with nicotinic acid and subsequently with statins. In the late 1970s, he developed angina pectoris, leading to coronary artery bypass graft surgery in 1981. In 2000, angina pectoris recurred, along with symptoms of heart failure and a loud murmur that led to the diagnosis of severe mitral insufficiency, with a cardiac ejection fraction of 40% to 50%. In 2000, a porcine mitral valve was implanted, and 3 coronary grafts were “redone.” Postoperatively, Mr S developed symptoms of severe heart failure, along with complete heart block that required a pacemaker. His ejection fraction decreased to less than 20% and remained in that range for the rest of his life. He lost considerable weight, his cardiac function ranged between New York Heart Association classes IIIB and IV, and he required multiple hospitalizations, culminating in implantation of a milrinone pump.

In the last 8 years of his life, Mr S developed increasing symptoms of prostatism, including 1 episode of urinary tract obstruction. Digital examinations revealed a prostate that grew to about 60 cm³, with the same area of induration persisting. His PSA increased to 12.8 ng/mL in 2008. His prostatic symptoms were managed with tamsulosin; he never received specific treatment for his prostate cancer. In 2008, 16 years after the diagnosis of prostate cancer, Mr S died of “cardiac cachexia.” His family declined an autopsy.

Mr S suffered greatly from his progressive heart failure. Widowed in 2003, he was hospitalized 17 times for heart failure after his wife's death. We followed his PSA off and on, but he showed less and less interest in it over the years. He remained fiercely independent, writing, drawing, reading avidly, and living alone without much help until the last few months of his life. He finally agreed to move from his large and lonely home to an apartment, and then 8 months prior to his death, into another apartment in an assisted-living community. Four weeks before he died, he agreed to hospice care. He died in his apartment with his children at his bedside.

Since the first Clinical Crossroads was published in July 1995, several studies have added to our understanding of the natural history of prostate cancer and the relative effect of treatment. The fundamental treatment dilemma confronted by Mr S in 1995 has not changed, nor have the treatment alternatives. Newer techniques are now advocated for performing radical prostatectomy and administering radiation therapy, but little additional information has emerged concerning the relative efficacy of these modes compared with conservative management. I suspect that if Mr S were confronted with the same problem today, he would make the same decision, possibly with increased confidence based on data from the past 14 years.

The most striking change in the past 14 years is the dramatic increase in the incidence of this disease and the stage shift away from metastatic toward localized disease.² Compared with 1995, the diagnosis of low-volume, low-grade disease (≤2 cores containing only Gleason 3 + 3 = 6 disease, with tumor involving <35% of either core) is much more common, leading to the recommendation of “active surveillance” protocols that have documented the slow progression of these lesions.³

Results from the finasteride chemoprevention trial published in 2003 documented a much higher prevalence of Gleason 3 + 3 = 6 disease than previously anticipated.⁴ When the trial was originally designed, the prevalence of prostate cancer was assumed to be 6%, and the trial was powered to detect a 25% reduction. After 7 years of follow-up, prostate cancer was detected in 24.4% of men in the control group and 18.4% in the treatment group. These much higher rates were the result of a decision to biopsy as many men as possible in each group, regardless of their clinical findings or PSA levels. The trial demonstrated that clinically significant prostate cancers were present even among men with serum PSA below 4.0 ng/mL. Equally important was the observation that most cancers detected were low-grade Gleason 3 + 3 = 6 tumors. The prevalence of these tumors appears to be much higher than suspected in 1995.

More recently, 2 large randomized trials addressing PSA testing for prostate cancer called into question the relative efficacy of prostate cancer screening, especially in older men. The Prostate, Lung, Colorectal and Ovarian trial sponsored by the National Cancer Institute concluded that after 7 to 10 years of follow-up, prostate cancer mortality remained low and did not differ significantly between the screened group and controls.⁵ The European Randomized Study of Screening for Prostate Cancer concluded after 8 years of follow-up that PSA-based screening reduced death from prostate cancer by 20% but was associated with a high risk of overdiagnosis.⁶ The study concluded that 48 men would need treatment to prevent 1 cancer death.

The last 2 decades have also witnessed a significant shift in the use of the Gleason scoring system.⁷ Prostate cancers are rarely graded as Gleason 2 to 5 tumors by contemporary pathologists. Furthermore, many tumors previously classified as Gleason 3 + 3 tumors are now classified as Gleason 3 + 4 disease. This shift may affect a patient's assessment of the risk posed by his tumor, especially if he relies on information published from data assembled in the pre-PSA era.^{8 - 9}

Treatment alternatives have not changed dramatically since 1995, but proponents of surgery and radiation have claimed superior outcomes associated with newer techniques.¹⁰ Laparoscopic and robot-assisted prostatectomy is advertised extensively, but reports on the relative efficacy compared with the traditional surgical approach do not suggest major differences in outcomes.¹¹ Intensity-modulated radiation therapy may have fewer complications than older radiation techniques, but no data are available from randomized trials that document any relative survival improvement associated with this technique.¹²

In 2008, the Scandinavian Prostate Cancer Group (SPCG) published 12-year outcomes from their trial comparing radical prostatectomy with watchful waiting.¹³ They continue to document a small prostate cancer–specific survival advantage for men undergoing surgery, but this advantage appears to be limited to men younger than 65 years at diagnosis. Men with well-differentiated tumors do well, but men with poorly differentiated tumors often have cancer recurrence requiring additional treatment. After 12 years of follow-up, the overall survival rate for the 2 groups appears comparable.

During the past decade, many clinicians have advocated treating men like Mr S with primary androgen deprivation therapy. Lu-Yao et al¹⁴ explored the effect of this approach using Medicare claims data linked to the Surveillance, Epidemiology, and End Results program. They found no survival advantage associated with early use of androgen deprivation and recommended that this approach be reserved for men who show clinical evidence of disease progression.

In the 1995 Clinical Crossroads, I discussed a decision analysis model published in JAMA in 1993.¹⁵ Data from the SPCG-4 study and the Connecticut Tumor Registry have validated this model, which estimated clinical outcomes for a 10-year period following diagnosis. For men similar to Mr S, who have small-volume, low-grade disease and a potential life expectancy of 10 to 15 years, the risk of complication from treatment probably outweighs any effect treatment may have on the risk of dying of prostate cancer. For men with higher-grade disease, and for men with life expectancy considerably greater than 15 years, the relative effect of screening and treatment remains uncertain but is more likely to yield a modest benefit.

Fortunately, a randomized clinical trial comparing the outcomes of surgery, radiation, and conservative management for screening-detected prostate cancer was initiated more than 5 years ago and has achieved the necessary accrual goals.¹⁶ Hopefully, this trial will provide critical information concerning the relative effect of treatment 14 years from now. Until then, men similar to Mr S will need to weigh their estimates of the benefits of treatment and the risks of treatment-related complications against their estimates of the likelihood of disease progression. If they are as lucky as Mr S, they will choose wisely and select the treatment most appropriate for them.