Prostate Cancer: To Screen or Not to Screen?

In a lively session at the 32nd Annual Meeting of the American Societyof Clinical Oncology (ASCO), Jerome P. Ritchie, MD, Brigham and Women'sHospital, Boston, and Steven H. Woolf, MD, MPH, Medical College of Virginia,Fairfax, debated the merits of screening for prostate cancer. In the firstof a three-part report on that session, E. David Crawford, MD, Universityof Colorado Health Sciences Center, Denver, frames the debate by exploringthe issues surrounding the screening and early detection of prostate cancer.Dr. Ritchie's and Dr. Woolf's remarks will be featured in subsequent issuesof ONCOLOGY.

"Prostate cancer is the most common cancer diagnosed in men, andit's the second leading cause of death, but more people die with prostatecancer than of prostate cancer," Dr. Crawford said. That is the cruxof the dilemma facing clinicians.

Some 317,000 new cases of prostate cancer will be diagnosed in 1996,and 41,000 men will die of prostate cancer. Also, the past 15 years havewitnessed a dramatic increase in both the incidence of and deaths fromthis cancer--almost a 500% rise in the number of new cases and nearly a200% increase in prostate cancer deaths. Given the graying of the US population,the fact that men are living longer, and the association of prostate cancerwith aging, these trends are likely to continue into the first few decadesof the 21st century.

Faced with a cancer that is rising in both incidence and mortality,clinicians have three options, according to Dr. Crawford. "The firstthing we can do is try to prevent it," he said. The second approach--tofind prostate cancer early, treat it, and cure it--has both pros and cons.Third, if a cure for advanced prostate cancer existed, clinicians couldwait until people developed advanced disease and then cure them.

Is Prevention Feasible?

"There is an opportunity over a long period of time to try to preventprostate cancer or just delay its progression," Dr. Crawford said.First, prostate cancer has a high prevalence, slow growth rate, and geneticheterogeneity. Another helpful feature is that prostate cancer is androgen-dependent.Also, there are some accepted precursors of prostate cancer, particularlyprostatic intraepithelial neoplasia, that can serve as markers. Lastly,like many tumors, prostate cancer has a long history before metastaticdisease and death of the patient occur.

If a chemoprevention strategy is to be effective, several criteria mustbe met, Dr. Crawford asserted. First, the strategy must interfere witha critical step in carcinogenesis. It should also interfere with tumorgrowth itself and have minimal toxicity.

With regard to chemoprevention and prostate cancer specifically, thereare many things we know, but a great deal that we don't know, Dr. Crawfordcommented. For example, "we know that retinoids promote cancer differentiation,[and] there is some evidence in prostate cancer that they may ameliorateincidence." It is also known that red meat increases incidence. Onthe other hand, the effect of beta-carotene is unclear, with some studiesshowing that it increases prostate cancer risk, others indicating no effect,and a recent study demonstrating that it decreases risk.

Recent studies also have shed some light on the relationship betweentestosterone and dihydrotestosterone (DHT) and prostate cancer, Dr. Crawfordsaid. Testosterone and DHT drive the cell cycle and may shorten it. Inthe presence of high testosterone levels, DNA repair may be altered, leadingto genetic alterations. Some studies have substantiated that African-Americanmen have higher baseline testosterone levels than white men, and this maybe one explanation for the higher incidence of prostate cancer in African-Americans.

In a rat model, Boslan from NYU found that testosterone and estrogens(E2), together, were the most mitogenic compounds, whereas DHT, by itself,did not produce much tumor growth. Thus, methods to interrupt testosteronemay be important, Dr. Crawford said.

Finasteride (Proscar) is another potential approach to chemoprevention.This drug is an 5-alpha-reductase inhibitor and inhibits the conversionof testosterone to DHT. "We know that DHT...has the most affinityfor the receptor. And so this may be a way to prevent prostate cancer becausewe know that men who are congenitally deficient in 5-alpha-reductase haveno prostate development or develop prostate cancer," Dr. Crawfordsaid.

Although finasteride decreases serum DHT by 75% and decreases intraprostaticDHT by 80%, it also increases serum testosterone by 10%. Its effects onserum testosterone may be problematic, given Boslan's findings that serumtestosterone and estrogens are the most potent drivers of prostate cancer.In the presence of higher serum levels, testosterone is aromatized to estrogens.

The soon-to-be completed NCI Chemoprevention Trial "should adda lot of information on various aspects of prostate cancer--not just theeffects of finasteride, but natural history and so forth," Dr. Crawfordnoted. In this trial, 18,000 men being between the ages of 55 and 70 withno evidence of prostate cancer were randomized to finasteride or placebofor 10 years. At the end of the treatment period, all patients will undergoa biopsy of the prostate.

Rationale for Early Detection

There are several reasons why early detection of prostate cancer isdesirable, Dr. Crawford said. First, "men die from prostate cancerand we can't cure advanced disease, and we know that most cancers startmicroscopically." A number of surgical series indicate that if a patienthas organ-confined disease and the prostate is removed, he can be cured.

Second, even if the chemopreventive agents under investigation, suchas finasteride and retinoids, do, in fact, prevent prostate cancer, itwould be years before such a benefit could be shown. With a cancer thatis increasing in mortality, one cannot afford the luxury of waiting forthose results, Dr. Crawford argued. "What's at stake here is the patient'slife, and I think we have some good tools...to at least detect prostatecancer earlier."

A 1989 survey showing that men didn't know great deal about prostatecancer or screening for that cancer was the impetus for the public educationand research campaign begun by the Prostate Cancer Education Council inthe United States. "Our goals were to increase public awareness, toencourage early detection and diagnosis and conduct community-based screeningstudies," Dr. Crawford said.

Calling the success of these efforts "phenomenal," Dr. Crawfordnoted that over 2-1/2 million men have been screened since 1989, and recordsare available on 300,000 men at some 800 sites around the United States.One of the first findings to emerge from this work was the value of prostate-specificantigen (PSA). In particular, the advent of PSA enabled clinicians to identifya new group of prostate cancer patients--those with an elevated PSA anda normal digital rectal examination (DRE). "PSA has made a big splashin the early detection of prostate cancer and it's here to stay aroundthe world, but it's also controversial."

Other important pieces of information coming out of the Prostate CancerEducation Council's efforts include the following:

• The age-related increase in PSA
• The higher incidence of prostate cancer in African-American males
• The positive predictive value of PSA testing
• The lack of an association between vasectomy and prostate cancer

Longitudinal Study

In 1992, in the context of Prostate Cancer Awareness Week, a multicenterlongitudinal study was begun in 1992 that followed 120,000 participantsto chart yearly changes in PSA, age ranges, and so forth. This study captureddata on about 50,000 patients each year and now has serial records from3 years of screening. The average positive biopsy rate is about 25%, headded.

What have we learned from this study? First, Dr. Crawford cited age-specificPSA reference ranges developed on the basis of a few hundred thousand PSAsobtained during Prostate Cancer Awareness Week. Dr. Osterling was one ofthe first to report the relationship of age and PSA based on data from1,000 men in Olmstead County, Minnesota. These data indicate that PSA increaseswith age.

"We think this is important because if you see a 40 year old manwho has a PSA above 2 that is abnormal, whereas a 75-year-old man is allowedto have a PSA of above 5," Dr. Crawford said, adding that PSA variesby race and age.

Also, some new data have been generated based on a comparison of prostatecancers that were detected at the initial screening vs subsequent screeningsat years 2, 3 and 4 (Figure 1). Thiscomparison shows that the majority of prostate cancers--70%--were detectedat the initial screening. The detection rate in the second year is approximately25%, and then it drops off substantially in serial screening.

Dr. Crawford also highlighted several other interesting findings thathave emerged from this large body of data. "If we look at age whenthe number of positive biopsies occurred in those that had prostate cancer,there is a large difference in men ages 40-49 in the first second yearof screening compared to men in their 50s and 60s. This suggests that menwith risk factors (ie, family history, African-American race) should havea PSA determination earlier than age 40 in order for the test to have animpact (Figure 2)."

Also, close examination of the initial, second, third, and fourth yearscreenings for prostate cancer shows that advanced (D2) prostate cancerhas been eliminated. "And if you believe that that may be valuable,then serial screening has been effective in eradicating advanced prostatecancer, which, in fact, the majority of patients presented with severalyears ago," Dr. Crawford said.

Determining free and total PSA also is important. It is known that approximately85% of PSA in the serum is bound to alpha-1-antichymotrypsin, whereas theremainder is either free in the serum (the active form) or completely surroundedby macroglobulin. It is also known that the ratio of free to total PSAis lower in patients with prostate cancer. Several studies have shown thatif a man has a PSA ratio less than .18, he has a higher risk of prostatecancer. This has increased the specificity of the test, according to Dr.Crawford.

Based on present knowledge, Dr. Crawford proposed that "after aninitial negative screen, you don't need to test for prostate cancer everyyear in all patients. Maybe...every other year or every third year [willsuffice]."

Why the Controversy?

"I think that we can clearly show that the [PSA] test is effectivein detecting prostate cancer...But why is it controversial?" Dr. Crawfordasked. "The screening skeptics say that our diagnostic methods areunreliable...that treatment isn't effective (radical prostatectomy, radiation,implants)." The PSA test also has been criticized on the grounds thatit is unethical to create anxiety, and that the test poses too great anexpense to society to be recommended on a widespread basis.

On the plus side, screening can detect more localized cancers, and haseliminated metastatic prostate cancer. However, part of the controversystems from the fact that more men die with, rather than of, prostate cancer.Particularly "eye-opening" in this regard, said Dr. Crawford,is the work of Wael Sakr from Wayne State. In a study of young men whodied of trauma at a young age, Dr. Sakr found that 31% of men between theages of 30 and 39 have prostate cancer.

"We don't want to find those prostate cancers in all those men,"Dr. Crawford asserted. "We want to find it in men [in whom] we caninterfere with the disease in such a way as to increase the survival rate."

When one looks at the results of screening over time, some importantfindings emerge, said Dr. Crawford. First, the initial year yields thehighest results (Figure 3). He emphasizedthat what is being seen is the prevalence, which is the incidence timesthe number of years the patients lives with the disease. Second, with serialscreening, metastatic prostate cancer is definitely on the decline. "Andhopefully with effective treatments, we're going to see the mortality fall.And, in fact, in some of the SEER data, we're seeing that."

Dr. Crawford concluded by noting that physicians need to keep in mindwhat patients with prostate cancer actually want. Several surveys indicatethat patients want the cancer to be slowed, their lives extended, and theirquality of life improved.