Conservative treatment of prostate cancer


Albertsen PC, Fryback DG, Storer BE, Kolon TF, Fine J. Long-term survival among men with conservatively treated localized prostate cancer. JAMA 1995; 274: 626-31.

Reviewed by

Henry Barry, M.D., M.S.

Clinical question

Do men between 65 and 75 with conservatively treated localized prostate cancer live longer than the general population?

Background

Whether or not to screen for prostate cancer is one of the most controversial dilemmas faced by primary care physicians. If this wasn't enough, we face the added problem of which treatment (if any) to offer once prostate cancer is diagnosed! The literature to date fails to provide support in favor of screening or early aggressive treatment. Decision analyses show that the magnitude of benefit does not appear to outweigh the additional morbidity and cost of screening. A Swedish study found that once diagnosed, watching and waiting was a reasonable option. This study, however, has been criticized because the patients studied were older, it used cytology histopathology to confirm the diagnosis, and the sample included a large number of men with low grade tumors. Finally, an analysis of men with localized prostate cancer found that men receiving therapeutic interventions actually experienced a poorer disease-specific quality of life!

Population studied

The study included all men in the Connecticut Tumor Registry (CTR) between the age of 65 and 75 with localized prostate cancer initially diagnosed between 1971 and 1976. Men were excluded from the study if the diagnosis was made at autopsy (n = 58), was an incidental finding during cystectomy (n = 5), or if they had radical prostatectomies (n=111). Men were also excluded if their medical records were unavailable (n = 158) or incomplete (n = 7), if their initial treatment could not be determined (n = 3), or if the diagnosis could not be confirmed by either pathology report or review of the original slides (n = 19). After exclusions, the authors analyzed results from 451 men. Note that this study looks at men diagnosed because they had either a palpable nodule or prostatic hypertrophy, because they were evaluated in the era before prostate specific antigen (PSA) testing was available.

Study design and validity

This is a retrospective cohort study. After identifying the subjects through the CTR, the hospital records were abstracted and pathology specimens (or reports) were reviewed. Death certificates were reviewed and the cause of death was recorded when the CTR identified that the man had died. The staff who abstracted the records and the pathologist who reviewed the slides and reports were all blind to the status of the subject. The authors then performed survival analyses for subgroups of men. Finally, the authors used standard life tables to compare the survival in the subjects with the age-adjusted survival rate for the general population.

Outcomes measured

The primary outcome measure was survival. Because survival is determined by many factors, the authors looked at confounders such as age at diagnosis, race, year of diagnosis, the method of diagnosis, the results of metastatic evaluations, and initial treatment (immediate or delayed hormonal therapy). The authors also measured the number and severity of comorbid conditions, and a pathologist assigned a Gleason score to each subject. Death certificates were used to identify the cause of death.

Results

The mean age at diagnosis for the 451 men was 70.9 years. The authors were able to follow up the subjects for an average of 15.5 years; 40 (9%) of whom were still alive at the time of last contact. Among the deaths, 154 (34%) were attributed to prostate cancer, 221 (49%) were attributed to other causes, and for 36 (8%) the cause of death could not be identified. Hormonal therapy was started immediately in 202 men and the remaining 249 men received no therapy during the first 3 months after diagnosis.The age-adjusted survival for men with low grade disease (Gleason score of 2 to 4) was not significantly different from the general population. As severity of disease increased, survival declined. The most powerful predictors of survival were tumor grade and comorbid conditions. Unfortunately, the authors do not directly address the impact of early versus delayed treatment on survival.

Recommendations for clinical practice

The incidence of prostate cancer and the rates for prostate surgery have risen since the 1980s; much of the increase in radical prostatectomy rates occurring in older men. Since the authors excluded men receiving prostatectomy, we should be cautious about extrapolating the results to today's practice. Nonetheless, the results of this paper lend support to the notion that conservative therapy without surgery or radiation is a reasonable option for older men with low grade prostate cancer (Gleason score 2-4).


copyright 1997, Appleton and Lange