Non-curative initial management of “low-risk” prostate cancer (PCa) in older men is associated with an increased risk of prostate cancer-specific mortality (PCSM) compared with those who undergo curative treatment, according to recently published findings.
Ayal A. Aizer, MD, of the Harvard Radiation Oncology Program in Boston, and collaborators studied a cohort of 27,969 men with PSA-detected low-risk PCa identified by the Surveillance, Epidemiology and End Results (SEER) program from 2004 to 2007. After a median follow-up of 2.75 years, 1,121 patients died, 60 (5.4%) from PCa. In adjusted analyses, non-curative treatment was associated with a significant 3.3-fold increased risk of PCSM, Dr. Aizer’s group reported online ahead of print in BJU International. Each one-year increment in age was associated with a significant 5% increased risk of PCSM.
Men older than the median age (67 years) experienced increased estimates of PCSM when treated with non-curative as opposed to curative intent; this finding was not observed in men at or below the median age. Among men older than 67, the three-year estimates of PCSM in those managed with curative versus non-curative approaches were 0.17% and 0.76, a significant difference between the management approaches. Among men aged 67 years or younger, the three-year estimates of CPSM in those managed with curative versus non-curative approaches, respectively, were 0.13% and 0.18%, a non-significant difference.
The clinical significance of the study’s findings is that older men with “low-risk” PCa may harbor occult high-grade disease that can be missed because of prostate biopsy sampling error despite extended biopsy, “placing them at increased risk of PCSM within a few years of the diagnosis when non-curative treatments such as active surveillance are used initially,” the researchers wrote.
Additionally, the authors noted that a possible explanation for the association between increasing age and PCSM risk may be derived from the established relationship between older age and high-grade disease. PCa undergrading due to limitations in current biopsy techniques “may, therefore, be particularly consequential in older men, in whom undergrading is known to be more common owing to increasing gland volume coupled with increasing prevalence of BPH [benign prostate hyperplasia].”
The new findings contrast with those of the randomized Prostate Cancer Intervention versus Observation Trial (PIVOT), whose results were published last year in The New England Journal of Medicine (2012;367:203-213). In PIVOT, researchers compared the effectiveness of radical prostatectomy (RP) versus observation in a group of 731 men with localized PCa. During a median follow-up of 10 years, 21 of 364 (5.8%) assigned to RP and 31 of 367 (8.4%) assigned to observation died from PCa, a non-significant difference between study arms.
Dr. Aizer’s group said their study, which had 27,969 patients, had greater power to detect a difference in PCSM than did PIVOT. Moreover, in PIVOT, approximately 20% of patients in the observation arm received definitive treatment, and the analysis was performed using intention-to-treat, Dr. Aizer and his colleagues pointed out. “Although analysis by intention-to-treat is most appropriate for a randomized trial given that randomization is preserved, such an approach further compromises the power of the study to detect a difference in PCSM.”
Ian M. Thompson, MD, Professor of Urology and Chairman of the Department of Urology at the University of Texas Health Science Center in San Antonio, commented that although the report by Dr. Aizer and his colleagues is intriguing, “the follow-up and numbers of deaths are far too small to reach any conclusions.”
“Given what we know about prostate cancer biology,” Dr. Thompson told Renal & Urology News, “what this indicates to me was that these  deaths were most likely due to two possibilities, occurrences that we see occasionally and rarely.”
Occasionally, he pointed out, a man who has had an initial biopsy finding of low-grade PCa and opts for surveillance is subsequently found on follow-up biopsy to have a dramatic increase in volume or grade of disease, which sometimes is heralded by an exponential increase in PSA. Rarely, clinicians will encounter a patient who simply has a skyrocketing PSA. “Both of these events are associated with what is referred to as an ‘interval cancer,’ or development of a very rapidly-growing tumor,” said Dr. Thompson, who also is Director of the Cancer Therapy and Research Center in San Antonio.
Dr. Thompson also pointed out that the new study had the limitations inherent in any retrospective study. In SEER, he said, bias almost certainly exists in how patients were treated, “and we will never know the extent of the bias.”
This treatment bias may relate to the fact that low-risk disease can encompass a wide range of biopsy findings. For example, Dr. Thompson said, one low-risk patient could have had one in 12 biopsy cores positive for cancer, with less than 5% of the core involved with Gleason 3+3 cancer and a PSA of 2.5 ng/mL, whereas another low-risk patient could have had a PSA of 9.89 and five of six cores positive with 50% or more of Gleason 3+3 cancer.
Another potential bias is patient age. Younger patients, who are more likely to be treated, are far less likely to die from PCa than older patients, who are at far greater risk of having high-grade disease, he noted.
Dr. Thompson stressed that researchers can never eliminate the biases of retrospective studies, and such studies cannot replace appropriately-powered clinical trials.