In the Journals

Low PSA levels may predict aggressive prostate cancer higher mortality risk

Very low PSA levels appeared associated with a greater risk for disease-specific death in men with high-grade prostate cancer, according to the results of a population-based study.

“An elevated PSA level is a well-known adverse clinical feature and is associated with a poor prognosis,” Paul L. Nguyen, MD, associate professor of radiation oncology at Harvard Medical School and director of prostate brachytherapy and clinical trials for genitourinary radiation oncology at Dana-Farber Cancer Institute, and colleagues wrote. “Nevertheless, high-grade disease is found even at PSA levels less than 4 ng/mL, and up to 10% of prostate tumors produce very small amounts of PSA.”

Nguyen and colleagues sought to determine whether a very low PSA level correlated with a greater risk for prostate cancer-specific mortality among men with a Gleason score of 8 to 10.

The researchers used the SEER database to identify 250,986 men (median age, 65 years; interquartile range [IQR] 59-72) diagnosed with high-grade prostate cancer between 2004 and 2010. They used a multivariate regression analysis to determine disease-specific mortality as a function of PSA level (≤ 2.5, 2.6-4, 4.1-10, 10.1-20, 20.1-40 or > 40 ng/mL) and Gleason score (8-10 or ≤ 7).

Median follow-up was 38 months.

Using a PSA level of 4.1 to 10 ng/mL as a referent, the researchers observed an adjusted HR for disease-specific mortality of 2.15 (95% CI, 1.65-2.79) among men with a PSA of 2.5 ng/mL or lower; 1.6 (95% CI, 1.22-2.1) among men with a PSA level of 2.6 ng/mL to 4 ng/mL; 1.6 (95% CI, 1.41-1.82) for men with a PSA level of 10.1 ng/mL to 20 ng/mL; 2.08 (95% CI, 1.81-2.38) for men with a PSA level of 20.1 ng/mL 40 ng/mL; and 3.23 (95% CI, 2.85-3.65) for men with greater than a 40 ng/mL PSA level.

A PSA of 2.5 ng/mL or less appeared to be the only level significantly associated with an increased risk for prostate cancer-specific mortality among men with a Gleason score of 8 to 10 (P < .001).

The researchers acknowledged study limitations. Because SEER replaces biopsy Gleason scores with prostatectomy Gleason scores among surgically treated patients, it is possible that the researchers underestimated the extent of prostate cancer-specific mortality in this patient population. Further, they noted that SEER does not report PSA density or doubling time.

“Despite its potential limitations, this study provides strong evidence to support the notion that low PSA levels in Gleason score 8 to 10 disease may be a sign of underlying aggressive and extremely poorly differentiated or anaplastic low PSA-producing tumors,” Nguyen and colleagues wrote. “Clinicians should be aware of these findings, and patients with low-PSA, Gleason score 8 to 10 disease should be considered for systemic staging and clinical trials studying the use of chemotherapy and other novel agents for very high-risk prostate cancers.” – by Cameron Kelsall

Disclosure: Nguyen reports consultant roles with GenomeDx and Medivation. The other researchers report no relevant financial disclosures.

Very low PSA levels appeared associated with a greater risk for disease-specific death in men with high-grade prostate cancer, according to the results of a population-based study.

“An elevated PSA level is a well-known adverse clinical feature and is associated with a poor prognosis,” Paul L. Nguyen, MD, associate professor of radiation oncology at Harvard Medical School and director of prostate brachytherapy and clinical trials for genitourinary radiation oncology at Dana-Farber Cancer Institute, and colleagues wrote. “Nevertheless, high-grade disease is found even at PSA levels less than 4 ng/mL, and up to 10% of prostate tumors produce very small amounts of PSA.”

Nguyen and colleagues sought to determine whether a very low PSA level correlated with a greater risk for prostate cancer-specific mortality among men with a Gleason score of 8 to 10.

The researchers used the SEER database to identify 250,986 men (median age, 65 years; interquartile range [IQR] 59-72) diagnosed with high-grade prostate cancer between 2004 and 2010. They used a multivariate regression analysis to determine disease-specific mortality as a function of PSA level (≤ 2.5, 2.6-4, 4.1-10, 10.1-20, 20.1-40 or > 40 ng/mL) and Gleason score (8-10 or ≤ 7).

Median follow-up was 38 months.

Using a PSA level of 4.1 to 10 ng/mL as a referent, the researchers observed an adjusted HR for disease-specific mortality of 2.15 (95% CI, 1.65-2.79) among men with a PSA of 2.5 ng/mL or lower; 1.6 (95% CI, 1.22-2.1) among men with a PSA level of 2.6 ng/mL to 4 ng/mL; 1.6 (95% CI, 1.41-1.82) for men with a PSA level of 10.1 ng/mL to 20 ng/mL; 2.08 (95% CI, 1.81-2.38) for men with a PSA level of 20.1 ng/mL 40 ng/mL; and 3.23 (95% CI, 2.85-3.65) for men with greater than a 40 ng/mL PSA level.

A PSA of 2.5 ng/mL or less appeared to be the only level significantly associated with an increased risk for prostate cancer-specific mortality among men with a Gleason score of 8 to 10 (P < .001).

The researchers acknowledged study limitations. Because SEER replaces biopsy Gleason scores with prostatectomy Gleason scores among surgically treated patients, it is possible that the researchers underestimated the extent of prostate cancer-specific mortality in this patient population. Further, they noted that SEER does not report PSA density or doubling time.

“Despite its potential limitations, this study provides strong evidence to support the notion that low PSA levels in Gleason score 8 to 10 disease may be a sign of underlying aggressive and extremely poorly differentiated or anaplastic low PSA-producing tumors,” Nguyen and colleagues wrote. “Clinicians should be aware of these findings, and patients with low-PSA, Gleason score 8 to 10 disease should be considered for systemic staging and clinical trials studying the use of chemotherapy and other novel agents for very high-risk prostate cancers.” – by Cameron Kelsall

Disclosure: Nguyen reports consultant roles with GenomeDx and Medivation. The other researchers report no relevant financial disclosures.