ORLANDO, FL—A six-gene blood test may help to improve the detection of prostate cancer, a case-control study suggests.
Used in conjunction with standard age-adjusted prostate-specific antigen criterion, the genetic assay outperformed PSA testing alone in discriminating between patients who had cancer and those who did not, said Robert W. Ross, MD, Attending Physician in the Lank Center for Genitourinary Oncology at the Dana- Farber Cancer Institute and Instructor in Medicine at Harvard Medical School.
When used with a PSA test, the gene screen accurately detected prostate cancer in more than 90% of cases, he reported here at the ASCO Annual Meeting.
In contrast, age-adjusted PSA alone had a sensitivity of less than 70% in detecting prostate cancer, the study showed.
The study was supported in part by Source MDx, which developed the new test.
Dr. Ross cited the two recent studies in The New England Journal of Medicine, the latest to seriously question the value of PSA-based methods for prostate-cancer screening, concluding that for every one life saved by PSA screening, 49 men suffer harm (Andriole et al: NEJM 2009;360:1310-1319 and Schroder et al: NEJM 2009;360: 1320–1328).
“Each year in the US, over one million men undergo the anxiety and pain of prostate biopsies that are negative for prostate cancer at a considerable psychological and social cost,” he said. The hope is that the new genetic test can help men avoid the pain, discomfort, and anxiety of unnecessary biopsies.
Developing the Test
To develop the test, Dr. Ross and colleagues obtained blood samples from three prospective cohorts of men who consented to whole-blood RNA transcription profiling: 204 men with newly diagnosed, localized, untreated prostate cancer; 170 healthy men without prostate cancer, and 30 otherwise healthy men with benign prostatic hypertrophy (BPH). The median age of the men was 61.
The RNA was converted to complementary DNA (cDNA). Then, using optimized polymerase chain reaction technology, 392 candidate genes associated with inflammation, prostate cancer, other types of cancer, and the epidermal-growth-factor receptor-1 were assayed in a training set of 76 healthy men and 76 with prostate cancer.
Using logistic regression methods, the researchers narrowed the pool to just six genes that as a group could discriminate between patients with prostate cancer and men without the disease. Five of the genes have lower expression in patients with cancer and one has greater expression, Dr. Ross explained.
The six-gene model was then validated in a cohort of 128 men with cancer and 94 healthy men. In the validation cohort, the six-gene test had a sensitivity of about 86%, compared with 69.5% for age-adjusted PSA.
The specificity of the six-gene test was 83%, compared with about 94% for PSA, he said.
PSA + Gene Test
The addition of PSA results to the six-gene model further improved the diagnosis of prostate cancer: The sensitivity increased to 87.5% and the specificity to almost 93%.
“Further study also showed that the six-gene model retains its superiority over age-adjusted PSA alone when BPH subjects are included with normal subjects,” Dr. Ross said.
In a second validation set of 128 men with cancer, 94 healthy men, and 30 men with BPH, the sensitivity of the six-gene test plus age-adjusted PSA was 87.5%, compared with about 70% for age-adjusted PSA alone. The specificity of the gene test was 91% vs 93% for PSA alone.
The best results were obtained when the researchers combined the training and validation datasets. In this model, use of the six-gene test plus age-adjusted PSA testing correctly predicted almost 94% of prostate cancer patients and almost 95% of healthy subjects.
“The six-gene whole blood-based diagnostic test can be a powerful tool in the management of prostate cancer to both reduce unnecessary biopsies in patients with prostate cancer and detect prostate cancer in patients with PSA values below the current cutoff,” Dr. Ross said.
Still, because this was a case-control study, the results need to be validated prospectively, he said.
To that end, his institution and others are planning a prospective study of 1,000 patients who meet PSA criteria for a biopsy, but have not yet had the test. The goal is to determine if the test can predict the results of biopsy, he said.
Howard M. Sandler, MD, Chairman of Radiation Oncology at the Samuel Oschin Comprehensive Cancer Institute at Cedars-Sinai Medical Center in Los Angeles, said that PSA testing remains “incredibly valuable” for prostate cancer screening, but that the gene assay “could be another test that helps to improve early detection of the disease.
“But the problem with all these tests,” Dr. Sandler said, “is their inability to answer the question we really want to know: Does the patient have potentially lethal cancer, or does he have cancer that will never kill him? This is one of several new tests that can detect prostate cancer better than PSA. But [the researchers] didn't show that the test improves the detection of clinically important prostate cancers.”
Dr. Ross said the test may be able to distinguish between slow and faster growing cancers, but that further testing is needed before that claim can be made.