SAN FRANCISCO – Genomic tests that help make treatment decisions for prostate cancer often pointed in different directions when the same patient was examined, as a small retrospective study showed
The Decipher and Prolaris tests yielded a third of the results Discordant The timing at which patients were evaluated who met the criteria of the National Comprehensive Cancer Network (NCCN) for active surveillance. The Prolaris and Oncotype DX tests did not agree a quarter of the time. To identify only two patients, the decipher and the oncotype DX agreed with one and disagreed with the other.
The tests also differed a bit in how often they supported the NCCN criteria for active surveillance. Although limited to a total of 22 patients, the results underlined the complexity of using genomic tests to select patients for active monitoring, Joseph R. Wagner, MD, of Hartford Hospital, Connecticut, said here at the American Urological Association ( AUA).
"What we can say from these results is that there are remarkable differences in the favorable prognostic results obtained by Oncotype DX, Prolaris and Decipher," said Wagner during an AUA press conference. "For patients who are candidates for active surveillance, Prolaris is most likely to support this option, and the results were most consistent with Prolaris and Oncotype DX." Stressing that the results did not suggest that one test is superior to the others, Wagner said, "This is less a scientific presentation and more a thought-provoking presentation."
The analysis had its origin in two changes to the recommendations in 201
Each test uses different types of information to produce results that affect various aspects of the decision-making process. Prolaris estimates the 10-year risk of prostate cancer death with active surveillance, the 10-year risk of distant metastases with definitive treatment, and the patient's standing in the AUA risk groups. Decipher and Oncotype DX provide information about a patient's 10-year risk for prostate death or distant metastasis after surgery and the likelihood that the patient has a high-grade disease (grade ≥3) or extracapsular extension by pathology.
Relatively Little Comparison Information about the three tests are available. Given that clinicians can order one of the approved tests (with the exception of the hospital or clinic guidelines), Wagner and colleagues attempted to collect data that could be helpful in the selection of tests.
The follow-up of medical records for newly added patients diagnosed prostate cancer from 2014 to 2017 and who was evaluated with at least two of the genomic tests. Multi-testing was on a patient request, not the attending physicians.
The review identified 22 patients, all but two of whom met the low or medium low risk NCCN criteria and were considered candidates for active surveillance. Records showed that 12 patients had ratings from the Decipher and Prolaris tests, which gave consistent results for active monitoring in eight of the 12 cases. Eight patients had both the Prolaris and Oncotype DX tests, which agreed on the adequacy of active monitoring for six of the patients.
The investigators also compared the results of each test with the NCCN risk assessment for each patient. The results of Prolaris and NCCN were consistent in 15 out of 20 patients (75%). The decryption and NCCN criteria were consistent in nine out of 15 patients (60%), and the Oncotype DX and NCCN were consistent in five out of ten patients (50%).
Wagner and his colleagues also performed a kappa analysis to improve the estimate match between each test and the NCCN criteria (kappa score ≥ 0.6 reflects moderate agreement). Comparison of Prolaris and NCCN (n = 20) gave a kappa value of 0.21 and the decipher NCCN comparison (n = 15) gave a value of 0.15. Too few patients (N = 10) were included in the comparison of Oncotype DX and NCCN to calculate the kappa value.
Looking to the future, Wagner suggested possible improvements to the test criteria to improve the performance of each test in the clinical setting. Examples: Include mutation patterns to determine if a patient can be better treated by a particular type of local therapy; Integrate results for radiotherapy and active monitoring into Oncotype DX and Decipher; and the addition of separate results for radiotherapy and surgery with Prolaris
Regarding the decision to use which test, Wagner said he explains to a patient the type of information that each test offers and then asks, "What do you want to know? "
" For most of us, we can choose which test to order or the patient can choose, but there is very little reason to choose one over another, "said press release host Stacy Loeb, MD, from NYU Langone Medical Center. "When they started, they had completely different endpoints … but over time, the reports have pretty much approached."
Wagner announced a relationship with Genomic Health, which markets Oncotype DX.
2018-05 -18T17: 45: 00-0400