An assay for nuclear-localized androgen receptor splice variant 7 protein in circulating tumor cells identified patients with metastatic castration-resistant prostate cancer who had better outcomes from treatment with taxanes, and those who did better with androgen receptor-signaling inhibitors.
Howard I. Scher, MD
An assay for nuclear-localized androgen receptor splice variant 7 (AR-V7) protein in circulating tumor cells (CTCs) identified patients with metastatic castration-resistant prostate cancer (mCRPC) who had better outcomes from treatment with taxanes, and those who did better with androgen receptor-signaling (ARS) inhibitors. AR-V7—positive patients had superior outcomes with taxanes while AR-V7–negative patents had better outcomes with ARS inhibitor treatment.1
For patients who were positive for AR-V7 according to the Oncotype DX AR-V7 Nucleus Detect test, developed by Epic Sciences, the median overall survival (OS) was 14.3 months with taxane-based chemotherapy compared with 7.3 months for those who received ARS inhibitors (HR, 0.62; 95% CI, 0.28-1.39; P  =  .25).
Conversely, patients who were negative for AR-V7 had a superior median OS when treated with ARS inhibitors (19.8 vs. 12.8 months; HR, 1.67; 95% CI, 1.00-2.81; P  =  .05).
Investigators concluded that the assay may help physicians choose the appropriate treatment for this patient population.
“This liquid biopsy test addresses a critical unmet need at a decision point in management to predict and select the therapy that is most likely to extend a patient's life,” lead author Howard Scher, MD, co-chair of the Center for Mechanism Based Therapy and head of the Biomarker Development Initiative at Memorial Sloan Kettering Cancer Center, said in a statement.
“During the treatment of metastatic prostate cancer, physicians will now be able to use AR-V7 status to determine when a patient's cancer has become resistant to androgen receptor—directed therapy and will respond better to chemotherapy, enabling the patient to live longer,” he added.
Previous study results have shown that AR-V7 is linked with resistance to enzalutamide (Xtandi) and abiraterone acetate (Zytiga), ARS inhibitors demonstrated to improve survival in this patient population.2 Approximately 10% to 20% of chemotherapy-naïve men are refractory to ARS inhibitors.
Scher et al included blood samples from 142 patients undergoing a change in systemic therapy for progressive disease. Seventy-two samples were collected before initiation of taxane therapy and 70 before the start of therapy with an ARS inhibitor. Seventy men were deemed high risk per standard prognostic factors.
Patients were not randomly assigned to treatment, but treating physicians were blinded to the patient’s AR-V7 status and CTC count. Physicians elected to use ARS inhibitors more often as a second-line treatment and use taxanes for later lines of therapy.
Scher et al found that median age, hemoglobin levels, albumin levels, and the presence of lung and/or liver metastases before therapy did not affect treatment choice. However, patients in the taxane arm had higher levels of lactate dehydrogenase (245.0 vs 203.0 U/L; P  < .001), PSA (133.4 vs 30.3 ng/mL; P  < .001), and alkaline phosphatase (126.5 vs 9.5 U/L; P  =  .01).
Rates of AR-V7 positivity were similar between the validation (34 of 142; 23.9%) and training (31 of 123; 25.2%) cohorts.
Overall, patients in the ARS inhibitor arm had superior median survival time (16.0 vs 12.9 months), but investigators did not observe a significant difference in survival rates between the treatment groups (P  =  .18).
Investigators developed a patient-specific risk score from the training cohort and calculated estimated median survival based on AR-V7 status, risk score, and choice of treatment. The median risk score was —0.632.
Among high-risk AR-V7—negative patients, the median OS was superior for those treated with ARS inhibitors (16.9 vs 9.7 months; HR, 2.38; 95% CI, 1.12-5.06; P  =  .02). Taxanes induced superior median OS among high-risk AR-V7—positive patients (14.3 vs 5.6 months; HR 0.35; 95% CI, 0.14-0.88; P = .03).
There were not enough low-risk, AR-V7—positive patients to evaluate survival rates by biomarker result.
Study coauthor Ryan Dittamore, chief of medical innovation at Epic Sciences, said in a statement that investigators developed this study specifically to help patients and physicians when making treatment decisions.
“This question weighs heavily on doctors and patients, but now, with the Oncotype DX AR-V7 test, we can provide them the confidence to know whether continuing with hormonal therapy or switching to chemotherapy will result in better survival outcomes,” he said. “In addition, the survival benefit gained through the utilization of our AR-V7 test could make the test as valuable to a patient's outcome as a blockbuster cancer drug.”
In an accompanying editorial, Stephen R. Plymate, MD, Adam Sharp, MD, PhD, and Johann S. de Bono, MD, PhD, wrote that the treatment landscape for prostate cancer is changing rapidly, with recent approvals for abiraterone for treatment-naïve patients with metastatic disease and apalutamide (Erleada) for nonmetastatic CRPC. The development of predictive biomarkers is “a high priority,” they wrote, but could not conclude that these results show that the assay is predictive.3
“Studies must correlate response to treatment with assay positivity, and not just survival data, to ensure that the assay is not simply a prognostic biomarker,” they wrote. “AR-V7 positivity, in this study, is associated with higher lactate dehydrogenase, alkaline phosphatase, and prostate-specific antigen levels, suggesting a higher disease burden in the taxane arm. This finding indicates that AR-V7 positivity by this assay may be more prognostic, associated with disease burden, than predictive.”
Results presented at the at the 2018 ASCO Annual Meeting suggest that the Oncotype DX assay may, in fact, be predictive.
Andrew J. Armstrong, MD, MSc, a medical oncologist with Duke Cancer Institute, presented findings from the PROPHECY (NCT02269982) trial. That study was designed to assess whether CTC AR-V7 detection was an indicator of aggressive disease and high tumor burden or a biomarker for clinical efficiency in men with mCRPC being treated with enzalutamide and/or abiraterone.4
A total of 118 men were included, 45% had ≥5 CTS per 7.5 mL of blood according to the Cellsearch CTC Test and 36% had previously received enzalutamide and/or abiraterone. A Cellsearch finding of ≥3 CTS per 7.5 mL of blood is an indicator for shorter survival.
In the study, 59 men received enzalutamide, 56 received abiraterone, and 3 received both.
As assessed by both the Hopkins AR-V7 AdnaTest AR-V7 assay and the Oncotype DX assay, AR-V7 positivity was associated with shorter radiographic progression-free survival (PFS) and shorter OS. With the AdnaTest, the median PFS was 3.1 months for AR-V7—positive patients compared with 6.9 months for AR-V7–negative status (HR, 2.4; 95% CI, 1.5-3.7). The median OS was 10.8 versus 27.2 months, respectively (HR, 3.9; 95% CI, 2.2-6.9).
With AR-V7 status assessed by the Oncotype DX assay, outcomes for median PFS (6.1 vs 3.1; HR, 2.5; 1.3-4.7) and median OS (20.3 vs 8.4 months; HR, 4.5; 95% CI, 2.1-9.8) also favored AR-V7—negative patients.
Concordance between the assays was 82%.
Armstrong said that both assays predicted poor outcomes in these patients, irrespective of treatment and number of CTCs, meeting the primary endpoint.