乌鸦传媒

ORLANDO -- Men with castration-resistant prostate cancer (CRPC) on treatment with enzalutamide had worse survival when given corticosteroids, regardless of their primary treatment, a post-hoc analysis of a randomized trial showed.

Patients on corticosteroids during treatment with enzalutamide (Xtandi) or placebo had a median overall survival of 10.8 months compared with 18.3 months for men not on steroids.

The data showed a survival decrement associated with corticosteroid use in both treatment arms, although patients allocated to enzalutamide fared significantly better, consistent with the trial's overall results.

Other key outcomes, such as progression-free survival (PFS), also were worse when steroids were used, and grade 3/4 adverse events occurred more frequently, Howard Scher, MD, reported here at the Genitourinary Cancers Symposium.

"The inferior outcomes of patients receiving corticosteroids may be due to either biologic properties of the tumor, such as 'promiscuous' androgen receptors, or unknown confounders," said Scher, of Memorial Sloan-Kettering Cancer Center in New York City.

Despite steroids' adverse effect on clinical outcomes, "enzalutamide was consistently superior to placebo with respect to overall survival, radiographic progression-free survival (rPFS), and time to PSA progression, after accounting for known prognostic and other factors."

The post-hoc analysis evolved from the results of a 1,200-patient randomized trial of enzalutamide versus placebo (2:1 randomization) for post-chemotherapy CRPC. The primary results showed almost a 5-month improvement in overall survival with enzalutamide (N Engl J Med 2012; 367: 1187-1197).

Investigators in the trial found that patients on corticosteroids at baseline had substantially inferior survival compared with patients who were not on corticosteroids at randomization. A multivariate analysis of the data showed that baseline corticosteroid use was an independent predictor of decreased survival.

Despite widespread use of steroids in cancer treatment, several lines of evidence suggest that corticosteroids might have the potential to stimulate prostate cancer growth, said Scher. Potential mechanisms include:

• Activation of promiscuous androgen receptors
• Stimulation of human SGK1 expression
• Promotion of IL-6 expression
• Activation of glucocorticoid receptor signaling

Scher and colleagues continued the evaluation of corticosteroids' effects on outcomes in the trial by comparing results in patients who received steroids during the study with those who did not. In particular, if they found a relationship between glucocorticoid use and survival, they wanted to determine whether patients on steroids still benefited from treatment with enzalutamide.

At randomization, about 30% of patients were already on corticosteroids. During the trial, the use increased to 47.6% of patients in the enzalutamide arm and 44.6% in the placebo arm, which included patients who were on steroids at baseline.

"Patients who received corticosteroids were generally sicker and had more advanced disease," said Scher.

On-study use of corticosteroids was associated with higher PSA values, more visceral involvement, and a greater burden of bone metastases.

Any use of corticosteroids (baseline + on study) was associated with worse overall survival (11.5 months versus not yet reached, P<0.0001), reduced rPFS (4.7 versus 8.3 months, P<0.0001), and a shorter time to PSA progression (5.4 versus 8.3 months, P<0.0001).

An analysis of corticosteroid use by treatment assignment showed superior median overall survival with enzalutamide among patients who received no steroids (not yet reached versus 18.8 months, HR 0.504, P<0.0001) and in patients who were treated with steroids (12.8 months versus 9.6, HR 0.685, P=0.001).

Enzalutamide also conferred a significant advantage with respect to rPFS: 11.1 months versus 3 without steroids, HR 0.225, P<0.0001; 5.6 months versus 2.9 with steroids, HR 0.598, P<0.0001.

Patients also fared better with enzalutamide with respect to the time to PSA progression, irrespective of steroid use: 8.6 months versus 2.9 without steroids, HR 0.145, P<0.0001; 5.6 months versus 3.1 with steroids, HR 0.443, P<0.0001.

Use of steroids during the study was associated with almost twice as many grade 3/4 treatment-emergent adverse events: 63.3% versus 34.4%. The most common adverse events were anemia, fatigue, spinal cord compression, and back pain.

Invited discussant William Oh, MD, of Mount Sinai Medical Center in New York City, suggested that much of the explanation for the inferior results in steroid-treated patients can be traced to poorer health.

"A surprisingly high number of patients were on steroids, and they were much sicker compared with patients who were not on steroids," said Oh.

As for the superior results with enzalutamide, irrespective of steroid use, Oh said, "If androgen receptors and glucocorticoid receptors interact, then enzalutamide should have a differential effect."

The Genitourinary Cancers Symposium is cosponsored by the American Society of Clinical Oncology, the American Society for Radiation Oncology, and the Society of Urologic Oncology.

Comment