Table 2. Secondary Efficacy End Points.
| End point | No. (%) | |
|---|---|---|
| Enzalutamide | AS | |
| Incidence of negative biopsy result at 1 y, No. | 114 | 113 |
| Biopsy result | ||
| Negative | 40 (35.1) | 16 (14.2) |
| Positive | 53 (46.5) | 74 (65.5) |
| Unknown | 21 (18.4) | 23 (20.4) |
| Missing | 11 (9.6) | 2 (1.8) |
| Discontinued | 10 (8.8) | 21 (18.6) |
| OR (95% CI)a,b | 3.5 (1.76-6.92) | |
| P valuec | <.001 | |
| Incidence of negative biopsy result at 2 y, No. | 100 | 83 |
| Biopsy result | ||
| Negative | 19 (19.0) | 10 (12.0) |
| Positive | 47 (47.0) | 40 (48.2) |
| Unknown | 34 (34.0) | 33 (39.8) |
| Missing | 11 (11.0) | 5 (6.0) |
| Discontinued | 23 (23.0) | 26 (31.3) |
| OR (95% CI)a,b | 1.6 (0.66-4.00) | |
| P valuec | .29 | |
| Percentage of cancer-positive cores at 1 y, No.d | 90 | 81 |
| LS change from baseline, mean (SE) [95% CI]e | −12.97 (1.99) [−16.9 to −9.03] | −2.90 (2.03) [−6.93 to 1.12] |
| Difference in LS, mean (SE) [95% CI] | −10.07 (2.40) [−14.79 to −5.34] | |
| P valuef | <.001 | |
| Percentage of cancer-positive cores at 2 y, No.d | 53 | 42 |
| LS change from baseline, mean (SE) [95% CI]e | −6.68 (2.37) [−11.36 to −2.00] | −1.53 (2.57) [−6.61 to 3.55] |
| Difference in LS, mean (SE) [95% CI] | −5.15 (3.17) [−11.40 to 1.11] | |
| P valuef | .11 | |
| Incidence of a secondary rise in serum PSA at 1 y, No. | 114 | 113 |
| PSA responseg | 28 (24.6) | 78 (69.0) |
| OR (95% CI)a,b | 0.1 (0.08-0.26) | |
| P valuec | <.001 | |
| Incidence of a secondary rise in serum PSA levels at 2 y, No. | 100 | 83 |
| PSA responseg | 92 (92.0) | 77 (92.8) |
| OR (95% CI)a,b | 1.1 (0.37-3.53) | |
| P valuec | 0.81 | |
Abbreviations: AS, active surveillance; LS, least squares; OR, odds ratio; PSA, prostate-specific antigen.
Calculated using an exact logistic regression model with treatment group, stratification factors, age, race, and time since prostate cancer diagnosis as fixed effects and study site and patient as random effects.
Calculated based on exact binomial distribution.
Calculated based on exact binomial distribution from the logistic regression model.
Analyzed using a mixed-mode–repeated-measures model, with treatment group, stratification factors, visit, visit-by-treatment, and baseline score as fixed effects and study site and patient as random effects.
Most recent biopsy taken during the 6 months before screening.
A Bonferroni-Holm test was used to adjust for multiplicity.
PSA response was defined as a secondary rise in serum PSA levels of 25% or more of the baseline, an increase of 25% or more than the nadir, or an absolute increase of 2 or more ng/mL (to convert to μg/L, multiply by 1).