Inaccurate Quote

In Reply Dr Kerley raises 4 concerns with the design and conduct of the MEAL trial: (1) evidence of dietary changes in the control group; (2) a potential placebo effect in the control group; (3) an inability to achieve predefined dietary targets in the intervention group; and (4) a failure of the intervention to incorporate fat intake. We believe these issues do not change the validity of our conclusions. First, while we observed small changes from baseline in some dietary components in the control group, most were small and insignificant compared with the intervention group and did not last beyond 12 months. Moreover, at least 1 change was counterproductive: after 24 months, control participants were eating fewer tomatoes compared with at baseline.1 Second, to our knowledge, there exists no proven placebo effect for any type of prostate cancer treatment; the cited study, which focused on PSA outcomes in patients with advanced prostate cancer, provided no evidence of it. Fluctuations in PSA values in the control groups of this study were small, consistent with expected event rates, and attributable to regression to the mean.2 Third, predefined targets for behavior change are aspirational, and the intervention produced clinically meaningful changes with very large between-group differences. Intervention participants increased vegetable consumption by 60%, including a 2.5-fold increase in cruciferous vegetables. Circulating lycopene levels in these patients increased by 95%. Fourth, although we did not target saturated fat, intervention participants nevertheless consumed 15% less of it after 12 months compared with controls, a significantly larger decline from baseline that was sustained over 24 months. Moreover, the 2 cited studies associating fat intake with prostate cancer progression incorporated outcomes of unproven clinical significance: for example, in vitro growth suppression of prostate cancer cell lines.3,4 Dr Csizmadi and colleagues suggest that vegetable consumption may not be linearly related to prostate cancer progression, and that patients with lower vegetable intake may have potentially benefited. However, post hoc analyses stratified by baseline vegetable intake demonstrated no differences in clinical outcomes among participants who consumed fewer vegetables. At baseline, 43% of participants reported less than 3 daily servings of vegetables, with no between-group differences (P = .64). For PSA-focused study outcomes (time to PSA >10 ng/mL or PSA doubling time), the unadjusted hazard ratio was 0.86 (95% CI, 0.65-1.13). This hazard ratio remained unchanged when adjusted for baseline intakes of total vegetables and fruits and vegetables. Thus, regardless of baseline dietary intake, our study does not support the hypothesis that vegetable or fruit intakes modify clinical progression.