Mediterranean diet, active surveillance, and low-risk prostate cancer: the details


As we advised readers yesterday, a newly published study by Gregg et al. from the M. D. Anderson Cancer Center has given some clear indications that men diagnosed with lower-risk forms of prostate cancer who are initially managed on active surveillance (AS) can benefit — in terms of time to disease progression — from what is known as the Mediterranean-type diet. However, now that we have read the entire article with care and spoken with the lead author, we feel we need to make some things very clear about this study.

First, this was a study based on data from a cohort of > 550 patients who had been followed, since 2006, exclusively at the M. D. Anderson Cancer Center (just like other large cohorts from centers such as Sunnybrook, Canada; the University of California, San Francisco; Memorial Sloan-Kettering Cancer Center; etc.). So although this was a study based on a prospectively enrolled and followed cohort of patients, it was not any sort of prospective, randomized trial.

Second, what Gregg and his colleagues did was to carry out a sophisticated assessment of the normal diets of about 500 of these patients at baseline, i.e., when they were initially entered into the study, and classify them as falling into one of 10 dietary subsets (explained in more detail below). They made no attempt to modify these patients’ diets over time. Nor did they encourage them to start using a Mediterranean diet (MD) after the patients were enrolled into the M. D. Anderson AS cohort.

As a consequence, what Gregg and his colleagues have been able to show as their key finding is that men diagnosed with lower-risk forms of prostate cancer (Grade Group 1 or 2), who were initially managed on AS, and who were followed over time for disease progression, and whose diets already met criteria for being MDs at the time of study entry, appeared to have longer times to disease progression than men whose diets did not meet the criteria for being MDs.

What Gregg et al. have not shown — although this may well be possible — is that among a similar set of men diagnosed with lower-risk forms of prostate cancer who changed their diets to meet criteria for MDs at the time of study entry have longer times to disease progression than men who remain on “normal” diets, with higher levels of red meat, dairy products, etc.

So let’s look at some of the details.

Gregg et al. started with a cohort of 560 patients enrolled into their AS study cohort. These patients were enrolled between February 2006 and February 2012. Among these 560  patients, 501 provided a sufficiently complete response to a baseline food frequency questionnaire (FFQ). After the exclusion of a number of these patienst for a variety of reasonable reasons, the actual study cohort included 410 patients who were followed for a minimum of 6 months and for whom a MD score analysis had been completed.

What was the MD score analysis? This is complicated, but basically — for each patient — the authors developed an individual dietary score based on responses to the FFQ and on a whole bunch of other standard dietary data. Those individual scores could range from 0 to 9, and the scores that best met the criteria for a Mediterranean-type diet were those in the “High” score group (i.e., 6 to 9) as opposed to those in the “Low” (0 to 3) or “Medium” (4 or 5) score groups. Your sitemaster is no expert on this sort of dietary analysis, but it seems reasonable to him having read the entire paper, and it was all carried out by experts in dietary sciences.

Gregg et al. then looked at the long-term outcomes of these 410 patients and compared them to each other based on the MD score groups computed at baseline. The patients were followed up to December 31, 2016. Here is what was found:

  • The overall average (median) MS score was 4.
  • The average (median) patient age was  64.4 ± 8.4 years.
  • 358/410 patients (87.3 percent) had Grade Group1 disease at diagnosis.
  • 297/410 patients (72.4 percent) had a single core that was positive for prostate cancer at diagnosis.
  • Low body mass indexes (BMIs), high testosterone levels, and low added sugar intakes were all associated with higher MD scores (on univariate analysis).
  • Over an average (median) follow-up of 36 months (range, 6 to126 months),
    • 76/410 patients (18.5 percent) experienced grade progression,
    • 12/410 patients died of causes other than prostate cancer and showed no documented signs of prostate cancer progression.

After adjustments for age and clinical characteristics, the authors also noted:

  • A “suggestive inverse association” between a high baseline MD score (6 to 10)  and a lower risk for Grade Group progression (hazard ratio [HR] for a high MD score vs a low MD score was 0.68).
  • Findings were consistent among men who had Grade Group 1 prostate cancer at baseline
  • For every 1-unit increase in the MD score, there was a >10 percent lower risk of Grade Group progression (HR = 0.88).
  • An increased MD score was associated with improved progression-free survival among
    • Men who did not have diabetes
    • Men who identified as non-White (HR = 0.64)
    • Men with high testosterone levels (HR = 0.83)
  • Men with a high MD score who did not use a statin (i.e., a drug to lower their cholesterol level) also had a potential improvement in their progression-free survival compare to men with low or medium MD scores

The authors state that, based on this study data:

Baseline adherence to the MD appears to be associated with a lower risk of GG progression in men with localized [prostate cancer] managed on AS. This would suggest that consistently following a diet rich in plant foods and fish with a healthy balance of monounsaturated fats may be beneficial for men diagnosed with early-stage [prostate cancer].

This is definitely an interesting set of findings, and it suggests that men who normally have healthy, Mediterranean-type diets, and then get diagnosed with lower-risk forms of prostate cancer will generally have longer progression-free survival times than men who eat “normal” diets with higher levels of red meat, dairy products, etc. Whether men diagnosed with low-risk prostate cancer on AS who change their diets to MDs at diagnosis can extend the time until disease progression remains to be proven.

Here are links to relevant resources, but one really needs to be able to read the full paper if one wants to fully appreciate the nuances of this study:

  • The abstract of the published paper can be found here.
  • A media release from M. D. Anderson Cancer Center about this paper can be found here, but this media release does not make clear that there was no attempt to change the diets of the patients.
  • A summary of the trial protocol for the M. D. Anderson AS cohort study can be found here.

Editorial note: The “New” Prostate Cancer Link thanks Justin Gregg, MD, Carrie Daniel, PhD, and Lany Kimmons at the M. D. Anderson Cancer Center for promptly providing us with a full-text copy of this paper. We also thank Dr. Gregg for kindly taking time on the phone to explain exactly what hadn’t been done over the course of this trial so that we could accurately interpret the findings.

4 Responses

  1. Curious if there was an age difference at initial diagnosis between MD scores. Or was there an association with high MD score and delay in onset? Very loose association, if any, I am sure with that study size.

  2. I wonder do they indicate what constitutes a high level of testosterone?

  3. Dear Mike H:

    The whole point of this paper is that there was quite a strong association between a high MD score at time of diagnosis and longer time to disease progression (which I think is what you mean by “delay in onset”). I do not believe there was any meaningful relationship between age at initial diagnosis and MD scores.

  4. John:

    What the paper says is that average serum T levels were higher in the men with the high MD scores.

    Specifically, the average serum T scores were: 379 ± 136 ng/dl among 141 men with the low MD scores; 408 ± 155 ng/dl among 171 men with the medium MD scores; and 444 ± 166 ng/dl among 98 men with the high MD scores.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: