Effect of Postdiagnosis Diet and Lifestyle on Clinical Outcomes in Prostate Cancer Survivors: A Systematic Review.

METHODS
The study protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) database (Record ID: CRD42023403249) and performed in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines.3
This review summarizes findings from publications evaluating postdiagnostic dietary patterns and PCa recurrence, progression, or mortality. Among these, if the paper also reported on all-cause mortality (ACM), we include comment on those results. Given that almost no studies evaluated postdiagnostic effects of diets before 2005, we used 2005 as the earliest year for our search. Two authors (V.L. + R.A.) reviewed titles and abstracts of papers published from January 1, 2005, to May 3, 2025, indexed by the databases PubMed (PubMed.gov), Embase (Elsevier), and/or the Cochrane Library (Wiley). We also reviewed reference lists of published systematic reviews for additional studies.
Eligible studies included adult men (18+ years old) diagnosed with PCa. In addition to trials that consisted of dietary changes only, trials that incorporated both dietary changes and supplements were allowed. Studies were excluded if they evaluated a single dietary factor or nutrient in isolation, if exposure was not at an individual level (eg, country level behaviors), or if they assessed prediagnostic diet or lifestyle exposures. Articles not written in English were also excluded.
Predefined data extraction tables were used to summarize the study design, participant characteristics, and dietary exposure(s). The risk of bias was evaluated independently by 2 reviewers (V.L. + R.A.). In cases of disagreement, a third reviewer (S.K.) was consulted to reach a consensus. The Revised Tool for Risk of Bias in Randomized Trials (ROB 2.0) was used for RCTs,4 while the Risk Of Bias In Nonrandomized Studies—Exposures (ROBINS-E) tool was applied to cohort studies.5

RESULTS
The search resulted in 648 citations after removal of duplicates. After screening the titles and abstracts, 65 full-text articles were assessed for eligibility. The Figure shows article identification, screening, and eligibility. Table 1 summarizes the postdiagnostic dietary patterns, lifestyle scores, and related indices. Table 2 summarizes the characteristics and statistically significant findings of the 21 unique studies included in this review (10 RCTs and 11 prospective cohort studies). Table 3 summarizes studies as “inverse,” “null,” and “positive” for each exposure. Tables 4 and 5 summarize the risk of bias for each study. Seven RCTs were rated as having a low risk of bias, while the remaining 3 RCTs were rated as having moderate risk of bias. All 11 cohort studies were rated as having moderate risk of bias.

Overall Healthful Diets
Four studies examined postdiagnostic consumption of overall “healthful diets”: the Healthy Eating Index (HEI-2015),14,23,24 alternative Healthy Eating Index-2010,24 Prudent diet,26 and Dietary Approaches to Stop Hypertension (DASH) diet.23,24 Findings were mixed.14,23,26
Observational studies by Schenk et al23 and Gregg et al14 evaluated the HEI-201528 in relation to risk of Gleason grade progression in men with low-risk PCa on active surveillance (AS). Gregg et al14 found a nonsignificant trend (HRtertile 3 (healthier diet) vs 1: 0.59; 95% CI: 0.32, 1.08; p-trend = 0.06) among 411 men, whereas Schenk et al23 did not find evidence of an association among 564 men in the Canary Prostate AS Study (PASS). The HEI-2015 and alternative HEI 2010 were not associated with PCa-specific mortality (PCSM) in the Multiethnic Cohort (MEC) study; however, each per SD increase was associated with a 10% (HR: 0.90; 95% CI: 0.83, 0.97) and 8% (HR: 0.92; 95% CI: 0.85, 1.00) reduced risk of ACM, respectively.24
An observational analysis by Yang et al assessed the relationship between a Prudent diet and risk of PCSM among 926 men diagnosed with PCa in the Physicians Health Study.26 There was an inverse but nonsignificant association between a Prudent pattern and reduced PCSM (HRquartile 4 vs 1: 0.46; 95% CI: 0.17, 1.24; p-trend = 0.11), and a significant 56% decrease in risk of ACM (HRquartile 4 vs 1: 0.64; 95% CI: 0.44, 0.93; p-trend = 0.02). Finally, Schenk et al23 evaluated the DASH diet in 564 men in PASS and found no association with Gleason grade progression, and the DASH diet was not associated with PCSM or ACM in the MEC.24

Conclusion:
Although some suggestive inverse (beneficial) associations were observed for the Prudent diet and HEI-2015 score and PCa outcomes, these did not reach statistical significance. However, both scores were statistically significantly associated with reduced risk of ACM. Further studies are needed to confirm these associations and potential benefits.

Mediterranean Diets
Four studies examined postdiagnostic consumption of Mediterranean diets (using standard scores)29–31 in relation to PCa outcomes, with varied findings.15,17,23,24
Prospective cohort studies by Gregg et al15 and Schenk et al23 evaluated grade-specific progression. Gregg et al15 observed a borderline inverse trend between high adherence to the Mediterranean diet score (MDS) and lower risk of grade group progression (MDS HRT3 vs T1: 0.67, 95% CI: 0.36, 1.25; p-trend = .05) among 410 men on AS.15 Schenk et al,23 however, observed no association between an alternative Mediterranean diet (aMED) score and grade reclassification in 564 men in PASS.
An observational study by Kenfield et al17 among 4158 men with nonmetastatic PCa observed no association between the MDS or an aMED score and risk of lethal PCa (metastasis or PCa-specific death) or PCSM. However, higher MDS and aMED scores were associated with a 22% lower risk of ACM (MDS HRT3 vs T1: 0.78; 95% CI: 0.67, 0.90; p-trend = 0.0007, aMED HRQ5 vs Q1: 0.78; 95% CI: 0.64, 0.94; p-trend = 0.001). For the aMED score only, the association was modified by southern European ancestry (pint = 0.01); the aMED score was associated with a 32% reduction in risk of ACM for nonsouthern Europeans only (see Table 2, footnote o). Similarly, there was no association for aMED score and PCSM in the MEC (Wang et al24), but aMED score was associated with 22% lower risk of ACM (HRper point increase: 0.88; 95% CI: 0.80, 0.96).

Conclusion:
These findings provide preliminary evidence that higher adherence to a Mediterranean diet may reduce risk of ACM among men with PCa, although there was no relationship with PCSM or grade-specific progression. Further research is needed to examine the Mediterranean diet with clinical PCa outcomes.

Plant-Based Diets
Two of 4 studies evaluating postdiagnostic plant-based diets in relation to prostate cancer outcomes reported potentially beneficial findings.10,20,22,24
Parsons et al22 conducted an RCT on 443 patients with early stage PCa on AS randomized to a 2-year counseling intervention promoting consumption at least 7 daily fruit and vegetable servings or a control group that received written information about diet and PCa. Intervention participants reported significant increases compared with controls in daily total vegetable servings (mean change, 2.43 vs 0.45, P < .001) that persisted at 24-month follow-up (mean change, 2.01 vs 0.37, P < .001). However, the study did not find evidence of reduced risk of clinical progression (PSA level of 10 ng/mL or greater, PSA doubling time (PSADT) less than 3 years, or upgrading defined as increase in tumor volume or grade on follow-up prostate biopsy) compared with the control group (adjusted HRint vs control: 0.97; 95% CI: 0.76, 1.25).
Liu et al20 examined 2 indices (plant-based diet index [PDI] and healthful PDI [hPDI])32,33 and risk of PCa progression and PCSM among 2062 men with nonmetastatic PCa in the Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE) (cohort). Progression was defined as biochemical recurrence, secondary treatment, bone metastases, or PCSM (190 progression events; 61 PCSM events). Participants in the highest vs lowest quintile of PDI had a 47% lower risk of progression (HRQ5 vs Q1: 0.53; 95% CI: 0.37, 0.74; p-trend = 0.003). There was no association with the hPDI overall, but among participants with Gleason grade ≥ 7 at diagnosis, higher hPDI score was associated with 55% lower risk of progression (HRQ5 vs Q1: 0.45; 95% CI: 0.25, 0.81; p-trend = 0.01). There was no association between either dietary index and risk of PCSM. Wang et al24 reported on these same 2 indices and the unhealthy PDI in the MEC study. There was an unexpected statistically significant 2-fold increase in risk of PCSM when comparing quintile 5 vs 1 for the PDI, however, the p-trend and per SD increase were not statistically significant, and hPDI association with PCSM was null. For ACM, higher PDI and hPDI scores were associated with a decreased risk (PDI: HRper SD increase: 0.90 [0.83, 0.97] and hPDI: HRQ5 vs Q1: 0.75; 95% CI: 0.58, 0.97; p-trend = 0.03, respectively), and a higher unhealthy PDI score was associated with an increased risk in the MEC. Inverse associations were also observed in African American and Latino men for PDI with ACM (Table 2 footnotes s and u).
Carmody et al10 conducted an RCT to investigate effects of a plant-based diet plus fish vs a wait-list control group among 36 men with recurrent PCa. There was no significant change in the rate of PSA increase between the groups.

Conclusion:
The results from CaPSURE are promising and support a role for a plant-based diet to slow disease progression, while the large RCT by Parsons et al did not show a beneficial effect among early-stage patients on AS. These data provide preliminary evidence that higher adherence to a plant-based diet may reduce risk of ACM among men with PCa and among racial/ethnic subgroups. Further research is needed examining plant-based diets with clinical PCa outcomes such as PCSM.

Low Carbohydrate Diet and Diets Varying in Quantity and Quality of Dietary Fat
Trials by Aronson et al6 and Demark-Wahnefried et al11 assessed low fat diets and PSA change from baseline to post-intervention as secondary outcomes. Aronson et al6 conducted an RCT of a low-fat, high-fiber, soy protein supplemented diet vs a Western diet for 4 weeks among 18 men on AS to assess the difference in serum stimulated LNCaP cell growth between groups. Subjects in each group were compliant with the prepared food dietary intervention. Demark-Wahnefried et al11 used a 2 × 2 factorial design to randomize participants to the presence or absence of 2 factors, flaxseed supplementation (30g/d) and dietary fat restriction (<20% total energy), at least 21 days before prostatectomy to assess tumor proliferation rate between groups.11 Adherence to flaxseed supplementation was supported by significantly higher lignan intakes and expression in urine and seminal fluid. While patient serum stimulated LNCaP cell growth (Aronson et al6 trial) and proliferation rates in the flaxseed groups (Demark-Wahnefried et al11) were significantly lower in the intervention vs control groups (not the focus of this review), there were no significant differences in PSA change observed between the groups.6,11
Two RCTs by Aronson et al7,8 examined reduction in dietary fat with reduced omega-6 fats (found in corn oil, safflower oil, fried foods, and chips) and increase in omega-3 fat with fish oil supplements. Compliance with the interventions was high in both trials. The first trial (2011), a 4 to 6 weeks preprostatectomy trial in 48 men, found no significant difference in PSA values between the low-fat/fish oil vs Western diet groups.7 In the second trial (2024) among 100 men on AS for 1 year, there was no difference in PSA or Gleason grade progression between the low omega-6, high omega-3 diet vs control group.8 Ki-67 index, a biomarker prognostic of PCa progression, was a secondary outcome in the first trial and the primary outcome in the AS trial and was significantly reduced in the low omega-6, high omega-3 vs control group in both trials.
Freedland et al12 examined a low carbohydrate diet (≤20 g/d) vs control and PSADT in a 6-month RCT in participants with biochemical recurrence after local treatment. The change in carbohydrate intake was significantly different between the arms. The trial stopped early after 45 of 60 planned participants enrolled and an interim analysis showed futility. While the primary outcome of PSADT was null, in a post hoc analysis adjusting for hemoconcentration and baseline covariates, PSADT was significantly longer in the low carbohydrate diet group than in the control group (30 vs 13 months, P = .007).12

Conclusion:
Studies evaluating varying fat and low carbohydrate diets did not find evidence for a delay in PCa progression or changes in PSA or PSADT (except in a post hoc analysis). Trials of longer duration and larger sample sizes will be required to evaluate the effect of these diets on PCa progression.

Proinflammatory + Proinsulinemic Diets
Langlais et al18 assessed postdiagnostic proinflammatory and proinsulinemic dietary patterns in CaPSURE in relation to progression defined as biochemical recurrence, secondary treatment, bone metastases, or PCSM. Men with higher inflammatory and hyperinsulinemia scores had increased risk of PCa progression (HRQ5 vs Q1: 2.61; 95% CI: 1.75, 3.90; p-trend< 0.01 and HRQ5 vs Q1: 1.63; 95% CI: 0.93, 2.86; p-trend = 0.05). However, a dietary pattern associated with insulin resistance had no significant association with PCa progression, and none of the diets were significantly associated with PCSM. Wang et al24 reported on these same patterns in the MEC, and while there were no associations with PCSM, men with higher hyperinsulinemia scores had increased risk of ACM and cardiovascular disease (CVD)–related mortality (HRQ5 vs Q1: 1.37; 95% CI: 1.02, 1.84 and HRQ5 vs Q1: 1.96; 95% CI: 1.15, 3.33, respectively). Dietary patterns associated with inflammation and insulin resistance were not associated with ACM or CVD-related mortality.
Yang et al26 investigated a postdiagnostic Western dietary pattern among 926 men in the Physicians Health Study and observed a positive association between the Western pattern and both PCSM (HRQ4 vs Q1: 2.53; 95% CI: 1.00, 6.42; p-trend = 0.02) and ACM (HRQ4 vs Q1: 1.67; 95% CI: 1.16, 2.42; p-trend = 0.01).

Conclusion:
These studies raise the possibility that diet, through its inflammatory and insulinemic potential, may play a role in PCa progression. Further trials are required to validate these findings.

Diets Combined With Lifestyle
Five observational studies evaluated postdiagnostic lifestyle scores or indices in relation to PCa outcomes in 4 distinct cohorts.9,13,18,19,24 In addition, 3 RCTs examined the impact of targeted dietary, physical activity, or more comprehensive interventions.16,21,25
Graff et al13 evaluated the 2015 health behavior score from Kenfield et al27 and a newly developed 2021 score in the Health Professionals Follow-up Study subcohort of roughly 4500 men with nonmetastatic prostate cancer. The 2021 health behavior score was not associated with risk of PCSM; however, each additional point of the 2015 health behavior score was associated with a 19% reduced risk of fatal PCa (HR: 0.81; 95% CI: 0.68, 0.97) in the fully adjusted multivariable model accounting for prediagnosis behaviors.
In the same report, Graff et al13 also evaluated a score based on Cancer Prevention Recommendations from the World Cancer Research Fund (WCRF) and American Institute for Cancer Research (AICR).34,35 The authors also created a score based on the American Cancer Society (ACS) Nutrition and Physical Activity Guidelines for Cancer Survivors.36,37 Among the same ~4500 men, neither score demonstrated an association with fatal PCa.
Byrne et al9 evaluated a lifestyle index based on 5 of 8 cancer prevention recommendations from the WCRF/AICR score plus smoking status, among 2715 PCa cases in the UK Biobank. A higher lifestyle index representing greater adherence to recommendations was associated with a 30% reduction in PCSM (HR: 0.70; 95% CI: 0.53, 0.93) and 27% reduction in ACM (HR: 0.73; 95% CI: 0.59, 0.89).
Langlais et al19 evaluated the 2015 and 2021 health behavior scores, WCRF/AICR score, and ACS score (+alcohol) in CaPSURE. For each 1-unit increase (ie, healthier) in the 2015 health behavior score, the 2021 score + diet, and the ACS score + alcohol, men had a borderline significant 11% and statistically significant 24% and 11% lower risk of progression (HR = 0.89; 95% CI: 0.80, 1.00; HR = 0.76; 95% CI: 0.63, 0.90; and HR = 0.89; 95% CI: 0.81, 0.98), respectively. The WCRF/AICR Score was also associated with a nonsignificant 17% lower risk of progression per point increase (HR = 0.83; 95% CI: 0.67, 1.02). Moreover, for each 1-unit increase in the 2021 score, the 2021 score + diet, and WCRF/AICR Score, men had a statistically significant 38%, 35%, and 29% lower risk of PCSM (HR = 0.62; 95% CI: 0.45, 0.85, HR = 0.65; 95% CI: 0.45, 0.93, and HR = 0.71; 95% CI: 0.57, 0.89), respectively, but no associations were observed with PCSM for the 2015 score or ACS score.
Wang et al24 also evaluated the 2015 and 2021 health behavior scores among men with PCa from the MEC. The 2021 score was associated with lower risk of PCSM among African American men (HRper point increase: 0.46; 95% CI: 0.24, 0.88) and Latino men (75-81% lower risk of PCSM in score categories of 2 and 2.5 points vs 0-1.5 points) but not for the other race/ethnic groups or in the overall population. The 2015 score was associated with a reduced risk of ACM (HRper point increase: 0.92; 95% CI: 0.87, 0.97), and the 2021 score was associated with a reduced risk of ACM (HRper point increase: 0.69; 95% CI: 0.63, 0.77) and CVD-related mortality (HRper point increase: 0.67; 95% CI: 0.56, 0.79), with similar estimates when adding dietary factors. Findings for racial/ethnic subgroups for ACM and CVD-mortality are presented in Table 2 footnotes s, t, u, and v.
Langlais et al18 evaluated 2 other lifestyle indices—the empirical lifestyle index for hyperinsulinemia (ELIH) and empirical lifestyle index for insulin resistance (ELIR)—in relation to risk of PCa progression in CaPSURE. Both indices were associated with increased risk of progression (ELIH HRQ5 vs Q1: 2.81; 95% CI: 1.78, 4.43; p-trend< 0.01; ELIR HRQ5 vs Q1: 2.43; 95% CI: 1.45, 4.07; p-trend < 0.01) but were null for PCSM. Wang et al24 also evaluated ELIH and ELIR with PCSM and ACM in the MEC and reported null associations.
Ornish et al21 conducted an RCT that investigated effects of a vegan diet supplemented with soy, fish oil, vitamin E, selenium, and vitamin C, moderate aerobic exercise (walking 30 minutes 6 days weekly), stress management techniques 60 minutes daily, and a 1-hour support group once weekly. Among 93 men with PCa on watchful waiting, PSA decreased 4% in the experimental group, but increased 6% in the control group (P = .016) from baseline to the 12-month timepoint. Adherence to the intervention after 1 year was 95% in the experimental group. Changes in serum PSA were significantly associated with degree of change in diet and lifestyle (P = .035).
A 6-month RCT among 117 men who elected AS by Wright et al25 evaluated diet and physical activity modification, which was based on the Diabetes Prevention Program and a goal of 7% weight loss. The primary outcome was change in glucose regulation; secondarily, the investigators measured the effect of the intervention on PCa progression through Gleason grade changes from standard-of-care prostate biopsies. Eighty percent of participants completed all nutrition sessions, and all participants attended both exercise physiology sessions. Although the study met weight loss goals and observed significant improvements in glucose-regulation biomarkers, there was no significant effect on Gleason grade progression.
A 6-month RCT conducted by Hébert et al16 examined the effects of a diet, physical activity, and stress reduction intervention (3-month intensive and 3-month maintenance program) vs control group on PSA levels at 3 and 6 months in 54 men with rising post-treatment PSA. Although significant diet and exercise changes were achieved, there was no effect on PSA levels.

Conclusion:
Observational studies provide evidence that a healthy diet combined with healthy lifestyle behaviors may have potential benefits on PCa outcomes and overall survival, although RCT data are mixed. AS interventions that focus on multiple factors (not only diet) have demonstrated effects on PSA. Future studies incorporating these scores/indices in varied PCa populations with long-term follow-up are needed to confirm these findings.

DISCUSSION
We sought to deepen our understanding on how dietary and lifestyle patterns impact PCa outcomes. Findings were generally favorable for diets emphasizing greater consumption of plant foods (eg, fruits, vegetables, and whole grains) and lower consumption of meats and processed foods. Eating plant foods, adopting prostate-focused recommendations (2021 health behavior score), adopting the ACS diet and lifestyle recommendations, and minimizing consumption of inflammatory foods and those with higher insulinemic potential may potentially lower risk of PCa progression. In addition, adhering to the Prudent and Mediterranean diets and adopting WCRF/AICR recommendations may lower the risk of ACM, while eating foods comprising the Western diet pattern and with higher insulinemic potential may increase risk of ACM. Overall, however, the current evidence for each specific type of diet is limited. Additional studies with larger sample sizes, a greater number of PCSM events, and longer follow-up periods are needed to address key knowledge gaps and gain a more comprehensive understanding of the effectiveness of these dietary + lifestyle approaches among men with varying stages of PCa. There is also a need for validated surrogate biomarkers to assess efficacy in RCTs. Until then, health professionals can guide patients on lifestyle practices that are associated with overall health benefits and that are potentially beneficial for PCa (see Table 6 for practical recommendations, based on this review article).
Significant barriers may prevent men with PCa from maintaining a healthy diet and lifestyle. Key obstacles include difficulty understanding lifestyle guidelines, a lack of knowledge and confidence regarding food choices, insufficient self-discipline, time constraints, a lack of social support for healthy activities, and physical limitations.40,41 Empowering patients to overcome these challenges is crucial for long-term survivorship. Providing credible recommendations, offering dietary counseling, and helping patients set manageable goals can support successful, lasting, dietary and lifestyle modification.41
Strengths of this systematic review are the inclusion of cohort and RCT study designs specifically focused on postdiagnostic diets, dietary + lifestyle scores, and indices in PCa survivors. In addition, the large number of PCa survivors analyzed enhances the generalizability of the findings, making them directly applicable to survivorship care. By excluding studies that assessed prediagnostic dietary intake or prediagnostic diet + lifestyle behaviors, this review isolates the impact of postdiagnostic diet and diet + lifestyle behaviors on PCa-specific outcomes.
Limitations include study differences in dietary assessment methods, exposure definitions, study designs, and racial/ethnic groups studied. Including dietary supplements within a randomized trial design makes it difficult to discern the impact (positive or negative) on lifestyle changes alone. An additional limitation is that, for the studies evaluating for metastases, only conventional imaging technologies were used since prostate-specific membrane antigen (PSMA) PET CT imaging was not yet available. Future studies should consider incorporating PSMA PET imaging as a measure of progression. Although many cohorts accounted for confounders, residual confounding from unmeasured lifestyle factors (eg, physical activity, comorbidities) may influence some of the observed associations. This is noted in our bias assessment (Tables 4 and 5). In addition, this review does not evaluate the specific contributions of individual dietary components to PCa progression. Finally, as with all systematic reviews of published literature, there is a potential for publication bias, as studies with significant findings are more likely to be published.