Abstract
Purpose:
Men on active surveillance for prostate cancer are extremely interested in dietary changes or supplements to prevent progression of their disease. We sought to determine if a high omega-3, low omega-6 fatty acid diet with fish oil capsules (D+FO) decreases proliferation (Ki-67) in prostate biopsies in men with prostate cancer on active surveillance over a 1-year time period.
Methods:
In this Phase II, prospective randomized trial, men (N=100) with Grade group 1 or 2 prostate cancer that elected active surveillance were randomized to the D+FO or a control group. Same site prostate biopsies were obtained at baseline and 1-year. The primary endpoint was the change in Ki-67 index from baseline to 1-year from same site biopsies compared between the groups.
Results:
The Ki-67 index decreased in the D+FO group by approximately 15% from baseline to 1-year (1.34% at baseline, 1.14% at 1-year) and increased in the control group by approximately 24% from baseline to 1-year (1.23% at baseline, 1.52% at 1-year) resulting in a statistically significant difference in the change of Ki-67 index between the groups (95% CI 2%, 52%, p=0.043). There was no significant difference in the secondary outcomes grade group, tumor length, decipher genomic score or PSA between the two groups. Four patients in the D+FO group were withdrawn from the trial due to adverse events related to the FO.
Conclusion:
A high omega-3, low omega-6 diet with FO for 1-year resulted in a significant reduction in Ki-67 index, a biomarker for prostate cancer progression, metastasis and death. These findings support future Phase III trials incorporating this intervention in men on active surveillance.
INTRODUCTION
Active surveillance (AS) is an established option for men with low risk or favorable intermediate-risk prostate cancer. However, approximately 50% of men that elect AS ultimately undergo therapy with either surgery or radiation within 5-years of the diagnosis1,2. Prostate cancer patients in all stages of the disease, including those on AS, are highly interested in dietary approaches and supplements they can take to delay the progression of their disease. So far, however, specific guidelines in this regard have not been established. The MEAL clinical trial evaluated increased vegetable intake over 2-years in men on AS and reported no effect on prostate cancer progression3. The CANARY prostate cancer AS study evaluated healthy diet patterns with a median follow-up of 7.8 years and reported no effect on upgrading of the cancer4. Thus, there remains a need for prospective randomized trials evaluating diets and supplements for men with prostate cancer to delay progression.
The Western diet is notably high in omega-6 fatty acids found in corn, soy, and safflower oil, and low in omega-3 fatty acids found in fatty cold-water fish such as salmon, mackerel, herring and sardines5,6. In preclinical models of prostate cancer, lowering omega-6 and raising omega-3 dietary intake delayed prostate cancer progression, however epidemiologic studies have shown mixed results7–11. In a cross-sectional study nested within a Phase II clinical trial in men on AS, higher prostate tissue eicosapentaenoic acid (EPA, omega-3) levels correlated with less clinical-upgrading on subsequent biopsies12. In a short-term pre-prostatectomy study, combining a low-fat diet with fish oil (FO) supplementation was found to decrease tissue biomarkers associated with prostate cancer progression (Ki-67-index and the cell-cycle progression score)13,14. Based on these prior preclinical and clinical studies, we conducted the CAPFISH-3 trial, a 1-year prospective randomized trial in men with Grade Group 1 or 2 prostate cancer on AS with randomization to either no treatment (control group) or to a high omega-3, low omega-6 diet with FO supplements (D+FO). The primary endpoint was the change in Ki-67 index from baseline to 1-year from same site biopsies with cancer compared between the groups. Ki-67 index is a marker of proliferation and is known to predict progression, metastasis, and death from prostate cancer 15,16.
METHODS
Trial Design and Conduct
The CAPFISH-3 trial was a single center phase II, randomized, open label, two-arm study in men on active surveillance for prostate cancer and was performed at the University of California Los Angeles. Men were randomized either to a control (no dietary intervention) group or to a group receiving dietary counseling to lower dietary omega-6 fat intake and increase dietary omega-3 fat intake combined with daily FO capsule supplements. Randomization (1:1) was by a permuted random block design stratified by Gleason score 3+3 or 3+4. The UCLA institutional review board approved the protocol and all amendments and the trial was registered with ClinicalTrial.gov (# NCT02176902). See Appendix (Trial Design and Conduct) for details.
Patients and Interventions
Eligible patients had biopsy-confirmed adenocarcinoma of the prostate (minimum 5% cancer in one core); Gleason score of 3+4 or less; clinical stage T2c or less; serum PSA of 25 or less; agree to not consume FO capsules if randomized to the control group; and were participating in the UCLA active surveillance program conducted by one of the authors (LSM). Exclusion criteria were intake of finasteride or dutasteride during the prior 6 months; prior treatment for prostate cancer and patient has allergy to fish or is a vegetarian.
Participants in the intervention arm received dietary counseling by the study registered dietician nutritionist (author PJ) consisting of monthly individualized counseling sessions in-person, via telehealth, or by telephone. Partners/spouses or those who shared in meal preparation in the home were included in the counseling sessions. Subjects were counseled to consume less than 30% of calories from fat with decreased consumption of foods high in omega-6 fatty acids such as fried foods, highly processed foods, chips and baked goods, and to increase intake of omega-3 rich foods (e.g. salmon, tuna) (see details in Appendix– Nutritional Intervention). The goal was to decrease the ratio of omega-6 to omega-3 fat intake to achieve a ratio of less than 4:1. FO capsules were provided by Pharmavite LLC (West Hills, CA) providing a daily dose of 2.2 g of omega-3 fatty acids (docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA)). The control group did not receive dietary counseling and refrained from taking FO.
Prior to enrollment, a prostate biopsy was obtained by LSM or one of his associates and the coordinates of each core were recorded and tracked using an image-fusion device (Artemis, Eigen Health, Grass Valley, CA)17. Twelve months from the initial biopsy, all participants underwent same site biopsies at the coordinates where cancer was previously located (MRI and/or non-MRI targets, minimum of 3 cores per site). In addition, systematic 12-core biopsies were performed at the discretion of LSM. Using the tracking technology, repeat sampling of a specific site is spatially accurate within several mm and has been employed successfully in previous studies17,18.
Assessments
Dietary intake was assessed at baseline and every 3 months (intervention arm) and every 6 months (control arm) by 3-day food records (see Appendix).
Fasting blood was collected at baseline, 6-months, and 1-year. Height was measured at baseline and anthropometrics (weight, waist and hip circumference) were measured at baseline, 6-months, and 1-year in both groups, and at 3 and 9 months in the intervention group.
Plasma lipids (total cholesterol, HDL-cholesterol, triglyceride), testosterone and PSA were measured in the UCLA central clinical laboratory. Red blood cell (RBC) fatty acid analyses were measured as previously described19. Serum cytokines were measured as previously described 20.
Slides from archived formalin-fixed, paraffin-embedded biopsy tissue were cut and either used for H&E staining, multiplex immunofluorescence analysis, or determination of Decipher genomic classifier score (Veracyte Inc. San Diego, CA), a 22 gene panel21. Ki-67 in prostate biopsy tissue were analyzed using multiplex immunofluorescence analysis as described in the Appendix, Table A1 and Figure A122. Ki-67 index was defined as the percentage of malignant cells staining for nuclear Ki-67.
Endpoints and Statistical Analyses
Patient characteristics and study variables were summarized using means/standard deviations or frequencies/percentages. The analyses were conducted with an intention to treat framework (as randomized) for patients who completed the study. The primary endpoint was the change in Ki-67 index from baseline to 1-year from same site biopsies with cancer compared between the groups. The primary analysis of this endpoint was a negative binomial mixed effects model which allowed us to use all available data from patients that completed the trial, even if a patient was missing follow-up or baseline data (due to non-evaluable biopsies). The outcome variable in the model was the number of malignant Ki-67 positive stained nuclei. In order to account for variation in the total number of cells evaluated in each specimen, we included an offset term to control for the number of malignant cells evaluated. The terms in the model included fixed effects for treatment (D+FO/control), time (baseline/12 months), and the treatment by time interaction with a patient random effect (to account for multiple biopsies per patient). Secondary outcome measures included pathologic features (Grade Group, maximum tumor length), Decipher 22 gene classifier score, serum PSA, testosterone, lipids and cytokines. For our primary PSA assessment, we utilized PSA values collected during the study (baseline and 12 months). Additionally, as a sensitivity analysis, we utilized PSA at screening. Other secondary outcomes included RBC omega-3 and omega-6 fatty acid levels, compliance with the diet and FO capsules and adverse events. Longitudinal secondary outcomes were assessed using linear mixed effects models with a similar structure to the primary outcome analysis. We presented these analyses using means/SDs at each time point as well as the p-value from the interaction term to assess for differential changes between groups. Prior to analysis, we assessed normality through visual inspections and drew on our clinical understanding of the variables. For cytokine data (Table A3) and PSA we ran these analyses after performing a logarithmic transformation to better meet statistical assumptions. For Grade Group progression, progression rate was compared between groups using the Cochran-Armitage Trend Test. The statistical analyses were run using the Glimmix procedure in SAS V9.4 (SAS Institute, Cary, NC) and p-values <0.05 were considered statistically significant.
Based on the treatment effect from a prior study incorporating a low-fat FO diet, we estimated a sample size of 35 evaluable subjects per treatment group would have 80% power for finding a 37% difference in log Ki-67 assuming a two-sample two tailed t-test with a 0.05 level of significance (means on log scale of 1.80 vs 1.43, SD=0.54)14. We planned to randomize 100 subjects anticipating a 10% dropout rate and anticipating 20% of subjects would have non-evaluable biopsy samples thus leaving 35 evaluable subjects per treatment group.
RESULTS
Demographics
The CAPFISH-3 trial was conducted from December 2014 to September 2022. One hundred and seven patients signed the consent form, seven of which failed screening, leaving 100 patients that were randomized (50 to the control group and 50 to the intervention group) (Figure 1). The majority of patients had Grade Group 1 (Gleason grade 3+3) prostate cancer and approximately one in four patients that entered the trial had Grade Group 2 (Gleason grade 3+4) prostate cancer (Table 1). The majority of patients in both groups had cT1C prostate cancer (elevated PSA with normal digital rectal exam – data not shown). The reasons for drop out are listed in Figure 1 and the reasons for not entering the trial are listed in the Appendix.
Figure 1. CONSORT Flow Diagram.
Table 1.
Demographic and Clinical Characteristics of the Participants that Entered the Triala
| Characteristic | Control (n=50) | D+FO (n=50) | Total (n=100) |
|---|---|---|---|
| Age | |||
| Mean (SD) - year | 64.5 (7.1) | 64.0 (6.6) | 64.2 (6.8) |
| Range - year | 46.0 – 83.0 | 50.0 – 79.0 | 46.0 – 83.0 |
|
| |||
| Race (%) | |||
| American Indian or Alaska | 1 (2.0%) | 0 (0.0%) | 1 (1.0%) |
| Native | |||
| Asian | 3 (6.0%) | 4 (8.0%) | 7 (7.0%) |
| Black or African American | 2 (4.0%) | 5 (10.0%) | 7 (7.0%) |
| White | 44 (88.0%) | 41 (82.0%) | 85 (85.0%) |
|
| |||
| Ethnicity | |||
| Hispanic or Latino | 5 (10.0%) | 3 (6.0%) | 8 (8.0%) |
| Not Hispanic or Latino | 44 (88.0%) | 47 (94.0%) | 91 (91.0%) |
| Unknown | 1 (2.0%) | 0 (0.0%) | 1 (1.0%) |
|
| |||
| BMI | |||
| Mean (SD) | 26.9 (3.7) | 27.6 (4.4) | 27.3 (4.0) |
| Range | 18.9 – 36.2 | 20.9 – 40.4 | 18.9 – 40.4 |
|
| |||
| Weight - kg | |||
| Mean (SD) | 84.0 (13.3) | 85.6 (13.2) | 84.8 (13.2) |
| Range | 61.2 – 117.5 | 57.6 – 122.9 | 57.6 – 122.9 |
|
| |||
| MAXIMUM Gleason score | |||
| 3+3 (%) | 35 (70.0%) | 35 (70.0%) | 70 (70.0%) |
| 3+4 | 15 (30.0%) | 15 (30.0%) | 30 (30.0%) |
|
| |||
| PSA | |||
| Median (IQR) – ng/mL | 5.6 (4.1–7.0) | 5.9 (4.9–7.7) | 5.8 (4.2–7.6) |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared).
See Appendix Table A4 for characteristics of patients that completed the trial and had evaluable Ki-67 index values.
Dietary and Fish Oil Capsule Intake
Based on analyses of the 3-day food records, subjects in the D+FO group had a significant reduction in dietary omega-6 fatty acid intake and omega-6 to omega-3 dietary ratio as compared to the control group, whereas there was no significant difference in dietary intake (excluding FO supplements) of omega-3 fatty acids between the groups (Table 2). There was no difference in weight change between the groups and there was no significant within group weight loss (Table 3). FO capsule compliance was computed for each patient as the number of pills consumed over the number of pills prescribed, resulting in an average overall compliance rate of 90.5% (SD 12.5%) for prescribed capsules consumed in the D+FO group.
Table 2.
3-Day Diet Records at Baseline, 6 and 12 Months
| Controla,b | D+FOa,b | Group x Timec | ||||||
|---|---|---|---|---|---|---|---|---|
| Time (months) | 0 | 6 | 12 | 0 | 6 | 12 | ||
| Calories (kcal) | 1872 (421) | 1832 (512) | 1798 (518) | 1767 (418) | 1669 (363) | 1434 (287) | 0.012 | |
| Protein (g) | 84.5 (30.6) | 84.2 (23.7) | 79.7 (26.8) | 83.3 (26.7) | 81.4 (22.1) | 73.0 (20.0) | 0.57 | |
| Carbohydrate (g) | 204.0 (50.3) | 195.9 (72.6) | 192.8 (74.1) | 176.5 (54.4) | 178.2 (65.5) | 157.4 (38.4) | 0.71 | |
| Fiber (g) | 20.4 (8.9) | 20.7 (9.7) | 19.1 (8.0) | 17.3 (7.2) | 19.1 (9.5) | 17.7 (8.3) | 0.86 | |
| Sugar (g) | 75.4 (30.8) | 68.5 (34.7) | 64.4 (35.7) | 61.2 (27.4) | 63.4 (30.5) | 54.6 (27.4) | 0.33 | |
| Fat (g) | 75.8 (25.7) | 75.2 (28.8) | 75.0 (27.7) | 73.6 (22.3) | 64.1 (16.9) | 53.2 (16.3) | <0.001 | |
| SatFat (g) | 23.3 (9.5) | 21.3 (8.9) | 22.1 (8.9) | 22.9 (9.0) | 19.4 (5.5) | 16.0 (6.5) | 0.008 | |
| MonoFat (g) | 28.6 (12.0) | 27.7 (11.4) | 28.5 (11.5) | 27.6 (9.7) | 24.2 (8.0) | 20.2 (6.8) | 0.001 | |
| PolyFat (g) | 17.3 (6.6) | 19.7 (11.9) | 17.9 (10.1) | 16.2 (5.8) | 14.3 (6.1) | 14.0 (14.9) | 0.07 | |
| Cholesterol (mg) | 278.6 (166.9) | 270.8 (152.1) | 281.8 (134.5) | 307.1 (169.3) | 263.9 (151.1) | 231.1 (132.3) | 0.08 | |
| Omega-3 (g) | 2.1 (1.5) | 3.4 (6.1) | 2.4 (3.6) | 2.0 (1.3) | 2.5 (2.6) | 2.3 (2.0) | 0.59 | |
| o-3+FOd | -- | -- | -- | 2.0 (1.3) | 4.7 (2.6) | 4.5 (2.0) | -- | |
| Omega-6 (g) | 15.0 (5.5) | 16.0 (8.4) | 15.2 (7.5) | 14.1 (4.9) | 11.8 (5.0) | 9.2 (4.3) | <0.001 | |
| Alcohol (g) | 9.7 (14.0) | 9.3 (14.6) | 8.7 (13.4) | 13.0 (16.2) | 11.7 (15.5) | 8.2 (12.5) | 0.25 | |
| % Fat Kcal | 35.8 (7.4) | 36.2 (7.7) | 37.4 (8.5) | 37.5 (7.3) | 34.5 (5.9) | 32.9 (6.4) | <0.001 | |
| Ratio n6:n3 | 8.5 (3.5) | 8.1 (3.3) | 8.5 (3.6) | 8.7 (3.3) | 6.8 (3.6) | 5.7 (3.2) | 0.017 | |
| Ratio n6:n3 +FOd | -- | -- | -- | 8.7 (3.3) | 2.8 (1.1) | 2.2 (0.7) | -- | |
Abbreviations: SatFat, saturated fat; MonoFat, monounsaturated fat; PolyFat, polyunsaturated fat.
Values in Parenthesis are Standard Deviations.
Control group baseline n=48, 6 months n=43, 12 months n=41. D+FO group baseline n=49, 6 months n=41, 12 months n=33.
p-values were assessed from the interaction term for differential changes between groups using linear mixed effects models.
Control group did not consume fish oil and therefore the fields in the table were left blank. The fish oil capsules provided 2.2 g of omega-3 fatty acids provided by docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA).
Table 3.
Weight, BMI, Serum Lipids, Testosterone, PSA and Grade Group Progression/Regression
| Controla,b | D+FOa,b | GroupxTimec | |||||
|---|---|---|---|---|---|---|---|
| Time (mon) | 0 | 6 | 12 | 0 | 6 | 12 | |
| Weight (kg) | 84.0 (13.3) | 84.9 (19.7) | 85.4 (13.3) | 85.6 (13.2) | 86.2 (13.4) | 85.6 (13.0) | 0.689 |
| BMI (kg/m2) | 26.9 (3.7) | 27.2 (5.4) | 27.3 (3.7) | 27.6 (4.4) | 27.9 (4.5) | 27.6 (4.5) | 0.681 |
| Triglycerides (mg/dL) | 105.2 (46.0) | 117.0 (45.4) | 115.9 (54.5) | 121.0 (53.0) | 107.4 (47.4) | 110.2 (43.5) | 0.016 |
| Total cholesterol (mg/dL) | 185.8 (34.4) | 190.3 (37.2) | 184.8 (46.6) | 191.9 (43.5) | 186.4 (38.0) | 193.6 (46.3) | 0.948 |
| HDL- cholesterol (mg/dL) | 56.8 (12.2) | 58.0 (13.1) | 57.0 (14.2) | 57.9 (15.2) | 58.8 (16.0) | 60.4 (17.6) | 0.09 |
| LDL-cholesterol (mg/dL) | 108.0 (34.4) | 108.9 (33.3) | 104.6 (41.7) | 109.8 (37.7) | 106.1 (33.3) | 111.1 (40.7) | 0.704 |
| Testosterone (ng/dL) | 464.6 (153.2) | 442.8 (165.4) | 435.7 (153.9) | 423.9 (162.2) | 407.0 (149.8) | 395.7 (150.1) | 0.824 |
| PSA (ng/mL)d | 6.6 (4.5–8.8) | # | 6.3 (3.7–8.4) | 7.2 (5.6–11.0) | # | 7.7 (5.0–11) | 0.854 |
| Grade Groupe,f,g | 0.499 | ||||||
| Progression | 17 (36.2) | 14 (31.8) | |||||
| No change | 23 (48.9) | 21 (47.7) | |||||
| Regression | 7 (14.9) | 9 (20.5) | |||||
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); LDL, low density lipoprotein; HDL, high density lipoprotein.
Values in Parenthesis are Standard Deviations or IQR.
Control group baseline n=50, 6 months n=46, 12 months n=46; D+FO group baseline n=50, 6 months n=37, 12 months n=43.
p-values were assessed from the interaction term for differential changes between groups using linear mixed effects models.
PSA not collected at 6-month time point(#).
Values in Parenthesis are Percent of Total Patients.
Control group n=47, D+FO group n=44.
Grade group progression rate was compared between groups using the Cochran-Armitage Trend Test.
Reduction in dietary omega-6 intake was confirmed by RBC fatty acid measurements, which are a measure of long-term dietary intake (Figure 2). DHA (omega-3), EPA (omega-3) and total omega-3 RBC levels were all significantly increased in the D+FO group but not in the control group (Figure 2, Table A2).
Figure 2. Red Blood Cell Fatty Acid Composition at Baseline, 6 and 12 Months as Percent of Total Fat.
Data in the control group were collected from n=46 (baseline), n=42 (6 months) and n=44 (12 months); D+FO group from n=45 (baseline), n=39 (6 months) and n=44 (12 months).
Abbreviations: DHA: docosahexaenoic acid; EPA: eicosapentaenoic acid; LA: linoleic acid; AA: arachidonic acid; compared to baseline; p≤0.001. p-values were assessed from the interaction term for differential changes between groups using linear mixed effects models.
Primary Endpoint
There was a statistically significant differential reduction in Ki-67 index in the D+FO vs control group when comparing change from baseline to 1-year in same site prostate biopsy cancer tissue (31% differential reduction, 95% CI 2%, 52%, p=0.043) (Figure 3). The Ki-67 index decreased in the D+FO group by approximately 15% (from 1.34% at baseline to 1.14% at 1-year) and increased in the control group by approximately 24% (from 1.23% at baseline to 1.52% at 1-year).
Figure 3. Prostate Ki-67 Index in Same Site Biopsies at Baseline and 12 Months by Multiplex Fluorescence Staining.
A. Dot plot of Ki-67 index values at baseline and 12 months for the control and D+FO groups; B. Mean change in Ki-67 from baseline to 12 months comparing control and D+FO groups. Among the 44 patients in the D+FO group and 47 patients in the control group, 40 patients in each group had evaluable cancer tissue for Ki-67 index analysis. Prostate Ki-67 index was assessed using a negative binomial mixed effects model. Error bars represent 95% confidence intervals estimated from the model.
Secondary Endpoints
Among the patients that completed the 1-year biopsy, there was no significant trend in grade group upgrading or downgrading between the groups (p=0.499, Table 3). For maximal tumor length (surrogate for tumor volume) there was no significant difference in change from baseline (Control group: mean 3.12, SD 2.28, D+FO group: mean 3.06, SD 1.94) to 1-year (Control group: mean 3.96, SD 3.12, FO group: mean 3.43, SD 2.81) between groups (p=0.40) for same site biopsies. For the Decipher genomic classifier score, there was no significant difference in change from baseline (Control group: mean 0.23, SD 0.19, D+FO group: mean 0.18, SD 0.10) to 1-year (Control group: mean 0.22, SD 0.11, D+FO group: mean 0.24, SD 0.18) between groups (p=0.09) for same site biopsies.
Testosterone, PSA, Lipids and Cytokines
Comparing baseline to 1-year, there was no significant difference between the groups in testosterone or PSA levels (Table 3). There was a significant decrease in triglyceride levels between the groups (Table 3). Comparing baseline to 1-year serum concentration of cytokines and macrophage colony stimulating factor, also known as colony stimulating factor-1, was significantly decreased in the D+FO group compared to control group (p=0.017, Table A3). Since PSA levels drawn during the trial were collected at patient entry (after the baseline prostate biopsy), and could potentially be affected by proximity to the biopsy, we performed a post hoc, sensitivity analysis of PSA values collected prior to the biopsy that were used for screening. For this analysis, when comparing baseline to 1-year, there was also no significant difference between the groups (p=0.203), with PSA values at eligibility measured as 5.6 (4.1–7.0) and 5.9 (4.9–7.7) for control and D+FO, respectively (median Q1-Q3).
Safety
Four of the 50 patients randomized to the D+FO group were withdrawn by the principal investigator due to adverse events attributed to the FO. One patient had grade 2 flatulence, constipation and bloating. Another was withdrawn due to grade 2 diarrhea, gastrointestinal discomfort and eructation. The two other patients were withdrawn due to grade 1 adverse events (loose stools in 1 patient and loose stools, eructation and nausea in the other).
DISCUSSION
CAPFISH-3, a prospective randomized trial, demonstrated that intake of a high omega-3, low omega-6 diet with FO supplements compared to a control group significantly decreased the change in Ki-67 index (primary endpoint) in same site tumor biopsy tissue from baseline to 1 year. In a meta-analysis by Berlin et al. of prostate cancer patients treated with curative intent, Ki-67 index was a predictor of biochemical failure, distant metastasis, and prostate cancer survival15. Moreover, a recent study by Kammerer-Jacquet et al. reported Ki-67 in prostate needle biopsy tissue was an independent predictor of prostate cancer survival in conservatively managed prostate cancer patients (patients that had no therapy 6 months after their biopsy)16. These authors concluded that Ki-67 in prostate biopsy tissue should be utilized as a viable biomarker for prognostication in patients on active surveillance16. There are no prior trials in the literature evaluating serial Ki-67 index values over time in men on active surveillance and there are no reports on Ki-67 index values in patients with Grade Group 1 and 2 prostate cancer. Kammerer-Jacquet et al. reported the mean Ki-67 index was 5% in conservatively managed patients with Grade Group 1 through 5 disease, however they did not report the Ki-67 index values for the patients with Grade Group 1 and 2 disease16. The Ki-67 index cutpoint in their study predicting prostate cancer specific death was 5%. In a meta-analysis of 21 studies, Berlin et al reported a mean Ki-67 index of 6.13% in prostate cancer patients treated with curative intent15. In our trial the baseline mean Ki-67 index values were lower (1.34% D+FO group and 1.23% control group) than previously reported, likely since we only enrolled subjects with Grade Group 1 and 2 disease. In a similarly designed and analyzed study in men at risk for lung cancer, Mao et al observed mean Ki-67 index values in bronchial biopsies in the range of 2–3% and observed a differential change in Ki-67 index in the celecoxib vs placebo arm (34% decrease in celecoxib and 3.8% increase in placebo)23. These changes in Ki-67 index are similar to our trial in which the Ki-67 index decreased by 15% in the D+FO group and increased by 24% in the control group. The Ki-67 index values from our trial will potentially provide useful data for planning future prospective trials evaluating diet and lifestyle intervention trials in men on active surveillance.
The intervention used in the present trial is based on prior preclinical studies and a prospective clinical trial14. Omega-6 polyunsaturated fat found in corn oil, soy oil, and safflower oil present in fried foods, highly processed foods, chips and baked goods is consumed in high quantities in the typical western diet and is known to stimulate prostate cancer progression in preclinical models6,11. Likewise, omega-3 fatty acids found in cold-water fish such as salmon and sardines are low in the typical western diet and are known to delay the progression of prostate cancer in preclinical models6,7,10. Moreover, a short-term randomized pre-prostatectomy trial combining a low fat diet low in omega-6 fatty acids with FO supplements reported a significant reduction in radical prostatectomy cancer Ki-67 and reduction in a cell cycle progression genetic risk score in the intervention vs the control Western group13. Noteworthy is the cell cycle progression score incorporates gene expression of Ki-67 in the 31 gene panel and, when combined with clinical data, is a predictor of prostate cancer progression24. To achieve a favorable ratio of omega-6 to omega-3 fatty acids in the present trial we combined the dietary intervention with FO supplements. The outpatient diet intervention achieved the goal of lowering omega-6 intake based on analyses of the 3-day food records and RBC fatty acid levels. However, FO supplements were required to raise the omega-3 levels as this was not achieved by the diet alone as measured by the 3-day food records.
There are numerous underlying mechanisms supporting the intervention used in the present trial. The omega-6 fatty acid arachidonic acid (significantly reduced in this trial) is known to stimulate prostate cancer growth through a number of mechanisms including increasing pro-carcinogenic arachidonic acid metabolite levels and activation of phosphatidylinositol 3-kinase signaling25–27. In addition, dietary omega-3 fatty acids have been shown to decrease the number of M2-like macrophages through binding to G-protein coupled receptors28. M2-like macrophages are immunosuppressive and promote angiogenesis, tumor progression and metastases28 and are associated with poor outcome and increased risk of nodal and distant metastasis29,30. Consistent with this mechanism, in the present trial, serum levels of the cytokine macrophage colony stimulating factor, which is known to play a role in stimulating macrophages and prostate cancer progression, was significantly reduced in the intervention group (relative to the control group)31.
Potential shortcomings of the present trial are that there was no significant changes between the groups in other tissue markers of prostate cancer aggressiveness such as the Grade Group, tumor length (a surrogate of tumor volume) and serum PSA level. There was also no change between the groups in the Decipher genomic classifier score, though data is limited on use of the Decipher in patients on active surveillance21. Our trial was not powered to detect differences in these secondary endpoints. Longer term and Phase III trials will be required to determine effects of the intervention on clinical endpoints. Another shortcoming is that our trial was 1-year in duration. Further studies are also required to determine the optimal dosing of omega-3 fatty acids. We chose the dosing for this trial based on dosing used in our prior prospective pre-prostatectomy trial that demonstrated inhibitory effects on Ki-67 and a cell cycle progression score14. A further shortcoming of the intervention is that 4 patients in the D+FO arm were withdrawn from the trial due to side effects from FO. Patients will need to be counseled about potential side effects prior to entering future trials incorporating FO. A strength of the trial is that it established the feasibility of performing same site prostate biopsies for future dietary intervention trials, thus allowing longitudinal evaluation of tissue biomarkers. Ultimately, longer term trials with larger sample sizes powered for clinical endpoints will be needed to provide clinical recommendations for our patients.
In summary, CAPFISH-3, a 1-year high omega-3, low omega-6 diet + FO intervention trial, resulted in a significant reduction in prostate cancer tissue Ki-67 index, a biomarker for prostate cancer progression, metastasis and death. Patients in the trial were compliant with the intervention which was well tolerated. Moreover, obtaining serial same site prostate biopsy tissue is feasible for future biomarker intervention trials. Based on the underlying anti-prostate cancer mechanisms of a high omega-3, low omega-6 diet with FO supplements, future trials are warranted evaluating this intervention in varying stages of prostate cancer.
Supplementary Material
CONTEXT.
Key Objective:
In this phase II prospective randomized trial we sought to determine if a high omega-3, low omega-6 fatty acid diet with fish oil capsules decreases proliferation (Ki-67 index) in prostate biopsies in men with prostate cancer on active surveillance over a 1-year time period. Same site prostate biopsies were obtained at baseline and 1-year using an image fusion device that uses coordinates to track the cancer sites.
Knowledge Generated:
The 1-year intervention resulted in a significant reduction in Ki-67 index (as compared to the control group), a biomarker for prostate cancer progression and metastases. These findings support future Phase III trials incorporating this intervention in men on active surveillance.
Relevance statement:
A diet intervention may have a role in modifying the biology of prostate cancer in patients undergoing active surveillance. Larger trials should expand this notion using more clinically impactful outcomes parameters.
Relevance statement written by Dr. Necchi
Funding/Support:
This study was funded by grant RO1CA231219 and P50CA92131 to WJA from the National Cancer Institute; R01AG069698, P30AG068345 and R01HL158691 to PC; UL1TR001881 to CTSI and P30CA016042 to Jonson Comprehensive Cancer Center. We thank Howard B. Klein and the Seafood Industry Research Fund (SIRF) for their generous support.
Role of the Funder/Sponsor:
The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Footnotes
Conflict of Interest Disclosure:
No disclosures were reported.
Availability of Trial Protocol: in Supplement
Data Sharing Statement:
Data available: Yes
Data types: Deidentified participant data that underlie the results reported in this article
How to access data: Researchers who provided a methodologically sound proposal can request data from corresponding author
When available: After publication of this article
Mechanisms of data availability: Data will be made available with a signed data access agreement.
Trial Registration: ClinicalTrial.gov (# NCT02176902)
REFERENCES
- 1.Timilshina N, Alibhai SMH, Tomlinson G, et al. : Long-term Outcomes Following Active Surveillance of Low-grade Prostate Cancer: A Population-based Study Using a Landmark Approach. J Urol 209:540–548, 2023 [DOI] [PubMed] [Google Scholar]
- 2.Simpkin AJ, Tilling K, Martin RM, et al. : Systematic Review and Meta-analysis of Factors Determining Change to Radical Treatment in Active Surveillance for Localized Prostate Cancer. Eur Urol 67:993–1005, 2015 [DOI] [PubMed] [Google Scholar]
- 3.Parsons JK, Zahrieh D, Mohler JL, et al. : Effect of a Behavioral Intervention to Increase Vegetable Consumption on Cancer Progression Among Men With Early-Stage Prostate Cancer: The MEAL Randomized Clinical Trial. JAMA 323:140–148, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Schenk JM, Liu M, Neuhouser ML, et al. : Dietary Patterns and Risk of Gleason Grade Progression among Men on Active Surveillance for Prostate Cancer: Results from the Canary Prostate Active Surveillance Study. Nutr Cancer 75:618–626, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Simopoulos AP: An Increase in the Omega-6/Omega-3 Fatty Acid Ratio Increases the Risk for Obesity. Nutrients 8:128, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Saini RK, Keum YS: Omega-3 and omega-6 polyunsaturated fatty acids: Dietary sources, metabolism, and significance - A review. Life Sci 203:255–267, 2018 [DOI] [PubMed] [Google Scholar]
- 7.Liang P, Henning SM, Guan J, et al. : Role of Host GPR120 in Mediating Dietary Omega-3 Fatty Acid Inhibition of Prostate Cancer. J Natl Cancer Inst 111:52–59, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lovegrove C, Ahmed K, Challacombe B, et al. : Systematic review of prostate cancer risk and association with consumption of fish and fish-oils: analysis of 495,321 participants. Int J Clin Pract 69:87–105, 2015 [DOI] [PubMed] [Google Scholar]
- 9.Lee KH, Seong HJ, Kim G, et al. : Consumption of Fish and omega-3 Fatty Acids and Cancer Risk: An Umbrella Review of Meta-Analyses of Observational Studies. Adv Nutr 11:1134–1149, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bilodeau JF, Gevariya N, Larose J, et al. : Long chain omega-3 fatty acids and their oxidized metabolites are associated with reduced prostate tumor growth. Prostaglandins Leukot Essent Fatty Acids 164:102215, 2021 [DOI] [PubMed] [Google Scholar]
- 11.Ngo TH, Barnard RJ, Anton T, et al. : Effect of isocaloric low-fat diet on prostate cancer xenograft progression to androgen independence. Cancer Res 64:1252–4, 2004 [DOI] [PubMed] [Google Scholar]
- 12.Moussa H, Nguile-Makao M, Robitaille K, et al. : Omega-3 Fatty Acids Survey in Men under Active Surveillance for Prostate Cancer: from Intake to Prostate Tissue Level. Nutrients 11, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Galet C, Gollapudi K, Stepanian S, et al. : Effect of a low-fat fish oil diet on proinflammatory eicosanoids and cell-cycle progression score in men undergoing radical prostatectomy. Cancer Prev Res (Phila) 7:97–104, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Aronson WJ, Kobayashi N, Barnard RJ, et al. : Phase II prospective randomized trial of a low-fat diet with fish oil supplementation in men undergoing radical prostatectomy. Cancer Prev Res (Phila) 4:2062–71, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Berlin A, Castro-Mesta JF, Rodriguez-Romo L, et al. : Prognostic role of Ki-67 score in localized prostate cancer: A systematic review and meta-analysis. Urol Oncol 35:499–506, 2017 [DOI] [PubMed] [Google Scholar]
- 16.Kammerer-Jacquet SF, Ahmad A, Moller H, et al. : Ki-67 is an independent predictor of prostate cancer death in routine needle biopsy samples: proving utility for routine assessments. Mod Pathol 32:1303–1309, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Natarajan S, Marks LS, Margolis DJ, et al. : Clinical application of a 3D ultrasound-guided prostate biopsy system. Urol Oncol 29:334–42, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Salami SS, Tosoian JJ, Nallandhighal S, et al. : Serial Molecular Profiling of Low-grade Prostate Cancer to Assess Tumor Upgrading: A Longitudinal Cohort Study. Eur Urol 79:456–465, 2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bagga D, Capone S, Wang HJ, et al. : Dietary modulation of omega-3/omega-6 polyunsaturated fatty acid ratios in patients with breast cancer. J Natl Cancer Inst 89:1123–31, 1997 [DOI] [PubMed] [Google Scholar]
- 20.Kim SJ, Mehta HH, Wan J, et al. : Mitochondrial peptides modulate mitochondrial function during cellular senescence. Aging (Albany NY) 10:1239–1256, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Press BH, Jones T, Olawoyin O, et al. : Association Between a 22-feature Genomic Classifier and Biopsy Gleason Upgrade During Active Surveillance for Prostate Cancer. Eur Urol Open Sci 37:113–119, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mori H, Bolen J, Schuetter L, et al. : Characterizing the Tumor Immune Microenvironment with Tyramide-Based Multiplex Immunofluorescence. J Mammary Gland Biol Neoplasia 25:417–432, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mao JT, Roth MD, Fishbein MC, et al. : Lung cancer chemoprevention with celecoxib in former smokers. Cancer Prev Res (Phila) 4:984–93, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Sommariva S, Tarricone R, Lazzeri M, et al. : Prognostic Value of the Cell Cycle Progression Score in Patients with Prostate Cancer: A Systematic Review and Meta-analysis. Eur Urol 69:107–15, 2016 [DOI] [PubMed] [Google Scholar]
- 25.Hughes-Fulford M, Li CF, Boonyaratanakornkit J, et al. : Arachidonic acid activates phosphatidylinositol 3-kinase signaling and induces gene expression in prostate cancer. Cancer Res 66:1427–33, 2006 [DOI] [PubMed] [Google Scholar]
- 26.Yang P, Cartwright CA, Li J, et al. : Arachidonic acid metabolism in human prostate cancer. Int J Oncol 41:1495–503, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Oktem EK, Aydin B, Gulfidan G, et al. : A Transcriptomic and Reverse-Engineering Strategy Reveals Molecular Signatures of Arachidonic Acid Metabolism in 12 Cancers. OMICS 27:127–138, 2023 [DOI] [PubMed] [Google Scholar]
- 28.Liang P, Henning SM, Grogan T, et al. : Effect of omega-3 fatty acid diet on prostate cancer progression and cholesterol efflux in tumor-associated macrophages-dependence on GPR120. Prostate Cancer Prostatic Dis, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hadimani SM, Das S, Harish KG: An immunohistochemical evaluation of tumor-associated macrophages (M1 and M2) in carcinoma prostate - An institutional study. J Cancer Res Ther 19:S300–S305, 2023 [DOI] [PubMed] [Google Scholar]
- 30.Han C, Deng Y, Xu W, et al. : The Roles of Tumor-Associated Macrophages in Prostate Cancer. J Oncol 2022:8580043, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Mougel A, Adriaenssens E, Guyot B, et al. : Macrophage-Colony-Stimulating Factor Receptor Enhances Prostate Cancer Cell Growth and Aggressiveness In Vitro and In Vivo and Increases Osteopontin Expression. Int J Mol Sci 23, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.