Background:

Hyperinsulinemia is considered to be important in the development of colon cancer, but few studies have investigated the associations of hyperinsulinemia with colon cancer survival via dietary scores.

Methods:

Empirical dietary index for hyperinsulinemia (EDIH) was derived to assess the insulinemic potential of daily diets reflecting the long-term insulin exposure, with higher (more positive) scores indicating higher insulinemic diets. We prospectively estimated the HRs and 95% confidence intervals (CI) to investigate the association of EDIH with disease-free, recurrence-free, and overall survival among patients with stage III colon cancer (1999–2009) enrolled in a randomized adjuvant chemotherapy trial (CALGB 89803).

Results:

Of 1,024 patients (median follow-up: 7.3 years), 311 died, 350 had recurrences, and 394 had events for disease-free survival. Compared with patients in the lowest quintile of EDIH, the corresponding HRs of patients in the highest quintile for disease-free survival events, cancer recurrence, and overall mortality were 0.80 (95% CI, 0.56–1.15), 0.76 (95% CI, 0.51–1.11), and 0.77 (95% CI, 0.52–1.14).

Conclusions:

Higher EDIH was not associated with the risk of colon cancer recurrence or mortality in this population of patients with stage III colon cancer.

Impact:

EDIH, as a measure of dietary insulinemic potential, may be associated with colon cancer risk but not survival in patients with late-stage colon cancer.

Hyperinsulinemia is considered an underlying mechanism linking diet and lifestyle to colon cancer risk and has been associated with inferior outcome among patients with colon cancer (1–3). Dietary factors can alter insulin levels, and there has been much interest in deriving dietary scores to predict hyperinsulinemia. The empirical dietary index for hyperinsulinemia (EDIH) was derived to assess the insulinemic potential of daily diets reflecting the long-term insulin exposure, and higher score was associated with a 26% greater risk of colorectal cancer (4). Few studies have investigated the relationship between EDIH and cancer outcome. In a prospective study of patients with stage III colon cancer, we investigated the hypothesis that higher dietary insulinemic potential, as measured by EDIH, would be associated with inferior patient outcomes.

As described previously (5, 6), 1,095 patients with stage III colon cancer were eligible in an adjuvant therapy trial (Cancer and Leukemia Group B, now the Alliance for Clinical Trials in Oncology, 89803) who completed the first survey of diet and lifestyle; 978 of these patients completed the second survey (1999–2001). Both surveys included the same validated food frequency questionnaire (FFQ; ref. 7). The study was approved by each site's institutional review board.

We excluded 38 patients who recurred before or within 90 days after the first questionnaire, or died within 90 days of the first questionnaire, and 33 patients who left ≥70 food items blank or had implausible calorie intake (men: <600 or >4,200 kcal/day; female: <500 or >3,200 kcal/day) on the first questionnaire, and finally included 1,024 patients.

EDIH has been described previously, with higher (more positive) scores indicating higher insulinemic diets (4). We calculated EDIH of each patient using the first FFQ, and updated EDIH on the basis of the results of the second FFQ using cumulative averaging as described previously (5). The primary outcomes include disease-free, recurrence-free, and overall survival since first questionnaire, as defined previously (5).

We investigated the associations of EDIH with survivals via Cox proportional hazard regression. Model 1 was adjusted for age at diagnosis (years) and time-varying calorie intake. Model 2 was additionally adjusted for sex, treatment arm, T-stage (T1–T2 and T3–T4), positive lymph nodes (1–3 and ≥4), performance status (fully active and restricted in strenuous activity), aspirin (yes and no), tumor location (left-sided, transverse, or right-sided), family history of colorectal cancer (yes and no), pack-years of smoking, and time-varying body mass index (BMI, kg/m2) and physical activity [metabolic equivalent of task (MET)-hour/week]. In addition, we conducted separate analyses by sex.

Data were collected by the Alliance Statistics and Data Center and frozen on November 9, 2009. All analyses were conducted using SAS statistical software, version 9.4 (SAS Institute, Inc.) and two-sided P < 0.05 was considered statistically significant.

Of the 1,024 patients with colon cancer (median follow-up: 7.3 years), 311 died, 350 had recurrences, and 394 had events for disease-free survival. Patients in the highest quintiles of EDIH were more likely to be female, younger, adipose, black, have more positive lymph nodes, and have left-sided tumors (Table 1). Also, they had higher intakes of all positive contributors to the EDIH except low-energy beverages and low-fat dairy products; for negative contributors to the EDIH, they had lower intakes of wine and whole fruits, but higher intakes of high-fat dairy products. Participants of different quintiles did not differ on performance status, stage, treatment arm, pack-years of smoking, physical activity, family history of colorectal cancer, or aspirin use. Patients in the highest quintile of EDIH appeared to experience a trend toward an improved disease-free, recurrence-free, and overall survival, although none statistically significant (Table 2). Moreover, the associations did not differ by sex (Pinteraction > 0.05).

Table 1.

Characteristics of patients with stage III colon cancer (N = 1,024) by quintiles of EDIH.

EDIH quintiles
12345
CharacteristicsaN = 204N = 205N = 205N = 205N = 205Pb
Male 117 (57.4%) 102 (49.8%) 91 (44.4%) 82 (40.0%) 56 (27.3%) <0.01 
Age, years 63 (28–85) 60 (24–78) 60 (31–82) 61 (29–83) 57 (21–80) <0.01 
BMI, kg/m2 25.9 (17.2–42.4) 27.3 (17.2–49.9) 27.3 (16.3–45.0) 27.5 (15.7–44.6) 28.2 (17.3–51.8) <0.01 
Race 0.05 
 White 180 (88.2%) 185 (90.2%) 187 (91.2%) 176 (85.9%) 181 (88.3%)  
 Black 8 (3.9%) 10 (4.9%) 14 (6.8%) 18 (8.8%) 18 (8.8%)  
 Other 16 (7.8%) 10 (4.9%) 4 (2.0%) 11 (5.4%) 6 (2.9%)  
Performance status 0.75 
 Fully active 152 (76.8%) 153 (75.7%) 154 (76.2%) 146 (73.4%) 145 (71.8%)  
 Restricted in strenuous activity 46 (23.2%) 49 (24.3%) 48 (23.8%) 53 (26.6%) 57 (28.2%)  
 Unknown  
Bowel wall invasion 0.65 
 Stage 1 and 2 28 (14.1%) 24 (11.9%) 34 (16.7%) 25 (12.7%) 26 (13.0%)  
 Stage 3 and 4 170 (85.9%) 178 (88.1%) 169 (83.3%) 172 (87.3%) 174 (87.0%)  
 Unknown  
Positive lymph nodes 0.05 
 1–3 or unknown 138 (69.3%) 134 (66.3%) 122 (60.1%) 135 (67.8%) 116 (57.4%)  
 ≥4 61 (30.7%) 68 (33.7%) 81 (39.9%) 64 (32.2%) 86 (42.6%)  
 Unknown  
Tumor location 0.04 
 Left-sided 68 (34.2%) 97 (48.0%) 90 (44.6%) 85 (42.7%) 95 (47.5%)  
 Transverse or right-sided 131 (65.8%) 105 (52.0%) 112 (55.4%) 114 (57.3%) 105 (52.5%)  
 Unknown  
Treatment arm 0.69 
 Fluorouracil + leucovorin 112 (54.9%) 103 (50.2%) 100 (48.8%) 99 (48.3%) 102 (49.8%)  
 Irinotecan, fluorouracil, leucovorin 92 (45.1%) 102 (49.8%) 105 (51.2%) 106 (51.7%) 103 (50.2%)  
Pack-years of smoking 0.3 (0.0–72.5) 0.6 (0.0–112.9) 0.6 (0.0–104.1) 1.3 (0.0–102.1) 0.6 (0.0–94.0) 0.10 
Physical activity, MET-hour/week 4.5 (0.0–119.9) 4.4 (0.0–122.9) 5.4 (0.0–114.2) 6.1 (0.0–147.4) 4.4 (0.0–125.2) 0.81 
Family history of colorectal cancer 49 (24.0%) 47 (22.9%) 50 (24.4%) 50 (24.4%) 60 (29.3%) 0.62 
Regular aspirin user 13 (6.4%) 18 (8.8%) 16 (7.8%) 17 (8.3%) 19 (9.3%) 0.85 
Calorie intakec, kcal/day 1,530 (617–3,666) 1,684 (649–3,691) 1,794 (860–3,534) 2,144 (1,032–3,578) 2,412 (1,181–4,119) <0.01 
EDIH-positive associationsd  
 Red meat, servings/week 1.1 (0.0–6.4) 1.6 (0.0–9.0) 1.9 (0.0–6.9) 2.5 (0.0–8.9) 3.7 (0.0–12.7) <0.01 
 Processed meat, servings/week 0.8 (0.0–3.8) 1.2 (0.0–7.5) 1.4 (0.0–6.2) 2.0 (0.0–14.0) 3.0 (0.0–27.0) <0.01 
 Low-energy beverages, servings/week 0.5 (0.0–18.2) 1.0 (0.0–17.9) 0.7 (0.0–28.4) 0.4 (0.0–27.6) 0.5 (0.0–55.1) 0.01 
 Cream soups, servings/week 0.2 (0.0–1.0) 0.2 (0.0–1.7) 0.3 (0.0–3.0) 0.4 (0.0–3.0) 0.3 (0.0–5.5) <0.01 
 Margarine, servings/week 0.7 (0.0–28.0) 1.9 (0.0–28.0) 3.0 (0.0–28.0) 3.0 (0.0–28.0) 3.0 (0.0–28.0) <0.01 
 Poultry, servings/week 1.1 (0.0–5.5) 1.2 (0.0–6.0) 1.2 (0.0–7.0) 1.6 (0.0–6.0) 1.8 (0.0–8.5) <0.01 
 Butter, servings/week 0.2 (0.0–17.5) 0.4 (0.0–17.5) 1.0 (0.0–18.4) 1.9 (0.0–28.0) 2.1 (0.0–28.0) <0.01 
 French fries, servings/week 0.2 (0.0–1.7) 0.4 (0.0–3.0) 0.5 (0.0–3.0) 0.5 (0.0–3.2) 1.0 (0.0–5.5) <0.01 
 Other fish, servings/week 0.9 (0.0–6.5) 1.0 (0.0–4.5) 1.0 (0.0–7.0) 1.1 (0.0–7.0) 1.0 (0.0–5.8) <0.01 
 High-energy beverages, servings/week 0.8 (0.0–13.3) 1.4 (0.0–14.3) 2.4 (0.0–20.5) 3.1 (0.0–25.5) 6.8 (0.0–38.5) <0.01 
 Tomatoes, servings/week 2.9 (0.0–10.7) 3.1 (0.0–13.0) 3.6 (0.2–13.9) 4.5 (0.2–18.0) 5.6 (0.2–18.0) <0.01 
 Low-fat dairy products, servings/week 0.7 (0.0–13.0) 0.5 (0.0–17.5) 0.4 (0.0–17.4) 0.4 (0.0–19.6) 0.3 (0.0–15.9) 0.01 
 Eggs, servings/week 0.9 (0.0–4.8) 1.9 (0.0–7.0) 3.0 (0.0–14.0) 3.0 (0.0–14.0) 3.0 (0.0–14.0) <0.01 
EDIH-negative associationsd  
 Wine, servings/week 0.4 (0.0–23.9) 0.1 (0.0–14.5) 0.0 (0.0–20.5) 0.0 (0.0–12.5) 0.0 (0.0–12.7) <0.01 
 Coffee, servings/week 7.0 (0.0–42.5) 7.0 (0.0–37.2) 7.5 (0.0–35.5) 6.3 (0.0–42.0) 6.1 (0.0–35.0) 0.02 
 Whole fruits, servings/week 11.1 (0.6–88.0) 10.4 (0.0–66.0) 9.2 (0.2–34.9) 9.6 (0.0–45.4) 8.1 (0.0–31.9) <0.01 
 High-fat dairy products, servings/week 3.4 (0.0–39.7) 4.2 (0.0–50.2) 5.3 (0.0–31.9) 6.0 (0.0–38.7) 6.9 (0.1–40.0) <0.01 
 Green leafy vegetables, servings/week 3.1 (0.0–17.5) 2.4 (0.0–14.7) 2.2 (0.0–15.3) 2.8 (0.1–15.4) 2.3 (0.0–10.6) 0.32 
EDIH quintiles
12345
CharacteristicsaN = 204N = 205N = 205N = 205N = 205Pb
Male 117 (57.4%) 102 (49.8%) 91 (44.4%) 82 (40.0%) 56 (27.3%) <0.01 
Age, years 63 (28–85) 60 (24–78) 60 (31–82) 61 (29–83) 57 (21–80) <0.01 
BMI, kg/m2 25.9 (17.2–42.4) 27.3 (17.2–49.9) 27.3 (16.3–45.0) 27.5 (15.7–44.6) 28.2 (17.3–51.8) <0.01 
Race 0.05 
 White 180 (88.2%) 185 (90.2%) 187 (91.2%) 176 (85.9%) 181 (88.3%)  
 Black 8 (3.9%) 10 (4.9%) 14 (6.8%) 18 (8.8%) 18 (8.8%)  
 Other 16 (7.8%) 10 (4.9%) 4 (2.0%) 11 (5.4%) 6 (2.9%)  
Performance status 0.75 
 Fully active 152 (76.8%) 153 (75.7%) 154 (76.2%) 146 (73.4%) 145 (71.8%)  
 Restricted in strenuous activity 46 (23.2%) 49 (24.3%) 48 (23.8%) 53 (26.6%) 57 (28.2%)  
 Unknown  
Bowel wall invasion 0.65 
 Stage 1 and 2 28 (14.1%) 24 (11.9%) 34 (16.7%) 25 (12.7%) 26 (13.0%)  
 Stage 3 and 4 170 (85.9%) 178 (88.1%) 169 (83.3%) 172 (87.3%) 174 (87.0%)  
 Unknown  
Positive lymph nodes 0.05 
 1–3 or unknown 138 (69.3%) 134 (66.3%) 122 (60.1%) 135 (67.8%) 116 (57.4%)  
 ≥4 61 (30.7%) 68 (33.7%) 81 (39.9%) 64 (32.2%) 86 (42.6%)  
 Unknown  
Tumor location 0.04 
 Left-sided 68 (34.2%) 97 (48.0%) 90 (44.6%) 85 (42.7%) 95 (47.5%)  
 Transverse or right-sided 131 (65.8%) 105 (52.0%) 112 (55.4%) 114 (57.3%) 105 (52.5%)  
 Unknown  
Treatment arm 0.69 
 Fluorouracil + leucovorin 112 (54.9%) 103 (50.2%) 100 (48.8%) 99 (48.3%) 102 (49.8%)  
 Irinotecan, fluorouracil, leucovorin 92 (45.1%) 102 (49.8%) 105 (51.2%) 106 (51.7%) 103 (50.2%)  
Pack-years of smoking 0.3 (0.0–72.5) 0.6 (0.0–112.9) 0.6 (0.0–104.1) 1.3 (0.0–102.1) 0.6 (0.0–94.0) 0.10 
Physical activity, MET-hour/week 4.5 (0.0–119.9) 4.4 (0.0–122.9) 5.4 (0.0–114.2) 6.1 (0.0–147.4) 4.4 (0.0–125.2) 0.81 
Family history of colorectal cancer 49 (24.0%) 47 (22.9%) 50 (24.4%) 50 (24.4%) 60 (29.3%) 0.62 
Regular aspirin user 13 (6.4%) 18 (8.8%) 16 (7.8%) 17 (8.3%) 19 (9.3%) 0.85 
Calorie intakec, kcal/day 1,530 (617–3,666) 1,684 (649–3,691) 1,794 (860–3,534) 2,144 (1,032–3,578) 2,412 (1,181–4,119) <0.01 
EDIH-positive associationsd  
 Red meat, servings/week 1.1 (0.0–6.4) 1.6 (0.0–9.0) 1.9 (0.0–6.9) 2.5 (0.0–8.9) 3.7 (0.0–12.7) <0.01 
 Processed meat, servings/week 0.8 (0.0–3.8) 1.2 (0.0–7.5) 1.4 (0.0–6.2) 2.0 (0.0–14.0) 3.0 (0.0–27.0) <0.01 
 Low-energy beverages, servings/week 0.5 (0.0–18.2) 1.0 (0.0–17.9) 0.7 (0.0–28.4) 0.4 (0.0–27.6) 0.5 (0.0–55.1) 0.01 
 Cream soups, servings/week 0.2 (0.0–1.0) 0.2 (0.0–1.7) 0.3 (0.0–3.0) 0.4 (0.0–3.0) 0.3 (0.0–5.5) <0.01 
 Margarine, servings/week 0.7 (0.0–28.0) 1.9 (0.0–28.0) 3.0 (0.0–28.0) 3.0 (0.0–28.0) 3.0 (0.0–28.0) <0.01 
 Poultry, servings/week 1.1 (0.0–5.5) 1.2 (0.0–6.0) 1.2 (0.0–7.0) 1.6 (0.0–6.0) 1.8 (0.0–8.5) <0.01 
 Butter, servings/week 0.2 (0.0–17.5) 0.4 (0.0–17.5) 1.0 (0.0–18.4) 1.9 (0.0–28.0) 2.1 (0.0–28.0) <0.01 
 French fries, servings/week 0.2 (0.0–1.7) 0.4 (0.0–3.0) 0.5 (0.0–3.0) 0.5 (0.0–3.2) 1.0 (0.0–5.5) <0.01 
 Other fish, servings/week 0.9 (0.0–6.5) 1.0 (0.0–4.5) 1.0 (0.0–7.0) 1.1 (0.0–7.0) 1.0 (0.0–5.8) <0.01 
 High-energy beverages, servings/week 0.8 (0.0–13.3) 1.4 (0.0–14.3) 2.4 (0.0–20.5) 3.1 (0.0–25.5) 6.8 (0.0–38.5) <0.01 
 Tomatoes, servings/week 2.9 (0.0–10.7) 3.1 (0.0–13.0) 3.6 (0.2–13.9) 4.5 (0.2–18.0) 5.6 (0.2–18.0) <0.01 
 Low-fat dairy products, servings/week 0.7 (0.0–13.0) 0.5 (0.0–17.5) 0.4 (0.0–17.4) 0.4 (0.0–19.6) 0.3 (0.0–15.9) 0.01 
 Eggs, servings/week 0.9 (0.0–4.8) 1.9 (0.0–7.0) 3.0 (0.0–14.0) 3.0 (0.0–14.0) 3.0 (0.0–14.0) <0.01 
EDIH-negative associationsd  
 Wine, servings/week 0.4 (0.0–23.9) 0.1 (0.0–14.5) 0.0 (0.0–20.5) 0.0 (0.0–12.5) 0.0 (0.0–12.7) <0.01 
 Coffee, servings/week 7.0 (0.0–42.5) 7.0 (0.0–37.2) 7.5 (0.0–35.5) 6.3 (0.0–42.0) 6.1 (0.0–35.0) 0.02 
 Whole fruits, servings/week 11.1 (0.6–88.0) 10.4 (0.0–66.0) 9.2 (0.2–34.9) 9.6 (0.0–45.4) 8.1 (0.0–31.9) <0.01 
 High-fat dairy products, servings/week 3.4 (0.0–39.7) 4.2 (0.0–50.2) 5.3 (0.0–31.9) 6.0 (0.0–38.7) 6.9 (0.1–40.0) <0.01 
 Green leafy vegetables, servings/week 3.1 (0.0–17.5) 2.4 (0.0–14.7) 2.2 (0.0–15.3) 2.8 (0.1–15.4) 2.3 (0.0–10.6) 0.32 

Abbreviations: BMI, body mass index; EDIH, empirical dietary index for hyperinsulinemia; MET, metabolic equivalent of task.

aContinuous variables presented as median (range), and categorical variables as number (percentage).

bP values calculated using the χ2 test for categorical variables and Kruskal–Wallis test for continuous variables.

cTime-varying calorie intake.

dThe food groups (servings/week) retained are defined in ref. 4.

Table 2.

Associations of EDIH with adverse outcomesa in patients with stage III colon cancer (N = 1,024).

EDIH quintiles
12345Ptrend
Cancer recurrence or death from any cause (disease-free survival) 
 No. of events/at risk #87/204 #76/205 #80/205 #71/205 #80/205  
 Model 1 HR (95% CI)b Reference 0.85 (0.62–1.16) 0.88 (0.64–1.20) 0.77 (0.55–1.07) 0.99 (0.70–1.40) 0.88 
 Model 2 HR (95% CI)c Reference 0.78 (0.57–1.07) 0.80 (0.58–1.09) 0.65 (0.46–0.92) 0.80 (0.56–1.15) 0.21 
Cancer recurrence (recurrence-free survival) 
 No. of events/at risk #78/204 #68/205 #72/205 #61/205 #71/205  
 Model 1 HR (95% CI)b Reference 0.84 (0.60–1.16) 0.86 (0.62–1.19) 0.71 (0.50–1.02) 0.91 (0.63–1.32) 0.53 
 Model 2 HR (95% CI)c Reference 0.78 (0.56–1.09) 0.79 (0.56–1.10) 0.62 (0.43–0.90) 0.76 (0.51–1.11) 0.13 
Overall mortality 
 No. of events/at risk #74/204 #56/205 #64/205 #52/205 #65/205  
 Model 1 HR (95% CI)b Reference 0.74 (0.52–1.05) 0.83 (0.59–1.17) 0.67 (0.46–0.98) 1.02 (0.70–1.50) 0.94 
 Model 2 HR (95% CI)c Reference 0.65 (0.46–1.08) 0.72 (0.51–1.01) 0.53 (0.36–0.78) 0.77 (0.52–1.14) 0.21 
EDIH quintiles
12345Ptrend
Cancer recurrence or death from any cause (disease-free survival) 
 No. of events/at risk #87/204 #76/205 #80/205 #71/205 #80/205  
 Model 1 HR (95% CI)b Reference 0.85 (0.62–1.16) 0.88 (0.64–1.20) 0.77 (0.55–1.07) 0.99 (0.70–1.40) 0.88 
 Model 2 HR (95% CI)c Reference 0.78 (0.57–1.07) 0.80 (0.58–1.09) 0.65 (0.46–0.92) 0.80 (0.56–1.15) 0.21 
Cancer recurrence (recurrence-free survival) 
 No. of events/at risk #78/204 #68/205 #72/205 #61/205 #71/205  
 Model 1 HR (95% CI)b Reference 0.84 (0.60–1.16) 0.86 (0.62–1.19) 0.71 (0.50–1.02) 0.91 (0.63–1.32) 0.53 
 Model 2 HR (95% CI)c Reference 0.78 (0.56–1.09) 0.79 (0.56–1.10) 0.62 (0.43–0.90) 0.76 (0.51–1.11) 0.13 
Overall mortality 
 No. of events/at risk #74/204 #56/205 #64/205 #52/205 #65/205  
 Model 1 HR (95% CI)b Reference 0.74 (0.52–1.05) 0.83 (0.59–1.17) 0.67 (0.46–0.98) 1.02 (0.70–1.50) 0.94 
 Model 2 HR (95% CI)c Reference 0.65 (0.46–1.08) 0.72 (0.51–1.01) 0.53 (0.36–0.78) 0.77 (0.52–1.14) 0.21 

Abbreviations: CI, confidence interval; EDIH, empirical dietary index for hyperinsulinemia; HR, hazard ratio.

aAs defined previously (5), disease-free, recurrence-free, and overall survival since first questionnaire could be derived from adverse outcomes.

bAdjusted for age at diagnosis (years) and time-varying calorie intake.

cOn the basis of model 1, additionally adjusted for sex, treatment arm, T-stage (T1–T2 and T3–T4), positive lymph nodes (1–3 and ≥4), performance status (fully active and restricted in strenuous activity), aspirin (yes and no), tumor location (left-sided, transverse, or right-sided), family history of colorectal cancer (yes and no), pack-years of smoking, and time-varying BMI (kg/m2) and physical activity (MET-hour/week).

Because hyperinsulinemia promotes colon carcinogenesis, a diet inducing an elevated insulin response may accelerate tumor growth, thereby contributing to worse survival (1). Previous studies reported higher EDIH was associated with higher risk of colorectal cancer and digestive system cancers (2, 8). In a prior analysis, an alternative measure of dietary insulinemic potential, dietary insulin index (DII) was associated with an inferior survival among patients with colon cancer (3). EDIH, an index intended to measure dietary insulinemic potential, would be associated with worse colon cancer survival. However, in our study, we observed a nonsignificant trend toward an improved outcome with higher EDIH. The unexpected results may be explained by different thoughts of constructing EDIH and DII to predict hyperinsulinemia that EDIH was to predict chronic insulin levels and DII was supposed to measure acute effects of food items on insulin secretion. Because we found insignificant interactions between EDIH and sex, we reasonably believe that sex may not drive these unexpected results.

Our study has many strengths, including ascertainment of cancer diagnosis and events, standardized treatment and follow-up, and detailed prognostic information. Dietary intake was self-reported and potentially subject to measurement errors. However, the FFQ in our cohort has been extensively validated both in healthy populations and cohorts of patients with cancer. In conclusion, the diet of higher insulinemic potential is not associated with worse survival in patients with stage III colon cancer.

K. Ng is a consultant for Tarrex Biopharma; is an advisory board member for Bayer, Seattle Genetics, and Array Biopharma; and reports receiving commercial research grants from Revolution Medicines, Evergrande Group, Genentech, Gilead Sciences, Trovagene, Pharmavite, and Tarrex Biopharma. R.B. Mowat is an MD at Daiichi Sankyo. R. Whittom has an advisory board relationship with AstraZeneca Canada, Roche Canada, Bristol-Myers Squibb, and Purdue. A. Benson has a consulting/advisory relationship with National Comprehensive Cancer Network (NCCN), Bristol-Myers Squibb, ACCC, ECOG-ACRIN, Amgen, Imedex, Artemida Pharma, Intellisphere OncLiv, American College of Radiology, China National Medical Association 6th Affiliated Sun Yet-San Hospital, Springer, Health Advances, Patient Resource (educational document review), Lexicon, Research to Practice, AVBCC, Harborside, Guardant, Merck Sharpe and Dohme, Terumo, and Lexicon and reports receiving commercial research grants from Celgene, Infinity Pharmaceuticals, Amgen, ECOG-ACRIN, SynCore, Merck Sharp and Dohme, Taiho, Rafael Pharmaceuticals, MedImmune/AstraZeneca, Xencor, Bristol-Myers Squibb, PreECOG, and Astellas. H. Kindler is a consultant for AstraZeneca, Inventiva, Merck, Paradox, Deciphera, Aldeyra, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Erytech, Five Prime Therapeutics, Ipsen, and Kyowa. J.A. Meyerhardt reports receiving other remuneration from Taiho and COTA Healthcare. C.S. Fuchs reports receiving other commercial research support from Agois, Bain Capital, Entrinsic Health, CytomX Therapeutics, Daiichi Sankyo, Eli Lilly, Evolveimmune Therapeutics, Genentech, Merck, Taiho, and Unum Therapeutics; has ownership interest (including patents) in CytomX Therapeutics, Entrinsic Health, and Evolveimmune Therapeutics; and reports receiving other remuneration from Amylin Pharmaceuticals. No potential conflicts of interest were disclosed by the other authors.

Conception and design: E. Cheng, S. Zhang, E.L. Giovannucci, C.S. Fuchs

Development of methodology: E. Cheng, S. Zhang, C.S. Fuchs

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L.B. Saltz, R.B. Mowat, R. Whittom, A. Hantel, A. Benson, D. Atienza, M. Messino, H. Kindler, C.S. Fuchs

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): E. Cheng, S. Zhang, F.-S. Ou, B. Mullen, K. Ng, L.B. Saltz, E.L. Van Blarigan, J.A. Meyerhardt, C.S. Fuchs

Writing, review, and/or revision of the manuscript: E. Cheng, S. Zhang, F.-S. Ou, B. Mullen, K. Ng, L.B. Saltz, D. Niedzwiecki, R.J. Mayer, R. Whittom, A. Hantel, A. Benson, H. Kindler, E.L. Giovannucci, E.L. Van Blarigan, J.A. Meyerhardt, C.S. Fuchs

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): E. Cheng, L.B. Saltz, C.S. Fuchs

Study supervision: C.S. Fuchs

Research reported in this article was supported by the NCI of the NIH under award numbers U10CA180821 and U10CA180882 (to the Alliance for Clinical Trials in Oncology; D. Niedzwiecki and F.-S. Ou); UG1CA233290 (to L.B. Saltz); U10 CA180863 CCTG (to R. Whittom); U10CA180820 ECOG-ACRIN (to A. Benson); UG1CA233180 (to J.A. Meyerhardt, S. Zhang, and R.J. Mayer); UG1CA233327, UG1CA233337, UG1CA189858, and U10CA180888 SWOG (to A. Hantel); K07CA197077 (to E.L. Van Blarigan); R01CA118553 (to C.S. Fuchs and F.-S. Ou); and R01CA205406 (to K. Ng). CALGB 89803 clinical trial was supported in part by funds from Pharmacia & Upjohn Company (now Pfizer Oncology; https://acknowledgments.alliancefound.org). K. Ng was supported in part by Department of Defense grant CA160344. C.S. Fuchs was supported in part by a Stand Up To Cancer Colorectal Cancer Dream Team Translational Research Grant (SU2C-AACR-DT22-17). Stand Up To Cancer is a division of the Entertainment Industry Foundation. The research grant is administered by the American Association for Cancer Research, the Scientific Partner of SU2C.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Giovannucci
E
. 
Insulin and colon cancer
.
Cancer Causes Control
1995
;
6
:
164
79
.
2.
Tabung
FK
,
Wang
W
,
Fung
TT
,
Smith-Warner
SA
,
Keum
N
,
Wu
K
, et al
Association of dietary insulinemic potential and colorectal cancer risk in men and women
.
Am J Clin Nutr
2018
;
108
:
363
70
.
3.
Morales-Oyarvide
V
,
Yuan
C
,
Babic
A
,
Zhang
S
,
Niedzwiecki
D
,
Brand-Miller
JC
, et al
Dietary insulin load and cancer recurrence and survival in patients with stage III colon cancer: findings from CALGB 89803 (Alliance)
.
J Natl Cancer Inst
2019
;
111
:
170
9
.
4.
Tabung
FK
,
Wang
WK
,
Fung
TT
,
Hu
FB
,
Smith-Warner
SA
,
Chavarro
JE
, et al
Development and validation of empirical indices to assess the insulinaemic potential of diet and lifestyle
.
Brit J Nutr
2016
;
116
:
1787
98
.
5.
Meyerhardt
JA
,
Niedzwiecki
D
,
Hollis
D
,
Saltz
LB
,
Hu
FB
,
Mayer
RJ
, et al
Association of dietary patterns with cancer recurrence and survival in patients with stage III colon cancer
.
JAMA
2007
;
298
:
754
64
.
6.
Saltz
LB
,
Niedzwiecki
D
,
Hollis
D
,
Goldberg
RM
,
Hantel
A
,
Thomas
JP
, et al
Irinotecan fluorouracil plus leucovorin is not superior to fluorouracil plus leucovorin alone as adjuvant treatment for stage III colon cancer: results of CALGB 89803
.
J Clin Oncol
2007
;
25
:
3456
61
.
7.
Rimm
EB
,
Giovannucci
EL
,
Stampfer
MJ
,
Colditz
GA
,
Litin
LB
,
Willett
WC
. 
Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals
.
Am J Epidemiol
1992
;
135
:
1114
26
.
8.
Wang
W
,
Fung
TT
,
Wang
M
,
Smith-Warner
SA
,
Giovannucci
EL
,
Tabung
FK
. 
Association of the insulinemic potential of diet and lifestyle with risk of digestive system cancers in men and women
.
JNCI Cancer Spectr
2018
;
2
:
pky080
.