Background:

The prognostic relationship between diet and endometrial cancer survival remains largely unknown. We sought to determine pre- and post-diagnosis dietary composition, glycemic load (GL), inflammatory potential (dietary inflammatory index) and quality [Canadian Healthy Eating Index (C-HEI) 2005] associations with disease-free (DFS) and overall survival (OS) among endometrial cancer survivors. In addition, we assessed associations between dietary changes with OS and explored obesity/physical activity effect modification.

Methods:

Survivors, diagnosed in Alberta, Canada between 2002 and 2006, completed past-year, food-frequency questionnaires at-diagnosis (n = 503) and 3-year follow-up (n = 395). Participants were followed to death or January 2022. Cox proportional regression estimated HR [95% confidence intervals (CI)] for dietary survival associations.

Results:

During 16.9 median years of follow-up, 138 participants had a DFS event and 120 died. Lower pre-diagnosis GL (HRT1vsT3, 0.49; 95% CI, 0.25–0.97) and greater post-diagnosis energy intakes (EI) from total- and monounsaturated-fat (HRT3vsT1, 0.48; 95% CI, 0.26–0.87) were associated with better OS. Higher pre-diagnosis C-HEI, less inflammatory diets and lower added sugar intakes were nonlinearly associated with better DFS. Consistently low pre- to post-diagnosis EI from carbohydrates and total-fats were associated with better (HR, 0.36; 95% CI, 0.18–0.72) and worse (HR, 2.26; 95% CI, 1.21–4.20) OS, respectively. Decreased pre- to post-diagnosis C-HEI was associated with worse OS. In stratified analysis, healthy diets were most beneficial for survivors with obesity and physical inactivity.

Conclusions:

Adherence to higher quality dietary patterns were associated with better survival.

Impact:

Our study provides novel evidence that both pre- and post-diagnosis diet are important prognostic factors for endometrial cancer survivors. Post-diagnosis survival associations with diet composition and quality highlight the potential for future interventions.

Since the late 1990s, the global incidence of endometrial cancer has continued to rise, particularly in North America, Europe, and Oceania (1). The increasing incidence, coupled with relatively favorable 5-year survival rates, has led to a growing population of endometrial cancer survivors (1, 2). Given that the risk of most endometrial cancers is increased with certain chronic conditions and adverse lifestyle behaviours, understanding the prognostic role of such modifiable factors is increasingly important to improve long-term health and survival outcomes in this growing population (2).

Diet and nutritional components play a role in the etiology of several cancers including endometrial cancer (3, 4). Epidemiologic evidence suggests that foods high in animal fats and sugars may increase endometrial cancer risk while high consumption of vegetables and fruits may have beneficial effects (5). However, to date only dietary glycemic load (GL), a measure of total dietary carbohydrate burden, and diet-related glucose elevation are considered probable risk factors for endometrial cancer (3). Because there is limited evidence for the role that diet has during cancer survivorship, cancer prevention dietary recommendations are frequently used for cancer survivors (6, 7). These dietary recommendations often focus on cultural ways of eating, adherence to national recommendations or, biological markers and processes related to carcinogenic mechanisms (inflammation, metabolic or hormonal disruption; refs. 3, 4). Furthermore, because macronutrient composition has been associated with overall and cancer-related mortality, some dietary interventions, such as ketogenic diets, manipulate the relative energy intakes (EI) from different macronutrients as a means of improving survival outcomes (8–11).

To date, our understanding of the role that diet has on endometrial cancer prognosis is sparse, but two previous studies provide some guidance (12–14). Pre-diagnosis dietary inflammatory index (DII), a literature-derived tool to assess the inflammatory potential of diet, was not associated with survival in the Australian National Endometrial Cancer study (13). For ovarian (n = 19), cervical (n = 54), and uterine cancer (n = 47) survivors in the Third National Health and Nutrition Examination Survey post-diagnosis diet quality, assessed via Healthy Eating Index (HEI), was associated with greater survival (14).

Our aim was to examine the prognostic relationships between diet and endometrial cancer survival at both pre- and post-diagnosis time points. Specifically, the primary objective was to determine associations between dietary composition (macronutrient and major food groups), GL, inflammatory potential (DII), and diet quality defined by the Canadian HEI (C-HEI) 2005 with disease-free survival (DFS) and overall survival (OS) outcomes. Our secondary objective was to assess associations between dietary changes from pre- to post-diagnosis with OS. Finally, we explored whether obesity and physical activity groups modify these relationships. We hypothesize that greater compliance to, and preservation of, a healthy diet will be associated with more favorable survival outcomes.

Study design and participants

The Alberta Endometrial Cancer Cohort Study is a follow-up of primary histologically confirmed endometrial cancer cases who participated in a population-based, case–control study for which full details have been previously described (15, 16). Briefly, the Alberta Cancer Registry (ACR) originally identified participants between 2002 and 2006 (15). Five hundred forty participants completed baseline assessments at cancer diagnosis and 425 survivors participated in follow-up assessments between 2006 and 2011 (16). Of the 522 and 404 participants who completed baseline and follow-up food frequency questionnaires (FFQ), 19 and 9 participants were excluded from these analyses because of implausible total EI (<600 or >5,000 kcal/day) leaving 503 and 395 participants, respectively (Fig. 1). This study was conducted in accordance with the Declaration of Helsinki. All participants provided informed written consent. The University of Calgary, University of Alberta, and former Alberta Cancer Board provided ethical approval for this study and follow-up assessments.

Figure 1.

Alberta Endometrial Cancer Cohort Participant Overview from 2002 to 2022. a *implausible total daily EI < 600 or > 5,000 kcal/day.

Figure 1.

Alberta Endometrial Cancer Cohort Participant Overview from 2002 to 2022. a *implausible total daily EI < 600 or > 5,000 kcal/day.

Close modal

Data collection

Baseline and follow-up interviews were conducted by trained personnel approximately 4.4 months [interquartile range (IQR), 3.4–5.7 months] and 3.4 years (IQR, 3.2–3.8 years) after diagnosis, respectively (16, 17). At both time points, participants self-reported past-year dietary intake using the Canadian version of the NCI's Diet Health Questionnaire–I (C-DHQ-I). The C-DHQ-I is a validated 124-item FFQ of foods, beverages, and dietary supplements adapted to Canada for brand names, nutrient composition, food fortification processes, and seasonal food intake routinely used in cancer populations (17, 18). Portion sizes were described with cups, ounces, tablespoons, and teaspoons (tsp) as well as metric equivalent (17). The Diet*Calc (2002 Canadian version 1) Analysis Program (v1.5.0 Sept 2010) was used to derive nutrient intake variables. Demographic information, reproductive and medical histories, hormone use, comorbidities, and family history of cancer were captured at baseline and follow-up interviews. Directly measured, standardized height, weight, waist and hip circumference were taken at both time points. Pre-diagnosis lifetime physical activity was ascertained with the Lifetime Total Physical Activity Questionnaire (19) and post-diagnosis physical activity was assessed with the Past Year Total Physical Activity Questionnaire (20). The Charlson Comorbidity Index was calculated to reflect participants’ post-diagnosis comorbidities (21). Cancer histology, stage (American Joint Committee on Cancer; ref. 22), primary and adjuvant treatment(s), and cancer recurrence were abstracted from medical charts by trained ACR health record technicians. The study pathologist determined cancer grade following the International Federation of Gynecology and Obstetrics guidelines (23).

Dietary exposures

Overall diet quality was measured with the C-HEI 2005, a dietary pattern score that ranges from 0 to 100 and reflects age- and sex-specific alignment with Canada's Food Guide (CFG) recommendations that were most relevant at the time of assessment (24). The C-HEI consists of eight adequacy components (total fruits and vegetables, whole fruit, dark green and orange vegetables, milk and alternatives, meat and alternatives, total grain products, whole grain products and, unsaturated fat) and three moderation components [sodium, saturated fat, and other foods not recommended (e.g., added sugar, alcohol, solid fats); ref. 24]. Details of the C-HEI 2005 scoring system for Canadian food servings according to CFG using the C-DHQ-I have been previously described (25). Briefly, scores were proportionally assigned to the servings/amounts between the minimum intake and meeting the recommendation for adequacy components while, inverse score was applied for moderation components (24).

Energy from macronutrients for protein, carbohydrates, total-, saturated-, monounsaturated-, and polyunsaturated-fat were measured as a percentage of participants’ total dietary EI. Given that most participants met the CFG EI from protein (10%–35%), carbohydrates (45%–65%), and total fat (20%–35%), alignment with these guidelines was not assessed (26). Major food groups of the CFG including vegetables and fruits, grain products, milk and alternatives, meat and alternatives and, other discretionary foods (high fat, sugar, or salt items) were assessed on the basis of the C-HEI 2005 definitions and measured in standard servings (24). In the current study, average intake of total sugar (g/day) and added sugar were considered. Added sugar was defined as all sugars added to, but not naturally occurring in, prepared or processed food items such as breads/cakes, beverages, jams, chocolates, ice cream, and sugars eaten separately or added to foods at the table (27). Canadian health agencies have adopted the World Health Organization's recommendation to reduce free sugar intake to <10% of daily EI and promote further reductions of added sugars to 5% of total EI/day or approximately 6 tsp/day (28). In alignment with the reporting recommendations (STROBE-nut), added sugar was assessed per serving (tsp/day) to facilitate interpretations (29).

GL was estimated from total carbohydrates dietary intake (g), multiplied by the food's glycemic index, and divided by 100. Dietary GL has often been assessed as the highest compared with lowest levels, however the Glycemic Index Foundation has recommended that, for optimal health, individuals should target a total GL < 100 g/day (30).

The inflammatory potential of participants’ diets was assessed with the DII, for which positive and negative DII scores (potential range, −8.87 to 7.98) reflect pro-inflammatory and anti-inflammatory potential, respectively (31). Participants’ overall DII scores were derived using 38 of the 45 potential food parameters including supplement intakes from iron, magnesium, zinc, niacin, selenium, thiamin, β-carotene, folic acid and vitamins A, B2, B6, B12, C, D, and E, following the standard methods (31). The seven items missing from this cohort's dietary data were: garlic, ginger, rosemary, saffron, thyme/oregano, turmeric, and eugenol (31). Participants’ DII scores were examined in reference to diets with neutral-inflammatory potential.

Outcome ascertainment

The ACR preformed record linkages with Vital Statistics Alberta to ascertain vital status. DFS and OS were defined as the time to the first recurrence/death and death from any cause, respectively. Participants were followed from cancer diagnosis until death or January 27, 2022, whichever occurred first.

Statistical analysis

Associations between pre-diagnosis and post-diagnosis dietary factors, as well as dietary changes from pre- to post-diagnosis, with survival outcomes were investigated using Cox proportional hazards models. Pre-diagnosis assessments measured survival time from cancer diagnosis while post-diagnosis and changes models used follow-up interview date. Multivariable-adjusted HR and 95% confidence intervals (CI) were estimated for participants with the healthiest compared with unhealthiest dietary measures with DFS and OS endpoints. As most recurrences are expected to occur within 3 years of diagnosis, DFS was not assessed in post-diagnosis analysis (n ≤ 5). Pre- to post-diagnosis change assessments used median groups when clinical references were unavailable. Unsaturated- and saturated-fat components were excluded from the CFG assessments to avoid duplication of results with the EI from dietary fat analysis. Ptrend was obtained using tertiles to reflect potential trends in improved dietary factors. Linearity of associations was assessed by fitting restricted cubic splines in the multivariable survival models (32). The proportional hazards assumption was evaluated via visual examination and statistical assessment of Schoenfeld residuals. Missing exposure data were addressed with list-wise deletion and characteristics of included participants were compared with those with missing data. All analyses were two-sided, considered statistically significant at P < 0.05 and performed with STATA 17 (RRID:SCR_012763).

Age at diagnosis (years), cancer stage (I, II, III/IV), grade (I/II, III, unknown or non-applicable), and primary treatment(s) (hysterectomy, chemotherapy, radiation, any adjuvant, missing), were selected as covariates a priori based on biological plausibility. All models were adjusted for total EI (kcal/day) except the percent EI from macronutrients models as these were already relative to participants’ total EI. Models were additionally adjusted for body mass index (BMI), family history of uterine or colorectal cancer, comorbidities [baseline: number of major comorbidities (0, 1, 2); follow-up: Charlson Comorbidity Index (score)] and smoking pack-years. Final models were derived using backwards elimination with a ∼10% change in the estimated HR considered meaningful when evaluating potential confounders one at a time with all other covariates in the model. There was insufficient evidence for confounding of the associations of interest for marital status, parity, menopausal status, hormone therapy, highest education attained, alcohol consumption, total physical activity, or nonlinear age. Potential modification of the associations between dietary exposures with DFS and OS by BMI (<30 vs. ≥30 kg/m2) and lifetime total physical activity [median threshold: <101 vs. ≥101 metabolic equivalent task (MET)-hours/week/year] groups were assessed with Wald tests. Sensitivity analysis excluding event within 2 years of diagnosis was conducted.

Data availability

Data described in the manuscript, codebook, and analytic code will be made available upon reasonable request to the corresponding author.

Participant characteristics

During a median 16.9 years of follow-up (IQR, 15.5–18.1 years), 138 of the 503 participants had a recurrence and/or died. Specifically, 67 participants had a recurrence and there were 120 deaths overall. Median age at diagnosis was 59.0 years, survivors were primarily of European ancestry (95.2%), achieved greater than a high school education (66.8%) and, were diagnosed with early stage (I: 78.9%) and grade (I/II: 76.7%) endometrial cancers (Table 1). Participants who completed follow-up had similar characteristics to those who did not. At baseline, the median C-HEI score was 52.3 (IQR, 45.9–59.5) and DII scores reflected anti-inflammatory diets. Pre-diagnosis median GL (90.3 g/day) was slightly greater than post-diagnosis measures (85.5 g/day). Pre-diagnosis EIs from protein (16.2%), total fats (32.9%), and carbohydrates (51.8%) were similar to post-diagnosis intakes. Compared with participants with pre-diagnosis dietary history data, those without these data (n = 37) were less likely to have undergone a hysterectomy for their primary treatment, less likely to have one comorbidity, and more likely to be current smokers (Supplementary Table S1). Participants without post-diagnosis dietary histories (n = 30) had higher levels of education and greater BMI than participants with these data (n = 395; Supplementary Table S1).

Table 1.

Characteristics of the Alberta Endometrial Cancer Cohort participants and dietary details pre-diagnosis (n = 503) and post-diagnosis (n = 395) by vital status, 2002–2022.

AllAliveDFSOverall deaths
CharacteristicMedian (IQR)/N (%)Median (IQR)/N (%)Median (IQR)/N (%)Median (IQR)/N (%)
Participants 503 383 138 120 
Age at Dx (year) 59.00 (53.00–65.00) 57.00 (53.00–63.00) 64.00 (58.00–72.00) 64.00 (59.00–73.00) 
Highest education 
 High school diploma 167 (33.20) 116 (30.29) 57 (41.30) 51 (42.50) 
 Non-university certificate 229 (45.53) 178 (46.48) 59 (42.75) 51 (42.50) 
 University degree 107 (21.27) 89 (23.24) 22 (15.94) 18 (15.00) 
Married or common-law 351 (69.78) 272 (71.02) 90 (65.22) 79 (65.83) 
European ethnic ancestry 479 (95.23) 363 (94.78) 130 (94.20) 116 (96.67) 
Overall AJCC stage 
 I 397 (78.93) 323 (84.33) 85 (61.59) 74 (61.67) 
 II 66 (13.12) 42 (10.97) 27 (19.57) 24 (20.00) 
 III/IV 40 (7.95) 18 (4.70) 26 (18.84) 22 (18.33) 
FIGO grade 
 I/II 386 (76.74) 311 (81.20) 85 (61.59) 75 (62.50) 
 III 68 (13.52) 40 (10.44) 32 (23.19) 28 (23.33) 
 Other 49 (9.74) 32 (8.36) 21 (15.22) 17 (14.17) 
Primary treatment 
 Surgery 493 (98.01) 377 (98.43) 128 (92.75) 116 (96.67) 
 Chemotherapy 40 (7.95) 25 (6.53) 16 (11.59) 15 (12.50) 
 Hormone therapy 6 (1.19) 6 (1.57) 2 (1.45) 0 (0) 
 Radiation therapy 158 (31.41) 113 (29.50) 52 (37.68) 45 (37.50) 
BMI (kg/m230.97 (26.49–36.78) 30.82 (25.99–36.97) 31.13 (27.43–35.99) 31.10 (27.59–35.78) 
Total lifetime physical activity (MET-h/wk/year) 100.89 (79.40–126.56) 100.04 (79.42–125.20) 107.43 (79.91–135.24) 108.42 (78.79–135.65) 
Pre-Dx EI from macronutrients (%) 
 Protein 16.23 (14.41–18.09) 16.36 (14.63–18.20) 15.85 (14.28–17.55) 15.75 (14.05–17.68) 
 Carbohydrates 51.77 (46.38–56.77) 51.20 (46.12–56.16) 53.18 (47.86–57.56) 53.44 (48.13–57.96) 
 Total fats 32.94 (28.13–37.25) 33.20 (28.58–37.51) 32.65 (27.98–36.80) 32.44 (27.23–36.80) 
Pre-Dx C-HEI05 (score) 52.26 (45.86–59.49) 52.48 (45.86–59.48) 51.12 (45.42–60.27) 51.19 (45.97–60.31) 
Pre-Dx DII (points) −1.53 (−4.12 to 1.31) −1.70 (−4.11 to 1.00) −0.95 (−4.15 to 2.32) −1.15 (−4.14 to 1.85) 
Pre-Dx GL (g/day) 90.30 (69.84–115.60) 90.05 (69.13–114.30) 89.88 (68.40–116.30) 91.74 (71.18–119.10) 
Pre-Dx total sugar (g/day) 90.32 (65.17–121.00) 88.99 (65.17–119.40) 92.38 (63.89–125.20) 95.67 (65.18–127.70) 
Pre-Dx added sugar (tsp/day) 8.05 (5.61–11.60) 7.92 (5.61–11.26) 8.27 (5.59–11.86) 8.52 (5.62–12.24) 
Post-Dx EI from macronutrients (%) 
 Protein 16.39 (14.72–18.22) 16.47 (14.78–18.30)  15.95 (14.10–17.40) 
 Carbohydrates 51.44 (46.09–56.01) 50.65 (45.88–55.40)  54.35 (47.83–58.36) 
 Total fats 33.50 (28.77–37.65) 33.59 (29.43–37.78)  31.45 (28.25–37.37) 
Post-Dx C-HEI05 (score) 53.70 (46.00–60.63) 53.86 (46.00–60.50)  52.49 (46.45–60.79) 
Post-Dx DII (points) −1.59 (−4.12 to 0.83) −1.52 (−4.06 to 0.78)  −1.79 (−4.60 to 1.26) 
Post-Dx GL (g/day) 85.46 (68.70–108.80) 85.16 (68.96–107.90)  87.31 (68.21–119.20) 
Post-Dx total sugar (g/day) 89.44 (66.38–114.90) 88.74 (66.58–113.10)  91.01 (64.51–129.50) 
Post-Dx added sugar(tsp/day) 8.11 (5.44–11.84) 8.11 (5.44–11.51)  8.06 (5.72–13.24) 
AllAliveDFSOverall deaths
CharacteristicMedian (IQR)/N (%)Median (IQR)/N (%)Median (IQR)/N (%)Median (IQR)/N (%)
Participants 503 383 138 120 
Age at Dx (year) 59.00 (53.00–65.00) 57.00 (53.00–63.00) 64.00 (58.00–72.00) 64.00 (59.00–73.00) 
Highest education 
 High school diploma 167 (33.20) 116 (30.29) 57 (41.30) 51 (42.50) 
 Non-university certificate 229 (45.53) 178 (46.48) 59 (42.75) 51 (42.50) 
 University degree 107 (21.27) 89 (23.24) 22 (15.94) 18 (15.00) 
Married or common-law 351 (69.78) 272 (71.02) 90 (65.22) 79 (65.83) 
European ethnic ancestry 479 (95.23) 363 (94.78) 130 (94.20) 116 (96.67) 
Overall AJCC stage 
 I 397 (78.93) 323 (84.33) 85 (61.59) 74 (61.67) 
 II 66 (13.12) 42 (10.97) 27 (19.57) 24 (20.00) 
 III/IV 40 (7.95) 18 (4.70) 26 (18.84) 22 (18.33) 
FIGO grade 
 I/II 386 (76.74) 311 (81.20) 85 (61.59) 75 (62.50) 
 III 68 (13.52) 40 (10.44) 32 (23.19) 28 (23.33) 
 Other 49 (9.74) 32 (8.36) 21 (15.22) 17 (14.17) 
Primary treatment 
 Surgery 493 (98.01) 377 (98.43) 128 (92.75) 116 (96.67) 
 Chemotherapy 40 (7.95) 25 (6.53) 16 (11.59) 15 (12.50) 
 Hormone therapy 6 (1.19) 6 (1.57) 2 (1.45) 0 (0) 
 Radiation therapy 158 (31.41) 113 (29.50) 52 (37.68) 45 (37.50) 
BMI (kg/m230.97 (26.49–36.78) 30.82 (25.99–36.97) 31.13 (27.43–35.99) 31.10 (27.59–35.78) 
Total lifetime physical activity (MET-h/wk/year) 100.89 (79.40–126.56) 100.04 (79.42–125.20) 107.43 (79.91–135.24) 108.42 (78.79–135.65) 
Pre-Dx EI from macronutrients (%) 
 Protein 16.23 (14.41–18.09) 16.36 (14.63–18.20) 15.85 (14.28–17.55) 15.75 (14.05–17.68) 
 Carbohydrates 51.77 (46.38–56.77) 51.20 (46.12–56.16) 53.18 (47.86–57.56) 53.44 (48.13–57.96) 
 Total fats 32.94 (28.13–37.25) 33.20 (28.58–37.51) 32.65 (27.98–36.80) 32.44 (27.23–36.80) 
Pre-Dx C-HEI05 (score) 52.26 (45.86–59.49) 52.48 (45.86–59.48) 51.12 (45.42–60.27) 51.19 (45.97–60.31) 
Pre-Dx DII (points) −1.53 (−4.12 to 1.31) −1.70 (−4.11 to 1.00) −0.95 (−4.15 to 2.32) −1.15 (−4.14 to 1.85) 
Pre-Dx GL (g/day) 90.30 (69.84–115.60) 90.05 (69.13–114.30) 89.88 (68.40–116.30) 91.74 (71.18–119.10) 
Pre-Dx total sugar (g/day) 90.32 (65.17–121.00) 88.99 (65.17–119.40) 92.38 (63.89–125.20) 95.67 (65.18–127.70) 
Pre-Dx added sugar (tsp/day) 8.05 (5.61–11.60) 7.92 (5.61–11.26) 8.27 (5.59–11.86) 8.52 (5.62–12.24) 
Post-Dx EI from macronutrients (%) 
 Protein 16.39 (14.72–18.22) 16.47 (14.78–18.30)  15.95 (14.10–17.40) 
 Carbohydrates 51.44 (46.09–56.01) 50.65 (45.88–55.40)  54.35 (47.83–58.36) 
 Total fats 33.50 (28.77–37.65) 33.59 (29.43–37.78)  31.45 (28.25–37.37) 
Post-Dx C-HEI05 (score) 53.70 (46.00–60.63) 53.86 (46.00–60.50)  52.49 (46.45–60.79) 
Post-Dx DII (points) −1.59 (−4.12 to 0.83) −1.52 (−4.06 to 0.78)  −1.79 (−4.60 to 1.26) 
Post-Dx GL (g/day) 85.46 (68.70–108.80) 85.16 (68.96–107.90)  87.31 (68.21–119.20) 
Post-Dx total sugar (g/day) 89.44 (66.38–114.90) 88.74 (66.58–113.10)  91.01 (64.51–129.50) 
Post-Dx added sugar(tsp/day) 8.11 (5.44–11.84) 8.11 (5.44–11.51)  8.06 (5.72–13.24) 

aParticipants with incomplete TNM stage (n = 8) were classified on the basis of available lymph node inclusion and metastatic information as stage I.

bThe frequencies for treatment are not mutually exclusive because participants could have multiple treatments.

cEthnic ancestry was based on participants self-report of the ethnic or cultural groups they, or most of their ancestors belong to.

dPost-diagnosis dietary data was available for n = 395 participants (alive: n = 321; overall deaths: n = 74).

eAJCC, American Joint Committee on Cancer; C-HEI05, Canadian Healthy Eating Index 2005; FIGO, International Federation of Gynecology and Obstetrics; Dx, Diagnosis.

Pre- and post-diagnosis dietary exposures

In multivariable adjusted pre-diagnosis models, lower GL were associated with better OS while moderate EI from saturated fat was associated with worse OS (Table 2).

Table 2.

DFS and OS outcomes for pre-diagnosis (n = 503) and post-diagnosis (n = 395) EI from macronutrients, sugar intake and, GL.

Pre-diagnosis DFSa,cPre-diagnosis OSa,cPost-diagnosis OSb,c
Events/CasesHR (95% CI)Events/CasesHR (95% CI)Events/CasesHR (95% CI)
EI from protein (%) 
T1: <15.1 51/168 1.00 47/168 1.00 T1: <15.1 27/132 1.00 
T2: 15.1–17.4 49/168 1.21 (0.81–1.82) 38/168 0.92 (0.59–1.44) T2: 15.1–17.5 31/132 1.53 (0.90–2.59) 
T3: >17.4 38/167 0.78 (0.51–1.21) 35/167 0.77 (0.49–1.21) T3: >17.5 16/131 0.74 (0.39–1.42) 
Ptrend  0.30  0.26 Ptrend  0.53 
EI from total fat (%) 
T1: <30.0 47/168 1.00 44/168 1.00 T1: <30.8 34/133 1.00 
T2: 30.0–35.8 49/168 1.12 (0.74–1.69) 41/168 1.02 (0.66–1.58) T2: 30.8–35.9 19/131 0.47 (0.26–0.85) 
T3: >35.8 42/167 0.89 (0.57–1.38) 35/167 0.87 (0.55–1.38) T3: >35.9 21/131 0.47 (0.27–0.82) 
Ptrend  0.61  0.57 Ptrend  0.01 
EI from saturated fat (%) 
T1: <9.4 44/169 1.00 41/169 1.00 T1: <9.3 25/132 1.00 
T2: 9.4–11.7 50/167 1.47 (0.95–2.25) 44/167 1.58 (1.01–2.48) T2: 9.3–11.5 30/133 1.08 (0.63–1.86) 
T3: >11.7 44/167 1.13(0.73–1.75) 35/167 1.10 (0.69–1.76) T3: >11.5 19/130 0.59 (0.32–1.10) 
Ptrend  0.58  0.52 Ptrend  0.11 
EI from monounsaturated fat (%) 
T1: <11.0 48/168 1.00 45/168 1.00 T1: <11.4 32/133 1.00 
T2: 11.0–13.3 48/168 1.24 (0.83–1.87) 40/168 1.07 (0.69–1.66) T2: 11.4–13.7 24/134 0.70 (0.41–1.21) 
T3: >13.3 42/167 0.82 (0.53–1.27) 35/167 0.88 (0.55–1.40) T3: >13.7 18/128 0.48 (0.26–0.87) 
Ptrend  0.41  0.62 Ptrend  0.014 
EI from polyunsaturated fat (%) 
T1: <6.4 47/170 1.00 42/170 1.00 T1: <6.4 29/133 1.00 
T2: 6.4–7.9 51/166 1.13 (0.76–1.70) 45/166 1.22 (0.80–1.88) T2: 6.4–8.2 24/131 0.68 (0.39–1.20) 
T3: >7.9 40/167 0.84 (0.54–1.30) 33/167 0.78 (0.49–1.26) T3: >8.2 21/131 0.57 (0.32–1.01) 
Ptrend  0.45  0.36 Ptrend  0.05 
EI from carbohydrates (%) 
T1: <48.2 38/168 1.00 31/168 1.00 T1: <47.9 19/132 1.00 
T2: 48.2–55.0 46/169 1.53 (0.98–2.41) 38/169 1.30 (0.79–2.12) T2: 47.9–54.6 20/132 0.97 (0.50–1.88) 
T3: >55.0 54/166 1.30 (0.83–2.03) 51/166 1.32 (0.82–2.13) T3: >54.6 35/131 1.91 (1.05–3.48) 
Ptrend  0.29  0.28 Ptrend  0.02 
Total sugar (g/day) 
T3: >106.5 48/167 1.00 45/167 1.00 T3: >106.7 29/131 1.00 
T2: 73.3–106.5 44/168 0.66 (0.41–1.06) 38/168 0.64 (0.39–1.05) T2: 73.9–106.7 23/132 0.72 (0.37–1.39) 
T1: <73.3 46/168 0.60 (0.35–1.03) 37/168 0.56 (0.31–1.00) T1: <73.9 22/132 0.69 (0.33–1.46) 
Ptrend  0.07  0.05   0.35 
GL (g/day) 
T3: >106.5 45/167 1.00 43/167 1.00 T3: >99.1 31/131 1.00 
T2: 77.9–106.5 47/168 0.79 (0.47–1.33) 39/168 0.56 (0.32–0.97) T2: 74.6–99.1 19/132 0.51 (0.25–1.02) 
T1: <77.9 46/168 0.66 (0.36–1.23) 38/168 0.49 (0.25–0.97) T1: <74.6 24/132 0.62 (0.27–1.42) 
Ptrend  0.20  0.06   0.32 
Pre-diagnosis DFSa,cPre-diagnosis OSa,cPost-diagnosis OSb,c
Events/CasesHR (95% CI)Events/CasesHR (95% CI)Events/CasesHR (95% CI)
EI from protein (%) 
T1: <15.1 51/168 1.00 47/168 1.00 T1: <15.1 27/132 1.00 
T2: 15.1–17.4 49/168 1.21 (0.81–1.82) 38/168 0.92 (0.59–1.44) T2: 15.1–17.5 31/132 1.53 (0.90–2.59) 
T3: >17.4 38/167 0.78 (0.51–1.21) 35/167 0.77 (0.49–1.21) T3: >17.5 16/131 0.74 (0.39–1.42) 
Ptrend  0.30  0.26 Ptrend  0.53 
EI from total fat (%) 
T1: <30.0 47/168 1.00 44/168 1.00 T1: <30.8 34/133 1.00 
T2: 30.0–35.8 49/168 1.12 (0.74–1.69) 41/168 1.02 (0.66–1.58) T2: 30.8–35.9 19/131 0.47 (0.26–0.85) 
T3: >35.8 42/167 0.89 (0.57–1.38) 35/167 0.87 (0.55–1.38) T3: >35.9 21/131 0.47 (0.27–0.82) 
Ptrend  0.61  0.57 Ptrend  0.01 
EI from saturated fat (%) 
T1: <9.4 44/169 1.00 41/169 1.00 T1: <9.3 25/132 1.00 
T2: 9.4–11.7 50/167 1.47 (0.95–2.25) 44/167 1.58 (1.01–2.48) T2: 9.3–11.5 30/133 1.08 (0.63–1.86) 
T3: >11.7 44/167 1.13(0.73–1.75) 35/167 1.10 (0.69–1.76) T3: >11.5 19/130 0.59 (0.32–1.10) 
Ptrend  0.58  0.52 Ptrend  0.11 
EI from monounsaturated fat (%) 
T1: <11.0 48/168 1.00 45/168 1.00 T1: <11.4 32/133 1.00 
T2: 11.0–13.3 48/168 1.24 (0.83–1.87) 40/168 1.07 (0.69–1.66) T2: 11.4–13.7 24/134 0.70 (0.41–1.21) 
T3: >13.3 42/167 0.82 (0.53–1.27) 35/167 0.88 (0.55–1.40) T3: >13.7 18/128 0.48 (0.26–0.87) 
Ptrend  0.41  0.62 Ptrend  0.014 
EI from polyunsaturated fat (%) 
T1: <6.4 47/170 1.00 42/170 1.00 T1: <6.4 29/133 1.00 
T2: 6.4–7.9 51/166 1.13 (0.76–1.70) 45/166 1.22 (0.80–1.88) T2: 6.4–8.2 24/131 0.68 (0.39–1.20) 
T3: >7.9 40/167 0.84 (0.54–1.30) 33/167 0.78 (0.49–1.26) T3: >8.2 21/131 0.57 (0.32–1.01) 
Ptrend  0.45  0.36 Ptrend  0.05 
EI from carbohydrates (%) 
T1: <48.2 38/168 1.00 31/168 1.00 T1: <47.9 19/132 1.00 
T2: 48.2–55.0 46/169 1.53 (0.98–2.41) 38/169 1.30 (0.79–2.12) T2: 47.9–54.6 20/132 0.97 (0.50–1.88) 
T3: >55.0 54/166 1.30 (0.83–2.03) 51/166 1.32 (0.82–2.13) T3: >54.6 35/131 1.91 (1.05–3.48) 
Ptrend  0.29  0.28 Ptrend  0.02 
Total sugar (g/day) 
T3: >106.5 48/167 1.00 45/167 1.00 T3: >106.7 29/131 1.00 
T2: 73.3–106.5 44/168 0.66 (0.41–1.06) 38/168 0.64 (0.39–1.05) T2: 73.9–106.7 23/132 0.72 (0.37–1.39) 
T1: <73.3 46/168 0.60 (0.35–1.03) 37/168 0.56 (0.31–1.00) T1: <73.9 22/132 0.69 (0.33–1.46) 
Ptrend  0.07  0.05   0.35 
GL (g/day) 
T3: >106.5 45/167 1.00 43/167 1.00 T3: >99.1 31/131 1.00 
T2: 77.9–106.5 47/168 0.79 (0.47–1.33) 39/168 0.56 (0.32–0.97) T2: 74.6–99.1 19/132 0.51 (0.25–1.02) 
T1: <77.9 46/168 0.66 (0.36–1.23) 38/168 0.49 (0.25–0.97) T1: <74.6 24/132 0.62 (0.27–1.42) 
Ptrend  0.20  0.06   0.32 

aPre-diagnosis models adjusted for: age at diagnosis, cancer stage, grade, primary treatment(s), at diagnosis BMI, family history of uterine or colorectal cancer, comorbidities (number of major comorbidities), at diagnosis lifetime smoking pack years.

bPost-diagnosis models adjusted for: age at diagnosis, cancer stage, grade, primary treatment(s), post-diagnosis BMI, family history of uterine or colorectal cancer, post-diagnosis comorbidities (Charlson Comorbidity Index), post-diagnosis lifetime smoking pack years.

cSugar and GL models were additionally adjusted for pre-diagnosis EI (kcal/day) and post-diagnosis EI (kcal/day) in pre-diagnosis and post-diagnosis assessments, respectively.

The relationships between added sugar, C-HEI and DII with survival outcomes were nonlinear (Fig. 2). Compared with consuming an average of 6 tsp/day of added sugar, intakes < 5.8 tsp/day were associated with better DFS while statistically significant harmful DFS relationships were observed for added sugar intakes between 6.3–12.7 tsp/day. Similarly, greater OS associations were noted for added sugar intakes < 5.1 tsp/day. Compared with the lowest C-HEI scores, better DFS was observed for scores ranging from 25.5 to 62.0 points. Relative to DII scores of zero, DII scores ≥ 0.3 were associated with statistically significantly worse DFS, while there was insufficient evidence of DFS associations for DII scores below zero.

Figure 2.

HRs (95% CI) for (A) pre-diagnosis added sugar and DFS (B) pre-diagnosis added sugar and OS (C) post-diagnosis added sugar and OS (D) pre-diagnosis C-HEI 2005 and DFS (E) pre-diagnosis C-HEI 2005 and OS (F) post-diagnosis C-HEI 2005 and OS (G) pre-diagnosis DII and DFS (H) pre-diagnosis DII and OS and (I) post-diagnosis DII and OS in the Alberta Endometrial Cancer Cohort Study, 2002–2022. Pre-diagnosis models adjusted for: age at diagnosis, cancer stage, grade, primary treatment(s), pre-diagnosis EI, at diagnosis BMI, family history of uterine or colorectal cancer, comorbidities (number of major comorbidities), pre-diagnosis lifetime smoking pack years. Post-diagnosis models adjusted for: age at diagnosis, cancer stage, grade, primary treatment(s), post-diagnosis EI, post-diagnosis BMI, family history of uterine or colorectal cancer, post-diagnosis comorbidities (Charlson Comorbidity Index), post-diagnosis lifetime smoking pack years. The reference for added sugar was 6 tsp/day; the reference for DII was 0 points (neutral-inflammatory potential).

Figure 2.

HRs (95% CI) for (A) pre-diagnosis added sugar and DFS (B) pre-diagnosis added sugar and OS (C) post-diagnosis added sugar and OS (D) pre-diagnosis C-HEI 2005 and DFS (E) pre-diagnosis C-HEI 2005 and OS (F) post-diagnosis C-HEI 2005 and OS (G) pre-diagnosis DII and DFS (H) pre-diagnosis DII and OS and (I) post-diagnosis DII and OS in the Alberta Endometrial Cancer Cohort Study, 2002–2022. Pre-diagnosis models adjusted for: age at diagnosis, cancer stage, grade, primary treatment(s), pre-diagnosis EI, at diagnosis BMI, family history of uterine or colorectal cancer, comorbidities (number of major comorbidities), pre-diagnosis lifetime smoking pack years. Post-diagnosis models adjusted for: age at diagnosis, cancer stage, grade, primary treatment(s), post-diagnosis EI, post-diagnosis BMI, family history of uterine or colorectal cancer, post-diagnosis comorbidities (Charlson Comorbidity Index), post-diagnosis lifetime smoking pack years. The reference for added sugar was 6 tsp/day; the reference for DII was 0 points (neutral-inflammatory potential).

Close modal

At follow-up, greater EIs from total- and monounsaturated-fat were associated with a two-fold increase in OS in multivariable adjusted models (Table 2). Post-diagnosis EI from carbohydrates > 54.6% was associated with worse OS (HRT3, 1.91; 95% CI, 1.05–3.48; Ptrend, 0.02). There was insufficient evidence of survival associations with total sugar, GL, C-HEI, and DII score for post-diagnosis intakes (Table 2; Fig. 2). The nine C-HEI food group components assessed were not associated with DFS or OS at either time point (Supplementary Table S2).

Reduced precision was observed in all sensitivity analysis estimates, yet only the GL OS association was no longer statistically significant (Supplementary Table S3).

Change in dietary exposures from pre- to post-diagnosis

Compared with being above the median at both time points, consistently low EIs from total fats and carbohydrates were associated with worse (HR, 2.26; 95% CI, 1.21–4.20) and better OS (HR, 0.36; 95% CI, 0.18–0.72), respectively (Table 3). A high-to-low change in C-HEI score from pre- to post-diagnosis was associated with reduced OS (HR, 2.39; 95% CI, 1.11–5.16) compared with consistently low scores.

Table 3.

OS outcomes for pre-diagnosis to post-diagnosis change in EI from macronutrients, sugar intake, GL, Canadian Healthy Eating Index 2005 and, DII (n = 395) in the Alberta Endometrial Cancer Cohort Study, 2002–2022.

Events/CasesOS HR (95% CI)a
EI from protein (%) 
 Unchanged high 23/141 1.00 
 Increased from low to high 9/63 0.97 (0.43–2.19) 
 Decreased from high to low 11/57 1.06 (0.51–2.20) 
 Unchanged low 26/119 1.01 (0.56–1.82) 
EI from total fat (%) 
 Unchanged high 17/139 1.00 
 Increased from low to high 11/63 1.34 (0.61–2.93) 
 Decreased from high to low 11/51 1.92 (0.87–4.23) 
 Unchanged low 20/127 2.26 (1.21–4.20) 
EI from carbohydrates (%) 
 Unchanged high 32/130 1.00 
 Increased from low to high 16/62 0.98 (0.52–1.88) 
 Decreased from high to low 9/59 0.55 (0.25–1.18) 
 Unchanged low 12/129 0.36 (0.18–0.72) 
Total sugar (g/day) 
 Unchanged high 27/127 1.00 
 Increased from low to high 9/58 0.53 (0.23–1.22) 
 Decreased from high to low 14/64 0.91 (0.42–1.94) 
 Unchanged low 19/131 0.65 (0.32–1.34) 
Average added sugar (tsp/day) 
 Unchanged high 40/223 1.00 
 Increased from low to high 10/44 0.99 (0.47–2.07) 
 Decreased from high to low 8/49 0.94 (0.41–2.20) 
 Unchanged low 11/64 0.79 (0.38–1.64) 
GL (g/day) 
 Unchanged high 18/80 1.00 
 Increased from low to high 10/43 1.05 (0.45–2.41) 
 Decreased from high to low 9/66 0.60 (0.23–1.55) 
 Unchanged low 32/191 0.54 (0.24–1.22) 
C-HEI 2005 (score) 
 Unchanged low 22/127 1.00 
 Decreased from high to low 11/42 2.39 (1.11–5.16) 
 Increased from low to high 12/59 1.67 (0.79–3.53) 
 Unchanged high 24/152 0.89 (0.47–1.67) 
DII (points) 
 Unchanged high (pro-inflammatory) 17/83 1.00 
 Increased from low to high 6/42 0.75 (0.28–1.96) 
 Decreased from high to low 11/54 1.16 (0.52–2.57) 
 Unchanged low (anti-inflammatory) 35/201 0.76 (0.39–1.47) 
Events/CasesOS HR (95% CI)a
EI from protein (%) 
 Unchanged high 23/141 1.00 
 Increased from low to high 9/63 0.97 (0.43–2.19) 
 Decreased from high to low 11/57 1.06 (0.51–2.20) 
 Unchanged low 26/119 1.01 (0.56–1.82) 
EI from total fat (%) 
 Unchanged high 17/139 1.00 
 Increased from low to high 11/63 1.34 (0.61–2.93) 
 Decreased from high to low 11/51 1.92 (0.87–4.23) 
 Unchanged low 20/127 2.26 (1.21–4.20) 
EI from carbohydrates (%) 
 Unchanged high 32/130 1.00 
 Increased from low to high 16/62 0.98 (0.52–1.88) 
 Decreased from high to low 9/59 0.55 (0.25–1.18) 
 Unchanged low 12/129 0.36 (0.18–0.72) 
Total sugar (g/day) 
 Unchanged high 27/127 1.00 
 Increased from low to high 9/58 0.53 (0.23–1.22) 
 Decreased from high to low 14/64 0.91 (0.42–1.94) 
 Unchanged low 19/131 0.65 (0.32–1.34) 
Average added sugar (tsp/day) 
 Unchanged high 40/223 1.00 
 Increased from low to high 10/44 0.99 (0.47–2.07) 
 Decreased from high to low 8/49 0.94 (0.41–2.20) 
 Unchanged low 11/64 0.79 (0.38–1.64) 
GL (g/day) 
 Unchanged high 18/80 1.00 
 Increased from low to high 10/43 1.05 (0.45–2.41) 
 Decreased from high to low 9/66 0.60 (0.23–1.55) 
 Unchanged low 32/191 0.54 (0.24–1.22) 
C-HEI 2005 (score) 
 Unchanged low 22/127 1.00 
 Decreased from high to low 11/42 2.39 (1.11–5.16) 
 Increased from low to high 12/59 1.67 (0.79–3.53) 
 Unchanged high 24/152 0.89 (0.47–1.67) 
DII (points) 
 Unchanged high (pro-inflammatory) 17/83 1.00 
 Increased from low to high 6/42 0.75 (0.28–1.96) 
 Decreased from high to low 11/54 1.16 (0.52–2.57) 
 Unchanged low (anti-inflammatory) 35/201 0.76 (0.39–1.47) 

aPost-diagnosis models adjusted for: age at diagnosis, cancer stage, grade, primary treatment(s), post-diagnosis BMI, family history of uterine or colorectal cancer, post-diagnosis comorbidities (Charlson Comorbidity Index) and post-diagnosis lifetime smoking pack years.

bSugar and GL models were additionally adjusted for post-diagnosis EI (kcal/day).

cMedian thresholds defined groups for EI from macronutrients and C-HEI 2005, 6 tsp/day defined added sugar groups, 0 points (neutral-inflammatory potential) defined DII groups, 100 g/day defined GL.

Stratified analysis

In BMI and physical activity groups, the relationships between lower added sugar and DFS were significantly different for participants who were inactive with ≥ 30 kg/m2 compared with active and <30 kg/m2 (Pinteraction = 0.02; Fig. 3). Modification was also seen between the active and < 30 kg/m2 group compared with the inactive and ≥30 kg/m2 group (Pinteraction = 0.04) as well as inactive and <30 kg/m2 (Pinteraction = 0.01) group in the DFS C-HEI assessments. The active and <30 kg/m2 group was not included in the assessment of the C-HEI OS associations due to insufficient events (n < 5).

Figure 3.

HRs (95% CI) for pre-diagnosis C-HEI 2005, GL, total sugar and, added sugar by obesity and lifetime physical activity status (A) with DFS (B) OS in the Alberta Endometrial Cancer Cohort Study, 2002–2022. aPre-diagnosis models adjusted for: age at diagnosis, cancer stage, grade, primary treatment(s), pre-diagnosis EI, at diagnosis BMI, family history of uterine or colorectal cancer, comorbidities (number of major comorbidities) pre-diagnosis lifetime smoking pack years, lifetime physically active. bMedian thresholds categorized physically active (≥ 101 MET-h/week/year) and physically inactive groups (<101 MET-h/week/year). cEstimates are not shown in B due to insufficient events (n <5). d*statistically significant interaction.

Figure 3.

HRs (95% CI) for pre-diagnosis C-HEI 2005, GL, total sugar and, added sugar by obesity and lifetime physical activity status (A) with DFS (B) OS in the Alberta Endometrial Cancer Cohort Study, 2002–2022. aPre-diagnosis models adjusted for: age at diagnosis, cancer stage, grade, primary treatment(s), pre-diagnosis EI, at diagnosis BMI, family history of uterine or colorectal cancer, comorbidities (number of major comorbidities) pre-diagnosis lifetime smoking pack years, lifetime physically active. bMedian thresholds categorized physically active (≥ 101 MET-h/week/year) and physically inactive groups (<101 MET-h/week/year). cEstimates are not shown in B due to insufficient events (n <5). d*statistically significant interaction.

Close modal

In this cohort, pre-diagnosis diets characterized by higher quality, lower sugar intakes and lower GL were associated with improved survival after endometrial cancer. Notably, better diet quality and limited added sugar were more important prognostic factors for survivors living with obesity and physical inactivity. Pre-diagnosis diets with greater pro-inflammatory potential were associated with worse DFS although, expected protective associations were not observed for anti-inflammatory diets. High post-diagnosis EI from carbohydrate decreased OS while maintaining carbohydrate EI < 51.8% at both time points improved OS. Moreover, post-diagnosis total- and monounsaturated-fat EI were positively associated with OS.

Diet composition

Dietary interventions generally aim to adjust cancer survivors’ macronutrient intakes to align with a prudent diet in hopes of improving outcomes. In practice, this recommendation often translates to reductions in total fat intake (6). Recently, a post-diagnosis diet intervention intending to reduce EI from total fat to 20% among breast cancer survivors reported significant decreases in cancer-related, although not total, deaths (33). Conversely, results from gynecologic cancer survivors who underwent a 12-week post-diagnosis low-carbohydrate, high-fat ketogenic diet intervention observed improved potential prognostic markers including reduced total, android and visceral adiposity, and lower fasting insulin compared with participants following a low-fat American Cancer Society diet (34). However, most evidence regarding the prognostic potential of ketogenic diets comes from animal studies and there is currently inadequate evidence of the efficacy and safety of these diets as an anticancer therapy for cancer survivors (11, 35). The specific fats targeted in dietary interventions may partially explained the conflicting aims seen among these studies. Increased polyunsaturated- and saturated-fat intakes have been associated with decreased [relative risks (RR)per5%, 0.96; 95% CI, 0.94–0.99] and increased (RRper5%, 1.04; 95% CI, 1.02–1.06) cancer-related mortality, respectively (9). To this end, meta-analyses have shown reduced OS when carbohydrates are substituted for animal fats and protein (HRpooled, 1.18; 95% CI, 1.08–1.29) but improved OS when replacement with plant sources (HRpooled, 0.82; 95% CI, 0.78–0.87; ref. 8). Our study results largely align with the prior literature, as we observed null survival associations with protein intake, reduced OS with elevated saturated fat and greater OS with higher EI from total- and monounsaturated-fat. While we did not observe the harmful associations with low EI from carbohydrate reported in past literature (HR<40%, 1.20; 95% CI, 1.09–1.32; ref. 8), these discrepancies may relate to the higher range of EI from carbohydrate in this cohort, compared with previous research.

CFG food groups were not associated with either outcome in this cohort. These results align with some, but not all, previous literature from other cancer sites. In previous studies, high pre-diagnosis vegetable and fruit intakes have been positively associated with OS in head and neck, ovarian, and breast cancer survivors (36). Yet, a review by Jochems and colleagues concluded that there is currently insufficient evidence of prognostic associations with pre- or post-diagnosis intakes of grains, protein, dairy, or oils and spreads among cancer survivors (12).

Greater total sugar and added fructose has been associated with reduced OS among women (37). In cancer populations, low sugar diets have been proposed as an adjuvant treatment based on their potential to interrupt the Warburg effect, a mechanism of cellular metabolic reprogramming leading to sustain cell proliferation, invasion, metastasis, and ultimately progression (38). Evidently, reduced cancer-specific and OS has been reported for breast cancer survivors with higher pre- and post-diagnosis sugar-sweetened beverages intakes, a major source of added sugar in North America (39, 40). Moreover, compared with breast cancer survivors with low pre- and post-diagnosis intakes of sugar-sweetened beverages, elevated cancer-related mortality has been seen with increased consumption after diagnosis (HR, 1.25; 95% CI, 1.04–1.50) and persistently high intakes (HR, 1.33; 95% CI, 1.12–1.58; ref. 39). Our results showing lower sugar intakes associated with longer survival support findings from previous research in other cancer sites (39, 40).

GL

Although GL has previously been associated with greater endometrial cancer risk (41) and there is biological plausibility of a prognostic relationship via elevated insulin pathways (41), the role of GL after endometrial cancer has not been investigated. Earlier studies from breast cancer survivors did not reveal any associations between survival and 2-year post-diagnosis GL (42), yet evidence from the National Health Service suggests that greater 1-year post-diagnosis GL was associated with increased cancer-specific (HRQ5vsQ1, 1.33; 95% CI, 1.09–1.63) and all-cause mortality (HRQ5vsQ1, 1.26; 95% CI 1.10–1.45; ref. 43). Our findings of an inverse OS association with lower GL at diagnosis but not 3-year follow-up aligns with past studies.

Dietary indices

Within cancer populations, a priori defined indices (e.g., HEI) are often used to investigate overall diet quality (6). Castro-Espin and colleagues conducted a systematic review that found better overall diet quality was associated with improved survival among breast and colorectal cancer survivors (6). Specifically, the meta-analysis showed that for breast cancer survivors higher versus lower diet quality was associated with better OS (HR, 0.77; 95% CI, 0.64–0.91; ref. 6). Improved survival has also been noted in 120 gynecologic cancer survivors with higher post-diagnosis HEI scores (HR≥70, 0.20; 95% CI, 0.10–0.43; ref. 14). In our cohort, we observed better pre-diagnosis diet quality and improved DFS, although there was evidence of nonlinearity.

Several dietary indices have been developed to assess dietary patterns on the basis of different proposed biological mechanisms (4). A review of inflammatory diets, measured with DII, showed an increased risk of cancer-related mortality (HRall sites, 1.16; 95% CI, 1.01–1.32; ref. 44), yet no cancer-specific or OS associations were seen among cancer survivors in the Australian National Endometrial Cancer study (13). Our study observed that pre-diagnosis pro-inflammatory diets were associated with shorter DFS. These results generally align with prior literature reporting cancer-specific (44) but not OS associations with inflammatory diets (13).

Impact of obesity and physical activity

Obesity and physical activity act on many of the same biological pathways as diet including insulin sensitivity, chronic inflammation, hypertension, and visceral adiposity (5, 45). Moreover, obesity and physical activity have been associated with mortality outcomes among cancer survivors including those of endometrial cancer (16, 46, 47), yet the role these lifestyle factors have in dietary survival associations remains unknown. Stratified analyses have reported positive mortality relationships with sugar-sweetened beverages and GL among breast cancer survivors with greater but not lower BMIs, although interactions were not statistically significant (40, 43). Elevated BMI has been positively associated with the overall postprandial glycemic response to food (48). Furthermore, some authors have proposed that, among individuals with obesity, the increased postprandial glycemic response coupled with obesity-related insulin resistance may lead to more substantial negative consequence of high glycemic diets (49). In this cohort, stratified analysis showed an inverse GL OS association only among participants with BMI ≥ 30 kg/m2, supporting earlier suggestions that these populations are at elevated risk from poor diet compared with healthy weight individuals (49). Moreover, healthy diets and reduced sugar intake were particularly relevant for survivors living with both obesity and low physical activity, because greater diet quality and lower added sugar consumption were associated with a 55% and 69% improvement in DFS, respectively.

Strengths and limitations

This study has several key strengths, including a substantial median 16.9-year follow-up during which qualified health record technicians collected outcome and follow-up treatment data. Participants were incident histologically confirmed endometrial cancers cases originally identified through the ACR, which has ≥ 95% case ascertainment. In addition, trained interviewers used cognitive interviewing methods and direct standardized anthropometric measurements to capture high-quality data for many potential confounders. However, social desirability and the ability to recall past-year dietary intakes may have led to under-reporting of unhealthy diets by participants, skewing the results towards the null. While measurement error in dietary assessment methods has been extensively investigated in past decades, the ability to rank individuals based on their relative intake of dietary factors using an FFQ is generally considered acceptable (4). In addition, as multiple associations were assessed in this study the risk of Type I error may have been elevated. Yet the alignment of our statistically significant results with prior research and biological plausibility gives confidence to the reported relationships. As participants in this cohort lived in Alberta, Canada and were predominantly of European ancestry, our findings may not be representative of more racially and ethnically diverse populations with different cultural diets.

Conclusion

Diet is a modifiable lifestyle factor involved in the risk or mitigation of morbidities and mortality. Among endometrial cancer survivors, limited evidence exists regarding the prognostic role of pre- and post-diagnosis diet during survivorship. This study observed that endometrial cancer survivors with better diet quality, lower sugar intakes and lower GLs at diagnosis had improved prognostic outcomes and importantly that post-diagnosis change in diet (reduced quality) negatively impact survival. Thus, health promotion efforts that highlight the importance of not only immediate healthy diet but also a healthy diet for life may be relevant for this population. Moreover, our results indicate that among endometrial cancer survivors, diet may have a larger impact for those with greater levels of obesity and low physical activity. Therefore, dietary interventions may be best targeted to inactive and obese endometrial cancer survivors. Our results support multi-focused interventions such as the Diet and Exercise in Uterine Cancer Survivors Parallel Randomized Controlled Pilot Trial, an 8-week randomized study of endometrial cancer survivors based on healthy eating and physical activity sessions (50), to improve long-term health outcomes in endometrial cancer.

No disclosures were reported.

The funders had no role in study design and conduct of the study, data collection and analysis, data interpretation, or manuscript preparation and decision to submit the manuscript for publication.

R.L. Kokts-Porietis: Conceptualization, formal analysis, visualization, writing–original draft, writing–review and editing. A.R. Morielli: Conceptualization, writing–review and editing. J. McNeil: Conceptualization, writing–review and editing. K.S. Courneya: Conceptualization, funding acquisition, investigation, methodology, project administration, writing–review and editing. L.S. Cook: Conceptualization, funding acquisition, investigation, methodology, project administration, writing–review and editing. C.M. Friedenreich: Conceptualization, resources, data curation, supervision, funding acquisition, investigation, methodology, writing–review and editing.

We would like to thank the participants and staff of the Endometrial Disease and Physical Activity Study and Alberta Endometrial Cancer Cohort Study for their contributions to the original case–control and follow-up cohort study.

C.M. Friedenreich was awarded three separate grants from the National Cancer Institute of Canada through the Canadian Cancer Society (NCIC No. 12018, NCIC No. 13010, NCIC No. 17323) and one grant from the former Alberta Cancer Board (ACB grant 22190). C.M. Friedenreich received career awards from the Canadian Institutes of Health Research and the Alberta Heritage Foundation for Medical Research/Alberta Innovates (AHFMR/Alberta Innovates). L.S. Cook and K.S. Courneya held Canada Research Chairs and L.S. Cook also received career award funding from AHFMR. L.S. Cook receives support from the US NCI (NCI 2P30CA046934–34).

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

Note: Supplementary data for this article are available at Cancer Epidemiology, Biomarkers & Prevention Online (http://cebp.aacrjournals.org/).

1.
Sung
H
,
Ferlay
J
,
Siegel
RL
,
Laversanne
M
,
Soerjomataram
I
,
Jemal
A
, et al
.
Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries
.
CA Cancer J Clin
2021
;
71
:
209
49
.
2.
Felix
AS
,
Brinton
LA
.
Cancer progress and priorities: uterine cancer
.
Cancer Epidemiol Biomarkers Prev
2018
;
27
:
985
94
.
3.
World Cancer Research Fund
/
American Institute for Cancer Research Continuous Update Project Expert Report
.
Diet, nutrition, physical activity and cancer: a global perspective
.
2018
.
Available from
: dietandcancerreport.org.
4.
Steck
SE
,
Murphy
EA
.
Dietary patterns and cancer risk
.
Nat Rev Cancer
2020
;
20
:
125
38
.
5.
Yasin
HK
,
Taylor
AH
,
Ayakannu
T
.
A narrative review of the role of diet and lifestyle factors in the development and prevention of endometrial cancer
.
Cancers
2021
;
13
:
2149
.
6.
Castro-Espin
C
,
Agudo
A
.
The role of diet in prognosis among cancer survivors: a systematic review and meta-analysis of dietary patterns and diet interventions
.
Nutrients
2022
;
14
:
348
.
7.
Keaver
L
,
Houlihan
C
,
O'Callaghan
N
,
LaVertu
AE
,
Ding
X
,
Zhang
FF
.
Evidence-based nutrition guidelines for cancer survivors in Europe: a call for action
.
Eur J Clin Nutr
2022
;
76
:
819
26
.
8.
Seidelmann
SB
,
Claggett
B
,
Cheng
S
,
Henglin
M
,
Shah
A
,
Steffen
LM
, et al
.
Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis
.
Lancet Public Health
2018
;
3
:
e419
e28
.
9.
Kim
Y
,
Je
Y
,
Giovannucci
EL
.
Association between dietary fat intake and mortality from all-causes, cardiovascular disease, and cancer: a systematic review and meta-analysis of prospective cohort studies
.
Clin Nutr
2021
;
40
:
1060
70
.
10.
Naghshi
S
,
Sadeghi
O
,
Willett
WC
,
Esmaillzadeh
A
.
Dietary intake of total, animal, and plant proteins and risk of all cause, cardiovascular, and cancer mortality: systematic review and dose–response meta-analysis of prospective cohort studies
.
BMJ
2020
;
370
:
m2412
m
.
11.
Yang
YF
,
Mattamel
PB
,
Joseph
T
,
Huang
J
,
Chen
Q
,
Akinwunmi
BO
, et al
.
Efficacy of low-carbohydrate ketogenic diet as an adjuvant cancer therapy: a systematic review and meta-analysis of randomized controlled trials
.
Nutrients
2021
;
13
:
1388
.
12.
Jochems
SHJ
,
Van Osch
FHM
,
Bryan
RT
,
Wesselius
A
,
van Schooten
FJ
,
Cheng
KK
, et al
.
Impact of dietary patterns and the main food groups on mortality and recurrence in cancer survivors: a systematic review of current epidemiological literature
.
BMJ Open
2018
;
8
:
e014530
.
13.
Nagle
CM
,
Ibiebele
T
,
Shivappa
N
,
Hébert
JR
,
Spurdle
AB
,
Webb
PM
.
Dietary inflammatory index, risk, and survival among women with endometrial cancer
.
Cancer Causes Control
2020
;
31
:
203
7
.
14.
Karavasiloglou
N
,
Pestoni
G
,
Faeh
D
,
Rohrmann
S
.
Post-diagnostic diet quality and mortality in females with self-reported history of breast or gynecological cancers: results from the Third National Health And Nutrition Examination Survey (NHANES III)
.
Nutrients
2019
;
11
:
2558
.
15.
Friedenreich
CM
,
Cook
LS
,
Magliocco
AM
,
Duggan
MA
,
Courneya
KS
.
Case–control study of lifetime total physical activity and endometrial cancer risk
.
Cancer Causes Control
.
2010
;
21
:
1105
16
.
16.
Friedenreich
CM
,
Cook
LS
,
Wang
Q
,
Kokts-Porietis
RL
,
McNeil
J
,
Ryder-Burbidge
C
, et al
.
Prospective cohort study of pre- and post-diagnosis physical activity and endometrial cancer survival
.
J Clin Oncol
2020
;
38
:
4107
17
.
17.
Csizmadi
I
,
Kahle
L
,
Ullman
R
,
Dawe
U
,
Zimmerman
TP
,
Friedenreich
CM
, et al
.
Adaptation and evaluation of the National Cancer Institute's Diet History Questionnaire and nutrient database for Canadian populations.
Public Health Nutr
2007
;
10
:
88
96
.
18.
Csizmadi
I
,
Boucher
BA
,
Lo Siou
G
,
Massarelli
I
,
Rondeau
I
,
Garriguet
D
, et al
.
Using national dietary intake data to evaluate and adapt the US Diet History Questionnaire: the stepwise tailoring of an FFQ for Canadian use
.
Public Health Nutr
2016
;
19
:
3247
55
.
19.
Friedenreich
CM
,
Courneya
KS
,
Bryant
HE
.
The lifetime total physical activity questionnaire: development and reliability
.
Med Sci Sports Exerc
1998
;
30
:
266
74
.
20.
Friedenreich
CM
,
Courneya
KS
,
Neilson
HK
,
Matthews
CE
,
Willis
G
,
Irwin
M
, et al
.
Reliability and validity of the past year total physical activity questionnaire
.
Am J Epidemiol
2006
;
163
:
959
70
.
21.
Charlson
ME
,
Pompei
P
,
Ales
KL
,
MacKenzie
CR
.
A new method of classifying prognostic comorbidity in longitudinal studies: development and validation
.
J Chronic Dis
1987
;
40
:
373
83
.
22.
American Joint Committee on Cancer AJCC
.
AJCC Cancer Staging Manual.
5th ed.
Philadelphia, PA
:
Lippincott-Raven
;
1997
.
23.
Amankwah
EK
,
Friedenreich
CM
,
Magliocco
AM
,
Brant
R
,
Courneya
KS
,
Speidel
T
, et al
.
Anthropometric measures and the risk of endometrial cancer, overall and by tumor microsatellite status and histological subtype
.
Am J Epidemiol
2013
;
177
:
1378
87
.
24.
Garriguet
D
.
Diet quality in Canada
.
Health Rep
2009
;
20
:
41
.
25.
Solbak
NM
,
Xu
JY
,
Vena
JE
,
Csizmadi
I
,
Whelan
HK
,
Robson
PJ
.
Diet quality is associated with reduced incidence of cancer and self-reported chronic disease: observations from Alberta's Tomorrow Project
.
Prev Med
2017
;
101
:
178
87
.
27.
MyPyramid Equivalents Database, 2.0 for USDA Survey Foods, 2003–2004 Food Surveys Research Group
.
Department of Agriculture
,
Beltsville, MD
2008
.
Available from
: http://www.ars.usda.gov/ba/bhnrc/fsrg.
28.
World Health Organization
.
Guideline: sugars intake for adults and children
.
Geneva, Switzerland
2015
.
29.
Lachat
C
,
Hawwash
D
,
Ocké
MC
,
Berg
C
,
Forsum
E
,
Hörnell
A
, et al
.
Strengthening the reporting of observational studies in epidemiology-nutritional epidemiology (STROBE-nut): An extension of the STROBE statement
.
PLoS Med
2016
;
13
:
e1002036
.
30.
Glycemic Index Foundation Low GI Explained 2022
.
Available from
: https://www.gisymbol.com/low-gi-explained/.
31.
Shivappa
N
,
Steck
SE
,
Hurley
TG
,
Hussey
JR
,
Hébert
JR
.
Designing and developing a literature-derived, population-based dietary inflammatory index.
Public Health Nutr
2014
;
17
:
1689
96
.
32.
Desquilbet
L
,
Mariotti
F
.
Dose–response analyses using restricted cubic spline functions in public health research
.
Stat Med
2010
;
29
:
1037
57
.
33.
Chlebowski
RT
,
Aragaki
AK
,
Anderson
GL
,
Pan
K
,
Neuhouser
ML
,
Manson
JE
, et al
.
Dietary modification and breast cancer mortality: long-term follow-up of the Women's Health Initiative randomized trial
.
J Clin Oncol
2020
;
38
:
1419
28
.
34.
Cohen
CW
,
Fontaine
KR
,
Arend
RC
,
Alvarez
RD
,
Leath
CA
III
,
Huh
WK
, et al
.
A ketogenic diet reduces central obesity and serum insulin in women with ovarian or endometrial cancer
.
Nutr J
2018
;
148
:
1253
60
.
35.
Li
J
,
Zhang
H
,
Dai
Z
.
Cancer treatment with the ketogenic diet: a systematic review and meta-analysis of animal studies
.
Front Nutr
2021
;
8
:
594408
.
36.
Hurtado-Barroso
S
,
Trius-Soler
M
,
Lamuela-Raventós
RM
,
Zamora-Ros
R
.
Vegetable and fruit consumption and prognosis among cancer survivors: a systematic review and meta-analysis of cohort studies
.
Adv Nutr
2020
;
11
:
1569
82
.
37.
Tasevska
N
,
Park
Y
,
Jiao
L
,
Hollenbeck
A
,
Subar
AF
,
Potischman
N
.
Sugars and risk of mortality in the NIH-AARP Diet and Health Study
.
Am J Clin Nutr
2014
;
99
:
1077
88
.
38.
Vaupel
P
,
Schmidberger
H
,
Mayer
A
.
The Warburg effect: essential part of metabolic reprogramming and central contributor to cancer progression
.
Int J Radiat Biol
2019
;
95
:
912
9
.
39.
Farvid
MS
,
Spence
ND
,
Rosner
BA
,
Chen
WY
,
Eliassen
AH
,
Willett
WC
, et al
.
Consumption of sugar-sweetened and artificially sweetened beverages and breast cancer survival
.
Cancer
2021
;
127
:
2762
73
.
40.
Koyratty
N
,
McCann
SE
,
Millen
AE
,
Nie
J
,
Trevisan
M
,
Freudenheim
JL
.
Sugar-sweetened soda consumption and total and breast cancer mortality: The Western New York Exposures and Breast Cancer (WEB) Study
.
Cancer Epidemiol Biomarkers Prev
2021
;
30
:
945
52
.
41.
Sadeghi
A
,
Sadeghian
M
,
Nasiri
M
,
Rahmani
J
,
Khodadost
M
,
Pirouzi
A
, et al
.
Carbohydrate quantity and quality affect the risk of endometrial cancer: a systematic review and dose–response meta-analysis
.
Clin Nutr
2020
;
39
:
1681
91
.
42.
Belle
FN
,
Kampman
E
,
McTiernan
A
,
Bernstein
L
,
Baumgartner
K
,
Baumgartner
R
, et al
.
Dietary fiber, carbohydrates, glycemic index, and glycemic load in relation to breast cancer prognosis in the HEAL cohort
.
Cancer Epidemiol Biomarkers Prev
2011
;
20
:
890
9
.
43.
Farvid
MS
,
Tamimi
RM
,
Poole
EM
,
Chen
WY
,
Rosner
BA
,
Willett
WC
, et al
.
Postdiagnostic dietary glycemic index, glycemic load, dietary insulin index, and insulin load and breast cancer survival
.
Cancer Epidemiol Biomarkers Prev
2021
;
30
:
335
43
.
44.
Zahedi
H
,
Djalalinia
S
,
Asayesh
H
,
Mansourian
M
,
Esmaeili Abdar
Z
,
Mahdavi Gorabi
A
, et al
.
A higher dietary inflammatory index score is associated with a higher risk of incidence and mortality of cancer: a comprehensive systematic review and meta-analysis
.
Int J Prev Med
2020
;
11
:
15
-.
45.
Lavie
CJ
,
Ozemek
C
,
Carbone
S
,
Katzmarzyk
PT
,
Blair
SN
.
Sedentary behavior, exercise, and cardiovascular health
.
Circ Res
2019
;
124
:
799
815
.
46.
Kokts-Porietis
RL
,
Elmrayed
S
,
Brenner
DR
,
Friedenreich
CM
.
Obesity and mortality among endometrial cancer survivors: a systematic review and meta-analysis
.
Obes Rev
2021
;
22
:
e13337
.
47.
Friedenreich
CM
,
Stone
CR
,
Cheung
WY
,
Hayes
SC
.
Physical activity and mortality in cancer survivors: a systematic review and meta-analysis
.
JNCI Cancer Spectr
2020
;
4
:
pkz080
.
48.
Carroll
JF
,
Kaiser
KA
,
Franks
SF
,
Deere
C
,
Caffrey
JL
.
Influence of BMI and gender on postprandial hormone responses
.
Obesity
2007
;
15
:
2974
83
.
49.
Jenkins
DJA
,
Dehghan
M
,
Mente
A
,
Bangdiwala
SI
,
Rangarajan
S
,
Srichaikul
K
, et al
.
Glycemic index, glycemic load, and cardiovascular disease and mortality
.
N Engl J Med
2021
;
384
:
1312
22
.
50.
Koutoukidis
DA
,
Beeken
RJ
,
Manchanda
R
,
Burnell
M
,
Ziauddeen
N
,
Michalopoulou
M
, et al
.
Diet, physical activity, and health-related outcomes of endometrial cancer survivors in a behavioral lifestyle program: the Diet and Exercise in Uterine Cancer Survivors (DEUS) parallel randomized controlled pilot trial
.
Int J Gynecol Cancer
2019
;
29
:
531
40
.