Abstract
Branched chain amino acids (BCAAs) are essential amino acids, and emerging evidence suggests that BCAAs may mediate pathways related to cancer progression, possibly due to their involvement in insulin metabolism.
We investigated the association between dietary intake of BCAAs with colorectal cancer (CRC) risk in three prospective cohorts: the Nurses’ Health Study I [(NHS), number of participants (n) at baseline =77,017], NHS II (n=92,984), and the Health Professionals Follow-up Study [(HPFS) n=47,255]. Validated food frequency questionnaires were administered every four years and follow-up questionnaires on lifestyle biennially. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using Cox proportional hazards regression models. Pooled HRs were obtained using random effect models.
After up to 28 years of follow-up, 1660 cases were observed in NHS, 306 in NHS II and 1343 in HPFS. In multivariable adjusted models, we observed a weak inverse association between BCAA intake and CRC [highest vs. lowest quintile, pooled HR including all three cohorts (95% CI): 0.89 (0.80–1.00), Ptrend=0.06, HR per standard deviation (SD) increment 0.95 (0.92–0.99)]. However, after including dairy calcium to the models, BCAA intake was no longer associated with risk of CRC [HR 0.96 (0.85–1.08), Ptrend=0.50, HR per SD increment 0.97 (0.93–1.01)].
We did not find evidence that higher dietary BCAA intake is associated with higher risk of CRC. As this is the first prospective study to examine the association between BCAA intake and CRC, our findings warrant investigation in other cohorts.
Introduction
Colorectal cancer (CRC) is the third most common cancer and the third leading cause of death from cancer in the US in both men and women [1].
Branched chain amino acids (BCAAs) namely leucine, isoleucine, and valine are essential amino acids, and emerging evidence suggests that BCAAs may mediate pathways related to cancer development and progression, possibly due to their involvement in insulin metabolism [2–4]. Some prospective studies have reported positive associations between plasma BCAA levels and risk of T2D [5,6], CVD [7–9], and pancreatic cancer [10] which may share etiological pathways (e.g., insulin resistance) with CRC [11].
Only few cross-sectional studies have examined the association between plasma BCAA levels and colorectal neoplasia, but did not provide support for a positive association between BCAAs and risk of colorectal neoplasia [12,13]. In one cross-sectional study from Japan, total plasma BCAA levels were inversely associated with risk of colorectal adenoma (a precursor for CRC [14]) in men, but not in women [12].
In several studies, albeit not all [15], dietary intake was weakly correlated with plasma BCAA levels [16,17], thus the contribution of dietary intake on plasma levels remains controversial. To our knowledge, no prospective study has examined the association between dietary BCAA intake and risk of CRC.
To test whether higher dietary intake of BCAA is associated with higher risk of CRC, we conducted a prospective analysis using data from three large prospective cohorts of US men and women, the Nurses’ Health Study, the Nurses’ Health Study II and the Health Professionals Follow-up Study.
Methods
Study population
The Nurses’ Health Study (NHS) began in 1976 in the US when 121, 700 registered female nurses aged 30 to 55 were recruited [18]. The NHS II was established in 1989 and included 116, 430 female registered nurses between the ages of 25 and 42 years. The Health Professionals Follow-up Study (HPFS) enrolled 51,529 male health professionals aged 40 to 75 years in 1986 [19]. Questionnaires were administered at baseline and follow-up questionnaires were administered biennially to collect updated information on lifestyle and medical history. The cumulative follow-up rates in these cohorts are over 90%.
The study protocol was approved by the institutional review boards of the Brigham and Women’s Hospital and Harvard T.H. Chan School of Public Health, and those of participating registries as required.
Assessment of dietary intake
Dietary intake was assessed repeatedly with validated semi-quantitative food frequency questionnaires (FFQs) every 4 years, starting in 1980 for NHS, 1989 for NHS II and 1986 for HPFS. Participants answered how often they, on average, consumed a specified amount of food during the previous year. The contribution of each food to nutrient intake was calculated by multiplying the frequency of consumption of each food by its nutrient content and summing up the nutrient intakes from all foods. These nutrient intakes were energy adjusted using the residual method [20]. Total BCAA intake (from now on referred to as “BCAA intake”) was calculated by summing intakes of valine, leucine and isoleucine. The reproducibility and validity of FFQs to estimate protein intake was examined previously [21–23]. In brief, de-attenuated Spearman correlation coefficients between FFQ and dietary records were 0.56 for animal and 0.66 for vegetable protein [24]. While correlations are not available for BCAA intake per se, total BCAA intake was highly correlated with animal protein intake (all partial Spearman correlation coefficients adjusted for age were >0.90), but not with vegetable protein intake (correlations ranging between −0.1 to 0.1 in all three cohorts). The individual branch chain amino acids valine, leucine and isoleucine were also highly correlated with each other (all partial Spearman correlation coefficients adjusted for age were 0.99).
For this analysis, follow-up started in 1984 for NHS and 1991 for NHS II, the first time a more detailed 131-item FFQ was administered in these cohorts. HPFS participants were mailed the detailed FFQ at baseline, thus follow-up for HPFS started in 1986. We excluded participants who left a high number of items blank (>=70) on the FFQs, those who reported unusually low or high energy intake, i.e., <800 or >4200 kcal per day for men and <600 or >3500 kcal per day for women, and those who had history of cancer (except non-melanoma skin cancer) or ulcerative colitis at baseline. After these exclusions, 77,017 women from NHS, 92,984 women from NHS II and 47,255 men were eligible for our analysis.
Assessment of other covariables
We collected information on age, body weight and height, smoking, physical activity, endoscopic screening, family history of CRC, use of multivitamin supplements, aspirin and nonsteroidal anti-inflammatory drugs (NSAIDs), and menopausal status and postmenopausal hormone use (for women). Body mass index (BMI) was calculated as body weight (kg) divided by height squared (m2).
Ascertainment of colorectal cancer cases
When participants reported a new diagnosis of CRC on their biennial questionnaires, we asked the participant or the next-of-kin for permission to obtain medical records and pathological reports. Study physicians blinded to exposure status reviewed medical records to confirm a diagnosis of CRC and extract information on stage, location and histology. Proximal colon cancers were defined as cancers located in the cecum, ascending and transverse colon, distal cancers as those in the descending and sigmoid colon, and rectal cancers as those located in the recto-sigmoid and rectum. Information on deaths was gathered from the National Death Index, state vital statistics records, the postal system, and the report from next-of-kin, [25,26] and cause of deaths were confirmed by study physicians who reviewed death certificates and medical records. For non-responders who died of CRC, we also contacted the next-of-kin to obtain permission to review medical records and confirm a diagnosis of CRC as described above.
Statistical analysis
Person-time was calculated from the date of return of the baseline questionnaire to the date of diagnosis of CRC, the date of death or end of follow-up, whichever came first. End of follow-up was May 31st, 2012 for NHS, May 31st 2013 for NHS II and January 31st, 2012 for HPFS. We calculated cumulative updated average of dietary intakes and lifestyle factors such as BMI or physical activity to better represent long-term patterns and reduce within-person variation. Cumulative updated exposure was estimated by calculating the average exposure from all available follow-up questionnaires up to the most recent follow-up questionnaire cycle [27]. In addition to cumulative updated intake, we also examined associations for BCAA intake at baseline and after a 0–4 year lag (simple updated intake) which represents the intake obtained from the most recent FFQ prior to each follow-up cycle [27]. Cox proportional hazard models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CI) for the association between BCAA intake (quintile) and CRC risk. We stratified jointly by age (in months) and 2-year questionnaire cycle to control finely for confounding by calendar time and age. In multivariable models, we additionally adjusted for other known and suspected risk factor for CRC (covariables were selected based on published data including previous findings from our cohorts [28–30]) including BMI (< 23, 23-<25, 25-<27, 27-<30, 30-<35, ≥35 kg/m2), smoking status (never, past, current smoking with 1–14 cigarettes per day and with ≥15 cigarettes per day), alcohol intake (0 to <5, 5 to <10, 10 to <15, or ≥15 g/day), physical activity (quintile), history of CRC in a first degree relative (yes or no), screening endoscopy (yes or no), multivitamin use (yes or no), regular aspirin or NSAIDs use (≥2 tablets/week vs. <2 tablets/week). For women, we also adjusted for menopausal status and use of postmenopausal hormone (premenopausal, never, past, current). Analysis was first conducted for each cohort separately and then pooled using random-effects model [31]. P for trend was calculated by including the median value of each quintile category as a continuous exposure in the models and utilizing the Wald test to assess statistical significance.
To test for potential modification by other CRC risk factors including BMI, smoking, physical activity, alcohol intake, animal/vegetable protein ratio and high vs. low fat intake, we included cross-product terms for exposure variables in the multivariable adjusted model. All analyses were performed using the SAS software (version 9.4, SAS Institute Inc.). All statistical tests were two sided and a p-value <0.05 was considered statistically significant.
Results
During up to 28 years of follow-up, 1660 CRC cases were documented in NHS, 306 cases in NHS II, and 1343 cases in HPFS. The age-standardized distribution of participants’ characteristics weighted by person-years according to BCAA intake in quintiles are shown in Table 1. In men, participants with higher BCAA intake were more likely to have higher BMI, but did not appear to differ appreciably with regard to physical activity. In women, those with higher intake were more likely to be physically active and had higher BMI. In all three cohorts, participants with higher BCAA intake were more likely to have higher intake of dairy products, protein, dietary fiber, folate, vitamin D and calcium and less likely to be current smokers.
Table 1.
Age-standardized* characteristics of participants according to quintiles of branched chain amino acid (BCAA) intake in Nurses’ Health Study (NHS), Nurses’ Health Study II (NHS II) and Health Professionals Follow-up Study (HPFS)
| NHS | HPFS | NHSII | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Q1 | Q3 | Q5 | Q1 | Q3 | Q5 | Q1 | Q3 | Q5 | |
| BCAA intake (g/d) | |||||||||
| Total BCAA | 10.1 (0.9) | 12.7 (0.3) | 15.6 (1.2) | 12.4 (1.1) | 15.6 (0.3) | 19.5 (1.8) | 11.6 (1.2) | 14.9 (0.4) | 18.4 (1.6) |
| Valine | 3.0 (0.3) | 3.7 (0.1) | 4.5 (0.4) | 3.6 (0.3) | 4.6 (0.1) | 5.7 (0.5) | 3.4 (0.3) | 4.4 (0.1) | 5.4 (0.5) |
| Leucine | 4.5 (0.4) | 5.6 (0.1) | 6.9 (0.5) | 5.5 (0.5) | 6.9 (0.2) | 8.6 (0.8) | 5.1 (0.5) | 6.6 (0.2) | 8.1 (0.7) |
| Isoleucine | 2.7 (0.3) | 3.4 (0.1) | 4.2 (0.4) | 3.3 (0.3) | 4.1 (0.1) | 5.2 (0.5) | 3.1 (0.3) | 3.9 (0.13) | 4.9 (0.5) |
| Age, (year) | 63.0 (10.9) | 62.3 (10.4) | 62.6 (10.0) | 64.0 (11.6) | 64.0 (11.1) | 62.6 (10.8) | 45.2 (8.4) | 45.2 (8.3) | 45.4 (8.3) |
| Body Mass Index (kg/m2) | 24.6 (4.4) | 25.6 (4.7) | 27.0 (5.1) | 25.3 (3.2) | 25.8 (3.3) | 26.4 (3.9) | 24.5 (5.3) | 25.5 (5.5) | 26.8 (6.0) |
| Alcohol drinkers (%) | 74 | 78 | 73 | 82 | 84 | 77 | 62 | 67 | 76 |
| Current smoker (%) | 18 | 13 | 11 | 7 | 5 | 5 | 12 | 9 | 9 |
| Family history of colorectal cancer (%) | 16 | 17 | 16 | 14 | 14 | 13 | 11 | 11 | 11 |
| History of endoscopy (%) | 18 | 21 | 21 | 23 | 25 | 23 | 12 | 13 | 12 |
| Regular aspirin or NSAID use (%) | 40 | 42 | 43 | 45 | 49 | 44 | 40 | 42 | 40 |
| Physical activity, MET-hours/wk | 14.7 (17.1) | 16.1 (16.6) | 17.8 (18.9) | 25.3 (24.0) | 26.4 (22.5) | 24.8 (23.3) | 21.4 (25.7) | 21.4 (23.2) | 23.5 (25.9) |
| Multivitamin use (%) | 50 | 55 | 58 | 45 | 48 | 45 | 29 | 32 | 32 |
| Postmenopausal (%) | 85 | 85 | 85 | - | - | - | 26 | 26 | 26 |
| Current PMH use (%) | 27 | 31 | 31 | - | - | - | 18 | 18 | 17 |
| Dietary intake | |||||||||
| Total energy intake (kcal/d) | 1733 (482) | 1766 (451) | 1696 (439) | 1963 (560) | 2003 (543) | 1927 (567) | 1799 (528) | 1826 (495) | 1761 (486) |
| Total protein (g/d) | 58.8 (5.3) | 72.5 (2.2) | 87.7 (6.9) | 72.6 (6.4) | 89.9 (2.6) | 110.5 (10.1) | 67.1 (6.6) | 84.5 (2.8) | 102.9 (9.0) |
| Animal protein (g/d) | 38.6 (6.2) | 51.9 (3.7) | 67.5 (8.2) | 46.3 (8.7) | 63.5 (5.2) | 85.2 (11.9) | 43.3 (9.0) | 61.0 (4.9) | 80.3 (10.6) |
| Plant protein (g/d) | 20.3 (4.4) | 20.6 (3.5) | 20.2 (3.8) | 26.3 (6.8) | 26.4 (5.3) | 25.4 (5.8) | 23.7 (6.5) | 23.3 (4.4) | 22.4 (4.5) |
| Dietary fiber (g/d) | 16.7 (4.9) | 17.8 (4.3) | 18.6 (4.7) | 21.4 (7.3) | 22.2 (6.1) | 22.5 (6.8) | 18.4 (6.2) | 18.8 (4.7) | 19.0 (5.0) |
| Folate (μg/d) | 410 (197) | 450 (191) | 495 (207) | 581 (324) | 617 (320) | 617 (343) | 445 (258) | 468 (244) | 504 (268) |
| Vitamin D (IU/d) | 290 (197) | 360 (199) | 456 (228) | 346 (240) | 425 (243) | 541 (309) | 317 (216) | 385 (208) | 481 (247) |
| Calcium (mg/d) | 866 (372) | 1032 (375) | 1230 (436) | 795 (318) | 928 (333) | 1081 (449) | 915 (384) | 1079 (383) | 1261 (451) |
| Unprocessed red meat (svg/d) | 0.5 (0.2) | 0.6 (0.3) | 0.6 (0.4) | 0.5 (0.3) | 0.6 (0.4) | 0.6 (0.5) | 0.4 (0.3) | 0.6 (0.3) | 0.6 (0.4) |
| Processed red meat (svg/d) | 0.3 (0.3) | 0.3 (0.2) | 0.2 (0.2) | 0.3 (0.4) | 0.3 (0.3) | 0.3 (0.3) | 0.2 (0.2) | 0.2 (0.2) | 0.2 (0.2) |
| Fish (svg/d) | 0.2 (0.1) | 0.3 (0.2) | 0.4 (0.3) | 0.2 (0.2) | 0.3 (0.2) | 0.5 (0.4) | 0.1 (0.1) | 0.2 (0.2) | 0.3 (0.3) |
| Chicken and Poultry (svg/d) | 0.2 (0.1) | 0.3 (0.2) | 0.5 (0.3) | 0.2 (0.2) | 0.4 (0.2) | 0.6 (0.4) | 0.9 (0.8) | 1.2 (0.9) | 1.5 (0.9) |
| Dairy products (svg/d) | 2.1 (1.4) | 2.4 (1.3) | 2.7 (1.4) | 1.8 (1.3) | 2.2 (1.3) | 2.5 (1.6) | 1.6 (1.1) | 2.1 (1.2) | 2.4 (1.4) |
Weighted by person years and age-standardized except for age.
Continuous variables shows as mean (standard deviation).
Abbreviations: BCAA: branched chain amino acid, NHS: Nurses’ Health Study, NHS II: Nurses’ Health Study II, HPFS: Health Professionals Follow-up Study NSAIDs: nonsteroidal anti-inflammatory drugs, PMH: postmenopausal hormone
Regular aspirin use was defined as use of 2 or more tablets per week. Dietary intake was energy-adjusted.
After pooling results from all three cohorts, we observed a borderline statistically significant inverse association between BCAA intake and CRC risk [Model 1: Q5 vs. Q1, HR (95% CI): 0.89 (0.80, 1.00), Ptrend=0.06; HR per SD increment (SD=2.59 g/d in HPFS, 1.98 g/d in NHS and 2.47 g/d in NHS II): 0.95 (0.92, 0.99)] (Table 2).
Table 2.
Hazard ratio (95% CI) of colorectal cancer according to quintiles of cumulative average intake of BCAAs in NHS, NHSII and HPFS
| Quintile category of BCAA
intake |
||||||||
|---|---|---|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | Q5 | P for trend | Per SD | P-value for perSD | |
| HPFS | ||||||||
| Cases (n) | 287 | 299 | 264 | 248 | 245 | |||
| Age-adjusted HR | 1.00 | 1.06 (0.90, 1.25) | 0.95 (0.80, 1.12) | 0.88 (0.74, 1.04) | 0.88 (0.74, 1.04) | 0.02 | 0.92 (0.87, 0.97) | 0.004 |
| Model 1 HR | 1.00 | 1.08 (0.92, 1.27) | 0.99 (0.84,1.18) | 0.93 (0.78, 1.11) | 0.93 (0.78, 1.11) | 0.17 | 0.94 (0.89, 1.00) | 0.04 |
| Model 2 HR | 1.00 | 1.10 (0.93, 1.30) | 1.02 (0.85, 1.21) | 0.95 (0.80, 1.14) | 0.96 (0.80, 1.14) | 0.34 | 0.95 (0.90,1.01) | 0.08 |
| NHS | ||||||||
| Cases (n) | 355 | 327 | 330 | 357 | 291 | |||
| Age-adjusted HR | 1.00 | 0.92 (0.79, 1.07) | 0.93 (0.80, 1.08) | 1.00 (0.87, 1.16) | 0.83 (0.71, 0.98) | 0.10 | 0.94 (0.90, 0.99) | 0.02 |
| Model 1 HR | 1.00 | 0.93 (0.80, 1.08) | 0.95 (0.81, 1.10) | 1.03 (0.89, 1.20) | 0.86 (0.74, 1.01) | 0.25 | 0.95 (0.90, 1.00) | 0.06 |
| Model 2 HR | 1.00 | 0.95 (0.81, 1.11) | 0.98 (0.84, 1.15) | 1.09 (0.93, 1.27) | 0.94 (0.79, 1.11) | 0.91 | 0.98 (0.92, 1.03) | 0.40 |
| NHSII | ||||||||
| Cases (n) | 66 | 60 | 71 | 55 | 54 | |||
| Age-adjusted HR | 1.00 | 0.87 (0.61, 1.24) | 1.05 (0.54, 1.24) | 0.84 (0.59, 1.20) | 0.94 (0.66, 1.35) | 0.70 | 1.00 (0.89, 1.13) | 0.94 |
| Model 1 HR | 1.00 | 0.87 (0.61, 1.23) | 1.04 (0.74, 1.46) | 0.82 (0.57, 1.18) | 0.90 (0.62, 1.30) | 0.52 | 0.99 (0.88, 1.12) | 0.87 |
| Model 2 HR | 1.00 | 0.89 (0.63,1.27) | 1.10 (0.78,1.55) | 0.90 (0.62,1.30) | 1.03 (0.70,1.51) | 0.90 | 1.05 (0.93, 1.19) | 0.44 |
| Pooled NHS, NHS II and HPFS | ||||||||
| Model 1 HR | 1.00 | 0.98 (0.87, 1.10) | 0.97 (0.87, 1.08) | 0.97 (0.87, 1.08) | 0.89 (0.80, 1.00) | 0.06 | 0.95 (0.92, 0.99) | 0.01 |
| Model 2 HR | 1.00 | 1.01 (0.90, 1.12) | 1.01 (0.90, 1.13) | 1.02 (0.91, 1.14) | 0.96 (0.85, 1.08) | 0.50 | 0.97 (0.93, 1.01) | 0.15 |
Abbreviations HR: hazard ratio CI: confidence interval SD: standard deviation BCAA: branched chain amino acid, NHS: Nurses’ Health Study, NHS II: Nurses’ Health Study II, HPFS: Health Professionals Follow-up Study
Model 1: Adjusted for age, smoking status (never, past, current <15 pack-years, current ≥15 pack-years), history of colorectal cancer in a parent or sibling (yes or no), history of endoscopy (yes or no), regular aspirin use (≥2 tablets/week, yes or no), menopausal status and postmenopausal hormone use (premenopause, never, past, current, in women only), body mass index (<23, 23 to <25, 25 to <27, 27 to <30, ≥30 kg/m2), physical activity (quintile), alcohol consumption (0 to <5, 5 to <10, 10 to <15, or ≥15 g/day).
Model 2: Further adjusted for dairy calcium intake (quintile).
P for trend was calculated including the median value of each quintile category as continuous variable to the models. Standard deviation (SD) was 2.59 g/d in HPFS, 1.98 g/d in NHS and 2.47 g/d in NHS II.
P for heterogeneity among studies was not significant in Model 1 and Model 2 (0.75 in Model 1 and 0.36 in Model 2).
To investigate whether observed associations may be explained by higher intake of major food sources of BCAA or by diabetes status, we also examined associations for CRC after including major food sources of BCAAs, i.e. dairy, red and processed meat, chicken and poultry and fish and other dietary factors such as intake of fiber, vitamin D and folate, separately (one-by-one) to the multivariable models. For the most part, risk estimates were similar when we adjusted for BCAA food sources and diabetes (Supplemental Table 1). However, after including total or dairy calcium to the models, BCAA intake was no longer significantly associated with risk of CRC (Model 2, adjusted for dairy calcium intake; Table 2).
When we examined associations separately by CRC location (proximal, distal and rectal cancers; Supplemental Table 2), BCAA intake was not associated with risk of proximal cancers. In multivariable models, we observed evidence for an inverse association between BCAA intake and distal colon (HPFS) and rectal cancers (NHS). Inverse associations between BCAA intake and left-sided cancers were slightly attenuated, but persisted after adding red and processed meat, chicken or turkey, or fish intake one-by-one to the models (Supplemental Table 3). After including dairy calcium intake to the multivariable models, inverse associations for distal colon cancer [HR per SD: 0.87 (0.78, 0.97)] in HPFS and for rectal cancers in NHS [0.89 (0.79, 1.00)] remained, but results for NHS did not reach statistical significance (Supplemental Table 2).
When we studied associations for BCAA intake at baseline or after a 0–4 year lag in the NHS and HPFS (for the 0–4 year lag analysis NHS II was excluded due to limited sample size), results were similar [baseline: Q5 vs Q1 pooled HR 0.95 (0.81–1.12), Ptrend =0.51, simple updated: pooled HR 0.97 (0.85–1.11), Ptrend =0.68]. Furthermore, we did not observe evidence for a statistically significant interaction with BMI, smoking status, alcohol intake or physical activity, high vs. low fat intake or animal protein/vegetable protein ratio. (Supplemental Table 4).
Discussion
In this large study using data from three prospective cohorts of US health professionals, we did not find evidence for a positive association between BCAA intake and risk of CRC. Although in multivariable adjusted models, we observed a weak inverse association between BCAA intake and CRC; after including dairy calcium intake to the models, BCAA intake was no longer associated with risk of CRC. Furthermore, in multivariable models that included dairy calcium, we observed an inverse association for distal colon cancers in HPFS and for rectal cancers in NHS.
BCAAs are important building blocks for proteins and function as nutrient signals mediating metabolic pathways, including those related to insulin and lipid metabolism [2,3]. Obesity and insulin resistance are established risk factors for CRC [11,28] and findings from both animal and human intervention studies suggest that higher intake or supplementation of leucine may have a favorable effect on glucose homeostasis, insulin metabolism and satiety [32–34]. In obese mice with hyperinsulinemia, BCAA supplementation was associated with fewer premalignant lesions of the colon (aberrant crypt foci) when compared with normal diet-fed mice. BCAA-supplemented mice also exhibited lower expression of insulin like growth factor-I (IGF-I) receptor on the colonic mucosa and decreased serum insulin and IGF-I levels [35]. Additionally, in both colon and hepatic cancer cell lines, higher exposure to BCAA was associated with lower insulin-induced proliferation [36].
Contrary to the mechanistic intake studies, several plasma based prospective studies have recently reported positive associations between plasma BCAA levels and risk of T2D [5,6] and CVD [7,8], which may share etiological pathways (e.g., insulin resistance) with CRC [11]. Similarly, in a prospective study which included participants from several US cohorts including HPFS and NHS, plasma BCAA levels were positively associated with risk of pancreatic cancer [10].
Epidemiological studies regarding the association between BCAAs and colorectal neoplasia have been limited to few retrospective plasma studies. Consistent with our findings on diet, these studies did not support that higher plasma levels of BCAA increase risk of colorectal neoplasia. For example, in a cross-sectional study based in Japan including 629 cases and 584 controls, total plasma BCAA levels were inversely associated with risk of colorectal adenoma [Q4 vs Q1, OR(95% CI): 0.68 (0.48–0.98), P trend 0.10)]. However, after stratification by sex, inverse associations were observed in men, but not in women [total BCAA, Q4 vs Q1, OR(95% CI): men 0.58 (0.37–0.93); women: 0.84 (0.45–1.57)] [12]. However, in that study, P for interaction by sex was not statistically significant (P interaction 0.55). Inverse associations also appeared to be most pronounced for leucine and restricted to colon (proximal and distal) adenomas, but the number of rectal adenoma cases was very small (19 cases in the highest quartile of BCAA concentration) and blood samples were collected after diagnosis, thus results may have been biased [12]. In another retrospective study which used metabolomics data from 64 CRC cases and 65 controls, results differed depending on the laboratory method used to identify metabolic profiles. For example, valine and leucine serum levels were lower in cases when compared to controls using the GC-TOMS analysis method, but did not appear to differ between cases and control when using the UPLC-QTOFMS method [13]. However, due to the retrospective design and the potential for reverse causation, these results need to be interpreted with caution. In addition, these two studies did not adjust for calcium or dairy calcium intake.
The reason for the differential associations observed in our study compared to previous prospective studies on plasma BCAAs and metabolic diseases, known to share pathways with CRC is unclear. However, in several studies, dietary intake was weakly correlated with plasma BCAA levels [16,17], which may at least in part explain these differential associations. Moreover, it is unclear whether higher plasma BCAA levels are causally related to risk of chronic diseases or are a marker for alterations in BCAA metabolism related to insulin resistance [2]. We did not examine plasma BCAA levels, thus we were not able to assess the proportion of variability in plasma BCAA levels explained by dietary intake in our cohorts.
Findings from our study did not suggest that higher BCAA intake increases risk of CRC. Specifically, after adjusting for total or dairy calcium, BCAA intake was no longer associated with risk of CRC, suggesting that calcium in dairy products may have likely contributed to the initially observed inverse association between BCAA and CRC risk. In meta-analyses and in our cohorts, higher dairy or calcium intake has been associated with lower risk of colorectal neoplasia [28–30, 37–40].
Red meat and poultry (−30%), dairy products (−15%) and fish (−8%) are main contributors of total BCAA intake in our cohorts. Red meat is a known risk factor for CRC [41], while dairy products and fish are generally associated with lower risk of CRC [42]. Dietary BCAA intake was highly correlated with animal protein intake therefore it was difficult to distinguish the effect of BCAAs per se from those of animal protein or its main food sources. It is unclear why BCAA intake was inversely associated with distal colon cancer in HPFS and rectal cancers in NHS. In our study, observed associations between BCAA intake and total or left sided cancers were slightly attenuated, but persisted after adding red and processed meat, chicken or turkey or fish intake separately (one-by-one) to the models. Thus, considering the collinearity between BCAAs and its major food sources, we cannot exclude that observed associations may be due to higher intake of other components in BCAA-rich foods.
Our study has other limitations which warrant further discussion. First, misclassification in estimation of BCAA intake cannot be excluded. However, misclassification of exposure generally leads to a bias towards the null association and we calculated cumulative average of dietary BCAA intake to reduce random error and better estimate long-term exposure. Second, we cannot exclude the possibility that residual confounding may have affected our results, however we have collected detailed data on numerous diet and lifestyle factors in our cohorts, which made it possible to assess and adjust for a wide range of potential confounders in our analysis. Third, the participants in the NHS and HPFS cohorts represent a more health conscious population than the general US population, which may limit generalizability to other populations. However, main sources for BCAA are dietary factors that can both increase (e.g., red meat) or decrease (e.g., fish) risk for CRC and as such can be part of “healthy dietary pattern” as well as an “unhealthy dietary pattern” and the relative homogeneity of our cohort participants may help to reduce residual confounding. Fourth, we cannot entirely exclude reverse causation, but when we examined associations for BCAA intake at baseline or after a 0–4 year lag results were similar.
One major strength of our study is the long follow-up and its prospective design. Repeated measurements of dietary exposure reduce random error and enable us to better estimate long-term exposure. Our large sample size also allowed us to perform stratified analyses by lifestyle characteristics and tumor location with sufficient power.
In conclusion, findings from this large prospective study do not support that higher BCAA intake, at least in the context of a US diet, is associated with higher risk of CRC. As this the first prospective study to examine this association, confirmation in other cohorts is warranted.
Supplementary Material
Acknowledgements
We would like to thank the participants and staff of the Nurses’ Health Study, the Nurses’ Health Study 2 and the Health Professionals Follow-up Study for their valuable contributions as well as the following state cancer registries for their help: AL, AZ, AR, CA, CO, CT, DE, FL, GA, ID, IL, IN, IA, KY, LA, ME, MD, MA, MI, NE, NH, NJ, NY, NC, ND, OH, OK, OR, PA, RI, SC, TN, TX, VA, WA, WY. The authors assume full responsibility for analyses and interpretation of these data.
Grant Support: This work was supported by U.S. National Institutes of Health (NIH) grants (P01 CA87969 to M.J. Stampfer; UM1 CA186107 to M.J. Stampfer; UM1 CA176726 to W.C.W.; P01 CA55075 to W.C.W.; UM1 CA167552 to W.C.W.; U01 CA167552 to W.C.W. and L.A. Mucci; P50 CA127003 to C.S.F.; R01 CA137178 to A.T.C.; K24 DK098311 to A.T.C.; R35 CA197735 to S.O.; R01 CA151993 to S.O.; R21 CA222940 to K.W. and R.N., R21 CA230873 to K.W. and S.O., K07 CA188126 to X.Z., and R00 CA207736 to F.K.T. Nodal Award from the Dana-Farber Harvard Cancer Center (to S.O.), and grants from the Project P Fund, The Friends of the Dana-Farber Cancer Institute, Bennett Family Fund, and the Entertainment Industry Foundation through National Colorectal Cancer Research Alliance. Research supported by a Stand Up To Cancer Colorectal Cancer Dream Team Translational Research Grant (Grant Number: SU2C-AACR-DT22-17). Stand Up To Cancer is a division of the Entertainment Industry Foundation. Research grants are administered by the American Association for Cancer Research, the scientific partner of SU2C (to C.S.F.). This work was also in part supported by an Investigator Initiated Grants from the American Institute for Cancer Research (AICR) to KW. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, and approval of the manuscript; and decision to submit the manuscript for publication.
Disclosure: JAM received institutional research funding from Boston Biomedical and has served as an advisor/consultant to Ignyta, Taiho, and Cota; ATC has received consulting fees from Bayer Pharma AG, Pfizer Inc, and Janssen Inc for work unrelated to this manuscript. CSF reports consulting role for Agios, Bain Capital, Bayer, Celgene, Dicerna, Five Prime Therapeutics, Gilead Sciences, Eli Lilly, Entrinsic Health, Genentech, KEW, Merck, Merrimack Pharmaceuticals, Pfizer, Sanofi, Taiho, and Unum Therapeutics. He also serves as a Director for CytomX Therapeutics and owns unexercised stock options for CytomX and Entrinsic Health. This study was not funded by any of these commercial entities. The remaining authors have no conflicts of interest to disclose.
References
- 1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA: A Cancer Journal for Clinicians 2018;68:7–30. [DOI] [PubMed] [Google Scholar]
- 2.Lynch CJ, Adams SH. Branched-chain amino acids in metabolic signalling and insulin resistance Nat Rev Endocrinol 2014;10:723–736. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nie C, He T, Zhang W, Zhang G, Ma X. Branched Chain Amino Acids: Beyond Nutrition Metabolism. Int J Mol Sci 2018;19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ananieva EA, Wilkinson AC. Branched-chain amino acid metabolism in cancer. Curr Opin Clin Nutr Metab Care 2018;21:64–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wang TJ, Larson MG, Vasan RS, et al. Metabolite profiles and the risk of developing diabetes. Nat Med 2011;17:448–453. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Guasch-Ferré M, Hruby A, Toledo E et al. Metabolomics in Prediabetes and Diabetes: A Systematic Review and Meta-analysis. Diabetes Care. 2016;39:833–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tobias DK, Lawler PR, Harada PH, et al. Circulating Branched-Chain Amino Acids and Incident Cardiovascular Disease in a Prospective Cohort of US Women. Circ Genom Precis Med 2018;11:e002157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ruiz-Canela M, Toledo E, Clish CB, et al. Plasma branched-chain amino acids and incident cardiovascular disease in the PREDIMED trial. Clin Chem 2016;62:582–592. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tobias DK, Mora S, Verma S, Lawler PR. Altered branched chain amino acid metabolism: toward a unifying cardiometabolic hypothesis. Curr Opin Cardiol 2018;33:558–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Mayers JR, Wu C, Clish CB, et al. Elevation of circulating branched-chain amino acids is an early event in human pancreatic adenocarcinoma development. Nat Med 2014;20:1193–1198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Giovannucci E Metabolic syndrome, hyperinsulinemia, and colon cancer: a review. Am J Clin Nutr 2007;86:s836–842. [DOI] [PubMed] [Google Scholar]
- 12.Budhathoki S, Iwasaki M, Yamaji T, Yamamoto H, Kato Y, Tsugane S. Association of plasma concentrations of branched-chain amino acids with risk of colorectal adenoma in a large Japanese population. Ann Oncol 2017;28:818–823. [DOI] [PubMed] [Google Scholar]
- 13.Qiu Y, Cai G, Su M, et al. Serum Metabolite Profiling of Human Colorectal Cancer Using GC−TOFMS and UPLC−QTOFMS. J Proteome Res 2009;8:4844–4850. [DOI] [PubMed] [Google Scholar]
- 14.Leslie A, Carey FA, Pratt NR, Steele RJC. The colorectal adenoma-carcinoma sequence. Br J Surg 2002;89:845–860. [DOI] [PubMed] [Google Scholar]
- 15.Rietman A, Schwarz J, Tomé D, Kok FJ, Mensink M. High dietary protein intake, reducing or eliciting insulin resistance? Eur J Clin Nutr 2014;68:973–979. [DOI] [PubMed] [Google Scholar]
- 16.Zheng Y, Li Y, Qi Q, et al. Cumulative consumption of branched-chain amino acids and incidence of type 2 diabetes. Int J Epidemiol 2016;45:1482–1492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Iwasaki M, Ishihara J, Takachi R, et al. Validity of a Self-Administered Food-Frequency Questionnaire for Assessing Amino Acid Intake in Japan: Comparison With Intake From 4-Day Weighed Dietary Records and Plasma Levels. J Epidemiol 2016;26:36–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Colditz GA, Manson JE, Hankinson SE. The Nurses’ Health Study: 20-year contribution to the understanding of health among women. J Womens Health 1997;6:49–62. [DOI] [PubMed] [Google Scholar]
- 19.Rimm EB, Giovannucci EL, Willett WC, et al. Prospective study of alcohol consumption and risk of coronary disease in men. Lancet 1991;338:464–468. [DOI] [PubMed] [Google Scholar]
- 20.Willett WC, Howe GR, Kushi LH. Adjustment for total energy intake in epidemiologic studies. Am J Clin Nutr 1997;65:1220S–1228S. [DOI] [PubMed] [Google Scholar]
- 21.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;135:1114–1126; discussion 1127–1136. [DOI] [PubMed] [Google Scholar]
- 22.Feskanich D, Rimm EB, Giovannucci EL, et al. Reproducibility and validity of food intake measurements from a semiquantitative food frequency questionnaire. J Am Diet Assoc 1993;93:790–796. [DOI] [PubMed] [Google Scholar]
- 23.Willett WC, Sampson L, Stampfer MJ, et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol 1985;122:51–65. [DOI] [PubMed] [Google Scholar]
- 24.Yuan C, Spiegelman D, Rimm EB, et al. Validity of a Dietary Questionnaire Assessed by Comparison With Multiple Weighed Dietary Records or 24-Hour Recalls. Am J Epidemiol 2017;185:570–584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Stampfer MJ, Willett WC, Speizer FE, et al. Test of the National Death Index. Am J Epidemiol 1984;119:837–839. [DOI] [PubMed] [Google Scholar]
- 26.Rich-Edwards JW, Corsano KA, Stampfer MJ. Test of the National Death Index and Equifax Nationwide Death Search. Am J Epidemiol 1994;140:1016–1019. [DOI] [PubMed] [Google Scholar]
- 27.Hu FB, Stampfer MJ, Rimm E, et al. Dietary fat and coronary heart disease: a comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements. Am J Epidemiol 1999;149:531–40. [DOI] [PubMed] [Google Scholar]
- 28.World Cancer Research Fund/American Institute for Cancer Research. Diet, nutrition, physical activity and colorectal cancer. 2017. Available at http://www.wcrf.org/colorectal-cancer-2017 [accessed Aug 2019]
- 29.Chan AT, Giovannucci EL Primary prevention of colorectal cancer. Gastroenterology 2010;138:2029–2043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wu K, Keum N, Nishihara R, Giovannucci EL Cancers of the Colon and Rectum in Cancer Epidemiology and Prevention 4th ed. New York: NY: Oxford University Press; 2018 [Google Scholar]
- 31.Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics 1982;38:963–974. [PubMed] [Google Scholar]
- 32.Holeček M Branched-chain amino acids in health and disease: metabolism, alterations in blood plasma, and as supplements. Nutr Metab (Lond) 2018;15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Macotela Y, Emanuelli B, Bång AM, et al. Dietary leucine--an environmental modifier of insulin resistance acting on multiple levels of metabolism. PLoS ONE 2011;6:e21187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Nairizi A, She P, Vary TC, Lynch CJ. Leucine supplementation of drinking water does not alter susceptibility to diet-induced obesity in mice. J Nutr 2009;139:715–719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Shimizu M, Shirakami Y, Iwasa J, et al. Supplementation with branched-chain amino acids inhibits azoxymethane-induced colonic preneoplastic lesions in male C57BL/KsJ-db/db mice. Clin Cancer Res 2009;15:3068–3075. [DOI] [PubMed] [Google Scholar]
- 36.Wubetu GY, Utsunomiya T, Ishikawa D, et al. Branched Chain Amino Acid Suppressed Insulin-initiated Proliferation of Human Cancer Cells Through Induction of Autophagy. Anticancer Res 2014;34:4789–4796. [PubMed] [Google Scholar]
- 37.Cho E, Smith-Warner SA, Spiegelman D et al. Dairy foods, calcium, and colorectal cancer: a pooled analysis of 10 cohort studies. J Natl Cancer Int 2004;96:1015–22. [DOI] [PubMed] [Google Scholar]
- 38.Massa J, Cho E, Orav EJ, Willett WC, Wu K, Giovannucci EL. Total calcium intake and colorectal adenoma in young women. Cancer Causes Control 2014;25:451–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Yang W, Liu L, Masugi Y et al. Calcium intake and risk of colorectal cancer according to expression status of calcium-sensing receptor (CASR). Gut. 2018;67:1475–1483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Zhang X, Keum N, Wu K et al. Calcium intake and colorectal cancer risk: Results from the nurses’ health study and health professionals follow-up study Int J Cancer. 2016. November 15;139(10):2232–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Carr PR, Walter V, Brenner H, Hoffmeister M. Meat subtypes and their association with colorectal cancer: Systematic review and meta-analysis. Int J Cancer 2016;138:293–302. [DOI] [PubMed] [Google Scholar]
- 42.Geelen A, Schouten JM, Kamphuis C, et al. Fish Consumption, n-3 Fatty Acids, and Colorectal Cancer: A Meta-Analysis of Prospective Cohort Studies. Am J Epidemiol 2007;166:1116–1125. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.