Jennifer Lovejoy, Arivale Chief Science Officer, PhD
Arivale responds to DIETFITS study in Journal of American Medical Association (JAMA)
As your trusted partner in wellness we’re always tracking new research related to genetics and wellness – and you probably are too. You may have seen a recent study published in JAMA called DIETFITS (“Diet Intervention Examining the Factors Interacting with Treatment Success”), which investigated the role of genetics and other factors in weight loss. Reports in the media reflect some confusion about the study outcomes. We believe the study had several limitations that make some of the results inconclusive, including the specific genes the researchers chose to investigate. To give you a clearer picture (as well as our perspective), this blog dives deeper into the study findings.
About the DIETFITS Study
First, some background. DIETFITS was conducted by investigators at the Stanford Prevention Research Center. More than 600 adults ages 18-50 years with a body mass Index of 28-40 kg/m2 were studied over 12 months. The study was designed to try and identify whether low-fat/high-carb or low-carb/high-fat diets are more effective for some people, based on their genetic makeup and other factors, including insulin secretion levels.
Individuals were randomly assigned to either a low-fat/high-carb or low-carb/high-fat diet and participated in 22 group weight loss sessions for one year. Both groups were told to maximize vegetable intake, minimize added sugars and refined flours, and focus on whole foods that were minimally processed and prepared at home as much as possible.
Similar to previous studies, the results showed no difference in weight loss between the low-fat and low-carb groups – both groups lost an average of 11-13 pounds over the year, although there was large variation in how much weight people lost (some people lost nearly 60 pounds, others gained as much as 20 pounds). Surprisingly, however, the investigators observed no difference in response to diet based on either the person’s genotype for the three variants or their insulin secretion.
Genetic Variants for Weight Loss in the DIETFITS Study
The DIETFITS study considered three genetic variants in their analysis: FABP2 (rs1799883), PPARG (rs1801282), and ADRB2 (rs1042714). It is unclear why the investigators chose these specific genetic variants given that other variants also had evidence of gene/diet interaction at the time the study was designed. Further, it appears that the investigators weighted each variant equally in their genetic model, regardless of the strength of the evidence or other factors. This may have affected the study outcomes, as which genetic variants are included in a genetic risk profile and how they are individually weighted is likely to play a role in what findings emerge.
Genetic Variants for Weight Loss at Arivale
Arivale has evaluated more than a dozen genetic variants for which published literature has suggested an effect on response to low-fat or low-carb diets. We currently report on five genetic variants related to diet and weight loss (at or near the genes APOA5, TCF7L2, FTO, TNF, and PLIN1). We report our results to our members as “polygenic profiles” this is like a snapshot of the sum of the effects of all the common genetic variants implicated for a quantitative trait or complex disorder that are observed for that specific person.
Many of the variants that we potentially could have included for our diet and weight polygenic profiles did not meet our bar for adequate evidence, including the ADRB2 variant used in the DIETFITS study. To date, only one case control study (159 cases) has investigated how the ADRB2 variant could impact body weight relative to fat or carbohydrate intake. Results from this smaller study showed that women who consumed more than 49% of calories from carbohydrates had a higher BMI. This effect was not found in men, perhaps because there were too few in the study, and the findings in women have not been replicated. For these reasons, Arivale does not include the ADRB2 genetic variant in its model for predisposition to gain weight with carbohydrate. In fact, we haven’t found compelling evidence yet for any genetic variants that increase risk for weight gain with high carbohydrate diets. Rather, the opposite may be more likely; there are several common variants that point to decreased weight gain from carbohydrates.
Arivale has found stronger evidence for genetic variants that convey increased risk of high body weight when total dietary fat is high. Interestingly, none of these genetic variants were included in the DIETFITS genetic “score.” The DIETFIT authors did state that analyses are underway to evaluate the impact of other genetic variants on weight loss outcomes, so this may lead to different findings in the future.
Key Takeaway #1: Genes Still Matter
Genes matter when it comes to weight management, but they should not be considered in isolation. In deciding on the optimal diet for weight loss, individuals should consider genetic background, lab values and other clinical markers, personal preferences, and lifestyle and environmental factors. A great example of why is someone who has genetic variants indicating they would do best on a low-fat/high-carb diet, but who also has high triglyceride levels, a condition that generally requires lower carbohydrate intake. In this case, the lab finding (high triglycerides) might trump genetics in terms of selecting a weight loss diet, or a more moderate-fat/moderate-carb diet to optimize both the triglyceride and genetic factors during weight loss may be more appropriate.
Insulin and Weight Loss
The role of insulin secretion or insulin resistance as a factor in weight loss has been a controversial area for many years. Although several studies have suggested that having a higher insulin level may make it harder to lose weight, other studies have found instead that people who have the lowest insulin levels have the hardest time losing weight. It’s well known that being overweight increases insulin secretion, so if high insulin levels were a barrier to weight loss, lots of people would be in trouble! This is clearly not the case, and the DIETFIT study is one more in the list of studies showing that increased insulin secretion is not a significant factor in the ability to lose weight.
Key Takeaway #2: Insulin Levels Don’t Predict Weight Loss
Calories and Weight Loss
Although the DIETFIT study itself doesn’t draw any conclusions about the role of calories in weight loss, the authors do note that participants in the study weren’t given a specific calorie guideline. The media has run with this, claiming that “calories don’t matter for weight loss” – after all, who wouldn’t love to believe that was true! Unfortunately, that’s not really what the study reported. On average, participants in DIETFIT reduced their calorie intake by 500-600 calories per day compared to baseline, which is why they lost weight.
So, why the confusion? While the study didn’t give a calorie target per se, they did prescribe other very specific dietary targets that were “proxies” for calories. For example, the low-fat group was initially asked to restrict its fat intake to 20 grams per day, and the low carb group to 20 grams of carbohydrate per day. These levels are quite low and not easy to achieve without significantly cutting calories. In addition, everyone was told to eliminate sugar and refined carbohydrates from their diet. For most people, this change alone would result in a reduction of at least several hundred calories per day. Finally, the participants were strongly encouraged to prepare all their meals at home. Restaurant and fast food meals are a major contributor to excess calorie consumption, so this practice also would have significantly reduced calorie intake for many participants.
Not surprisingly, the calorie reduction achieved in the study (500-600 calories/day) is what is typically recommended in weight loss plans – a reduction that leads to a 1-2 pound/week weight loss for most people. Furthermore, the concept that people are more successful losing weight when they eat lots of vegetables, fruits and other “high volume, low calorie” foods is also not new. The notion of “volumetrics” as a proven method of weight loss has been around for several decades and is widely recommended to reduce hunger while reducing calories, making a low-calorie weight loss plan easier to stick to.
Key Takeaway #3: Calories Matter a Lot
Unfortunately, it’s a myth that you can “eat all you want and still lose weight.” What doesn’t matter is how you choose to count your calories. You can count actual calories, fat or carb grams, servings or exchanges, or points – the bottom line is that you have to significantly reduce calories to lose weight.
Other DIETFIT Study Strengths and Weaknesses
We think the DIETFIT study makes an important contribution to the scientific literature on weight loss. One major strength of this study was that both diets placed equal emphasis on food quality and a consuming a whole foods-based diet, so their results can’t be attributed to differences in diet quality. In addition, the investigators used a behavioral weight loss approach based on frequent group sessions with a Registered Dietitian that has been validated as effective for producing and sustaining weight loss in many previous studies, regardless of diet type.
However, while using a proven weight loss method is a strength, it may also be a limitation in that it may have “washed out” differences between diets that might be seen with a less rigorous behavioral program, such as when people try to follow these diets on their own.
Another limitation is that the study didn’t randomize or stratify the study groups by genotype or insulin resistance status, which, as the authors note, limited “the level of causal inference to be drawn.” In other words, the study wasn’t designed in the best way to test the hypothesis that genes matter for weight loss.
While we recognize that the science of weight loss genetics is still in early stages, there are many well-conducted studies that have looked at different genetic variants and found significant impacts of dietary fat versus carbohydrate on weight loss. These are the genetic variants we report to our members. Arivale scientists continue to curate new studies as they emerge to keep us – and our members – up to date on this exciting new field.