Talking precision nutrition with Dr Dariush Mozaffarian: 'The microbiome is going to be the absolute foundation of personalized nutrition'
As PeakBridge leads a $15m round in Cambridge, Mass-based InsideTracker, which integrates biomarker data from blood, DNA, activity trackers, and other user-generated data to create personalized recommendations to optimize health, we revisit a recent fireside chat* on precision nutrition at Branchfood’s virtual Food Edge 2022 summit moderated by FoodNavigator-USA (FNU) featuring Dr Dariush Mozaffarian, M.D, Dr.P.H., Jean Mayer Professor at the Tufts Friedman School of Nutrition Science and Policy; and Professor of Medicine at Tufts School of Medicine (DM).
FNU: What is precision nutrition?
DM: There is no accepted definition. I think personalized nutrition is the idea that we want to personalize something to the person, but I think precision nutrition goes further and applies data and quantitative science and artificial intelligence, machine learning and metabolomics. But they're fairly broad concepts, and we need to understand the evidence supporting those concepts.
FNU: What biological data might inform a precision nutrition strategy and where does DNA fit into this?
DM: 20 years ago, all the attention was on genetics and the Human Genome Project, and there was an optimistic belief that if we figured out genes we could solve disease. And we've learned over the last 20 years that that was overly optimistic, that for complex chronic diseases like diabetes, obesity, heart disease, stroke, and most cancers, there aren't a lot of major genetic causes; the major preponderance of impact is from the environment and lifestyle.
And we know that because we've had an obesity epidemic that has exploded globally in 30 years and our global genes haven’t changed. So I think genetics are going to play a pretty minor role… I think epigenetics - which is how the environment changes the transcription of the genes and how it changes the phenotype that comes out of the genes - that will be important for sure, and that can be changed in weeks or months in a person and can also be transmitted from mother to child.
So I wouldn't put a lot of stock in measuring your genetics, but stay tuned for what we might learn from epigenetics.
FNU: A study from the Weizmann Institute in Israel a few years ago found big differences in the rise of blood glucose levels among people after consuming identical meals... Can you talk a little more about the gut microbiome and precision nutrition?
DM: I think that the microbiome is going to be huge. The microbiome is going to be the absolute foundation of personalized nutrition, both how nutrition changes the microbiome and then how the microbiome’s response to nutrition influences our health.
We don't have enough science now to customize a person's entire diet based on personalized factors, so I think focusing on little bite sized pieces at a time makes sense, and I think the area that has the most evidence so far is the link between a person's microbiome and their glucose and insulin response to carbohydrates.
That paper and others have shown that if you measure the microbiome and add information about the person, about their age and weight and medical conditions, you can start to predict that for a similar carbohydrate, you might have a different glucose response.
Now, of course we could just take a pure public health approach and say, ‘We need more beans and nuts and seeds and healthy oils and less refined sugar and carbs,’ which is true. At the same time, it’s part of the diet and culture to have starch and sugar, so I think knowing which might have the biggest glucose and insulin response in you and what might be better, could be helpful.
So it's terrific early science, but that now that needs to be translated to show that it actually makes a difference in health and in healthcare.
FNU: Why might people have a different glycemic response to the same carbohydrate?
DM: The microbiome interacts with food in at least two ways. One is the types of bacteria in there, so there could be more of one species vs another species, and the second is the function of the bacteria, so you could have similar composition of bacteria, but in different people, those bacteria might produce different compounds that get into the bloodstream.
Based on animal models, somewhere between a fifth to a quarter of all the circulating compounds in our bloodstream come from our microbiome, and so if you think about hormones and other things in our bloodstream that are influencing the body, maybe a quarter come from the microbiome, not from our own cells, and so there's huge influences of the microbiome on our health.
So I think that what defines a healthy diet, in large part, is what gives us a happy healthy microbiome versus an angry vindictive microbiome.
Imagine you eat one type of whole grain. If you have bacteria that eat that up and digest some of it including its starch, your body is going to digest fewer of those calories. If you don't have those bacteria, maybe your body's going to digest more of those calories.
So when you look at a calorie count on a package and it says this has 150 calories per serving, maybe your gut bacteria are going digest 50 of those 150 calories and you're going to only get 100, while another person’s gut bacteria will only digest five and they will get 145 calories.
So that's something we haven't even started to scratch the surface of in terms of science.
FNU: How actionable is this kind of information right now?
DM: There are companies like Day Two, for which I have served on the Scientific Advisory Board in the past, and they are trying to work with healthcare providers to make it actionable. In their studies and in other studies, they've shown that giving this information to people with diabetes specifically helps them stay in normal glucose range longer. So that's important. If you can stay in normal glucose range longer that can be important. Now, whether that translates into better health outcomes and lower health care spending, that's what we have to show now.
And that's why we need the private sector and government to invest in this because at the current pace, it might take us 30, 40, 50 years to get to where we want to be, and my goodness, we don't have 50 years to wait.
FNU: Can you provide more examples of how differences between individuals might inform a more personalized approach to nutrition?
DM: I think there are some early data to suggest for example, that different people may have different blood cholesterol responses to different dietary fats. In weight loss there's evidence that people who have higher insulin secretion in response to carbohydrate do much better on low carb, high fat diets. If you know your insulin in response to a carbohydrate, you know how important lowering carbohydrates is for yourself for weight loss.
A third area that’s interesting is the timing of meals. There is research that shows we should have three square meals a day and never skip a meal and always have breakfast; there's research that shows that we should be grazers and have small meals throughout the day; and there's research that shows intermittent fasting is advisable.
Those three lines of research are completely contradictory, but they are all backed up by some evidence and so very likely there's personalized differences between people who may do better with grazing vs intermittent fasting vs three square meals a day. But we’re not there yet.
FNU: What are the research priorities in the precision nutrition field going forward?
DM: If you look at the number of diseases and health outcomes that are influenced by nutrition and or the microbiome, from autism and depression, all the way to gastrointestinal health , celiac and other diseases, cardiovascular disease, diet-related cancers, even Alzheimer's, immunity and autoimmune diseases… where do we even start? All of it needs to be understood.
So I think that's what the new NIH precision nutrition initiative is trying to do, to enroll people in both large observational studies with careful monitoring and apps and extensive phenotyping, and also do carefully controlled trials to try to disentangle these things.
But $300m is absolutely nothing if you think that the US government spends $160bn a year just on direct medical care for diabetes, which is almost entirely preventable by good nutrition.
FNU: What would you like to see from the food industry?
DM: One, make sure what you're doing is directly focused on nutrition, equity, or sustainability, because if you're not really mission oriented, you're going to fail in in the modern world. You can't just be going from Doritos to Cool Ranch Doritos and calling that innovation.
Second, be sure your marketing and your plans don't exceed the science.
I also want to emphasize equity. If this is only for those who are high-income and highly educated, we're only going to worsen health inequities with some personalized nutrition approaches. So we have to be sure to focus on communities that have been traditionally marginalized and disadvantaged to be sure that the science is accessible to everyone.
FNU: So the National Institutes of Health is putting some money behind precision nutrition?
DM: The top cause of poor health in the world is poor nutrition, so it's striking that of the 27 centers and institutes that NIH has, none are focusing on nutrition. We have called on Congress to create a national institute of nutrition so we can have a nutrition moonshot and in the next five or 10 years really dramatically expand our knowledge of the science.
That requires congressional action, but in the meantime, the NIH has funded a $300m precision nutrition initiative.
At Tufts, the Human Nutrition Research Center on Aging, which is funded by USDA, has one of the six clinical research centers that's going to push forward this work. The NIH has also created a new office of nutrition research within the office of the director, so these are baby steps toward where we need to go; we really need a global nutrition moonshot.
FNU: Our grandparents didn't know anything about personalized nutrition, but levels of obesity and diabetes were way lower then than they are now. Aren't the real problems causing diet-related disease more structural?
DM: 100%, I agree that the biggest problems are structural, and President Biden has announced the White House Conference on Hunger, Nutrition and Health to bring together all of the agencies of government to really try to tackle hunger, nutrition and diet related chronic diseases by 2030, which is the first time in 50 years that the government is going to be focusing on nutrition.
So for sure, there are structural changes that have to happen. But in the meantime, if a person has the means to access some of these personalized nutrition services, it may help change their behavior, that’s a good thing, right?
Broad approaches to motor vehicle safety, smoking, sugar sweetened beverages work for some things, but they haven’t worked for everything. There will be key discoveries that will help individuals improve their health, and also help us understand how to motivate people better. Maybe part of the power of precision nutrition will be that personalization has an incremental health benefit, but also a huge behavioral benefit?
*The transcript has been edited for clarity and brevity.
