Transcript

Flavor Aversion Learning in Cancer Treatment | Robert Batsell | TEDxKalamazooCollege

[Music] [Applause] [Music] [Applause] I love pizza, but I won't eat frozen pizza, and I haven't eaten it since 1986. I was in graduate school then, and like many graduate students, I was living on a limited budget, so frozen pizza was a regular staple of my diet. I still remember it was Tony's frozen pizza pepperoni and I was looking forward to it that night, but I was also coming down with the flu. So, nothing against the good people at Tony's, but after I ate that pizza later that night, I became ill and my brain associated the taste of pizza with illness and I have not eaten that frozen pizza since then. This is an example of flavor version learning and I imagine many of you have a similar story as in most reports of large populations about 70 to 80% can tell you about a flavor version that they have learned. Flavor version learning is a type of classical conditioning or Pavlovian conditioning. In my example, the pizza is the conditioned stimulus or the CS. the illness producing agent, the flu, was the unconditioned stimulus or the US, which produced the unconditioned response of illness. In many cases, the source of the illness may be the food itself, fish that's been left out too long or or derves that got on the table a little too early, but that's not the case today. In other situations, the source of the illness could be unrelated to the flu, like in my example of physical illness. In today's talk, I'm going to cover four properties of flavor version learning. First, one trial learning. Robust flavor aversions can be learned with a single pairing of taste and illness. From an evolutionary standpoint, this makes sense. For the majority of animals on our planet, and for many humans as well, finding food is at the top of their daily to-do list. If you are a foraging animal and you come across a potential food substance, it behooves you to eat a small amount of it because you don't know if it's going to be safe or not. If later that day you start to feel poorly, it's important that you can remember what that food was so that you don't eat it again. Because the next time you encounter that food, if you eat a much larger amount, it could be the last meal that you ever eat. Long delay learning. Typically, when illness follows consumption of food, you're easily able to make that association. However, many toxins are slow acting and they might not have their physiological effect for a number of hours after they've entered the body. Luckily, flavor aversion learning can overcome these delays. Lab studies have shown that delays of two hours, four hours, six hours, some cases 12 to 24 hours uh can be interposed between consumption of the food and illness and the aversion is still learned. However, as I noted, the closer the food is to the illness period, the stronger the aversion will be. Three, novelty matters. In other words, you're more likely to learn aversions to foods you haven't consumed before, foods that you haven't learned to be relatively safe. Often when we are enjoying meals, we're eating multiple foods. Some of them may be familiar. One or two might be novel. If later that evening you get ill, you're more likely to learn the aversion to the food you've never had before. The novel foods. Your familiar foods are somewhat protected. Now, as I've been talking about these properties of flavor aversion learning, I've been couching them in evolutionary perspective. Indeed, as taste aversions have been examined across the animal kingdom, uh the ability to learn flavor aversions has been seen in every vertebrate in which it's been tested and in many invertebrates as well. I'm sure you can easily see what an advantage this is for an organism who's trying to learn about what they're eating. And for that reason, this mechanism has been conserved across evolution. However, in our modern world, there are circumstances where individuals might be experiencing frequent illness episodes and as a result, there's a very powerful learning mechanism goes ary. And the example I'm going to focus on today are individuals with cancer who are undergoing chemotherapy treatment. Consider the story. A former student of mine was diagnosed with Hodkins lymphoma while she was in graduate school. She needed to take a leave of absence from her program and she moved home to live with her family while undergoing treatment. She's of Lebanese descent and on the day of her first treatment, her mother, wanting to cheer her up, made her favorite Lebanese dish. My student being the beautiful daughter that she is. When she went home, she was already starting to feel poorly, but she didn't want to disappoint her mother after all the work that she had put in. So, she ate her food. She then later became ill and like in my story of Tony's pepperoni pizza, she associated her favorite Lebanese dish with illness and wasn't able to eat it for another three to four years after that first treatment. As she said to me later, she knew that cancer was going to have a big effect on her life, but she was unprepared that it also could steal her favorite food from her. Now, let's break down this scenario using some of the terms we introduced earlier. The chemo agents can be typically radiation therapy or drug chemotherapy. Unfortunately, both of these types of chemotherapy can produce illness as a pronounced side effect. Oncologists refer to this as PVN for post treatment vomiting and nausea. It can occur in acute phase where it occurs for a couple of hours after the treatment. But in some cases, it may be delayed and the patient may experience it for a couple of days following the treatment. Now, a patient is likely to eat some food before going in for treatment. They might have a muffin in the morning before they then go in for that day treatment. If they later then experience illness, they might be associating this familiar food with illness. Later when they re-enounter that food they might experience AVN for anticipatory vomiting and nausea. This is the learned effect that has come about because of flavor aversion learning. It's important to note that AVN isn't necessarily linked to just food cues. One of my best friends from college uh successfully underwent chemotherapy treatment five years ago and he told me about one of the most powerful triggers he had for AVN. It was a smell of rubbing alcohol. He had his infusion port implanted in his chest and when he went in for each of the treatments, they needed to clean the port with rubbing alcohol. Then he had the drug infusions and as he was later starting to feel ill at the end of treatment, they'd also re-wab the port with rubbing alcohol. Later he realized that rubbing alcohol could be a very pronounced trigger for him to experience nausea and sometimes vomiting. He said that this persisted at least two years after chemotherapy happened. And he said every time it did, it made him think of me. I probably should take that more personally than I do. The point is a number of food cues can become triggers for Aven, but it doesn't even have to be food cues. Some patients have told me how the smell of their deodorant then becomes one of the triggers. Others have actually said the treatment center itself and seeing the medical equipment can trigger a bout of AVN. So the focus for the my remainder of my talk is going to be talking about our understanding of flavor aversion learning and how we can understand some of these properties to address the issue of anticipatory vomiting and nausea. The best way to prevent anticipatory vomiting and nausea is to prevent PVN prevent post-treatment vomiting and nausea. If the patients are not getting sick, then they cannot make the associations. Luckily, over the last couple of decades, there's been tremendous improvement in the uh development of anti-imetic or anti-nausea drugs. Indeed, some recent reports suggest about 70 to 80% of the patients who take these drugs experience a decrease in their vomiting and nausea symptoms. However, that also means there is a chunk of the population who doesn't get any relief from these drugs. And moreover, for many of the patients, the drugs are not effective every time. It oftentimes takes a couple of trials to find the right cocktail, if you will, of these anti-imetic drugs. Well, that will actually help that patient reduce their nausea. And as I noted earlier, it only takes a single trial for aversion learning to occur. There are some other concerns about eliminating PVN and that some patients might choose not to take the anti-imetic drugs and this has been reported due to a number of different reasons. In some cases, it might be a philosophical approach of I'm already taking enough drugs into my body. If this one isn't exactly essential, do I need to take it as well? But then the patient might not realize the other consequences that will occur from experiencing the nausea. In some cases, the patients have the mistaken notion that the illness is a necessary component of the treatment. In other words, if I'm not getting sick, I'm not killing the cancer. But that isn't the case as well. They do not need to personally experience the illness. A third reason that patients oftentimes report is finances. Going through chemotherapy treatment can be expensive. And for individuals of more limited means who are having to make choices, they might say, "Is this drug necessary for the treatment?" And if they can't afford it, they might eliminate it, which then gives the chance for PVN to occur and then maybe create the cycle for anticipatory vomiting and nausea. Unfortunately, once AVN has been learned, these antiimetic drugs have no effect on AVN. So instead behavioral scientists using these properties of flavor aversion learning have introduced developed techniques to redirect the learning to take it away and protect the patients familiar diet. One such technique is known as the scapegoat flavor technique which was developed by Dr. Eileen Bernstein. It takes advantage of some of the information the properties I mentioned earlier about novelty being important and proximity to illness being crucial for the learning. So this is how it occurs. As I mentioned before, the patient would eat their normal breakfast, say the muffin as they're going in and later as they experience chemotherapy, they have the potential to experience illness and maybe they would have learned an aversion to that familiar muffin. However, in the scapegoat flavor technique, the patient is given a second food, a novel food, one they've never had before, and one they don't care if they ever eat again. And it is given to them right before treatment. So, it is very proximal to illness. And because it is novel, it will take on most of the learning, thus protecting the familiar food. Indeed, in one study that Bernstein and her colleagues did, they simply use lifesaver candies. Lifesavers come in a variety of flavors and you can probably find some that individuals have never had before. In this case, they were working with children who were undergoing chemotherapy treatment and they asked the children later to report what foods they had learned aversions too. And on the trials where the scapegoat lifesaver candy wasn't given, the children did tend to report that I don't want to eat those foods I had that morning anymore. However, on those trials where the lifesaver candy was given, the children said, "I don't like that lifesavers anymore." But they would still eat the foods that they had for breakfast that day. Now, the scapegoat flavor technique is an effective and obviously with the use of lifesavers, very coste effective way to redirect this form of learning. But before somebody tries to implement this themselves or with their oncology team or recommend it maybe to somebody who you know is going through chemotherapy treatment, I want you to remember a fourth property of flavor aversion learning and it is that tastes and odors augment each other. In my own lab working with rat subjects, my students and I explore a phenomenon known as augmentation. Augmentation is a two-stage design. In the first stage, we're going to give the rats a sweet taste, saccharine, which they like to drink. And this is followed by illness. So, they'll learn a saccharine aversion. In the second trial, we now give them that same saccharine, but we're going to introduce a new cue, say in one study, almond odor, and this compound is now followed by illness. You might think from the earlier information on the scapegoat flavor technique that the previous learning the aversion of the saccharine might somehow protect the almond odor, but it does the exact opposite. If you test the almond over odor, what you're going to see is a super odor aversion. In other words, the presence of the aversive taste acts to increase or augment the learning of the odor aversion, thus potentially making it an even stronger trigger of AVN. So let's take our augmentation design and apply it back to the scapegoat flavor technique. We have an individual who is going through treatment. They are going to use the lifesaver technique before each treatment session they have. And let's say on that first trial they do learn an aversion to that flavor of lifesaver. On the next trial they decide I know what I'm going to do. That flavor was so effective last time in protecting my regular diet. I will use it again. I already dislike it. However, if some other cue is present, say the odor of coffee, which tends to be present often in treatment centers, or some other cue that the individual might experience, and again, this is followed by illness, the potential exists for the aversive flavor to augment these other cues and make them stronger triggers. So, the implication from the augmentation studies for the scapegoat flavor technique are clear. only use each scapegoat flavor once. In conclusion, taste aversion learning is a very old, very adaptive mechanism that has served animals on this planet well to help them identify and avoid things that might kill them. However, in our modern world where sometimes we need to harness the power of chemicals that can induce illness, this learning mechanism may be counterproductive in some ways to patients quality of life. By keeping in mind these properties of flavor aversion learning that I've mentioned to you today, researchers and encologists working together can build better treatment plans and improve patients quality of life. Thank you very much.