Mouse magic, or How lab mice learned to stop worrying and trust the healing energy

I frequently call homeopathy The One Quackery to Rule Them All, but there are times when I am not so sure that that’s the case. You see, there is…another. I’m referring, of course, to what is referred to as “energy medicine.” What energy medicine modalities have in common is that they postulate that there is some sort “energy field” around humans that can be manipulated for therapeutic intent or that somehow practitioners can channel “healing energy” from elsewhere. For example, as I’ve discussed many times before, reiki is based on the concept that reiki masters can channel this fantastical healing energy from something called the Universal Source. That’s why I frequently liken reiki to faith healing, because, at its core, that’s what it is. Substitute God for the Universal Source, and it’s easy to see why. So what’s different about therapeutic touch (TT)? Basically, it’s the same thing, except that TT practitioners claim that they can wave their hands over patients (touching is actually usually not involved) and manipulate the human “energy field” to therapeutic intent. It’s a specialty so ridiculous that even a 11-year-old girl could show that TT practitioners cannot detect “human energy fields,” much less manipulate them.

None of this, of course, stops advocates from not only practicing “energy medicine” but designing nonsensically quackademic experiments testing “energy medicine.”

I hadn’t seen a particularly silly bit of seemingly “basic science” on energy medicine published in quite a while; that is, until now. The other day a study was brought to my attention. It’s a study by someone we’ve met before, Gloria Gronowicz at the University of Connecticut Health Center. Depressingly, she’s in the Department of Surgery there, and, equally depressingly, she did a study entitled Therapeutic Touch Has Significant Effects on Mouse Breast Cancer Metastasis and Immune Responses but Not Primary Tumor Size.

Basically, this is a mouse tumor model study. Amusingly, it works with a tumor model very similar to mouse tumor models I’ve worked with in my laboratory before. Specifically, the tumor model used is the 6-thioguanine-resistant 66cl4 cell line, which was derived from an aggressive 4T1 mouse mammary carcinoma that can metastasize from the primary tumor to popliteal lymph nodes (lymph nodes behind the knee joint). I’ve used the 4T1 tumor before. Basically, “syngeneic,” means it’s derived from the same mouse strain and is transplantable. That means that it can grow in mice with intact immune systems as long as they’re the same strain from which the tumor was derived. In fact, I know the man, Dr. Fred Miller, who derived the cell line and supplied the cells to Dr. Liisa Kuhn, the corresponding author of this publication.

I always love reading the introductions and methods to papers like this, particularly the explanation of TT and the justification for doing the study. For some reason, authors of pretty much every study of this type can’t resist an appeal to popularity, claiming that because so many people use “complementary and alternative medicine” (CAM), then killing mice to study magic is justified. In this case, the authors claim that, of all CAM, patients report the most benefit from “energy medicine” or “biofield” therapies. Drolly (unintentionally, of course), they note, “Scientific evidence for the possible reasons for this benefit is needed. As a first step we undertook a cancer study in animals to remove psychosocial factors.”

Gronowicz et al owe me a new keyboard for that one, as I spit up the iced tea I was drinking when I read that part. I suppose that on a strictly literal basis it’s true. Animals don’t have psychosocial factors. Well, that’s not entirely true. Mice don’t like being alone, for instance. They don’t like being too crowded either. They’re sensitive to how many mice are placed in a single cage. In any case, what the investigators did was to inject tumor cells into the footpads of the mice. To be honest, I’m not sure at all why they chose this route. Usually, with 4T1 cells, we inject them into the flanks of the mouse or into the mammary fat pad. However, these are mere quibbles with respect to methodology when it comes to mouse tumor model experiments. To truly appreciate the hilarity of the methods, you must read how the investigators administered TT to the mice:

TT treatments commenced 24 h after cell injection and were repeated twice a week for the entire period. Two mice at a time were placed into large tissue culture flasks (Sarstedt, Newtown, NC, 18 cm × 11.5 cm × 4 cm) with bedding by a technician through a premade hinged door. Previous studies from our laboratory had shown that tissue culture plastic did not impede human biofield treatments [36]. Flasks were clamped two feet in the air in a ring stand at the end of an L-shaped room. Practitioners alternated treatments so that each practitioner treated mice once a week. Treatment lasted 10 min with hands kept 2–10 inches from all sides of the flask without touching (TT1). Briefly the treatment sequelae were centering, assessment, treatment, and evaluation and followed previously published protocols [29, 36]. The control/mock group consisted of placing two mice in a similar flask and setup for ten minutes twice a week (CA1) at the other end of the same L-shaped room with a non-TT person standing next to the flask. The third group of mice was PBS-injected and received no treatment (PBS1). On the 26th day, mice had developed large tumors in their foot pad and were euthanized.

I love the bit about how prior studies showed that putting the mice in a tissue culture flask “did not impede human biofield treatments.” I was half-tempted to look up the paper referenced to see exactly how they had determined this critical bit of information, but then I figured I already had enough amusement for one night sitting right in front of me in the form of this paper. I also didn’t want to risk another keyboard or have to go dry while I wrote this. In any case, for full ridiculousness, just try to visualize what is going on here. Healing touch practitioners apparently stood over the large tissue culture flasks holding two mice each. The flasks were suspended two feet in the air by being clamped to a ring stand. They then held their hand over the flask containing the mouse and thought real hard or did whatever it is TT practitioners do to manipulate the human biofield, except they did it on mice. I couldn’t help but chuckle. A skilled filmmaker could easily make a comedy out this. How, for instance, does a TT practitioner trained on humans detect the mouse energy field? It is, after all, presumably so much smaller and squeakier than a human energy field.

Even more hilariously, the controls consisted of mice placed in the same flask but just having a non-TT practitioner just stand next to it. This must be because the healing power from a real TT practitioner is so awesome that it can’t be the TT practitioner just standing in the same room with the mouse but not exercising his or her skill. Note how the point is made that the control mice were placed at the other end of an L-shaped room. It’s as though the TT practitioners think that the residual energy of their awesomeness might affect the control mouse if they were placed in the same part of the room. Or maybe they did the controls and the TT mice at the same time. It’s not discussed. They also had a no-tumor arm, in which the mice just got a saline injection not containing any tumor cells, presumably for normal values for blood levels of everything they were measuring. Then, investigators repeated the experiment, except that mice were treated with TT or fake TT for two weeks prior to being injected with tumors. I’m telling ya, ya can’t make stuff up like this up. At least, I can’t.

Not surprisingly, the TT had no effect on the growth of the primary tumors. Indeed, the tumor volumes were about as close to each other as I’ve ever seen. Similarly, tumor cell proliferation and apoptosis were unchanged in the primary tumors, as one would expect. One thing I did notice is that the authors let these tumors get rather large, and surely they must have become painful. 220 mm3 is not that big a tumor when it’s in the flank or mammary fat pad, but it’s quite large for a mouse to have on its footpad, which is only a few millimeters in thickness, at most. In the second experiment, the tumors were allowed to grow to grow to as much as 330 mm3. Where was the University of Connecticut’s IACUC during all this? I realize that in this model the tumor has to be on the foot pad because the lymphatic vessels from the foot pad drain to the popliteal lymph nodes being assessed for metastases, but come on.

In any case, the authors’ positive results ended up being not particularly impressive. The authors did clonogenic assays, in which the popliteal lymph nodes were dissociated into a cell suspension and plated to see how many colonies of tumor cells grow. There does appear to be a modest decrease in the number of metastases to the popliteal lymph nodes reported, but the variability is high. Indeed, the authors had to do this to get their result to be statistically significant:

For the metastasis assay, most mice had 2–9 cancer cell colonies/lymph node. In the contralateral control limb (C), no tumors developed and no metastatic colonies were found (Figure 2). In contrast, every mouse had metastatic colonies in the mock-treated group (CA). In the TT-treated group (TT), three mice had no metastatic colonies while the remaining mice had some colonies. One mouse had 7-fold more colonies (76 colonies) than the mean. If this extreme outlier is excluded since it is greater than two standard deviations from the mean, TT significantly decreased metastasis compared to the mock-treated group (Figure 2).

Sorry about that outlier, but if you had to remove the outlier to get a statistically significant result, your result was probably not significant, and you still have to explain the outlier. Believe me, I know. I’ve had this sort of result before. Just because a value is two standard deviations from the mean is not sufficient reason to discard it. I will give the authors some credit, though. They did state that they removed the outlier and that without its removal, their results were not statistically significant. Now if only they didn’t keep referring from that point on to the differences in metastases to the popliteal lymph nodes as being statistically significant when in reality their own results show that TT didn’t affect the size of the primary tumors or the number of popliteal metastases. That‘s the real result: No effect on tumor growth or lymph node metastasis.

So all we’re left with is a fishing expedition among cytokines.

The authors look at a whole slew of cytokines and find some differences, none particularly striking, specifically decreases in IL-1α, IL-1β, MIG, and MIP-2. Of course, they used a commercial kit to check 32 cytokines, of which 11 were elevated in cancer and four were reported as decreased by TT therapy. All differences reported are modest. Also some differences in T-lymphocyte populations were noted. It’s hard not to wonder if this means anything at all, given that TT had no effect on primary tumors or metastases, as we would expect there to be no effect from such magic.

One thing I noticed as I read this study is that it all sounded rather familiar. In fact, so it was. I had written about this study before, only at the time it was just a poster presentation. Here we are, several months later, and the paper is published not in a good journal but in a quack journal, Evidence-Based Complementary and Alternative Medicine. At the time, I noticed that no mention of blinding was made. In the paper, the same thing is almost true. No mention of blinding is made other than that the core facility doing flow cytometry was blinded to experimental groups to which the samples belonged. Imagine my relief. That doesn’t change that the people measuring the tumors, counting the clonogenic assays, and measuring the cytokines almost certainly were not blinded to experimental group. Cytokine measurements can be very sensitive to how the sample is collected. If, for instance, the mice being euthanized were subject to different levels of stress right before death, that could produce differences in cytokine profiles. Remember, the blood was collected immediately postmortem by sticking a needle into the mouse’s heart.

So what could explain these results? Leave it to the authors to take a stab at it. First, let’s recall their justification for choosing TT:

Several reasons for choosing Therapeutic Touch for this study were the method of practice, which is an uncomplicated, well-defined protocol consisting of four steps, easily amenable for reproducibility of practice in a research trial and simple to perform in any setting [29]. The rigorous training program and credentialing process for practitioners, mostly nurses in all of our studies, was also important for consistency. There are no religious ties to the practice, so issues such as the role of prayer or religion are not involved in the interpretation of results. The first step in the practice is to set an intention, which is for the “highest good” of the subject. Finally, TT treatments do not require physical touch, so there is no heat transfer or variable handling of the subject being studied.

So the mice didn’t have to practice a specific religion to benefit. Imagine my relief.

Except that we know there was differential handling. The TT group held their hands over the flasks containing the two mice for ten minutes, while the control mice simply had some lab tech stand next to their flask for ten minutes. Similarly, the “control” mice, the mice with no tumors, who were used to produce control cytokine profiles and white blood cells were not regularly removed from their cages and placed in tissue culture flasks at all, unlike both the TT and sham TT groups. I would therefore question whether they were a proper control group for the cytokine measurements, particularly in light of the authors’ cluelessness:

A possible explanation of our findings is that the mice recognize and respond positively in a psychosocial manner to the biofield practitioner [46]. In studying psychosocial stress with inflammation and cancer, mouse models have shown that specific psychosocial stress factors produce generalized immune dysfunction, which particularly affects cytokine production resulting in changes in the numbers and function of specific leukocytes [47]. An alternative explanation of our findings is that the opposite of stress, such as exposure to a familiar and nonthreatening individual, may promote normal immune function. Mice attribute human contact with food, water, and positive environmental stimulation. Recently, rodents have been shown to detect and respond to the state of their social partners [48], and perhaps rodents may also respond positively to repeated human interactions. Thus, mammals may be capable of “felt affective experiences” [48]. On the other hand, mice that were placed in a similar apparatus by the same non-TT individual (CA group) did not manifest these changes in immune function suggesting that the TT treatment itself was responsible for the significant effects.

The problem is that the mice in the non-TT group were not handled the same way as the mice in the TT group, as I described above. Nor was the no-tumor control group handled the same way. More importantly, if it is true that differences in handling explain the results, then there is no need to invoke magic mouse energy fields manipulated by TT practitioners as the reason why there were modest changes in the cytokine profiles of the mice treated with TT. Just chalk it up to differences in handling, no magic required. Oh, and, big surprise, pretreating with extra TT didn’t change anything.

Papers like this simultaneously amuse and appall me. They amuse me because, well, they are ridiculous. Just the vision of earnest TT practitioners holding their hands between two to ten inches from a large tissue culture flask containing two mice are inherently ridiculous. However, it is appalling that many mice were forced to endure tumors growing on their footpads and then death to test whether TT practitioners can magically manipulate their mousy energy fields to cure their cancers. It’s also appalling that money was spent on this that could have been used in real cancer research.

On the other hand, I took a look at the foundation that funded this study, the Trivedi Foundation, which touts itself as giving “Scientific Research Grants to Raise the Consciousness of Living and Nonliving Materials Through the Authenticity of Modern Science”:

The Trivedi Effect® is a natural phenomenon that is harnessed from the universe and is capable of transforming living organisms and non-living materials to operate at a higher level and serve a greater purpose for the welfare of humanity. This phenomenon was discovered through the powerful energy transmissions of Mahendra Trivedi. Through the transmission of this energy, the recipient establishes a deep connection to the Creator, or Universal Intelligence, and awakens their Divine potential. He has since been able to transform three other individuals into Trivedi Masters™ who now have the ability to harness this energy.

The Trivedi Effect® has been scientifically proven to transform all living organisms, such as animals, seeds, plants, crops, fungi, bacteria, viruses, cancer cells and humans. Further scientific exploration has revealed that this energy has no limitations because it has the ability to transform the very structure of the atom: the building blocks of life itself. This means that the Trivedi Effect® is able to transform the very thing that this world is built upon. The Trivedi Effect® has the intelligence and profound capability to transform anything and everything. It is the ONE thing that CAN transform ANY thing.

On second thought, maybe paying to test whether waving hands over mice with breast cancer tumors in their foot pads can cure their cancer isn’t the worst thing a foundation like this could spend money on. Just peruse the website. This is Deepak Chopra-level woo. Besides, who wants people like this transforming the very structure of the atom and the building blocks of life itself. In any case, regardless of its proclaimed ability to “transform anything and everything,” one thing it can’t transform is woo into science.