Microbe Mania with Melancholic Microbes
Recently in a train station book shop I stood gaping in astonishment in front of a thematically highly specialized book display. It was the bowels-brain table. The books piled up on it promised enlightenment about how the bowel and in particular its contents influence us – yes – how, they verily steer our emotions. A selection of book titles: “Shit-Wise – How a Healthy Intestinal Flora Keeps us fit”; “Bowels heal brain heal body”; “Happiness begins in the bowels”, or “The second brain – How the bowels influence our mood, our decisions and our feeling of wellbeing”. Newspapers, magazines and the internet can also tell us this. The wrong bowel bacteria make us depressive – but the right ones make us happy … which is why yogurt helps against depression.
Whence doth come this sudden enthusiasm for the right bowel flora? This social acceptance of the bowels, their contents and discussion about them even at the dinner table was certainly due to Giulia Enders. In 2014 while a medical student, she published in the million-seller “Darm with Charme – Alles über ein unterschätztes Organ”, (The Gut: The Inside Story of Our Body’s Most Underrated Organ) translated in the meantime into umpteen languages. But this did not come from nothing, for science set the foundation for a Microbe Mania and continues feeding it today, with persist “spectacular” discoveries. All wonderfully suited for capturing a broad public. Studies in which it is shown that you can predict the cognitive development of a one-year-baby from its diapers … Timid mice suddenly turn bold and daring after transplantation of feces from rambunctious rodents… or a mouse model in which the animals develop symptoms only after application of stool from Parkinson patients… Autistic children who become social again with prebiotic therapy. And so on and so forth. Doesn’t that remind us all a bit of the stem cell hype?
The whole bowel business is actually old hat. Even Hippocrates knew: “All diseases start in the bowels”. In 1917 the military surgeon Nissle cultivated a quite special bacterial strain which he immediately named after himself: E. coli Nissle. He found it in the stool of a soldier who was especially resistant to the brutal conditions in the trenches of the First World War in the Balkans. Adolf Hitler was so impressed by that, that he prescribed himself (and his German Shepard Blondie) two capsules E coli Nissle a day. Even today still, anyone who wants to can gobble down bacteria from the bowel of the unknown soldier: They are sold under the name Multaflor as a probiotic in every drugstore. The soldier however is not named on the package insert. But neither Hippocrates nor Nissle could explain the comet-like ascent of the microbioms. Because it is based more than anything else on the advances of gene sequenced technology, which makes it possible to identify bacteria genetically in high throughput. Without previous cultivation. Previously, in the culture dish, the oxygen destroyed the anaerobes; after that, no anaerobes are detected. Only the sequencing of the bacterial genomes brought to light the enormous variety of the microbiome in humans and mice. The rest is history.
In the meantime, Pubmed is spitting out 30 new microbiome papers per day. No area of specialization that has escaped the grip of Microbiome Mania. Neurology and psychiatry are only the peak of the iceberg. Cardiology, endocrinology, nephrology on their trail. From the main German research funder’s (DFG) Fachkollegien (‘Study sections’) we hear that a substantial portion of the applications now include a section on microbiome. And this from scientists many of whom have neither experience in microbiology and procaryotes nor bioinformatics of bacterial genomes. And this is where things start getting problematic.
In addition, many microbiome animal studies have the same shortcomings as those we know from other fields: frequently non-randomized, not blinded, no preceding definition of criteria for inclusion and exclusion, low number of cases, no preceding formulation or registration hypotheses, and so on. That opens all doors to the selective usage of data, including omission of uncomfortable findings or changing hypotheses in the course of experiments. Spectacular findings – but no independent confirmation. Publishing such studies is relatively straighforward, because many scientific journals are very keen right now on publishing microbiome studies… but only those that deliver positive results. Not any that contradict previous findings or simply render no microbiome effects. Since I have published a few papers on the microbiome (some even with a lot of non-significant differences, i.e. NULL results), I receive invitations from reputable journals (not predators!) to submit microbiome-brain studies.
Unfortunately, things do not look much better in most clinical microbiome studies. They are usually quite small, poorly controlled, often not pre-registered, patient groups show high heterogeneity and statistics and bioinformatics are questionable. Due to the high number of studies published, a whole string of meta-analyses has been performed. Their usual conclusion is: It is not possible to conclude anything, because the studies available do not have the minimum quality requirements; they are too small and reveal a massive publication bias. The few well-conducted studies are either neutral or even negative, as was the case in a recent, relatively large study on patients with Irritable Bowel Syndrome (IBS). The patients in the stool transplantation group faired worse than did those treated with placebo treatment (Halkjær et al. 2018). The biggest problem in the clinical microbiome literature is, however, that many observational studies mistake correlation for causation and vice versa. Depressive patients have a microbiome that is different from that of happy probands. Ergo it must be the microbiome that causes depression. The idea that people with depression have other eating habits and other life styles, which of course impact the microbiome, is blithely overlooked. Here we can replace depression with autism, Parkinson, etc.
We thus once more experience, as previously in the stem cell field, a classical so-called Gartner Hype Cycle: After the introduction of a new technology and the subsequent euphoria, expectations soar ever higher. All studies find what they were supposed to find. This is dampened down only when many scientists working in the field find nothing new and exciting to report. Then the “noise” gets so high and the evidence base so instable for want of quality that the hopes give way to disappointment. Although Microbiome Mania has not yet reached this phase, the first signs are showing and growing. I have mentioned some of these signs above. When disappointment falls to a new low, disappointment sets in; the grain separates from the chaff, the solid results harden into a deepening understanding of the mechanisms and a stable level is reached, the “plateau of productivity”. The microbiome field has yet a long way to go.
Does research really have to undergo these cycles, reeling from exaggerated expectation to bitter disappointment? There is nothing wrong with enthusiasm and hope in a new discovery – those are, after all, the essential drivers behind successful research. However, the plateau of productivity could be reached much faster and in in a way that would not devour resources so greedily. But for that we must learn from the cycles run to date. Researcher, journalists, funders, and the media should simply keep in mind a few basic rules of good science. And what would they be? Reduction of bias through measures such as randomization, blinding, pre-registration; reduction of the rate of false positive results by using larger groups sizes (and thus statistical power). Results that contradict our hypotheses should be published, and important findings reproduced independently.
When we read that stem cells heal paraplegic mice, or that lactobacilli make depressives happy, we should remember Carl Sagan’s dictum: Unusual claims demand unusual evidence!
Further reading:
A German version of this post has been published as part of my monthly column in the Laborjournal: http://www.laborjournal-archiv.de/epaper/LJ_18_12/20/index.html
To infinity, and beyond! 