‘We are very close to the time when…the only treatment was your own immunity’

India’s infant mortality rate, although in steady decline, continues to be a black spot, putting it on a par with some of the poorest and most malnourished countries in the world. A recent study led by Pinaki Panigrahi, professor at the department of epidemiology and pediatrics, University of Nebraska Medical Center, has added a potent new weapon in the fight against infant mortality (go.nature.com/2iVAyyo).

As widely reported, Dr Panigrahi and his team have identified a probiotic strain, which, when combined with a fructo-oligosaccharide (FOS, a kind of sugar found naturally in breast milk and some fruits and vegetables), forms a solution (called a synbiotic) that prevents neonatal sepsis.

Sepsis is a life-threatening complication of a bacterial infection that occurs when the body’s immune system, overwhelmed by the infection, turns on itself, leading to tissue damage and organ failure. It kills around 600,000 babies around the word every year. The exact number for India is not known, though it is estimated to account for a quarter of newborn deaths (Source: Lancet Global Health study, 2016)

Dr Panigrahi’s team has been running the largest ever clinical trial involving probiotics since 2008 at multiple sites in Odisha, administering the synbiotic solution to thousands of infants. It was found to be strikingly effective. The solution did not just significantly bring down the incidence of sepsis, it also reduced the risk of infections overall, including respiratory ones.

The data safety and monitoring board, an independent group of experts monitoring the study, in fact stopped the trials midway because the results showed such startling efficacy that it would have been unethical to continue depriving the “control group” of the treatment.

The synbiotic offers an astounding promise: a very cheap, easy to manufacture and easy to administer oral vaccine against one of the biggest known killers of babies—for the trial, the infants were given the synbiotic for only seven days.

Dr Panigrahi, who did his postgraduate training in pediatrics in Odisha, before shifting to the US in 1985, spoke over the phone, giving in-depth details about his breakthrough. Edited excerpts:

You began trials in 2008, but your work, studying neonatal diseases in India, began much earlier. How did that lead to these findings?

I was studying necrotizing enterocolitis (a disease where the tissue in the intestine dies. It mostly affects premature infants, and its exact cause is not known) and we published some seminal observations more than 20 years ago. The consensus then was that bacteria played a major role in the disease, but what we showed was that it was not the presence of bacteria but the absence of good bacteria or bacterial diversity that was causing disease, and that some good bacteria could have prevented it. We knew from that time that it’s the bacterial diversity and combination, the presence of good bugs in the right numbers, that’s critical.

Then we did a basic surveillance study of infection in 2001, going out to the community, trying to find out why babies die, when do they die. Is it early sepsis or late? If it’s early-onset sepsis, it comes in the first three days of being born, it comes from the mother, from the vaginal tract or the (umbilical) cord-cutting. If the infection goes to the bloodstream, then the infant will show symptoms and unless treated, the baby will die. But, we also found that most of the babies who died in the first week of life were not dying of infection but due to birth asphyxia. Late-onset sepsis (occurs between 4 and 90 days of life), on the other hand, was responsible for 92% of infection-related deaths. And these are apples and oranges—no, not even that—apples and meat curry. A treatment for early-onset sepsis will not work for late onset.

We wanted to see how this sepsis was happening. The results of that have not been published yet, but we have mentioned this in the Nature paper and shared the data with peers. When we did the blood culture for the babies, we also took the baby’s umbilical swab, stool and skin samples. We also took samples from the vaginal tract, and the mother’s blood. We compared those and saw that the bulk of the sepsis was happening through bacterial translocation through the gut: the bugs were going from the intestine, from the stool and going through the mucosal layer and into the bloodstream. We thought that if that’s the case, then we can definitely block it with good bacteria introduced to the gut. It’s only after that, that we started intensely looking for a candidate strain.

Probiotics is difficult territory—they have rarely proven effective in clinical studies.

The previous studies that you find in the literature, sometimes the probiotics work, sometimes they don’t and it’s all over the place, that’s why you can call probiotics either a miracle drug or snake oil, and for valid reasons. In the past, most people have taken what is available already in the market and used it to treat disease. If you ask them, why did you use this strain, they have no answer. That is extremely discomfiting for people in medical research.

In our trial, we had identified a particular disease, we knew its mechanism and we were looking for a very specific thing. We screened around 280 strains of probiotics, around 70-80 off the shelves, from US, India, Russia… but as expected, none of them did anything in our model. Sometimes they would attach to the intestinal cells but wouldn’t be able to block, say, the E Coli from attaching as well. Sometimes they could not prevent the translocation of bacteria in the tissue-culture model.

Then we started isolating bacteria from the diapers of healthy babies. We found two strains that worked, Lactobacillus Plantarum and L Salivarius. The Salivarius was doing the job that we wanted it to do, but it was also translocating by itself, which means it would cause disease, so we stuck to Plantarum.

One of the major problems with the efficacy of probiotics is that these good bacteria fail to colonize the gut.

Yes, in India we had done clinical trials but colonization just didn’t happen. So we thought perhaps if we could give it some prebiotic sugars it would help the bacteria grow. The bacteria need food, but regular sugar gets absorbed in the small intestine so it’s not available to the gut bacteria. But FOS is non-absorbable, so that’s why we added it. We were not sure if that would actually work. We did a small clinical trial with that combination and we found that Plantarum was colonizing wonderfully for two months and up to four months, if given for the first seven days after the baby is born. Then the colony remains for the first 60 days and after that babies don’t get the disease because the bulk of sepsis happens within the first month and the second.

Were you nervous when the trial was stopped early?

I have been on many data-safety-monitoring boards and I haven’t had any experience where we stopped any study due to early efficacy. The first thing we thought was that there was something wrong. We were worried though we knew there were not too many deaths, but we did not know how many babies were spotted with disease. We were delighted when we finally got to know why the trial was stopped.

And the positive effect on respiratory disease was another surprise.

Yes, we were absolutely surprised when we realized that there were positive results. How can respiratory infection go down like this? There is literature that myocardiopathy and other neural diseases can be affected by good bacteria, but that was not the design of our trial. But then the intestine is such a large organ, it is the largest lymphoid organ (lymphoid organs form part of the body’s immune system). If you open it up, all the intestinal microvilli will cover a tennis court, and it is a similar size in a child as well. Also, diarrhoea also went down a bit, umbilical cord infection had also gone down, all of those things helped a little.

But we don’t know the exact mechanism of how this happened and that’s what we want to find out next—which component of the immune system was affected, and then hopefully, if we know the mechanism, it can have many more implications in the overall management of disease.

It has been reported that it had an effect on stunting (impaired growth and development in infants) as well.

No, we haven’t done any study yet, but that is the next thing to do. Stunting is extremely worrying, much more than sepsis. Children are being born stunted in large numbers, especially in South Asia and sub-Saharan countries. Why they were being born stunted was a mystery, but is no longer a mystery. First it was thought that it was because of undernutrition in utero—the mother is not well fed—and that is partly true in sub-Saharan countries. But the biggest reason is environmental. Stunting happens in utero and in the first two years of life; we call it “thousand day”. When we were in medical school (in the 1980s in Odisha), this was not something we actually saw, we only read about it in textbooks. For example, even dengue or chikungunya, I never saw a case of dengue when we were in school, and we used to make jokes about how the word chikungunya sounded.

The biggest problem is population density and that the environment has gotten much worse. People are literally living in a latrine, so babies right now are getting bombarded constantly by millions and millions of bugs within hours and days of being born. You will find that most babies have some enteric infection every month. One goes away and another sets in. During this period there is a dysfunction: no matter how much you feed the baby, the baby can’t absorb it. So, along with intrauterine events and this, they would grow up cognitively impaired.

For your trials, you picked healthy babies who were being breastfed. Are you planning trials with more diverse subjects next?

Yes, we are. The chances of getting any infection, including sepsis, is much higher if you are premature or of low birth weight, and the degree of prematurity and low birth weight is proportional to the problem. We will now go back to that population and make sure that it works on a wider group of subjects and that it doesn’t cause any problems itself. It won’t be another 15-year ordeal. These will be smaller studies, different birth weights and prematurity.

How far away are we from actually seeing the synbiotic being adopted for clinical use?

We are a few steps away from that, because we have only done trials in one part of the country. Though we did spread it out over different areas—tribal, coastal, etc.—and spread it out over months so seasonal variations were controlled for. But these have to be repeated. We are not too far, we need to do trials in three-four other places and then we are good.

Did you see a lot of antibiotic-resistant microbes during your trial?

I have goosebumps on my arms even as we speak. It’s the only thing I’m afraid of for my relatives and friends in India. In the previous surveillance study that we did in Delhi, Mumbai and two sites in Odisha, we found that even babies born at home who had never seen an antibiotic, never seen a hospital, came to the hospital with pretty bad antimicrobial-resistance infections. About 45% of them E Coli, etc. Where are they getting them from? Most probably from the family, from the surroundings, because what we are eating and drinking is full of bacteria. We have taken antibiotics a hundred times being Indians, and we have resistant bugs that haven’t killed us because they haven’t translocated from our intestines, but we pass them on to the environment from where the babies get it.

We have to have other modalities against infection and not just antibiotics. If you tell me I’m using neem oil, I’ll say please do.

In our own study we have some evidence, very small, that acquisition of the antimicrobial-resistant bugs is lower in babies getting the synbiotic than in the control groups. It lasts for the first three months and then that effect washes away.

This is a very common problem now in US and Russian hospitals as well. (Former US President Barack) Obama called it a threat to national security. It’s a problem for everyone. It threatens the return of a pre-antibiotic era in Indian NICUs (neonatal intensive care units). We are very close to the time when there was no penicillin, and the only treatment was your own immunity.

Do foods with good bacteria, like dahi, actually help in any way?

Dahi or any sour product like kefir or kimchi are a good source of some bacteria but are they a wonderful source of probiotic bacteria? No.

In olden days, there was no tropical enteropathy, less bombardment of bugs, less environmental pollution. The kind of toxins we are now exposed to, the chemical fertilizers and pesticides we are eating and drinking, none of them were there. In a cleaner, healthier environment, to maintain good health, you take some sour product, get some lactobacillus, and it works. If you need probiotics for clinical reasons, that is a different thing altogether. For yogurt companies, their intention is not to save you from sepsis or infection or diarrhoea. The primary purpose is to make a nice tasting yogurt which has nice consistency and has stability. We did not find a single strain from a yogurt that will work in our model. Most strains of probiotic bugs don’t colonize the gut at all, they get in and get out. You can feed someone billions of bugs, but most of them will die in the stomach acid, and many will die in the alkaline duodenal juice. The ones that survive need to be able to get their food, multiply, and stick to the intestinal cells and stay there, which doesn’t happen.