#458 - [Journal Club] - 📌 Is this the trial that brings probiotics back to the NICU?

Probiotics left US NICUs after the FDA reclassified them, and NEC rates went up. This week Ben reviews the Connection Study, a 2,158 infant phase 3 trial of IBP-9414, a live biotherapeutic product containing Limosilactobacillus reuteri, run under an IND with the regulatory rigor probiotics never had. Neither primary endpoint reached significance, but all cause mortality fell 27%, and surgically confirmed NEC favored treatment after 14 days of dosing. The discussion may be the real story, including a kappa of 0.36 between radiologists reading the same NEC films. Ben and Daphna talk through what a regulated probiotic would mean for the babies waiting now.

----

Live biotherapeutic product IBP-9414 (L. reuteri) in very low birth weight infants: the Connection Study. Neu J, Del Moral T, Guthrie SO, Hudak ML, Indrio F, Kim JH, Kronström A, Martin CR, Modi N, Rastad J, Schnitzer TJ, Singh R, Strömberg S, Szajewska H, Thuresson M, Caplan M.Pediatr Res. 2026 Feb 20. doi: 10.1038/s41390-026-04826-7. Online ahead of print.PMID: 41721054

Transcript

Ben Courchia, MD (00:00.72) Hello everybody, welcome back to the Incubator Podcast Journal Club. It is Wednesday, I believe, and we are going over more recent papers. This is kind of an old paper that I've been deferring until now.

Daphna Yasova Barbeau, MD (00:05.902) Yeah.

Daphna Yasova Barbeau, MD (00:14.55) Okay. Well, it's summertime. Everybody's catching up. Okay. Okay, so what made you pull all the way to the recesses of January? Why this paper? That's true.

Ben Courchia, MD (00:27.536) Because I felt like if I didn't do it now, it was never gonna get done this year. Yeah. This is actually a paper on probiotics. It's published in, my god, Pediatric Research. And it's called "Live Biotherapeutic Product IBP-9414 (Limosilactobacillus reuteri) in Very Low Birth Weight Infants, the Connection Study." First author, Josef Neu, who you're familiar with, one of my former

Daphna Yasova Barbeau, MD (00:34.829) Mm-hmm.

Daphna Yasova Barbeau, MD (00:52.334) Mm, very much so.

Ben Courchia, MD (00:56.454) Teresa Del Moral is second author. Cammy Martin's on there. Mark Hudak is on there. Lots of big, big names. Big players. Actually, since we're talking about probiotics, I just want to give a plug. Our good friend Jen Canvasser is organizing the NEC Family Summit, which is going to be taking place September 1st to September 4th, 2026,

Daphna Yasova Barbeau, MD (01:02.84) So big names, big players, yeah.

Daphna Yasova Barbeau, MD (01:10.99) Mm-hmm.

Ben Courchia, MD (01:24.506) going to be taking place in Davis, California. This is truly what they're preaching about, involving family in how we can build a world without NEC. I highly encourage everyone to go check out the NEC Family Summit for registration and information on necsociety.org. So

Daphna Yasova Barbeau, MD (01:43.803) And then their biennial, the next symposium in 2027. Yeah.

Ben Courchia, MD (01:47.472) That's right. That is right. So this paper is a very interesting study, especially in the ongoing conversation about probiotics. I think that what prompted me to go back and review this is the fact that you reviewed data a couple of weeks back on the fact that since probiotics have left the scene, specifically in the US, we have had increased

Daphna Yasova Barbeau, MD (02:10.721) Right.

Ben Courchia, MD (02:16.631) rates of NEC. And I think that this is where this paper picks right up. Following instances of confirmed translocation of bacteria with the use of probiotic agents, warnings of probiotic induced sepsis have been issued by North American health authorities. Unlike probiotics, live biotherapeutic products, LBPs, are developed under regulatory oversight requiring rigorous clinical trials and manufacturing controls to document their safety, efficacy and purity prior to marketing approval.

Regulation of these products aims to increase their quality and help prevent serious consequences of bacterial contamination while preserving the widely documented benefits to vulnerable preterm infants. I think this is a key part of the introduction, because it is a legal distinction more than a quality one. The probiotics that we were using until the FDA came down on them were regulated as foods or dietary supplements, which means that really nobody had to prove, before they reached the baby, that

Daphna Yasova Barbeau, MD (02:57.293) Mm-hmm. That's right.

Daphna Yasova Barbeau, MD (03:03.521) Yeah.

Ben Courchia, MD (03:15.545) what was on the label was actually in the packet, right? That verification really did not need to be done, or that it was made to pharmaceutical standards. Live biotherapeutic product is basically the consequence of this FDA memorandum on probiotics, which is that probiotics are now going to be regulated as a biologic. So you'll need an investigational new drug application, an IND as it is commonly known,

just to study it in humans. You'll need a license to market it. And the sponsor has to submit full genomic characterization of the strain, its antibiotic susceptibility profile, a defined cell bank, a validated potency assay, et cetera, et cetera. And while it doesn't really rehabilitate probiotics as a category, it is the way in which probiotics are going to come back in the US, where each individual strain is going to need to go through this.

So the IBP-9414 is a candidate that contains live Limosilactobacillus reuteri, formerly known as Lactobacillus reuteri. Here they present the results of a prospective randomized placebo controlled phase 3 trial comparing the efficacy and safety of IBP-9414 with placebo in very low birth weight infants.

So the Connection Study is a multicenter prospective double blind placebo controlled randomized phase 3 trial. It was conducted under the US investigational new drug and EU clinical trial exemption oversight. The primary outcome measures were reduction in the incidence of necrotizing enterocolitis and the time to sustained feeding tolerance, SFT, as it is known throughout the paper.

With secondary outcomes including all-cause mortality. The trial was conducted in 10 countries. In terms of the patients, the eligibility for randomization included birth weight of 500 to 1500 grams, gestational age at birth of 23 to 32 weeks, and being less than or equal to 48 hours of life, and consent from.

Ben Courchia, MD (05:29.699) The first 300 infants had birth weight between 750 and 1000 grams. At a pre-planned 300 infant safety evaluation, the recruitment of participants was broadened to include smaller babies, birth weight 500 to 749. Enrollment of infants with birth weight 1000 to 1500 was open next, with a pre-specified restriction on this group to only 10% of the study population. So they wanted to include that category as well, but limit how much it would compose of the entire cohort,

in order, in my opinion, not to skew the data too much. After each dosing of this... I'm gonna call this probiotic. I know that they say that we have to call this whatever, live... what is it? Live...

Daphna Yasova Barbeau, MD (06:10.485) I think we can all agree on the lingo for today. Yeah.

Ben Courchia, MD (06:13.162) All right, thank you very much. The live biotherapeutic product. But basically after

fine. So the enrolled infants received a single daily enteral dose of 1 x 10⁹ colony forming units of powdered IBP-9414, reconstituted with sterile water, or sterile water placebo, within 48 hours of birth, and continued receiving the daily dose until they reached 34 and six weeks of gestation. After each dosing, the NG or OG was flushed with breast milk,

formula, air, sterile water, basically whatever the investigators wanted at that time. Follow-up was conducted at 40 weeks postmenstrual age. The trial was closed when 2,158 infants had been randomized. This is a large, large study. The study dose, body weight, total enteral feed volume, and administration of any human milk, any parenteral macronutrients, and concomitant medications were recorded daily during the treatment period.

They had a randomization schedule, very rigorous. One-to-one ratio. Treatment assignment was blinded for the investigators and any personnel involved with the study, and so that's important as well. They had safety assessments at 300, 600, and 1400 participants completed treatment. They had an independent DMC convened to assess the unblinded data for safety and futility.

In terms of efficacy endpoints, the primary endpoints of the study were the incidence of NEC and time to reach SFT, sustained, what did it say, sustained feeding tolerance. So, full feeds, I guess, would be another way of putting it, but we're gonna define it. So, NEC, and that's very important. So how do we diagnose NEC? And I want you to pay attention to the surgical versus the medical, because the surgical.

Daphna Yasova Barbeau, MD (08:13.996) Paying attention.

Ben Courchia, MD (08:15.148) Yeah, I'm talking to the audience in general as well. I know you're paying attention. But the idea here being that surgical is not going to be so controversial, because you go in and there's lots of information you gather during that process, but it's the medical that's gonna be interesting. So NEC was defined in one of two ways. Either number one, independent evaluation of abdominal radiographs obtained due to investigator perceived clinical signs of NEC, by two radiologists unaware of the infant's clinical course,

that identified findings of pneumatosis intestinalis and/or portal venous gas. A third radiologist could resolve any disagreement. Or it could be diagnosed via pathway number two, which is direct documentation of intestinal necrosis on laparotomy. So that's less subjective. The time to sustained, my god, I keep forgetting, sustained feeding tolerance. So what do they define sustained feeding tolerance as?

It was defined as the time to the first day of the first 10-day period during which the baby received more than 120 ml per kilo per day of enteral feeds without parenteral nutrition, and that their mean daily body weight increase was more than 10 grams per kilo. They had secondary endpoints that they looked at, including the incidence of NEC confirmed by radiograph, NEC confirmed by surgery/autopsy. They looked at all-cause mortality, weight gain during the postnatal weeks three to four,

duration of hospitalization, days with investigator reported signs of feeding intolerance, et cetera. A pre-specified sensitivity analysis of NEC excluded infants with an adverse event during the first 14 days. They're gonna bring up this idea of the first 14 days, because it's important. They're gonna make the case saying, hey, if it happened in the first 14 days, could we attribute that to a failure of the probiotic?

Because they say there's a time period where you need to colonize the gut, and there's a process, and maybe 14 days. They've elected to use 14 days based on prior data, to say if it happens in the first 14 days, maybe the probiotic didn't really have a chance to do very much. And so maybe we should exclude them. And that's why they did this pre-specified analysis. And you'll see that as I'm gonna go, I'm gonna skip the statistics. I mean, we're

Ben Courchia, MD (10:35.339) they did a modified intention to treat and they have a good power calculation. I'm gonna skip that. It's a very rigorous study. There's no funny business here. But it's important to look at that, because everyone is gonna pick apart this study and try to look for what they want. So without further ado, let's go into the results. So in total, 2,158 infants were randomized, of whom 2,117 received at least one dose of the study drug

and comprised the modified intention to treat efficacy population. The comparison and baseline characteristics between the intervention group receiving the probiotic and the placebo were well balanced. The median birth weight was 850 grams versus 860 grams. The median gestational age, 27 weeks in both groups.

Infants were treated for a median of 49 days and received human milk on a median of 95% of treatment days. Kudos to these hospitals for such a high rate of breast milk exposure. And let's get into the primary endpoint. So the incidence of necrotizing enterocolitis. The efficacy analysis found that the incidence of NEC, confirmed by radiograph or surgery/autopsy,

was 8.7% in the probiotic treated infants versus 10.2% in the placebo treated infants. While the difference is slightly higher in the control group, this did not reach statistical significance, with a relative risk of 0.85, a 95% confidence interval of 0.66 to 1.11, and a p-value of 0.24.

A pre-specified sensitivity analysis of this primary endpoint focused on events occurring only after 14 days of treatment. And we talked about that, that's one of the pre-specified analyses. And it showed a 0.74 relative risk of NEC with the probiotic. Again, this did not reach statistical significance, with a p-value of 0.067. Almost, but not there.

Ben Courchia, MD (12:57.671) Treatment effects as a function of time demonstrated a divergence in NEC incidence between the probiotic and the placebo group after 14 days by Kaplan-Meier curve. What about NEC confirmed by surgery or autopsy? So the pre-specified secondary endpoint of NEC confirmed by surgery/autopsy, which is unequivocal histopathological confirmation, demonstrated a 0.8% incidence of NEC

in the probiotic group compared to 1.4% in the placebo group. Again, a difference favoring the probiotic, but that did not reach statistical significance. Relative risk of 0.52, a p-value of 0.126. When such events occurred after 14 days of treatment, there was a relative risk of 0.29, and that was statistically significant with a p-value of

with the probiotic treatment. A sensitivity analysis examining the effects in infants who are small for gestational age, SGA, and across multiple subgroups based on birth weight, gestational age, sex, and method of delivery, demonstrated similar results. In terms of time to sustained feeding tolerance,

the time to sustained feeding tolerance was achieved at 21 days in the probiotic group and at 23 days in the placebo group. Not statistically significant.

Mortality. The probiotic reduced mortality, a key secondary endpoint, with findings demonstrating an incidence of 6.2% in the probiotic group compared to 8.5% in the placebo group. This was statistically significant, relative risk of 0.73, a confidence interval that goes from 0.5 to 0.98, and a p-value of 0.036. The mortality advantage of the probiotic was stronger

Ben Courchia, MD (14:59.621) when evaluating death after 14 days of treatment, with a relative risk of 0.5, a p-value of 0.007. A positive treatment effect of the probiotic was also observed for investigator reported death attributed to NEC or intestinal perforation, with a relative risk of 0.53 of such death with treatment. Again, p-value of 0.07. The relative

risk of death by NEC or intestinal perforation dropped to 0.39 when considering events occurring after 14 days of treatment. So when we are looking at mortality specifically, table four is really where you want to focus your attention. All cause mortality, excluding infants dying during the first 14 days, significantly less deaths in the placebo group,

2.2 versus 4.2%. When we are looking at deaths by NEC or intestinal perforation, excluding infants dying in the first 14 days, again, 0.5% with the probiotic, 1.3% with the placebo, p-value of 0.05. What about babies who were very small, born less than 1,000 grams? These efficacy endpoint findings were consistent among infants with birth weight

less than a thousand grams. In this subgroup, the relative risk of NEC confirmed by radiograph or surgery/autopsy was 0.87 with probiotic treatment. And time to sustained feeding tolerance was 23 versus 24 days between the groups. All cause mortality incidence was 6.9% in the probiotic group versus 9.3% in the placebo group, and that was statistically significant. So

a very interesting paper, because we keep saying that we need our probiotics. And so now we say, okay, what about the rates of NEC? So I want to dive into the discussion, because it's very important. In terms of mortality reduction, the authors in the discussion do mention that the group treated with IBP-9414 demonstrated a significant 27% reduction in all-cause mortality compared with the placebo group. And this reduction represents a number needed to treat of about 44, meaning that widespread use of this particular

Daphna Yasova Barbeau, MD (17:01.196) Yeah.

Ben Courchia, MD (17:24.488) probiotic could save more than a thousand patients annually in the US alone, which is a very striking statistic. So how do we reconcile the fact that mortality was improved, mortality due to NEC, but somehow the rates of NEC, we had such an issue teasing that apart? They say, and I quote, it was evident during the course of the study that there was a large rate of discordance between the two radiology reports submitted for each diagnosis, with an interrater reliability, a kappa of

Daphna Yasova Barbeau, MD (17:30.594) Mm-hmm.

Ben Courchia, MD (17:54.169) 0.36, which is considerably lower than what is generally accepted as adequate, usually above 0.6. This finding suggests that many of the infants who were reported as having NEC may have been misdiagnosed. Including such infants in the analysis of overall NEC makes demonstrating the effect of IBP-9414 much less likely, highlighting the insufficient diagnostic accuracy of radiographic diagnosis of NEC, even with controlled parameters.

This is likely to be inherent in any prospective randomized trial using current ambiguous definitions and criteria for diagnosing NEC, which goes back to some of the papers that we reviewed on the podcast about maybe introducing abdominal ultrasound as an additional parameter to try to help with that.

Daphna Yasova Barbeau, MD (18:35.566) People can pay attention to tomorrow's paper.

Ben Courchia, MD (18:40.764) That's right. That's right. About using broadband. In terms of the rationale for the 14 day time point, I'm not gonna, it's been 18 minutes, I'm not gonna get into that, but in the discussion is where they talk a little bit about the rationale for saying, hey, we need to wait 14 days to get a sense. The analysis focusing on data starting after 14 days of life was pre-specified for the primary NEC outcome, but was also performed post hoc for all other NEC and mortality outcomes.

Daphna Yasova Barbeau, MD (18:44.76) That's right.

Ben Courchia, MD (19:10.908) The rationale underpinning this approach was to accommodate the known time delay in microbial colonization of the gut, as clinically effective IBP-9414 colonization of the immature intestinal tract of preterm infants is anticipated to develop during the initial weeks of dosing. Similar to findings observed with other prophylactic treatments, such as vaccines for COVID-19 and statins for cardiac disease prevention, the pharmacodynamic effects of IBP-9414 on intestinal immunocompetence and function

are expected to require some time to develop, something that I was alluding to earlier in the discussion. The other thing that they addressed was the composition of the primary outcome. They said that radiograph confirmed and surgically/autopsy confirmed NEC cases were combined in the primary endpoint, as there were likely to be a smaller number of surgically/autopsy confirmed NEC.

And no data exists in the literature to determine an appropriate sample size for this endpoint alone. Thankfully, by the way, because can you imagine if that was so prevalent? In fact, they say 88% of the infants with NEC were diagnosed clinically, requiring investigator reported clinical signs or symptoms and confirmation by radiographic findings, which is, I think, most of our experience, thankfully. Thus, the analysis of overall NEC events was almost entirely weighted by this group and highly dependent

on the validity of the radiographic confirmation of NEC signs and symptoms, which was shown to be unreliable. The additional stricter metric for evaluating the impact of the probiotic on NEC incidence, such as the surgically/autopsy confirmed NEC endpoint, points to a favorable treatment effect, especially after 14 days of treatment. I'm going to mention two more things. One is the adverse events. And the one thing that I think people may have on their mind is that the active ingredient

was not detected in any blood cultures performed during the study using standard hospital procedures capable of detecting L. reuteri. Obviously this is an important point, because this is what is leading us to where we are today. And so, yeah. So the conclusions of the authors are that although the primary endpoints did not reach statistical significance, IBP-9414 treatment resulted in a clinically relevant reduction in all cause mortality and in the incidence of surgically

Ben Courchia, MD (21:30.424) autopsy confirmed NEC. Safety profiles, including the risk of sepsis, were similar between the two groups. These results substantiate the potential of IBP-9414 treatment to support vulnerable, very low birth weight infants and improve their outcomes. I'm done.

Daphna Yasova Barbeau, MD (21:47.917) Ooh, but such an important paper. And I think the discussion was really valuable, especially because of the stars on the paper, right? So hearing their thoughts about it is useful.

Ben Courchia, MD (22:02.076) Yeah, I think the discussion is often a marker of the authors' courage, right? It could either be an author trying to put it together to make it look nice, or it could be authors really tackling the hard questions that their data raises, which in this case is exactly what they did.

Daphna Yasova Barbeau, MD (22:06.946) Right.

Daphna Yasova Barbeau, MD (22:10.35) Mm-hmm.

Daphna Yasova Barbeau, MD (22:14.926) Mm-hmm.

Agreed. Agreed. Yeah, they know what everybody's gonna ask about, so they're able to predict what to discuss ahead of time. But I think what's really interesting is, I think the data is coming out about the impact on NEC, but it's interesting that it's this reduction in all cause mortality, right? And we think it's mostly NEC, right? That we're keeping babies from dying because of NEC.

Ben Courchia, MD (22:24.651) Mm-hmm.

Daphna Yasova Barbeau, MD (22:47.64) But there are potentially these other, the gut microbiome impacts every organ system. So I think I'm hopeful that there will be even more studies that are looking at the other long term comorbidities of these very little babies and the use of probiotics. So it's really exciting, because

maybe, maybe we can take all this information and convince the FDA to have a product for babies.

Ben Courchia, MD (23:23.416) I believe that this is precisely why this paper was designed the way it was designed. That's why it's going through an IND, because this might very well be the first probiotic that's gonna be FDA approved in the US. And we might all end up having that as the only option for our patients. So

Daphna Yasova Barbeau, MD (23:27.202) Yeah.

Daphna Yasova Barbeau, MD (23:31.544) Right.

Daphna Yasova Barbeau, MD (23:45.55) And some standardization of those products is not a bad thing, you know. I don't think anybody's arguing that we should just have over the counter, you know.

Ben Courchia, MD (23:50.916) It's not a bad thing. It's not a bad thing, but I do believe that it does take a long time. Clearly you can see this is a lot of work. And in the meantime, there are some babies who would definitely benefit who are not getting it. So

Daphna Yasova Barbeau, MD (24:10.703) Absolutely. Yeah, I think that's the hardest thing. Every, every case of NEC, you say, gosh, if this baby had been on probiotics, could this have been avoided? That's tough to watch and to think about on a day-to-day basis. Okay, buddy. Sounds good.

Ben Courchia, MD (24:24.285) Yeah, agreed. Agreed. All right, see you tomorrow.