All right, well, it is my great honor as the president of the Arthroscopy Association of North America to introduce the Anna Innovations Lecture and this year's presenter. Specialty Day has been integral to Anna's educational program each year, and recently the Board of Directors created an annual lectureship that recognizes a leader, an innovator who has been instrumental to the development of our craft and who has advanced our field. Steve Burkhart was selected as the inaugural lecturer, and his presentation was initially scheduled to be delivered at the 2020 annual meeting in Orlando. Unfortunately, the COVID pandemic prevented that meeting from happening, and his presentation was delayed. We were fortunate to have Dr. Burkhart present his lecture at the AOSSM Anna Combined meeting in Nashville last year, and he set the bar really high, and we are excited to hear future innovation lectures at each Specialty Day moving forward. Being president of Anna has been the highlight of my professional career, and developing this program and working to make a real difference for diversity and inclusion has been a cornerstone of my term. Anna is celebrating its 40th anniversary year, and in honor of that, we are developing a 40th anniversary monograph edited by John Ticker with co-editors Mike O'Brien and Jessica Ryu. It will be released at the annual meeting in May. Creating and reviewing the monograph has allowed me to reflect on how we arrived at where we are today and where we strive to be in the future. This collaborative session is a great step in that direction. I'm reminded of the many influential pioneers of Anna, the past presidents in innovation and advancement in technology that were instrumental to grow Anna to 6,000 members today. As Dr. Cole has stated, our goal with this innovation lecture is to highlight and reinforce the innovative nature that many surgeons inherently possess and to underscore the significant contributions that Anna surgeons and others have made to arthroscopy and related surgery over the past 40 years. There are leaders, mavericks, and innovators that stand out in orthopedic history, and there are those who are making history as we gather today, like Dr. Martha Murray. In honor of a historic collaboration with the Ruth Jackson Orthopedic Society, we are honored to have a pioneering innovator who, fittingly, is female, to give today's 2022 lecture, the first to be delivered in person on specialty day. Dr. Murray is the orthopedic surgeon-in-chief at Boston Children's Hospital and professor of orthopedic surgery at Harvard Medical School. She has a bachelor's in science from mechanical engineering from the University of Delaware and a master of science in material science and engineering from Stanford. She received her doctor of medicine from the University of Pennsylvania and then completed her orthopedic surgery residency and fellowships in both pediatric surgery and sports medicine at Harvard Medical School. It is fair to say that treating ACL injuries is commonplace for most of us in this room today. Dr. Murray has spent her career thinking about and scientifically studying the ACL, and as we heard earlier this morning, she is a pioneer. She has successfully translated her passion and research involving new ways to treat ACL injuries from the lab to FDA marketing approval with grant funding from several sources, including the NIH and the NFLPA. She has 11 issued patents and over 100 peer-reviewed publications around the development of bridge-enhanced anterior cruciate ligament region repair. This technology is excellent, and her outstanding work on innovative ACL treatment has garnered great respect, and she has won the prestigious AOS Kappa Delta Award as well as the AOSSM O'Donoghue, Cabal, and Grana Awards. At this very meeting, she is also the recipient of the OREF Clinical Research Award. Dr. Murray is a surgeon, an educator, an entrepreneur, and an outstanding and award-winning innovator. Ladies and gentlemen, please welcome the 2022 Anna Innovations Honoree, Dr. Martha Murray. Thank you very much for that very kind introduction and for the invitation to speak today. So today I'm going to talk about, can rocket science help orthopedic innovation? These are my disclosures. We'll go through some of these in more detail in the presentation. So when we think about surgical treatment of ACL injuries, we typically started with repair back in the day. That's how we used to try to treat these. But unfortunately, this had a 50 to 90 percent failure rate, and the results were found in randomized controlled trials to be no better than non-operative treatment. So repair has been largely abandoned now in favor of what we all know as ACL reconstruction. The outcomes of ACL reconstruction are good, but they're not perfect. So patients still experience some pain and weakness, typically from where the graft is taken. A return to pre-injury level of activity is only around 65 percent. There's an elevated risk of getting osteoarthritis of the knee, whether you have surgery or not, whether you have a reconstruction or not. And interestingly, there's also a 50 percent increase in the risk of myocardial infarction later in life for people who have an ACL tear as a young person. So we wondered, was there something we could do differently? Could we innovate in some way to make this better? So what we did was we went back and tried to figure out what is the basic science behind why the ACL doesn't heal. And we went over this this morning, so I'll go more quickly through this. But basically, the MCL tears, MCL ends bleed, they form a nice clot between the torn ends that the ligament can grow back together through. In contrast, when the ACL tears, the ends bleed, but the blood turns into a hemarthrosis rather than a clot between the torn ends. So there's never that provisional scaffolding that can help the ACL to heal. So with that understanding, we thought, well, what if we could put something in the gap? What if we could fill that gap with something that would induce healing? Could we restore the ACL? So in collaboration with Kurt Spindler and other folks down at Vanderbilt, we worked on a central defect model in the ACL, where we made a central defect in the ligament, and it was supported biomechanically on both sides by intact fiber. So this is a mechanically stable defect. And what we found was when we put a mixture of collagen and blood in that defect, we could induce healing of the defect, see it both histologically as well as biomechanically. So we thought, great, you know, we're doing awesome. We've got this basic science. We made this prototype. And this is what we do in academics, right? So we've got this idea. I thought, we'll go to industry. They love this, right? They want to take this now, and they're going to want to run with it and get it through to commercialization. We are all set. But industry didn't really feel that way. They were like, we want to be over here, right? We don't want you where you are. We want you to get through preclinical assays and the FDA. So we were kind of facing this valley of death. We don't know how to do all that. So what we did was we started a startup company. So this was a company that we founded in 2007. We hired an awesome CEO. We raised $5 million in a Series A. We thought, we are just killing it. It closed its doors three years later, right? So what we learned from that was that startup companies are challenging. $5 million isn't a lot of money to start a company. And definition of our regulatory pathway is critical, which I'll speak more to in a minute. But we were still stuck facing this valley of death. So we thought, how do other fields handle innovation? And we thought, well, gosh, how does NASA do it? I mean, they put a rocket on the moon. Like, that is crazy. So it turns out that NASA actually has a very organized way of handling innovation. They have these nine technological readiness levels that they work through from basic principles through to optimization and all the way down to the product is used routinely. And the key parts of each phase when they do this is they have a couple things they do in every phase. So the first thing they do is when they start a phase is they develop the team. They then design an experimental plan that's pre-specified before they start that phase. And they decide what success will look like before initiating any of the experiments. They then generate the data and then all get together at the end to review the data as a team to decide, did we knock it out of the park for this phase or do we need to go back to the drawing board and do this phase again? So when we think about this in terms of orthopedic innovation, when we think about the technological readiness levels, basic principles, a possible solution, and proof of concept. So one, two, and three, we're pretty used to doing in academics. The last two levels, so when we actually do manufacturing of a proven product or the products used routinely, we also do pretty routinely for it by industry or in clinical trials. So where we were kind of stuck was in these middle phases. How do we optimize the product? How do we test it in a model environment? Can we build an actual prototype that's the same thing we would use in humans? And can we then test it in the real environment or the human knee? So what we wanted was, could we do these in academia? So rather than relying on industry to take this over, could we do these in the lab? So we were here at proof of concept. So our next phase was this optimization. We wanted to optimize both the scaffolding that we'd put between the torrents, as well as what biologics should we add to that scaffold to stimulate healing? And then how do we stabilize the knee while healing's occurring? So to do this, we developed a team. We had a team of surgeons, material scientists, a hematologist, a biomechanics expert, a biologist, and a statistician. And our plan was to optimize the best option in vivo. And then our pre-specified primary outcome was whichever repair had the best maximum load when we tested it. So optimizing the scaffold, first off, we found a lot of things that didn't work. So we found that if we put platelet-rich plasma, a clot of that, in between the torrent ends of the ACL, that didn't work at all. If we used a collagen scaffold by itself, that didn't work either. And if we used a gel, as we used in our early experiments for a full defect, that also didn't work. It just kind of flowed out of the knee. So what did work best was a combination of using platelets and collagen and using a large sponge-like scaffold to hold that biologic in the right place so it could do its thing between the torrent ends of the tissue. We then wondered, all right, what do we put in this scaffold to make things heal? And we worked to optimize the biologic component. And what we found was that concentrating platelets into PRP wasn't necessarily better, and that platelets and plasma were both important, as were red blood cells were important in getting collagen to be made in the defect. So we basically came to the conclusion that all parts of blood play a role. We then wanted to optimize the early stabilization of the knee, and we found that bone-to-bone fixation using a simple polyester suture actually worked really well to stabilize the knee. But if we attached the suture in the notch with a suture anchor, about 20 percent of the time in our animal studies, that pulled out. So we moved to actually making extra cortical buttons on the front of the tibia and on the side of the femur rather than having something intra-articular. So basically, this was the prototype that we came up with after optimizing the different phases. So it's a collagen scaffold that's a little bit bigger than my thumb, so it's pretty big. It's larger than the native ACL. We use whole blood for our biologic stimulus. We use a non-absorbable suture to help initially stabilize the knee, which goes from an extra cortical button on the femur to one on the tibia. So now we've optimized our implant, now we want to test it in the model environment. So for this, we developed a different team, because now we're going to move to large animal in vivo surgery. So now we need a surgeon, we need a veterinarian, we need a still biomechanics expertise, but now we're starting to need manufacturing expertise as well as radiology and pathology to help us understand our results. And for this phase, our pre-specified outcome that would mean success was that the mechanical properties of our repair would be similar to that of an ACL reconstruction in this large animal model. And we also wanted to make sure that our implant wasn't going to cause any significant adverse reaction either for the synovium or for the cartilage. So we developed a fit-for-purpose model in collaboration with Braden Fleming down at Brown, and we wanted an animal model that had a similar size to people. Doing a mouse ACL is pretty hard, so we wanted something bigger. And it turns out that pigs also have very similar blood components to humans. So the pig had relevant size and anatomy, relevant hematology, very similar biomechanics. And in that model, we were able to validate that if we cut the ACL and didn't do anything, an untreated ACL transection would not spontaneously heal, similar to people. If we did a suture repair alone, the results of that were inferior to if you do an ACL reconstruction. Again, similar to what we find in our human patients. And probably most importantly for us was that an untreated ACL transection led to the development of post-traumatic osteoarthritis in a very similar pattern to what we see in our human patients. So in that model, we could do a large animal study where we could cut the two ACLs, and on one side we could just do a suture repair, and the other side we could do a suture repair plus our platelets in the sponge. And then we could look at basic outcomes. And this slide I also showed this morning, you can see that in an intact pig ACL, it's going right through the notch, just like in people. If we did a simple suture repair without any biologic augmentation, the sutures tear and the ACL resorbs and we don't have anything left in the notch at three months. But if we do a suture repair and we combine that with our biologic sponge that's holding the platelets there, we could get the ACL to heal. And when we look at the mechanical properties of that tissue, it actually looks pretty good. So the mechanical properties of that healing tissue is much better than if we just do a suture repair alone. And it's also equivalent to what we see when we do an ACL graft in this animal model. When we look at the mechanical properties one year later, we find similar findings that the mechanical properties of the repaired ACL are very similar to that of a graft in this model. Lastly, we also looked at the cartilage surfaces. So as I mentioned previously, in a porcine ACL, if you cut the ligament and let the animal walk around for six months, they develop this breakdown of the articular cartilage, particularly in the medial part of the joint, similar to what we see in human patients. When we do an ACL reconstruction, we have similar findings of changes, particularly in that medial side. But when we did the bridge-enhanced ACL restoration procedure, we actually did not see those osteoarthritic changes. So this was very surprising to us, but it was nice. So now we're at this phase, we're in the model, are we going to go or are we going to no-go? So we've got mechanical properties that are similar to an ACL reconstruction, we didn't see any significant adverse reaction to the scaffold or the implant, and we had less osteoarthritis. So we said, okay, let's go, let's go to the next phase. So now we have to make a prototype. So now we're talking about building a functioning prototype that's something that you could actually put into a human being, and that's a very different thing than just putting something into an animal. So now we needed a project manager, we needed manufacturing expertise, we needed regulatory expertise, we needed a quality consultant, and we needed help from the FDA. Our plan for this phase was then to verify that our manufacturing process produced the same thing every time. So each lot looked the same, each scaffold within a lot looked the same, and we wanted to verify the biocompatibility and sterility of the product. So our pre-specified outcome for this phase was that all the scaffolds in each lot would meet these lot release criteria, and they would be found to be biocompatible. We put together a manufacturing facility for these early clinical implants. We could make about 100 implants a year. This is just a room in our lab that we outfitted to do this GLP manufacturing with design control. We were able to verify the biocompatibility, packaging, and sterility using outside consultants as well, and we had some key internal personnel who really ran this for us. So with that data, we went to the FDA. This is our FDA application for our first in-human study, and we said, can we do this in people now with all this data? We submitted that in August of 2014, and 29 days later we got the approval letter from the FDA saying we were a go. So with that, now we've got to test our prototype in the actual environment. So now we're starting to talk about something called an investigational device exemption surgical trial. So to run this, this is a trial that's run in conjunction with the FDA, where they're overseeing it as well as our IRB and the other normal study things that you would have. So we need a clinical research nurse. We needed study monitors, statisticians, research administrators, data safety monitoring board, research coordinators, and then we also needed a clinical team that was willing to try this new procedure for the first time. And so this was surgeons, anesthesiologists, radiologists to help with the MR, OR staff, etc. So our plan for these trials was to do a small trial first, to just do 20 patients, 10 patients with the investigational procedure and 10 patients with standard. And our pre-specified outcomes for this was really to look at safety in that first trial, to make sure there weren't going to be any really bad things that were happening there. And then in our second trial, we were planning if the first trial looked okay, we would move to a larger second trial and begin to look at efficacy. So with that, we opened what we call the BEAR-1 trial, so Bridge Enhanced ACL Restoration or BEAR. So in this study, Dr. Lyle Michele was the surgeon for this study. This was a cohort study. So we had 10 patients that had the BEAR procedure, 10 patients with ATL reconstruction with autographed hamstring. These patients had a complete mid-substance ACL tear. They were 18 to 35 years of age, and they had to be within 30 days between their injury and surgery to qualify for the BEAR group. The patients turned out to actually be fairly similar in both groups, even though this wasn't an RCT. In age, the average age was 24 years of age. Their mechanism of injury was primarily non-contact, and their average MARC score was 12. These are just showing the 10 MRIs for the patients in their injury film. So this is just showing that these were mid-substance ACL tears. So the results of this study showed that we had no infection, rejection, or failure. There was no gap formation on our post-operative MRIs, so they all had a continuously healing ACL. And at two years, the patients that had BEAR had good patient-reported outcomes, they had good knee stability, they had better hamstring strength than the ACL reconstruction group that had a hamstring autographed, and we had better restored the ACL anatomy. So this was good, but we're at Boston Children's Hospital, and we worry a lot about the adolescents. We all know that these patients are at high risk for graft failure, high risk for post-traumatic arthritis. So with that in mind, we opened the BEAR 2 trial, and in this trial, we were able to get the FDA to drop the minimum age for the trial to 14. So Dr. Yen was the PI on this study, and this was designed as a double-blind, randomized control trial. We had three surgeons on the trial, 100 patients. We did a two-to-one enrollment with 65 patients having the bridge-enhanced ACL restoration or the BEAR procedure, 35 having autographed ACL reconstruction. We opened the trial in May of 2016, we were fully enrolled a year later, and we were able to get two-year follow-up on 99 of the 100 patients. So in terms of the BEAR 2 outcome measures, these were our pre-specified primary outcomes. So we worked with the FDA, we got this all approved, yes, this is what it was going to be. So we had co-primary outcomes, and we had to show non-inferiority of both patient-reported outcomes in terms of an IKD subjective score, as well as KT1000 testing for instrumented apulaxity of the knee. In this trial, not surprisingly, as it was a randomized control trial, we had very similar demographics in the two groups. Median age in this trial was 17 years. Gender was almost evenly split between men and women. And then the median marks activity score was 16, which, again, is the maximum score on this. So very young, active cohort. The results in this trial, the most important thing for us really was still safety. And fortunately, we didn't have any infections and no immune response or allergic reaction to the scaffolding. And no scaffolds needed to be removed. In terms of our primary outcomes, patient-reported outcomes at two years, again, our first primary outcome that was pre-specified was a patient-reported outcome of IKDC subjective score, where age-matched norms have a mean of 88. The ACL reconstruction group came in at a mean of 85. Bayer patients came in at a mean of 89 and was found to be not inferior, thus meeting our first primary outcome. When we looked at KT1000 testing, uninjured knees typically have less than 1 millimeter side-to-side difference. Other trials for ACL reconstruction show a range of 1 to 3. This came in for the ACL reconstruction group with a mean of 1.8, consistent with those prior studies. Bayer came in at a mean of 1.6 and, again, was found to be not inferior to ACL reconstruction, thus meeting our second primary endpoint. Hamstring strength at two years out, the ACL reconstruction group, which was largely autographed hamstring, had a persistent 60 percent deficit, and the bridge-enhanced ACL repair patients recovered their strength. How about retear rates? So we've spoken a lot today about retear rates, particularly in adolescents. If we look at primary repair, simple suture repair, the Gagliardi study would say that the failure rate is about 49 percent for that. In MOON studies and other studies, 15 to 20 percent is the typical retear rate for ACL autographed. In the Bayer II trial, our retear rate for the ACL autographed reconstruction group was 6 percent. Bayer patients were 14 percent. So with that, we also wanted to look at what was the effect of the second ACL surgery. So compared to primary ACL reconstruction, we know that revision ACL reconstruction has inferior outcomes, so lower IKDC scores, lower Kuh scores, lower Marx activity levels. And what we wanted was for this 14 percent of the patients, did the same thing happen for them? So if you had a Bayer and then retear and had to have a graft, would you also have a worse result than if you just had a primary procedure? When we looked at this, we found augmented repair, then ACL reconstruction, and the IKDC scores, the mean in this group was a mean of 85 points, which was very similar to if they just had a single ACL reconstruction. And when we look at second ACL surgery results and we compare revision ACL reconstruction, so two grafts versus Bayer, then a graft, we can see that the median IKDC scores in the Bayer to ACLR group are much more similar to a primary ACL reconstruction than they are to a revision ACL reconstruction. When we look at KT1000 testing, augmented repair, then ACL reconstruction, again, similarly to what we saw for the subjective results, when you have a Bayer that has to then go to an ACL reconstruction, the apilaxy to the knee looks very much like a primary ACL reconstruction rather than a revision ACL reconstruction. What about the effect of sex on augmented repair? Now we know that women after ACL reconstruction have been reported to have lower Kuh scores, and particularly this relates to quality of life. They have greater apilaxy at two years than men. They have a greater persistent strength deficits as well. And so what we hypothesized was that females would have inferior outcomes after augmented ACL repair, just like they do after ACL reconstruction. And we looked at patient-reported outcomes as well as apine laxity and muscle strength. So in this study, we looked at 65 augmented ACL repair subjects in the Bayer II trial. This was 28 males, 37 females. The women were younger and, in general, had a lower BMI, but there was no difference between the sexes and tibial slope, race, or level one sports participation. So what we found for this in terms of patient-reported outcomes was actually there was no difference in the IKDC or Kuh scores between men and women. Apilaxy showed no difference. And actually, the women had better return of muscle strength at six months. They had less of a quad deficit, less of a hamstring deficit, lower hop test deficit. And this difference was still present even when we accounted for the differences in age and BMI. And there was also no difference in either the ipsilateral or contralateral ACL injury. So our Bayer II trial results from the randomized control trial, we had no infections, rejection, or immune reaction. At two years out, the augmented ACL repair subjects had comparable patient-reported outcomes, comparable knee stability, improved muscle strength, a re-injury rate similar to that seen in ACL reconstruction, and that surgery after re-injury may be as successful as primary ACL reconstruction and better than revision ACL reconstruction. However, we did find also that we may need to look at rehabilitation more carefully. We also found that slower rehabilitation early on led to an improved MR appearance and that women recovered muscle strength faster than men, suggesting we may need to look at rehabilitation as a function of sex. So where are we with bridge-enhanced ACL restoration? Well, we started off with preclinical studies in a large animal model, where we found that scaffold material was resorbed and gradually replaced by ligament tissue, and that repair was durable in longer-term studies. We also then moved and checked to make sure things were biocompatible. We looked at a sterilization validation to make sure we could make a product that was sterile to use in patients. And then we ran a level one randomized control trial. And this showed us that the patient-reported outcomes, functional outcomes, and re-tear and re-operation rate were similar in the two groups. And with the collection of that data, we were then able to get FDA-authorized marketing – sorry, FDA approval for pre-marketing approval. So this product can now be commercialized. So basically, we were able to go from where we were, our comfort zone, which was technological readiness levels one, two, and three, and move through the rest of it to get to where we can actually move it into an actual system and hopefully use this routinely someday. So I think the future of orthopedic innovation, you know, we do struggle with this valley of death that's here. But what I'd really like to see is, can we now turn this experience into a translational pipeline where it's fairly routine for us in academia to work on in vivo models, pre-clinical assays, and clinical trials to get us to the point where industry might want to pick this up? So in summary, what can we learn from rocket science? Well, if we look at the technological readiness levels, they really do give us a defined pathway for innovation. We recognize the importance of team. So we need varied interest and expertise, and these may change as you go through each phase of the technological readiness levels. But in all phases, we needed surgeons. We need surgeons for their clinical and technical insights. We need them for their surgical skills to run large animal studies, and their clinical trials are really dependent on surgeons being willing to participate. This also requires rigorous science. So just like NASA requires rigorous science, we should require that here in orthopedics. And thinking about a go-no-good decision based on data is really helpful. But in summary, I think that orthopedic innovation may benefit from a little ingenuity and perseverance. So thank you very much for your attention. I'd like to acknowledge our pre-clinical team, our clinical team over the last 15 years, and then also our financial support from NIH as well as NFLPA and other sources. So thank you very much for your attention. Thank you, Martha. That was a wonderful lecture, and we appreciate all you've done. We're going to take a minute for a question. Dr. Cole? Thank you. Your body of work with the individuals you collaborate with is clearly monumental, and all of us are watching the ball really carefully. I have a question about just the basic statistics that you provided on the 6% versus 14% because success and failure in ACL surgery is not just about failure in terms of the graft fails. I think we all are pretty comfortable saying there's other variables, intangible or not, that are important to a successful outcome. But I assume that difference was not statistically significant, right, at least in your conclusion slide, 6% versus 14%. You said there was no differences? Yeah, but I would qualify that by the study was probably not large enough to really tell it, yeah. So I think, would you agree that that might be part of the narrative moving forward? When we manage ACL patients going through 6 to 12 months recovery time, I think when you look at the risk aversion to failure, it's so exceptionally high that that's a really difficult conversation to have. And in light of that, do you think from the data you have in terms of ACL tear location that maybe when you get enough data, you'll get a sense of who might be best where you can get down to that 6%, that ambient 6% rate for this reconstruction? Because one potential concern is that it just may be generalized to all ACLs at all times, which I don't think will happen. But maybe there's a best case scenario where we can get things close to 6% if that's the aspirational goal. Yeah, certainly. We would love to study more patients with this technique. This is very early on. I will say we've looked at the Kaplan-Meier curve for the re-tears in the light of a larger data set. So you can take a very large data set, apply the inclusion and exclusion criteria that we used for Bayer II, and look at that Kaplan-Meier curve. And actually, with larger numbers of patients, the curves exactly overlap for autograft versus Bayer. We also have been working hard with the limited number of patients we have right now, which is only like 110, to look at those factors that might imply that they're at a higher risk level for re-tear. And not surprisingly, I'm sure to you, the risk factors are medial tibial slope, age, activity level. Right? So if you have a combination of those things, you're going to be at a much higher risk for re-tear. If you're over 18, have a tibial slope that's under 10, you're going to probably do fine. But that's on an N of 130. We need an N of 1,000. Right? So that's the next step, is we've got the Bayer III trial, which is enrolling, that's aimed to enroll another 300 patients. The Bayer Moon trial is going on, which is a randomized control trial looking at Bayer versus bone patellar tendon bone autograph to find that out. So there's so many questions to ask. So this is hopefully just the beginning. Thank you. Well, in honor of being our first in-person innovation lecturer, we congratulate you and appreciate your participation. So thank you very much.

bridge-enhanced ACL restoration

ACL injuries

innovation in ACL treatment

preclinical studies

clinical trials

orthopedic innovation