Prophylactic Antibiotic Regimens in Tumor Surgery International Randomized Controlled Trial

Prophylactic Antibiotic Regimens in Tumor Surgery International Randomized Controlled Trial

Dr. Michelle Ghert is a full Professor in the Department of Surgery, Division of Orthopaedic Surgery at McMaster University. She is Director of MacOrtho Research and co-Chair of the Sarcoma Disease Site at the Juravinski Cancer Center in Hamilton. Dr. Ghert completed her undergraduate and medical degrees at Stanford University and Vanderbilt University respectively. She trained in orthopaedic surgery at Duke University and completed a 2-year fellowship in orthopaedic oncology at the University of Toronto. She has been on Faculty at McMaster since 2005. Dr. Ghert has been selected to the Presidential Line of the Musculoskeletal Tumor Society and will be President-Elect in October of this year.

Christina Ip-Toma: All right, welcome everyone to us you bytes. So happy to see you here. My name is Christina Ip-Toma. I am mom to Osteo Angel Dylan and I'm the director of scientific programs for MIB agents.

Today on Osteobites we are talking with Dr. Michelle Ghert of McMaster University. Dr. Ghert will be sharing results from the prophylactic antibiotic regimens in tumor surgery otherwise known as PARITY, an international randomized control trial, which is the first-ever prospective randomized trial in orthopedic oncology. And we have Junior Advisory board members Vicki and Andrew as our panelists today. Just a heads up before we start that Dr. Ghert's presentation today will include interoperative images. So in case you get queasy from those, you might want to just listen to the audio.

We are very excited to have Dr. Ghert on Osteobites today. She is a full professor in the Department of Surgery, Division of Orthopedic Surgery at McMaster University. She's the director of ortho research and co-chair of the sarcoma disease site at the Juravinski Cancer Center in Hamilton. She completed her undergraduate and medical degrees at Stanford University and Vanderbilt University respectively. She trained in orthopedic surgery at Duke University, and completed a two year fellowship in orthopedic oncology at the University of Toronto. She's been on the faculty at McMaster since 2005. Dr. Ghert has been selected to the presidential line of the musculoskeletal tumor society, and will be the president-elect in October of this year. Dr. Gertz's other research interests include the promotion of diversity and inclusion within the orthopedic surgery workplace.

Then before we get started, I just have a couple of events that we wanted to share with you. For those of you in the Boston area, we invite you to join us for our first ever mingle event of 2022. We promise it will be fun and there will be free drinks and appetizers. Everyone in the osteosarcoma community is invited. It's kind of like Osteobites but without the speaker and lot more conversation. Well, but the invitation and registration link in the chat and you can please share it with everyone you know in the Boston area. While we're there that weekend, we're also going to be hosting a Healing Hearts retreat for bereaved osteosarcoma parents on April 1st. Please email Casey, I will also put her email in the chat for more details and we'll put you in touch with her.

We'd like to thank our sponsor of this episode, BTG Specialty pharmaceuticals. BTG provides rescue medicines typically used in emergency rooms and intensive care units to treat patients for whom there are limited treatment options. They are dedicated to delivering quality medicines that make a real difference to patients and their families to the development, manufacture and commercialization of pharmaceutical products. Their current portfolio of antidotes counteracts certain snake venoms and the toxicity associated with some heart in cancer medications. Their drug, Voraxaze is for high dose methotrexate toxicity. Alright, Vicki, do you want to go ahead and introduce yourself?

Vicki: Yeah, hi, everyone. My name is Vicki Boy[?] and I was diagnosed in July of this year. I'm currently still going through treatment but hopefully, we'll be finalizing and beating cancer by May. I will be starting at Villanova University in the fall.

Andrew: Hi, my name is Andrew. I'm a junior in high school and a survivor. I was diagnosed four years ago. Now I'm a junior board member.

Dr. Michelle Ghert: Thank you. Thank you for that incredible introduction. I just want to say that it is an honor to be speaking to this group. I and my research group are very, very impressed, deeply impressed with your community, your organization. We're really glad that actually helped us learn about your organization after you reached out to me. So we love working with patients, we really want to know what your experiences are so that we can make your care better.

So I'm going to be presenting the results of a trial. There are some scientific elements to what I'm going to say but I will try to make it as clear as possible to those of you who are not physicians. Also I would very much like to explain that this trial is one of many that's published on a daily basis in journals, but there's there's some things that are special about it. One, it's for people like you. It will help guide treatment for people with osteosarcoma. Also, it was the first time that people in my field so you guys have had surgeons and the surgeons, they don't work together so much in research because as you know, osteosarcoma is very rare. So you end up working just in your center but if you really want to do a trial and answer some really important research questions, you need to actually work as a team. Not just with those in your center, but not even those in your region or country, you really need to go international because these tumors are so rare that that we it's hard to find out what the right thing to do is if you don't have the data or evidence. So yes, I will be describing the PARITY trial or the prophylactic antibiotic regimens and tumor surgery trial, I am presenting on behalf of the PARITY investigators, which is a very, very large group.

I would like to thank the sponsors of this study, you can see from the fairly long list of funders that a major aspect of being an academic surgeon or physician is the attempt at funding to do your research. So we spent a lot of our time writing grants, and so very grateful to these funding agencies and also to highlight Silver Hearts Foundation, which was actually a patient patient advocacy group that had funded this study.

So why do we do this study? When we do limb salvage surgery or surgery where we save the leg and don't have to amputate the leg. We have to replace to take the bone that was resected with a big metal implant. You can see on this X-ray that there are some changes, especially at the femur at the top where it looks like it might be loosening. And why would that happen? That's usually because of infection. So why are the infection rates so high with the surgeries that we do? It's part of it's because the patients are immunocompromised because as you know, on chemotherapy, that affects your immune system. Also these are big, long surgeries. The implants or the metal surfaces very comfortable for bacteria to set up their shop there. So there's lots of things affecting and against us in trying to prevent infections. So what is the actual infection rate? How often does this happen? So we did a systematic review, which is where we scoured the literature to see what was reported so far. You can see that in retrospective studies, so these are not trials, these are just people writing up their experience at their individual centers, the infection rate was 10%. And 10% is not 90, it seems like it's low but if you consider the fact that they are disastrous and very difficult complications, this is very high. If a patient has a total hip or knee replacement unit for arthritis, the infection is less than 1% risk. So this is very high.

So if you have an infection, you might need to be on antibiotics a long time, you'll have to have multiple surgeries to try to get rid of the of the bacteria. And you often end up with an amputation after after a long journey of trying to eradicate the infection. This can affect your function and your quality of life. And if you have an infection and you need to be on chemotherapy, you might not be able to have the chemotherapy because you're actively infected. So these are things that we really, really try to avoid. So how do we do that? Well, we know that if patients get intravenous antibiotics around the time of their surgery, so within an hour before the surgery starts, the risk of infection decreases. So we know that perioperative antibiotics are critical.

If you're going to have a knee or hip replacement, the recommendations by the various associations in orthopedics generally recommend 24 hours, maybe up to 36 hours of antibiotics after surgery. This is not the dose that you get while you're in surgery. This is after. So we asked the surgeons in the field that do what we do. Do you prescribe 24 hours like the surgeons that do hip and knee replacements or do you do something else? And you can see from this slide that surgeons are all over the place. About a third will give 24 hours based on what they know about the hip and knee literature, but many will go much longer even over a week and some people go longer than a week. So this is a lot of antibiotics for those people. The reason why we do that or some of us do that is because we really don't want to have infections. So you have to balance that with you know, are antibiotics bad in some ways or do we need to make we practice antibiotic stewardship, which means we minimize the chance of bacteria becoming resistant? The more antibiotics we give, the more chance that they will become resistant.

So this just goes back to the point that I brought up before I started, we looked at the level one evidence in the field and level one evidence means studies that are done as a trial prospective trial, meaning you randomize patients. It's a lot of organizations a tremendous amount of work and how much of that evidence do we have in our field? There weren't any of these trials that we could find in a systematic review. So in that survey that we found out how much surgeons give antibiotics, we asked, "Would you participate in a trial, even though we've never done one, would you participate to find out what the best thing to do for our patients is?" And almost 90% said yes. So this was something that made us think "Okay, let's go ahead, and let's do this." It's never been done but they're all other surgical specialties and medical oncologists and cardiology, all the different types of doctors do trials. So this is something that's time for us to at least start thinking about doing if not moving on it.

So the objective of the PARITY trial was to determine if long duration or five days of antibiotics intravenously after surgery, reduced surgical site infections compared to 24 hours. So this is in patients that are having a tumor resection of the femur or tibia, and that are having an endoprosthetic reconstruction. Endoprosthetic means the big metal implants that are used to replace the bone that was taken up. So this study was designed by a group of experts in various fields is orthopedic oncology. Also we had to have infectious disease experts work with us and explain what the how long antibiotics were reasonable. We thought maybe one week for the longer arm, but they felt that it was that was not reasonable, it's too long. We need people to help us design a randomized trial. So we had specialists click to clinical epidemiology and surgical trials and also biostatistics. So this even from the very beginning, this was a large group figuring out how to move forward.

So we designed the study to be international, because as I mentioned, we need to have lots of sites involved in it with multicenter and randomized parallel-arm design, that means there'll be two groups, and the patients will be allocated by a randomization program online, meaning I can't choose what my patients would have. I would have to just log into the randomization system and and the the randomization system that has been developed to balance everything out, each site would tell me what arm. Now I didn't do that because it was the pharmacists at each site that randomize the patient and shrouded the bags because it's important to blind, anybody taking care of the patient to the outcome. This is because if I knew how much antibiotics a patient was getting, I might change my opinion of how their wound is looking. I would just be affected by knowing too much. It's better to focus on their care, and all the other factors that are important, as opposed to being stuck on what antibiotics they're on. Blinding is something that is considered fairly critical to randomised trials. It basically overcomes a lot of important biases. However, it's not always possible, but in this case, it was a possibility.

So when you determine if infection occurs, it's not always straightforward because some patients have an obvious infection with they have febrile, they have redness, they got pus, sometimes it's more subtle. So we needed a committee to look at each of the cases of infection and to determine whether it actually did occur or not. Now this central adjudication committee saw the data saw the information but were blinded to whether the patients were in the one or five day regimen. So this was important because at the end of the study we had, we had to feel comfortable that the decision on whether an infection occurred or not, was not based on whether they had a lot of antibiotics or just a little.

Also, we had to make sure that the patients were safe. So we had an independent data safety and monitoring board. This committee is independent of the study, and was the only group that saw the actual data of the both groups that were unblinded. So in intervals, they met yearly and quarterly reviewed all the data to make sure that there was no concern related to one of the one of the arms was doing too much, much, much better than the other. So, this study was organized at McMaster, you can see the blue box, but we did coordinate with a lot of people. So we coordinated with the data safety monitoring board, the central adjudication committee and a steering committee, all the committees that I described. Plus, we also coordinated with the participating sites.

So in the trial, patients were randomized to cephalosporin, or cefazolin, or cefuroxime. These are antibiotics that we know are the best against the the bacteria that tend to infect the arms and legs in reconstruction surgery. So the regimen one, the patient received antibiotics for one day, and then four days of placebo, meaning they had an injection of saline but they didn't know that because it was blinded. The other end regimen was five full days. So a total of an extra 12 more doses because they were at every eight hours. So this group of patients received five full days of intravenous antibiotics every eight hours.

So who were in this trial? So they were patients that had primary bone tumors like osteosarcoma, and there are also some soft tissue sarcomas that can eat through bone just the way osteosarcoma does. And they could have been included if they were undergoing the procedures that osteosarcoma patients undergo. There's also patients that have metastatic bone disease from other organs, such as renal cell cancer, and if they are undergoing a similar kind of surgery, then they would be eligible. So the primary outcome, meaning what we were really looking for, was a surgical site infection that occurred within one year of the date of surgery. We use the CDC classification and this does and at one year after surgery after which an infection is not considered to be related to the actual surgery.

Christina: I just have a question for you there on timing. Is there any correlation between time post-surgery and risk of infections? So are you more you know, is it within the first month that you're most likely develop an infection? Does that risk decrease over time?

Dr. Ghert: It's a great question and what we did find and I'll show a graph, most of the infections were earlier on. It's not the next day, it's not a week later, it tends to be a month to two or three months later. There were some later infections, and that's a really good question. Our outcome was a time-dependent analysis, meaning the amount of the time taken to get infection was also taken into consideration. So does that answer your question?

Christina: Yes, thank you. Okay, great, okay.

Dr. Ghert: Okay, great, okay. So there were many data points collected for each patient, and so you need a really robust data management system that was managed centrally at McMaster. And you could only see each site can only see that data for their own patients. So even though McMaster had access to all of the data, it was all blinded. So this is important that you collect the data prospectively, because if you just ask the site two years later, "Oh, what did you do for this patient? And what kind of surgery did they have?" They're not going to remember, and even the medical records generally are not very helpful. So it's very important to add the data to the database right away. That way, you have data that is correct, accurate and timely. So we had to work with each of the sites to make sure that the data was accurate, valid and timely.

So there were four levels of data validation. So there's a lot of communication back and forth with the sites to make sure that they're keeping up with their data. So that at the end of the study, after the years of the study, we didn't have to go back to the medical charts, which would mean which would result in data that you don't feel so comfortable with. So this was truly international collaboration. The majority of the sites were in North America, but we had sites in Africa, we had Europe, South America and Asia as well as Australia. So this was really truly international. There were 55 clinical sites that went through the startup process, meaning they were able to get through ethics approval and contract negotiations with our site, and 48 of those 55 ended up enrolling at least one patient. This was across 12 countries, across six continents and there were over 150 investigators working on this.

So, you can see from this graph, the blue bars are the enrollment at each site per month. So that is you can see overall over the course of from January 2013 until September 2019. So this is six years, at least, that the enrollment basically increased. There were some months, particularly near the end that we had lots of enrollment and there were some that were less than this, this is pretty standard for any trial. Overall, you can see there's a trend towards more enrollment and that's because the gold line shows the number of sites that were collaborating. So you can see the beginning it was less a little drop-off. Then when people realized the trial was actually happening, they wanted to participate. So we were happy to have more and more sites work with us. And that's why the enrollment would-- that's why we were able to do this study, because of that gold line showing that people said, "Okay, let's all work together, let's do it."

And so, after a very long period of time, we finally finished all the follow-ups of one year in October of 2019. Oh, sorry, that was... I need to move this. Enrollment completed in October 2019. 2020 is when we finished actual follow up but then because of the pandemic, there was some delay in getting the final data from some of the sites. So we finally finished all of the data accumulation in early 2021. So these are the results, we had 611 patients that were enrolled, so they were randomized, and it was 312 to the 24 hour regimen and 299 to the five day. So about equal, the reason why they're not exactly equal is because some of the sites did not enroll very many patients and the randomization system was programmed to balance each patient at the site with another at the site. So if you only enrolled one patient or three patients, there would be an imbalance but this is considered acceptable given a large number of overall patients. Seven patients were reviewed by the adjudication committee and found to be ineligible and again, this adjudication committee was blinded, and so they were removed. So we had 604 patients that were included in the analysis, that it was 311 and 293.

The randomization session did a phenomenal job of making the baseline characteristics of risk factors in each group equal. So that includes patient age, gender, race, location of the tumor, comorbidities, and surgical variables such as the time in surgery, and the amount of soft tissue removed and the amount of bone removed and various other factors. So we wanted to see if patients actually stuck to the regimen that they were randomized to and if you don't, if you go to the other side, it's called a crossover. So this actually was very uncommon, only two patients crossed over and those that went from the 24 to the five-day regimen.

So the other thing is that we found that there was a decent number of patients that left the hospital before five days, and you can't get intravenous infusions at home, in general, unless it's 100% necessary. So and this was not considered for a trial in the real world, we weren't going to either keep patients in the hospital or send them home with an IV, just for one or two doses. So, however, we were happy to find that the majority about 83-84% received at least four days, and that was equal between groups. So... oh, sorry.

Okay, so the primary outcome infections, 15.9% of patients had at least one surgical site infection, so that is even higher than what we had thought it was going to be. So one finding in this study is that these indeed, are very common complications. So, in between the two groups, there were 52 of 311, or 16.7% of patients in the 24 hour regimen had an infection, and in the five day regimen, 44 of 293, or 15%. So as I mentioned before, we calculated the hazard ratio, that's a time-dependent risk outcome. And that determined that there was no difference. So all those statistics and all those P values and confidence intervals indicates that there was really no statistical difference. If a hazard ratio is one, that means there exactly is no difference.

However, there's a little bit of noise in the system there. We're not sure 100% but we're very fairly close. And you can see from this graph the number of every time the graph drops a bit, a patient has had an infection. So you can see that the infections are more common in the first three months and then level out, but if you look at the green and the blue lines, you can see that really, there's nothing there that will tell you that there's a major difference between the groups. So we did find to see if there were different types of infections that were different between groups, whether they were superficial, deeper organs face, those are the different depths of the infections. And again, there was no difference.

We also did sensitivity analyses, these are done to make sure that you're comfortable with the results and you take into account things such as death and amputation because patients that are no longer live or have had an amputation cannot get an infection. So that's what we call sensitivity or competing, risk analysis. And also per protocol, meaning the patients that crossed over versus those that stayed in their group. And all of those sensitivity analyses showed us that we were comfortable with the results. We also did subgroup analysis is meaning wonder, were men more likely to do better with five days? Were osteosarcoma, more likely to do better with five days? And there was no difference between the treatment groups. Now these are subgroup analyses and the study means is not powered, or it's not directed at answering these questions. So these can only be hypothesis-generating, it doesn't really give you a definitive answer.

We did look at in fact, at antibiotic-related complications, and this is something that was also adjudicated, so 7% of patients had an antibiotic-related complication, and this is a complication for which they were admitted to the hospital or had to go under some intervention. Therefore, these are fairly serious antibiotic-related complications. 7% is almost half the number of infections however, they are very serious. They were significantly more common in the patient group that got five days, so they 5.1% of patients in that group had a serious antibiotic-related complication versus 1.6% in the 24 hour. The hazard ratio is well above one in this, even with a confidence intervals meeting where at least 95% confident that there's a significantly higher risk in that group. The most common complication was c-diff associated colitis, and also other types of diarrhea. So c-diff colitis is a very serious condition that requires long-term antibiotics and has a significant risk of relapse as well.

Why does c-diff colitis occur? Well, the antibiotics that you get intravenously affect the microbiome or the the happy coexistence of the multiple bacteria in the colon, and the GI tract. So when you disrupt that balance, some of the more difficult... Sorry about that, my computer's doing this. Some of the more aggressive bacteria can overcome the weaker bacteria ad then that's how you get this Clostridioides-difficile colitis. And again, this just shows that there's a statistically significant difference.

Also, we looked at unplanned reoperation. So you do the surgery, but you're not planning on doing another one. 25% of patients had to go back to the operating room at least once in the first year. This didn't differ between treatment groups with a hazard ratio of 1.06. And not surprisingly, the most common types of reoperation were irrigation debridement. That means washing out and implant exchange and implant revision. These are the things that we do to try to get rid of bacteria. So most of these reoperations were done for infection. And again, the reoperation rates were not different across groups. The most common were implant exchange, irrigation, implant revision, they had enough events that we can do a statistical comparison. And you can see that there was no difference between groups.

Unfortunately, almost 30% had an oncologic event that means they either had a local recurrence or metastatic recurrence, and almost 13% died within the first year after surgery. These oncologic outcomes such as recurrence, and mortality were not different across groups. And the most common cause of mortality was disease progression. And again, this just shows the data between the treatment groups, the hazard ratios are all just very, very close to one so something that if it's one then the one group doesn't do better than another.

So what is important about this study? We definitely believe that infections and avoiding complications from antibiotics are very important to patients and the healthcare system. So this is the outcome that we're looking at is something that's important, pretty much everybody that's involved. We design this study with all of the experts that helped us to reduce bias. And bias means that you're moving away from the truth because you have various factors that affect what you're doing in the trial, such as knowing what what what your patient is given, or knowing or being able to pick what you want for your patient or not following your patient. So you don't really know what happened.

So we concealed randomization so that nobody knew, everybody's blinded. So the patients, the caregivers, and even the people that assess the outcomes, which was the adjudication committee were blinded, and even the statistician, the data analysis, did not know what group was what. We used objective study outcomes. So we had the adjudication committee, which was blinded, determined infection objectively and blinded. So that if the committee decided there was an infection, then that's the closest you can get to the truth. We also interpreted the study results blinded, so we looked at the data, and when it was analyzed, and we only saw group A, group B, we didn't know whether it was the long duration or short duration. Then we decided how we were going to write the final paper based on if A is one day or if a is five days. So we did that ahead of time. So there was no bias in the interpretation of the results and then we broke the randomization code.

Probably the most important strength of this study is that we had a multinational collaboration in orthopedic oncology. This is something that, as I mentioned, hadn't been done before and it required a lot of work and sacrifice on the part of my colleagues and their research personnel to actually get the study done. We didn't lose a lot of patients to follow up. Aside from those that passed away, we only lost about 5% of patients, which is very good for a prospective trial over the course of a year. We included lots of patients that would be eligible, meaning these results could be extrapolated to a lot of patients that are being treated now or going forward. And we had representation from diverse healthcare systems. So we had that in Canada, we have more of a socialized health care system. In the United States, it's more private or insurance-based, and then in South America was different. So we were able to run this trial in lots of different kinds of health care systems so again, we can extrapolate the results going forward to lots of different hospitals around the world.

So there were some challenges. There were many in getting this study done, but we overcame them. But the things to think about where that deciding an infection occurred is always not straightforward sometimes. And the sites reported to us if they thought there was an infection, but there was also the problem with wound healing, which can lead to an infection. So we had to have the adjudication committee review in-depth many cases, to decide in the end whether there was an infection or not, as that was the primary outcome. Of course, that was critical. So as I mentioned before, some of the sites as osteosarcoma is very rare. So they did may not have had time or the ability to enroll as many patients as other sites. So this did result in somewhat of an imbalance in the number of patients in each group, because true randomization should end up with equal. However, because there were so many patients, 600, that the known and unknown variables were at least the known variables were equally matched.

We didn't follow patients after a year, you can get an infection or called a late infection after one year. Why does that happen? Well, the CDC doesn't think it's related to the operation. It could be related to bacteria in your system, that when you brush your teeth or you get a cut, bacteria gets into your system, and then it finds the implant, and then it sets up shop there. So we did not follow patients after a year and we don't know if the antibiotic regimen would have affected that. And there were some wide confidence intervals meaning we weren't the precise difference between groups the hazard ratio was 0.93 but it did have some room before and after 1. Meaning, there's still a chance that we just didn't detect the difference but the absolute difference was 1.7% which is compared to an overall 15.9% is very small. If you want to reach statistical significance with that, to find a difference of 1.7, being what we call statistically significant, you need thousands and thousands of patients. And in our field, we don't really feel that that would be necessary because the information that we have now should be enough to help you make a decision. And to do a trial of thousands, you can see this trial took six and a half years of enrollment, thousands of patients would take more than most of the time that were actually in practice. So this is the best that we could do within our field.

So the key findings of the PARITY trial that we did not demonstrate a benefit of a five-day regimen of antibiotics, intravenous cephalosporins, compared to 24 hours in reducing surgical site infections after the surgery to remove a sarcoma surgical resection on endoprosthetic reconstruction. However, there was a significantly higher risk of clinically serious antibiotic-related complications if you were randomized to the five-day regimen. So our message to clinicians is that, when they're writing post-operative orders for this day in the hospital, they can consider the uncertainty the benefits, so we don't really know if antibiotics help, but we certainly didn't get any evidence to show that they did. There is relative confidence in the harm, meaning that we know there's a significantly higher risk for antibiotic-related complications.

So I would like to acknowledge these are the sites that participated in Canada. I have two slides of American sites because they were very, very active. So this is the first slide and this is the second. We're very grateful for our American collaborators and working together with them internationally. These are the sites that participated internationally.

And just as an end note that these are pictures of I and my team. We traveled around the world to conferences and to different sites to ensure that trials going well, but at the same time, we developed friendships. And there's a very unique bond between surgeons that take care of patients that have challenging problems, which really try to help each other through their processes and to learn from each other and to make to improve our care of our patients. So these are just some pictures of us having fun and learning from each other, and basically becoming friends. So it's almost brought an international circle of people very close together. It's probably the most enjoyable part of this kind of research as working with people that speak many different languages, but we all speak the same languages of helping our patients.

So the PARITY trial was published in January of this year in JAMA Oncology. It's actually full, open access, so you don't have to have a library registration to pull the paper. This is a first... this is what it looks like, you can download a visual abstract and editorial. There's also quite a bit of supplemental content. And just for fun, we thought we'd show the authorship list. So these are the basically hundreds of people that participated and contributed to this study. They all have, they are what we call indexed on PubMed, meaning they get credit for actually being part of this study. Thank you.

Christina: Thank you so much, Dr. Ghert. It was like a cliffhanger until the end.

Dr. Ghert: I know, I know and I presented it initially to our society no one had seen the data except for the writing committee so it was pretty exciting, after all those years.

Christina: That's amazing and also amazing enrollment. 600 patients is massive around the world.

Dr. Ghert: Thank you.

Christina: So, please if you have any questions, feel free to put them in the q&a and then I think our panelists have some questions.

Is there a reason why the study focused on lower extremity reconstruction instead of upper extremity?

Dr. Ghert: Very good question. Yes, we looked into that in the upper extremity, the infection rate is actually much, much lower than in the lower extremity you can get an infection in the upper extremity but the risk is so low that there's not just nowhere near as much I guess overall stress about it. We did consider that but the event rate is so low which is good. Why is that? Part of it is probably the larger surgery in the lower extremity. The bacteria that hang out on the skin and in the groin and in the leg are different than in the shoulder region and the shoulder tend not to be as much of an issue. So, the steering committee just, you know, of course, all these things came up as a great question and that the steering committee decided that there wasn't enough clinical evidence to include upper extremity.

Christina: And then, Ryan, did you have a question?

Ryan: Oh, yeah, I do. So my question is does it make a difference whether it's an oral or IV antibiotic?

Dr. Ghert: Good question. So we do know that intravenous antibiotics have better penetration of the tissue that goes right into your bloodstream, goes right to the surgical site. When you have oral antibiotics, there's a period of time where it's in your stomach, and then it gets absorbed. It doesn't have as rapid or intense accumulation at the surgical site. So at the time of surgery, for sure, when the risk is highest for bacterial contamination, I would say that intravenous antibiotics would be standard of care across the board. Some people discharge patients on oral antibiotics again, because they're so scared of infection. There is some discussion if you already have an orthopedic infection, can you be treated only with oral there was a recent study published in New England journal that felt that maybe you didn't need to have intravenous, but this is for people that already have infections. So intravenous antibiotics have a higher risk for complications, but they also are much more impactful at the time that you actually need to have them.

Christina: You know Dr. Ghert, I remember, actually Dylan had because he did have to be on antibiotics, I remember for an extended period of time, and he actually has those great balls that, you know, since most patients have a port that you can administer at home. I knew you'd mentioned some patients didn't finish all their doses because they were discharged for five days but I don't know if this particular drug that's a fossil and is available in that format, but was that something or is that?

Dr. Ghert: Yeah, you could have done that. Though, not all our patients had ports, because there are some bone sarcomas that aren't treated with chemotherapy. So it would have been only selected for some patients that had a port. I think that that's a reasonable possibility. It just couldn't have been, it wasn't what we call real life or pragmatic that, you know, to arrange for a couple of doses of antibiotics at home would just never happen in real life, meaning outside of the trial. Unless you had an infection and you have long-term antibiotics, then you either get a PICC line or a port so you can get antibiotics at home. Does that answer the question? It's a good question. It's just the design of the trial was meant to be, the word we use is pragmatic. So it's not so much of an experiment as what happens in real life if you prescribe one day? What happens in real life if you prescribe five days?

Christina: Yeah. Well, and especially now that we know that the 24 hours is sufficient.

Dr. Ghert: Yes, exactly. Exactly.

Vicki: The next question was, what are ways that you can minimize antibiotic resistance?

Dr. Ghert: Great question. I guess the simplest answer is don't prescribe antibiotics unless they're needed. So this unless they're needed is a pretty broad expression. If a patient has an infection, and it's obvious that they're infected, and there's a bacteria that you've been able to culture that you know you can treat with antibiotics, and then the patient should get antibiotics. But a lot of doctors give antibiotics just because they're worried about something that's going to happen. So if around the world, we said, okay, you know, we really need to steward our antibiotic prescription if our governments or regulating bodies said, "Okay, we're going to review your antibiotic use every year" I think that there would be less of an issue. So the resistance develops because you give too much antibiotics, you give them for too long and you give to too many different kinds all at the same time, so that the bacteria, really start figuring out how to get around. So in the United Kingdom, they stopped using cefazolin because of this c-diff problem, and they actually won't let the surgeons use cephalosporins there so they have less c-diff. It has to come from the government. We are allowed as doctors to pretty much do what we want if we all believe we're doing the right thing, but there's nobody you know, we're not regulated in that way.

Christina: So kind of following up on that, because I know there are some other issues like, for example, Dylan had some allergies to certain antibiotics. So we're always kind of trying to figure out though, again like clindamycin was I think actually the one that was used most frequently. So I am curious and maybe this will be like a feature study, but like, what is the standard for the type? I mean, you'd mentioned cefazolin, though I don't think, I don't remember that one ever being in our arts and all. So what class of antibiotics work best and are you looking for broad struct spectrum antibiotics or something that you know works specifically on these infection sites?

Dr. Ghert: Right, and that's a really great question because it wasn't just the standard bacteria that we were finding in this study. So most orthopedic infections, certainly not all are, but most are staphylococcus. There are staphylococcus epidermidis or staphylococcus. I've totally blanked. Anyways, this stuff, they are susceptible to cephalosporins. Cephalosporins is basically it's like penicillin, but way, get much more developed. So those are the best antibiotics for those bacteria. Also, in they also cephalosporins also have a broader spectrum. They actually have an impact against gram-negative, which stops our gram-positive. So it has a reasonably good spectrum. It's not as broad as some other ones, but cephalosporins are considered very, fairly broad. And we did, or not us, but our colleagues at McMaster did a survey of North American surgeons that do hip and knee replacement, asked them what antibiotic to use, and 95% use cephalosporins. So as a pre-op, or inter-op dose, that is the standard of care. I prescribe clindamycin if the patient has a true penicillin allergy. It does have really good absorption in bone but the spectrum is not quite as good as cephalosporins.

Andrew: Did you and the other investigators find that patients were hesitant to consenting to this study knowing that they could be randomized to receive less antibiotics?

Dr. Ghert: Good question. Very good question. I think it varied by site. So at our site, we had 90% acceptance. I'm trying to remember the reasons. You know, most people were like, Yeah, you know, you just consented somebody to rearranging the anatomy of their leg, and then you ask them for one more consent. And for them, it's so small compared to what you're actually doing. I do believe that the patients that declined are the ones that did want more. So there were some sites and I think it's it is different. So as we said, different health care systems, the United States of patients, you know, want, they're paying for their care. So they want everything. And in Canada, you know, socialized and they don't have an issue related to... they're paying for it so they want the best. They're just happy to get care because they don't have to pay for it. And it's, you know, Canada, we have excellent health care. So I think that's why we had not as much of an issue.

There were some patients that didn't want five days because my standard of practice is one day. So I mean, this was before the trial. So they all asked, "What would you do?" And so some of them said, "I don't want five days." So overall, the consent was, most sites did very well, with patients agreeing to participate. We particularly had a really good experience. I find that patients, in general, are really interested in participating in research, which is great because if we don't do research, we can't improve the care of patients. Can you imagine the COVID vaccine, if people just didn't want to participate in those trials, where would we be now? So patients have to be comfortable. There's no way, we're not pushing them into anything. But for the most part, the ones that I've worked with feel that they want to give back to the healthcare system. So, good question. It's always a challenge because I know both arms are safe. I mean, I thought five days was pretty safe but obviously, maybe not now. And any patient that I enroll in a trial, I'm comfortable with all of the arms, I know that they're you know, that there's nothing that I know of at the time that's gonna make a difference as far as their safety.

Ryan: What's the antibiotic protocol for treating surgical site infections?

Dr. Ghert: So, what you do is if you have an infection, you're most likely you're going to need surgery. So we would go in and we would culture the tissue to see what bacteria is there and then direct the treatment based on the bacteria. In the interim, you would probably get either a cephalosporin or vancomycin, which is a very strong antibiotic against gram-positive, which is what you're thinking it's most likely going to be. Then once you get the culture results, you decide what to go do from there. So and if it is a true surgical site infection like the implant is infected, then the surgery will involve a lot of removing of the implant. You put cement in, and then you'd probably need several weeks of intravenous of those antibiotics. So once you have a deep infection, that's when you say, "Okay, we really need to be aggressive now because something happened and the patient needs aggressive treatment." e really want to save the leg, or in this case, leg or arm. In other cases, our job is to you know, we always tell the residents, you save a life, save a limb and give them good function. So obviously, function is very important, but you want to save their life, you want to save their limb, and you want them to be happy with what they have left.

Vicki: I saw that you had colitis on one of the sides, and the one who ended up having that as the complication of long-term use, just how long does it take for your microbiome to like, fully recover after this disruption of long term antibiotic use?

Dr. Ghert: That's a great question. I believe it could take a year or two until you stop relapsing from c-diff. I guess it depends on how you can be treated. You know, we do treat c-diff but if you have other complications related to chemotherapy or surgery, you may have to go off them for the treatment for a while. But I would say that the microbiome will take a long time, one to two years to recover.

Christina: Dr. Ghert, I'm curious, this seemed just like a very successful trial effort in terms of how many sites enrolled and physician stated and it being the first one, does this kind of set you up to be able to do another type of study like this?

Dr. Ghert: Yeah, great question.

Christina: And what's the next question, I guess, that you would have addressed?

Dr. Ghert: Yes, we have started another trial and a lot of the sites are participating. We did a Delphi process, which is a consensus-based process a few years ago, where we asked everybody that we knew around the world in our field, and patients in different centers, what's a research question that is important to you that can affect our practice and that we can answer? And the process took, I think, 18 months it was the last step was an in-person meeting. then we published the results ad the number one priority was for soft tissue sarcoma, so not osteosarcoma. How how to follow patients? How often should we see them? Should we do chest X-rays, CAT scans? Because soft tissues or coma is different than bone sarcoma, it doesn't respond to chemotherapy. So you can have metastases, but knowing about them is not really very helpful. But we still follow our patients very intensely, because we're used to that with the bone sarcomas. So that was the number one question. So we started that trial. It's called the surveillance after extremity tumor surgery trial. We have 21 sites that are open so far, and we don't need as many sites because soft tissue sarcomas are more common than bone sarcomas. So each site will have more soft tissue than bone.

So we've enrolled about 115 patients so far, so it's not 600 but we're doing pretty well. And I do think that PARITY was successful, but there were many days and weeks and months where it was very difficult. The challenges did not seem like they were surmountable. There were many days that I didn't know if it was going to actually finish. So the same thing with the current trial, we have tough days or tough weeks, and we just keep pushing, because there's no option. You know, if we don't do these trials, and I retire at some point, and nothing happens, then there's nothing to help the patients, because we're just going to keep doing what we're doing, and as opposed to maybe making things better.

Andrew: What is the risk of infection if a patient has other types of surgeries like lung surgery?

Dr. Ghert: It would be much lower. I think we're trying to tease out the risk factors like really what causes infection, but in general, the more complex, the longer the surgery, the higher the risk for infection. So thoracic surgery when you're resecting metastases or a rib, it's fairly short. The other thing is we put in implants. You know, these big metal implants are just a magnet for infection. If you're not putting an implant and the risk is pretty low. So I would say that if you're having a metastatic if you have a nodule in your lung and you're having it taken out, it's got to be less than 1%. It's not very common. Yeah, the infection risk is pretty unique to our specialty to our patient population and I think that's why the research question resonated with so many people around the world, and they were willing to participate in trying to figure out the best way to prevent them.

Christina: So are there you know, I mean, I think they're like coatings and stuff like that but are there different materials and now with like 3D printing, that are just less susceptible to infection?

Dr. Ghert: Yeah, there are some implants that are silver coated, or some are trying to antibiotic coat the implants. The evidence for that being effective is just not there. A lot of people believe, they'll swear up and down that's the answer but there's no evidence to support it. I do think that the risk for infection is so multifactorial, that the implant surface probably is not the key factor in reducing infections. They're also more expensive, and so certain healthcare systems will not support them. And yeah, they're only available in Europe. So I don't know, it just hasn't... the FDA is just not comfortable with them and if the FDA is not comfortable, then then I'm not. But it's a good question. Like what can we do? Do we do something on the implants? Maybe that's the answer, but we haven't found it yet.

Christina: Okay, next trial. Next trial is like about [crosstalk].

Dr. Ghert: Yes, someone else is here. We're trying to teach people how to do these things so that, you know, this becomes part of what we do as a society, as opposed to just one center, and I think it's happening. It's working. It's going well.

Christina: Yeah, well, especially, as you said, for rare disease, it's otherwise it's hard to get the numbers to actually have something as statistically significant. So thank you so much, Dr. Ghert for spending an hour with us today and for your commitment to osteosarcoma patients everywhere and making it better. More information on this episode and all osteophytes can be found on YouTube, our YouTube channel, our website,, and your favorite place to find the podcast. And if you did register for this session, you will get an email with all the links mentioned today. I'm including just a link to the paper on Dr. Ghert that you mentioned and we can also share information about the other trial as well.

Next week, we are going to be talking with Elena Gerasimov from Kids v Cancer about compassionate use of investigational drugs. She will give an overview of the compassionate drug use program compassionate use program and explain how to apply for experimental drugs and discuss the resources available to both patients and doctors throughout the application process. Thank you again Dr. Ghert and our panelists, Vicki, Andrew, and Ryan and to our sponsor, BTG Specialty Pharmaceuticals. Thank you all for joining us today. We hope to see you next week.


For all past and future episodes of OsteoBites, click here

Other blog posts