Earlier this month, Dr William Musick, Director of the PGY2 Infectious Diseases Pharmacy Residency program at Houston Methodist and Dr. Ryan Keul, Clinical Pharmacy Specialist for Antibiotic Stewardship at Texas Children’s Hospital shared their experience with DoseMeRx for vancomycin monitoring and a case study on busulfan dosing during a vizient member webinar.
- The benefits of Bayesian kinetics (DoseMeRx) and why it’s relevant in clinical practice
- How DoseMeRx works and fits within the clinical workflow at Houston Methodist and Texas Children’s Hospital for vancomycin dosing and monitoring
- AUC-based decision making for vancomycin at Houston Methodist Hospital
- Analysis of a busulfan case study and DoseMeRx as a clinical decision support tool
- Available drug models in DoseMeRx for adults, pediatrics and neonate patient cohorts
Kristi Kuper: Hi everybody and welcome. My name is Kristi Kuper and I’m the Senior Clinical Manager of Infectious Diseases for Vizient. I’m excited to welcome you to this webcast entitled Utilizing Real-Time Precision Drug Dosing Software to Improve Patient Care and Reduce Lab Utilization. The purpose of this webinar is to provide you with an overview of a new software that is now available in the US called DoseMeRx. Vizient is pleased to be the first GPO in the country to be able to offer this technology to our members under a special contract. DoseMeRx is a useful dosing tool for pharmacists and clinicians to help guide the improved dosing of a variety of medications which will be discussed in more detail today.
What is most exciting is that it can be integrated into your electronic health record and is available right at the point of care. So, here are just a few housekeeping notes. As was mentioned, this call is being recorded and a link will be distributed after the webinar to everyone who has registered. You can also access that link for the recording at www.vizientinc.com, and we do have a copy of the slides that are available for download today. Although this is an educational presentation, we do not have continuing education available for this webinar. And as was mentioned, all lines will be kept in listen only mode, but you can answer your questions or ask your questions in the chat box at the lower right hand side of the screen.
If you have any issues or questions after the webinar, you can always email firstname.lastname@example.org, and then we’ll also send you a follow-up email not only with the link that I mentioned but additional information on DoseMeRx which is Vizient Contract IT0340 and information on how to obtain a free trial. So, let me introduce today’s presenters. Our first presenter is Mr. Andre Pontin, who is Vice President for Business Development for DoseMe. Andre is based in Houston, Texas, and he will be providing us with a short introduction to the company. Our second presenter is Dr. William Musick, who will be initially providing a clinical overview of DoseMeRx, and we’ll return at the end to specifically discuss the functionality within DoseMe that can assist with Busulfan dosing. He has been with Houston Methodist Hospital for 13 years as their clinical specialist in infectious diseases.
In addition, he is also Director of the PGY2 Infectious Diseases Pharmacy Residency program at Houston Methodist. And during his tenure at Methodist, he has played a large role in expanding the scope of pharmacy practice and research, and was instrumental in the development and implementation of their antibiotic stewardship program. Dr. Musick specializes in the care of the critically ill and immunocompromised patients, and he has various organizational roles at Methodist. And in addition, he was recently appointed to the Texas Department of State Health Services, Healthcare Safety Advisory Committee.
Our third presenter is Dr. Ryan Keul. He will be providing an overview of the functionality of DoseMeRx to assist with antimicrobial dosing with a pediatric emphasis. Dr. Keul is currently the Clinical Pharmacy Specialist for Antibiotic Stewardship at Texas Children’s Hospital, where he has been since August 2017. He is the pharmacy leader for the antimicrobial stewardship program at both the main hospital and its satellite campuses which are located in Houston and surrounding areas. He is very involved in teaching pharmacy students, residents, and physician trainees, and participate in other collaborative groups, focus on promotion of antimicrobial stewardship and the treatment of pediatric infectious diseases such as sepsis.
Finally, we’ll be joined in the question and answer by Dr. Rob McLeay, who is the Founder and Chief Scientific Officer of DoseMe. And with that, I’ll turn it over to Mr. Andre Pontin who will give us the overview of DoseMe. Andre?
Andre: Thank you, Kristi. I really appreciate everyone’s attention today and for jumping on the call. DoseMe was founded in 2014 with a focus on providing the best in class clinical decision support systems to help optimize medication use and clinical dosing practices. We constantly strive to provide effective solutions. We’re dedicated to the support, training, and service of our clients and partners. We believe in continuous research and development in the PK dosing optimization, really, effectively trying to make an easy-to-use tool for the clinicians out there. As was mentioned earlier, I’m based here in Houston, Texas. This is our headquarters down here near the Texas Medical Center.
We have a secondary office in Brisbane, Australia, and currently we’ve helped over 125 hospitals across the world in the US, South America, Europe, and in Australia. So, really three overarching goals for DoseMe, kind of our mantra is, number one, making dosing easier. Our goal is to help clinicians be able to do that. Our second focus area is making dosing more cost effective, enabling the clinical team out there to use first line regimen drugs effectively to treat various conditions without having to move to second and third line. And thirdly, really making doses processes more rigorous and reproducible. This is obviously very important not only for small individual hospitals but also across large systems that are looking to do standardization of care, for example.
So, we offer 24/7 support and I want to hone in on that. Having an office here in the US and also in Australia, we are available for any discussions that occur at any time of the night, 24/7/365. DoseMe prides itself on being easy to use PK dosing software. We are HIPAA and FDA-compliant and we’re registered Class 1 medical device in EU and Australia as well. Vizient has given the contract members up to 30% discount available to DoseMe and we’re a Vizient-awarded supplier as well. So, we’re here to answer any questions afterwards. Thanks, Kristi.
Kristi: Okay, and with that, we’ll turn it over to our first presenter, Dr. Will Musick.
Will: Hello, everyone. My name is Will Musick here at Houston Methodist Hospital. I want to dive in quickly and give just a brief overview of DoseMe, the tool, what it is and why it’s become so important recently. So, this is a screenshot of the probably the Ferrari screen of DoseMe. Once you have patient information in the database, be it either manually, entered, or via data streams from your EHR, this is where you’re able to compare various doses and simulated outcomes. You’re able to individualize the PK target that you want to shoot for, whether it’s target peak, or target trough, or an AUC for example. The software can also be customized to your individual institution if you have institutional dosing guidelines that provide a starting point. Those can be integrated into DoseMe as well.
DoseMe works off a cloud-based shared patient database. It can either be for your team, or your hospital, or your health system if you choose. It is completely HIPAA-compliant. Our hospital is notorious for contract and data safety issues, as I’m sure are yours, but we had no problems with data safety agreement issues with DoseMeRx. DoseMe works primarily off published clinically-validated drug models that are pre populated inside of DoseMe. You’ll get a better peak at what dose models are available in Dr. Keul’s section of the talk. The models that DoseMe uses are customizable and they are able to help you develop custom drug models for your hospital or drugs that may not exist in the de facto libraries.
DoseMe and Bayesian kinetics. Bayesian kinetics may be a new word for some of you, and I want to take about two slides and explain why Bayesian kinetics has kind of jumped to the front of therapeutic monitoring. Bayesian kinetics really tries to look at a population model that exists in published literature and data from your individual patient and trying to decide how desperate they are. When we talk about Bayesian PK, it’s really not new. It’s gotten a lot of press lately with a lot of computer-assisted dosing programs that exist on the market at present; other computer-assisted platforms have existed for a long time but they were not very user-friendly. There are a lot of publications that popped up in the early 80s about Bayesian kinetics being a very effective way to target specific aminoglycoside exposures in patients.
Compared, and these were compared to traditional therapeutic drug monitoring methods. I mentioned specifically on this slide the Sawchuk and Zaske method, which is probably what most of us were taught in school. You start with a population model, feed some patient-specific data in it to get a clearance rate and volume of distribution, start your dosing your patient, get a trough or two levels and calculate an elimination rate constant, and then feed that back into the system in an iterative process to get to the outcomes that you want, basically simulating the outcomes of different dose choices on pencil and paper. What Bayesian does is really a hybrid of a priori and posteriori pharmacokinetic concepts.
A priori being population PK, before you have any data about your individual patient, and a posteriori being data that you’ve obtained about the patient on the chosen drug, basically drug levels in this case. It’s also a hybrid of statistics and pharmacokinetics. You start with Bayesian-level analysis with an assumption that your patient is part of the population from whom the population model was derived. The word ‘model’ freaks a lot of people out, but it’s just an equation. You use them every day. This starting population model is known as the Bayesian prior. Each piece of patient level data that you obtain is then compared to that population model or the prior and asses for its fit into that population model.
If it fits into the population, great, we continue on with the population model. If there’s a statistically significant variance from population model, the software can then start to create a patient specific model and move forward with that. One of the things that I like about Bayesian statistics is that it actually assumes that nothing is true. Not only does it take into account variability and patient PK parameters, but it also takes into account variability and error in laboratory specimen.
The way the interaction with DoseMe works is really two different ways. You can input a lot of information into the patient database, dose and schedule, patient characteristics and demographics, genomic results if they’re available for certain drugs like Warfarin. It’s then fed into the system, compared to a PK/PD model, the DoseMe algorithm and the Bayesian analysis kicks in, and you end up with a simulation of your individual patient that then, through the first screenshot that I showed you and you’ll see more in Dr. Keul’s presentation, you can start to simulate individual dosing outcomes within the DoseMe platform. It can be integrated into your EHR, definitely through Cerner and Epic, and there’s actually an iOS and Android app as well.
Now, in the presence of EHR integration, all of the data on the left hand side of this slide automatically populates into your patient database. There’s no loading in individual doses, there’s no loading in individual levels, and then DoseMe takes over and does the same thing and allows you to simulate potential outcomes in DoseMe, and then can actually generate notes that can then be populated to a charting section of your EHR. So, why all this talk? I think at least from my perspective, being an infectious diseases person, this is all centered around vancomycin. Bayesian isn’t any drug thing that you can find a population model for. But really, in 2004 and moving onto 2014, we became acutely aware of how important the area under the curve of vancomycin exposure was as opposed to the historical trough concentrations we had targeted.
I want to give you just a quick example that makes a nice visual representation, I think. If you look at the study that was done in 2014, the red box I have drawn around the outcomes, these are all trough levels drawn in patients. And if you look specifically at those patients with troughs between 15 and 20, you can see that the dotted line across the chart on the Y axis represents an AUC of 400, what we think to be the most appropriate target for vancomycin TDM right now. You can see that we have both patients above and below that AUC target at a trough of 15 to 20 which we consider these days fairly aggressive. You can also see that there are patients below the AUC required to treat this particular infection, even it levels out to 20 to 25.
These are patients in whom you should probably use another drug. Vancomycin is going to be too toxic in this range. Likewise, there are also patients who, at troughs well below what we would consider therapeutic may actually have therapeutic area under the curves. The current set of vancomycin dosing guidelines does recommend not accepting troughs less than 10 for concern for development of resistance, but this is like an area of vancomycin dosing that will be explored in the future once we fully transition to AUC monitoring.
As I’m sure many of you are aware, there was a commentary published just a few weeks ago on the American Journal of Health Systems Pharmacy. I would direct you all to that. It’s a fairly quick read. It talks about most likely what the recommendations from a set of guidelines would look like in the current era, targeting specifically AUC-based monitoring of vancomycin and how to implement that. And really, when it comes down to it, you’ve got two options. You can do fairly complicated math, and you can go from checking trough levels only to checking two or three vancomycin levels from which you calculate your AUC via pencil and paper, or you can do it inside of your EHR via DoseMe. I’ll turn it over to Dr. Keul for an example, specifically elucidating vancomycin and in the pediatric population.
Ryan: Hello, everyone. My name is Ryan Keul and I’m the Clinical Pharmacy Specialist for Antimicrobial Stewardship at Texas Children’s Hospital. What I’m going to talk about today for my portion of the talk is really some of my past experiences with DoseMeRx, and then ultimately utilizing real-time precision drug dosing software to dose antimicrobials. I wanted to start out with my history of DoseMeRx. A couple of years ago, I was a PGY2 infectious disease pharmacy resident at Houston Methodist Hospital, where Dr. Musick, you just heard him speak, was my residency director. My project for my PGY2 infectious disease project was focused on basically developing and implementing an AUC target of vancomycin dosing nomogram in adult hospitalized patients that could really empirically hit AUC target prior to having trough levels or knowing what the organism is.
Basically, we built this nomogram and then we dosed about 100 patients with the nomogram dosing. And you can see the nomogram on the slide, and the way it works is you have the patient’s creatinine clearance to the left, and then you have the actual body weight at the top. And you basically come across, and the nomogram will recommend a dose that’s highly likely to be able to hit your AUC target. And then we took this nomogram and we compared it to about 100 patients who were dosed via clinical pharmacist. Basically, we need a way to take a single trough level and basically work backwards to be able to determine [INAUDIBLE 00:19:44]. This is where I was first introduced to DoseMeRx.
So, after finishing residency at Houston Methodist, I moved to Texas Children’s Hospital about a year and a half ago. Texas Children’s Hospital is a 684 bed, quaternary care teaching hospital. We’re affiliated with Baylor College of Medicine. We have a couple other campuses outside of the Medical Center, so West Campus and Woodlands, and I work mainly at Main Campus which is located in the heart of the Medical Center. So, shortly after moving to Texas Children’s, my boss call me into her office and said that, “We were thinking about getting this new pharmacokinetic, pharmacodynamic dosing software.” And she wanted me to basically be the point person for the software because of my experience with infectious diseases, and I thought, “Okay, that sounds great. What’s the name of the software?” And she said, “The name of the software is DoseMeRx.”
So basically, DoseMeRx kind of followed me from residency to my new job working as a pediatric clinical pharmacy specialist. So, starting in September 2017, this is when we started implementing DoseMeRx at Texas Children’s Hospital. Around November of 2017, we had to look at some of the onboarding limits such as dose limits, default calculations. We had to really customize DoseMe to fit our pediatric population. Around January 2018, we reviewed some of the different drug models with the people at DoseMe. One of the things we’re really interested in was neonatal aminoglycoside as well as vancomycin dosing, both of which are currently offered by the DoseMeRx platform. And then just recently about two months ago in August 2018, we trained all of our clinical pharmacy specialists and we have now gotten live with the DoseMeRx platform.
And I just quickly wanted to highlight some of the different drug models that are available for the pediatric models. So, to the left on this slide, you have what’s currently available, and then to the right, you have what’s available upon request. Down to the bottom of this slide, you have kind of the key, so it’s showing different models that are available for pediatrics, which is highlighted with a P, and then neonatal with the N, and then pre-term with the PT. And I would say currently at Texas Children’s Hospital, the two drug models that we’re using most frequently are gentamicin and vancomycin. And going forward, I’m going to be focusing on vancomycin dosing using DoseMeRx.
So, as I’m sure many on the phone are aware, the guidelines for treatment of MRSA infection which were published by the IDSA in 2011, as well as the vancomycin therapeutic drug monitoring guidelines which were endorsed by ASHP, IDSA and SIDP in 2009, both of these guidelines basically mentioned that the pharmacodynamic parameter that best predicts efficacy of vancomycin is AUC/MIC, and you heard Dr. Musick talk about many of these points. And we know from these guidelines that troughs are really recommended as a surrogate marker. So, the current thinking or the thinking with those guidelines is that a trough of 15-20 for severe MRSA infection should be able to get you a target AUC of approximately 400 mg-h/L, and that’s assuming an MRSA with a minimum inhibitory concentration or MIC of less than or equal to 1.
There’s been numerous mostly small retrospective studies that have tried to identify an AUC cutoff that correlates with some type of improved patient outcome. So, this goes all the way back to the first one in 2004. Most of these studies have looked at either S. aureus – pneumonia, S. aureus – bacteremia or MRSA – bacteremia specifically. And looking to the right in this table, you can see that the vast majority of patients that have some type of improved outcome had an AUC cutoff of approximately 400. And then most importantly, looking at the Prybylski’s 2015 studies which is looking at the very bottom, that was a meta-analysis of approximately 1,700 patients. They found that an AUC of approximately 418 led to reduced persistent bacteremia and mortality, and that troughs greater than 15 which were endorsed by the current guidelines had no impact on efficacy.
So, speaking to the vancomycin targets in the adult population, most of the studies have shown a minimal association with target troughs and clinical outcomes. Like I mentioned on the previous slide, there have been some studies that have showed that as trough targets increase, as we get to trough targets of approximately 15 to 20, there is an association with increased nephrotoxicity. And I would say in large, many hospitals are currently making the change away from trough-based dosing towards AUC/MIC dosing. Most of these hospitals are targeting an AUC cutoff of approximately 400.
As far as looking at pediatric vancomycin dosing, there really isn’t a whole lot of data out there, and this same target of 400 has been largely extrapolated from the adult literature. Similar to the adult population, we know that when we increase trough targets, there’s been an association with nephrotoxicity, and there’s been numerous studies that have shown that troughs of 9-12 are often times sufficient to be able to achieve an AUC target of 400.
And then like Dr. Musick mentioned, the traditional AUC calculation is not ideal for a number of reasons, and one of those being that it requires multiple vancomycin levels. So, our experience here at Texas Children’s Hospital has been very similar to many of these studies that have been published in the past. so, we have a 2016 study that was published by our infectious disease physician group and they were looking at approximately 120 patients with S. aureus bacteremia, and you can see on this table that these different outcomes are broken down by — or stratified by trough level.
So, the first group has a trough less than 10. In the middle, you have a trough of 10 to less than 15, and then all the way to the right you have a trough greater than 15. And what they found in this study was that there was no association with trough level and median duration of bacteremia for S. aureus – bacteremia, nor was there any correlation between median time to resolution of fever. But interestingly, what they found was that they found this association with increase in trough level on a kidney injury. So, the patients with the highest trough levels greater than 15 had a much higher rate of nephrotoxicity compared to patients with a trough less than 10.
So often times, at least at Texas Children’s Hospital where I currently practice, our providers are very hesitant to get vancomycin levels because they do not feel that in pediatrics, that vancomycin levels basically correlate with efficacy. But they feel like if they give a vancomycin level, that someone may want to increase the dose to be able to target the higher vancomycin trough. So for the most part, we dosed the vast majority of our patients 15 [INAUDIBLE 00:26:37] without really any kind of mechanism to be able to monitor for efficacy or safety.
So really, one of the big advantages or one of the things that we thought would be important for us with DoseMeRx was, one, the ability to offer timely, targeted, individualized dosing. Number two, reduce reliance on the timing of blood samples. Like I mentioned, Texas Children’s Hospital is a teaching hospital, so I would say probably anywhere between 30-50% of the time, the vancomycin levels are not ordered or drawn at the correct time. With DoseMeRx, you don’t need a level, it’s a steady-state. Like I mentioned on the previous slide, there’s only one blood sample required to be able to calculate AUC. And then lastly, in my opinion, having used some of the Excel data sheets that are available that use basically two levels to be able to calculate AUC, I feel like DoseMeRx is much easier to utilize.
So, my last 5 to 10 minutes, I wanted to go over a patient case and just kind of show you some of the screenshots and how DoseMeRx actually functions. So in this particular case, this is a 16 year old male, 79 kg. He’s admitted with a Tmax of 104 and he’s admitted to the pediatric intensive care unit in septic shock. Cultures are obtained in the emergency room, and the started on ceftriaxone and vancomycin 1 gram Q8 hours. And ultimately, this patient’s blood cultures are positive for Methicillin-resistant Staphylococcus aureus.
So, this is what you see when you first log into DoseMeRx and you click on patients. So, there’s a place to be able to plug in the patient ID, the patient’s last name and first name, date of birth, sex, weight and also height. Once you plug in that patient information, you need to select the actual drug model that you’re going to utilize. So, in this case, I’ve selected the pediatric vancomycin model. Once you plug in the model that you’ll be utilizing, it’s time to add in the vancomycin doses, so you would click Add Multiple Doses down at the bottom. And this basically brings you to a spot where each vancomycin dose can be plugged in. So for instance, you can see that this patient on October 22nd in the evening around 6:00 p.m, they received a 1g dose. And then on October 23rd, shortly after midnight, so around 12:45 a.m., this patient also received another 1g dose.
And then next, you would click on lab results, which is to the right of the dosing tab, and you would plug in the patient’s vancomycin concentration. So, in this case, you can see that the patient received two doses and then the team decided to check a vancomycin trough, which came back at 6.8 micrograms/mL, although the patient wasn’t currently a steady state and the serum creatinine was 0.64. So then you can click Review Historical Plot down at the bottom, and this will basically bring up a dose response curve. So, you can see both of the doses that were given so those patients. You can also see the trough level and what time that was ultimately checked. So, this trough level was drawn roughly at the appropriate time. It just wasn’t necessarily drawn at a steady state or before the appropriate dose.
And then the other thing that this historical plot will show is that it will show the population model. So, the population model would normally be shown in red, and it’s kind of hard to see on the slide, but it is there. Basically, what’s shown in blue is the individual patient. So, this individual patient is basically conforming very well to the population model, and his dose response is basically sitting right on top of the population model. So then down here at the bottom, you can click Calculate Next Dose, and DoseMeRx will ask you, “What time do you want to simulate this dose for?” So, in this case, our trough was checked at approximately 8:30, so I’m going to simulate the next dose for 9:30 a.m.
And then you can see that DoseMeRx will come back with a recommendation, and the recommendation is largely based on hospitals default individualized target. So, the default target for Texas Children’s Hospital is an AUC24 of 400. So, you can see in this case, DoseMeRx is coming back and saying, “You can start 770 mg four times daily for three days, and this will give you an exact AUC of 400. So, in this particular case, I wanted to see, instead of doing Q6 hour dosing, is there any way we can do Q8 hour dosing which the patient is already on and still be able to hit that AUC target of 400?
So, in this case, I go to the customized dosing which is shown on the right. I keep the target at AUC24, I plug in, the AUC24 that I’m targeting is 400. I plug in the dosing interval at 8, and I ask DoseMe to extend the number of doses out to the next 12 doses. And then I click Calculate down here at the bottom, and now DoseMeRx is coming back with the recommendation of 1250 mg given three times daily for four days. This is going to give me a peak concentration of approximately 35, a trough of approximately 10, and most importantly an AUC24 of 487 which is just above what our target is at 400.
And then lastly, over to the right, it will give you once again a dose response curve. So, what you can see, with the solid black line, those are the historical doses that have been given. And then you can see with the yellow line that comes after that, it’s a yellow dash line, that is the dose response curve for the customized dose that I just put down. So, in summary, with this patient case, hopefully you were able to see some of the benefits of using DoseMeRx to be able to dose your patients antimicrobials.
And really, like I mentioned previously, what it offers us here at Texas Children’s is the ability to offer timely, targeted, individualized dosing. We have less reliance on the timing of blood samples as well as steady-state concentrations. We can take one sample and we can calculate AUC and be able to make those dosing adjustments much more quickly. And then lastly, like I mentioned previously, DoseMeRx is very easy to use. So, in a real-time setting, going through this whole patient case, I could probably do all that in less than 5 minutes. And what we’re really looking to going forward is being able to integrate DoseMeRx into our electronic medical record so that this process can become even a little bit quicker.
So with that, I’m going to turn it back over to Dr. Musick and he is going to be talking about a case with Busulfan therapeutic drug monitoring.
Will: Okay, the speakers group today thought it was important to be sure that everybody, while Dr. Keul and I are both ID people, that you have some examples of this isn’t just a vanc dosing computer, that it can really handle a broad array of drugs that are able in which you are able to use therapeutic drug monitoring to achieve certain outcomes. So, I borrowed a case example from our bone marrow transplant group. As I mentioned before, I work at Houston Methodist Hospital. I work on the Main Campus in the Texas Medical Center right next door to Texas Children’s. We’re actually an 8-hospital system composed of over 2,250 operating beds, and our bone marrow transplant unit is located here in the main campus in Texas Medical Center.
So, the case in question is a 47 year old male with a history of FLT3+ AML. No questions on the AML or Busulfan per se. I usually get involved after the chemo is given. The patient was in complete remission after receiving 7+3+Rydapt induction, followed by four cycles of HiDAC consolidation. He was then moved over to the bone marrow transplant unit for an allogeneic matched unrelated donor stem cell transplant, and the plan for conditioning for this payment was Busulfan, Cyclophosphamide, and Alemtuzumab conditioning regimen. You can see how the regimen plays out below.
So, historically, the way Busulfan therapeutic drug monitoring has been done here at Methodist is also via a software package. It’s a software package that’s been around for a long time. Its current name is Phoenix. It used to go by the name WinNonLin. It’s a PK software package that was really designed for bench-top Phase I/II drug development. The data entry into Phoenix is a bit fiddly and actually a data entry or import from Microsoft Excel documents. And I mentioned it’s not particularly user-friendly.
And this is software that isn’t really intended for individualized patient pharmacokinetics. Its true intent was to derive population PK models or PK equations from large therapeutic drug monitoring databases. And the screenshot that’s on there is an example of the output of a set of Busulfan drug levels for the patient in this case. Phoenix can draw pretty graphs, just like lots of software that’s out there to give you a visual representation of what’s going on in your patient.
As we move forward with DoseMe implementation here at Methodist, we’re currently using it targeted at vancomycin dosing. Our BMT unit is in the process of implementing and validating its use compared to their historical dosing methodologies, but this is the historical plot from the patient in question here. Historically, via our standard protocol here for Busulfan dosing, the patient gets dosed one and then nine levels. And then one those levels come back, the regimen is altered, and then a few doses later the patients get nine more levels and the process occurs again to make sure that the dose suggestion was appropriate and the desired drug exposure was achieved in the individual patient. And you can see the levels and the individual doses mapped out here.
Like Dr. Keul was alluding to previously, here, you can see a nice example of the population model or the population equation for Busulfan in the red lines and then the data from the individual patient. DoseMe decided that maybe this patient deserved their own customized pharmacokinetic parameters and then you can see how it pulled the blue line up out of the red line and created the individual patient model that then you can use to target specific TDM outcomes in your patient. For those of you like me that have been practiced for a while, graphs are great but we still want to see the numbers and we still want to see some data. So, the PK parameters for your individual patient that DoseMe has calculated is available for you to compare back to known population models to assess differences between your patient and population.
And these are the parts of the individual patient simulation screen in DoseMe where you spend a lot of time making decisions. Dr. Keul showed you multiple individual parts of the screen blown up for clarity sake, but this is the whole screenshot. And you can see here that down in the left hand corner is your customized dosing window where you would select targets, number of doses per day, infusion time, and then you simulate the exposure of that patient in DoseMe and then look at potential outcomes. And then you can go back and resimulating again just by going back in to setting your target, changing the individual dose, and then seeing the simulated outcomes in that individual patient.
And at this time, I’ll turn it back over to Kristi Kuper from Vizient.
Kristi: Great, thank you. Thanks to both you and Dr. Keul. Those were excellent presentations and a great overview of the DoseMeRx software platform. So, at this time, we do have some time for some questions and answers, and I have a couple that have come in. But just as a reminder, if you could put your questions in the chat box at the lower right hand side of your screen, that would be great. Also, you can see here, if you go to DoseMe-Rx.com, you can actually register for a free trial, and you’ll also receive an email after this webinar with a link to that website along with a listing of all of the drugs that are included in the software platform.
So, the first question I’m going to start with is probably, I think, more for Andre. Andre, the question specifically says, “How many Cerner clients to do you have and is the program compatible with Cerner model, or must it be customized?” Maybe my request for you is in your response, if you could also speak to some of the other EHR platforms that you worked with, I think that would be beneficial.
Andre: Yeah, thank you Kristi. Great question. So, we currently have about 17 Cerner hospitals that use it either as a standalone or as an integrated version. And the question here around the Cerner model, I think you may be referring to if you’re a millennium platform, and the customization aspect, they kind of go hand in hand. So, we do do some individual customization for clients depending on what is needed for their Cerner platform that they actually have, but that’s pretty straightforward and we do it across the board with most of our clients anyway. So, regardless of the model you have, I’d say you’d be fine. And then secondly, we work with Epic. We work with Allscripts. We work with a variety of other EMR vendors as well. And as you saw from Will and Ryan’s presentation, the optimal use is for it to sit inside the EMR. It’s a lot less clicking and so forth. So, hopefully that helped answer that question.
Kristi: Yeah, and I think there’s another question that just sort of piggybacks off that. I don’t know if you want to add anything else to this question, but is patient information autopulled into DoseMeRx? Can you talk a little bit about that process and how it might differ between when you’re using the standalone application versus that that’s embedded in the EHR?
Andre: Yeah, so when it’s embedded inside the EHR, all the information is pulled across. We will launch inside of the patient profile as you saw from one of Will’s screen grabs there. And that pulls in the characteristics of the patient. It pulls in any dosing protocol, regiments that the patient is on, pulls in any lab results that have been taken. So, that is all pulled in, already preorganized by Cerner, Epic, Allscripts as I said before. The ones that aren’t inside the EHR, you do have to enter some of the data and then that’s what Ryan focused in on as well. The standalone, we’ll do that complicated mathematical equation for you very quickly, but once you get really comfortable with the standalone version, it takes about three to five minutes to enter the patient information. Optimally though, inside the EMR, it’s literally three to five seconds inside the patient profile.
Kristi: Okay, great, thanks. And I think the next question is probably best answered by the Founder and Chief Scientific Officer of DoseMe, Dr. Rob McLeay. So, are you there? Just to make sure we have your audio.
Rob: I am, thank you.
Kristi: Okay, great. Wonderful. So, here’s the question: Do you know how many patients in the studies being presented were oncology? There have been multiple studies performed showing cancer patients have different vancomycin pharmacokinetics. Was the same outcome observed with AUC?
Rob: Yeah, that’s a really great question, and firstly, I want to mention that DoseMe has many different models. One of the presenters earlier mentioned that you can select a particular model, in their case, selecting the pediatric vancomycin model. At all times inside DoseMe, we have full documentation on what these models are, and they’re all published literature models. Now, for adults, the default vancomycin model is one, it does contain patients with hematological malignancies.
And in fact, in this particular paper, it’s the Buelga-Adal vancomycin model, they developed two different models, one which showed a different rate of clearance in AML patients, and then a broader model across a more general population had a slightly different rate of clearance. Now, we used the broader, more general model in DoseMe by default. But what I need to make really clear is that when you enter in those patient specific results, and that red line and that blue line separate away and you have an individual model, that individual model is reflecting your single individual patient and accurately determining their specific individual rate of clearance.
Kristi: Great. I am going to ask Will this next question because I know you have some personal experience with this [INAUDIBLE 00:45:05]. So, one of the questions that came in says, “Does this think with data mining programs,” and they specifically call it vigilance, but I would say just in general, could you maybe talk about the relationship between this software, the EHR, and then any clinical, third-party clinical decision support system you might be using?”
Will: Yeah, we are actually a Vigilant user at Houston Methodist and our stewardship program is, in many ways, run by Vigilant auto alerting and auto bloodstream infection notification to our on-call stewardship pharmacist. The integration that we’re presently considering at Houston Methodist is the integration with our EMR to make the data entry of patient demographics, vancomycin dosing or drug doses given drug levels obtained that much easier for the person making decisions. We don’t really have any experience at present about the interplay between Vigilant and DoseMe, but for example, I get a page about a patient who’s got an MRI, say, bacteremia in the middle of the night.
The way I deal with that is, number one, acknowledge the alert in Vigilant and then the remote link to Epic, I log in and see what’s going on with the patient: look at the culture results, the vancomycin if it’s available, how the patient is doing clinically, as much information as I can. And now at that point with DoseMe running in integrated EHR mode, it would really be just as simple as opening a patient profile and looking at the vancomycin dosing. So, at 3:00 in the morning, I’m not trying to plug in doses, and times, and levels, and times, and creatinines, it’s all there already when it’s integrated with the EHR.
Kristi: Okay, great, thanks. This next question, no, Andre, just to confirm, the question is: Is the trial version able to be integrated into EHR? Can you only use the standalone? I believe that’s just the standalone, correct, for the trial version?
Andre: Yeah, that’s correct. The two to four week trial is the standalone version for obvious reasons just because of some of the integration that needs to be done.
Kristi: Right, okay. I don’t see any other questions. Will, can you just advance it to the last slide? So, I want to thank everybody again joining us. And as I mentioned, we will be sending you out a little virtual gift package after the webinar with the email that you registered for this webinar for. But if you ever have any questions about this topic or anything else related to Vizient, just make a note of this email address: email@example.com, and that’s a great one point of contact for you for any pharmacy or Vizient-related questions. So, I’d like to thank our presenters today and I’d like to thank the representatives from DoseMeRx for joining in and thank you to our participants for dialing in today. We appreciate your time.
Try DoseMeRx – Free Trial
Try DoseMeRx for yourself, with a free 14 day trial including your choice of drug packages from a wide range of medications, customize dose or target functions to individualize treatment for every patient, and access our mobile and tablet apps for iOS and Android. During your free trial, you’ll also have access to our support team who are here to answer any questions you have. Start free trial now >>