JOE: Hello, everyone, and welcome to today's webinar: Leveraging Genomic Data to Transform Rare Cancer Treatment. My name is Joe Jacher, and I'm a field application scientist here at Genomenon. I will be moderating today's session.
Before we begin, a couple of housekeeping notes. Today's webinar will cover content available in our Cancer Knowledge Base solution, that we call CKB for short. Genomenon's product portfolio also includes the Mastermind platform. While these are separate products, both support gene and variant interpretation workflows. You can explore these solutions firsthand by registering for a free account, using the links provided on the screen. For your reference, this session is being recorded and will be available on our website. After the presentation, we'll host a Q&A. Feel free to submit your questions in the Q&A section at any time during the presentation. Lastly, before we wrap up, we'll share a brief optional survey. Your feedback is invaluable in helping us shape future webinars.
So, let's get started! As mentioned, I'm Joe Jacher, and I'll be moderating today's session. I'm thrilled to be joined by my colleague Dr. Cara Statz and Dr. Jens Rueter from the Maine Cancer Genomics Initiative.
Dr. Rueter is the chief medical officer at the Jackson Laboratory, the Medical Director at the Maine Cancer Genomic Initiative, and the Associate Director for Translational education at the Jax Cancer Center. As a member of the Jax senior management team, he works with a number of Jax clinical genomics and education experts, as well as several national leaders, on advancing the field of precision medicine with the ultimate goal of individualizing cancer treatments for patients and improving their outcomes. Previously, Dr. Rueter was a hematologist/oncologist at EMMC Cancer and the Medical director for EMMC Biobank for translational research in Brewer, Maine. Dr. Rueter graduated from medical school in Berlin, Germany; his residency in internal medicine at Tulane; and fellowship training in hematology/oncology at the University of Pennsylvania.
Dr. Cara Statz is part of the oncology curation team at Genomenon with nearly 10 years of experience in clinical genomics, and has been contributing her expertise to interpretation of data related to genomic variants and targeted therapies in oncology and curation of those data into CKB since 2015. In addition to her curation-related activities, Cara regularly meets with users to provide additional insights into the capabilities of CKB. She received her PhD From Uconn in genetics and genomics, and has prior experience as a clinical research associate, and leverages her clinical and research experience in her work on CKB with the ultimate goal of enabling better patient outcomes. Thank you both for being here today.
All of us at Genomenon are excited to be hosting this webinar in awareness of Rare Disease Day. I'd like to begin by sharing a few details on this very important day. Rare disease day is observed every year on February 28th, or the 29th during leap years — the rarest day of the year — and is dedicated to raising awareness for people with rare diseases, including rare cancers. In fact, one out of five cancers is rare, and there are over 500 different rare cancers. I'm honored to be part of this webinar joined by experts in this field and helping to raise awareness towards such an important cause.
A bit of background on who we are at Genomenon. We provide genomic intelligence in the areas of clinical diagnostics and precision therapeutics. We simplify complex genetic data into actionable insights in both diagnostics and therapy development. A brief history of our timeline: Genomenon was founded just over 10 years ago in 2014. Several years were spent developing and refining our data before we launched our flagship product Mastermind in 2017. In 2019, we began selling into the clinical, diagnostic and pharma markets, and since then, have taken on larger-scale and ambitious projects like newborn screening sequencing pilots, and in 2023, we announced the release of the curated clinical exome at the gene level, two decades ahead of the NIH. In 2024, we expanded our work from the germline into the somatic space through the acquisition of CKB. So, on that note, I will pass it over to Cara to cover CKB.
CARA: Great. Thank you, Joe. All right. So, Genomenon is proud to offer a comprehensive suite of solutions that can transform complex genomic data into actionable insights. This is in both the fields of clinical diagnostics as well as precision therapeutics, so this includes Mastermind and the Cancer Knowledge Base, also referred to as CKB, as Joe mentioned. These are solutions that provide researchers and clinicians a powerful tool for gene and variant curation as well as interpretation. These insights are available to researchers and scientists both as a software. It provides a user friendly interface, but it's also available as data, and that data can be integrated within your clinical workflow. In addition to our software and data, Genomenon also offers services around curation and interpretation, data accuracy, as well as regulatory compliance.
As mentioned, today's webinar will focus on CKB and its unique value within the field of rare cancer diagnostics and therapeutics. CKB is referred to as the cancer knowledge base. What is it? It's a relational database that can be used to interpret and identify complex cancer genomic profiles, and then, in the context of those profiles, identify potential treatment options. These treatment options could be related to evidence from the literature or recruiting clinical trials.
CKB is the standard in evidence-based interpretation of complex cancer profiles, and with CKB, you can search comprehensively. You can quickly query the database, access updated evidence-based information on variants, FDA drug labels, looking for information with those drugs, especially those related to biomarkers. You can also identify targeted therapies, or look for clinical and preclinical evidence. With the preclinical evidence, you can even narrow it down to different types of preclinical evidence, such as cell line xenograft models, patient-derived xenograft models, cell culture, etc., so it's very helpful in terms of research.
Then, also clinical trials. We curate clinical trials from clinicaltrials.gov weekly, and so you can look at trials based on variant requirements, recruiting tumor types, and then, of course, the therapeutic arms. All of this information is across 2,100 genes, so just over 2,100 genes are available in CKB. Content in CKB is up to date and timely. We do have a team of PhD-level scientists curating into CKB daily, again ensuring that we're providing the most current scientific evidence to support clinical decision making.
Lastly, CKB is structured for scalability. It leverages standard protocols, standardized ontologies. For instance, we utilize the disease ontology for tumor types when we're curating in CKB. With these standard protocols and ontologies, it helps to enable mapping and identification in a consistent, scalable, and reproducible manner.
So with access to CKB, you can find immediate value. What does this look like? For instance, to reduce turnaround time, you go into CKB, you search for a variant. You can view and interpret that variant information instantly, as opposed to having to search PubMed and sifting through lots of papers, especially when you're going through several potential variants of unknown significance. It really minimizes the need for additional research. Then, of course, you can unlock therapeutic interventions. You want to identify the best treatment options for a patient. With CKB, you can use the evidence to identify potential drug approvals for that patient.
In the context of molecular profiles, molecular profiles is what links to the different types of evidence, and a molecular profile can consist of one or more variants. Molecular profiles can also be used to link to clinical trials, so you could identify potential trial options for your patient as well. This is all done with ease, with that user friendly interface, of course. Lastly, supercharge your team of experts. Like I mentioned, we have a team of curators that are curating daily. These are PhD-level scientists making sure the data is up to date, and we're getting the newest content in there as soon as possible, so you have access to premium curated content.
Let's look at some key statistics in CKB. As I mentioned, we have curated over 2,100 genes, the majority of these genes relating to cancer, and then genes that might be part of a fusion. Connected to those genes are, of course, variants. Right now, we have just over 48,000 variants in CKB. When we're curating evidence, we're curating targeted therapies, and so that includes just over 3,000, currently. As I mentioned, we curate clinical trials weekly, and right now we're just over 15,000. Then, of course,we're making changes to the UI and data implementation, ensuring that our customers have the best user experience.
But what I really want to hone in on is the unique efficacy evidence in CKB. This is really what distinguishes CKB from other databases that are out there, and this is how you can actually use this type of evidence to help identify potential treatment options in the context of rare tumor types. Unique efficacy evidence. What does that actually mean? Efficacy evidence in CKB means that we're taking a molecular profile, which I said could consist of one or more variants, connecting it to a tumor type, a potential drug or a treatment, and then a response. All of that information, as I mentioned, is evidence-based within the literature, and then connecting that type of response to an efficacy evidence annotation.
The annotation is basically a one sentence summary of the results of a study. This is especially useful if you're looking at clinical trial information, you want to know, okay, in this clinical trial, what were the primary endpoint results? What were the secondary endpoint results? We curate that information into that one sentence summary, and so our efficacy evidence ranges from high level. As I mentioned, we have FDA drug approvals, so we have evidence of the clinical trials that were used to support those approvals.
Then, we also curate professional from the professional guidelines. That includes NCCN and ESMO. From there, we curate information related to published clinical trials, so ranging from phase three to phase one, then clinical study data, and then, of course, preclinical evidence, as I mentioned earlier. Right in between that preclinical evidence and the clinical study data are case reports, and that's where you can utilize those different types of information, in terms of supporting rare tumor type therapeutics options.
Currently, for case studies, we have just over 5,500 case study efficacy evidence annotations. When we think about looking at a patient who has a rare tumor type, a very specific molecular profile, and identifying treatment options, having to sift through PubMed, with thousands of papers being published every day and looking for that specific information that might be relevant — you can just actually use CKB in that case, and so with all of those case study annotations related to just over 500 genes and over 2,800 molecular profiles. It's a lot of information at your fingertips that can be easily queried.
In addition to all of that, we also include the AMP/CAP/ASCO tiering in CKB. This is done at the evidence level, so looking at a efficacy evidence annotation, including that molecular profile, and then, of course, the level of evidence. Looking at level A evidence would correspond to evidence that's specific to FDA approvals, FDA contraindicated guidelines, and then that would correspond to tier one, and then so on. In terms of phase three, phase two, clinical study, level C, and then finally, that level D evidence, and that's where you see the case reports and case series, and those would correspond to tier two.
Alright. I talked a little bit about using the web-based interface of CKB, which is CKB Boost. Now, I just want to bring up CKB Flex. That's that data option I mentioned earlier. CKB Flex data can be used for licensing. It's scalable and flexible content integration. You can integrate that content into your clinical pipeline, use it for clinical reporting, and that Flex data will include all of the content in CKB. Again, over 2,100 curated genes, over 48,000 cancer-related variants, and then over 15,000 clinical trials. Those clinical trials span across 38 countries. Currently, we are covering 38 countries for trials. That also includes therapies and global drug approvals, so looking at other drug approval agencies such as EMA, and then, of course, drug classes. When we curate drugs in CKB, we can group them into a specific drug class, and then you can pull out evidence related just to that drug class, if you wanted. All of that evidence, of course, relates to the efficacy evidence.
Okay, this slide sort of just highlights the different leading cancer centers and genetic testing labs around the world that are currently using CKB content for their work. It kind of gives you an idea that it's quite a variety. At this point, I'm now going to turn it over to Dr. Jens Rueter, who's going to present on the Maine Cancer Genomics Initiative, highlighting how using the CKB can help identify treatment options for rare cancers.
JENS: Thanks, Cara, and good morning, everyone. I will give an introduction to the Maine Cancer Genomics Initiative, and we've been working with CKB ever since we started back in 2016. At that time, CKB was still at the Jackson laboratory, but especially now that it's part of Genomenon, we continue to have a great partnership.
So what is the Maine Cancer Genomics Initiative? It's a program that we started at Jax, the Jax laboratory being a biomedical research organization headquartered in Bar Harbor, Maine. This was a program that we started, first, as a truly clinically focused program that aimed to take advanced genomic diagnostics, make them available to community practices in very rural settings, and really help the clinicians along that pathway to providing targeted therapies and biomarker-informed, -omics-informed therapies to their patients. It was enabled by a large philanthropic donation from the Harold Effon Foundation, that really saw the benefit of having a partner in the state of Maine, for this very rural state, to actually deliver on this promise. When we started in 2016, this was the early beginnings, I should say, of using genomic information and genomic testing to inform patient treatments. We did that by not only providing the interpretation, but also by providing testing which we did for the initial four years.
How did we go about doing this? Being a research organization, we designed a study protocol that enrolled both, initially, the clinicians, because we really wanted to better understand what are the clinicians' barriers to using precision oncology approaches? and the clinicians, once enrolled, could then actually enroll their own patients. We left it open to the clinicians, really, to decide which patients would be the most appropriate. The only criteria they had to fulfill is that they were still able to receive some form of cancer-directed therapy.
Since this is in the context of rare cancers and rare disease day, the rare cancer group, even though it's relatively small by definition, within oncology, is heavily enriched within the group of patients that benefits most from targeted therapy, simply because there are usually not as many standard of care options available. We'll talk about that a little bit later. The way we then went about it, once these patients were enrolled, we provided the tests initially through Jax, free of charge, but really the cornerstone, highlighted here in yellow, is that it was an experiential education program for the clinicians to better understand how to use the information.
We developed this program of genomic tumor boards, which is really where CKB played a significant role in this project, from the very beginning. We have collected data and we continue to collect data, both from the clinicians, better understanding how it's useful to them. How is testing and genomic tumors useful to them? And how do their perspectives change over time? Patients, we also follow them for measures about attitudes and perceptions, but also quality of life, and of course, their clinical outcomes. In Maine alone, since 2017, we've enrolled more than 2,000 patients, so it's a growing endeavor.
The genomic tumor boards, this is sort of the main focus of what we do, and I think the most impactful program that we have. We basically convene this. We used to convene it in person. If you click one more time, it is now a fully virtual experience, as you can imagine. We bring together the clinician that presents the case, and then we provide and essentially orchestrate that one or more of these, what we call, precision oncology experts dial in and provide input on this specific profile. It's really a global community, now, as you can see. Most of our advisors come from the U.S., but we have representation from Australia and Scotland and Germany. It's really a very global enterprise at this point, and I can tell you that it's always a very unique experience. If we're hosting a genomic tumor board at 4:00 PM on a Thursday in Maine, and the outside expert that calls in is in Australia, and there it's already Friday morning, 6:00 or 7:00 AM. It's a very unique experience.
So in these genomic tumor boards, what are the key characteristics? Well, they're physician-driven, so we leave it up to the physician, really, as to which patients they would like to discuss. Oftentimes, and especially as they become more experienced, they will have very specific questions on the case. We, at this point in time, actually allow for a patient with any type of genomic, or omic testing, really, to be presented. It does not have to be a Jax test, it can be a commercial test or another academic laboratory test. CKB really continues to provide what we call the evidence backbone through the discussion, and I'll show you in a case in just a minute as to how what we mean by that and how that's used.
The experts, then, to get everyone together, then provide interpretation of the evidence, including, by the way, also an interpretation of the need or the requirement for germline genetic testing, which, of course, is very closely linked to somatic testing, and therefore needs to be addressed in a genomic tumor board. We run these genomic tumor boards on a very high volume of approximately 500 cases per year.
I just wanted to walk through a specific example. Just so, you see, and this is actually one that really applies to a rare cancer, cholangiocarcinoma, which is in that category of rare cancers. This is a fantastic example to show how a genomic tumor board can help in in identifying treatment options. This 71 year old man, incidentally, he was found to have two masses in the liver, and that demonstrated Cholangiocarcinoma. He was felt, unfortunately, not be a surgical candidate or resection candidate, which unfortunately happens very often with this type of cancer. He then went on standard frontline therapy with chemoimmunotherapy, so platinum-based regimen with a P1 inhibitor. NGS testing was obtained on this initial biopsy to identify next-line treatment options, and there were three specific — well, there were actually more than three variants identified, but these were the ones that were highlighted specifically on the report. There was an IDH2 variant, a PIK3CA variant and an ERRFI1 variant.
One of the things we always discuss is, how frequent are these mutations in the cancer? Because it gives us some idea as to, are we looking at a truly unique case? Or is this a very common picture? That can help decide, are we on the right track here? Is what we think we're looking at actually what we're looking at? You can see, again speaking to the theme of rare tumors and rare cancers, having this combination of therapies is actually quite rare. In fact, if you look at this plot, which is taken from the MSK-IMPACT cohort, you can see that there is really usually no overlap in these different variants, especially the ERRFI1 variant. It's actually quite rare to be seen. The average that's given behind it in the range is sort of all cancer types, but in C]cholangiocarcinomas it's a very rare event. This is sort of highlighting that we need to think about this in a very comprehensive way.
This is how we then display the initial evidence from CKB, we actually look at all clinically relevant, and sometimes even preclinical, evidence that we can identify. This is not the entire evidence in CKB, so we do actually a pre-filtering step. If we were to display everything that's in there, this table would become very difficult to manage. We focus on the things that we think will be most relevant for the discussion, and then we basically highlight exactly what Cara mentioned earlier. We highlight what we call the evidence level or the approval status, as it's called also in CKB, and show both the tier coding level, as well as what phase of clinical investigation or FDA approval has been assigned to this.
Then, as you can see on the right hand side, we use a color code to identify the different options that CKB provides. You can see right away that in IDH1 or IDH2, actually, which is what the patient has, is an interesting target, in that there are obviously IDH inhibitors available, and in cholangiocarcinomas with IDH1 mutations, ivosidenib is FDA-approved, but this patient has an IDH2 mutation for which there is actually an approved drug as well, not in cholangiocarcinoma, but in AML. The next line down, you can see there is a new investigational drug that currently has some phase one data, but that is not yet readily available, which is a broader IDH1/IDH2 inhibitor.
Then we turn to this other variant, the ERRFI1, which is part of the whole EGFR signaling pathway. It's an inhibitory protein, actually. The thought is that tumors that have a mutation in this gene might be sensitive to the EGFR inhibitors, and in fact, CKB was able to identify a case report. It had a case report actually curated that we were able to pull from, it's a single patient that had a cholangiocarcinoma and had an ERRFI1 mutation, and was treated with Erlotinib.
Last, but not least, PIK3CA is an important aberration nowadays, because obviously, we have specific PI3 kinase inhibitors, but if you look outside of breast cancer, where the current approvals are, it becomes a bit more challenging to look at. There is some phase two data that in other solid tumors and not in cholangiocarcinoma, Temsirolimus might be helpful with the PIK3CA mutations, and then there are some other drugs in clinical trials. We provide this overview to the genomic tumor board, and then a discussion ensues.
In this particular case, the discussion very quickly focused on the ERRFI1 mutation and the IDH1 mutation as well. We then go into a bit more detail and provide some of the background information. I already alluded to kind of what the level of evidence is. The genomic tumor board then concluded, if you go to the next page, that we should also look at clinical trials, and there are some clinical trials that are enrolling. This is also directly pulled from CKB. But you can also see, we were based in Maine. There is no clinical trial in the state of Maine, and the patient and the oncologist then relayed that actually, the patient is not interested in traveling anywhere for clinical trial, so that was down-prioritized, simply based on this information.
We can see what then happened: the GTB then concluded with the fact that well, you could consider using an IDH1 inhibitor or an IDH2 inhibitor, IDH1/2 inhibitor based on the available evidence. You could also consider using Erlotinib, and we basically give it back to the clinician in that format. We don't actually say you should treat. We say, these are options for you, and then we leave it up to the physician. In this particular case, the initial first-time therapy started to fail after about six months, and then actually, this was a unique case, the patient was very interested in the genomic tumor board and the genomic testing.
After a very important conversation between the physician and the patient, the decision was made to actually start on the Ivosidenib based on the GTB discussion. The disease actually started to stabilize, and for a fairly reasonable amount of time, for about four months, tolerability was fine, but then there was unfortunately slight progression, so the physician switched to the other drug, because obviously Erlotinib, being a common drug in other tumor types in lung cancer, in particular, there's a lot of familiarity with the drug and with the side effects. So the physician decided to start Erlotinib, based on the ERRFI1 mutation, and based on the discussion that we had of the GTB.
That all went well for a while, but then there was some additional progression, and then, interestingly enough, the clinician had the insight to just add the other drug back in, which we don't do very often, but if there's no overlap in toxicities, especially not an amplifying effect in toxicities, it can certainly be tried. The physician actually was very thoughtful and implemented some dose reductions. That treatment worked actually for quite a long time. It worked for approximately a year, and then, unfortunately, the patient, of course, eventually had progression, and then passed away from cancer. It's very difficult to treat, cancer remains that. But I hope this example just kind of clarifies how how GTBs can work, and how relevant it is.
If we zoom back out a little bit, we have been able to publish some of the initial findings from this this program. Really, the two key findings that we had are, one, we are able to reach most rural patients in the state. On the left hand side, you can see the map, that insert shows the population density in the state of Maine. You can see there are many areas that are essentially not inhabited at all, but there's only very few clusters of metropolitan or more urban sites. A lot of the state is very rural, and the most rural areas are really up in the northern part of the state. You can see on the right hand map, that's where we have the darkest shade of red, and that's where we had our per capita highest enrollment, as well as in some rural areas along the coast.
On the right hand side, and we published this as well, we see that patients on target therapies live longer. This is the one year overall survival, the hazard ratio there is .69, so it's quite significant, even though the curves don't seem to separate very much. The risk of death at one year is actually significantly lower in the patient population that started on a targeted therapy, which is quite remarkable, because it's a very heterogeneous population and many different cancer types, but still we were able to show this impact.
We also know from the physicians that the GTBs really helped them increase knowledge, interpret results, and improve patient outcomes. This was based on a survey that we conducted with the clinicians over time. Based on all of these findings, which were just recently published, but we came to these conclusions earlier on, and based on this, we decided to launch a nationwide study. This was done in the context of the SWOG Cancer Research Network. We were invited to basically design a cluster randomized trial with a collaborator, Meghna Triveti at Columbia. We used a very similar model to what we found in Maine, but instead of offering this to anyone, and for any case that the physician decides, we became a bit more structured for the study and turned it into a cluster randomized trial, where the randomization occurs at what we call recruitment center level, meaning at the practice level, essentially. This study has, as a primary endpoint, the proportion of patients that started an evidence-based genome informed therapy at six months. The study was activated in August of 2022. We closed accrual in November of 2024, and we enrolled, in that roughly two year time period, 1,282 patients.
It was a very successful study, and for me, it was actually a very important milestone, because we were able to show that you can do this in different constituencies as well. We're still analyzing the data, of course. What's interesting, though, in that particular context, is that we mandated that every patient case that is enrolled in one of the intervention arms, which is the genomic tumor board plus education (the E here stands for education), that every patient is presented at the genomic tumor board. The distribution of tumor types is a bit more skewed towards the common cancers, where you do get genomic testing, lung cancer, colon cancer, but we also had some very unique and rare cancers in that group as well.
This is my concluding slide. Where do we go from here, and what are our conclusions? We think that GTBs are an important tool for oncologists, and it's especially important in rare and difficult-to-treat cancers. CKB really provides the backbone, the critically important background evidence that enables the GTB decision making. We've shown that these types of efforts are relevant throughout the country, and they're probably actually relevant on a global scale as well. We do know that genomic or molecular tumor boards are run in other parts of the world. Thinking about the future, we will need to develop scalable solutions. I think that's critically important, because there's a lot of scale here, and currently, our process is is still very manual, and it's very much an n=1 as well.
What do we need to scale, and what are the issues we need to address? Well, there's an increasing complexity, really of genomic and proteomic and transcriptomic, etc., biomarker testing. While CKB is going to grow as a tool, as we heard, and is scalable by itself, figuring out how we can extract information and bring it to people at scale, it's going to be really important, especially as the amount of biomedical literature continues to just exponentially rise over time. We have some initial ideas on how to go about doing this. As I said, we're currently working very closely with the with the folks at Genomenon to continue to improve this and to grow it over time. I'm happy to answer any questions later. Thanks very much for your attention. I'll turn it back to Cara.
CARA: Great. Thank you, Dr. Rueter, for that wonderful presentation on MCGI, and how it's utilizing CKB. Dr. Rueter presented a case with a little snippet of evidence from CKB. What I'm going to do is actually walk you through a quick demo of CKB Boost highlighting how you would find that piece of evidence. I'm going to share my screen.
What you're seeing here is the homepage of CKB Boost, and if you're not familiar with it, I'll just go through the different types of searches you can do. You can explore by a specific gene, following HGNC nomenclature, explore by variant. You're going to come in at the protein level when you search for a variant, again, following HGVS nomenclature. You can also explore by drug class, drug, or tumor type. Tumor type might be especially useful in the context of rare tumor types, where you want to maybe see all of the evidence for a specific rare tumor type. You could click on that and then search for that tumor type, and it will bring up the evidence.
We also have a couple advanced searches as well that you can do. If you wanted to enter multiple parameters related to evidence or even clinical trials, you can do that. Perhaps you're interested in a certain biomarker for a clinical trial, a certain location, you can search by state within the U.S. You can search based on the other countries that we cover, or maybe it's a specific phase. Same thing for evidence, if there's a complex molecular profile, you could try to see if that's available in CKB, and then add a tumor type as well as a drug, and even go in through specific response data.
I'm going to actually highlight the ERRFI1 gene that was mentioned in the previous presentation. I'm going to go to explore by gene, and you would just start typing in the gene symbol. Once it comes up in the dropdown you'd click on it, hit submit, and it'll bring you to the gene detail page. This gene detail page contains a lot of information. I'm just going to walk you through and highlight some of the information that might serve as really good utility. Looking at the gene description alone can be very helpful, especially in a gene that's maybe not as common, as we saw in Dr. Rueter's presentation, with variants in this gene. This particular gene, based on the gene description, it indicates that it's really an inhibitor of EGFR. In that first sentence, we include information related to its normal function.
In the second sentence of the gene description, we include its relationship to cancer. We'll look to see if there's common types of mutations within that gene in a specific tumor type, or maybe expression-based data. So we can see here that decreased expression of this particular protein has been identified in lung squamous cell carcinoma and breast cancer. Below that, you'll see the gene ID, chromosome map location, the canonical transcript that we use for the variants when we curate this gene, and then the gene role. To the right, there's just some information, a couple visuals related to the protein effect of the variants for this gene, and then the impact of the variants. We do curate all different types of variants within CKB, but we also, as mentioned, include expression, amplification, positive, negative, and so on.
Down below, you're going to see five different tabs, and this is where you might want to come, especially if you're looking for evidence for a rare tumor type in the context of a specific profile. The first tab includes your gene variants. Again, CKB is protein-centric. You're going to search for specific variants based on the protein level, and then you can also filter down to impact. If you want to just only look at frame shifts or nonsense variants, you could do that. The protein effect, when we curate a variant within CKB, we focus on the functional characterization of that variant. You can look at the controlled vocabulary that we use, so loss of function, gain of function, it could have no effect, or it could be unknown. The variant description is there, and our variant descriptions highlight the location of the variant within the protein, and in the second sentence, it indicates whether or not it's been characterized or it's predicted to result in a certain protein effect.
We also include a column here called associated with drug resistance, so if a variant is associated with resistance, you could find that information there. It would indicate why in the column which you can see right here for this particular example.
Briefly, the second tab is category variants. Category variants are especially useful when you're trying to identify potential treatment options, but you don't have any evidence for the specific variant that's found in your patient's tumor. We curate content related to category variants, often general category variants. So if it's act mut, or maybe it's at the exon level, we curate that information and link it to that type of molecular profile. When we curate the specific variants, we can then group them under that category variant, so it sort of acts as a parent term.
Molecular profiles. I mentioned how we have these molecular profiles in CKB, and that's what connects to the tumor type, the drug, the response, and then that's all linked to the efficacy evidence annotation that we write. Molecular profiles can consist of one or more variants. Especially when it comes to case studies, we see multiple variants that play a role in terms of the response to the drug. You can see down here the different types of molecular profiles that include at least one ERRFI1 variant.
This fourth tab is the evidence. As I mentioned, this is probably where you would want to come to look to see, is there any evidence whatsoever in the tumor type of my patient related to this particular gene? Down below you can see that table. It's sort of like the screenshot that was in Dr. Rueter's presentation, how that looked. In the first column you have that molecular profile, in the second column, the tumor type. I mentioned when we curate, we use the disease ontology terms, so you could actually filter down in this column to that specific tumor type. In this case, if it was cholangiocarcinoma, you could filter down to cholangio.
In addition to that, you can also filter down by response type. Perhaps you want to look at only resistance or sensitive data. You could do that as well, and then the therapy name. Therapy name could be a combination of drugs. It could be a single drug. Then, approval status. That approval status is sort of like the level of evidence. You can see there's at least four different case studies that we've curated from the literature, and then following that would be the preclinical data. When you end up on the evidence tab on a gene detail page, it actually sorts with the highest level of evidence at the top and then down below would be your preclinical data.
The next column is the evidence type. We do have different types of evidence within CKB. The majority is actionable evidence. However, we do curate diagnostic, prognostic and risk factor evidence as well. The efficacy evidence column is that annotation or one-sentence summary I was describing before. You can see, with this intrahepatic cholangiocarcinoma example, this was a case study of a patient who actually had a truncation in ERRFI1 and demonstrated a partial response, and that was due to treatment with Tarceva, or Erlotinib, in this case, which is an EGFR inhibitor. This makes sense because if we look at it mechanistically, you know, ERRFI1 inhibits EGFR, so if you have loss of that, you would have increased EGFR.
The two columns on the right are the evidence level and inferred tiers. I mentioned how we integrate the AMP/CAP/ASCO evidence levels, and then the corresponding tiers. This is at the level of the evidence. With case reports or case series, you're going to see evidence level D, and then that's going to be inferred tier 2 in that case. We do say inferred, just because if you were perhaps looking at FDA drug approval evidence, and that would be tier one, but your tumor type is different from what's on the drug label, you know, it would be off-label, and in that case you'd have to downgrade that tier. So this is just a quick way to look for information related to rare tumor types, especially in the context of a specific genomic profile.
The last tab here is clinical trials. Currently, there are no trials recruiting on patients that actually have ERRFI1 mutations, but if there were, those would be listed there, and you would see the NCT ID, the title of the trial, its recruitment status. You could actually go into the clinical trial that's been curated in CKB, and find more information there in terms of variant requirements and locations.
All right. With that, I'm going to stop sharing. Hopefully, this helped better understand how CKB can be used in terms of rare cancers. I will hand it off to Joe, who's going to start the Q&A.
JOE: Great! Thanks so much, Cara, appreciate all of your insights. Same to you, Dr. Rueter, really appreciate it. We did have a few questions come in throughout the webinar, so I'll start from the first one and go from there. If anyone has any questions in the audience, feel free to put them into the Q&A box. Starting off with the first question: How often is a patient recommended for a clinical trial based on their genomic profile? And is the trial usually available in Maine, or do patients have to travel?
JENS: That's a great question. I don't have the exact number of how often clinical trials are recommended. But if I were to guess, it's a very, very high percentage, because clinical trials are available for many genomic markers that are identified on test reports or other omic markers, of course. The problem really, at the end of the day, is access. There are currently only a few of these trials available in Maine, including the larger ones through the NCI. The NCI comboMATCH trial, for example, is available, which actually currently has a nice number of different cohorts that are based on genomic markers. We're actually also, as the Jax coordinating center, are also running the taper trial for the state of Maine, which is also a precision oncology trial, where patients are matched to specific cohorts based on their genetic profile. Both of these trials probably come up, I would say, in 30-40% of cases, would be my guess, as a potential option.
The remaining trials are phase one studies, usually at one of the large academic centers in Boston, or they're in other places, too, but Boston would be closest. We have developed some nice relationships with the physicians in the phase one units, where we then can very specifically ask, is the trial actually currently enrolling? Is there a slot? And then we can establish that referral. But it's very difficult for patients, especially if you're in the northern part of the state, it's a six to seven hour drive down to Boston. If you're in the southern part of the state, it's maybe an hour and a half, but it's very, you know, large. It's still a big gap. That's a great question, though.
JOE: Thank you for that. Are pediatric cancer cases ever discussed in your tumor board?
JENS: Yes, they are actually, for about five years now we've been running a pediatric genomic tumor board as well. Those are very, very rich discussions. We also have, and you saw on the slide, we have specific advisors for the pediatric space. We have a pediatric pathologist from Boston Children's, and then a church, and then we also now have a pediatric oncologist that we work with. So those are very rich discussions, even though, interestingly enough in the pediatric cancer and pediatric oncology world, there's a lot of diagnostic talk as well about specific fusions that identify cancers, and then not so much the clinical. Clinical trials do come up, but actually, pediatric oncologists, I've found, are much more open, or sometimes more open towards designing sort of investigational plans for these children, because they just know that the options are limited.
JOE: Thank you. Have immune-based therapies influence the treatment of rare cancers?
JENS: Oh, that's a great question. They have. Interestingly, the one that immediately comes to mind is a study that was run by the SWOG Early Therapeutics group. I'm currently blanking on the acronym, but they treated specifically rare cancers with combination immunotherapy approaches, and they were very successful in that. That was not biomarker-based, as far as I know. They could be enrolled simply as having rare cancers. That's one area where I do know that this immunotherapy is important, and then, the fact that we have two tumor diagnostic approvals based on immunotherapy markers, including MSI-high/Pembrolizumab, that combination, and then TMB-high as well, so you can only identify those if you actually do the appropriate sequencing and can identify those markers, so absolutely. That's where immunotherapy is very relevant.
JOE: Thank you, and just a reminder to those in the audience. I did see a hand go up. We're not able to get audio privileges, but if you could, please put your questions into the Q&A text box! That'd be super helpful, and then we could read it from there.
We have a few other questions that came in, and this would be for both of you, the attendee is asking about central nervous system tumors. Are those included in the Tumor Board, and are they included in CKB? Asking about CNS tumors.
JENS: Cara, you wanna do you wanna answer for us?
CARA: Sure. Yes, we do include CNS-tumor-related content in CKB, so you can search again, either by tumor type based on terms within the disease ontology. We also, as I mentioned, curate content from the guidelines, the central nervous system guidelines as well. All of that content should be available and up to date in CKB.
JENS: From the tumor board perspective, we actually have a specific genomic tumor board for CNS cancers. You know, I didn't go into much detail there, but the majority of our genomic tumor boards are open to any cancer type, and they're not specialized, but we have three that are specialized. One is a pediatric cancer one. The second one is for primary CNS tumors, and the third one is for gynecologic cancers. The reason that we have that specialization — well, pediatric is sort of, I think, almost self-explanatory. It's just, you know. It's just a different field, if you will, of oncology. But we have some sub-specialized oncologists in the state of Maine that are very active in our program. One is a neuro-oncologist, and there's a special gyno group. It just made organizational sense to have specific GTBs around those tumor types.
JOE: Great, and as a followup to that, we had another question come in that's very related. How do individuals submit tumors or information to your tumor board to get some insight? Is that possible? How do they submit? How does that communication work?
JENS: That's a great question. Joe, you can advise on how to best do this, but I'm happy to provide a link to a tool that we have, where you can submit a case very directly. I'm also happy to provide my my email address, and you can just send me an email and we can get the ball rolling. We don't need much information. We ask for de-identified information, because this is obviously an educational conference, and so we want to keep that in mind, but it's a fairly straightforward process. We're happy to discuss cases from any anyone that wants to bring it to us. Joe, do you want me to find that information and put in the chat, or what is the best way to do it?
JOE: Perhaps we could put it in our email we'll send out to our users. We always send out a recording of the webinar, and we could maybe try to put that information into the email separately.
JENS: Yes, absolutely. I will provide that information, if that's a good way to do it.
JOE: Perfect. It looks like the last question that has come in this would be for, I think, Cara and I, asking about the relationship between CKB and Mastermind, how they work together, if they're individual resources. Cara, if you wanted to take a stab at that first, and I could follow maybe more information on Mastermind.
CARA: Sure. Currently, yes, CKB and Mastermind are separate resources. If you wanted access to CKB, plus Mastermind, you'd have to have a separate subscription to that. There is no integration. However, they are definitely complementary, so having both, especially if you're doing a lot of germline interpretations, looking at hereditary cancers, because Mastermind includes the hereditary information in terms of ACMG criteria, and looking at pathogenicity. But then, also having that somatic piece as well for CKB in terms of treatment options. But yes, currently are separate, and, Joe, if you wanted to add to that, feel free.
JOE: Yeah, they're separate, but both very, very powerful. For somatic, in Mastermind, you're able to add in specific keywords like somatic, and other cancer-related terms. What Mastermind would do will search all the available literature for your gene and variant for the term somatic, and for other cancer related terms, and so you're able to get all that literature available, and you'll see the sentence fragments. So they really are complementary products for cancer.
With that, I just want to give a big thank you, again, to both Cara and Dr. Rueter for your time today, like I mentioned before. We'll send out a link to the recording of this webinar and some information on Dr. Rueter's tumor board. If everyone wouldn't mind staying on just for an extra second to take our two question survey. That'll be popping up on the screen momentarily. A big thank you for your time and for tuning in, and we look forward to seeing your feedback. Thank you!
