Dr. Suzanne Kamel-Reid: Thank you very much. I’d like to thank Sequenom for asking me to speak today. What I’m going to do is just give you a bit of background, and then introduce how we’ve worked with the Sequenom MassARRAY platform, first to working with the OncoCarta KIT that they had available a couple of years ago, and then moving on to develop our own panels, and then how we’re using these panels in the clinical trial setting

So, just as a bit of background, as you know, there have been great advances in the past few years both in next generation sequencing and genotyping technologies and there also has been a great increase in the number of targeted agents that are available for therapeutically. The cost of doing these analyses is going down. And so, it’s certainly a good time to be doing this type of work. As you know there are a lot of mutations--acquired mutations that are common across many different types of cancers. And so, when you think about that and think about the agents that are out there and available now for therapeutic use, it really begins to move what we do in the clinic from looking at single mutations and single genes to beginning to want to look at multiple mutations and multiple genes to give our patients more options

So, I guess probably in 2010, we purchased a Sequenom MassARRAY system, and we’ve been using it since that time for quite a few studies, and I’ll be describing some of those today. So, as Fritz [sp] said, the Sequenom uses MALDI-TOF type of technology, and it’s based on allele-specific primer extension and it differentiates between the wild-type and mutant alleles based on mass differences in the nucleotides

So, today I’m just going to be talking about the genotyping, that, and MassARRAY is capable of. It is capable of doing other things, as I’ve listed here, but we’ll be focusing on the genotyping

So, as I mentioned a couple of years ago or maybe in 2009 we started working with the OncoCarta Panel v1.0, and, as you probably know, it does look at 25 mutations in 19 different oncogenes, as listed here. And our experience with that was actually extremely good. So, what I’ve listed here is just some of the assays that--some of the tissues that we tested. And you can see that we tested actually a lot of different types of tissue, a lot of different tumor sites, and we were able to detect many different kinds of mutations in those different types of tissues. So, this system is good for using material from paraffin, from FMAs, from blood, bone marrow, etc., so very robust that way

As of the time that this slide was made, we’d screened over 845 samples using the OncoCarta v1.0. And in fact, we had a mutation detection rate of about 40 percent. So, in the cases that we were looking at, we detected mutations in about 40.5 percent of the cases. And during our validation of the platform, we confirmed the mutations in 200--well in over 295 cases. And for this slide, I’m just showing ones that were at above a level of 10 percent. We find that the system sensitivity’s about six to 10 percent. So, if your mutant alleles present six to 10 percent and we’re able to detect it. And as you can see here, very high concordance rate using other technologies in terms of mutation detection. So, a really robust platform for use, and we were actually very happy with its performance characteristics

So, because we like to save money in Canada, we thought that we would develop our own panel and also because we wanted to put a few more mutations on a panel that were relevant to the tissue types that we were testing. So, the OncoCarta v1.0 had some genes and mutations that we weren’t really interested in for solid tumors that we were testing. So, we designed our own panel with the help of Sequenom and Hannah [sp], especially, who’s extremely good at this and my test development technologist, Tong. So, we developed our own solid tumor panel, and we have been using this panel now for the last year and a half to two years

This is just a little more detail as to some of the mutations on that panel. So, we added some more mutations, for example, in PIK3CA we added beta-catenin, and we added other genes that we were interested in

And so, when we were doing our validation, we actually tested the same sample using both the OncoCarta and our solid tumor panel that we designed, and we also used other technologies. So, we amassed quite a lot of data. So, again, when we used our own panel, what we found was that in the tumors that we were screening, we were picking up mutations in about 42 to 43 percent of cases. And when we confirmed mutations that we found in over 131 cases, we found that in fact the concordance was extremely good. So, again, the system was performing very well for us

So, because of these data and because of our confidence in this system, we--actually one of the things that we did as well was we compared what we’re currently doing in the routine lab. And so, we had a routine lab test in the clinical lab looking at mutations in exomes [sp] 19 and 21 of the EGFR gene in non-small cell adenocarcinomas, and our mutation detection rate in those tumors was about 15 percent and that is completely consistent with what you see in the literature

So, what we thought was, “Let’s try our solid tumor panel instead and see if we can actually increase our mutation detection rate,” knowing that there were more mutations available in that solid tumor panel, for example in EGFR as well as in other genes

And what you can see from the data here is that we were actually able to increase our mutation detection rate from 15 percent to 53 percent. So, you can see that using a panel that has more mutations that you can screen, a more comprehensive panel, certainly gives you more information. The mutations that we were detecting that we weren’t detecting using our routine lab tests were mutations such as the T790M, which is of course important in terms of resistance, as well as KRAS, BRAF, NRAS and other mutations. The new LungCarta panel that the next speaker will describe has P53 on it, and we were able to detect those mutations as well. Our panel has these genes on it that are not on the OncoCarta panel, and, again, we’re able to detect mutations in these genes. By far the mutations that we detected the most frequently in lung cancer are KRAS mutations. And, again, use of this type of technology really increases your ability to pick up mutations that may be relevant in terms of patient care

So, because of our experience with the Sequenom with using it for lots of different tissue types and our belief that the technology’s extremely robust, we then went on to do a study to look at the feasibility of actually using this type of technology in the real-time clinical setting. Were we going to be able to actually get the information that we needed in a timely enough fashion to actually be able to affect clinical care

So, this was a study that was done in collaboration with the Ontario Institute of Cancer Research in Toronto. And what we wanted to do was to actually demonstrate the feasibility of this type of process and optimize the process using procedures so that we could actually use this technology and be able to impact patient care. So--and, of course, the background for that was that we thought that because we see a lot of acquired mutations in oncology, and because a lot of these mutations are recurrent, and because we have a lot more agents available, hopefully we’re going to be able to use this information in the clinic and therapeutically

So, to start with our plan was to screen 50 to 80 eligible patients, get a fresh biopsy, and then analyze it. At that time we were using the PacBio at the OICR that high throughput sequencing system. So, we were comparing the results that we were able to get from the PacBio and from the Sequenom, and then using our own lab developed panel. And then, any mutations that we detected we would then confirm by Sanger sequencing, and then report those--that information

So, the objectives were to demonstrate the feasibility, optimize the processes, and there wasn’t a lot of optimization to be done. It’s not until you actually begin to get these fresh biopsies and then have to put it through the system that you find where all the pitfalls and the limitations are. And then, the secondary objectives were to actually see how good the PacBio system was in the same setting and to do a preliminary evaluation of the impact of the results, were they actually affecting clinical care

These are the types of mutations that we detected, and this is not up to date at all. So, we have detected many more, you know, as we screen these patients, but the one mutation I wanted to just point out was this RET mutation just because I have a case that I can show you where this information was actually able to be used therapeutically. So, this was a RET mutation that we found in the thyroid tumor of a 58-year-old man who had metastatic disease--quite extensive metastatic disease, and he was oxygen dependent. He had been on a number of previous therapies. We detected this mutation, and he was then put on sorafenib, which is a multi-kinase inhibitor in November. And he really had quite a dramatic response to this inhibitor

And so, when you look at his scans here, you can see, for example, how his lungs are full of whatever this material is. And this was prior to treatment. And after treatment you can see what kind of a response he had and how much less oxygen dependent he would be. So, this is just another level. You can see this big mass here, and you can see that it’s just gone--completely cleared up using treatment. And that treatment was completely directed--was completely because of the genotyping results. So, it was a good example of how the results actually did have an impact--a very positive impact on clinical care

So, because of that feasibility study, and the great success that we had with that, we have now started a couple of trials at Princess Margaret Hospital, where I work, using our genotyping capabilities in the phase I setting. So, we have two trials, IMPACT and COMPACT. I’ll be talking about IMPACT first

And the objectives of IMPACT were to actually provide molecular profiling information for a set of common mutations and hopefully use them to guide therapy. So, these patients were actually patients with advanced disease. They’re on--currently on therapy. We did the genotyping of their tumors to then hopefully be able to more rationally put them on a therapy subsequent to the current therapy that they were on. So, in the hopes that when they progress on the current therapy they’re on, we will have some genotyping information that may help rationalize what next treatment they should go on. Secondary objectives were to track the utilization of this information and see how physicians were actually using it and also to then track the accrual rates into the various clinical trials that are available at Princess Margaret Hospital

So, Princess Margaret Hospital actually has the biggest phase I program in the country in Canada. And so, we do have a lot of clinical trials that are available. We’re also part of the NCI Phase I Program, The Drug Development Program. So, we have a lot of drugs that we get first crack at. So, it’s a very good setting to be able to do this type of work

So, the inclusion criteria were patients with advanced disease, and we focused at the beginning on breast, non-small cell lung cancer, colorectal, ovarian and phase I candidates. We’ve actually expanded the patient cohort now to include other sites, because we have--really had extremely good accrual into this trial. And so, we had a lot of demand from other tissue sites--physicians from other tissue sites asking that they be included as well

So, the end points are that we would like to have the molecular profiling information available in the patient chart. And that is so that once the patient progresses the information is available to the treating physician and, again, secondary to look at utilization rates and so on

So, this is the workflow for this trial. And, again, this was quite a learning experience for us because there are a lot of different steps, and, you know, you think that in the routine lab you’ve got it all figured out, but once you begin to do something different, you know, it--there are a lot of bottlenecks that you need to figure out

So, the patients are consented, the tissue is collected. It goes to the lab--the clinical lab for DNA extraction. It then goes for the profiling. Any mutations are verified by Sanger sequencing or another appropriate technology that’s present in the clinical lab. And then, the results go to an expert panel that determines whether they are actionable or not, whether they’re recordable or not, and then they are interpreted and reported and go into the clinical chart

So, we to begin with had said that we would only open it up to 500 patients. As of the end of September, we already had over 400 patients enrolled. And so, we expanded the cohort now to a thousand patients. I mean, everybody was extremely enthusiastic about using this genotyping information and having it available. And so, we decided to double the enrollment cohort

So, what we do is we actually use paraffin material, and we also have blood that’s available for biobanking and for future purposes, and we analyze the data using our Sequenom PMH solid tumor panel and, as I said, we do report the data into the clinical chart

So, this is an example of what our report looks like, so patient information. And you’ll note here that we do still say molecular profiling clinical research study report, so very careful about that because we really want to make sure that we’re following all guidelines. We are a Kappa [sp] accredited lab. And so, we want to make sure that we’re not breaking the rules here. And so, we do still say that this is a clinical research study report. And we do indicate whether a mutation has been identified, and then I’ll put an interpretation in here depending on what the mutation is. And if it was unsuccessful for any of the assays, then we will indicate that here as well. We indicate what methodology is used. So, we do say Sequenom primary extension technology verified by Sanger sequencing or another validated test available in the clinical lab, and then we describe the methodology, and then we actually list all the genes and mutations that were analyzed

Now, you can understand why we’re still saying that these are research results in the clinical trial setting because sometimes we do get a mutation showing up in a tissue site that’s unaccepted. And so, you know, you’ve got to do your best in terms of trying to interpret that

We’ve had a lot of success, a lot of enthusiasm at our center. And these are the results--just interim results. So, far you see that about 39 percent--well in the tumors that we’ve tested we’ve identified actionable mutations in 39 percent of the cases. So, that’s quite a lot of cases overall. In some places some tissue sites, of course, you will see more mutations that are actionable than in other tissue sites. And this is for example--over a bit. This is really just a reflection of what we’ve got on our panel, and we know that we have to modify our panel

So, our future plans, as I said, is to increase the enrollment. We’ve added the additional tissue sites. And we are going to be integrating other technologies as well. Another study that we’ve also put into place is called COMPACT, and that acronym all it stands for is that we are now extending impact to the community. So, rather than just having the genotyping information available to patients at PMH, we’re now extending this so that we can have it available to patients that may go to hospitals in the community. Knowing that the information will go back to the community medical oncologist, and we will follow up with them so that if there are trials that their patients are eligible for based on the genotyping information that we will make sure that they know that so that the patients are enrolled in the appropriate clinical trials. So, we’ve rolled this out to other hospitals in the greater Toronto area and will be extending it out starting in November

Now, using this technology because we like it so much we are actually designing more panels that are specific for some of the other types of testing that we do. So, we have actually created a hereditary panel, and we’ve created a leukemia panel as well, and we’re also modifying our solid tumor panels. So, what we’re actually doing is for our solid tumor panel we’re shrinking it a bit so that there are--so that all the mutations that are on there are actually just actionable--clearly actionable, and we’re going to influent that in the routine clinical lab starting--I’m saying November, but the lab is pushing back a bit. So, it may be December

And we’re designing this leukemia panel. We’ve got 27 genes, 180 mutations. And, again, that’s going to be specifically for the leukemia group who has a very specific idea as to exactly what they want to be able to look at for patients that have AML ALL and so on

So, we’re designing quite a few panels that we’re hoping to use in the near future in the clinical lab, and I’d be very interested in people’s ideas on how to report this information in a concise way so that the reports aren’t pages and pages long

So, I’d just like to acknowledge a lot of individuals that have been involved in these studies, so for the feasibility study the OICR as well PMH OCI, a lot of individuals involved. The funding a lot of it has come from the PMH Foundation, which means from patient donations, and we have had other centers in Ontario involved in biopsying patients and sending tissue and in the expert panel decision making. For the IMPACT and COMPACT studies this is truly a PMH UHN initiative. And so, there are an awful lot of people involved in that from the different tissue sites, from pathology, from all aspects of medical oncology

And with that, I’d just like to thank you, and I’ll be glad to take any questions