Alex Baker, PhD candidate
University of Cardiff, Wales, UK
Alex's research is in the nascent field of oncolytic virotherapy, utilizing re-engineered adenoviruses to specifically target and destroy cancer cells. He gained his first degree at the University of Aberdeen in Scotland. Having published research on novel mechanisms of bacterial antibiotic resistance, he joined Absolute Antibody where he developed sophisticated antibody engineering techniques which continue to address a variety of industrial needs. He then returned to Aberdeen to work with Dr Frank Ward on the immune co-checkpoint inhibitor sCTLA-4.
In 2015, Alex moved to lab of Dr Alan Parker at Cardiff University where he combined his experience of protein engineering and cancer immunology in the development of new virotherapies. He believes that the interdisciplinary approach available in the lab will accelerate the innovative treatments under development there to a translational benefit. As well as his Cardiff based research, Alex maintains an active collaboration with Dr Mitesh Borad at the Mayo Clinic developing next generation antibody therapies for genetically defined cancers.
Aside from his research responsibilities Alex is passionate about science communication and public engagement. He has run patient engagement days for those with the diseases he works on and organized the first ever Welsh Pint of Science festival to a sold-out audience. Most recently, Alex has been appointed to the board of the British Society of Gene and Cell Therapy (BSGCT) as Early Career Representative where he is responsible for engagement with scientists starting out in the field.
Learn about Alex’s research
Title: Engineered viruses as precision cancer therapeutics
- Understand how protein engineering can be used to generate tissue specific viruses
- Learn how re-engineered viruses can be used as cancer therapies
Oncolytic virotherapy, the use of viral vectors to treat cancer, holds huge promise. Viruses are natural DNA delivery vehicles evolved to target specific tissues and transform them. Oncolytic virotherapies harness these abilities for therapeutic rather than pathological results. By engineering the virus to target cancerous cells rather than healthy cells we can create virotherapies which self-amplify at the point of need. Whilst historically safety focused, the field has now pivoted to enhanced efficacy following the first approved oncolytic virotherapy, T-VEC, for melanoma.
Our laboratory develops Adenoviruses (Ads) as oncolytics. Ads are versatile platforms offering large transgene capacity, ease of manipulation, and lytic potential with an excellent safety profile. However, current Ad-based therapies are hampered by high levels of pre-existing immunity within the population and off-target effects caused by the promiscuity of Ads’ canonical receptor: CAR.
We address these issues by a “bottom up” engineering approach to enhance the well characterized Ad5 serotype combined with a “top down” investigation of understudied Adenoviruses with advantageous phenotypes. By engineering Ad5 we can ablate natural tropism and facilitate specific infection of cancer cells; demonstrated by both in vitro, and in vivo models of cancer. Concurrently, we can develop rare Adenovirus serotypes devoid of pre-existing immunity, namely neutralizing antibody activity. Integrating proteins from these serotypes into therapeutic vectors enables us to radically improve cancer cell transduction.
Once targeted, the viruses must be capable of efficient cancer cell killing. We have developed Ad vectors with a variety of transgenes to manipulate signaling pathways for therapeutic benefit. Current research focuses on combining the above aptitudes into a single virus with a 3-pronged therapeutic action:
- Inherent viral immunogenicity
- Direct cancer cell lysis
- Stimulation of anti-tumor immunity
Watch the webinar
Presenter: Alexander T. Baker – PhD Researcher Biography
0:00:00 – Slide 1
Moderator: Hello everyone and welcome to today's live broadcast, Engineered Viruses as Precision Cancer Therapeutics presented by Alexander T. Baker, a PhD candidate in the Parker Lab at Cardiff University. I'm Christina Jewell of LabRoots and I'll be your moderator for today's event. Today's educational web seminar is brought to you by Labroots and sponsored by Thermo Fisher Scientific. For more information on our sponsor please visit Thermofisher.com. Now let's get started. (0:00:39)
Before we begin I would like to remind everyone that this event is interactive. We encourage you to participate by submitting as many questions as you want at any time you want during the presentation. To do so is simply type them into the ask a question and click on the send button. We'll answer as many questions as we have time for at the end of the presentation. (0:00:39) Also please notice that you will be viewing the presentation in this slide window. To enlarge that window just click on the arrows at the top right-hand corner of that side window. If you have trouble seeing or hearing the presentation click on the support tab sound at the top right of the presentation window or report your problem by clicking on the answer a question box located on the far left of your screen. (0:01:24) This presentation is educational and thus offers continuing education credits. Please click on the continuing education credits tab located at the top right of the presentation window and follow the process to obtain your credits. (0:01:39)
I would now like to introduce our speaker, Mr. Alexander T. Baker. Alex's research is in the nascent field of oncolytic virotherapy utilizing re-engineered adenoviruses to specifically target and destroy cancer cells. (0:01:56) He gained his first degree at Aberdeen University completing a project on the soluble isoform of the immune coats checkpoint inhibitor, CTLA4, in the lab of Dr. Frank Ward. In 2015 Alex moved to the lab of Dr. Alan Parker at Cardiff University where he combined his experience of protein engineering and cancer immunology in the development of new virotherapies. (0:02:24 )As well as his Cardiff based research Alex maintains an active collaboration with Dr. Mitesh Borad at the Mio Clinic developing next generation antibody therapies for genetically defined cancers. For a more complete bio on our presenter please click on Alex's biography tab. (0:02:44)
Alex, you may now begin your presentation.
Okay, thanks very much for the introduction there Christina. Yes, thank you very much to gibco for inviting me to talk as part of this seminar series and for Labroots for hosting today. Before I kick off, I just want to thank all the people involved in the research that I'm going to be talking about today and also to all of our funders, Wellcome Trust, Cancer Research UK, LSRNW and Tenovus as well, and of course Cardiff University. So with that I will just get to the talk. (0:03:22)
I'm here to talk you to today and give a broad overview of oncolytic virotherapy and to do this I'm going to be talking both about the engineering research that we are doing in our lab to create specifically targeted viral vectors to go off to cancer. (0:03:39) And also I’m going to be using examples from elsewhere in the field as well. And this really is just a broad overview, so if anybody has specific questions please do get in touch with myself or the lab generally.
0:03:51 – Slide 2
So I'd be remiss if I didn't mention everyone who's been involved in this because it's obviously isn't just me. On the left we've got a picture of the lab group that's based here in Cardiff in the Henry Wellcome building and then on the right there's people from all over the place. (0:03:39) Big thanks to Linda Coughlan of Virat, Mount Sinai in New York. The guys at the top left there have all been instrumental in a bunch of cryptography work that I've been doing and of course thank you to Alan, Gavin and John who make up my supervisory time across my PhD and a big thanks to Sarah and James who are post-docs here.
0:04:30 – Slide 3
So a question I usually get when I start talking about oncolytic virotherapy is what is it? It's actually a very old idea based on classical biochemical exploitation of the differences between cancer cells and healthy cells. It predates penicillin in fact. The only idea being that cancer cells have kind of highly replicative protein factories if you will, an ideal host for the viruses which require protein replication in order to make copies of themselves. Since this very old idea in the field has morphed beyond all recognition of course and modern oncolytic virotherapy is more like gene therapy. And we're also beginning to exploit the immunology mechanisms involved and there was a recent review actually that was talking about how antiviral immunity is actually a key part of how oncolytics can actually work to therapy. (0:05:27) And the vectors that we use are no longer a accrued wild-type viruses or even simply passage viruses as is the case by a traditional vaccine methods. They're much more heavily engineered both rationally and irrationally as I'll be talking about later today.
0:05:45 – Slide 4
So the next thing people say is that these people already have cancer and you really want to give them a virus. They make you ill, you know. Well, A, not all of them do. Most viruses and in fact the best ones you never knew you had and the reason for that is it's not in a viruses interest to harm its host, because if it harms its host it no longer has anywhere to grow. (0:06:07) So adenoviruses which is what I'm going to be focusing on today most people have been infected with. And most people at worst had a cold. A virus's main job is simply to make copies of itself. That's all it's for. It's not there to make you ill and it does this by finding the correct type of cell to infect. It then enters that cell, makes copies of itself and spreads out. So it's nature's own DNA delivery vehicle.
0:06:37 – Slide 5
So what do I need to do to make a virus a therapeutic? Well, first of all we've got to get up to target our virus and we can do that at the genetic level by using tumor specific promotors which are promotors which are particularly active in cancerous cells. (0:06:53) We can go off the cell surface level regulation looking at cancer specific markers on the cell surface and we can also make our viruses replication incompetent except in the presence of certain cancer mutations, for example a mutated P53. And it's no good being able to just target our cancers if we can't also destroy the unhealthy cells so we can do this by immune stimulation. (0:07:19) We can rely on the virus's own lytic captivity so physically bursting the cancer cells like a balloon. And we can also do this by deliver of toxic transgenes which is something I'll just touch on at the end today. And all of this can be achieved through the engineering methods that we have both inhouse and among our collaborators.
0:07:38 – Slide 6
So what's the catch then? Why aren't we all treating cancers with dura therapy already? And obviously there are some issues of the field which are being addressed. So first of all we have to address the problems with the native tropism and not the virus. (0:07:54) So adenovirus interacts with factors in the blood. It tends to get traffic towards the liver and naturally it's a respiratory infection as well, so it interacts with the lungs and esophagus. And we need to be able to avoid all of these things less our therapeutic virus get mopped up in these tissues before it can reach the tumor and have it therapeutic action. (0:08:19)
We also have a big issue surrounding host immunity. So like I mentioned before most people have actually been infected with adenovirus before and the body doesn't know that the virus that we're giving it is a therapeutic. It sees it as a foreign pathogen, so it tries to neutralize it. And if it does this too efficiently and neutralizes it before it can reach its cancer cell, it's target, then it can't actually have a therapeutic action. (0:08:41) And finally we have the problem that you have with pretty much any drug which is issues with off target. So off target infections in this case could be taken damaging because if we go off the cancerous tissue, even with a targeted virus and we share a receptor with another tissue, then we can actually have targeted destruction of something which is not our cancer. So we have to be very careful with our targeting practices.
0:09:10 – Slide 7
So before I move into the nuts and bolts of the engineering, just want to give you a quick rundown of what adenovirus is and how it works. So our adenovirus you can see here on the left has three primary capsid proteins. It's got the fiber protein which is divided into the N-terminal region, the shaft and the knob domain with that knob domain forming the primary interaction with the cell. (0:09:34) That N-terminus links to the Penton protein at the base and then that interacts with the hexon protein which is the most abundant protein in the adenovirus and forms also a bunch of different interactions and is a big part of the immune response.
0:09:51 – Slide 8
So this is the adenovirus infection pathway as it's understood for adenovirus serotype V. There are many, many different adenoviruses. There are 52 canonical ones right now but I think we're actually up to more like 87 if you include the ones which aren't fully characterized. So adenovirus Type V is the most well studied. The fiber knob domain comes along and attaches to the CAR receptor so it names its coxsackie adenovirus receptor. This shaft domain then flexes out of the way and allows interaction with alpha B integrins. This stimulates endocytosis of our virus which then immediately gets trafficked towards the lysosomal pathway and results in a pH shift which activates a viral protease cleaving off that fiber protein at the shaft. We then have adenoviral escape from the endosome and then trafficking along the micro trivial network to the nucleus.
0:10:56 – Slide 9
So we have a bunch of different strategies we can actually employ to engineer these viruses and broadly speaking there are two approaches that we take in our lab. First of all we have our bottom-up strategy which is where we start with a well characterized adenovirus such as adenovirus type V. We then work to improve its tumor selectivity. (0:11:18) We then look to de-target it so it doesn't go after its native tropisms anymore, re-target it to go after cancerous cells and resulting in an improved Ad5 based selector. Next up we have our top-down strategy which is the inverse of this. So instead of starting with a well-characterized vector we start with a adenovirus which has not been well characterized at all. And then we pan through these viruses looking for beneficial phenotypes. (0:11:49)
So we can start off with something and then we can improve its selectivity again. We can then look at the viral interactome to find ways that we can go off to cancer cells that are new and we can also look for new viral mechanisms and eventually we can create a chimeric factor which incorporates the therapeutic effect.
0:12:10 – Slide 10
So we can actually achieve all of this through a mechanism that's known as recombineering which is a technology which is developed inhouse here at Cardiff by Gavin Wilkinson and Richard Stanton. So what we do to alter our virus seamlessly is we start with a vectorized version of the viral genome in something called a BAC, a bacterial artificial chromosome which to all intents and purposes here is a giant plasmid. (0:12:37) We then look for the region of the viral gene in which we want to alter and we create hemology arms flanking a DNA insert which we want to make and then we rely on lambda phage genes which are engineered into our E coli host strain for homologous recombination into the BAC to allow swapping over of that region that we want to alter. (0:12:58) It's essentially a two-step process, but if anyone has any further questions please do ask me at the end. And then we can go ahead and culture our virus in 293 cells. This system is enormously powerful as it allows us to seamlessly alter almost any region of the viral genome.
0:13:17 – Slide 11
So I'm going to start off by giving some examples of the bottom-up strategy where we work with the well characterized Ad5 virus.
0:13:25 – Slide 12
So I mentioned earlier that we have problems with native tropism so one of those is integrins as I mentioned in the Ad5 pathway. And this does assist with trafficking towards the spleen. So we can actually create a mutation within the penton protein in an RGD motif and make that into an RGE motif instead and prevent proper interaction with those integrins. (0:13:53) And it's kind of difficult to illustrate this one succinctly because this is the effectiveness of this mutation has been built up across the literature for a very long time. But on the right here you can see that when we put the RGE mutation into the Ad5 penton protein we have strongly reduced splenic transduction.
0:14:13 – Slide 13
Another issue is Ad5's native interactions with the blood. So it can interact with blood clotting Factor X and if it does this, this forces the virus to traffic towards the liver where it will get mop topped there and if we have all of our virus in the liver then it's not getting to our therapeutic location. So on the right in the blue you can see the basis in the Ad5 Hexon protein which are mutated in order to prevent this Factor X interaction. (0:14:43) And on the bottom there is some biocore data. You can see an orange the Ad5 comes along. It binds very strongly to Factor X and it doesn't really come off. It's extremely tight interaction. However, if you look at the purple line that's the Ad5HVR7 mutation and we've got almost no interaction at all. So we've successfully ablated both integrin and Factor 10 binding.
0:15:05 – Slide 14
So if we look at how these mutations work in combination we can see here an illustration of a virus which has both the penton and hexon RG and HVR7 mutations and this is an image of some mouse organs who have been treated with the indicated viruses. So you can see that with regular Ad5 there's a very strong blue color in the liver and that's indicative of viral interaction and there is not very much in the spleen. (0:15:33) But if we add that Factor X ablating HVR7 mutation we can see the liver returns to a yellow color and there's no real viral interaction there because that Factor X interaction is no longer trafficking our virus into the liver. However, what we've done is drive our virus to the spleen. But if we then combine our mutations we can see that we've managed to completely avoid liver infiltration and also spleen as well. So those two mutations are working together to de-target our virus.
0:16:03 – Slide 15
So I've talked about the secondary interactions. What about that primary interaction with coxsackien, the adenovirus receptor? So if we make something called the KO1 mutation which is a two residue substitution SP to EA in the fiber knob domain, we can successfully block that interaction with CAR. (0:16:24) So if you just pay attention to the black bars for now on the left of that graph. On the left we've got Ad5 showing transduction of CHO-CAR cells. CHO-CAR cells have no adenovirus receptors except for (inaudible) the virus receptor. So as you would except the well typed virus capsid successfully interacts and you've got a tall black bar. On the right with the KO1 mutation it's almost completely ablated. You're really looking at background noise in the assay there. And the others are control switch all happily go into a fast.
0:16:58 – Slide 16
So we managed now to ablate our virus's primary and secondary receptor tropisms, so we have successfully crippled our virus now. So we need to start building it back up again and finding a therapeutic target. So what are we looking for in a therapeutic cancer target? We're looking at something which is highly expressed and highly specific. So this is distribution data, so a number of positive tumors for a receptor called aVB6 integrin. (0:17:31) And you can see in an awful lot of cancers it is very highly expressed although it does vary by the cancer itself. aVB6 integrin is an oncofetal antigen so it's only expressed in the initial development and it's also expressed in wound healing which means that in regular healthy tissues in an adult, anyone's who's going to have cancer, it's not really present at all. (0:17:54) However, since it's on our cancers it makes it a very, very good marker and a very selective marker as well. So this is a model of aVB6 integrin which was published a short while ago. In the green we have the aV component and in the purple the B6. We also know the aVB6, it's not on that table there, is highly upregulated in pancreatic cancer which makes one about very exciting, has exceptionally poor prognosis right now.
0:18:23 – Slide 17
So what we can do is work with something called A20 peptide. And the A20 peptide is actually derived from certain mouth disease virus and it uses aVB6 integrin as its native target in cattle. So what we've done is we have disrupted the loop on the end of the adenovirus loop which is seen in the cyan here on the right and we've inserted this A20 peptide which is pitched in the dark blue. So we end up with three copies of our A20 peptide per fiber knob protein.
0:19:02 – Slide 18
By doing this we successfully retargeted our virus towards some cells which are otherwise completely resistant. So these are BT-20 triple negative breast cancer cells. Triple negative breast cancer has probably got the worse prognosis of—yes, the worst prognosis of all the breast cancers for sure and probably a lot of the gynecological cancers. So by inserting this A20 loop into our fiber knob domain we can actually retarget the Ad5 here so that it's capable with infecting these BT-20s when otherwise it's incapable and we can further stimulate that infection by combing it with the K01 mutation. So we've now broken down our virus so it's incapable...
0:19:47 – Slides 19
...of interacting with its native cells. And then we've provided it with a new receptor and new cell type which is of therapeutic benefit. So the native tropism is gone by the ablation mutations and then our off target infection is eliminated by going after a highly cancer specific marker. But we are still left with the issue surrounding host immunity.
0:20:11 – Slide 20
So an example of how we can avoid host immunity is embodied in some of the top-down strategy. So this is where we work with a poorly characterized adenovirus instead of the well characterized Ad5. So the question is if we're going to work with a poorly characterized adenovirus...
0:20:27 – Slide 21
...which one because this is phylogeny of all of the prime adenovirus which exist at the moment. There are more than just this though. They have recently been described aquatic snail adenovirus and adenovirus for birds of prey and sea lions and all sort of things. There really is a huge diversity in this family. And the vast majority of them are extremely understudied. So these Species C's are probably the best studied of these and Ad5 is a Species C virus. They are very well tolerated but they have very limited efficacy. (0:21:02)
The E's, and there was actually some work done here in Cardiff a long time ago by Gavin Wilkinson and his team generating an anti-ad for vaccine for the U.S. military. Species B's, they have low seroprevalence rates and there is some exciting work happening with Species B’s actually with the Enadenoteucirev, ColoAd1 story out of Len Seymour's lab in Oxford, as Species D are by far the most diverse of the adenovirus species. (0:21:28) And you can probably tell by the amount of information and I'll be coming back to those. The Species A's are a small species. They're quite unstudied, potentially CAR utilizing but not much information out there for them. The F are kind of bunkers, they've got two different fiber proteins. We really don't know what they're doing in terms of receptors, but they are quite common. And the Species G's, there are quite a lot of simian and rhesus serotypes which are newly described in here.
0:21:58 – Slide 22
So where do we look? We're going to look in alternative species, Species C because as I mentioned they have limited advocacy. We're kind of getting to understand the B's. We've got a lot of information about the C's. There's not much diversity in the other species which leaves us with the D's which are extremely poorly characterized. And there's quite a bit of confusion in the literature there. There is not really much information known about what receptor those subgroup D's use. Some of them have been shown to use sialic acid but most of them, it's completely unknown.
0:22:33 – Slide 23
So those D's are where we are going to look. And actually they contain 35 of the 52 canonical adenovirus serotypes. So it's kind of baffling that they're so understudied given that they make up the bulk of the viral family.
0:22:33 – Slide 24
So why are they useful? What are we looking for? So what we have here is seroprevalence data. So these are neutralizing antibody responses against the adenovirus infections in various people, sampled across Sub-Saharan Africa. So Ad5 you can see that pretty much everyone in that population has some degree of preexisting immunity to Ad5 with 48% of them showing very, very strongly neutralizing antibody activity. (0:23:19) Ad35 there is species B. You can see that there is much, much lower seroprevalence there and then the rest of those viruses are all Species Ds and you can see all of them have very low seroprevalence, especially the Ad48 there, the 26 and the 49. So that means that we can start evading the problems with preexisting neutralization. So if we treat people with a vector based on these things then not already going to clear it before it can actually have it's...
0:23:48 – Slide 25
...therapeutic action. So this is an assay done in VSMC, vascular smooth muscle cells on the left and then the endothelial cells on the right. And it shows that the species B viruses are much more capable of infecting the cell types than the Ad5 is. It's a varying degree studies.
0:24:10 – Slide 26
So what we can do to start exploring some of these tropisms is we can do something called pseudo typing. So that's the structure of our fiber protein here on the top there and that's a model that I made. It's actually chimera but it gives you an idea about what the protein may look like in reality. So what we can do is we can cleave genetically that knob domain sequence out and replace it with that of our Species D virus. We can also do something similar in the hexon protein, the hexon being so abundant is a major part of pre-existing immunity.
0:24:44 – Slide 27
So this is some really interesting data which came out a while back now where they replace the fiber knob domain with that of canine adenovirus just to demonstrate the power of the pseudo typing approach. And it actually showed 15-fold of percolation in the adenovirus infection when they pseudotype the canine adenovirus and they showed this is in both neuroblastoma cells so U-118MG's here but they also showed it in a bunch of different cells lines which are shown in that table achieving up to a 33-fold improvement in infectivity than they did with the regular Ad5. So just by changing that one protein we've drastically improved our cancer infectivity.
0:25:31 – Slide 28
So seroprevalence there is a bit of confusion in the literature about where that neutralizing antibody response is targeted and it actually seems to depend slightly on the root of administration. It looks like in a natural infection most of that neutralizing antibody activity will be against the fiber protein; however, if you do an intramuscular injection and also partly in a wild-type infection it starts to get very muddled here. (0:25:59) There is also a strong neutralizing antibody response against the hexon protein and that hexon protein here is what we've got here on the right. In the gray we have the structural region which really forms the capsid and sort of locks together to form that classic virus icosahedron that you get with adenovirus. And that blue region is the externally exposed parts and that contains what we call HVR's which are hypervariable regions. (0:26:25) And these vary wildly between the different adenoviruses. So, since that's the exposed region that's where we get most of our antibody response. So Ad5 though will be a lot of people with anti-Ad5 HVR antibody responses whereas if you expose them to a rarer adenovirus they won't secede any hexon proteins which have HVR's like that before so they won't be able to neutralize the viruses effectively. So by mutating those HVR's we can start to avoid that seroprevalence.
0:26:59 – Slide 29
So one example of this came out of the (inaudible) lab in the states and they actually progressed this virus all the way through Phase 1 clinical trial now as a vaccine vector for HIV. What they've done is they've taken Ad5 and they've mutated its HVR regions to look like those of adenovirus 48. And I won't go into the clinical trials because it's not really my data to present, but they seem to have some really exciting stuff there. And I'll just be talking about a slightly...
0:27:29 – Slide 30
...earlier paper from the beginning of the development of this Ad5.ENVA-48 virus. So on the left here we have mice which are naïve for adenovirus infection. Apologies. On the right we have mice which have been pre-exposed for Ad5 so they have anti-Ad5 immunity. So at the top what they've done is they've made these viruses so that they express HIV Gag protein and they're trying to assess how much of an immune response they raise against Gag after administration with these viruses. (0:28:04) So on the left we have regular Ad5. In the middle we have Ad5 which has only one of the HVR's changed for that of Ad48. And on the right we have Ad5 which has all of its HVR's changed for that of Ad48. So in the naïve mouse all those viruses produce a strong anti-Gag immune response as assessed by an ELISPOT assay with interferon gamma response and on the right we can see that the pre-existing Ad5 immunity is completely neutralized the Ad5 and the HVR48 1 virus but the HVR48(1-7) virus, that's the one with all of the changed HVR's has successfully provided an anti-Gag immunity because it's not been neutralized before it can have a therapeutic action. (0:28:50)The bottom two here show two doses of virus, again expressing that but these are measuring cytotoxic T-cell responses. So in the naïve mice we have a very strong response at the high infectious dose in all the viruses and at the low infectious dose 10 to the 7 VP at 5 and Ad5 HVR48(1-7) are having that successful immunogenic activity with Gag protein. (0:29:19) However, if we look again at our pre-stimulated mice, the ones which have already seen Ad5 we can see the high enough dose we managed to get a response against the two control viruses Ad38 and Ad48, so it's 35 and 48. We also managed to get one with HVR48 virus but Ad5 is completely neutralized even at the high dose. Excitingly though because you don't want to go giving people more virus than you need to at the lower dose Ad5 HVR 48 is still immunogenic in producing that cytotoxic T-cell response against HIV-Gag protein. (0:29:58) So it looks like the HVR pseudo typing approach is a very successful method of evading immune seroprevalence. And this is actually something which we are exploring in our lab with other viruses as well, both for cancer applications and to vaccine applications as well.
0:30:16 – Slide 31
So there we are. We've managed to ablate all viruses natural tropism and provide it with a therapeutic...
0:30:22 – Slide 32
...one and now we've found a way to avoid the problems with pre-existing immunity as well.
0:30:27 – Slide 33
So we've managed to target our virus. We've managed to deliver it to our therapeutic point. So how are we going to actually destroy these cancer cells? So I'm only going to touch on this briefly because it's a whole presentation in and of itself, but there are three broad modalities we can go for here. We can get enhanced viral lysis by using tumor specific promotors and we rely on the virus's preferential replication in those cancer cells under that control of that tumor specific promotor to pop those cancer cells. (0:30:59) We can also recruit the immune system by having our viruses in code stimulatory antigens and then we can actually change the immunological makeup of our tumor as well by expressing antibodies against immuno-oncology targets.
0:31:15 – Slide 34
So we can actually turn a virus into a triple threat. We can restore healthy phenotypes by targeting those immuno-oncology targets. We can go for selective replication of the therapy at the point of need by using two separate promotors and the virus's innate preference to cancer cells. And we can also express toxic transgenes. Then we can go for enhanced immune recruitment as well, both relying on the viruses own nature as a foreign pathogen so it's going to present viral antigens on MHC but also we can bring in transgenic antigens to further recruit the immune system.
0:31:53 – Slide 35
So to summarize everything that I've been talking about today, we've got this recombineering technology which is enormously powerful. It makes us able to engineer our virus almost without limit. And the only real consideration is that the virus has to still be able to assemble itself and it's not able to package DNA much in excess of 105% of its original genome in size which is around 30,000 base inaudible. We've hopefully have convinced you that these viruses can be engineered to target tissues of our predetermination. (0:32:29) And also, and this I think is really important one, we require a very detailed knowledge of the basic virology that underpins these viruses in order to go through rational design. There are irrational designs using directed evolution and that's how that inaudible one story works, but I've not really gone into that one today. However, those still require proper knowledge of the viruses even to decide just decide which ones you're going to perform a directed evolution experiment with. (0:32:59) I've also discussed how we can really target any mechanism on the cancer cell with these viruses. If it's cancer specific it can be a virotherapy target because we don't just have to relay on extra cellular targets. We can also manipulate the virus's genetic regulation to force proper regulation and up regulation of our virus within the cell itself. And this is strategy which is employed for the lot of the SV viruses which are in development and that's the fasciculus dimittis virus. (0:33:28) And I've talked about how these viral vectors can have their therapeutic effect in several different modalities, both as a physically lytic agent replicating within these cancer cells, as delivery vehicles for therapeutic transgenes and as a tool for immunotherapy as well.
0:33:48 – Slide 36
With that I just wanted to leave everyone with a list of the citations that I've put throughout this talk so please have a look at those papers if anything is of particular interest.
0:33:57 – Slide 37
And finally I'd just like to thank our lab funders and everyone who's been involved with the work that has been going on in our lab that I've been discussing today. And with that thank you again to Labroots for hosting and thank you for the invitation by gibco. I'll be taking questions now, so that's great. Thank you very much.
Moderator: Thank you, Alex for that informative presentation. We are now ready to start our live Q&A portion of our event and I just want to remind our audience members how to do so. Just simply type them into the ask a question box that is beneath Alex's bio picture and type your question in and hit send. We'll answer as many of our questions time permitting. (0:34:48) And if we do run out of time before we get to your questions, it will be answered via e-mail following the presentation. Okay, let's get started.
Alex, we have so many questions coming in, but I'd like to start with this one.
(0:35:04) Question: Where do you see the field heading in the future?
Yes, so that's a pretty common question. I think it sounds very cliché but in some ways we're already there actually because for a very long time the field has been focused only on safety. There was a very high profile death in a clinical trial in the late 90s which set the field back a very long way, but now we've kind of been vindicated. (0:35:34) We all went back to the drawing board. I say "we". I wasn't really in the field then, but we all went back to the drawing board. We rebuilt these viruses from the ground up for extreme safety and that's now been achieved. And we've actually had a clinically approved oncolytic virus for the first time last year. That's talimogene laherparepvec (T-VEC). (0:35:57) So we're shifting focus now towards efficacy. So there's a lot of trials right now in combinations of oncolytic virotherapies with existing chemotherapy agents, but what we're working towards in our lab is towards monotherapies with virus. So I think we're just going towards enhanced efficacy, enhanced specificity for our cancers and what's of course maintaining that safety aspect. (0:36:22) Though I think the direct evolution approach has showed a lot of promise, but with the enhanced understanding that we're gaining from these new studies we can really move towards rationally designed and extremely efficacy viruses in our next generation therapeutics. And with the understanding of clinical trials and our increasing understanding of cancer genetics immunology we can combine these immunological mechanisms with our viruses to fully exploit the host immune system and create that triple threat virus that I mentioned at the end there.
Moderator: Alex, I have a question from Nigel.
(0:37:02) Question: He would like to know where are these viruses sourced from?
That's actually kind of a broad question. So, yes, what's its name. Yes, the viruses themselves are isolated in the wild. There are a handful of people who go around characterizing rare and weird viruses. When they do they apply for taxonomic identification of them and they get fully classified. (0:37:32) So we've got 52 fully classified adenoviruses at the moment but there's 86 candidates as well. And they all get deposited in various viral banks around the world. So if we want to work with a specific strain, then we either have to go with the people who have just discovered it and isolated it or we have to go and just buy it from a bank. If you mean where are the therapeutic viruses sourced from, they all have to go through GNP grades production so as go to manufacturing practice to make sure that they're manufacturing in the absence of any toxins in a fully traceable so that they really, the drug grade virus. I hope that answers the question.
Moderator: Thank you, Alex. Now this question has two parts.
(0:38:21) Question: First did you or anyone else ever try this treatment in pancreatic cancer, because in pancreatic cancer the stromal barrier is the major hindrance for the immune system on the attack. Also since it's a CD8 and the CD4 T-cell mediated immune response, do you see perforin or Granzyme response along with the IFN gamma response? (0:38:45)
Right. I'll be 100% honest. I have not been running the pancreatic cancer experiments myself. It's been performed in both our lab by a post doc and also in the lab of Gunnel Hallden. It's pituitous timing actually. We published in combination with Gunnel Hallden today on the use of virus which we made using our vector technique in combination with Gunnel's approach using tumor specific promotors. So please have a look at that paper. It literally just went up this morning and that will definitely talk more about this. (0:39:28) As far as the immune response goes, we're still early days, so we need to best characterize these viruses in the presence of the immunology co-markers and we need to perform T-cell killing assays to proper assess this. As I say, I'm not the best person to answer that question, so if you would like to know more about that I think my LinkedIn is attached to my profile on Labroots, please send me an e-mail and I'll get you a better answer.
Moderator: Okay, our next question.
(0:40:06 ) Question: Cancer therapy is administrated over long periods addressing pre-existing immunity through a mutation. It's all well and good, but these viruses are still prone to acquiring immunity. How is adapted immunity addressed?
Yes, that's a really good question. So there are a couple of different answers to that. (0:40:28) First off, acquired immunity takes time to develop, so if we have a virus which is sufficiently efficacy then it should be able to infect everywhere that we need it to have its primary response very quickly. So that will be way faster than we could possibly raise in the data response against it. so we'll be able to get to our therapeutic action site long before that develops and it will have many, many days in order to actually have its therapeutic effect in lies those cells. (0:41:06) The other side of it is that as I've said we've moved on from this idea of viruses is a purely lytic agent. So we are talking about engaging host immune response as well. So hopefully what we can do is we can promote immune action against the cancerous tissue which will then create a more lasting anti-cancer effect long after the initial infection has subsided. (0:41:32) The other answer is that we can evade immune response in many other ways as well, so that follow up treatments can be performed with different viruses as well. So if we really perfect this methodology of cloaking from the immune system then we will end up with a battery of different viruses which can all act in the same way but will have different immune epitopes. Yes, so it's a combination effect basically.
Moderator: Very good. Okay, let's see.
(0:42:09) Question: Do you think there are any dangers involved in administering a virus capable of evading the immune system?
That's another really common question and it's obviously very valid. It depends on what you mean by evade because there are many considerations taken into account, so a lot of people will add immune stimulatory agents and I know people have in the past expressed concern with combining an immune evading combining—sorry—an immune evading virus with an immune stimulatory virus because the potential indication of that is run away auto-immunity. (0:42:53) So these things needs to be very specifically targeted and carefully related. The way that we retain safety is by making sure that our viruses are ultra-specific. So as long as we are not seeing prolific off target effects, then we should isolate that immune response to the tissues where it's actually wanted, i.e., the cancerous or unhealthy ones. The other thing we can do is render these viruses replication incompetent. (0:43:19) So instead of becoming a repetitious virus and acting like a virus, they act purely as an DNA shuttle. So we have a virus. It doesn't even have to have a virus genome if we really packaged it right, although that's not been tried just yet. That's my speculation. But we can use these things as a magic bullet so they just turn up at the cancer sites specifically, deliver their genetic payload and express therapeutic transgenes and we can include anything we like in that really within the size constraints placed on a inaudible viral genome. So, yes, there are dangers if it's done badly, but all of these will be addressed long before these things are ever see clinic.
Moderator: Alex, we have some attendees who they'd like to know...
(0:44:13) Question: How do you enhance immune response to the transgene but not the virus?
Right. So that can be achieved under its—so these viruses are very new but by encoding inaudible antigens you can raise an immune stimulatory response. So the obvious one will be the new immuno-oncology targets, so you could add immune checkpoint inhibitors. (0:44:42) So you could have your virus perhaps expressing antibodies or peptides which will inhibit this effect or you could actually encode proteins which will allow MHC to express immune stimulatory peptides, so there are many inaudible antigens which have been described such as 5T4 which has been trialed by yet. (0:45:06) So there's a drug called TroVax which was recently tried in Cardiff as well by the Godkin group who work upstairs for me and they saw some really exciting results with that. So hopefully by combining some of these existing treatments and also by exploiting some new mechanisms we can achieve that. (0:45:26) Also the virus itself is not always completely exposed, so the only part of a virus which the immune system can see is that external capsid. So I showed that image of the hexon protein and I'll just pull up the slide. There it should be loading. So that blue section of the hexon protein is exposed to the external environment. So you would expect that our immune system to raise a response again that. (0:45:56) However, that gray section is hidden from the immune system when it actually has an intact virus. However, once it goes into a cell it's still a foreign protein but then it gets spliced up and expressed on MHC and everything else, so we're still seeing expression of borrowed antigens from the cell surface even though the virus itself is cloaked.
(0:46:22) Question: Alex, do you believe that virus therapy is suitable for all cancer types?
That's a difficult question to answer because every cancer is unique. I'm sure I'm preaching to the choir here, but cancer is not one disease. Cancer is many diseases so being able to make a broad statement saying that every cancer is going to be treatable with one mechanism I think I would probably be lying. That said, with sufficient amount of engineering we can pick cancers off one by one by using mechanisms which are specific to each of them. (0:47:03) So I think so long as there is a mechanism by which we can specifically target that cancer type, then we should be able to raise a virotherapy against it. We just have to be careful in how we do it. I am sure, and well I know for a fact some cancers are more resilient than others. So some of them will lend themselves very nicely towards targeting so for example where I showed on one of the slides the list of various cancers which express avB6. I'll just pull that slide up. Yes, so I know pancreatic cancer, many of them express it but then others don't—sorry, I'm not being—there we are. (0:47:53) Okay, so it's cervical cancer. The vast majority of them express avB6 so perhaps a virus targeted in that method will be very efficacy in most cervical cancers. However if we look at liver cancer there, we see none of them were avB6 positive, so that virus would be an inappropriate therapy. Yes, so the answer is honestly it depends.
Moderator: Okay, thanks Alex.
(0:48:21) Question: At what stage of the cancer do you think this treatment will be more efficient?
Right. So there's a few things to consider there. I'm tempted to say as always earlier is better because the earlier you detect cancer the fewer cancer cells you have to destroy giving your therapy maximum efficacy. One of the really nice things about oncolytic virotherapy though is that a virus is a replicative agents so what can happen is—so this is the dream scenario where we've completely evaded neutralization. (0:49:02) We will be able to inject a person intravenously with a virus. It should self-spreads all around the body but is only capable of infecting the cancerous cells. Once it infects our cancerous cells it's probably going to go to the main—the primary tumor first of all because that's the biggest site so most of the virus is going to get mopped up there assuming we have sufficient extra vascularization in order for our virus to actually infect it. (0:49:31) Once it's there, we should see widespread lysis within that primary tumor which will then produce more and more progeny variance which will re-enter the blood stream and hopefully would be able to then whizz off around the blood stream and hit any metastases as well. So assume your cancer's not so advanced that the therapy is traumatic or anything like this. (0:49:58) Then I said therapy, assuming the disease is not so advanced that the therapy is in itself traumatic, we should be able to go off the metastases. So I would say earlier is better but not necessarily a barrier.
Moderator: Thank you, Alex. And we have some great questions coming in from our attendees. Let's go with this one.
(0:50:21) Question: Will this treatment be capable to achieve a cure as a monotherapy or would it be used as a supplement to other traditional cancer treatments such as surgery, chemo, radiation?
Right, so yes. At the moment the therapy is largely being considered in combination. That's because currently the field has been so focused on safety for so long that we've actually made our viruses kind of too safe. They aren't necessarily that lytic at the moment in the cancer on their own; however, they do seem to be very good at making our cancer cells vulnerable. So if we hit them with complimentary chemotherapy agents we can achieve that oncolytic effect but not of the monotherapy. (0:51:10) That's starting to change now because as I said earlier the field has sort of shifted focus and we're talking about using this as a monotherapy now. Again, this is going to be a cancer specific effect because the mechanism by which you target it and everything is going to impact the rate of replication of the virus, the specificity of replication, etc. So it will be a cancer dependent response. (0:51:38) For example if we target things using self-surface markers we can probably force a greater number of viruses into our tumor achieving a more potent lytic effect but perhaps, and this is very much a perhaps with genetic targeting, we would see that entry in many other cells but only replication in our cancer ones which means that less actual viral copies enter our cancer cells in the first place. So perhaps it might be a more protracted effort in order to achieve sufficient viral replication for lysis.
(0:52:15) Question: What could be the mode of drug delivery?
Right, so at the moment we generally—I say we, I don't run clinical trials. I'm the bench monkey. But, yes, we tend to use intratumoral delivery because we don't have to worry about these issues that I described with pre-existing immunity as much because these things are never exposed to the blood stream prior to their entry into the tumor. (0:52:48) But as I mentioned where we're really heading is towards intravenous delivery because that allows us to get a systemic administration of our therapeutic virus and to really reach every aspect of this. And I can't say too much about this at the moment, but watch our lab publications in the next couple of months.
Moderator: Thank you, Alex. We have time for just a few more questions
(0:53:19) Question: Can you tell me a little bit more about the recombineering technology that you mentioned previously?
Yes, sure. So I'm actually quite glad for an opportunity to talk about this because it's incredibly powerful technology and as far as I'm concerned everyone should be using it. So traditionally viruses are engineered by vectorization of the genome in a plasmid and then standards cloning techniques restriction enzymes digests are used to alter the viral genome. However, restriction digests are only available against certain sites and while there are Type II methods available they don't really solve the problem when you have such a massive viral genome. (0:53:19) So what this allows is a homology based cloning technique whereby we can generate a cassette which is specific against any aspect of—I'm sorry, against any DNA site within our giant plasmid or bacterial artificial chromosome. So that means that any region of our viral genome is up for modification with no additional steps. (0:54:30) So first of all we create homology arms so short sequences of DNA which are homologous to the area of the virus which we want to modify and between those arms we have a cassette which allows us to make sure that we've integrated our DNA to the correct place. So we put that setting, we use antibiotic selection to make sure that our cassette has gone in and then we use the same homology arms again but this time with our edited sequence, whatever that may be, to insert a mutation at the site we wish—sorry, at the site whether that cassette now is. (0:55:10) So it's first cut and then stick and then we do that again to cut and stick out a marker and stick in our mutant gene. And we can do that at any site. We can change the size of the genes as well, etc, etc. Yes, that was actually developed some years ago now by Dr. Richard Stanton and Gavin Wilkinson at Cardiff and it's since been adopted by labs all over the world actually.
(0:55:39) Question: What applications do you think these vectors have beyond cancer?
So beyond cancer is an interesting question and I start to tread on the toes of the gene therapists then. There is an emerging—well, some people probably resent me saying emerging field, but there's a field called cancer gene therapy as well which works more like treating cancer as a magic bullet as I mentioned earlier. So using these viruses as shuttle vectors rather than oncolytic agents. The sky's the limit with viral engineering. (0:56:16) There are enormous assets out there at the moment to engineer DNA delivery vehicles from the ground up. And the closer they get to an effective DNA delivery vehicle the more the constructs start to look like viruses with things like decorated liposomes and DNA which is encased in various chemicals and proteins. (0:56:43) So if you consider that a virus is really just a mechanism of delivering DNA to where you want it to be it opens you up to everything, because you can then change the very makeup of your cells from the outside. Obviously that is potentially dangerous. You don't want to introduce degenerative mutations, you don't want to cause disease but you can have corrective mutations as well which is what people are doing with adenoassociated viruses and many others and lentivectors and everything at the moment. So any application where you want some DNA and engineered virus is probably your best candidate.
(0:57:22) Question: Alex, chemotherapy is known as you know to make many patients immunosuppressed. Would that effect the efficacy of the viruses in its attempt to hijack the immune system if used on conjunction with chemotherapy?
Right. So that's where the triple threat that I was talking about comes in. So if you are working with a virus which works purely by mean stimulation, so it's one of those magic bullet therapies that I was talking about although even that's not entirely a monotherapy in itself. (0:58:03) Then you would have some problems there because you would have no method of cancer cell killing without the immunity. However, the virus itself is also a lytic agent so if there is absolutely no immune response then our virus is still going to be replicating it in those cancerous cells, specifically in the cancerous cells as well might I add. (0:58:29) So it's only going spread from there unchecked. So you're not going to see—sorry, I'm not being very eloquent here, right. So what you're going to see is your virus is still continuing to replicate in lies those cancerous cells even in the absence of an immune response and there's no immune response to suppress the viral spread but that's not a problem because our virus has been rendered specific for cancer cell replication. (0:58:57) So it would still have therapy detection even in the absence of the immune system but again this is where we come back to that idea of cancer specific virotherapies because it depends entirely on your cancer's makeup how your virus is going to actually impact it.
Moderator: Alex, we have time for one more question and I'd like to end with this one.
(0:59:24) Question: Do you think that there are any dangers involved in administering a virus capable of evading the immune system?
I think I answered a similar question earlier actually. Yes, so if your virus was poorly engineered, yes. But your virus also, let's start with these viruses are not super pathogenic agents. We're not talking about the stuff that you see on the news. We're talking about the stuff which maybe gives you a sniffle. (0:59:58) So once of course these things are more dangerous if you're completely immunocompromised and everything else because they can continue replicating unchecked. That would be a problem. However, our viruses are engineered in such a way that normal healthy cells are completely able to either suppress these viruses but more to the point they don't even get infected in the first place. So if the virus is not able to infect healthy cells and your virus is not able to replicate in healthy cells you've got a double layer of protection against these things even in the absence of an immune response. (1:00:34) So all of these things are addressed pre-clinically as well but the safety mechanisms are already in place and we are vindicated in that these things do work with the now approved virotherapy T-VEC.
Moderator: Thank you again, Alex for this great live Q&A session. Do you have any final comments that you would like to leave our audience with?
No, not really. I'd just like to say thank you very much for the opportunity to speak today and it was brilliant to get some really interesting insightful questions there. Please do get in touch with us if you have any further questions. I'm more than happy to answer any other questions by e-mail and if you have any other ideas please do comment for us. It's been brilliant, thank you.
Moderator: I would like to once again thank Mr. Alexander t. Baker for his presentation and his important research. I'd also like to thank Labroots and Thermo Fisher Scientific for making today's educational webcast possible. Now before we go I want to remind everyone that today's webcast will be available for on demand viewing through July 2018. You will receive an e-mail from Labroots alerting you when this webcast is available for replay. We encourage you to share that e-mail with your colleagues who may have missed today's live event. That's all for now and we thank you for joining us and hope to see you again soon. Goodbye.
End Presentation: 62:13
Get to know Alex
Why did you choose cancer research?
Cancer research enables me to chase my interest in biological sciences and contribute to helping those who otherwise wouldn’t have any recourse. Also, cancer is one of the most difficult scientific problems we face (as it isn’t really just one problem), and I can never turn down a good challenge!
What motivates you to succeed in your field?
Being the first person to discover something new will always excite me; but seeing work that I’m involved in start to positively impact people’s lives is always incredibly heartening. Even when all the experiments are going wrong, knowing that their results can really go on to save lives makes it easy to try one more time.
What are your top 3 favorite things to do outside of the lab?
White water kayaking, mixed martial arts, cooking
What role have the mentors you’ve had in your passion for basic research?
I’ve been very fortunate to have mentors who encourage me to chase my own ideas. This has really let me broaden my interests and make connections I otherwise wouldn’t have been able to, both in terms of scientific ideas and people all over the world. None of what I have achieved would be possible without them.
Which scientist current or past would you most like to meet and why?
Can I meet a scientist in the future? It would save me a lot of time!
Why did you become a scientist?
Cliché as it may sound, I don’t think I ever quite grew out of the phase of the curious kid asking “Why?” to everything. So, when science was able to help me answer some of those questions it completely captured my imagination and interest. With every answer it gives come a dozen new questions, so the question quickly became not whether I wanted to do science, but just what kind of scientist I wanted to be.
If you didn’t have to sleep, what would you do with the extra time?
I like to think I’d learn to play the piano, and how to program… but there’s a real chance I’d become the world expert on Netflix.
How do you relax after a hard day of work?
Head to the gym for an hour, cook a big dinner, and collapse with a book or TV show on the sofa.
I want to be the next Gibco Cell Culture Hero
Do you want your research to be highlighted as our next Gibco Cell Culture Hero? Complete the form below for a chance to have your PhD or Post-Doc research featured in our webinar series.
Links to content or other Internet sites should not be construed as an endorsement of the organizations, entities, views or content contained therein. The opinions and/or views expressed on social media platforms represent the thoughts of the individual and online communities, and not those necessarily of Thermo Fisher Scientific.
For Research Use Only. Not for use in diagnostic procedures.