Tocagen Inc (TOCA)

[gr8db8]

Understanding Patient Response to Improve Therapies with Dan Hogan

https://cofactorgenomics.com/wk-34-patient-response-therapies/

David: I’m really excited about the mission of Tocagen and hearing more about Dan’s work. Tocagen develops immune-oncology candidates for therapeutics and they’ve got a great tagline, “no one should die of cancer.” Dan, it’s great to have you here.

Dan: Thank you. At Tocagen, we utilize a replicating retrovirus that selectively infects tumor cells. Then we encode a gene within that retrovirus that we can inject into a patient’s tumor or even through an IV. The virus enters tumor cells, selectively infects them, and inserts that DNA into the cancer cell genomes and expresses whatever gene we put in there.

Our lead therapy is at the tail end of a registrational phase III trial in recurrent Glioblastoma, is called Toca 511. In this case, we have the replicating retrovirus encoding a gene or cytosine deaminase, which will convert an antifungal drug, which the patient takes in the form of a pill, 5-chlorocytosine, into a chemotherapeutic, 5-fluorouracil. This will happen specifically in the tumor. It then leads to a patient response of direct self-killing of the tumor cells. In addition to that direct self-killing, it elicits an immune response due to the direct cell killing, as well as a release of viral antigens, killing of some of the suppressor immune cells in the tumor. In a nutshell, that’s what Tocagen does.

My role is primarily to lead the bioinformatics group. Part of that is doing the usual genomics correlation studies as a part of clinical trials, to better understand or characterize the tumor molecular profiles of the patients and relate that information to clinical outcomes. The goal is to utilize that information to better understand the mechanism and the mechanism of action, to improve the therapy. We also try to understand patient response – what distinguishes patients that respond very well versus those that don’t?

David: What are some of the things that you are learning? Both through research in general, but also through these clinical trials identifying patient response and who’s responding to what. That’s the big challenge of precision medicine, figuring out who’s in the 30% and who’s in the 70% and how you can dial in treatments to better serve those different populations.

Dan: Most of the work we’ve done has been with phase I studies. This is tough for a variety of reasons. In our phase I study, about 20-25% of patients showed an objective clinical response by MRI, meaning after treatment, the tumor would start to grow back but then regressed or stop growing. Those that showed durable patient response lived for quite an extended period of time, so we’ve spent a lot of time trying to understand what’s special about those patients. We have some leads from tumor profiling that we’ve done that suggest there are some intriguing differences in the immune cell populations.

The tumors are different in the patients that responded than in the ones who didn’t. It is important to remember, it’s a small sample size, and it’s phase I, so there’s no control arm. There are many caveats to that, but we are taking that data and the framework for distinguishing patient response. Now we’re going to apply that to the phase III trial. In that trial we have about 400 patients that are split one-to-one on Tocagen treatment versus the standard of care. We have done the RNA and DNA sequencing on all those patient’s tumors, as well as peripheral blood monitoring, so we’re in a much better position with the phase III study to have the power to find something significant.

David: Let’s talk a little bit about where samples are being derived from, whether it’s for diagnostic or monitoring purposes. Looking at liquid biopsy and a solid tissue allows for different angles. So how can we tie what we learned at the side of the tumor to what we see in the biofluids, and how are you approaching these different sources?

Dan: Glioblastoma is a particularly tough one to correlate what’s happening in the tumor with the blood. There’s been headway made in other solid tumor indications such as lung and melanoma. They longitudinally profile the blood for specific mutations as a way to monitor disease progression, which shows a heck of a lot of potential. Glioblastoma is difficult for a few reasons. We don’t have complete understanding of how the immune system works in the brain, either in a healthy patient or one with Glioblastoma. On top of that, Glioblastoma doesn’t seem to shed much DNA, so there is no way now to monitor either specific mutations or anything else peripherally. It’s an uphill battle. We’re not able to apply any methodology that’s being developed in other solid tumors for Glioblastoma. At least for us, it’ll be a question of trying to carefully connect dots.

David: What about some of the other tumor types? You mentioned lung and melanoma being two areas where oncology has made huge progress over the years, but with the addition of breast and colorectal cancer, they make up the majority of the literature in the field. There’s a huge number of other cancer types. I know work is being done, but where are we with these approaches to other cancer types?

Dan: Therapeutically, there’s been success in melanoma, lung, and colorectal cancer. I think the great breakthrough was checkpoint inhibitors, which essentially take the brakes off T-cells, but those are just a tiny part of what we can do. In tumors like lung and melanoma that have high mutation levels and fairly high levels of T-cell infiltrates that are being suppressed, taking the brakes off is enough. That’s why we’ve seen great success stories there. Even within those tumors, of course, there’s a lot of heterogeneity from patient to patient, and no pre-existing sets of T-cells in the tumor. How are we going to get the immune system going there? There are lots of trials going on with various approaches to solve this.

CAR-T cells have a lot of potential, but you you need specific antigens. Some of the myeloid suppressor cells such as the MDSC, are very sensitive to 5FU, which is one of the reasons our therapy works. I think there’s a lot of promise in other solid tumor types. We have the ability to collect so much data within the tumor, as well as peripherally, with next-gen sequencing and other emerging methods. This lets us monitor the immune system and measure patient response at an unprecedented level.

David: There’s a shift from focusing on DNA to RNA through RNA-seq in a more holistic, omics approach in nucleic acids, genomics, proteomics, metabolomics and others. Let’s talk about sequencing and its role in today’s immune oncology landscape.

Dan: I think of RNA as a diagnostic, but it’s lagged behind DNA. There’s just more complexity to it. Honing in on specificity and signal over noise takes a larger sample size and more time. Therapeutically, we’re starting to utilize RNA-based therapeutics, oligonucleotides mRNA-based therapeutics, and gene therapy, all of which have great potential.

David: There’s so much talk about big data, machine learning, and artificial intelligence. What are some of the tools and approaches that the scientific community needs to be using in order to handle this kind of exponentially growing massive data?

Dan: We do all the heavy lifting on cloud based servers for processing the data, and the subsequent analyses on a local Linux server. Integrating data from different sources and performing QC is half the battle. It’s the boring half, but an important one. One challenge is that the public data from other labs tends not to be that helpful for us. We’ve spent so much taxpayer money generating these giant data sets, we need to figure out how to do it in a way that’s more universally useful. Part of that is doing a better job collecting metadata on the patients and their clinical profiles. It’s very difficult because with clinical trials, there’s no obligation to share patient data, and there’s also concerns with patient privacy and intellectual property. We need to ensure that information is useful not just for us and for our therapy, but for the broader community. Then, how can that information be incorporated into subsequent trials by others to help move the field forward? Hopefully FDA will start requiring certain elements of the trials to be published or accessible to others.

David: I’d like to bring it back to biomarkers and the specific things and we may see in the future. Is there anything that you’re really excited about in terms of specific biomarkers or pathways?

Dan: I’m in the brain cancer space and I’m hopeful about merging imaging techniques with genomic data sets. I think it’s even a term called radio genomics where you can go back and do correlation studies between MRI, PET, or other types of scans with genomic features, then use that to sort the imaging as a way to get proxies on the molecular profiles of the tumors. There’s a lot of promise there. Also, some of the spatial transcriptome profiling methods may hold a lot of insights into tumor heterogeneity and the role of the architecture fit within the tumors and how that may be affecting patient response to immunotherapies. I’m also excited about the ability to cheaply and quickly monitor the immune system at any given time. This would be valuable for disease and healthy states to provide a lot of information that we will be able to utilize for next generation early and noninvasive diagnostics.

David: I’m excited to see this next iteration of immune-oncology and precision medicine. Thanks for all of your insight and it’s been a lot of fun. Tell people where they can go to find out more about you and Tocagen.

Dan: Tocagen has a pretty rich source of peer reviewed papers that others can check out. Keep an eye out for our results on the phase III trial over the next month or two. It’s going to be a little big inflection point for the company and hopefully the field. In this field, the needle hasn’t moved so patients and physicians are desperate for something that shows promise and fills that progress, so hopefully that’s what we’re doing.

David: We’ll be looking forward to seeing those results. Thanks again for your time. It’s been a real pleasure.