Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
It is an interesting study indeed and the small sample size still yields tantalizing data. The combination that was used, ie radiotherapy and FLT3-ligand, is a rational approach. The radiotherapy induces low level cancer cell death which ends up releasing tumor specific antigens. The FLT3-ligand leads to a proliferation of dendritic cells, and in the presence of the released antigens, you would theoretically observe greater numbers of mature antigen presenting cells. This would improve priming for CTLs and, in my opinion, would help to overcome immune suppression caused by CTLA-4 on Tregs. I think you can numerically overwhelm CTLA-4 on Tregs intratumorally with the presence of high densities of mature dendritic cells.
Also, radiotherapy, when used the right way in a very targeted manner, can be a friend to in-situ vaccine approaches. Oncosec's electroporation platform by itself appears to accomplish a similar release of tumor specific antigens; this effect was observed in many of their preclinical datasets. By themselves, radiotherapy and electroporation don't typically accomplish much, but they definitely contribute to the very early stages of the immunity cycle. Without adequate tumor specific antigen release, there wouldn't be much of an immune response. Their recognition is essential for any in-situ cancer vaccine.
The 4-month PFS endpoint seems short, but I imagine disease progression is common within that timeframe for advanced stages of NSCLC.
This is a great find danger - thanks for sharing!
The Epac plus Keytruda failure was predictable. There won't be much improvement in PFS or OS with Epac if it is being used with an anti-PD-1 agent alone. The problem is timing: patients need to have a preexisting immune response at baseline in order for these agents to work well, and neither one of the two drugs primes anything.
Here is what I had previously discussed regarding IDO inhibitors:
IDO1 inhibitors will always fail in patients who are not already primed. If there is no interferon gamma production to begin with, then there is not going to be much upregulation of IDO. It is like using a PD-1 antagonist in patients who don't have high percentages of PD-1 expressing CTLs at baseline. So many of these combination immunotherapies aren't going to make much of a difference if they don't address the priming phase in the immune cycle.
Timing is everything, and this really holds true with IDO1 inhibitors.
Some perspective on patient responses to pembrolizumab monotherapy following treatment failures. GLTAETS
https://meetinglibrary.asco.org/record/152162/abstract
In case anyone missed it, O'Connor mentioned treating lymphoid tissue with a costimulatory agent. Why do this? Because that is where antigen presentation occurs to naive T cells.
The forthcoming TNBC data remind me of the Immunopulse IL-12 monotherapy trial where patients going directly from IL-12 to anti-PD-1 agents had profound responses in advanced melanoma. That trial demonstrated around 75% ORR and I think 50% complete responders in the subset. I'm not saying the TNBC study will observe those same percentages - after all, the trial is tiny and only designed for immune monitoring data, but there might be some very tantalizing data coming out next month. To maintain perspective, approximately 5% of heavily pretreated TNBC patients respond to anti-PD-1 or anti-PD-L1 agents and the complete response rate is well below 1%.
There is likely a very good reason why O'Connor wants to advance TAVO quickly through a combination study in TNBC.
I think the way we ought to look at OX40 is that it serves to protect the longevity of T cells; it helps effector and regulatory T cells survive after they are activated.
The good news is that OX40 antibodies facilitate the development of memory T cells, much like what we are observing in some cases with IL-12.
The bad news is that if you don't already have adequate effector cells present, then you are basically allowing Tregs to survive longer than normal and accumulate. This makes T cell priming rather difficult when those Tregs accumulate and express CTLA-4.
The timing with an OX40 antibody is going to be essential, and you have to make sure you have adequate CD8 and CD4 effector cells present before administration of the antibody. This is why you'll tend to see that vaccines coupled with OX40 antibodies are more effective than OX40 antibodies with anti-PD-1.
You have been here for quite some time Wait - I appreciate your historical perspective. There have indeed been several notable platform shifts over those past six years, but I am glad they were flexible enough to make the transitions brief instead of advancing worthless programs through late stage studies.
At the present moment, the company has locked into an intratumoral electroporation platform that I think will create tremendous value over upcoming months. Immunopulse IL-12 is undoubtedly converting many cold tumors hot, i.e. priming immune responses through in situ vaccination. THEY ARE TAKING ADVANTAGE OF TUMORS' COMPLETE REPERTOIRE OF ANTIGENS. They aren't able to convert all patients into responders yet, but the more they tweak the multigene construct the closer they will get. And they are getting extremely close to addressing what I think has been the only limitation - how to prime immunity when there are Tregs present intratumorally and in tumor draining lymph nodes. Their first multigene construct may in fact be one solution, and another being the administration of anti-CTLA-4 intratumorally and/or in tumor draining lymph nodes. Systemic administration of an anti-PD-(L)1 will always be needed in advanced stage solid cancers. For them to consider licensing an encoded anti-PD-L1 affimer for perhaps the systemic component of their immunotherapy tells me that they are looking to make current checkpoint inhibitors obsolete. Imagine that for a moment!
Thanks for sharing those links Jon! I completely agree that checkpoint inhibitor therapies can and need to be used intratumorally, but for the purposes of driving a strong systemic immune response. That is, in my opinion, going to be the optimal way to develop an abscopal effect that results in adequate CTL in all lesions, both local and distant.
Once patients drive adequate CTL into all distant lesions they would then be 'primed' for a response. Because those CTL are activated, they will have PD-1 surface expression and will be releasing interferon gamma intratumorally in those distant lesions. If significant, that interferon gamma will lead to upregulation of PD-L1 on tumor cells as well as IDO enzymes on APC and tumor cells, thus blunting responses in those distant lesions. That's why I say that those particular checkpoint inhibitors need to be administered systemically. While not all patients seem to have that blunted effect with distant lesions, there are many that do.
Again, this is not to say that those checkpoint inhibitors shouldn't be used intratumorally as well. I have been saying this for nearly two years - if you want a strong abscopal effect and faster response time, then administer checkpoint inhibitors intratumorally as well. But I think the intratumoral therapy needs to precede the systemic administration of anti-PD-(L)1 and perhaps IDO inhibition. Again, intratumoral therapy is used to prime immune responses and systemic therapy is used to 'release the brakes' on existing CTL that are being suppressed.
Bottom line is I agree, but with the caveat that a couple checkpoint inhibitors also need to be administered systemically, especially in solid tumors.
IDO inhibitors systemically delivered in combination with anti-PD-1 agents are no doubt helping some patients achieve responses, especially those with advanced melanoma. I think IDO inhibitors absolutely need to be administered systemically just like anti-PD-(L)1 agents. Any place there are CD8 T cells present intratumorally, there will likely be some IDO as well. For some patients the IDO will just be a negligible effect, but for others it will definitely lessen effects from CTLs.
Anti-CTLA-4 is one checkpoint inhibitor where it makes sense to deliver intratumorally or in primary tumor draining lymph nodes. I say that because those are the places where cancer antigens are presented to T cells for priming, and those are also the places where CTLA-4 on Tregs will hijack CD80 and CD86 on mature dendritic cells. Overcoming this initial immune suppression will allow for T cell priming to occur. Once that happens, activated CD8 T cells seek out and destroy cancer cells with matching antigens. This becomes the abscopal effect. Without that priming, there would be no abscopal effect. To affect distant lesions, you have to prime the immune system locally, specifically where all cancer antigens are being presented to naive T cells. Eliminating the Tregs or specifically inhibiting CTLA-4 expression on the Tregs in the TME or primary tumor draining lymph node will be much safer and far more effective than a systemic route. Oncosec has the opportunity to encode anti-CTLA-4 on a plasmid with other genes. I am not certain they will do it anytime soon, but if they do get to it I think they will discover that protracted release of CTLA-4 antibodies encoded and delivered intratumorally or in the primary tumor draining lymph node will lead to complete responses across many solid tumors.
There are many studies coming out now that show how predictive CD8 TIL densities can be for many different solid tumor cancers. The only way to drive an increase in those CD8 TIL is to allow T cell priming to occur - that is so essential yet seemingly still overlooked. The priming phase in the immune cycle definitely should not be ignored, nor should the CTLA-4 on Tregs that appear to be preventing that priming. I've said this before on this forum, and I'll say it again and again and again, any company that can deliver an anti-CTLA-4 agent intratumorally and in a protracted manner will be worth many billions of dollars. Will Oncosec do it?
My thoughts exactly. Also, no one is encoding a DNA version of anti-PD-L1 affimer and delivering it systemically via electroporation. This definitely appears to be an Oncosec move. If it works well, then I could see Oncosec getting into a licensing agreement with Avacta.
I believe the main reason Oncosec isn't just encoding a DNA antibody version of the PD-L1 or PD-1 on its own is due to Inovio's IP for systemic delivery of DNA-encoded antibodies. An affimer version would get around this IP. I had originally thought minicircles would be their approach to systemic delivery of an anti-PD(L)1, but an affimer license agreement with Avacta may be more feasible and more readily available. Affimers definitely appear to have some advantages over recombinant antibodies, which you summarized well.
Hey Danger, thanks for contributing to the topic of Tregs. My interpretation of the immunosuppressive factors (i.e. depletion of tryptophan) associated with IDO1 is that the enzyme becomes upregulated in the presence of interferon gamma - your articles also support this notion. There appears to be a very strong positive correlation between IDO1 enzyme production and interferon gamma levels. This is also true of PD-L1 upregulation on tumor cells and dendritic cells in the presence of interferon gamma. If indeed interferon gamma is driving the upregulation, then it stands to reason that there needs to be a pre-existing immune response present prior to administration of an IDO1 inhibitor. The same is true for anti-PD-1 and anti-PD-L1 checkpoint inhibitors as well - they simply won't work if there is no pre-existing immune infiltrate and upregulation on tumor cells.
Because CTLA-4 is constitutively present on Tregs, I think its immune suppressive qualities are found at the onset of an immune response. I am very confident that a CTLA-4 inhibitor delivered intratumorally and/or in a primary draining lymph node would drive significant immune responses for most patients and allow for expansion of CD8 Tcells intratumorally (this may also be achieved through a numerically-superior intratumoral presence of both mature and immature dendritic cells as well). Once an immune response really starts to get going, that's when IDO1 and PD-1/PD-L1 become overexpressed. I think CTLA-4 on Tregs are ultimately preventing the expansion of cytotoxic t-lymphocytes. Again, without the CTLs present to begin with, there would be no justifiable reason to use anti-PD-1/anti-PD-L1 or IDO1 inhibitors.
As for PD-L1 presence on dendritic cells, I have read several accounts where the effect is negligible in contrast to PD-L1 expression on tumor cells. Also, if there are patients with adequate PD-1 on immune cells and relatively little to no expression of PD-L1 on tumor cells, then there would be no checkpoint; therefore, patients would respond as long as there were adequate CTLs present. An anti-PD-L1 agent might receive credit for responses in patients with no PD-L1 present on tumor cells - and some might attribute the response to PD-L1 present on dendritic cells - but I don't think that's what's really going on. I think patients with limited or no PD-L1 expression on tumor cells are responding because there are adequate numbers of CTLs - not all patients will have upregulation of PD-L1 on tumor cells in the presence of CTLs.
I think Oncosec is interested in the affimer platform more for the systemic delivery of anti-PD-L1. Oncosec's intratumoral multigene product(s) would be complementary to the anti-PD-L1 affimer.
Oncosec has the capability to encode checkpoint inhibitors on plasmids, but I am pretty sure they would run into IP issues if they commercialized them for systemic delivery- I think Inovio has patents regarding systemic delivery of DNA encoded antibodies. Oncosec's strength and IP portfolio is in intratumoral administration rather than systemic. Their multigene product isn't going anywhere and I continue to believe it will be extremely disruptive.
If a DNA encoded anti-PD-L1 affimer shows efficacy and a good safety profile when administered systemically, imagine the potential loss of revenues for Keytruda, Nivolumab, Tecentriq, etc. This creates a position of strength in terms of negotiations with Merck, BMY, Roche, etc.
Could get really interesting around here if Avacta and Oncosec start encoding an anti-PD-L1 affimer protein and delivering it systemically. Boy is it gonna make heads turn if they achieve proof of concept with a systemic application. Systemically delivered encoded anti-PD-L1 affimer with intratumoral multigene?
It looks like there is some serious interest in affimers, partly because they are one-tenth the size of antibodies, yet still might achieve the same therapeutic benefits and a safer profile. Now, encoding them and allowing cells to manufacture them following electroporation would take the affimer platform a step further. This is quite interesting - I didn't see this coming at all.
I have been wondering for quite some time if and how Oncosec was gonna try to encode a systemic anti-PD-1 - thought it might have been with a minicircle DNA vector, but this might be the answer.
You hit the nail on the head.
Anti-PD-1 agents won't work if you don't have appropriate percentages of TIL and PD-L1 expression on tumor cells. They are probably driving cancer antigen, i.e. neoantigen, specific immune responses. This is essential to any response using immunotherapy. The IL-12 also leads to upregulation of PD-L1 on cancer cells. This sets the stage for anti-PD-1 agents. Now it appears to work with TNBC.
The reported information is tiny and anecddotal, yet very intriguing.
Perhaps it has more to do with the fact that they would also be using a new electroporation device with the second product candidate (which is definitely a PIIM construct). We have seen some preclinical data on one PIIM, but it still isn't clear if that is actually going to be the second product candidate. Who knows, maybe they are performing preclinical studies in multiple cancer models.
I don't suspect the IST designs are holding things back on the announcement of a second product candidate - these seem to be straight up Immunopulse IL-12 with keytruda, with the exception of the SCCHN trial which will have a third, unknown immunotherapy agent.
Yeah, it's completely under wraps for whatever reason.
You bet. I think an anti-CTLA-4 agent's greatest contribution to combination cancer immunotherapies will be observed when it is applied directly in the tumor microenvironment and/or in a tumor draining lymph node - those are the places you find the mature dendritic cells in the company of Tregs, and that's where antigen presentation gets hijacked by CTLA-4 on Tregs.
To maximize anti-CTLA-4 efficacy, it likely needs to be applied in a protracted manner. I say that because these patients could miss a "window of opportunity" when neoantigens are released from dying cancer cells. If there are CTLA-4 Tregs present in proximity to dendritic cells when they mobilize, then an opportunity could be missed to activate CD8 T cells.
As I mentioned before, there is a possible way around this without relying on an anti-CTLA-4 agent: you increase the number of dendritic cells - both mature and immature - intratumorally to 1) numerically overwhelm CTLA-4 on Tregs (requires mature dendritic cells) - a diversion tactic, and 2) to capture more cancer neoantigens following cancer cell death with immature dendritic cells.
Oncosec has the opportunity to accomplish both approaches, and potentially even in a combined fashion.
Thanks for sharing. CTLA-4 is constitutively expressed on FOXP3-positive Tregs as well. We know CTLA-4 has a greater affinity to CD80 and CD86 on dendritic cells than CD28. In fact, there is evidence of CTLA-4 physically removing CD80 and CD86 on dendritic cells, thus leaving them completely unable to interact with CD28 on CD8 T cells. This ultimately prevents activation/stimulation of immune responses specific to the antigens presented by dendritic cells. CTLA-4 hijacks antigen presentation and they mostly do it in the tumor microenvironment and tumor draining lymph nodes.
Hargrove, I really do believe this first multi gene construct is going to be disruptive. The single issue I have with Immunopulse IL-12 is that it doesn't address overcoming the initial stages of immune suppression. That is, having too many Tregs at the start of an immune response isn't a good thing. Obviously, this isn't a problem for perhaps the majority of patients as we have seen remarkable complete responses so far. However, for those patients who haven't responded to the combination of Immunopulse IL-12 and Keytruda, it seems to be clear that they aren't activating enough CTL. I attribute this to too many intratumoral Tregs expressing CTLA-4.
I think this first PIIM construct will allow for APCs, like dendritic cells, to numerically overwhelm Tregs and their associated CTLA-4-derived suppressor effects. In theory, you may not actually need an anti-CTLA-4 agent. This new multi gene construct should lead to improvements in CD8 T cell activation and proliferation. I am actually so optimistic that I think it will convert nearly all anti-PD-1 nonresponders into responders, not just in metastatic melanoma but also in many other solid tumor cancers.
There is a very strong rationale for in situ vaccines. This is a good overview of the advantages and logic:
https://academic.oup.com/annonc/article/28/suppl_12/xii33/474l
Oncosec appears to be developing their multigene combinations through rational design, i.e. addressing multiple components of the immunity cycle intratumorally to ultimately drive systemic immune responses.
I continue to think Oncosec is going to create a multigene product that will work off the shelf with any solid tumor whether it be cutaneous, subcutaneous or visceral. They might even be able to achieve such a feat with its first multigene product used with GENESIS.
There might be other sources of funding UCSF could tap into as well if it is an IST.
IL-2 probably has an opposite effect on FOXP3 Tregs, i.e. it tends to increase the number of FOXP3 Tregs. Th1 responses, which are promoted by IL-12, are characterized in part by their release of IL-2. This suggests that IL-12 not only leads to a proinflammatory response, but also contributes to increases in Tregs through IL-2 production vis-a-vis Th1 phenotype promotion.
In addition, Th1 responses produce interferon gamma, which is also a double edged sword; it too not only contributes to an inflammatory immune response, but also leads to the upregulation of PD-L1 on tumor cells and quite possibly CTLA-4 on T cells, including Tregs. Therefore, driving immunity also means you are driving increases in Tregs. This isn't much of a problem once the number of cytotoxic T cells have reached an adequate threshold, but in the early stages of immunity, if you have too many Tregs that are also expressing CTLA-4 you will likely observe inadequate cytotoxic T cell activation. And I believe this occurs because the CTLA-4 on Tregs is outcompeting CD28 for ligands (CD80/CD86) on dendritic cells.
Assuming there are adequate numbers of intratumoral cytotoxic T lymphocytes present, and these cells are of the partially exhausted phenotype, e.g.expressing PD-1, and the tumor cells express PD-L1, then an anti-PD-1 agent like Keytruda will generally work very well. But you absolutely need these CTLs present in order for Keytruda to work.
Oncosec is driving immunity through an in situ vaccination approach. This platform is generally well tolerated because the proteins encoded by the plasmid products are being released locally, in the tumor microenvironment. These proteins synergize to drive a systemic immune response.
Is anyone here familiar with liver and/or ocular immune-privileged properties?
https://www.medpagetoday.com/reading-room/asco/immunotherapy/69883
Whole-body immune priming with in situ vaccine improves response in refractory patients
SAVESAVED
by Mark L. Fuerst
Contributing Writer, MedPage Today
[quote
Advanced melanoma patients who do not respond to programmed death-1 (PD-1) inhibitors show promising responses to a combination of an immune primer plus anti-PD-1 therapy.
The study demonstrates that it is possible to generate whole-body immune priming using an in situ vaccination approach with minimal systemic side effects.
"Injecting an interleukin[IL]-12 plasmid and then electroporating tumors led to an increase of anti-cancer immune signals and immune cells in the tumor and throughout the body," the study's lead author, Alain Algazi, MD, of the University of California San Francisco, told MedPage Today. "This effect was even more pronounced when tavokinogene telseplasmid was administered in combination with the PD-1 antibody pembrolizumab, even though we selected patients for the study who were predicted to be non-responders to pembrolizumab monotherapy."
The PD-1 antibodies pembrolizumab and nivolumab have enabled thousands of metastatic melanoma patients to achieve long-lasting remissions with relatively few side effects, but these agents do not work in the majority of metastatic melanoma patients, he explained. "The bottleneck for response in many patients is a lack of intratumoral immune cells and immune signals -- that is, an immunologically 'cold' tumor. If we can convert these 'cold' tumors into 'hot' tumors with anticancer immune cells and signals, we can help patients who would not otherwise respond to PD-1 monotherapy achieve long-term remission."
There is an unmet medical need in patients with melanoma who are refractory or have relapsed on anti-PD-1 therapies. "Patients who do not respond to PD-1 antibody monotherapy have few options, with a significant chance of inducing long-term remission," Algazi continued. "BRAF-inhibitor combinations have a median progression-free survival of less than 1 year in BRAF-mutant melanoma patients. The combination of nivolumab and ipilimumab may capture additional responses, but there is a high risk of severe toxicity -- 70% grade 3 and 4 adverse events -- so this option is not appropriate for many patients."
Previously, studies showed that intramural delivery of plasmid IL-12 by electroporation can reshape the tumor microenvironment, transforming both treated and untreated lesions into CD8+ T-cell inflamed tumors. Earlier studies showed that the combination of pembrolizumab and intratumoral plasmid IL-12 held promise in stages III/IV melanoma patients.
Algazi presented data from recently completed phase II monotherapy and combination therapy studies at the 2017 World Congress of Melanoma (Abstract P524). The OMS-I100 monotherapy study included 51 patients and the OMS-I102 combination study included 22 patients, all with metastatic melanoma. The combination-study patients were selected based on their baseline biomarker data, which predicted that patients would not respond to anti-PD-1 therapy. Patients must have had histological or cytological diagnosis of melanoma with locally advanced or metastatic disease that was not amenable to definitive local therapy and at least one measurable tumor accessible for intratumoral injection.
Monotherapy patients were treated with IL-12 injections alone on days 1, 5, and 8 of every odd cycle at 6-week intervals. In the combination study, patients received IL-12 injections plus 200 mg intravenous infusions of pembrolizumab every 3 weeks.
At 24 weeks, the best overall response rate in the combination arm was 50% (11 of 22 patients). This included 41% of patients with a complete response and 9% of patients with a partial response. Another 9% of patients achieved stable disease, for a total disease control rate of 59%. In the monotherapy arm, about one-third of patients achieved a best overall response rate.
The progression-free survival rate at 15 months was 57%; among responders, the duration of response was 100%.
Fewer than 10% of patients in both studies reported treatment-related serious adverse events (9.8% in the monotherapy arm and 8.7% in the combination arm).
"We have demonstrated that intratumoral therapy with pembrolizumab can help many patients who would not be helped by PD-1 antibodies alone, including patients with prior exposure to ipilimumab, PD-1 antibodies, or both," said Algazi.
In addition, biomarker data highlights the positive impact of the combination therapy on immunological mechanisms. "Responding patients had a significant treatment-related increase in the density of intratumoral CD8+ T cells coupled with a significant increase of intratumoral Th1-related gene expression."
Intratumoral T-cell receptor (TCR) clonality significantly increased after a single cycle of treatment. "Responders who had low peripheral TCR clonality had significantly higher intratumoral TCR clonality after treatment, but in non-responders, this relationship was inverted with no significance noted," he said.
Durable clinical responses continue to demonstrate that this combination therapy is a promising treatment modality. "The updated correlative immune-focused biomarker data further highlights that this IL-12-based therapy can drive intratumoral Th1 polarization and an increase in CD8+ tumor infiltrating-lymphocytes."
The increased treatment-related intratumoral TCR clonality and proliferating exhausted T cells in the periphery of responding patients extends the concept that intratumoral injections plus pembrolizumab can reshape the tumor microenvironment, Algazi noted. "Collectively, these data suggest that combination intratumor therapy plus pembrolizumab directs the tumor microenvironment towards a Th1 immune phenotype with reduced immune suppression. The robust intramural and systemic anti-tumor response supports improved clinical outcomes in patients predicted not to respond to anti-PD1 therapy."
The next step, he said, is a multisite phase II clinical trial focusing exclusively on patients with documented resistance to PD-1 antibodies. "A definitive randomized phase III trial is also in the planning stages. In parallel, we are developing an additional study for patients with advanced squamous cell carcinoma of the head and neck, with more details regarding this to be available soon. Our findings need to be confirmed, but this new treatment approach could help metastatic cancer patients to achieve long-term remission with an acceptable side-effect profile/quote]
Oncosec's PISCES trial is enrolling patients who have failed prior checkpoint inhibitor therapy, ie anti-PD-1, in stage III/IV melanoma. Patients who were enrolled in the P2b trial - combination pembrolizumab and Immunopulse IL-12 - weren't treatment naive, and the majority of them failed prior checkpoint inhibitor therapy. And these patients were selected because they had such low percentages of a partially exhausted CTL phenotype - they could have selected predicted nonresponders or actual nonresponders with higher percentages of the exhausted phenotype, a marker of CD8 T cell activation, but nope, they didn't go that route. These were patients with very few CTL. These types of patients have very limited options once they fail anti-PD-1 therapies.
So what is stopping them from achieving better results in the P2b trial? Well, we can start with describing the patient population - again, not treatment naive patients and not just any predicted nonresponders, but patients who have essentially little to no activated T cells at baseline! Also, based on spatial analyses of the tumor microenvironment, the P2b nonresponders to the combination had much higher quantities of intratumoral FOXP3 Tregs in close proximity to CD8 T cells; this suggests to me that those Tregs, which might also be expressing CTLA-4, are preventing T cell activation by outcompeting CD28 for ligands on dendritic cells.
So, adding an additional encoded protein to a plasmid that just abates Treg-induced immune suppression, would theoretically lead to improvements in T cell activation. And improvements in T cell activation then drives the percentage of partially exhausted CTL phenotype, which are - in addition to PD-(L)1 on tumor cells - a prerequisite for anti-PD-1 responses.
ITT, getting back to your question about Halo's PEGPH20...
I think hyaluronan's role in the progression of pancreatic, breast, colorectal cancer, etc is misunderstood by Halozyme and clinical researchers. I don't think it is creating the physical barrier at all to immunotherapies or chemotherapies. In fact, it probably facilitates cell motility. Hyaluronan likely facilitates motility through binding with CD44 and RHAMM, both of which are overexpressed on pancreatic, breast and other cancers' cells. Having more hyaluronan available intratumorally means that when it binds with certain variants of CD44 and RHAMM on cancer cells, it will help to spread the cancer cells. This is where PEGPH20 comes in - the enzyme breaks down hyaluronan, thus mitigating the risk of metastases. I don't think PEGPH20 is driving immunity or turning cold tumors "hot" at all, it is simply reducing the spread of cancer cells. That isn't a bad problem to have necessarily, but there is going to be collateral damage with the PEGPH20 enzyme and it won't really improve antigen presentation, T cell activation, nor CTL proliferation.
What collateral damage? Hyaluronan also binds with CD44 on activated T cells, thus improving motility of those cells into the tumor microenvironment. Improving motility contributes to an increase in the number of tumor infiltrating lymphocytes and allows for systemic immune responses and an abscopal effect. By introducing PEGPH20 systemically, they are likely abrogating immune responses and contributing little to nothing when an anti-PD-1 agent is used. For advanced diseases, the addition of PEGPH20 won't add much value to combination therapies in my opinion.
And here is another big problem: systemic administration of the enzyme affects hyaluronan all over the body. The presence of it in skeletal muscle, bones, connective tissue, etc suggests to me that PEGPH20 will lead to very significant SAEs. You can already see this in the data when it is combined with chemo drugs. These patients experience severe pain, peripheral edema, along with several other serious side effects. The peripheral edema tells me that the fluid that accumulates with hyaluronan is being released when the enzyme breaks up the hyaluronan. The hyaluronan is found in these places for a good reason. It seems insane to me to have patients subjected to systemic administration of the enzyme.
ONCS's first PIIM construct encodes proteins that will be expressed intratumorally and in the presence of all available neoantigens. This will very likely lead to strong immune responses, an abscopal effect, and memory T cell production. It increases the number of CTLs, and due to their activation they start expressing a partially exhausted phenotype. This partially exhausted state is then ameliorated with an anti-PD-1. But without those CTLs, there is no point in using an anti-PD-1 agent. You absolutely need a significant number of tumor-antigen specific CTLs to drive robust and lasting responses, i.e. tumor cell destruction. PEGPH20 probably won't accomplish any of that, but it will perhaps slow down the formation of metastases in some early stages of cancer.
I would think first line therapy is very much in the cards for stage III/IV melanoma. It would be most prudent for them to identity an optimal biomarker threshold in the first-line setting as you're alluding to, so that patients who might respond with anti-PD-1 monotherapy don't receive more treatment regimens than they need.
I certainly think EP pIL-12 with pembro would work in other solid cancers, but they would have to use the new GENESIS EP device if they're directly treating subcutaneous and visceral lesions. The first new PIIM construct is very likely going to be much better than Immunopulse IL-12 due to the fact that the genes are p2a linked and the PIIM includes encoded genes that will vastly improve antigen presentation and T cell activation while mitigating Treg effects intratumorally.
I will read up on HALO at some point and provide some feedback.
Yes, in-situ cancer vaccination (vis-a-vis DNA-encoded plasmids delivered intratumorally) can probably be characterized as a paradigm shift.
Not sure, many BPs would be incentivized to partner for various reasons.
No, I'm just familiar with cellular biology and data analysis. I'm not a medical professional.
I think Oncosec is approaching immuno-oncology in the most rational and elegant way from a biological perspective. The only way to generate robust and lasting immunity against cancer is to get T cells to recognize the cancer cells, proliferate, and build memory. This can be done through adoptive cell transfer, vaccination (through predicted recombinant protein or non-intratumoral DNA/mRNA-encoded means), or in-situ vaccination. The most streamlined and perhaps the most effective, safest, and cheapest approach would be to take advantage of a primary tumor's full repertoire of neoantigens in situ. An intratumoral in situ platform using DNA-encoded plasmids eliminates the need for algorithms (i.e. no guess work to determine which neoantigens might drive immune responses - the host immune system will make those determinations - algorithms are required for recombinant protein and non-intratumoral DNA/mRNA-encoded vaccinations); removes the need for engineering and expanding T cells in the lab (engineering and T cell expansion are required in the lab for adoptive cell transfer); allows for a protracted rather than transient (as in the case of mRNA and recombinant proteins) release of proteins, thus reducing the number of repeat administrations and dosing level requirements; requires potentially only one systemically administered drug - an anti-PD-1 agent; and dramatically reduces the risk of systemic toxicities.
And when you add all of these advantages up - no tumor biopsies, no predictive software, no engineering, no T cell manufacturing/expansion in the lab, dramatically reduced SAEs, lasting immunity, etc. - you end up with enormous cost savings.
The first PIIM construct is a disruption waiting to happen.
Yes, I really do think they will be partnering on the first PIIM construct. The only way to drive true personalized cancer immunity is to take full advantage of a primary tumor's complete set of neoantigens. Proliferation of the activated CTLs then allows for an abscopal effect, thus eliminating distant lesions as well. And to accomplish such a feat by allowing your own cells to manufacture the DNA encoded proteins...
I wouldn't be too surprised to see some partnerships beginning to develop with their first PIIM construct. If they are able to achieve what they have already in stage III/IV melanoma with just Immunopulse IL-12 and pembro, you can imagine what they are capable of doing when they add in two additional encoded products in a single plasmid. Moreover, the IL-12 expression profile using the P2A structure is an improvement on IRES. This first PIIM construct should improve neoantigen presentation intratumorally while simultaneously negating the immune suppression caused by Tregs. Improvements in presentation will lead to improvements in T cell activation, thus contributing to higher numbers of intratumoral CTLs. Not only will the treated tumors shrink, but so too will distant lesions.
I can see the company going the route that we have seen other companies take: partner for multiple solid tumor types and enroll patients concurrently. IND-enabling work for the first PIIM construct should be wrapping up and I anticipate clinical rollout soon after discussions with FDA. I think the rollout could be tied to an upfront payment to move things expeditiously through a phase I trial for multiple solid tumors. This is, of course, in my opinion.
I simply cannot imagine that a company developing an anti-PD-1 drug will just wait on the competition for potentially multiple solid tumor types. There is a LOT to lose if the anti-PD-1 combination with the first PIIM construct ends up achieving what I think it will. And if Oncosec demonstrates it can drive robust tumor-antigen specific immunity in patients across multiple solid tumor types, it will setup an adapt or die scenario for potentially hundreds of companies working on treatments for solid tumors. We aren't just talking about one single cancer indication, but an in situ platform that could be used for many solid tumors. If a company developing an anti-PD-1 allows their competition to partner on multiple solid tumors, guess who has the enormous advantage if the combo works in multiple solid tumors- right, it would be the first mover. It is a technological platform, not a single drug being tested in a single indication.
Jeff, I believe that was in reference to other trials performed by UCSF. That is exactly what can happen when you systemically deliver anti-CTLA-4 and anti-PD-1 drugs together or even with anti-CTLA-4 alone - they can lead to autoimmune reactions. UCSF has published several articles discussing these autoimmune SAEs.
Organs can also be targeted by immune cells if you engineer T cells to attack antigens that aren't in fact neoantigens. Your body has a built in mechanism that distinguishes between what is a "foreign" antigen, e.g. neoantigens, and what is a "self" antigen. If you circumvent that built-in regulatory mechanism by introducing loads of CAR-t cells that target "self" or shared antigens, for example, you'll end up with t cells that indiscriminately start attacking healthy tissues.
That gold standard is one that will take advantage of tumors' full repertoire of neoantigens - the only way to do it is to create the intratumoral conditions that improve neoantigen presentation and T cell proliferation.
Billions of dollars are being spent creating the technology that best predicts the antigens that might lead to responses. On top of that, billions more are being spent by companies and institutions that are manufacturing recombinant protein vaccines. Then there is Moderna Therapeutics that is taking a shotgun mRNA approach that also requires antigen predictions and customized vaccine combinations. While Moderna's approach allows patients' cells to manufacture the encoded antigens - a step way above recombinant protein vaccines - tumors still need to be sampled, algorithms need to be used to identify neoantigen targets, the mRNA therapy has to be highly customized, and after several weeks a patient may receive the vaccine. And then the encoded antigens have to lead to adequate numbers of CTL.
Oncosec has added an encoded antigen to their first multigene construct, but the point of encoding it goes beyond vaccine personalization. The multigene combination should in theory increase intratumoral antigen presenting cells and increase the number of local, mature dendritic cells. This not only kick starts an immune response, but it's a way to overcome T reg suppression intratumorally. When you have more dendritic cells available intratumorally that can't all be suppressed by T regs, you end up with improved presentation IN THE PRESENCE OF ALL AVAILABLE TUMOR NEOANTIGENS.
Oncosec's intratumoral in situ vaccination platform allows for an off-the-shelf product that doesn't require algorithms or highly customized combinations that might not target the right tumor antigens. Moreover, it allows patients' cells to manufacture the proteins. This platform is already observing significant results with just the encoded IL-12 with Pembro. The first PIIM construct is going to be completely disruptive in my opinion. You can probably say goodbye to those tumor biopsies, the algorithms, the manufacturing and high costs of production, the time required to generate the vaccines, the risks of inadequate immune responses, and the toxicities.
First, some general statements regarding these other treatments:
1. Why go through the expensive and timely process of developing recombinant proteins when you can have a patient's cells manufacture those proteins.
2. Recombinant IL-12 proteins can also lead to major SAEs if not administered locally or if they end up going off target.
3. Combining IL-12 gene therapy with chemotherapy is just plain silly. Chemotherapies destroy the very machinery that IL-12 creates. Chemotherapies can help to release tumor antigens, but they become counterproductive when they start destroying immune cells.
4. Electroporated plasmids encoding IL-12 and an antigen administered intramuscularly lead to a vaccination that targets a single antigen. While the antigen-specific immune response might be sometimes effective when used in combination with an anti-PD-1, it doesn't take advantage of tumors' full antigenic repertoire. It is a guessing game in terms of which antigen(s) will work for individual patients.
5. Combination IL-12 therapies that don't include an anti-PD-1 will more likely than not have mediocre PFS, DOR and OS rates compared to those that do incorporate an anti-PD-1.
There are several things Oncosec is achieving currently and/or in its FIRST multigene construct that none of these companies are addressing, including: mitigating T reg suppression, increasing the number of APCs like dendritic cells in the presence of ALL potential neoantigens, maximizing IL-12 expression through a P2A-linked construct, driving interferon gamma production intratumorally, and incorporating a proprietary device that can work with tissue heterogeneity. And all these things can be accomplished using a single electroporated multigene plasmid construct.
The FIRST multigene construct data not only show successful expression of multiple encoded genes in a single product, but the selected genes highlight a rational approach to increasing intratumoral CTLs. The flt3l-encoded gene fused with an antigen is going to accomplish more than just drive a single tumor antigen-specific immune response.
Flt3l fused with an antigen will lead to endogenous expansion of dendritic cells and maturation of those cells (through activation with the fused antigen). The maturation of dendritic cells through activation with an antigen increases the costimulatory molecules CD80 and CD86. So, not only does the fusion of Flt3l and an antigen lead to dendritic cell expansion and activation, but it vastly increases CD80 and CD86 intratumorally.
And what do we know about CTLA-4 on T regs? They outcompete CD28 binding on T cells for CD80 and CD86 on mature dendritic cells. By significantly increasing the number of CD80 and CD86 costimulatory molecules on dendritic cells, you're shifting the balance of power in favor of T cell activation. You end up with more mature dendritic cells per T reg, thus altering ratios that would otherwise suppress T cell activation and proliferation.
Bottom line: the combination will lead to more intratumoral CTLs. It will accomplish this by improving neoantigen presentation, increasing mature antigen presenting cells intratumorally, and mitigating the overall effect T regs have on immune suppression.
I would have to research where that 30% threshold is originating, but you can also visibly see it in one of the figures in the JCI publication - all patients in the discovery and validation cohorts who had more than 30% of the partially exhausted phenotype in CTL responded to anti-PD-1 monotherapy.
Yeah, when you combine a systemic anti-CTLA-4 agent with anti-PD-1, the thresholds can be lowered. While the combination leads to many SAEs, it is still more effective than either agent used alone. This begs the question: how low can Immunopulse IL-12 plus pembrolizumab go in terms of the percentage threshold for responses? It looks like they were getting responses out of patients who had no detectable exhausted TIL at baseline - that is quite amazing. This suggests to me that the addition of Immunopulse IL-12 is sufficiently elevating the percentages of the partially exhausted CTL in approximately half the patients who are predicted nonresponders to anti-PD-1; this elevation ultimately sets the stage for anti-PD-1 treatment.
Also, I am absolutely convinced, beyond any reasonable doubt, that addressing the CD4-positive T regs expressing FOXP3, will lead to complete responders across the board. When I saw the data from the SITC presentation, I immediately thought the T regs in close proximity are inhibiting T cell activation in nonresponders, and I think they are doing it by overcompeting with CD28 on T cells. I think this issue could be addressed by adding either an encoded anti-CTLA-4 agent to the multigene construct or adding an antibody that blocks ICOS-ICOSl interaction.
To answer your other question, I think the fact that interferon gamma can promote dendritic cell maturity, suggests to me that antigen presentation is enhanced - there are going to be more CD80 and CD86 expressions on DCs, thus activating more CD8 T cells. So, bottom line, I think you're exactly right. And absolutely, at the end of the day it's all about the CTLs.
Titan, there are a few things I want to quickly point out in regards to Oncosec's first multigene construct:
1. They are essentially creating a plug-and-play platform that would allow for the creation of personalized cancer vaccines. They have demonstrated in this first multigene construct that they can successfully express three encoded genes on one plasmid. One of these genes can be an encoded antigen that gets expressed with IL-12 and an flt3 ligand. The encoded antigen is part of a fusion protein that, together with flt3 ligand, facilitate the maturation of dendritic cells. These mature dendritic cells then present the antigen to T cells for activation.
2. Flt3 ligands lead to the proliferation of dendritic cells - this is a major deal. In theory, this would improve antigen presentation, therefore improving immune responses. The presented data show that CD8-positive T cells targeting the encoded antigens do in fact proliferate significantly in vivo.
3. They are using the new GENESIS EP device.
4. The genes are all P2A-linked.
More to come later...
Here are the data for ONE multigene construct...
http://oncosec.com/wp-content/uploads/2017/11/SITC_preclinical_PIIM_poster.pdf