NorthWest Biotherapeutics Inc (NWBO)

[eagle8]

Andrew Caravello, DO
@andrewcaravello
🧬 The Cancer Vaccine Era Has Arrived: How Jaffee’s Map Converges with $NWBO #Murcidencel #DCVax, and Why $MRK #KRAS Stack and #KEYTRUDA Need Dendritic Cell Instruction in #PanCAN

🧠 TLDR

Dr. Elizabeth Jaffee’s title is a structural assertion, not a marketing one. The cancer vaccine era arrives when vaccines are treated as immune teaching tools, combined with the right immune helpers, and judged by long term control and by what changes inside the tumor, not just by whether a scan shows shrinkage.

Jaffee’s core point is that PD 1 therapy mainly amplifies what already exists, and that in many solid tumor settings only a minority of patients, often framed on the order of 20 percent, present with enough naturally primed tumor reactive T cells for PD 1 alone to convert into durable control. That is why vaccines exist, to create those T cells in the first place.

Two PDAC settings make this easy to see.

First, the best case setting: after surgery, when tumor burden is low. In that setting, an ex vivo dendritic cell proof of concept study set a meaningful durability goal, at least 60 percent recurrence free at 2 years, and reported 64 percent recurrence free at 2 years and 83 percent alive at 2 years. The study showed vaccine specific immune activation and evidence consistent with vaccine linked T cells reaching a lung metastasis that was later removed. The study used autologous dendritic cells pulsed with an allogeneic mesothelioma tumor cell lysate containing antigens also expressed in PDAC.

This validates the category of ex vivo dendritic cell instruction in low burden PDAC. The remaining commercial question is whether a patient specific lysate platform improves the instruction payload, consistency, and breadth.

Second, the hardest setting: late metastatic disease after chemotherapy, where suppression is maximal. In that setting, a randomized phase II study showed the immune teaching step can occur, with tumor infiltrating T cell clones that recognized mesothelin and mutant KRAS, yet the clinical response rate was only 4 percent. The authors blamed the barrier Jaffee emphasizes: suppressive myeloid cells dominating the tumor environment.

Together, these two settings illustrate Jaffee’s point in numbers.

Teaching the immune system is necessary.

Long term control depends on whether the tumor environment is remodeled enough for those taught cells to function and persist.

🧠 Introduction

This is not a write up of one clinical trial. This is a systems level reading of Dr. Elizabeth Jaffee’s Grand Rounds message about why cancer vaccines are becoming viable now, especially in pancreatic cancer.

Her core claim is structural. Many immune therapies fail in immune cold tumors because the immune system was never properly taught what to attack. Even when it is taught, the tumor environment often prevents those immune cells from organizing, functioning, and persisting.

The talk is not framed as a catalog of products. It is framed as a troubleshooting log built over decades of iteration. Pancreatic cancer is treated as a systems problem, and the studies are treated as instruments to measure which system layer is missing, rather than as one shot bets.

🎙️ What Jaffee’s Title Really Means

When Jaffee says The Cancer Vaccine Era Has Arrived, she is not saying every vaccine works now. She is saying the field finally understands what was missing.

For decades, researchers tried to activate immunity, but the immune system needs two things:

Specific training so T cells know what to attack.

A workable battlefield inside the tumor, so those T cells can enter, survive, and keep working.

Her talk argues that vaccines matter because they provide training, and the new era exists because the field now pairs that training with the right immune helpers and measures the real effects inside tumors.

🏜️ Why Pancreatic Cancer Is Different

Pancreatic cancer is difficult because it is not only a tumor. It is a whole ecosystem built to block immunity.

Jaffee describes a tumor environment packed with suppressive immune cells and dense stroma. The tumor might contain T cells, but they are often the wrong type or too exhausted to kill. A useful baseline anchor is that only 15 percent to 20 percent of patients present with resectable disease, and even after surgery and modern chemotherapy, recurrence is common. That is why a 64 percent 2 year recurrence free signal is treated as a foothold rather than a cure.

This is why immune checkpoint drugs, like PD 1 blockers, generally do not work well in pancreatic cancer. Those drugs can take the brakes off, but if the immune system was never properly trained or cannot enter the tumor, taking brakes off does not create a functioning response.

This is also why a checkpoint leader like $MRK cannot rely on #KEYTRUDA alone in #PanCAN. PD 1 drugs amplify existing tumor specific T cells. If the teaching layer never produced enough high quality T cells, or the tumor environment prevents them from working, brake release cannot create durable control.

🧪 The Key Method: Window Studies

A key idea in Jaffee’s work is what researchers call window studies. Patients are given a treatment for a short, planned time, often around two weeks, before surgery. Then the removed tumor is examined in detail.

This matters because scans do not tell the whole story in immune therapy. Immune therapy can be working without immediate shrinkage, and it can also look promising on scans while failing to build long term control. Looking directly at the removed tumor lets researchers answer the real questions:

Did immune cells enter the tumor?
Did they become real killing cells or just show up?
Did the tumor start building organized immune structures?
Did suppressive myeloid cells stay dominant or weaken?

This is how Jaffee chooses the next combination to test. It is not guessing. It is troubleshooting a system.

🧱 What GVAX Taught Her, and Why It Is Not #DCVax

#GVAX and #DCVax are both called vaccines, but they are fundamentally different.

GVAX is a whole tumor cell vaccine engineered to secrete GM-CSF. Its purpose is to recruit the body’s own dendritic cells and encourage them to pick up antigen and start teaching T cells. GVAX is primarily a recruitment and priming stimulus.

$NWBO #Murcidencel #DCVax is a controlled ex vivo dendritic cell instruction approach. It aims to take the patient’s monocytes outside the tumor environment, differentiate and mature them under defined conditions, load them with tumor antigen reality, then return instruction grade antigen presenting cells back into the patient. That is deliberate teaching rather than relying on endogenous dendritic cells in a PDAC microenvironment to perform consistently.

This distinction explains why GVAX still matters in Jaffee’s talk. GVAX proved that even in pancreatic cancer, a vaccine driven stimulus can push the tumor from immune absent toward immune organized, inducing tertiary lymphoid aggregates, TLS. It proved the battlefield can change. Ex vivo dendritic cell platforms exist to make that change more controllable and repeatable.

🛠️ The Modality Gap: Biology vs Logistics

Jaffee’s current trials also include peptide and neoantigen approaches, and she describes a collaboration using a second generation CD137 pathway agent. That does not refute the dendritic cell thesis. It reflects the real world constraints the field lives under.

Top tier labs use peptides and neoantigens because they are logistically fast, scalable, sponsor friendly, and compatible with multicenter trial operations.

The biological tradeoff is that in immune cold tumors like PDAC, off the shelf antigen delivery often generates signals of infiltration without reliably generating durable killer licensing and persistence in the face of myeloid suppression. That gap is exactly why the field keeps reaching for heavier downstream levers like CD137 agonism, sequencing logic, and microenvironment remodeling.

So the presence of those approaches is not a contradiction. It is a tell. The field is trying to solve the same missing layer from multiple angles, under real world constraints.

🧩 TLS Are Not All Equal

Jaffee warns that tertiary lymphoid aggregates, TLS, are not a simple yes or no marker. They can exist in different forms, including immature, resolving, or true germinal center structures. In her work, different patterns correlate with different outcomes.

TLS are basically mini lymph nodes the body builds inside or next to the tumor. A mature TLS is a local immune command center with three jobs:

Keep presenting tumor antigens.

Pull in new immune cells with chemokine traffic signals.

Help killer T cells stay active long enough to matter.

She goes further than saying TLS appear. She describes heterogeneity inside those structures and highlights a machine learning derived germinal center signature, Pattern 9, a multi gene immune and inflammatory program associated with longer disease free survival. In plain terms, the tumor is not just trying to stop immune cells from entering. It is trying to stop the immune system from organizing.

🔓 Why Checkpoint Drugs Alone Are Not Enough

The metastatic CRS 207 and GVAX study is a real world example of Jaffee’s argument.

The study detected tumor infiltrating T cell clones that recognized mesothelin and mutant KRAS. Yet the clinical response rate was only 4 percent.

The paper names the barrier. Myeloid cells were enriched, and high myeloid and Treg signatures tracked with poor response. This is consistent with the broader literature on PDAC myeloid dominance and dendritic cell impairment. The lesson is not that vaccines cannot teach. The lesson is that teaching is not enough if the tumor environment remains dominated by suppressive cells that prevent killing.

⚡ Why Jaffee Keeps Talking About CD137

Jaffee’s vaccine only arm showed immune cells could increase, but real killing programs did not always appear. She then noticed CD137 rising when PD 1 was added, which signaled that T cells were hungry for additional activation.

CD137 is an extra accelerator. It can push T cells toward stronger killing and persistence when the system is already primed.

In plain terms, pancreatic cancer often needs three layers:

Teach the immune system what the tumor is.

Remove enough suppression so immune cells can function.

Provide a strong activation signal so they become durable killers, not temporary visitors.

🧬 KRAS Is the Immune Lock

KRAS in pancreatic cancer is not only a growth switch. The literature increasingly treats it as an immune lock.

KRAS signaling supports recruitment of suppressive myeloid cells, reshapes fibroblast behavior, and undermines antigen presentation. That is why KRAS inhibition matters. When KRAS inhibition works, immune permissiveness can increase, more CD8 activity becomes possible, and myeloid pressure can decrease.

But resistance pathways emerge and immune checkpoints rise. The window closes.

So KRAS inhibition should be treated as a gate opener. The real question becomes what fills the opened window before the tumor adapts and shuts it again. Here, KRAS stack means the emerging combination design space around KRAS pathway inhibition and immunotherapy in PDAC.

🧬 Why IL 12 Keeps Coming Up

Jaffee mentions IL 12 for the same reason pancreatic cancer humiliates checkpoint therapy.

Pancreatic tumors often suppress and distort the instruction system. Endogenous dendritic cells inside PDAC are frequently pressured into impaired maturation, apoptosis, or tolerogenic programs. While some endogenous cDC1 can still function, which is why strategies like Flt3L plus CD40 agonism show promise, the point is that without rescue the endogenous instruction layer is too inconsistent to bet a whole therapeutic strategy on.

IL 12 is one of the clearest signs that the teaching apparatus has shifted into Type 1 mode. It is the signature of instruction strong enough to produce durable killer programming rather than weak priming or tolerance.

This is why dendritic cell platforms exist as a concept. A platform is not just a way to inject cells. It is a way to control the instruction state outside the tumor’s suppressive logic, then return instruction grade antigen presenting cells back into the patient.

More dendritic cells is not necessarily better. Simply differentiating monocytes into dendritic cells creates quantity, not quality. The maturation recipe determines whether those cells become instruction grade Type 1 teachers or become phenotypically mature but functionally IL 12 deficient. This is the difference between firepower and instruction.

🧾 Why Durability Shows Up Most Cleanly After Surgery

After surgery, the visible tumor is removed. The immune system is not trying to storm a fortress. It is trying to prevent microscopic residual disease from reigniting.

In that setting, the resected PDAC study aimed for at least 60 percent recurrence free at 2 years and reported 64 percent, with 83 percent alive at 2 years.

The authors acknowledge selection and timing limitations. That is appropriate. It keeps the result in the right category: an early foothold signal that matches the architecture thesis.

🧠 Conclusion: The Design Regime

Jaffee’s title becomes persuasive because the field now has boundary conditions and footholds.

In metastatic PDAC, immune teaching can occur and still fail clinically when suppressive myeloid biology dominates. In resected PDAC, ex vivo dendritic cell instruction can produce a durability foothold when tumor burden is low and immune memory has time to consolidate.

That is the turning point she is naming. The problem is no longer can the immune system be activated. The problem is can the immune system be taught, then allowed to organize, then kept functional long enough to become memory.

Pancreatic cancer does not mainly lose because the immune system lacks weapons. It often loses because the immune system’s instruction layer is suppressed and unreliable inside the tumor. In that context, more firepower does not solve the problem. It amplifies a system that may be miswired.

That is why a controlled dendritic cell platform is a direct implementation of the architecture she is describing. It is an architectural match, not a claim of endorsement. If endogenous instruction is inconsistent, a controllable workaround is to manufacture instruction ex vivo under defined conditions, load it with relevant tumor antigen, then return instruction grade antigen presenting cells back into the patient.

The convergence:

Checkpoint blockade is brake release.

KRAS pathway inhibition is gate opening.
Innate agonists are ignition.

Dendritic cell instruction is the missing layer.

So the cancer vaccine era is not a promise. It is a design regime.

Durability is constructed, not wished for.

https://x.com/andrewcaravello/status/2014716867244404774