NorthWest Biotherapeutics Inc (NWBO)

[Rkmatters]

Two things. Study design and immunotherapy mechanism of action. Linda Liau already told us that Trial design removed some of their best responders by first inclusion criteria. They also had a crossover arm, which means we will get to see whether this trial has “late effect” and can murky the statistical outcome. She also told us with immunotherapy it takes time to mount a substantial immune response. The effects of immunotherapy can be seen in Overall Survival. For GBM overall survival is the standard. They have waited this long to unblind, it makes sense to add OS as a primary endpoint, and adjust the SAP imho.


Off my Jan 27, 2017. Hope this helps to explain immune therapy mechanisms of action

He can cough all he wants and rewrite PFS history, using his eyesight to but he can't change the prior trial's OS. He can plant suggestions that investigator made mistakes in recording of prior events and study median but without proof it's just accusations and conjecture. He can talk all he wants about futility, but his low PFS theory is based on PsPD passing size perimeters, that means the therapy is working well. If it's working well they should see long-tail. And if the therapy were not working well we wouldn't be waiting for a primary endpoint update as Linda Powers told annual meeting shareholder attendees, scans need to read and tumors carefully measured. Well death requires no MRI reading.

But let's chat about changes in tumor size on an Immunotherapy and the clinical signals of ilpilumumab:
-- it demonstrated a relatively low conventional (shrinking of tumors) response rates: (10-15% in melanoma)
But then as they Investigators were going through their observations, efficacy was distinctively different:
-- they noticed survival improvement more than their 10-15% response rate initially was suggesting as.
That suggested that there must another benefit beyond the conventional response (same observation as other immunotherapies).

Case studies showed mixed examples of efficacy benefit. Axel Hoos described (verbatim):

1)Included conventional response patterns: after initial treatment: tumors shrink
2)included case patterns where after initial treatment: Tumors grow, and then subsequently shrink; predominantly because Lymphocytic infiltration into the tumor; makes the lesion larger, for the T cells can then acts against the tumor cells that leads to shrinkage
3) included case with patterns where the immune system is in some form of equilibrium state with the tumor; and leads to prolonged stabilization of disease. That may over time lead to a conventional response but it may take time to a year or more for that to happen.
4)included cases with patterns where actually new lesions appear while other lesions disappear or shrink. Those patterns are particularly interesting because suggest they that micro metastatic tumor cell deposits in different tissues that once detected by the immune system may actually lead to a visible lesion on a CAT scans, that previously did not exist, because T cells accumulate around the micro metastatic tumor cell and make that a visible lesion. Those lesion are usually small and they are transient. They go away with time, once the T cells do their job.

Conventional Responses is:
-- Tumor shrinkage

Immunotherapy benefit go above and beyond conventional responses of "shrinking" mass that might be there. To look only at conventional response rate is faulty as it doesn't give a true picture of response benefit.

"5. Delayed vaccine effect
As a consequence of their immunological mechanisms of action, cancer vaccines may require considerable time after administration to induce immunity. Therefore, tumors in subjects treated with cancer vaccines may show early progression followed by subsequent response. This potential phenomenon should be considered in the design of later phase clinical trials, particularly if nonclinical data or early phase clinical trials suggest that the phenomenon exists and time-to- event endpoints are used. Due to delayed effect of the vaccine, the endpoint curves may show no effect for the initial portion of the study. If the vaccine is effective, evidence of the effect may occur later in the study. This delay in the effect may lead to an average effect that is smaller than expected and thus may require both an increase in sample size to compensate for the delay and a careful assessment of trial maturity for the primary analysis. In addition, possible violation of the proportional hazards assumption should be considered when selecting a statistical method for the primary analysis. " -- FDA on vaccines

But trial was not enrolling the patients with established tumors. Not to mention this trial more than doubled the 110 events it initially was set to review in PFS, so they appear to cover the "sample size" to compensate for any potential delay effect. And so expected to see more cases of #3 and some cases of #4, when dormant cells wake up elsewhere.

3) included case with patterns where the immune system is in some form of equilibrium state with the tumor; and leads to prolonged stabilization of disease. That may over time lead to a conventional response but it may take time to a year or more for that to happen.
4)included cases with patterns where actually new lesions appear while other lesions disappear or shrink. Those patterns are particularly interesting because suggest they that micro metastatic tumor cell deposits in different tissues that once detected by the immune system may actually lead to a visible lesion on a CAT scans, that previously did not exist, because T cells accumulate around the micro metastatic tumor cell and make that a visible lesion. Those lesion are usually small and they are transient. They go away with time, once the T cells do their job.

Remember with DCVax-L Phase III there shouldn't be established tumors, as it's an intent for gross total resection patients that they remove. This trial was designed to remove ALL EVIDENCE OF DISEASE PROGRESSION at baseline. And this study design was to find patients who qualify for a GTR. It doesn't mean they will always get it, but with state of the article surgical equipment, the rates of GTR should be high. Tumor debunking is known to decrease immunosuppressive cytokines such as TGFB. As such, this study will more than likely experience a higher disease stabilization as it WILL not be enrolling the type of patients that the ilpilumumab trial enrolled --melanoma patients with established tumor mass. And so there shouldn't be a need to "shrink" tumors and cause immune "swelling" first effect like the DCVax-Direct trial encounter leading to possible "false" progression. The PsPD in thar made it passed main trial screening should not have tumors. They have Blood Brain Barrier (BBB) enhancements, of a therapy working well, which isn't the same thing as a TUMOR that is growing. And PsPD after enrollment can be determined by other imaging modalities, which this trial uses. It is a "enhancement" condition that clears itself in time. It is not disease progression.

Here is an example of what we should expect to see as far as MRI imaging on disease progression. This is a recurrent DC enrolled study. The patients that had lower disease tumor volume had significantly longer PFS and OS. The recurrent trial OS median was at 8 months (range: 5–107 mo). IF a KM Curve for all patients were created, there would be a long-tail response for a nice percentage of the GBM patients, as the median OS doesn’t tell the story. Had KM curve of all survival those who lived longer would represent the 35% of long tail survival curve (OS times of the 15 patients (7 of which had signs of disease progression at first vaccination).


The natural killer cell response and tumor debulking ARE associated with prolonged survival in recurrent glioblastoma patients receiving dendritic cells loaded?with autologous tumor lysates

Recurrent glioblastomas (GBs) are highly aggressive tumors associated with a 6–8 mo survival rate. In this study, we evaluated the possible benefits of an immunotherapeutic strategy based on mature dendritic cells (Dcs) loaded with autologous tumor-cell lysates in 15 patients affected by recurrent GB. The median progression-free survival (PFS) of this patient cohort was 4.4 mo, and the median overall survival (OS) was 8.0 mo. patients with small tumors at the time of the first vaccination (< 20 cm3; n = 8) had significantly longer PFS and OS than the other patients (6.0 vs. 3.0 mo, p = 0.01; and 16.5 vs. 7.0 mo, p = 0.003, respectively). cD8+ T cells, cD56+ natural killer (NK) cells and other immune parameters, such as the levels of transforming growth factor ß, vascular endothelial growth factor, interleukin-12 and interferon ? (IFN?), were measured in the peripheral blood and serum of patients before and after immunization, which enabled us to obtain a vaccination/baseline ratio (V/B ratio). an increased V/B ratio for NK cells, but not cD8+ T cells, was significantly associated with prolonged PFS and OS. Patients exhibiting NK-cell responses were characterized by high levels of circulating IFN? and e4Bp4, an NK-cell transcription factor. Furthermore, the NK cell V/B ratio was inversely correlated with the TGFß2 and VeGF V/B ratios. These results suggest that tumor-loaded DCs may increase the survival rate of patients with recurrent GB after effective tumor debulking, and emphasize the role of the NK-cell response in this therapeutic setting.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3661164/pdf/onci-2-e23401.pdf

Several lines of evidence indicate that the immune system is capable of interacting with cancer cells to prevent their growth as well as to destroy established tumors.2 However, attempts at utilizing the immune system to treat established tumors are confronted with consistent limitations, largely due to the immunosuppressive environment generated by malignant cells.3 The induction of anti-GB immunity has been documented in vitro as well as in animal models.4 Results from several early clinical trials using dendritic cell (DC) vaccines to initiate antitumor immune responses were promising,5 indicating that antitumor immunity was induced in a fraction of patients and that immunological responders exhibited a prolonged survival rate as compared with control patients. Furthermore, increased levels of interferon ? (IFN?) in the peripheral blood as well as in peripheral blood mononuclear cells (PBMCs) of GB patients have been associated with prolonged survival, and tumor debulking is known to decrease the expression of immunosuppressive cytokines such as transforming growth factor ß (TGFß).6,7 Severe side effects have never been associated with DC-based vaccines, and the quality of life of patients treated with this immunotherapeutic intervention has been deemed acceptable.8

Although several GB-associated antigens have been identified, it is possible that the use of whole tumor-cell products as antigens (i.e., lysates, tumor-eluted peptides or fusion products between DCs and GB cells) may reduce the risk of tumor escape due to antigen-loss variants. An example of such escape has been provided by the recent results of a clinical trial targeting a tumor-associated antigen created by a large deletion of the epidermal growth factor receptor (EGFR)-coding gene (EGFRvIII), which is expressed by 25–30% of GB patients. Vaccinated patients demonstrated an increased survival rate that was correlated with increased anti- EGFRvIII antibody titers. Notably, recurrent tumors were devoid of GB cells expressing EGFRvIII, due to tumor immunoediting.9

Most clinical studies have emphasized the role of CD8+ T cells in antitumor immune responses as elicited by DC-based immu- notherapy.6,10 Although it has been suggested that CD56+ natural killer (NK) cells play a role in such responses,11 the capacity of these cells in exerting bene cial effects against gliomas (and possibly other tumors) has not been fully evaluated. NK cells are large, granular lymphocytes belonging to the innate immune system. Unlike T or B lymphocytes, NK cells do not possess rear- ranged T-cell receptors or immunoglobulin genes and instead kill target cells based on the absent expression of MHC Class I molecules.12 DCs have been recognized as major players in the regulation/initiation of both innate and adaptive immunity.13,14 Moreover, resting NK cells can be primed by the production and trans-presentation of interleukin (IL)-15 by DCs.15

In this study, we report the results obtained with 15 patients affected by recurrent GB receiving a DC-based vaccine and stress the relevance of NK cells in inhibiting tumor growth in the con- text of DC-based immunotherapy.

The clinical features of the patients are summarized in Table 1, and their main genetic features are summarized in Table S1. The median age at the time of vaccination was 46 years (range: 16–64), and the median KPS was 70 (range: 30–90). DC immunotherapy started after the third surgery in 4 patients, after the second surgery in 10 patients and at the time of MRI relapse in 1 patient (patient 6). After the first histological diagnosis of GB (according to WHO criteria), all patients received standard therapy, which consisted of conventional radiotherapy (standard dose 60 Gy) and chemotherapy (temozolomide according to Stupp’s protocol or nitrosoureas), followed by second- or third-line treatments (fotemustine, mito- xantrone, enzastaurine, diphtheria toxin). Within 72 h after the surgery, all patients were subjected to MRI examination with contrast enhancement. The exam showed residual tumors in 14/15 patients. Leukapheresis was performed within 4 weeks of the surgery. During this time, patients received a maintenance dose of glucocorticoids (4 to 12 mg of dexamethasone) and anti- epileptic treatments.

BUT THIS IS AN IMPORTANT POINT showing that theories about MRI size for non-progression patients (even in recurrence, typically do not increase above progression thresholds unless it’s PROGRESSION; Magnetic resonance imaging (MRI). MRI was scheduled within two days before the first vaccination, every 2 mo thereafter or when clinical conditions were worsening.

Two months after the first vaccination, a clear increase in tumor volume (ranging from 180% to 620%) was observed in patients manifesting disease progression, whereas a slight increase in tumor volume (ranging from to 2.3% to 23%) occurred in most of the remaining patients. For the latter group, because the increase in tumor volume was < 25%, the disease was defined as stable according to the MacDonald criteria.

Nine months after the first vaccination, an MRI performed on patient 3 showed a 187% increase in tumor volume, indicating disease progression. Although the neurological condition of the patient was unchanged, she received two further vaccinations. Without changing the steroid dosage, 5 mo later, the tumor volume had decreased by 58% and remained unchanged for the following 30 mo.

PATIENT #9, only received SoC (S, R, C (per Stupp) prior to recurrence. And with surgery, and no residual tumor (a.k.a, this Phase III nGBM intent to treat population) witnessed a response to DC vaccination, a total of 7 injections (though dosage changes, per the study: Patients 7–15 were scheduled to receive seven vaccinations. Vaccinations 1–4 were given at 2-week intervals followed by two monthly vaccinations and a nal vaccination 2 mo after the sixth vaccination. The first, fifth, sixth and seventh vaccines contained 10 × 106 DCs, while the other courses contained 5 × 106 DCs. ) Patient #9 PFS2 (recurrent progression) was 14 months PFS from debulking recurrent surgery. Patient #9 OS was 17.5 months, calculated from second surgery. PFS was calculated from the last surgery until disease progression and death/last follow-up, if censored. OS was calculated from the last surgery to death/last follow-up, if censored. This is a patient who failed newly diagnosed Standard of Care, and did not record a progression with NO RESIDUAL TUMOR, while on a DC therapy.




Table 2. Outcomes of dendritic cell-based immunotherapy in RECURRENT GBM PATIENTS. DC immunotherapy started after the third surgery in 4 patients, after the second surgery in 10 patients and at the time of MRI relapse in 1 patient (patient 6). After the first histological diagnosis of GB (according to WHO criteria), all patients received standard therapy, which consisted of conventional radiotherapy (standard dose 60 Gy) and chemotherapy (temozolomide according to Stupp’s protocol or nitrosoureas), followed by second- or third-line treatments (fotemustine, mito- xantrone, enzastaurine, diphtheria toxin). Within 72 h after the surgery, all patients were subjected to MRI examination with contrast enhancement. The exam showed residual tumors in 14/15 patients.


Patient No. of vaccinations Amount of DCs (× 106) PFS (mts) OS (mts)
1 3 vaccines 40 (× 106) 3 PFS 7.5 OS

2 3 vaccines 40 (× 106) 8 PFS 12.5 OS

3 5 vaccines 60 (× 106) 7 PFS 51 OS

4 3 vaccines 40 (× 106) 4.5 PFS 7 OS

5 6 vaccines 60 (× 106) 4 PFS 11.5 OS

6 5 vaccines 60 (× 106) 22 PFS 25 OS (died of heart failure) — DC therapy started at MRI first progression

7 4 vaccines 25 (× 106) 5 PFS 16.5 OS

8 4 vaccines 25 (× 106) 1 PFS 5 OS

9 7 vaccines 55 (× 106) 14 PFS 17.5 OS — THE ONLY PATIENT WITH NO RESIDUAL TUMOR at first injection

10 5 vaccines 35 (× 106) 12 PFS 16.5 OS

11 5 vaccines 35 (× 106) 2 PFS 7 OS

12 5 vaccines 35 (× 106) 2 PFS 7 OS

13 4 vaccines 25 (× 106) 3 PFS 6 OS

14 6 vaccines 45 (× 106) 4 PFS 7 OS

15 4 vaccines 25 (× 106) 5 PFS 8 OS


Patients 1, 2, 3, 4 and 6 received three biweekly intradermal injections of 20 × 106 (first vaccination) and 10 × 106 DCs (second and third vaccinations). Patient 3 was treated with two further monthly injections (with 10 × 106 DCs each) 9 mo after the third vaccination. Patient 6 also received one further injection of 10 × 106 DCs at progression, which was 9 mo after the third vaccination. Patient 5, who was 17 years old at the time of recurrence, was treated at the Pediatric Hemato-oncology and Neuro-oncology University Hospital Gasthuisberg and received four weekly injections (with 10 × 106 DCs each) followed by two monthly vaccinations (with 10 × 106 DCs each).

Patients 7–15 were scheduled to receive seven vaccinations. Vaccinations 1–4 were given at 2-week intervals followed by two monthly vaccinations and a final vaccination 2 mo after the sixth vaccination. The first, fifth, sixth and seventh vaccines contained 10 × 106 DCs, while the other courses contained 5 × 106 DCs.

PFS and OS were not affected by the following clinical parameters: age > 40 y; age > 60 y; immunotherapy schedule; steroid dosage at the rst vaccination and the presence of progres- sion at the first vaccination (5.7 ± 4.1 mg vs. 0.5 ± 0.9 mg; p = 0.0036). Furthermore, patients with a tumor volume < 20 cm3 (PFS: p = 0.01; OS: p = 0.003) and a Karnofsky Performance Score (KPS) > 70 (PFS: p = 0.005; OS: p = 0.4,) survived longer than the rest of the patient cohort (Fig. 1B and C).

DC immunotherapy leads to NK-cell responses in a fraction of patients. The first vaccination led to an increase in the frequency of circulating NK cells in patients 2, 3, 6, 9, 10 and 14 (Fig. 2A), which further augmented after the second vaccination. The peripheral blood lymphocytes (PBLs) of these patients exhibited increased expression levels of E4BP4 and interferon ? (IFN?), as measured by real-time PCR (Fig. 2B and C).

This Figure 2. patients with increased frequencies of natural killer (NK) cells showed an increased expression of e4Bp4 and interferon ?. (A) The time course of NK-cell frequency evaluated by flow cytometry shows a significant increase in some patients (n = 6, blue line) but not in others (n = 8, red line).(B and C) The time course of e4Bp4 and interferon ? (IFN?) expression levels as evaluated by real-time pcR shows a signi cant increase in patients with increased NK-cell frequency (n = 3, blue line) but not in others (n = 6, red line) (*p < 0.01, **p < 0.001, ***p < 0.0001 vs. first vaccination). (D and E) Flow cytometry histograms of IFN? production by cD3+cD8+ T cells and cD3-cD56+ NK cells from patient 9 and patient 11.


Our study also reported a significant, positive influence of NK-cell responses (i.e., high V/B NK cell ratio) on survival, particularly PFS, which is an interesting observation given that OS may be influenced by subsequent treatments, notably the anti-VEGF antibody bevacizumab. The patients who survived longer (patients 3 and 8), however, did show signs of CD8+ T cells responses. This intriguing finding suggests that CD8+ T cells may be involved in DC-mediated anti-glioma responses. The mechanisms that favored CD8+ T cell responses in these two patients remain unclear. Furthermore, recent data suggest that the genetic signature of GB may influence the outcome of DC immunotherapy,43 and systems biology tools may increase our understanding of the interactions between tumor cells and immune system, as recently reported.44 In particular, it is likely that systems vaccinology, which combines transcriptional profiling with flow cytometry, proteomics and transcriptomics, may allow for the identification of appropriate biomarkers that are associated with therapeutic immune responses.45,46


Encouraging data have emerged regarding the possibility that the combination of immunotherapy and chemotherapy might result in optimal antitumor responses, particularly T cell-based responses.47 In GB, the potential efficacy of combining immuno- therapy with the anti-VEGF antibody bevacizumab, which has previously been shown to exert signi cant effects on recurrent GB,48 was highlighted by the observation that VEGF inhibits DC maturation and tumor-infiltration by lymphocytes.49,50 The anti- CTLA4 antibody ipilimumab may also result in increased T-cell responses when used in association with DC immunotherapy, as recent data provided indirect evidence that ipilimumab may permeate the blood brain barrier and have an effect on melanoma brain metastases.51

Furthermore, DC immunotherapy combined with temozolomide-based chemotherapy has provided promising results in the clinical setting. Temozolomide is already a part of the standard treatment for GB,52 and both preclinical observations and preliminary clinical data suggest that the combination of temozolomide with immunotherapy may result in the development of antitumor T cell responses.53,54

The tumor volume at the time of vaccination, as measured in T1-weighted images after contrast enhancement, was correlated with survival rates. As expected, patients with larger tumors required more dexamethasone during the first cycle of vaccination, which presumably had a negative impact on DC maturation and activity.23 The negative, dose-dependent influence of bulky GBs has previously been reported during DC immunotherapy.24 Moreover, antitumor immunity may develop more effectively in the presence of small tumors, a theory that has been mathematically modeled in the case of adoptive immunotherapy for GB.25


And as we also know the Indeterminate PsPDs that were removed at baseline (prior to disease confirmation) using post surgical scan "enhancement" changes and remain alive in the Indeterminate Arm have not had a PFS event. The SIZE clock was then reset for all main arm study patients to baseline.

The "baseline" scan is where all randomized study arm patients are being measured for PFS events in this trial. And so any patient that didn't fail prior baseline, well their tumor size growth measure is reset to nadir of baseline scan, not the post surgical scan. PsPD scans tend to stabilize after the initial "enhancement". And most PsPD do not pass size limits.

Fact is that Brad (phase III patients GBM 1-3) was not removed for his PsPD size changes from his Phase I/II baseline scan of his therapy working well. If he had been classified to have a "false progression" event his DCVax-L therapy would have stopped. Here in this trial it would not.

The 10 Indeterminate patients who remain alive scans did not grow > 1 cm after the BBB distribution, when measured from baseline to month 2 scan, after being evaluated by the same Central Review Imaging then no reason to believe that PsPDs who do not have true progression will be wrongful removed for progression. Patients are removed for "unequivocal" progression after enrollment or if a new lesion shows up (size matters their too). Prior to enrollment they were removed for both "equivocal" (PsPD) and "unequivocal" progression. And as Linda Liau said, some PsPDs were excluded from main enrollment and the study design. She felt that they should have been enrolled, as they could possibly make the difference on PFS medians, but post surgical scan to baseline scan, the Central Review blinded removed simple based on size change at a time where they easily get mixed in with rapid progression. There are no rapid progression patients scans in the main arm at baseline. And again they will have baseline Nadir (lowest size point at baseline) from which to base tumor progression changes. Patients will have already gone through 6 weeks of Chemoradation to capture BBB enhancement at baseline scan. They didn't have those difference at post surgical scan and that was why many false progression patients were excluded from the main trial.

Assuming the vaccine works, even if there are "false positive" progression determination after main trial enrollment that doesn't mean patients go on to have a real event. The skeptics fail to recognize that long tail survivor in Phase I/II were all Mesenchymal. That gene group makes up a large percentage of GBM. And that gene type has historically bad OS on standard of care. The nomograms can give patient life expectations, but knowing the gene type picture helps with accuracy.

I imagine that there will be a provable statistical relevance overall survival difference between percentage of Mesenchymal patients randomized to the vaccine and the placebo regardless of crossover or not. Remember Linda Liau mentioned that Mesenchymal standard of care Chemo responsive patients, like Brad Silver (patient GBM 1-3) are expected to live around 16 months. The Phase I/II was able to show statistical significance for that Gene Type without a randomized control and many skeptics felt that was not a true significance. This Phase III study could blow Mesenchymal's historical survival data out of the water. This Phase III has a placebo control and even with crossover design, the Mesenchymal who are treated early on in the study at a time with no evidence of disease can do better then those treated after disease progression. It won't even matter if the vaccine at crossover improved the Mesenchymal in the overall placebo cohort overall survival, there should still be a statistical difference between that gene type given that fact that 47% of the placebo group did not crossover. That can be a straight vaccine verses placebo comparison that regulators can look at.

There will be placebo patients who won't crossover who fall in that Mesenchymal gene type group. Those patients will represent standard of care Mesenchymal's. If the Mesenchymal randomized to vaccine patients live years longer than may be all the study needs to show statistical significance. Now imagine the placebo Mesenchymal's crossover lives 3 years, that to me would make a strong case for approval even stronger. Those who bet against the company and the company stock may try to get investors to believe that the Mesenchymal data will simply be perceived as post hoc. And those who short the stock certainly want investors to believe that Linda Liau is wrong and that historical and placebo control Mesenchymal data both within and outside the study won't count. Meanwhile I perceive that if this study repeats both what is known about Mesenchymal's historical survival data (by way of the placebo non-crossover arm) and what is known about Mesenchymal's DCVax-L historical (by those patients in the treatment cohort) that it will be approved for at least that gene type subgroup. But hey, I like survival facts, as does the regulators as they speak for themselves. :)