Conduct of Study
Patients were enrolled from five sites in California and one each in Kentucky and New Jersey. Tumors were collected over 18 months between August 2018 and January 2020. ITT enrollment took place over 18 months between September 2018 and February 2020. TPP vaccine injections took place over 24 months, from November 2018 to October 2020. As of 9 Dec 2021, 18 of the 21 surviving patients had been at risk of death for a minimum of 15.4 months from ITT enrollment. (Of the remaining three, two did not receive AV-GBM-1 and the third withdrew at three months, after two doses of vaccine therapy.) The flow from screening to cell collection to enrollment and treatment is shown in Fig. 1. Enrollment for GBM collection continued until it was certain that at least 55 patients would be enrolled per ITT. As a result, products were actually made for 60/60 of those who completed ITT enrollment. The mean and median time from surgery to ITT enrollment date was 1.0 months (range 0.2 to 3.0), and from the enrollment date to first treatment date was 2.2 months (range 1.4 to 3.4 months).
Feasibility
As shown in Fig. 1, the success rate for establishing a tumor-cell line for each patient was 97%. The success rate for the collection of adequate numbers of monocytes per patient was also 97%, but a second pheresis procedure was required for 10. For the 63 patients who had a cell line and adequate monocytes collected, the AV-GBM-1 product was manufactured for 60/60 (100%) of ITT-enrollees. The numbers of cells for intermediate and final products are shown in Table 1. Of the 60 tumors, 46 (77%) were in culture for 28 days or less; 11 were in culture for 30 to 35 days, and the remaining three were cultured 46, 54, and 55 days. Of the 60 cultures, 58 (97%) yielded more than 1 million TICs for irradiation prior to creating the tumor-cell lysate, and more than 10 million cells were irradiated for 36 (60%) cultures. More than 450 million monocytes were cryopreserved for differentiation into DC for 52 patients (87%). There were more than 1 million cells per DC-ATA dose for 51 patients (85%), with viabilities greater than 50% for 59 (98%) and greater than 70% for 48 (80%). Of the eight patients who had fewer than 450,000 monocytes from the leukapheresis products, only three had fewer than 1 million viable DC-ATA per dose based on the final product.
Table 1
Characteristics of AV-GBM-1 products manufactured for patients (n = 60)
Variable
|
Mean
|
SEM
|
Median
|
Lower Limit
|
Upper Limit
|
Total irradiated tumor cells x 106
|
14.0
|
1.2
|
11.0
|
0.78
|
63.4
|
Number of monocytes frozen x 109
|
1.7
|
0.15
|
1.5
|
0.075
|
5.2
|
Number of DC to incubate with ATA x 106
|
750
|
108
|
560
|
38
|
5720
|
Total DC-ATA per dose x 106
|
7.9
|
0.68
|
7.4
|
0.26
|
27.0
|
% DC-ATA viable at cryopreservation
|
80.8
|
1.7
|
84.0
|
48.9
|
100
|
Patient Characteristics
Table 2 shows the characteristics of patients and their tumors. Methylation of the O-6-methylguanine-DNA methyltransferase (MGMT) gene promoter was present in 21 patients (35%), and mutations of the gene for isocitrate dehydrogenase (IDH) were present in 7 patients (11.7%). Local sites reported gross total resection of GBM for 55 patients (91.7%); however there was no central review of postoperative MRI scans. Sixty patients were enrolled per ITT, but three withdrew before starting treatment. Their ages were 50, 63, and 66; KPS were 70, 90, and 90; two were white and one Hispanic, and all three were male.
Table 2
Variable
|
N = 60 (100%)
|
Age
|
|
Median
|
59 years
|
Range
|
27 to 70 years
|
Gender
|
|
Male
|
42 (70.0%)
|
Female
|
18 (30.0%)
|
Race
|
|
Asian
|
1 (1.7%)
|
Black
|
2 (3.3%)
|
Hispanic
|
9 (15.0%)
|
Other/Unknown
|
5 (8.3%)
|
White
|
43 (71.7%)
|
KPS at baseline
|
|
70
|
14 (23.3%)
|
80
|
17 (28.3%)
|
90
|
25 (41.7%)
|
100
|
4 (6.7%)
|
MGMT classification
|
|
Methylated
|
21 (35.0%)
|
Not Methylated
|
38 (63.3%)
|
Equivocal
|
1 (1.7%)
|
IDH gene
|
|
Mutated
|
7 (11.7%)
|
Not Mutated
|
53 (88.3%)
|
Anatomic Location of Tumor
|
|
Frontal
|
Left 11 (18.3%), Right 5 (8.3%)
|
Parietal
|
Left 6 (10.0%), Right 8 (13.3%)
|
Temporal
|
Left 7 (11.7%), Right 13 (21.7%)
|
Occipital
|
Left 2 (3.3%), Right 1 (1.7%)
|
Fronto-Temporal
|
Left 0 (0.0%), Right 1 (1.7%)
|
Parieto-Occipital
|
Left 1 (1.7%) Right 0 (0.0%)
|
Temporo-Occipital
|
Left 0 (0.0%) Right 1 (1.7%)
|
Temporo-parieto-occipital
|
Left 1 (1.7%), Right 0 (0.0%)
|
Temporal and Amygdala
|
Left 0 (0.0%), Right 1 (1.7%)
|
Insular
|
Left 1 (1.7%), Right 0 (0.0%)
|
Corpus callosum
|
1 (1.7%)
|
Vaccine dosing and other therapy
The 57 patients who started treatment received 392 injections, an average of 6.9 injections per patient. Thirty-nine patients received all eight injections (68.4% of treated patients, 65.0% of ITT patients), three received seven doses, five received six doses, two received four doses, five received three doses, and three received two doses. All patients who discontinued treatment before the planned eight doses did so because of symptoms and MRI findings consistent with progressive disease, except for one patient who discontinued because of seizures and one patient who died from an unknown cause.
Concurrent therapy
During the first three weeks of vaccination, no concurrent anti-cancer therapy was given. Subsequently maintenance TMZ was administered to 54/57 (94.7%) of patients who had received at least one vaccine dose. Maintenance TMZ-based therapy began one week later in 54 patients: 29 (53.7%) TMZ alone, 14 (25.9%) TMZ plus bevacizumab, 11 (20.4%) TMZ plus alternating electrical tumor treating fields (TTF). Of the remaining three patients, one received bevacizumab and two received no additional therapy. Forty (70.2%) took corticosteroids concurrently with AV-GBM-1.
Safety
Injections were well-tolerated. Table 3 summarizes the most frequent treatment-emergent adverse events (TEAE) that were documented during and following the first vaccine injection. Only 26 AE were attributed to AV-GBM-1, 24 grade-1, and two grade-2. (Supplementary Table 1). The most common AE attributed to AV-GBM-1 were injection site reactions (15.8%), flu-like symptoms (10.5%), and bone pain (7%). An additional 74 AE were classified as possibly due to AV-GBM-1, including many that are commonly observed in GBM patients being treated with chemotherapy (Supplementary Table 2). One patient discontinued AV-GBM-1 because of seizures. There were 55 serious AE (SAE) reported for 29 patients (Supplementary Table 3). These included hospitalizations for 16 episodes of seizures in 13 patients, seven falls in six patients, six episodes of focal weakness in four patients, and three patients with cerebral edema. One patient was discovered deceased at home; he had experienced a fall two days earlier, but he had refused to go to the hospital, so the immediate cause of death was unclear.
Table 3
Most frequent treatment emergent adverse events (n = 57 patients)
Adverse Event
|
Grade 1
n (%)
|
Grade 2
n (%)
|
Grade 3
n (%)
|
Grade 4
n (%)
|
Total
n (%)
|
Fatigue
|
20 (35.1)
|
10 (17.5)
|
1 (1.8)
|
0
|
31 (54.4)
|
Headache
|
13 (22.8)
|
6 (10.5)
|
2 (3.5)
|
0
|
21 (36.8)
|
Seizure
|
2 (3.5)
|
10 (17.5)
|
7 (12.2)
|
0
|
19 (33.0)
|
Nausea
|
11 (19.3)
|
6 (10.5)
|
0
|
0
|
17 (29.8)
|
Focal weakness
|
3 (5.3)
|
8 (14.0)
|
5 (8.7)
|
0
|
16 (28.1)
|
Thrombocytopenia
|
5 (8.7)
|
4 (7.0)
|
2 (3.6)
|
2 (3.6)
|
13 (22.8)
|
Insomnia
|
8 (14)
|
4 (7.0)
|
0
|
0
|
12 (21.0)
|
Vomiting
|
7 (12.2)
|
4(7.0)
|
0
|
0
|
11 (19.3)
|
Abdominal pain
|
8 (14.0)
|
2 (3.5)
|
1 (1.8)
|
0
|
11 (19.3)
|
Fall
|
6 (10.5)
|
1 (1.8)
|
1 (1.8)
|
2 (3.5)
|
10 (17.5)
|
Dizzyness
|
8 (14.0)
|
2 (3.5)
|
0
|
0
|
10 (17.5)
|
Cerebral edema
|
0
|
1 (1.8)
|
6 (10.5)
|
2 (3.5)
|
9 (15.8)
|
Injection-site reaction
|
8 (14.0)
|
1 (1.8)
|
0
|
0
|
9 (15.8)
|
Myalgia
|
5 (8.7)
|
3 (5.3)
|
0
|
0
|
8 (14.0)
|
Depression
|
4 (7.0)
|
4 (7.0)
|
0
|
0
|
8 (14.0
|
Neutropenia
|
0
|
2 (3.5)
|
4 (7.0)
|
1 (1.8)
|
7 (12.2)
|
DVT/PE
|
0
|
3 (5.3)
|
4 (7.0)
|
0
|
7 (12.2)
|
Flu-like symptoms
|
6 (10.5)
|
1 (1.8)
|
0
|
0
|
7 (12.2)
|
Confused/Forgetful
|
2 (3.5)
|
3 (5.3)
|
1 (1.8)
|
0
|
6 (10.5)
|
Bone pain
|
6 (10.5)
|
0
|
0
|
0
|
6 (10.5)
|
Pruritis
|
6 (10.5)
|
0
|
0
|
0
|
6 (10.5)
|
DVT/PE = deep venous thrombosis and/or pulmonary embolus |
Efficacy
At analysis (12/21), 49 of the 60 ITT patients had progressed and 39 had died. Figure 2 shows OS and PFS survival curves, and Table 4 shows median survival and survival data at 6-month intervals. The patients who withdrew without starting treatment were censored from analyses at 1.6, 2.5, and 5.7 months from ITT enrollment.
Table 4 Overall Survival and Progression-Free Survival Dating from Enrollment and, Separately, First Injection: Median and Six-Month Intervals (Dec 21)
If no date of progression, observation for progression ends on last date known alive
|
From
Enrollment
|
From First
Injection
|
|
OS
|
PFS
|
OS
|
PFS
|
N (censored)
|
60 (21)
|
60 (11)
|
57 (19)
|
57 (10)
|
Median (months)
|
16.0
|
10.3
|
14.0
|
8.1
|
95% CI on Median
|
12.9:21.4
|
8.6:11.7
|
10.1:18.9
|
6.5:9.3
|
|
|
|
|
|
Percent Surviving*
|
|
|
|
|
6 months
|
89.5
|
77.7
|
87.5
|
71.4
|
12 months
|
70.2
|
34.5
|
55.4
|
28.6
|
18 months
|
45.4
|
22.4
|
40.7
|
19.6
|
24 months
|
30.0
|
17.3
|
27.7
|
13.4
|
OS=overall survival, PFS=progression-free survival
Exploratory Analyses
Using data from the 57 patients who received at least one vaccine dose, we compared OS curves from the date of the first injection between groups defined by patient or tumor characteristics. These comparisons were unadjusted, and results must be regarded as exploratory. Concerning the number of doses injected, the median OS was 20.9 months for the 39 patients who completed all eight doses compared to 7.7 months for the 18 patients who received two to seven injections (p < 0.0001). This relation may reflect dose response or simple reverse causality. Better KPS is typically associated with better OS. Median OS from the first dose was 18.8 months for the 27 patients with KPS of 90 or 100 at the time of ITT enrollment compared to 10.2 months for the 30 patients with KPS of 70 or 80 (p = 0.0643). Younger age is typically associated with better OS. Median OS was 16.3 months for the 32 patients less than 60 years old, compared to 11.5 months for the 25 participants aged 60 and over (p = 0.2403). Age at enrollment did not correlate with OS time (r = -0.07, p < 0.61). GBM patients with MGMT promoter methylation typically have a better OS. Median OS was 20.9 months for the 20 patients with a methylated MGMT promotor, compared to 12.1 months for the 36 unmethylated (p = 0.0431). The presence of IDH1 mutations is also associated with better OS. Median OS was 20.9 months for the seven patients whose tumors were IDH-mutated, compared to 12.7 months for the 50 with IDH wild-type (p = 0.1885). TTF has been shown to improve survival when administered with maintenance TMZ. From the date of the first injection, the addition of TTF during maintenance TMZ (n = 10) or later (n = 4) was not associated with better OS compared to no TTF (n = 43; p = 0.4867). Concomitant medication with corticosteroids might inhibit vaccine effects. There was no significant difference in OS between the 40 patients taking corticosteroids concurrently with the vaccine and the 17 who did not (p = 0.3258). Concomitant TMZ might inhibit the vaccine effects, or it might be of benefit because of its effects on regulatory T cells. Among patients who received at least three vaccine injections, there was no difference in survival for the 44 patients who delayed restarting TMZ until after the first three injections, compared to the 10 patients who started TMZ during the first three injections (p = 0.9568). There was no correlation between OS and the number of ITC irradiated during vaccine production, the number of cryopreserved DC-ATA per dose, or viable number per dose based on post-cryopreservation cell counts (all p > 0.18).