CVM Clinical Trial
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
My personal view on Overall Survival in the Study population is summarized in the following post I have made on I-Hub
2
3
4
How it works
The goal of these sheets is to get to compute the date where endpoint (298 "events") is reached knowing enrolment timeline and making assumptions on H&N survival (OAS statistics)
5
I have chosen a few OAS statistics (after having made it with many available) and made a log regression to get likely survival figures at any point in time. As this sheet is copyable, anyone can add a tab and put your own stats and share with the CVM investing community !
6
The way to do this is :
7
- Please enter a date when you think P3 will end (you can enter today's date if you haven't heard from Cel SCI about end of P3 yet !)
8
- Then have a look at the number of deaths statistically likely to this date with the computed efficacy in red (bottom of the sheet)
9
- if deaths are different than 298, than you can compute/change the efficacy number to reach 298. Efficacy number is assumed % of improved survival in MK arm
10
Once you reach 298 you can estimate Multikine benefits VS SOC based on the date you think clinical trial will reach its endpoint
11
If efficacy is better than 10%, then there are good reasons to think that the drug is efficient and therefore is likely to be approved. If it is much much better, than time to bet the farm !
12
Enjoy !
13
Sheet 1
This model is taken from Cel SCI Clinical Trial. It is believed that Cel SCI initially thought survival (OAS) at 3 years would be of 55% and 43% at year 5
14
Then Cel SCi stated that as overal survival was much better than anticipated hence the trial would take much more time than initially thought
15
The way to read this sheet is that if Cel SCi had been correct in their assessment of OAS, then the end point would have been reached by 14 Jan 2019 with a 46% improvement VS SOC or way before !!! (like early 2018) with less efficacy
16
Probably too optimistic (or pessimistic for the patients, depending on how you see this)
17
18
Sheet 2
This model is taken from some stats gathered from a quite complete Taïwanese study. Figures are from Phase III OAS data and seem more realistic with current SOC survival
19
Let's keep in mind that inclusion criteria for patient include loco-regional (local nodes) but not metastatic cancer, therefore all stage III and as well IVa cancer where prognosis is worse for patients
20
Also most countries where most patients where recruited don't have same quality of SOC than US or western Europe or Taïwan where healthcare follow up is efficient, therefore we would expect a worse OAS for those unfortunate people
21
So "we" (I mainly) believe that this study is based on more realistic OAS than Cel SCI, however we believe that it shows very conservative numbers and that in the real world we should expect worse OAS in this clinical Trial
22
This model shows that a date of Jan 14th 2019 for reaching endpoint means the Multikine improves survival by 12% VS standard of care (provided no or little drop outs) and in conclusion good hope for a happy ending
23
24
Sheet 3
This model is taken from some stats gathered from a UK study. Survival is strong. End point should be reached by May 2019 with a 10% improvement in MK arm. We believe this soc is too strong as cancer treatment in UK is much better than much of trial sites
25
26
Sheet 4- Model ...
This model is same as model 2, calibrated with 2 known leaked events : feb 2017 and feb 2018
27
Source to Feb 2017 and Feb 2018 Events table of 133 and 208 (Taïwan) :
28
29
30
31
32
33
34
35
36
37
38
Sheet 5- Dropouts
Now we added some drop-outs in this model from initial realistic model
39
6% observed drop-outs seems OK as most drop out will be observed from chemo patients (being long and painful) in SOC. No drop outs expected from fast and non toxic MK taking
40
One can calibrate with known events (feb'17 and '18, see above) and will show that mk benefit is still good and end point shouldn't be reached before October
41
Or one can try to compute 10% benefit and see what happens. I give you a hint : in order to achieve 10% we need trial to end at least by end of May '19
42
Sheet 6- CVM with Dropouts
Huge number of dropouts and still efficacy is shown very strong using Cel Sci soc survival model !!!
43
Sheet 7- Sandbox
This sheet uses realistic SoC and is unlocked for date modification and % survival improvement so that any user can input any data
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100