Vascular Biogenics Ltd (VBLT)

[wcopeland]

Why do you people continue to listen to Staccani? This person is incorrect and irrational, and sadly is too arrogant to recognize they need to listen to others. Oren and others, you are misinterpreting the meaning of median OS. Reasonable people on this board have told you this once, but you choose to not listen. I will try once more using FACTS.

Median overall survival among 17 patients is not computed by looking at the 9th longest survival time. That way of calculating mOS is only correct if every patient has died. With incomplete observations (ie. not all patients have deceased), this way of thinking is wrong.

Median overall survival defined

Reference: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3059453/
“The time ‘t’ for which the value of ‘L’ i.e. total probability of survival at the end of a particular time is 0.50 is called as median survival time.”

Reference: https://www.verywell.com/median-survival-ms-2252158
It is the time—expressed in months or years—when half the patients are expected to be alive. It means that the chance of surviving beyond that time is 50 percent. It gives an approximate indication of the survival as well as the prognosis of a group of patients with cancer.

Reference: https://www.cancer.gov/publications/dictionaries/cancer-terms?cdrid=655249
“The length of time from either the date of diagnosis or the start of treatment for a disease, such as cancer, that half of the patients in a group of patients diagnosed with the disease are still alive. In a clinical trial, measuring the median survival is one way to see how well a new treatment works. Also called median overall survival.”

Given these definitions, the point where the curve first drops below a 50% survival probability (mOS) is at 810 days. So the median overall survival is properly listed at 810 days.

1. Survival Curve Interpretation 101: A step-drop in the curve means (at least 1 patient died). A tick on the curve means that a patient has lived at least that long, but we have yet to confirm their death (maybe because they are alive, maybe because they stopped talking to VBLT).

How many drops do you count on the survival curve in the figure? I count 7. And how many ticks/crosses do you count? I count 10. (Do not forget the one a day=0). This means that the curve truly accounts for 17 patients. But only 7 of the 17 patients have died!

If you used the flawed method of taking the 9th longest as median, we might get an estimate of 15.7 months. But then what happens if 4 more months go by and still no more patients die. The 9th longest might become 19.7 months. Do you see the problem with trivially selecting the 9th? It does not make sense to do that until AFTER all patients have deceased. While we wait for that to happen, you use a measure of survival probability.... that is, when does the curve cross below 50%?

2. The survival curve and the numbers in the table ARE NOT the same. That figure has been re-used since their ASCO 2016 presentation (http://ir.vblrx.com/static-files/f9e8a890-f2da-409d-b180-3302bc886f32). I would imagine the numbers in the table continue to get updated as the company follows up with patients.

How do I suspect this? The longest living patient on the survival curve is ~850 days (~28 months). VBLT would have to be really stupid to mix up 32.2 months and 28 months (Slide 14 in http://ir.vblrx.com/static-files/9b1c10ac-65b5-4867-9100-0c25ab7a1a9c). If you go down the list of other individuals, you can see the same thing. Most have months that are much further along that the plot suggests.

Why else do I suspect that the table simply has more recently updated data than the figure? Because the survival curve suggests that the longest survivor is still alive, the 2nd longest survivor is deceased, and the third longest survival is still alive (remember: alive, deceased, alive). However, the table shows the order of patients as deceased, alive, deceased. What does this mean? It likely means that many of the people continued to live for several months past the ASCO 2016 presentation.