Also, could you please cite a current source for no PR or CR? Yes, at the last time the announced details, nothing had met the RECIST criteria for those states, but several patients were well on the way, and none had resumed disease progression, which is the point where RECIST criteria can actually be applied. At this time, I don't believe any additional information has been provided on these patients. Therefore, you are making an argument from ignorance.
Apoptosis is not consistent with an immune response, so I'm not sure why you would expect to see it. Apoptosis is when a mechanism inside the cell causes it to die. The failure of this sort of mechanism is on of the possible causes of cancer.
Cytoxic T-cell activity, such as DCVax is intended to evoke, is an external actor on the cells, and therefore will be identified as necrosis, not apoptosis. I don't know why you would expect otherwise.
I think I'll assume that all of the patients that are still alive are rapid progressors. After all, statistically, that's just as valid as your approach.
You can only look at the groups as actually defined, not pick and choose to make your desired narrative.
As my post said, you only reach the median patient on the indeterminates if you don't also roll in the definite pseudoprogressor and the long tail on the unclassifieds (2 of three being survivors). You can't logically do one without the other.
To go into your argument in a little more detail: For the moment, I'm going to ignore unclassifies and just count the indeterminates. Don't worry, I'll get back to them.
Let's assume you're right, and that 50% of the total population is rapid progressers, and even just transfer over the lowest 3 people from the indeterminate group. That drops the median OS by 1.5 people. That still leaves you with a 14 month median OS. On the indeterminate group, it pushes median OS up by 1.5 people, that puts the median OS at 27 months if and only if the surviving person died right after this was published. If, on the other hand, all three of the next group of survivors make it just another few months, the next person who has actually died is at 31 months, so we have an OS range between 27 and 31, which sounds pretty good relative to your 21 months. Not exactly what you are claiming.
Let's on the other hand, assume that only 35% of the population are pseudo progressors. In that case, we take the lowest 9 patients from the unclassified group and move them over. In that case, the median OS moves back up to about 14 months, still pretty good compared to 10.5. At the same time, the median OS of those remaining 16 pseudo progressors is indeterminate, since the median patient is still alive. It's at least 27 months, and might be significantly higher.
The only when you get the numbers you're playing with is if you add the unclassified as well as the indeterminate to the low end, but not to the high end. Yes, if you add in the low three unclassified, the median OS for rapid is about where it is in the literature. But you also have to add in the 30 month death and the 30 month and 22 month survivors to the remaining group of "pseudos". If you do that then your median OS goes to at least 27.5, and that's only if the shortest three remaining patients died right after the study was published. It could be infinitely higher, since the median patient hasn't died.
And once again, your ability to understand logic appears to be about as good as your ability to understand how the groupings were formed.
The only definitive answer you can get from this is that the patients that were definitively identified as rapid progressors before any results were known survived longer than the best available data suggests they should have. Any attempt to reclassify "indeterminate" patients is invalid.
No, there is not conclusive evidence out of this, and I wouldn't expect there to be. As you say, that doesn't happen until the Phase III completes. That doesn't mean there is zero evidence. It's not conclusive, but it is suggestive, and it does suggest a benefit.
Historical norms for non-rapid progressers is about 15-18 months, not 30. Check your data.
Define what you mean by compassionate use? Is it only approved for use after all other treatments? Yes, but that's when it's being targeted for anyway.
Unlike a US compassionate use program, however, DCVax in Germany is specifically approved for their insurance programs to reimburse, so it is not dependent on the patient's ability to pay for it, like it would be here. It will be compensated at whatever price NWBO negotiates with these insurance companies, which will certainly be profitable, if probably not quite as profitable as a fully approved product (that being said, even payment for a fully approved product will still be negotiated by the insurance companies).
Also, unlike the US Compassionate use program, it is approved for broad use, not on a case by case basis.
So, no, it is not just "compassionate use" and the nature of the approval will have little effect on the level of compensation.
No, clueless is supporting your argument by rearranging patients by the variable you are testing for. Once you start doing that, it really doesn't matter how you do it.
While it's perfectly likely that some to many of the uncategorized patients are rapid progressors, it's completely invalid to assume that they follow any particular distribution. The most likely case is that the rapid progressors in the uncategorized group follow roughly the same distribution as those in the categorized group, meaning that while you are possibly correct to move some of the shortest surviving patients to that group, you have to do the same to some of the mid-survival patients,. Since you can't be certain, you shouldn't do either, and instead take the data as presented.
Assuming that only the low end should be moved is exactly as invalid as assuming that the high end should be moved, which was my original point.
No, you cannot include people in selection groups that were done beforehand by using the results that didn't exist until afterwards. They reported the sorting exactly how they did it.
Did you actually read my entire post? Apparently not, since, as I said I would, I got to the unclassifieds by the end. You are arbitrarily switching classifications with no justification, you are treating median OS as a final number when the median patient is still alive. In fact, in general, you are abusing data in ways that no rational person would accept.
Really? Because it seems to elude you rather more severely. You are making the argument that the short end of the indeterminates are rapid progressors. That not a valid argument, but I understand why you would make the argument. However, you cannot make that argument and, at the same time, claim that only the short end are rapid progressors. It's equally likely that some of the mid range are rapid progressors, and possibly even some of the long tail.
There is simply no valid way to reclassify after the fact like you are trying to do.
I'm not saying 3, I'm not saying 5. I'm saying you can't know, and it could be anywhere from 3 to 25. If all three of the lowest are, then almost definitely someone higher up is as well, and there's no way to pick which is which after the fact. This is based entirely on the fact that it is a distribution, and you can't cut off a single tail of the distribution to make the changes you are trying to make. That's the reason your approach is flawed. Please acknowledge that you understand that it is just as probable that patient 10 from the indeterminate is a rapid progresser as patient 3?
And no, I did mean 38. I'm not saying it is, I'm defining the possible limits. Right now, if you count all patients including those still alive after you transfer the first five, you come up with 27 months as the median, with 3 of those still alive. Therefore, if 3 of them died as soon as this data was recorded, you end up with 27. If you count only patients that have died and assume all of the remaining patients outlive them, then the patient who died at 38 months is the median patient. Neither of those extremes is likely, but it's almost definitely longer than 27, and it has to be 38 or less.
And no, I can't pick a study. Take the median of all patients in all relevant studies, and you're probably coming into the 22-26 month range. 27 Months over 26 isn't statistically significant even in the full scale trial. 29 months over 22 most definitely is. Nothing in this few patients, unblinded without a control group is going to be statistically certain, but once again, the evidence is neutral to good, not completely negative as you are suggesting.
First, I'm not the one cherry picking. I'm simply taking the reciprocal of your argument. If you are going to claim that the shortest lived are rapids, then the the remainder must be pseudo.
I'll admit you're right on the median patient, I apparently miscounted by one. While that does mean the median patient is dead, it still doesn't tell us who it is. If we take out the five you suggest, then the median patient is either one of the still living patients or the one that died at 38 months. That gives us a minimum of 27 months (if we count the still living, specifically it is the second still living patient from the indeterminate group. So a range between 27 and 38, not your hard and fast 26.
Of course, if instead of just cherry picking the lowest ones, we also take a scattering of the higher numbers, which the distribution of the positively identified rapid patients suggests we should, then we are easily back into the indeterminate range.
Let me repeat this in a shorter response to make sure you get it. It is invalid to assume that only the shortest lived patients from the indeterminate and unidentified groups are rapid progressors. While it is likely that some or all of them are, it is equally likely that some of the survivors beyond 115 months are.
Someone playing games around the announced financing. Considering woodford bought in at $7.40, I'd say that's closer to the real price.
The point is terminology, not function. Reporting that the tumor shows necrosis doesn't imply that the cells died by necrosis. The two usages of the word are not the same thing.
Necro just means death. Cell Necrosis means death in a specific way, tumor necrosis does not.
The immune system also has mechanisms which induce necrosis, so your statement isn't entirely correct.
Regardless, upon going back and reading more carefully, the reports refer to tumor necrosis, not cellular necrosis. All that means is lots of dead tumor cells. The mechanism is not specifically reported on. You are reading to much into a report that was released to the general public.
Realistically 27+. Under your scenario, and adding in the longer lived indeterminates and the one definite pseudo progresser, the median patient hadn't died as of the time of this study.