Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
I provided numerous posts explaining the calculations back after the ASCO presentation and then Larry Smith's article reiterated similar arguments. Here is the short form:
- The trial halt in Aug 2015 mathematically provides a treating endpoint of the 331 patients (actually Nov 2015 is when the last actual vaccination took place, which was ~2 months after the halt).
Therefore we know, taking Nov/Dec 2016 as being when the 248 event occurred, that 331 - 248 = 83 patients were PFS free for at least 12/13 months. However, we know that only so many patients received the vaccine Nov 2015, and that the majority of the 83 patients would all be PFS free between 14 - 16 months. That is 25% of the entire trial population. So either all 83 patients were recently treated (meaning ~2015), or a significant number of them are > 2 - 4 years PFS free.
- If recently treated, then we are seeing at least 25% PFS free ~ 14 - 16 months. Using SOC for the other 248, 15% - 20% should also be PFS free between 14 - 16 months (and that is overly conservative). So then add another 248 * 0.20 = 50 which gives 133 patients MINIMUM at ~ 14 - 16 months. That is 40% of the total trial. AND THAT IS ASSUMING THE 248 PATIENTS SHOW SOC RESULTS. The fact that we KNOW there is > 25% of the patients showing 14 - 16 months PFS free from the 83, then there will be more than 50 patients of the 248 showing 14 - 16 months PFS free. My estimate is mPFS will be > = 14 - 16 months. And that is HUGE!!! Optune only showed ~7 months (improved from 4 months).
And repeat this with OS, in that 100 patients to date are alive > 20 months. Add to that SOC results for the other 231, and you should see ~ 40% at 20 month OS, which is 0.40 x 231 = 92 patients, meaning there are ~ 192 patients at least with OS of 20 months or more. That's nearly 60% and would mean that the mOS of the entire trial is somewhere greater than 21 months!
There you go, which strengthens my case! I recalled when they were in the ~$2B range. Didn't realize they jumped in recent months. That in itself blows my mind that NWBO is $0.20 a share. The only reasoning I can think of is that Big Pharma, or Big Investors somewhere were targeting NWBO to short their stock and force them to dilute in the hopes they would go bankrupt before results.
I don't believe that there is information out there telling the shorts that NWBO's P3 is a bust, mainly because of the results that we know to date. 2/3 of the patients would have to show significantly worse results than SOC in order to off set the results of the 83 PFS free over 14 months, and 100 patients alive greater than 20 months.
NWBO Stock Evaluation
To be honest, and I understand that there has been substantial dilution of the stock in the last couple years, however, I believe that the $4 -$5/share ($2 Billion Market Cap) estimate of Chris LaCoursiere is a reflection of what NWBO should be right now, awaiting P3 results, not if P3 results are positive.
Here is why;
1) #1B - $2B is the market caps of KITE & JUNO, both without any approved products.
2) NWBO would have access to both US & European markets right away; so say 50% of each which gives ~ 14,000 cases (14,000 US cases and 14,000 European cases).
~$130K per treatment; ~$2B in revenue potential from GBM alone.
This means that their market cap should be significantly higher than $2B.
3) Thirdly, a P3 positive result, and ultimate FDA approval would be a huge breakthrough for Immunotherapy. And with DCVax Direct and the Opdivo trial means there could be additional diseases and products that would provide revenue potentials > $10 Billion per year. If P3 results are positive (and especially considerably positive, as my estimates of mPFS is between 12 - 16 months, and mOS of 26 months or more), this would put a lot of optimism that there will be additional fast tracked DCVax products like Direct and combination treatments.
Therefore, and because stocks are all about emotions, over bought, and over sold, I believe NWBO shares could easily jump to ~ $20 - $30 a share with a market cap of $12 Billion. Or higher. Of course you may have that initial jump, like up to $40 or $50 and slow correction down to $25 or $30 a share, but positive P3 results could easily see a huge jump, especially if there are still a lot of naked shorts.
SmithonStocks, reiterates my Mathematical Arguments.
As I stated numerous times, and it has been reconfirmed through Larry Smith's latest article; the mathematical data that is known to date regarding the 100 patients still alive, as well as the 83 patients that were PFS free late in 2016/early 2017. The fact that nearly 1/3 of the patients are showing better results than SOC, and that doesn't include any data from the other 2/3 of the patients, strengthens the case that the trial should have extremely good mPFS and mOS results.
Tick tock tick tock....
SmithonStocks, reiterates my Mathematical Arguments.
As I stated numerous times, and it has been reconfirmed through Larry Smith's latest article; the mathematical data that is known to date regarding the 100 patients still alive, as well as the 83 patients that were PFS free late in 2016/early 2017. The fact that nearly 1/3 of the patients are showing better results than SOC, and that doesn't include any data from the other 2/3 of the patients, strengthens the case that the trial should have extremely good mPFS and mOS results. Tick tock tick tock....
Not sure about CLDX but I believe that IMUC P2 did not do a crossover, and their P2 results showed statistical significant increase in PFS, albeit a small one of 1 month, from 10 months to 11 months, and the OS was not statistically significant with mOS SOC of ~19 months and vaccine mOS ~21, it was ~1.9 months difference which is less than the required 2 months. I didn't provide the decimals, so its a round about number for each.
Also, from what I have heard, IMUC's P3 trial, which also doesn't have a crossover, is having problems enrolling patients, and is having trouble getting patients enrolled. It is difficult to get enrollment when a patient is given a 33% chance of getting a placebo and rather than knowing that no matter what, they will crossover and get the vaccine, like in the case with DCVax-L.
I will have to recheck if the 90% is overall, or 90% of SOC crossed over. Either way, that means that there will be ~11 - 33 patients that would not have crossed over. Moreso, I would think a good portion of those patients either passed away prior to their tumor progressing (meaning early on after enrollment), or were PFS free at the time of the presentation. Either way, it wouldn't really provide a representative sample of SOC patients.
Thanks. Yes, I wonder how many people are utilizing the enrollment graph to look more closely at the potential results.
I disagree with a SOC mPFS result being higher due to cherry picking. If anything has shown consistency between numerous GBM trials with SOC arms, it has been the mPFS, which has typically been 7 - 9 months. Actually, I just realized, I used 11 months to be conservative as the ICT-107 P2 trial showed SOC of 11 months for HLA patients. Even MGMT patients that typically have longer OS, show mPFS results less than 9 months. So assuming 11 months median PFS, 25% of the patients would need to be PFS free for 20 months or more.
AND, we know that the median PFS of the final 83 patients was atleast (and most likely much larger) 16 months. Which is 5 months more than the 11 months.
I did work them back. I used the enrollment schedule. That is why I have the 83 starting back in Feb/Mar 2015. And therefore there is at least some patients of the 83 that are 20-21 months PFS free. And the median value of these 83 would be around 16 months.
I don't think you can evaluate SOC without crossover. only 10% haven't crossed over yet. And those one could be either they passed away before they could crossover, or they are part of the PFS free patients, still alive today. Remember, about 15% of the GBM population live longer, and PFS free.
So really there is no way to compare SOC without vaccination in this trial. One can only compare SOC + late vaccine to SOC + early vaccine. The comparison will show if taking the vaccine earlier is more effective than taking it later. And if there is a 2 month difference, then that secondary endpoint is met. That is why the PFS is the primary endpoint, and the most important aspect of this trial. Strong results with PFS will allow this to move to accelerated approval, as it provides a surrogate endpoint to a very aggressive, lethal disease.
Not saying its necessarily an issue, but if DCVax-L is truly effective, the secondary endpoint trial evaluation, an OS increase of 2 months, could theoretically fail. Meaning the crossover arm could have an mOS of 26 months, and the DCVax-L initially treated arm could be 27 months. The secondary endpoint fails to meet the 2 month difference, BUT, the fact that both median OS values are significantly greater than the historical mOS of the SOC. That is why the PFS results are extremely important. And is what will drive the approval in my mind if the OS endpoints are not met.
I will explain exactly how this is calculated, and it doesnt really change much by using Nov 2016 instead of Dec 2016.
From the enrollment graph, we know how many patients were enrolled, in each specific month since 2012. By laying this out numerically in a spreadsheet, and utilizing a median PFS event time of 10 months, the only way to have 85 patients without a PFS event by Dec 2016, would require numerous patients to have a much longer PFS free eventing.
For example, if you assume mPFS of 10 months, 30% PFS free at 18 months, and hold 20% of all patients to have > = 24 months, this would trigger 85 PFS free events by June 2016. Using this assumption, which is the SOC statistics, ~ Nov 2015 would be the month in which the most enrolled patients were PFS free at ~ 146.
2nd scenario; take the worse case scenario; all of the last 83 enrolled patients are the 83 patients that were PFS free in late 2015. If that is the case then we would have the least possible mOS (of those 83 patients), of ~ 16 months. The shortest PFS length would have been those treated in Nov 2015, so even if you assume the 248th event occurred in Nov. 2016, all 83 patients had a PFS event free time of at least 12 months.
That is why I am saying that mathematically, it is showing that the PFS results are incredibly high, and if the SOC population (which is a 1/3 of the trial) shows mPFS eventing in the 8 - 10 month range, this means that the vaccinated population must be showing an even greater result. MEANING THE 83 PATIENTS THAT ARE PFS FREE ARE MOST LIKELY NEARLY ALL VACINNATED NON CROSSOVER PATIENTS!!!
From the data, I believe that the PFS data is overwhelmingly positive, and that is why I am so positive about the P3 trial. However because of the crossover arm, it could be very likely that the OS endpoint may not achieve an increase, however, with the cross-over aspect, depending on the mOS, this may not matter much. Like you say, if the SOC mOS is actually more like 26 months, and due to the crossover, even if the vaccinated group shows a mOS of only 27 months, but the mPFS is a significant increase, the FDA would be hardpressed not to approve DCVax-L.
I actually do not think that the ICT-107 P2 mOS results and DCVax-L P3 mOS will be similar (i.e. ~ 20 months), I was only using the scenario that the SOC could have been moved up closer to 20 months mOS, as was seen with the ICT-107 P2 trial. Optune showed a mOS considerably smaller, and they had a much larger population than the ICT-107 P2 trial. However, the shorts like to argue that the DCVax-L is cherry picking the patients or removing patients from the trial that would effect the mOS negatively.
If in truth we know that mOS of SOC is actually going to be closer to 14 - 16 months, then the data shows that DCVax-L is overwhelmingly positive, however, because it is a crossover, the SOC population might actually have a higher mOS if the vaccine actually works.
Latest Calc - ~40% with PFS of 20 months or more.
Sorry, been busy, and haven't been able to crunch numbers since the ASCO presentation, however, the enrollment schedule graph which was provided, and was extremely similar to my original estimation (you have to take my word on that as I didn't provide my excel spreadsheets, but my numbers came out similar if you back calculate).
However, HERE IS THE BIG CALCULATION: Using that enrollment schedule, in order to have ~83 patients without progression in December 2016, and using a high estimate of ~10 months for the median PFS (using a smaller median length would show even higher/better results), the entire population would need to see ~ 40% of the patients with PFS of 20 months or more. Anything less, and there is no mathematical way that this could occur.
Moreover, if you try and say that the 98 PFS free patients were the last 98 treated, and calculated the least PFS scenario possible, then you would have a median PFS of ~ 19 months, with the last 10 - 20 patients having a PFS of 4 - 5 months (but the first 10 -20 patients would need to have a PFS of 23 - 24 months).
When you look at OS, it isn't as convincing, but still shows a very high result, with ~40% of the patients requiring an OS of 25 months or higher.
What I cant believe is that investment firms aren't using mathematicians to crunch this data to provide worse case, scenarios.
Granted, mathematically it is true that both endpoints could fail, PFS & OS, meaning there is no difference in the two populations, (at least 2 - 3 months; 2 months OS & 3 months PFS), HOWEVER, with the data we are seeing, THAT WOULD REQUIRE THE ENTIRE POPULATION TO HAVE 40% OF THE PATIENTS WITH A PFS OF 20 MONTHS AND 40% OF THE POPULATION TO HAVE AN OS OF 25 MONTHS OR MORE.
Lastly, my thoughts on this. With just the OS numbers, I would not state that the evidence is overwhelmingly positive, its a little too tight (due to the ICT P2 results, though I wonder if there is a reason the median OS was 20 months), however, the PFS results are OUTSTANDING. THERE IS MATHMATICAL PROOF THAT THE ENTIRE POPULATION HAS A HIGH PFS FREE LENGTH, AND THIS MEANS THAT EITHER THE ENTIRE POPULATION IS SEEING THIS (WHICH NO OTHER PFS RESULT > 10 MONTHS HAS EVER BEEN RECORDED), OR THERE IS A SIGNIFICANT DIFFERENCE AND EFFICACY, MEANING THE PLACEBO PATIENTS ARE SEEING 10 MONTHS AND THEREFORE THE VACINATED GROUP IS SEEING 15 - 20 MONTHS PFS FREE IN ORDER TO AVERAGE THAT OUT.
And lastly, I have to state that I really dislike crunching the numbers, in terms of these patients are people, that have or had lives, families, and they aren't and should never be just another number or statistic. I just see that there seems to be efficacy with this vaccine, even if it is mainly PFS, and it blows my mind that for whatever reason, people, investors, shorts, blog reporters, would attack and profit off a company that could have a vaccine that could improve the short life of someone that has this horrible, aggressive, heartbreaking disease.
What I am calculating is the OVERALL population, meaning both placebo and vaccine. And yes, without unblinding data, I can never know what the difference will be; HOWEVER....
If there is a significant number of OVERALL population (at least 30% with PFS > 15 months) and when you compare this to EVERY OTHER GBM Trial, where no mPFS > than 11 months has been seen, AND you can calculate that there is somewhere between 55% - 68% of the OVERALL population with an OS > 19 months, then THE ONLY WAY THAT THE VACCINE IS NOT SHOWING EFFICACY, IS IF THE PLACEBO POPULATION IS ALSO SEEING PFS > 15 months AND OS > 20 months.
Yes, if someone could post that enrollment curve, that would help out significantly.
ALSO IN RESPONSE TO PSPD: When I stated that I do not know if the DCVax-L arm will show 2 months greater OS than the placebo arm, that is saying that without unblinded data, I can not mathematically prove that. The same goes with PFS, HOWEVER, with PFS, there isn't a single trial that I have seen, even with subsets, that shows a median PFS value > 12 months, let alone 15 months. There is the ICT-107 study that showed median OS SOC values of ~ 20 months, so its difficult to show that the SOC OS results will be significantly less than ~20 months.
BUT, the better the SOC patients do, the better the number of 231 patients that have sadly passed away will have lived 19 months or longer, meaning that there could be > 60% of the 331 living 19-20 months.
FACTS:
- 100 patients still alive
- All 100 patients have ATLEAST 19 months since their enrollment and first treatment
- 165 patients were treated between May 2014 and early November 2015
- That means the average number of patients enrolled per month in this time period is ~ 165 / 18 months = ~ 9 patients per month.
- At least 300 patients were enrolled on Aug 15, 2015
- No more than 31 patients were enrolled between Aug and Nov 2015
So what it comes down to, is using MATH/STATISTICS to figure out how many of the 233 patients ALSO lived 19 months or more.
I mean the fact that WE KNOW 1/3 of the entire population has > 19 months OS is in itself HUGE!!!!! I would even say that 25% of those will have been treated over 22 months ago.
Question: In Linda's Dec 15 presentation, she stated "85%" had received DCVax-L vaccine, right? I just want to confirm, as that coupled with today's 90%, is some good information.
I am not sure about that in terms of legality. One could argue (like a certain Adam F) that if they were to do that, it could be used against them as just trying to falsely generate "good news".
In my mind, the end results beat out Optune's result considerably. From my calculations, PFS results will be outstanding, somewhere > 15 months median value. OS results could be a bit different, I believe OS results will be much better overall (as they have stated), and median value will probably be > 21 months. What could hurt the trial, is that both placebo and vaccinated populations increase > 21 months, but there may not be a 2 month difference. HOWEVER, I think that with a HUGE PFS increase, and an OVERALL OS increase > 21 months, even if DCVax-L is only 1.9 months better (~22.9 months), there will be a very good chance that it will be approved.
Because PFS events were already met. Was just waiting on OS events.
I would say that by the end of summer, there should be Phase 3 results. But the numbers they provide, give a lot of information to calculate what the entire population is seeing. One could almost model the results.
All I can comment about is the numbers that I have to work with. But to me, these numbers suggest that their is significant increase in overall survival AND PFS with the overall DCVax-L patient population. Which means, since 1/3 is using a placebo, so SOC, that either the entire population is showing an increased result (which hasn't been seen in any trial, even subset trials in terms of mPFS, 9 - 11 months is constantly being seen). SO that means that most likely the PFS of DCVax-L compared to the placebo population might actually be > than 15 months.
Another key part that I estimated was that using the ramp up, the median point of the entire trial would have been hit around August 2014. I missed the actual date, which was 37 months ago, at ~ June 2014. So now we know that 167 patients were treated from June 2014 until Nov. 2015. When you divide those patients by those 18 months you get an average of 9 patients per month.
So that means that there is an extremely high probability that 10% of the 100 patients still alive, has an OS of ATLEAST 28 months. AND AGAIN, depending on how many patients in the 231 lived 28 months or more, it will only increase that percentage.
100 patients are still alive. Last patient was treated 19 months ago. MATH DOESNT LIE!
Estimated the last treatment to be early November 2015. AND showed that there was a ramp up, which shows that ~ 11 patients per month were being treated. So we know for sure that 30% of their patients are alive 19 months or more.
That doesn't include the data from the other 231 patients. So it means that unless the other 231 patients show TERRIBLE OS values, then there MATHMATICALLY, has to be > 60% of the entire patients population has OS of 19 months or more.
THAT MEANS THE OVERALL MEDIAN OS VALUE NEEDS TO BE MUCH GREATER THAN 19 months.
At least 30% of the patients (100) have OS > = to 19 months. NOW assume that 50% of the 231 patients lived 19 months, even assume 40% of them lived 19 months which is less than what ICT-107 showed in P2, that would mean ~40% * 231 = 92 + 100 = 192 PATIENTS lived 19 months or higher. That is ~ 60% of their patients. And if the 231 patients actually show a longer OS, then there is greater than 60% living longer than 19 months.
SHOWS EXACT NUMBERS OF WHAT I WAS CALCULATING!!
100 still alive. AND last treatment was 19 months ago, which is what I had calculated. That means they should have similar results to what I was estimating!!!
Yes, Marnix is talking about their products, and going over their technology, and their differences, DCVax-L vs. DCVac Direct for example.
I think its difficult to measure how long after PFS events have been met, that the same number of OS events will occur. Using 12 - 13 months as seen on SOC, is really just between the median values, patients at the beginning of the curve, with lower PFS values, do not see 12 - 13 months of life, and reversely, patients with extremely high PFS lengths > 15 months, may actually live much greater than another 12 - 13 months.
It all depends on the number of patients/treatments per month back in early 2015. If there was ~ 11 patients per month, that would mean that 233rd patient would potentially be alive almost 3 years by now (~ 28 months).
seconds away from knowing hopefully!
Atleast ~60% of DCVax-L patients OS >=18 months.
In actuality, it should be 58% - 68% of the 331 patients. Here is why:
If today, Dr. Bosch announces that the 232 OS events have been reached, then take a month ago it actually was reached, May 5, 2017 . That's 18 months from Now 2015 the last month any patients should have started treatment. Assuming ~ 11 patients a month, which is more conservative than the ~25 patients a quarter, and compares to the 31 patients treated between Aug 2015 and Nov 2015 (there was > 300 patients treated as per Aug 15, 2015, which means ~ 11 patients per month for the 31. If you use an increased SOC statistics (like what the shorts use to negate the effects of DCVax-L in that SOC itself has slowly increased the median OS), you would get > 50% of the 232 patients also showing 20 months, which would make an even greater % of overall patients living past 18 months ~68%.
Using conservative SOC GBM numbers, the 233 should see at least 191 patients reaching 18 months, which means then unless the study underperforms SOC standards, there should be 98 + 191 = 289.
NOW THIS IS THE KEY PART TO THIS: THIS IS ASSUMING THAT ALL 98 PATIENTS STILL ALIVE ARE THE MOST RECENTLY TREATED ONES. THAT ISNT THE CASE, WHICH MEANS THAT MANY OF THESE WILL ACTUALLY HAVE OS SIGNIFICANTLY GREATER THAN 18 - 20 months.
If NWBO failed the PFS endpoint, then it would be very very unlikely that they would be able to reach the OS endpoint. Almost impossible.
With data even from other immunotherapy trials showing PFS increases, it is much more likely that the PFS endpoint would be met and the OS endpoints (particularly with a crossover) would not be met.
NICE stats & Info;
So it looks like DCVax-L is on an STA, Single Technology Appraisal.
81% of decisions made by NICE (549 of 674) were recommended, optimised or recommended for use in the CDF.
The fact that DCVax-L has a STA in process gives it good chances, but lets look more into what this means. Moreover, 67% of cancer drugs were recommended by NICE.
Now it shows that DCVax-L is in the REFERRAL PROCESS for STA's.
"During the referral process of an appraisal, NICE asks the National Institute for Health Research – Health Technology Assessment programme (NIHR HTA programme) to formally commission the ERG or AG to produce a report.
After formal referral, NICE plans the topic into the work programme, and normally publishes the detailed timelines on its website within 6 weeks."
Therefore, NICE is still waiting for the report, but once it is provided, and this may be waiting for P3 trial data; then they begin the appraisal process which takes best case scenario, just over half a year (~38 - 43 weeks).
Not sure if this was directed to me, or iclight. Facts remain that just over 300 patients were treated before Aug 2015, and the remaining 31 were treated between then and Nov 2015 (which tells us they were at < 10 patients a month). That means that using math, we KNOW: ATLEAST
ATMOST ~31 patients of the 89 PFS free patients were PFS free 13 - 14 months (27%)
This also means that ~58 patients (18%) were ~ 15 - 16 PFS free minimum
And if you extrapolate using common sense, and ~ 10 to 15 patients a month (use 13 for this assumption);
~45 where PFS free for 16-17 months (14% Of all 331)
~32 where PFS free for 17-18 months (10% of all 331)
19 where PFS free for 18-19 months (6% of all 331)
And this doesn't include at least 230some patients that would have been treated prior to April 2015, that would increase those numbers to (using SOC PFS results):
From 27% to 47% for 13 - 14 months PFS free
From 18% to 30% for 15 - 16 months PFS free
From 14% to 26% for 16 - 17 month PFS free
From 10% to 20% for 17 - 18 months PFS free
From 6% to 16% for 18 - 19 months PFS free
Now graph those numbers next to STUPP, OPTUNE, or OPTUNE's SOC, even a graph of an MGMT sub-group that has increased efficacy vs. THE GENERAL POPULATION, these results are BETTER!!! Math doesn't lie.
iclight, SINCE THE VACCINE IS MADE FROM TUMOR CELLS, IF THERE ISN'T ENOUGH TUMOR, THEN THE VACCINE CAN NOT BE MADE. SO THOSE WITH PARTIAL RESECTION OR JUST BIOPSY, WOULDNT BE ABLE TO GET THE VACCINE, BECAUSE IT COULDNT BE MADE.
So please enlighten me just HOW you could include those patients? And you state that I don't know about trial design?
More over, no matter if you looked at the Biopsy subgroup, partial resection, etc... The median PFS doesn't jump several months, nor does the median OS value.
And the median PFS surely doesn't jump up to 15 months.
No my point that the entire population was included was not wrong. In any clinical trial, the main population body is targeted. You can say entire, main, bulk, whatever the wording you feel is correct, but no matter what, they will remove any outliers. And outliers ARE NOT SUBSETS, as they are small portions of the population that could have drastically different results than the MAIN population. SO NO I WASNT WRONG, you just don't understand Statistics. This just further proves your inability to understand Statistics.
2) THERE IS NO SUBSET IN THE DCVAX_L TRIAL. Dr. Liau stated that they are seeing OVERALL the ENTIRE trial population (meaning both placebo and vaccinated) is living longer. That provides NO INFORMATION REGARDING THE PFS RESULTS! It means that placebo patients are seeing similar PFS timelines as SOC patients, but then, once crossing over, their lives are extending longer than if they were JUST on SOC.
Wow!
Sorry, never in a large, double blind, P3 trial. My bad!
I agree. The only change I would do to RK's enrollment, and this would be using intuition, and no math, so would be a very subjective assumption, but it would be that there would be some kind of a ramp up of patients that would peak in the last several months. So I could see anywhere from 20 - 30 patients per quarter, or 10 - 16 patients a month. But one cant quantify that, only from intuition. Plus that also tells me that the ~30% we know of that is 15 - 18 months PFS free, could actually close toe 50 - 60%.