Where we stand after the 2nd line update is a bit uncertain. While the 11.7 month MOS (p<0.217) is very impressive, the reality is that BPs may want more confirmatory data before signing any agreement due to the nature of bavi's MOA and target. I could be wrong on this, and perhaps any potential BP is just waiting for the FDA go-ahead on PIII. However, I am sure that potential BPs are waiting for the final OS results from the 1st-line trial. If we show something similar to 2nd-line, maybe a 40% or more increase in OS vs. C+P (e.g. C+P at 10.5 mo, B+C=P at 14.7), then this would certainly provide enough validation to partner on both indications (would drive SP north of 6 IMO).

However, if we come in at something clearly not beneficial in 1st-line, lets say <20% improvement, then I think we will have a hard time moving bavi forward without better PII results (possibly other indications).

Also, the decision by the FDA is crucial, and I believe this will go in our favor. As Garnick points out, the trial did accomplish dose identification for PIII (3mg) with continued excellent safety profile and good OS benefit. The only way the FDA asks for a re-do IMO is if they do not believe the studies done by pphm were adequate enough to rule out 3mg involvement. If no questions on investigation procedure and results are accepted, then I believe there is nothing to worry about.

Although, just because we get a go from the FDA doesn't mean we have the funds to run a PIII ourselves. This is were the 1st-line trial could help push the balance in our favor and lock down the license agreement.

IMO

What statistical significance means in Phase II trials. One needs to put some perspective on the rhetoric by all the 'so-called' analysts that without a s.s. p-value any trial is a failure. In reality, Phase II trials are not powered to prove a 5% chance of random failure (p<0.05), but instead are designed to give the developer a good signal that the therapy has the potential for success and has no significant safety concerns.

Peregrine's Phase II 2nd-NSCLC has proven exactly that. The study shows a p-value <0.217 on a 60% improvement in OS from 11.7mo to 7.3mo for control. To put this in perspective, I will quote authors Ratain and Sergent's article, "Optimising the design of phase II oncology trials: The importance of randomisation" that appeared in the European Journal of Oncology in 2009:

"We and others (Rubinstein et al.) propose that a one-sided test of the null hypothesis that the true primary outcome is no different between treatment and control with a false-positive rate of 0.20 (type I error) is appropriate. In theory, if every phase III
trials were required to be justified by a randomised phase II trial with p < 0.20, this would lead to a phase III success rate of 80%, a significant improvement from the status quo."

Although we are not quite at the p<0.2, the point is that there is enough here to warrant further investigation and we should see a major pharma challenging this in Ph III once FDA signs off on the request.

Yes the trial has discpreancies, but randomization was still maintained despite the incorrect drug given, which should strengthen the results. Additionally, a strong 1st-line NSCLC number of 40% or more increase in survival would further validate bavi's use in this indication, and potential partners will take notice.

IMO, major pharma needs new drugs, especially high priced, lucrative oncology drugs. Pipelines are unusually bare in Ph III across multiple major pharmas that have an oncology platform, and one of these companies will take a chance on bavi.

Linked to article: http://old.ecco-org.eu/binarydata.aspx?type=doc/6-Optimising-the-design-of-phase-II-oncology-trials-The-importance-of-randomisation.pdf

Meaning if pancreatic would report this month we would not be guaranteed a benefit over GEM. I agree that this trial was doomed from the outset looking at ECOG scores, and it may just be that this disease is too progressive for an immunotherapy drug to have efficacy. As FTM mentioned, even the great Avastin was unsuccessful in this tumor.

I believe lung and pancreatic tumors and disease progressions are different, and I will be doing further research to verify. Initially, I have read a study showing M2 macrophages had no impact on PFS or OS in pancreatic ca patients. This does not appear to be the case in lung ca.

iMO

My belief was that the pancreatic trial was too early to project any benefit. However, the 1st-line NSCLC MOS appears to have more room for a low side surprise. The trial is around 17mo post enrollment, and even a reading of 12-13 mos would be a marginal win, and enough to move to Ph 3. With mngt indicating 1H13 as expected read-out, we likely have a couple of months of cushion yet to come.

The 2nd-line results are a bit of a mystery with the censored patients and new control arm, but I think they still will warrant Ph 3 go by the FDA.

Also, IMO this was an overreaction to one trial outcome in one tumor type. Just as one can overestimate efficacy from a small ph 2, one can read too much into a negative result and project it on to the entire MOA.

What us ihubers should do is try determine if there is a difference in the tumor microenvironment between lung and pancreatic tumors that might explain the differences in bavi efficacy.

Yes, my chance of being below $1 at year-end (30%, vs. 70% of above $9) is based on bavi's clinical fate alone. If Cotara would be partnered and brought into Ph III, then that should keep prices above $2-$3 IMO.

The imaging potential of the PS program is interesting, but this is a long way from impacting pphm's top-line. Consensus analysts I have seen give 0 value (in terms of valuation) to programs pre-Ph II due to low rate of commercial success.

Just because a trial is running longer, that doesn't always equate to patients living longer in the treatment arm. You can take a look at the Stimuvax Ph III for a good example of this. However, I would say for our 1st-line NSCLC, the extension is likely attributed to bavi as the control group (carboplatin+paclitaxel) is unlikely to produce a MOS to justify the length of this trial (enrollment completed in Sept 2011). Meaning, bavi will likely have a benefit here (>+3-4mos MOS vs. control), but the street again can't be certain. There is a chance that the trial has completed and mngt is reviewing the data and taking longer than expected (intentionally or not). Honestly, we need an update here and mngt will address this in March (earnings and/or investor conf's).

The pancreatic trial is too early to tell anything based on end of enrollment. Although we can speculate on 10 mos of MOS, I think the separation between bavi+gemcit and gemcitabine alone is not a guarantee at this point.

Finally, I believe the street is still betting the the Ph II 2nd-line NSCLC trial will have to be redone, and for the time being, this is still bavi's lead indication (could switch to 1st-line NSCLC or pancreatic if data is supportive).

My opinion is that the street is taking a slightly negative side on all of these scenarios, which is not irrational after the monumental collapse of the SP in Sept 2012.

In terms of SP in the short-term, look for appreciation in end of Feb as we come closer to investor conf's and earnings. IMO, the current action is betting on no news until then.

Agreed, to be consider for breakthrough therapy or an accelerated approval, I would presume you would need stronger data than what the final 2nd-line results will provide. Perhaps I am wrong, but we likely lost our opportunity of a near-term approval after the discrepancies were found. A fantastic pancreatic result could change this, but this trial seems too small for the FDA to be convinced (IMO).

I don't believe that there currently is negative insider news floating on pphm to keep it around $2.00. The general investment community lost faith in mngt and bavi after the trial discrepancies, punishing the stock to lows around $0.8. Basically, they (investors) felt like pphm could be years away from moving bavi to a ph III if the 2nd-line trial needs to be redone and 1st-line NSCLC and pancreatic miss on MOS endpoints. If that is true, then $1 or less is warranted, however, mgnt has indicated they believe the data supports Ph III advancement (thus the $2 price), but investors are not yet convinced.

So what happens next? If the next PR indicates green light for Ph III 2nd-line or 1st-line NSCLC MOS is 4-5 mos better than control or pancreatic is 3+ mos better than control, then we will see prices returning to $5 in a hurry. If any of the above is coupled, we should clear $8 with ease. Finally, a BP announcement adds even more credence to the bavi story, also pushing SP higher.

Ok, that is the positive. If we don't here that 1st-line NSCLC or pancreatic is trending nicely at the upcoming investor conferences in March and no word on 2nd-line decision by FDA, I think we could easily stay at $1.50-$2 for the entire 1H13.

My prediction for an end-of-year SP for 2013 will be >$9 (70%) or <$1 (30%)

IMO

I agree that Cotara as "partnership candidate" indicates nothing specific, but I disagree with you on the 1Q to 1H as minor. You should recall that we were thinking a MOS readout for this 1st-line trial would be available in 2H12. Then we are moved to 1Q13, and now to 1H13. If we are looking at march (1Q) or april (2Q), this should be very good for bavi. Around that timing we would be looking at 19 mos -20 mos post enrollment completion of Sept 2011.

With carboplatin+paclitaxel likely not capable of >15-20 mos survival, I say this is an indication patients are living longer on bavi.

Also your timing is incorrect. We will likely have final 2nd-line results, 1st-line MOS result, and preliminary pancreatic MOS at ASCO in June 2013. These could be significant catalysts for the stock.

We will see $4 or more this year or we will be around $0.5 or less due to multiple negative PRs. I would bet on the former.

IMO

I understand the importance of the Stimuvax example as it shows one that just because a trial is lasting longer than expected, this doesn't always equate to longer survival for the treatment arm.

However, I would say Stimuvax's target patient group differs from bavituximab in that Stimuvax’s NSCLC patients had stage III stable disease, with an expected MOS of somewhere between 25 - 33 months. The literature shows significant variance in MOS depending on baseline characteristics (e.g. ECOG score).

With stage III-IV locally advanced or metastatic, untreated NSCLC, you have much lower expected survial times. Carboplatin+paclitaxel in the ECOG trial (2002) showed a MOS of 8.1 months. In the Avastin Phase III trial, C+P control had a MOS of 10.3 months and C+P+Avastin has 12.3 months MOS. Differences in the ECOG C+P and control C+P in Avastin trial could have been attributable to better ECOG PS scores in Avastin trial (40/60) vs. ECOG trial (28/67/5) and a greater % of women in Avastin (50/50) vs. ECOG trial (42/58), as women had longer MOS times in both trials.

Either way, the main point is that bavi’s treatment population in the 1st-line NSCLC trial, is stage IIIB-IV with locally advanced or metastatic disease, a more advanced patient group with lower expected survival times and thus lower chance of variance in MOS output. In other words, this means that the control group is much less likely to produce a much higher MOS than its historical averages. With bavi’s 1st-line trial being extended by multiple months, and using the enrollment completion date of Sept 2011, we are much safer in assuming it is bavi extending the survival compared to the C+P control group (we could be looking at a >18month MOS readout in this trial if reported in Q213) than those investors of Oncothyreon in Stimuvax.

IMO

Yes, RCJ, this is actually a key point. If this were a Ph III trial, the mistake would have been much more detrimental as you need to establish statistical significance (p<0.05). However, in a Ph II trial, the trials are usually not powered (not enough subjects) to establish this certainty, or you would need some outstanding results (as we actually established with the original results in Sept 2012).

Either way, where we are now is that pphm just has to show that the one arm, the 3mg, has a positive survival signal. This is a grey area, but I recall an article read previously on oncology endpoints where the authors recommended a p value of <0.2 as a threshold to move to Ph III. They indicate that this would provide a 80% chance of Ph III approval. Of course this p<0.2 is not a requirement by the FDA. I think if we came out with anything less than p<0.3 we would get an easy green light from the FDA (and even this is not a cut-off)

Furthermore, I believe the FDA understands that most drugs fail to make it to market, and it is not their job to predict which ones will and which will not. The FDA's job is to protect patients from unnecessary risks.. meaning, if the efficacy is not clear in mid-stage development, why put more patients at risks in a larger trial if there are significant side effects. However, with bavi, you don't have significant grade 3/4 toxicities, and that should allow the FDA more room in interpreting the efficacy data in determining if more testing should be conducted (i.e. going fwd to Ph III).

I encourage you to look at this FDA presentation on oncology endpoints, effectiveness, and approval. See especially slide 49, where he talks about statistical vs. clinical significance. He even mentions that Phase II results are a case where statistical significance is not necessarily required for approval if the drug would show clinical significance (I am not indicating this means bavi gets AA, but it should mean that Ph III green light is possible with our clinically significant results, but not statistically significant).

Here is the link:
http://www.fda.gov/downloads/AboutFDA/CentersOffices/CDER/ucm103366.pdf


IMO

I would actually take the opposite view. If King was correct in is statement back in late 2012 that pphm believes that at 1mg the target is saturated, then you would not see a dose-dependent effect in survival, meaning 1mg and 3mg would theoretically produce the same MOS.

Taking this further, if you missed one or two doses of 1mg due to the discrepancies, perhaps your MOS was not affected as clinical impact was already established from the other properly administered doses.

This may be entirely incorrect, but it is a feasible scenario. Nonetheless, with bavi's superior side effect profile, why take a risk of under-dosing patients in subsequent trials...in other words, just give the 3mg to be safe.

IMO

That is specifically to the point though, as "likely" doesn't work with the FDA. So even though the control arm is "likely" not affected by the discrepancies (assuming they were not frequent), you can not be certain. Thus, pphm has no choice other than to combine placebo and 1mg.

Sure we can speculate, but that is all it would be. If I had to speculate, I would say the placebo arm's MOS was not affected by the discrepancy. In a sense we are talking about two different impacts.

1. What impact, if any, on docetaxel's MOS be if one or two, etc.. of bavi 1mg doses were given?

2. What impact, if any, would bavi 1mg MOS be if it missed one or two, etc.. doses?

As noted, I would say adding a dose or two of bavi to docetaxel (if that is what happened) would likely not impact survival, meaning the 5.6 month reported survival is likely a valid number.

As for bavi's missed doses.. it is hard to speculate, but perhaps this had more of an impact than the situation listed above.

IMO

It is not possible to understand the clinical impact of the incorrect 1mg and/or placebo doses. Even if details on how many doses were actually given incorrectly (placebo arm receiving 1mg bavi or vice versa), it is really futile to try discern the clinical impact of such errors.

This is because we do not have a good understanding of bavi's pharmacodynamics to the level of predicting each dose's impact on survival. Let's say we find out that 3 weekly injections were misgiven to each treatment arm (1mg and placebo), we can't attribute a survival impact to each misdose. Meaning, we don't really know how many doses of 1mg or 3mg bavi it takes to achieve 6 mos, 9mos, 12mos, etc.. of survival.

This is likely why pphm is combining the 1mg and placebo arms. They can't believe that they can walk into the FDA and say, "well, the 1mg arm only got 2 weekly injections of placebo, therefore we think the MOS is impacted by 20% (or pick a number)".

I am not saying that the MOS reading for the 1mg arm should be completely discredited, but it should only be used as an indicator and not used in any statistical analysis.

Corporalagan, you can calculate a MOS with your example. In your example you can actually obtain a MOS with the two 'evented' patients of 9 mos and 11 mos (MOS would be 10 mos), because the two that are alive are censored, and MOS reported is only on evented patients.

Kaplan-Meier is a statistical method of approximating chance of survival by multiple probabilities of survival between time intervals (e.g. every month). In your example, the first time interval of 0 mos - 1 mos would have a survival rate of 100%. The rest goes as follows:

time interval(i)___At risk____ dead____censored___survival for(i)
0-1 mos_________ 4________ 0_______ 0______ 100% (4/4)
>1-2 mos_________ 4________ 0_______ 0______ 100% (4/4)
>2-3 mos_________ 4________ 0_______ 0______ 100% (4/4)
>3-4 mos_________ 4________ 0_______ 0______ 100% (4/4)
>4-5 mos_________ 3________ 0_______ 1______ 100% (3/3)
>5-6 mos_________ 3________ 0_______ 0______ 100% (3/3)
>6-7 mos_________ 3________ 0_______ 0______ 100% (3/3)
>7-8 mos_________ 2________ 0_______ 1______ 100% (2/2)
>8-9 mos_________ 1________ 1_______ 0______ 50% (1/2)
>9-10 mos________ 1________ 0_______ 0______ 100% (1/2)
>10-11 mos_______ 0________ 1_______ 0______ 0% (0/1)


You obtain the survival data point for each time interval by multiplying the survival for that interval by each preceding interval. So, this is easy for up to <9 mos as it is 100% in each interval, so 100% survival. Then, patient dies at 9 mos, and the chance of surviving that interval is 50%, so cumulative survival would be 50% times the preceding intervals of 100%, would give you 50%. Of course, the next interval is 100% chance of surviving, so K-M data point is 50% * 100% or 50%. And only in the next interval, >10-11 mos, do you get 0% of survival.

As mentioned, the MOS would be calculated by taking the median of the two 50% K-M data points, 9 mos and 11 mos, so 10 months.

The key here is the censored patients at 5 mos and 8 mos. If brought back into the study, they could impact MOS calculation (remember censored is also patients lost to the study because of various reasons and are not eligible to be brought back in). However, that is why companies wait to report interim results to prevent multiple censored patients that are in the study and are currently to the left of the MOS estimation

IMO

Just a simple question... how can AF's source obtain this information on these supposed 18 patients? Why would any of pphm mngt purvey any one particular 'investor' to insider information, opening them up to, yes that's right, more lawsuits. Hmmm, that seems more of a stretch than believing this ridiculous story AF has cooked up.

If pphm has indicated that 3mg is clean and they are going to the FDA to prove it, then that should be enough until further info is provided (i.e. full results).

Excellent work FTM, this is the type of research we can do to try stay ahead of the market. I agree that the Avastin C+P control arm was not a good comparison to the Phase IIa for bavi. I believe it was LifeTech Capital that mentioned this difference in dosing as well: "Investors should note that these results were achieved using less Paclitaxel (175mg/m2 vs. 200mg/m2) and less Carboplatin (AUC=5 vs. AUC=6)", but you have found a much better historical control


Granted, both C+P controls (bavi and talactoferrin) have small pt samples, and thus a risk for variance in the OS number (let's say compared to what an avg. figure would be over multiple trials or just a larger, Ph III trial). However, I feel comfortable with your 46% increase number mentioned, and believe it can be replicated in the current 1st-line trial.

Let's say bavi is already at 14mo MOS if reported today, a 46% increase would mean C+P would be 9.6 mo. That gives us some wiggle room from the 8.5 mentioned in the talactoferrin control.

Either way, we are probably going to see a headline PR of something like '50% improvement in median OS with bavi when added to C+P in 1st-line NSCLC'

Thanks for your work FTM, appreciated as always.

IMO

RRdog, can I ask how you came to the 11 month estimate for MOS on the pancreatic trial? I have tried to estimate this as well using start of trial, date of completion of enrollment, and comparing to 2nd-line NSCLC timelines.

For example, we know that the 3mg/kg arm in the 2nd-line NSCLC had a 13 month MOS reported in Sept 2012. Start of that trial was June 2010, and completed enrollment in Oct. 2011.

[url][/url][tag]NSCLC[/tag]
Months post enrollment: 11 (at data release, Sept 2012)
Months post start of trial: 31 (at data release, Sept 2012)
MOS: 13
Ratio of MOS to months post start of trial: 13/31 = 0.419

[url][/url][tag]Pancreatic cancer:[/tag]
Months post enrollment: 7
Months post start of trial: 24
calculated MOS based on months post enrollment: 8.3
calculated MOS based on NSCLC Ratio of MOS to months post start of trial: 10.1

So, I would say that we are looking at between 8.3 and 10.1 MOS, or to round, 8 to 10 mo MOS.

That was if we were to report this month. If I would have to give a final estimate now, I would say 10 months and it would be reported in late Feb. Either way, I truly believe that a partner would want to move bavi to a Ph III at a 3 month incremental benefit to the control arm or better, regardless of stat sig.

BioBS2012, I think you are spot on in painting big pharma as rigid, and stuck in their processes. This is actually a big problem in the relationship between them and biotech with disagreements on how to proceed in clinical development (biotech usually wanting to accelerate). I actually watched BMS end a licensing agreement with cabozantinib over this type of friction (by the way, that drug just got approved).

Regarding your OS question on pancreatic and 1st-line NSCLC, I think the way they are trending it is likely that at least one comes out positive enough to move to Ph III. One thing to remember is that you do not need to show statistical significance to move to Ph III (even though some bashers indicate so). In fact, a p value of <0.2 has been indicated as a good cut off in cancer trials to warrant transition to Ph III (yet this is not a requirement either). This is why your assumption on Ph III in 2nd-line is probably a good one (just depends if FDA accepts comparison arm).

My guess is we will get 14 mos in 1st-line NSCLC and prob 8-9 mos in pancreatic cancer. That may be enough to move in both indications.

yes, i think all 3 would have to miss to drive down the price to around $0.5

CP, you obviously didn't read my post. I don't think pphm will redo a Ph II to convice a partner. What I said was that sometimes phases are redone after a license agreement has been put in place. For example, if the Ph III will be large, and costly trial, the BP may want to redo Ph II to ensure it should be moved to Ph III.

That said, it is not common, and if the FDA give the go-ahead for Ph III, I don't see this happening with bavi.

As far as $500M 'pocket-money', sure that isn't much compared to the cash on hand for a lot of these major pharmas, but that doesn't mean you can just assume any risk. $500M could buy anywhere from 5-10 upfront payments for licensing agreements (much less risky).

I am familiar with IMGN as I have been watching this company for the last 3-4 years. You are absolutely right, Roche took them to the cleaners with the T-DM1 deal. The only reason why their market cap is so high now is that they are considered one of two leaders in antibody drug technology, which is a huge focus right now in pharma. I think if Roche had worse terms they would have already bought IMGN, but that doesn't mean that it won't happen still.

As for pphm vs. imgn, I think the two companies couldn't be further apart. IMGN has a technology platform and a robust pipeline, whereas pphm's fate is really tied to bavi (although the imaging potential is there). If 1st-line NSCLC and pancreatic OS miss and they have to redo Ph II, then I think we will be looking at a $.50 SP. Fortunately, I don't see this happening.

I do think that provides credence, especially if Roche is the partner. Although, looking at from the BP perspective, they are looking to poke holes so they can get better terms and reduce risk in Ph III failure.

If I recall, Medarex was bought by BMY for around $2.5B, which at the time was thought of as a risky bet. The difference is that BMY and Medarex were already partnered on ipilimumab (Yervoy), so BMS was certain in the results.

As for Amylin, Bydureon is projected to be a blockbuster and the company already had marketed products (e.g Byetta).

Also, I don't think pphm will be bought for $500M because bavi will likely be licensed before it is acquired. My guess is that if pphm is eventually acquired, it will be in the multiple of billions ($2B-$4B), and it will be likely at least 2 years from now.

IMO

CP, I am not suggesting that this scenario is likely at all, for two reasons. First, pphm mngt probably wouldn't take a buy-out now based on the depressed SP. If they actually were being courted, then the data for 1st-line NSCLC and/or pancreatic would be fantastic. Why not put a PR out there and see the SP rise to $6-7 or more. Then they can entertain offers. The second reason is that (as mentioned before) no BP will plow $500M or $600M into pphm before bavi is tested further.

I don't think Cotara provides significant value, and although Avid's business is strong, it is not something a major pharma needs (they have their own mfr resources). So, the only way a BP comes on in my opinion is if:

1. pphm is able to get FDA nod to move into Ph III in 2nd-line NSCLC
2. pancreatic OS is 9 mo +
3. 1st-line NSCLC OS is 13 mo +

That said, I think at least one of the three criteria will be met, and we will get a deal done in 1H13.

IMO

CP, it is very true that any partner would be doing a thorough DD on the company, their pipeline, and specific trials. However, when I was at BMS, when we would bring in a Ph III ready candidate we would likely have to redo Ph II to ensure that Proof of Concept has been met as the biotech would have poorly powered trials, with potential geographic focus. Additionally, the BP has to believe pphm management that proper inclusion/exclusion criteria was done when screening patients.

This is why there is a risk and why many licensed products ultimately fail in Ph III. I think bavi's potential makes up for most of the question marks, but nevertheless the BP will have to acknowledge the risk associated. As I mentioned in my original post, many major pharmas are disparate to fill late-stage pipelines, and NSCLC and pancreatic cancer are two areas of unmet need.

Good pt jakedogman1, I don't think any potential partner wants to share development responsibilities with pphm. A buyout, despite the great potential in bavi, is too risky for any big pharma to make. They would have to give at least a 70-80% premium, so you would be looking at up to $500M if an offer was given today.

Of course, just because the companies could be sharing development costs, the major pharma would be driving all decision-making and conducting the trials. I would anticipate a smaller upfront payment and a generous profit sharing for pphm. If bavi would to be a success in Ph III, then the major pharma would make the buyout offer (even though it would be likely in the billions).

IMO

FTM - here is my take on the top pharmas and their propensity to pick up a product with bavi's profile.


First, we must put bavi in terms of how a major pharma would view the opportunity. This of course will vary by major pharma, but generally it will be that bavi is

1) immuno-oncology drug
2) target of PS has been well verified in role of apoptosis, but mostly unknown role in cancer
3) extremely high risk and reward potential due to:
a) management is clearly a weakness - how much can we trust any results previously reported?
b) no clear benefit seen in response rate and PFS
c) potential for excellent OS results - carries more weight than b)

With that said, I will go through each of the top pharmas likelihood of taking on a product with bavi's profile (ordered by most likely to least likely)

1. Pfizer - has a history of taking on risky candidates, and has said previously that immuno-oncology is an interest. Recent loss of exclusivity with Lipitor has eroded top line, and mngt may want to role the dice to see if they can replace some of that revenue. Additionally, oncology is a key therapeutic area for Pfizer, but they have lagged Roche in this area. If you look at their last-stage oncology pipeline, it is anemic. They failed with axinitib in NSCLC, but it appears dacomitinib will make it to market (erlotinib competitior). Bavi would actually potentially compete with dacomitinib in 2nd-line NSCLC, but I don't think this will be a deterrent. Finally, Pfizer has one of the largest R&D budgets in pharma, and has the cash to risk on bavi.


2. Astrazeneca - All of the top pharma companies are facing patent expirations, but AZ is particularly troublesome as they will lose 80% of their revenue by 2016. With a few recent late-stage failures, AZ needs to license late-stage candidates or do some kind of M&A over the next few years. There are rumors that AZ could look to do a large scale acquisition (e.g BMS), but the more likely scenario is more licensing and small to mid-size deals (see Amylin deal). Oncology pipeline is especially thin, and bavi could fill a big gap. This would be the first move into the immuno-oncology space for AZ (also could be a strategic move as they would be playing into BMS strategy and many pharma experts believe AZ is interested in BMS and is just waiting for the right time to pull the trigger).


3. Roche - clearly Garnick's connection with Roche bodes well for a possible pphm partnership, as he can potentially alleviate some of the uncertainty around bavi's profile with his backing. Additionally, Roche is the clear leader in oncology right now, in both having the top selling product and breadth of portfolio. Roche has made a major move to work on antibody drug conjugates (e.g. T-DM1), and has little experience in immuno-oncology. Additionally, with Avastin, Tarceva, and onartuzumab, they may feel that their late-stage lung cancer pipeline is already stacked. Like Pfizer, Roche spends near the top in R&D, and licensing and partnerships are a major component of their strategy.


4. Merck - over the years Merck has shown its strength to be in the cardiovascular/metabolic space. That continues to be the case (e.g. Januvia franchise), and the bulk of their Phase III candidates continue to support this area. As one of the largest pharma companies, it is a bit shocking to see that they only have 6 cancer products in their entire pipeline (4/6 in Phase II). Merck could be making a decision to deprioritize this TA, but I doubt that they would step out completely. Bavi could be a way for Merck to get back into the oncology game.


5. BMS - arguably the leader in immuno-oncology (e.g. Yervoy, anti-PD1, etc..), and the company made this a priority years ago. BMS will try hold onto that title and I could see them doing a licensing deal on bavi as a strategic move. However, this hinges on whether management believes the bavi story, and my inclination is that this is not the case. Ranked as 4th most likely, it really could be 1st or last based on the above.


6. GSK - historically, one of the more reckless R&D organizations in major pharma as their failure rate is among the highest. Their strategy has been to take on a number of small potential products and throw as many shots at the goal as possible and hope some stick. Vaccines are a different story with GSK, as they are considered one of the best in this category. In fact, they now have a armory of 'antigen-specific cancer immunotherapeutics', including MAGE-A3 and PRAME (both being tried in NSCLC). Although there is an interest in immuno-oncology, vaccines seem to be the path GSK is taking (IMUC's ICT-107 dendritic cell vaccine for brain cancer fits them better).


7. Lilly - sometimes referred to as the sister company to BMS, Lilly shares a lot of the same approaches to drug development. Both BMS and Lilly focus on only a few therapeutic areas to try build expertise and improve success rates (as opposed to Sanofi or GSK). However, Lilly has taken a different approach in recent years by relying on internal development as a significant source of new products. This has not treated them well as they have had multiple high-profile, late-stage failures in the past few years. It will be a stretch for Lilly license a risky bavi and risk another Ph III failure.


Obviously it could be any number of companies I didn't mention (Amgen, Sanofi, etc..), but I feel these are the more likely candidates for taking on bavi. Also, this is how I see it prior to any additional data being presented. If pphm would come out with excellent pancreatic OS result and/or 1st-line NSCLC OS, then I think there will be multiple companies clamoring to sign the deal.

IMO

A little history from my side - the way i first came on to bavi was working at BMS. I was part of a R&D Strat group and I identified bavi and Cotara from a list of 100 most promising biotech molecules in clinical development. I FWD bavi onto our transaction group as I noticed they were both unlicensed. Never heard anything from them (at the time, pphm had a $100M MC). Although they are arguably the leader in immuno-oncology, I can't see them pulling the trigger on bavi now. BMS has a history of a somewhat conservative licensing strategy, and the BP that will take on bavi will need to acknowledge the major risk associated with this molecule.

Disclosure: I do not currently work for BMS anymore, although still in the pharma business.

Note that FOLFIRINOX is not necessarily a competitor to bavi, only a different chemotherapy regimen than gemcitabine. In fact, as we know, chemotherapy should enhance bavi MOA and theoretically the combination of bavi+FOLFIRINOX could produce even better survival data.

Additionally, as you noted, the severe side-effect profile of FOLFIRINOX limits its use to ECOG performance status 0-1. NCCN cancer guidelines currently recommend gemcitabine, gemcitabine+erlotinib, and FOLFIRINOX at the same level for patients with metastatic disease and good performance levels (gemcitabine for poor performance). Clearly, there is a need for new therapies for patients with poor performance - an oppy for bavi+gemcitabine.

I think this is a good point that pphm and any partner should consider when designing a Ph III trial. I would presume that two trials could be run, one with gemcitabine and one with FOLFIRINOX (or a doublet therapy such as FOLFIRI or FOLFOX).

The Lancet produced a response to the FOLFIRINOX study, which gives one an understanding of how oncologists may view the two alternatives:
http://download.thelancet.com/flatcontentassets/pdfs/S1470204510702370.pdf

Also - thanks FTM for giving me the full list of gemcitabine trials and your work on showing ratio of gemcitabine to gemcitabine+CT.

Highly cited paper on pancreatic cancer recommends a 2 month minimum improvement in PFS and OS before moving to Ph III. See link below.

Due to bavi MOA, we can forgo the PFS requirement with a strong OS signal. I would suggest that if we see something around 9mos or better for MOS, potential partners will be interested in moving bavi to Ph III in pancreatic (typical OS for gem is around 6 mos, but has been as high as 7 mos).

This 9 mos threshold is also important as bevacizumab (Avastin) showed 8.8 mo MOS in its Phase II prior to entry into Ph III (note that it ultimately failed in Ph III). There have been others that have brought forward candidates with less of an improvement in OS (e.g. axinitib with 6.9 mos MOS), but all have failed in Ph III.

Although 9 mo MOS would not guarantee success in Ph III, it should act as a threshold to get big pharma's attention. Currently, we are 7 mo post completion of enrollment (June 2011), and 24 months since start of trial. My estimation is that we are already likely looking at 8-9 mos OS if results were reported this month - hope they are not.

It is important to recall that pphm has already given an update on pancreatic trial in June 2011 with AACR presentation. Recall that enrollment was completed later that month. The poster included 66/70 patients (bavi+gem = 32 patients, gem = 34 patients). Only part of patient data was disclosed for response, with Bavi having 4/15 (27%) response rate vs. gem 2/17 (11%). Compare bavi's 27% response rate to Avastin's 21% in its Ph II. As for OS, bavi's responders ranged from 5.6 mo - 12.7 mo (median of the 4 was 7.5).

The only other pancreatic trial that pphm has done was a preclinical in 2006. Compare the safety profile of this trial to current Phase II and you notice similarities in vascular effects. From researchers in PC trial: "The lack of toxicity to normal vasculature of 3G4-based therapy is likely due to the specificity of PS exposure on tumor endothelial cells" The study also showed reduction of metastases, tumor weight reduction, and 'marked differences in the clinical course of the animals...' (bavi arm much healthier, although OS not mentioned). At the molecular level, there were signs of improved macrophage infiltration with bavi arm, as well as some alteration of tumor vasculature (although minimal).

insert-text-here
http://www.24hmb.com/UpLoad/Editor/2010/1/27/2010012761108861.pdf

insert-text-herehttp://onlinelibrary.wiley.com/doi/10.1002/ijc.21684/full

Bavi 1mg + Placebo is estimated to be between 7.4 mo and 9 mo based on original KM curve. Back in early Sept when we were given the Kaplan Meier curve for the NSCLC 2nd line I tried to recreate the # of evented (died), at risk (still alive), and censored (dropped out of study for some reason or are still alive and possibly added back to KM curve). I did this to try estimate the final MOS numbers.

Now that pphm will combine the placebo and 1mg treatment arms, I combined my previous estimates and came up with a likely MOS of between 7.4 mo and 9 mo, with the variability attributed to censored patients. Meaning, if all censored patients were added back to the KM curve, the MOS of placebo+1mg would be around 9 mos (I don't anticipate this to happen - likely the opposite).

Ironically, this 7-9 mo range for MOS contains the avg. MOS of docetaxel from historical studies (7.4 mos from 12 trials - believe it was FTM's data). I presume this is one reason why pphm is going with this method.

My estimation is we will see final numbers of around 13-14 mos for 3mg treatment arm vs. 7-8 mos for control arm in this trial, resulting in at least a ~60-80% improvement in OS. This will not be statistically significant, but should be enough to move to a Ph III trial due to the safety profile of bavi 3mg arm.

Additionally, the FDA will likely want to see how the 1st-line trial reports in final MOS and safety profile as it may help validate safety/efficacy risk-balance of bavi in NSCLC, regardless of line of treatment.

Major pharma co. BMS highlights tumor microenvironment on immuno-oncology page (see link below). Bristol is one of the leading major pharmaceutical companies in immuno-oncology right now following the approval of their highly touted melanoma drug, Yervoy. Their next big immuno-oncology drug is anti-PD1 (made headlines at ASCO 2012), which they have in Ph III right now (in lung and melanoma).

I am not necessarily indicating BMS is going to take on bavi, but it is a good sign when they are talking about the tumor microenvironment. On the link below they discuss immunosuppresive factors, including TGF-B and IL-10, which bavi helps reduce.

Although not validating the PS target, this helps pphm sell the bavi story to other major pharmas. What this means is if major pharma is given positive phase II data, MOA should not be a limiting factor.

IMO (i have worked for BMS R&D Strategy)







http://www.bmsimmunooncology.com/tumor-immunosuppression.aspx

Nice post ftm, and just to add that the FDA does not require a placebo controlled IIb to move into phase III testing in NSCLC. Meaning, even if the FDA discredits the new 'control' arm of 1mg/placebo mix, the 3mg arm results of 13+ mo of MOS compares very favorably to the historical average MOS for docetaxel in 2nd line setting (7-8 mo), granted it is a small sample with just the 3mg treatment group.

Still, with an unaltered, clean 3mg treatment arm, I think bavi gets the green light to ph III trials, and a BP should be quick to sign once the go-ahead is given. However, FDA may want to see 1mg and 3mg tested in ph III to ensure 3mg is correct dose.

IMO

How PS is a validated target despite bavi being the only drug in clinical development:

As recently mentioned with the Stimuvax cancer vaccine (NSCLC) failure, a clear mechanism of action is important when investing in risky biotech companies (Martin Shkreil, Seeking Alpha). I have to agree to some extent (although there are examples of anticancer drugs or other drugs that have unclear or unknown mechanisms that have been approved by the FDA). Nevertheless, I have spent a significant amount of time researching phosphatidylserine, the apoptotic pathway, and the tumor microenvironment. Just to add context on my background: I have a BS in Biology (minor chemistry) and a Doctorate in Pharmacy.

Before I go to the reason it is validated, let me just review our understanding of tumor-associated macrophages:

As discussed on this board before, macrophages (which comprise up to 50% of solid tumors such as those found in NSCLC) are controlled by cancer cells to produce immunosuppressive cytokines (chemical signals) such as IL-10, IL-6, PGE-2 and vascular growth factors (VEGF, endothelin-2, EMAD-11) important for new blood vessel growth and metastasis of tumor.

These cancer-supporting macrophages are referred to as tumor associated macrophages (TAMs) and exist primarily in the 'M2' state, which means they are producing immunosuppresive cytokines.

Because of this immune suppressive tumor microenvironment, it has been previously thought that inflammation in cancer is correlated with reduced survival. However, recent evidence indicates that tumor associated macrophages are 'plastic' and can be switched to a 'M1' state, where they will secrete inflammatory cytokines (e.g. IL-12, TNF-alpha, nitric oxide). Studies have shown that tumors with their TAM balance shifted to the M1 inflammatory state have increased patient survival (including NSCLC)

Now, the reason why PS is a validated target is two-fold: its role in the apoptotic pathway is well documented/understood and the cytokines impacted by blocking PS (whether increased or decreased) have established roles in shaping the tumor microenvironment, including shifting TAM state from M2 to M1. Additionally, there are other drugs in approved or in development that influence this M2 to M1 pathway at different targets. The best example is the already approved bortezomib (Velcade) for multiple myeloma. Bortezomib has NF-KB inhibitory effects, which is believed to help reprogram TAMs to M1 state (as well as inhibit degradation of pro-apoptotic pathways). Another example is STAT3, which is a transcription factor that when activated leads to production of immunosuppresive cytokines IL-6 and MFG-E8 that help support a M2 environment. Otsuka has a STAT3 that just completed Ph I in solid tumors.

These are just two examples of attacking the immunosuppressive microenvironment. Other targets include IL 10, CCL2, and Toll receptors. Pharma is just now realizing the potential of shifting the type and function of immune cells by way of influencing what cytokines are secreted. Macrophages, dendritic cells, T cells, and NK cells have all been indicated to be able to change role and function in cancer based on what cytokines are present. PS is one target, but it is a good one as it is found in multiple cell types and is consistently immunosuppressive in the microenvironment.

I don't see why bavi could not be active in multiple tumor types, starting with NSCLC. Looking forward to 2013 OS results.

Peregrine updates timing of NSCLC trials @clinicaltrials.gov... I hope this is not a repeat post, but I didn't see it listed after doing a brief scan.

NSCLC 1st-line trial:

Previous expected primary completion date: April 2012--> now March 2013

Previous expected study completion date: June 2012--> now July 2013


NSCLC 2nd-line trial:

Previous expected primary completion date: April 2012--> now October 2012

Previous expected study completion date: June 2012--> now March 2013


Can't wait to hear the reaction from this board on implications. Obviously we are looking at great news here as these are time-to-event trials, and this means bavi patients are living longer! Great news for the disputed 2nd-line trial as Peregrine doesn't expect the study to mature with OS until 1Q13, which means discrepancies likely not a huge impact on OS signal.

Even more exciting is that there is likely going to be a clear signal in 1st-line setting as we are looking at possibly a 15 mo OS.

We should see a PR Monday as mngt will not likely update clinicaltrials.gov site with this type of info and not communicate with investors. All IMO, but if this is as good as I see it, PPHM will be making new highs very soon.

Comments? :)

Interestingly, Peregrine has added language around OS in the prostate cancer. They now will be continually reviewing OS throughout the trial (see below). This supports the continued belief that OS is vital for determining bavi efficacy. Also note that primary endpoint has changed from overall PFS to the probability of PFS.

From clinicaltrials.gov:

Overall survival [ Time Frame: 24+ weeks ] [ Designated as safety issue: Yes ]
To estimate the overall survival in subjects with CRPC (previously treated with docetaxel) following cabazitaxel + bavituximab therapy. Overall survival will be assessed continually during the duration of the study.


The primary objective of this study is to determine the probability of progression-free survival (PFS) after 12 weeks of therapy in subjects with CRPC treated with cabazitaxel + bavituximab.

Thanks for the clarification, I wasn't aware of that info. Still seams strange to me that they are presenting an update on bavi clinical program in the state they are in right now. Why give this type of presentation until investigation is complete? I guess assumption is that there will be a PR prior to Dec. 5, which could be pushing SP higher.

Just to argue with myself...one could argue that Phase II could be considered late-stage (especially Ph IIb), but the wording "advancing" indicates we are transitioning to a new stage. If we have been in Ph II, then this should indicate we are moving to Ph III