We have reports from KOL's that results from sequential use of Abi-MDV are disappointing. Apparently the drugs confer cross resistance. I've mentioned it once before, I think it would be an interesting experiment to test Abi/Cabo in MDV refractory patients and see if that anticipated synergy can meaningfully extend the time to chemo treatment. In that setting, regulators would be more tolerant of Cabo's side effect profile.
"abi + cabo still can be better than Xtandi. Or why don't you think this comination has a chance for better results in pre chemo ?"
Anything is possible. I'll give you my reasons for skepticism. Many of Cabo's toxicity issues were resolved by lowering the standard dose from 140 to 100 and finally to 60 mg/day. However, it is still a toxic drug. There is a drop in efficacy from 60mg to 40mg and a sizable drop from 40mg to 20mg. Patients treated with Abi/Cabo in the pre-chemo setting have about a 3 year survival expectancy and treatment would presumably last 1-2 years in most cases. That duration is on the outer edge of what we've seen in other indications. Granted, nearly all Cabo trials to date are in salvage or refractory settings. So one of my concerns is tolerability of the drug combination for a long duration treatment. Even if there is a measurable survival improvement, how will the combination affect QOL and how will that tradeoff be perceived?
In the refractory settings, most patients' disease develops Cabo resistance in about 6-12 months. Certainly there is a minority that does better and we will have a better handle on Cabo resistance soon, but I think that is a reasonable guess for now. How long will it take for Abi/Cabo patients to develop resistance to the Cabo side of the combination? Is there a synergy wherein the Abi side of the combination prevents Cabo resistance? Will this synergy be sufficient to overcome the 7-8 month deficit that Abi now has compared to MDV? In the wake of the Comets we assume Cabo will be approved post-chemo. So the treatment sequence would be MDV-Chemo-Cabo. You can throw in Abi, Provenge, Alpaharadin, Cabaz, Mito, in different slots, but bare with me. If we change this sequence to Abi/Cabo-Chemo, do we confer resistance to new generation ADT and Cabo simultaneously at the expense of being able to use these drugs sequentially?
As long as we are looking at other drugs in the space, the Xtandi (aka MDV3100 and Enzalutamide) pre-chemo results are out. Usually it is problematic to draw comparisons across different trials, but in this case, the Xtandi numbers absolutely slayed prior abiraterone results. OS Xtandi HR=.7 and Abi (aka Zytiga) HR=.8. The real telling statistic is that Xtandi delayed chemo treatment by 17 months vs Abi by 7-8 months.
There is little doubt in my mind but that Xtandi will displace Abi as the first line drug of choice. The implication for EXEL is that Xtandi is the drug they need to test in combination with Cabo if they are serious about moving into the pre-chemo space. I'm still skeptical that this will be a viable indication for Cabo, but we'll see. Maybe an interesting trial would be a Cabo/Abi combo vs Abi monotherapy in Xtandi progressed patients.
So I'm starting with a feeling here that PGNX may have a penchant for putting the best face on the data at the expense of full disclosure. All companies do this to some extent, but my antennae are tingling more than normal. Back to the data. A dose reduction from 2.5 mg/kg to 2.3 mg/kg is hardly significant. So much so that I bulk all the data together as a single set. I still see sepsis as a player, and a concern that only further data can compensate for. There was also a ph 1 grade 5 for liver toxicity, so this is not a benign drug. The question is whether its anti-tumor activity will offset toxicity enough to show a positive impact on survival. The CTC counts and PSA reductions were the best data points they had and those formed the core for the efficacy argument. At the end of the prepared remarks, I was left wondering how the drug affected soft tissue. There were no observed responses and there was no waterfall plot showing tumor growth/shrinkage. There was mentioned prolonged stable disease, but without comparative data this is borderline meaningless. I also wanted to see before and after bone scans. In the Q&A analysts hit them on both concerns and the answers struck me as confirmation that the drug does not shrink tumors nor result in a significant degree of bone disease regression. So the story is good, but it is centered on PSA, CDC data and to a lesser degree, stable disease observations; all offset by observed toxicity. There may still be dosing issues to work out and design and implementation of a ph 3. Also, PGNX is not targeting the entire post-ADC/Chemo indication. They are planning on a using biomarkers to narrow the indication which would make its footprint smaller than Cabo's. I just do not see PSMA ADC as a near enough threat to Cabo to cause a revaluation of the asset. PGNX still has a lot of work to do before they even roll into ph 3. Any delay, they may have to target a post-Cabo pop, use Cabo as control, or trial ex U.S.
"Last night there was a conference call that was a bravura performance and should lead to another nice bump in the stock today ( stay tuned as the market looks like it may implode sooo...)."
Not so fast. I just listened to the presentation and I see a mixed bag. I actually like the imaging agent more than the ADC. I think they need to make a better argument for how the imaging agent will actually change how doctors treat patients. The obvious example is preventing unnecessary prostatectomies. I think there are good answers to my question, but management needs to answer that question for the benefit of regulators and investors alike.
Now about PSMA ADC. First, I have not done any in depth dd on this drug. I read a few abstracts and listened to the company presentation. I've seen no analyst reports and I've not read the annual report nor a history of press releases. First about the two sepsis deaths. It needs to be stated that a very low rate of grade 5's in refractory indications is acceptable as long as the overall therapeutic index is positive. The good must outweigh the bad. In its EXAM trial there were a handful of deaths attributed to Cabo, but the drug was still given full approval. It bothers me that there was no mention of sepsis deaths in any of the prior abstracts. Annual reports satisfy SEC requirements for disclosure, but it should not be a surprise that the market reacted negatively to the disclosure when it was made in a milieu with better dissemination. more...
"Simple question is, is this a good thing for those long and who are not a party to the convertibles, or holding short etc..?"
I don't think there will be any effect on pps. I actually expect a low conversion rate. As bonds are converted, dilution increases and long term debt decreases and it is not a secret that this is going to happen.
"Can someone also comment on whether or not this condition was met?"
Not yet, and we will need to get closer to the end of the quarter before we will know. The pps closings starting mid-Feb will be the ones that count.
"In regard to the difficulty of making assumptions about control arm performance, doesn't the prevalence of prostate cancer and its relative indolence in comparison to other forms of cancer offer a more predictable subject for speculation?"
You just can't get a perfect match for your recruited population. EXEL used the survival duration for patients coming off Abiraterone in the 301 trial as their measure. The Abi trial was post chemo. The Comets do not specify the order in which the patients received their Abi/MDV and Chemo regimens even tho the group used as a yardstick all had chemo first and then Abi. The Comet patients did not receive their Abi under the auspices of a clinical trial. Their oncs may have used different criteria for going off treatment than those used in the trial. We have no idea how the MDV pretreated patients will perform relative to the Abi group. The 301 trial was international as is the Comets, but I suspect the geographic footprints differ somewhat. I think you get my drift. The powering survival estimates are just an informed guess.
"Given that we are talking about OS and not PFS, how much scrubbing, interpretation, and evaluation is necessary when the event is death?"
You know, I've never participated in an analysis and I don't work in the industry, All I know is what I've observed and read so I'm not exactly sure why it takes so long. Let's look at EXAM. There was an unscheduled OS analysis done at the direction of FDA. The cutoff date was 6/15/12. It wasn't until 8/29 that FDA informed EXEL that an ODAC board was not necessary for the NDA. Granted, the results might have in an in basket for a few weeks, but I think that a realistic representation of how long it takes. I suspect that in addition to confirming the mortalities the status of the survivors is also confirmed to make sure an event is not missed. Then too, I don't know if the DSMB convenes physically or virtually.
You mentioned OS vs PFS and you are right that OS is good deal less complex. The actual cutoff date for the EXAM PFS analysis was 4/6/11, yet it was on 9/7/11 that EXEL announced that the event trigger had occurred and on 10/24 they announced the topline result. I assume the long lag factor is attributable to having scans read by a third party.
"Why shouldn't I conclude that the average OS is expanding with the life spans of those patients who have yet to die? My apologies in advance if this question reveals my ignorance about this subject."
Don't apologize, you have a grasp of the issue. The basis for this argument is that the control performance is a constant and that any undue lengthening in the wait for the anticipated result is due to the increased survival of the treatment arm. I'm pointing out that first, we don't know how overdue the analysis really is. The triggering even may have occurred weeks or even months ago, but the company has not received those results from the DSMB yet. Second, presuming that the control arm survival is a fixed constant is tenuous. Companies have to make assumptions about control performance in order to properly design the trial. These assumptions are based on the best historical information available, but in the real world, they are often wrong by a substantial margin, and the error is usually in favor of better control performance.
Just as mb's always assume more off label use than actually occurs, mb's also assume that any perceived or real delay in topline results is attributable to treatment efficacy. I don't want to hurt feelings or step on toes, but biotech is the Wild West of Wall Street and it pays to maintain a healthy degree of skepticism when looking at your portfolio.
"Probably a more realistic time frame than what I was assuming."
I think you are spot on in assuming we are due for some topline data, I just think it is unwise to lose objectivity and assume that more time equals more drug efficacy.
"The conclusion may be optimistic..."
I've watched this happen before. The boys over at ONTY had developed some involved math models showing that the Stimuvax trial had to be successful based on duration. Same thing with Thermodox and at CTIC a prior CEO even made an investment argument based on this presumption. There are just too many variables. Companies sit on data, discrepancies in enrolment statistics, especially inaccurate assumptions about control arm performance, all contribute to variability. You can make a fair argument for the EXAM trial. We have sequential analyses 14 months apart during which the placebo arm held pretty steady at 20 months, but even then projecting that out positively requires assumptions and we as investors (me too) tend to favor our prejudices when we make those assumptions.
"I make no assumption about the experimental arm."
"....we've had at least half the events in this arm since median OS is 10 months if we're being generous. This means relatively few events have happened in the trial arm if we still haven't gotten a signal for the interim analysis yet..."
You are making assumptions about the control arm and assuming that any departures from those assumtions is due to better performance on the experimental arm. This is almost the very definition of confirmation bias. Start with a belief and look for evidence to support that belief.
Comet 1 powering assumptions used by EXEL were for a control median of 7 months and treatment 9.3 months. Those assumptions could be way off, but we'll see. For today, lets assume a blended overall median of 9 months. So average time on trial is 8+ months and a 9 month median OS would anticipate at least 480 events by the end of February. The interim is triggered by 387 events.
It is not uncommon for powering assumptions to be off by considerable amounts, We may still be weeks or months away, but my best guess is that the CRO is cleaning the data and the DSMB is scheduled to meet or teleconference to crunch the numbers with the blind lifted.
I'm guessing EXAM first and then Comet 1.
"...it's likely that much of the placebo group will have died by now since we started accrual well over a year and a half ago now. We're likely down to mostly Cabo patients on 60 mg. The longer we don't hear news the better results are likely to be."
This is typical message board logic. Starting with the assumption that the treatment arm will outperform and ending with the presumption that it is outperforming because results aren't out yet. I wish it actually worked that way. That results have not been released does not mean that the triggering events haven't occurred. My best guess is that for both the EXAM final analysis and the Comet 1 interim, both triggering events have already occurred or will shortly. The CRO and DSMB for both can do quite a bit of foot dragging. The data collection and processing period can be lengthy.
EXAM reached full enrolment in February 2011. 330 comprised the ITT population. Final analysis is triggered at the 217th event. As of 6/15/2012, 162 had died leaving 168 from which another 55 events are required to trigger the final analysis. The median survival duration for placebo was 20 months and 26 months for Cabo. That was 20 months ago. Anyone left on trial has a minimum duration of survival of 35 months.
Comet 1 announced full enrolment of 960 on 9/26/13. EXEL has said that the majority of patients enrolled in the final 12 months. Trial enrolment is typically backended, my conservative assumption is that median enrolment was reached at the end of May 2013. That would put median time on trial at 8 months and the mean average at something more than 8 months. Powering assumptions used by EXEL were for a control median of 7 months and treatment 9.3 months. more.....
"I believe we have a Phase 2 trial recruiting to remedy that exact concern for abiraterone. Wonder if this data will speed up recruiting :)"
First lets comment on the correlation that was made. It's sort of a nobrainer isn't it? Patients who progress sooner by almost any measure whether it be PSA, new lesions, soft tissue disease progression suffer shorter survival. Adding spreading bone disease to that list just confirms that it is another manifestation of overall disease progression and worse prognosis. So it's kind of obvious, but I see his point.
So let's talk about Cabo and Abi, which is where he's going with this. The Comet 1 results may offer some insight. In addition to tracking survival, bone scan results are a secondary endpoint. This trial will likely answer whether Cabo delays the progression of bone disease and for how long this effect will last. If bone response is durable, it adds to the argument that Cabo/Abi is a potentially synergistic combination. The perception being that Abi primarily exercises its effect on soft tissue disease whilst Cabo is more bone targeted. We will wait for the Comet 1 results and hopefully have more data on which to speculate.
"No new safety signal -- those deaths were noted, I believe, and previously reported in the updated annual report at the 2.5 mg/kg dose;..."
Fair enough, although not everyone is familiar with what is buried in annual reports. What's been reported so far looks fine, but I just don't see this having much impact on EXEL. The drug is toxic, it still has a ways to go to get into ph 3, and while active, there isn't any comparative data yet on which to base a guess as to its impact on survival. If they take too long, they may wind up with a Cabo control arm for the pivotal trial.
"Down a dollar 10 after hrs. to $5.10."
It is PGNX that is down, not EXEL. Their data was somewhat disappointing and a safety signal cropped up with two deaths due to sepsis.
"I have heard a rumor from my onc doc/biotech investing friend that EXEL may announce some results at ASCO in the next couple of days."
Let me set the table and you can decide. The only applicable result would be Comet 1 interim. I frankly think it unlikely that the interim will succeed. If I am correct, my guess is that EXEL will seek a small stage to make this announcement. As of Jan 23, EXEL filed an SEC statement averring the company was not in possession of any undisclosed material information. MMM stated in December that he considers the C 1 interim a material event. EXEL has never used a scientific conference as a platform to make a major announcement. Their M.O. is to issue a press release and hold a conf call.
"I sold this pos. When management has an offering without telling the stockholders........and at well below the current stock price...."
First I believe you are sincere and I hope you are walking away break even or better. I am puzzled though. The last offering was handled much worse than the current secondary. This was done overnight and the damage was fairly minimal. After waiting years (like me) and with three pivotal trials about to report out, I would think that would be the focus behind holding or selling, not impatience with management. It strikes me as if you've run 25 miles and then dropped out of the marathon. Regardless, GL.