Over the last few weeks, Celsion's share price has skyrocketed from below $1 to nearly $2 per share, largely on hopes that the recently presented post-hoc analysis shows that the Phase III ThermoDox HEAT trial was not a complete failure. We have spoken with key opinion leaders as well as doctors who participated in Celsion's (NASDAQ:CLSN) Phase III ThermoDox HEAT study. Their opinions were clear: the HEAT trial was a total failure. Even the post-hoc subgroup analysis that Celsion presented last month failed to show a statistically significant benefit in patients and should be viewed with extreme skepticism. Having reviewed the Phase III results and subgroup analysis, all the doctors we spoke with unanimously believed that Celsion has no chance of regulatory approval. Furthermore, our research leads us to believe that the road forward for Celsion is perilous, especially if they pursue another Phase III trial for ThermoDox.
We believe Celsion is a compelling short at these levels and have a price target of $0.73, which is the current net cash balance per share. We believe the ThermoDox asset is worthless and that the company's ongoing cash burn, which we estimate at $14 million this year, means that the company's value will continue to decline. In fact, it is not unusual for companies to trade below cash when they have no viable assets and are burning cash.
Nothing Compelling from the Phase III Results
What makes us believe Celsion is a compelling short? First, the Phase III trial was a failure. It failed to meet its primary endpoint of a 33% improvement in Progression Free Survival (PFS). In fact, there was no benefit at all! Patients who received RFA plus ThermoDox had a PFS of 13.97 months versus 13.87 months in the control arm. Look at how closely the two lines overlap with each other.
Second, while Celsion is touting the results of its post-hoc subgroup, experienced biotech investors know that such analysis is very suspect. Last month, Celsion presented a post-hoc subgroup analysis that showed ThermoDox may work better when RFA is given for greater than 45 minutes (see link here).
This is an extreme form of data-mining. When biotech companies fail in their Phase III trials, they invariably conduct a post-hoc subgroup analysis to dig for any positives. This is akin to Scott Boras representing a free agent who just finished the season hitting .200, and going around to teams and presenting data that shows the baseball player averaged .300 against lefties in 25 games. While the .300 average may seem impressive, the data set is too small to draw meaningful conclusions, is not indicative of the real world, and ignores the fact that the baseball player is hitting 0.180 against righties. In Celsion's case, by believing the subgroup analysis, investors are also believing that the ThermoDox treatment actually harms patients if it's given for less than 45 minutes. This demonstrates the absurdity of the subgroup analysis.
In addition, there's no scientific justification for Celsion's subgroup classification in the first place. A doctor who participated in the clinical trial told us that the 45 minute cut-off time was more or less arbitrary. He said that while he thought the 45 minute time frame made sense, he admitted that the 45 minute mark was a company-chosen time frame that made the data look the best.
Lastly, it's important to note that even the cherry-picked subgroup failed to achieve statistical significance. Despite Celsion's best efforts at data mining, they could not even find a subgroup that showed statistical significance! The log-rank p-values for both Progression Free Survival and Overall Survival failed to reach statistical significance and were 0.1379 and 0.2213, respectively. These levels are nowhere close to being statistically significant, which typically requires a p-value below 0.05.
When we asked key opinion leaders for their thoughts on the subgroup analysis, they all agreed that no conclusions could be drawn from the Phase III trial, nor did they find anything remotely compelling in the trends in the subgroup data. It was their unanimous opinion that Celsion cannot obtain regulatory approval in any country based on the results of the Phase III ThermoDox HEAT study, even on the results of the post-hoc subgroup analysis. They all said it would not pass regulatory muster with the FDA, the EMA in Europe, or the CFDA in China.
Risky and Expensive Road Ahead for Celsion
Celsion only option if they want to continue to pursue ThermoDox they is to run a new Phase III trial. This is not an easy task, and our research indicates that Celsion will face significant challenges if they choose to conduct another trial.
First, a new Phase III trial will take at least 5 years to complete. Celsion needs to design an entirely new trial with new parameters, obtain FDA approval for the new trial design, enroll patients, and then follow the patients through PFS and OS. As reference, the ThermoDox HEAT Phase III trial began enrolling in the second quarter of 2008 and read out in the beginning of 2013. That's 4.5 years.
Compounding matters is that Celsion may have to conduct a larger clinical trial. The doctor who participated in the HEAT study believed Celsion should go for a smaller clinical benefit than the 33% improvement in PFS that Celsion originally targeted. However, in order to achieve statistical significance for a smaller benefit, Celsion will need to enroll a greater number of patients to make the statistics work out. The HEAT trial took 4 years to enroll. A larger trial would take even longer, perhaps 5 to 6 years to enroll.
In addition, there are other significant challenges that a new Phase III trial will face that may prevent it from even getting off the group. The doctors we spoke with all said they would be wary about participating in a trial that originated from such data mining, especially if the data mining did not result in a statistically significant finding. As a result, we believe Celsion will face difficulties getting centers and clinicians on board for a new trial, especially one that will require a lot of patients and potentially take 6-7 years to complete. The clinicians also told us that RFA times are declining and that fewer and fewer procedures last as long as 45-90 minutes. This is because the base RFA procedure is getting more efficient due to advancements of the probes. Clinicians naturally feel that less surgery time is better, so Celsion will be fighting an uphill batter if they ask these clinicians to extend the RFA procedure time for longer periods.
Lastly, running a new trial is extraordinarily expensive. Celsion will need to raise a lot of money, which will heavily dilute existing shareholders. Over the course of the HEAT Phase III trial, Celsion spent $76 million in R&D and $22 million in G&A, suggesting that Celsion will need at least $100 million to run a second Phase III trial. Celsion has roughly $50 million of cash today, so they will need to raise at least another $50 million, which represents another 30 million shares on their existing share count of 62 million. That's nearly 50% dilution. We believe current shareholders don't fully understand just how much dilution they are facing if Celsion decides to pursue a second Phase III trial for Celsion.
To illustrate what this means to investors, based on the new share count of 62 million, even if Celsion gets back to its peak market cap of $327 million in early January before the failed HEAT data was released, that means Celsion shares will reach $3.50 per share 6 years from now. Biotech companies require a very high discount rate due to the risks they face, so discounting the 3.50 by 30% per year over 6 years gets a present value today of $0.75 per share. There's no upside to shareholders if Celsion decides to pursue a second Phase III trial!
Our share price target on Celsion is $0.73 per share, which is where it was before the recent increase after it was featured as one of Jon Najarian's picks for ASCO. Humorously, the company didn't even attend ASCO, so making it a pick for ASCO is odd to say the least. Even at $0.73 the stock would probably have downside due to the continued cash burn and lack of promising products.