ImmunoCellular's ICT-107 Phase II: Patience Will Pay Off

ImmunoCellular Therapeutics (NYSEMKT:IMUC) is currently in Phase 2 trials for ICT-107 for glioblastoma multiforme, GBM. The intention of this article is to discuss the timing and likelihood of results given known information along with certain assumptions. For a background of the company and ICT-107, check out an article by Dr. Dee Kotak at PropThink and the clinical trial database.


  • Phase 1 results were impressive. Sixteen patients with 80% survival after 2 years, over 50% after 3 years in a disease where survival under current care is well under 40% at 2 years and under 25% at 3 years.


  • Uniform enrollment into each arm, treatment or placebo, given the odds (67% chance into treatment arm; 33% into placebo).

  • ~90% of patients in this trial (nearly complete or complete resections) live at least one additional year after diagnosis.

  • For the placebo group, assume 22 month overall survival. This is three months higher than the company's assumption of 18.8 months (source: Q4 '12 conference call), which was based on standard of care for patients with complete resection. However, if a tumor returns, the patient is allowed to pursue other clinical trials; this, plus some inclusion criteria (e.g., KPS score of at least 70%) adds to survival time.

  • <25% three year survival in the placebo group.

  • The treatment arm has the same efficacy regardless of manufacturing center. If this is not true, it could seriously impact top-line results.

Events: 32 and 64

The company remains blinded to the results; however, they announced on June 7, 2013 that the Data Monitoring Committee had reviewed the data after 32 deaths and recommended continuation of the study. This is an important data point: 32 deaths reached in May 2013.

The graph below shows the survival curve (black) of the placebo arm with the above assumptions (22 month overall survival). In order to have achieved 32 deaths by May, the treatment group likely has a median overall survival of over 30 months. If the treatment group resembled the placebo group, the 32nd death would likely have occurred at least two months earlier. So assuming the placebo arm has no better survival curve than what is shown in the graph below, it is likely that the ICT-107 is having some meaningful impact.

The survival curve for the treatment group was backed into based on the assumed survival curve from the placebo group and 32 deaths reached in May. With 22 month median survival, the placebo curve above can be considered a conservative case. If the median survival for the placebo group is less than 22 months, it would mean even better expected results for the treatment arm.

Patience Pays Off

Some deaths will necessarily have to come from the treatment arm, so the longer the results take to come back, the better the efficacy of ICT-107. Many biotech analysts and investors have warned against believing this. They fear a repeat of Vical, whose trial for Allovectin continued for an extended period of time, lifting investor's hope that the therapy worked; in fact, the eligibility requirement resulted in securing the healthiest patients in both arms, and the trial eventually failed. In the case of ICT-107 trial, the GBM survival curve is too steep and the percent eligible for the trial is too high (50%) for an Allovectin-like repeat. This is an important point and some biotech analysts have ignored the math in favor of broad skepticism.

Using the survival curves in the graph above, the 64th death isn't predicted to occur until March 2014, with top-line reported a month or so later.

As a reference point, the curve for the treatment group in the above graph indicates a 2-year survival of 65%; comparing this to the 16 patients in Phase 1, it would mean that instead of 13 of 16 surviving after 2 years, 11 of 16 survive. Moving farther away from that becomes more and more statistically unlikely without significant differences in eligibility criteria or treatment procedures between Phase 1 and Phase 2.

But Know the Risks...

The company had originally guided the 32nd death for Q1 2013. It could be that the treatment group is doing better than expected, but in reality, the company may have miscalculated the survival of the placebo group.

Since hiring their current CEO, Andrew Gengos, the company has been tempering expectations. While proper expectations-setting is generally wise, it does give investors pause, particularly with a few items. They have stated that top-line results could be available as soon as Q4 2013; however, really good results on overall survival would likely require a late Q1 2014 or even Q2 readout. Additionally, the company has announced that it will perform a 3-4 month data analysis. Is this an attempt to keep something in the back pocket, salvaging the data with a post-hoc analysis? Perhaps. Slicing the data by HLA status, tumor site, demographics, and even between the two vaccine manufacturing sites are possibilities.

Dilution is a definite risk. The company has sufficient cash into Phase 2 readout ($29m as of September 30, 2014 with a burn rate of $2-3m per quarter), however, the company may choose to raise funds if there is a run-up in the stock price.


Taking the aforementioned risks into consideration, an investment in ImmunoCellular is worth considering. The payoff here could be extraordinary. Depending on the difference in overall survival, the stock could jump four times or more from current levels. One approach would be to wait until early Q2 2014 to purchase shares. While the likelihood of failure diminishes sharply the later we get into Q1 without news, the share price will probably be much higher than the current $2-3's. A balanced approach to investing in ImmunoCellular would be to scale in, picking up shares over the next several months.

Disclosure: I am long IMUC. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.