Oncolytics Biotech (ONCY) is currently running a two-stage Phase III clinical trial examining Reolysin in combination with paclitaxel and carboplatin in patients with platinum-refractory head and neck cancers (REO 018). This trial includes a go/no-go decision to continue enrollment in the second stage of study. Investors had originally expected a decision by early to mid-August, but one month later, they are still left in the dark. We see a number of red flags that lead us to believe the current trial is unlikely to be successful.
Oncolytics is in the process of deciding whether to move forward with the Phase III REO 018 study. It passed the initial safety hurdle, which was expected. Now the decision whether to continue enrolling in the second stage of the study will be based, in the company's words, on the following:
An evaluation of the predictive probability of success in the second stage, to be performed by an independent statistician. This evaluation will be based on second scan (12 weeks) post initiation of treatment data for all 80 patients from the first stage.
The trial completed enrollment around or before the company's PR on April 2, 2012. With regard to the efficacy analysis, 12 weeks from around this date puts us around June 25 when the 80th patient completed treatment. Suppose the trial was heavily back-end-loaded (the bulk of the patients enrolled later), which the trial was, the efficacy analysis should take roughly eight to 10 weeks. If the trial took a full eight weeks to analyze the data, this puts a completed efficacy analysis around Aug. 19 at the latest. Obviously, we are well past that date.
Management previously guided that all radiographic scans of PFS would require independent analysis to complete the interim PFS analysis. Once completed, Oncolytics contract research organization (CRO) will unblind the data and send it to an independent statistician. The statistician will run a variety of scenarios to determine the probability of success for Phase III. We believe we've seen 70% as a number quoted by the CEO. The statistician will calculate the PFS and send the information to the CEO. Only then will the company be able to make a determination on how many patients are required in the second stage of the trial. To move forward in the trial, Oncolytics likely needs to see roughly 40%-50% improvement in PFS from the Reolysin group over the control arm to realistically move this to the next stage. Oncolytics modeled the control arm to have PFS of around two months and OS of around 4.5-6.5 months.
However, we note that Oncolytics is using a very unusual two-stage design. We spoke briefly with CEO Dr. Brad Thompson on Friday, Aug. 24, about this design. When OS is so short, there is generally a high correlation between PFS and OS. The decision to move formally into stage two depends on a Bayesian analysis to determine whether under the current PFS data on the 80 patients (40 on paclitaxel/carboplatin plus Reolysin, 40 on paclitaxel/carboplatin), whether the trial can meet statistical significance on overall survival in stage two based on the historical correlation between PFS and OS.
Why Are We Skeptical?
Personally, we feel that Oncolytics Phase III trial could fail for a large number of reasons. First, Oncolytics has never run a randomized, placebo-controlled study in head and neck cancer, let alone any indication. Since inception, Oncolytics has largely run inconclusive single-arm Phase I studies and has no experience with randomized trials. This typically leads to very poor modeling of the placebo-arm and companies have a tendency to overestimate the efficacy of their own drug. For example, they might believe their drug adds more than three months to the standard of care. However, in pivotal trials this difference is notably smaller, leading to a failed study. Therefore, we believe it's possible the efficacy goal is unattainable.
Second, the REO 018 trial is in a quite heterogeneous indication, which creates a great amount of uncertainty due to varying responses amongst different histological subsets of disease. We view the catch-all nature of the trial to be conducive of a poorly designed study. Head and neck squamous cell carcinoma (HNSCC) is highly heterogeneous; the current trial includes patients with oropharynx, oral cavity, larynx, hypopharynx or squamous cell nasopharynx cancer (NPC) with distal metastasis(es). The problem with such a mixture is that different types of tumors have different responses, some better/worse than others. Oncolytics has never shown data supporting similar response rates of Reolysin across HNSCC types.
Third, Oncolytics did not screen for HPV status or RAS mutation status, both of which have been shown to have an effect on patient outcomes. This was more of a cost-cutting measure on Oncolytics' part; however, this would have been a meaningful way the company could have enriched the trial. Fourth, current data to date lacks peer-reviewed publications of efficacy with regard to hard endpoints, like PFS and OS. Early data was based on investigator-assessed PFS, not centrally reviewed PFS. Trial investigators have a tendency to overstate efficacy; hence centrally reviewed data like the current REO 018 trial tends to be less than previous studies. This is a significant risk to Reolysin showing efficacy. We have yet to see it state the PFS or OS data from the Phase I/II study conducted in the U.S. (REO 015), which completed several months ago. We even contacted one the principal investigators from the trial, who would not comment on the data or the Reolysin program in general.
Additionally, we have not seen the peer-reviewed PFS data from the REO 011 study; note, this study was only in 24 patients, all with varied conditions and previous therapies. We've heard PFS was quoted at around five months with OS at 7.1 months. However, a brief review of their investor presentations shows these values as not available. Without hard data from these trials showing supporting Reolysin, it's hard to take management's word on it. Oncolytics often talks up the objective response rate (ORR) from these two trials; however, this is an extremely subjective endpoint. Lastly, HNSCC has been a graveyard for drug development in large part because of the difficulties in treating the disease, but also highly heterogeneous.
This trial initially began in late May 2010 and completed enrollment close to the company's PR on April 2, 2012. In our view, this trial had an extremely slow accrual rate, which can be indicative of a few red flags: small potential patient population, no investigator interest, and poor execution by management. Management has previously spoken of the problems enrolling patients and it added a number of clinical sites last year and this year, however with diminishing quality. Around May 2011, the study was being conducted at a number of quality institutions, but enrollment was painfully slow. With the start of 2012, Oncolytics began enrolling patients at clinical sites in Russia, which has been a hot spot for bad clinical data in recent years.
We've spoken with a few investors that met with management in July; they were told data was expected in early August. We've heard that Oncolytics' CRO is currently double-checking all data sources and clinical forms, but this delay seems to be taking far too long. We're now sufficiently past that time window and investors should start to raise significant questions about what's exactly going on.
Previous Failures of Viral Therapies
Viral approaches to treating cancer have not gone well. Amgen's OncoVEX program, acquired through the $425 million acquisition of Biovex, uses a modified oncolytic herpes simplex virus 1 that incorporates the granulocyte-macrophage colony stimulating factor gene. OncoVex unfortunately committed the same development mistakes: single-arm Phase I/II studies, then moved into Phase III without rano. Lo and behold, last year Amgen disclosed "its decision to terminate its current OncoVEX Phase III trial in patients with squamous cell carcinoma of the head and neck to permit significant modification of clinical trial design mandated by the changing therapeutic landscape for patients with SCCHN." Rather, we suspect efficacy was the issue.
This is one of the more high profile cases of such therapies. There are about a dozen others that have been in trials and performed well below expectations. Previous to this there were two high-profile viral therapy programs that were approved in China, Oncorine and Gendicine, both of which have largely been busts because of small datasets and lack of reproducible results. Even Onyx Pharmaceuticals attempted to pursue a viral therapy program, ONXY-015, which was eventually ceased and sold off. We've not heard this program mentioned by the acquirer in many years.
The reality is that oncolytic viruses have largely been a science pipe dream. There have always been signs of efficacy, but always in small trials that are never backed up in larger studies. In our opinion, Oncolytics' Reolysin has shown some signs of efficacy; however, we believe the company is testing the drug too late in disease progression. We don't view Reolysin as more special than other programs. Previous attempts utilized engineered viral approaches that sought to increase potency and added additional firepower. Reolysin (reovirus type 3 Dearing) is a wild-type reovirus that is commonly found in humans, which is claimed to preferentially lyses (kills) cancer cells.
Oncolytics has roughly $35 million in cash currently, with 77 million shares outstanding and a fully diluted share count of 84 million. If the current trial does not move into Phase III and is scrapped, Oncolytics could easily trade down to cash value, which is less than $0.50 a share. It also has an open $150 million shelf that can be used at anytime. Oncolytics entire pipeline is built on Reolysin, so this failure would look cast some doubt for the whole platform and ongoing trials.
There are clearly a large number of red flags that could point to a failed study. That said, we believe there is a significant possibility the trial won't move to the second stage. But never underestimate the incompetence of management at some biotech companies to keep on chugging along. We expect a decision before the end of September.
Disclosure: I am short ONCY.