Undervaluation Of Vical's Full Pipeline Allows A Cheap Bet On Cancer Trial

| About: Vical Incorporated (VICL)

By Dee Kotak, MD

Vical Incorporated (NASDAQ:VICL) expects to announce top-line data mid-2013 from the pivotal Phase III trial (NCT00395070) of Allovectin in metastatic melanoma, some two years later than first predicted. The uncertainty created by the delay, complexity of melanoma immunotherapy trials, and focus on the cancer trial has weighed heavily on Vical's share price and caused investors to overlook the advancement and value of the rest of the Vical pipeline. Institutions, with their longer-term perspective have not been discouraged and capitalized on the opportunity to be positioned (institutional ownership of 44%) for the advancement of the Vical infectious disease vaccine pipeline, notwithstanding the awaited melanoma data. Not long ago, melanoma was considered a graveyard indication, but the approvals of Yervoy and Zelboraf in 2011 and recent encouraging melanoma vaccine data from Amgen (NASDAQ:AMGN) and OncoSec Medical (NASDAQ:ONCS) among others has set the stage for potentially remarkable results. Understanding immunotherapy, the Vical melanoma data and Phase III trial design are challenging but worthwhile, as the market has significantly underestimated the probability of success, creating a near-term, high-return opportunity coupled with a pipeline safety net. Vical has a diversified development program with substantial commercial opportunity, validating partnerships, solid cash position, and no debt, allowing longer term investors the luxury of benefiting from the near term oncology catalyst with a pipeline backdrop. Of particular note is that the Vical vaccines are for infectious diseases, hence, the development timelines are very much shorter than for the melanoma immunotherapy.

Core DNA Delivery Technology

The Vical approach involves designing and constructing closed loops of DNA called plasmids, or pDNAs. These pDNAs contain a DNA segment encoding the protein of interest, as well as short segments of DNA that control protein expression. Plasmids can be manufactured quickly and cheaply, and Vical has improved efficiency of gene expression and immunogenicity. The benefits of the DNA delivery technologies include broad applicability, convenience, safety and repeat administration.

Vical's Pipeline

The two major value drivers of Vical's pipeline are the Cytomegalovirus (CMV) and the Herpes Simplex Virus-2 vaccine programs.

CMV can cause serious complications in two distinct patient populations: immunocompromised transplant patients and children born to women initially infected during pregnancy. Vical is developing two CMV vaccines: TransVax and CyMVectin. TransVax is designed to prevent CMV reactivation or infection in transplant recipients. In 2011, Vical licensed TransVax to Astellas for $130m total upfront and milestone payments, double-digit royalties on net sales, and the option to co-promote in the U.S.; Astellas bears all developmental costs. A Phase III registration trial in hematopoietic stem cell transplant recipients and a Phase II trial in solid organ transplant recipients are scheduled to start in 1H 2013. Although a small patient population, the market opportunity is about $500m. CyMVectin serves a much larger patient population by preventing primary CMV infection during pregnancy and thereby precluding maternal-fetal CMV transmission, congenital CMV infection, and related birth defects. Congenital CMV represents a major commercial opportunity and could potentially lead to widespread vaccination in a similar patient population to Gardasil, representing a >$2B market opportunity. CyMVectin is ready to enter the clinic, and on completion of Phase I testing will be a very attractive partnering opportunity.

HSV-2 is the leading cause of genital herpes worldwide, with 40-60M in the U.S. infected. Twenty percent are symptomatic with multiple ulcer recurrences. After completing preclinical safety/toxicology and biodistribution studies, a Phase 1/2 therapeutic vaccine trial is expected to begin in 2H13 with safety, immunogenicity and clinical endpoints designed to establish clear proof of concept and support potential future partnering discussions and development progress.

Additional value is added to the Vical pipeline from their proprietary Vaxfectin adjuvant and a number of validating partnerships. The Vical pipeline, excluding Allovectin and cash (~$75M at March 31), could support a valuation in the $200-250M range, thus we believe that Vical's pipeline provides a real backdrop to the Allovectin trial (details below).

Stage III/IV melanoma

The American Cancer Society estimated that approximately 77,000 new diagnoses of, and approximately 9,500 deaths from, melanoma would occur in 2013 in the United States. The lifetime risk of getting melanoma has been growing and is now approximately 2% for Caucasians in the United States. It is one of the more common cancers in young adults. Metastatic melanoma is a fatal disease with a rapid systemic dissemination; the 5-year survival rate is less than 15%. While the minority of patients, which constitute 4% of newly diagnosed melanoma patients, present with distant metastasis at initial diagnosis, the majority who present with early stage initially eventually develop metastatic disease as a consequence of disease progression. Approximately one-third of all melanoma patients will experience disease recurrence. In patients with advanced metastatic melanoma, median survival typically ranges from six to ten months.

Primary treatment of melanoma includes surgical excision and usually involves clinical staging. Treatment is based on the extent of the disease and may include single agent or combination chemotherapy and radiation therapy. FDA approved biologic agents include, but are not limited to, kinase inhibitors, interleukin-2 (IL-2) and interferon alpha. Two new drugs for metastatic melanoma were approved in 2011, both on the basis of increased survival. Yervoy (in March), a monoclonal antibody marketed by Bristol-Myers Squibb (NYSE:BMY), increases the effectiveness of T-cells that can seek out and destroy melanoma cells. Zelboraf (in August), a BRAF inhibitor marketed by Roche (OTCQX:RHHBY) and Daiichi Sankyo, interrupts a key process in melanoma growth in patients with a particular melanoma mutation (<50% of patient population). Both drugs are associated with significant side effects, and neither is considered a cure for melanoma.

Currently, there are no melanoma vaccines approved by the FDA. Clinical trials investigating the safety and effectiveness of different types of melanoma vaccines (approximately 100) are ongoing.

Cancer vaccines are considered biological response modifiers. The biological response modifiers stimulate or restore the ability of the individual's immune system to fight infections and disease. The broad categories for cancer vaccines are:

  • Preventive (or prophylactic) vaccines, which are intended to prevent cancer from developing in healthy people
  • Treatment (or therapeutic) vaccines, which are intended to treat existing cancers by strengthening the body's natural defenses against cancer.

Cancer treatment vaccines are designed to treat cancers that have already occurred. Therapeutic cancer vaccines try to refresh the immune system's memory to recognize cancer cells for removal by the body. The cancer vaccine may contain inactivated cancer cells or viruses that express tumor antigens.

Melanoma vaccines attempt to stimulate and enhance the individual's own immune system to respond to tumor related antigens (unique proteins or protein bits that sit on the surface of cancer cells and can trigger some immune response), or any antigens that are over expressed by cancer cells. They function by creating antibodies or activated T lymphocytes with intended destruction of tumor cells and regression of the melanoma. The intent of cancer vaccine therapy is to delay or stop cancer cell growth; cause tumor shrinkage; prevent cancer from coming back; or eliminate cancer cells that are not killed by other forms of treatment, such as surgery, radiation therapy, or chemotherapy. Melanoma vaccines can be generally categorized or prepared in a number of different ways - purified antigen vaccines, cell lysate vaccines, whole cell vaccines, autologous whole-cell vaccines, heat-shock protein-peptide complexes, shed antigen vaccines, dendritic cell vaccines, genetically modified tumor vaccines and anti-idiotype vaccines.

Allovectin's Novel Immunotherapeutic Approach in Metastatic Melanoma

Allovectin is a plasmid/lipid formulation containing the DNA sequences encoding HLA-B7 and ß2 microglobulin, which together form a MHC Class I complex that is important for antigen presentation. Injection of Allovectin directly into tumor lesions directs a local, regional and systemic immune response against metastatic tumors through three interrelated mechanisms of action:

  • Allogeneic response against HLA-B7 in patients who are HLA-B7 negative. Intratumoral injection of Allovectin results in the expression of the B7 MHC class I protein on 57-79% of tumor biopsies. The tumor cells expressing this foreign antigen are rejected. Additionally, there is an increase in frequency of HLA-B7reactive cytotoxic T lymphocytes and generation of specific antitumor responses.
  • Upregulation and/or restoration of normal expression of tumor antigens. Loss of ß2-microglobulin is a known mechanism of immune escape in melanoma. Increasing intracellular production of ß2-microglobulin restores proper tumor antigen presentation, initiating a systemic immune response against other tumor cells.
  • Induction of a pro-inflammatory response after intratumoral injection with the pDNA/lipid complex, contributing to systemic immune surveillance.

The goal of all three of these mechanisms is to initially cause recognition of the tumor at the local site (break tolerance) to allow a then-sensitized (boosted) immune response to recognize un-injected tumors at distant metastatic sites. Allovectin can be used safely in all patients whether HLA-B7 positive or negative. The Allovectin mechanisms of action overcome many of the potential shortcomings of therapeutic cancer vaccines. Most cancer vaccines target one or at most a few tumor-associated antigens due to practical limitations of manufacturing complexity and cost. Stimuvax (not a plasmid vaccine) targeted only a single antigen and this study failed to reach its primary endpoint last year. Other vaccines like Immunocellular's (NYSEMKT:IMUC) ICT-107 cell based vaccine can target a broader range of tumor-associated antigens, but the cell-based approach is even more complicated and expensive. The genius of the Vical approach is that by injecting the melanoma lesion and enabling the immune system to recognize it as foreign, the immune system can target many of the tumor-associated antigens expressed. So although the plasmid codes for two proteins that are not tumor-associated antigens, Allovectin induces an immune response to multiple melanoma associated antigens. This approach sidesteps the issues of choosing which tumor-associated antigens to use, what combination, and how to induce durable immunity to any such combination. Allovectin is not a cancer vaccine in the classical sense, but it induces a similar immune response indirectly by the mechanisms outlined above. Furthermore, the plasmid approach allows repeated administration without inducing an immune response against the vaccine, decreasing its effectiveness.

Phase II Data is Much Better than Top-Line Data Suggests

The Allovectin high-dose Phase II trial was an open-label, multi-center, single-arm study in Stage III (52%) or IV (48%) metastatic melanoma. Patients had no brain or liver metastases and normal lactate dehydrogenase (LDH) levels (elevated is a poor prognostic indicator). The data showed that the trial had a total of 15 responders among the 127 patients receiving the high dose, an 11.8 % objective response rate (RECIST-Response Evaluation in Solid Tumors), with four of the patients having complete responses and 11 having partial responses (confirmed and reviewed by the sponsor and not adjudicated by an independent review panel). The Kaplan-Meier estimated median duration of response was 13.8 months, and all responses were durable with a range of 6 months to 66 months. Over 50% of the responders were still alive more than 6 years after the trial started. The updated Kaplan-Meier median survival was 18.8 months. The safety profile was excellent, with no reported Grade 3 or Grade 4 adverse events associated with Allovectin. At the time of the Phase II study the median overall survival for the trial population would not have been expected to be more than 6 to 11 months.

Beneath the surface, the Phase II data is far more impressive than top-line data suggests. The key to the Phase II data is to understand the objective response rate of 11.8% on an intent-to-treat basis. This response rate is clinically meaningful on its own, but when analyzed in the context of the Phase II study and Phase III trial design it points to potentially stellar results:

  • Greater than 60% of subjects did not complete more than one treatment cycle, yet there was still an objective response rate of 11.8%.
  • 10 of the 15 responders (66.7%) required at least two cycles of Allovectin before responding. The median time-to-response was two cycles (18 weeks), with a range of one to seven cycles.
  • An intact immune system is required to gain the greatest benefit from immunotherapy; chemotherapy impairs the immune system. When looking at the Phase II survival data, the chemo-naive patients treated in a single lesion (the Phase III target population) had a 22.5 month median overall survival (control for Phase III expected to be 11-13 months median overall survival).
  • The overwhelming reason for trial dropout was "disease progression" according to the traditional Response Evaluation in Solid Tumors (RECIST) criteria that was designed for chemotherapy trials and is less valid and applicable to immunotherapy.
  • Remarkably, especially compared to the new melanoma immunotherapies, no patient withdrew due to serious adverse events associated with Allovectin.

Why was the dropout rate 60%?

Immunotherapeutic agents produce antitumor effects by inducing cancer-specific immune responses or by modifying native immune processes. Resulting clinical response patterns extend beyond those of cytotoxic agents and can manifest after an initial increase in tumor burden or the appearance of new lesions (progressive disease). Response Evaluation Criteria in Solid Tumors or WHO criteria, designed to detect early effects of cytotoxic agents, may not provide a complete assessment of immunotherapeutic agents.

  • Patients can have transient worsening of disease, manifested either by disease progression of known lesions or the appearance of new lesions, before the disease stabilizes or tumor regresses.
  • The immune response induced by Allovectin can make the melanoma lesions appear to enlarge and progress due to the immune response causing swelling.
  • Immunotherapeutic effects are not seen early like traditional chemotherapy, and continued disease progression can be observed well after the induction period.
  • Patients progressed and dropped out even though they had not had adequate time to benefit from the treatment as the peak effects of immunotherapy occur after six months.

Response rate in context

The goal posts of immunotherapy trials changed in 2011 with Yervoy and Zelboraf improving survival, but response rate is still a valid marker of efficacy as responders live longer.

  • Amgen recently announced positive Phase III trial results with Talimogene Laherparepvec having a primary endpoint of a durable response rate at six months of 16%; this was sufficient news to add a couple of billion to Amgen's market cap. OncoSec Medical's Immunopulse therapy had a six month durable response rate of 38% in a small Phase II study.
  • In chemo-naive patients Yervoy and Decarbazine had approximately a 15% response rate, and 11.2- versus 9.1- month median overall survival, a two month improvement; and in chemo-experienced patients, Yervoy alone has an 11% response rate and a 10- versus 6.4- month overall survival. For Zelboraf monotherapy in chemo-naive patients, the response rate was much higher, at 48%. But median overall survival was 13.6 versus 9.7 months for the control, an increase of only 4 months. The studies are not directly comparable to the Allovectin patient population as this excluded liver and brain metastases.

If the response rate for Allovectin of 11.8% were extrapolated to a trial population without a 60% dropout rate, then it would be at least 22-24%.

Although it is too early for the overall survival data from the Amgen and the OncoSec studies, what is clear from the Yervoy and Zelboraf data is that response rates of 15-48% are not improving overall survival by more than 2-4 months, albeit in a sicker population. What is particularly impressive from the Allovectin study is the profound survival advantage in responders, with greater than 50% of those responders alive at six years after the trial started.

Phase III trial

Enrollment in the Phase III trial began in 2007 and has involved approximately 90 clinical sites (45% US, 45% EU, 5% Brazil and 5% Israel). Patients were previously treated with surgery, adjuvant therapy, and/or biotherapy, but could not have been previously treated with chemotherapy. The patients were randomized on a 2:1 basis with 260 patients treated with Allovectin (maximal treatment period 2 years) and 130 treated with their physician's choice of either of two chemotherapy agents, Dacarbazine or Temozolomide. The primary endpoint is overall response rate that compares the two trial arms for objective responses that are ongoing or commence at 24 weeks or more after randomization. The study will also evaluate safety and tolerability, as well as survival as a secondary endpoint. The study was initiated in 2007 and the final patients were treated in February 2012. The patients in the control arm do not receive an intralesional placebo injection, as this was considered to be too burdensome by the FDA.

Vical was expecting to announce the trial results by mid-2011, but this has now been pushed back to mid 2013. The trial is event driven (deaths) and Vical has not revealed the number of pre-specified events necessary to unblind the trial. The inescapable conclusion is that either the control group is living much longer than expected; Allovectin is far more efficacious than the 11.8% overall response rate and 18 month median overall survival achieved in the Phase II study; or both. The market response to repeated delays in announcement of the expected trial data has been to move to the sidelines, as well as a loss of confidence in a positive trial outcome. For biotech sophisticates, there is sufficient data available to build models to predict the outcome under various assumptions.

Why the Phase III trial is likely to be a success

A number of the Phase III trial characteristics support the most likely explanation: that Allovectin works and that the overall response rate will be very much higher than 11.8%:

  • Patients are healthy enough to receive a minimum of two treatment cycles
    • Good functional status and no visceral metastases except lung
    • 10 of the 15 responders from the high dose Phase II study required a minimum of two treatment cycles
    • Survival in patients who only received one cycle or less in the Phase II study had a median overall survival of 12.2 months (N=78), evidence of a threshold effect
  • Patients have functional immune systems
    • Only chemo-naive patients, median overall survival was greater in the Phase II chemo-naive patients (22.5 versus 18.8 months in the entire intent to treat population; no control arm but 22.5 months would represent a doubling of survival compared with historical results for chemotherapy)
    • Normal levels of lactate dehydrogenase (not the x1.5-2 upper limit of normal used in some trials)
  • Patients on study for more than 2 treatment cycles
    • FDA agreed to modified RECIST criteria for "progressive disease" up to week 16, which should drastically reduce dropout due to "progression"
  • Designed to capture immunotherapy advantages
    • Primary endpoint-objective response rate at 24 weeks or more after randomization allows time for immunotherapy to work
      • 90% powered to detect a greater than 10% difference in response rate
    • Secondary endpoint - overall survival
      • 90% powered to detect a survival difference of 18 vs 11 months

Is there independent evidence to validate the mechanism of action of Allovectin?

Most of the melanoma therapies in development utilize very different mechanisms of action compared to the Allovectin approach of primarily inducing the immune system to attack the melanoma as if it were a foreign tissue, and releasing (exposing) multiple melanoma antigens to which the immune system then becomes sensitized and can thus target other melanoma lesions. In March 2013 Amgen announced top line data from its Phase III study in a randomized, open-label, Phase III trial to evaluate the safety and efficacy of Talimogene Laherparepvec (a cancer-killing virus that also carries a gene encoding GM-CSF, an immune system stimulant) compared to a control therapy with GM-CSF in over 400 patients with unresected stage IIIB, IIIC or IV melanoma. Talimogene Laherparepvec is injected directly into tumor tissue and then replicates until the membrane of the cancer cells rupture, thereby destroying the cells, in a process known as cell lysis. The virus that was contained in these cells is then released locally in the tumor tissue along with GM-CSF. This is intended to lead to the activation of a systemic immune response to kill tumor cells throughout the body. The study met its primary endpoint of durable response rate (DRR), defined as the rate of complete or partial response lasting continuously for at least six months. A statistically significant difference was observed in DRR: 16% in the Talimogene Laherparepvec arm versus two percent in the GM-CSF arm. The analysis of overall survival (OS), a key secondary endpoint of the study, is event-driven. A pre-planned interim analysis conducted with the analysis of DRR has shown an OS trend in favor of Talimogene Laherparepvec as compared to GM-CSF. The OS data is expected to mature in late 2013 and further data will be presented at ASCO 2013. This study was in a sicker population than the Allovectin trial because patients were allowed visceral metastases, so it is not directly comparable. However, the trial did meet its primary endpoint and it is not unreasonable to expect that Allovectin could have a more profound effect in a more healthy population. The focus will now be on the secondary endpoint of overall survival.

The Phase III trial has been designed for Allovectin success by having a patient population expected to live long enough with an intact immune system to receive sufficient doses to benefit from the immunotherapy as well as having the modified RECIST criteria that should allow more patients to stay on treatment longer. The response rate of 11.8% seen in the Phase II study is likely to be very significantly higher in the Phase III study because the dropout rate will be very much lower for the reasons set out above (patients were not dropping out for adverse events). I think it is not unreasonable to estimate the response rate could be in the 20-25% range that will compare favorably with Talimogene Laherparepvec.

Why are the patients living longer than predicted?

There is clearly a mismatch in the outcome and trial assumptions as evidenced by the two-year delay in expected data. I discussed the reasons for this, amongst other issues, with the Vical CEO, Mr. Samant, and outline some of the arguments and evidence below:

  • Assumption options
    • 11 month for Dacarbazine (DTIC)/Temozolamide (TMZ) control is incorrect, or
    • 18 month for Allovectin treatment is incorrect, or
    • both assumptions incorrect
  • How best to refine control arm survival assumption?
    • Conduct a review of historical survival data for DTIC/TMZ
    • Calculate survival for patients in the Phase II study who received minimal Allovectin therapy (≤1 cycle)
    • Examine survival data from previous Phase III trial of DTIC ± Allovectin
  • Historical survival data of DTIC and TMZ in metastatic melanoma
    • No published study with exact Phase III population
    • DTIC compliance has improved-positive impact on survival?
    • Prior to 2006, no data on LDH impact on survival
    • DTIC/TMZ (<2009) median overall survival (in months) in metastatic melanoma
      • 10 studies: range 5.6-11.5 months
    • DTIC/TMZ (>2009) median overall survival (in months) in metastatic melanoma
      • 3 key studies: 8-9 (Zelboraf control), 9.1 (Yervoy control) and 13.1 months in the failed Genta study appears to be an outlier, likely due to below-normal LDH requirement
    • Based on review of historical data
      • Not substantially greater than the assumed 11 months
  • Review of high dose Allovectin Phase II data
    • Open label, multi-center, single-arm study (N=127)
      • Stage III (52%) and VI (48%) metastatic melanoma
      • No brain or liver mets and normal LDH
      • 78 (61%) completed 1 cycle or less
      • 12.2 months median overall survival (N=78)
    • Based on Phase II subset analysis for patients who received ≤ 2 cycle of treatment, median survival +12 months and DTIC/TMZ median survival extrapolated at ~12 months.
    • Published stage III and IV melanoma survival data from the American Joint Committee on Cancer Staging Database cannot be extrapolated to the present study and is misleading because:
      • They are survival times for people with an initial diagnosis at each stage and do not include patients who had earlier disease that eventually progressed
      • The survival times are overall survival curves regardless of the treatment provided
      • The time from diagnosis can be years before a patient enters a clinical trial
  • Review of low dose Allovectin Phase III data
    • Controlled, multi-center, randomized trial (N=202)
      • DTIC + Allovectin (N=98) vs DTIC (N=104)
      • 98 out of 202 patients with normal LDH, no liver mets, ~88% Stage IV
        • 11.9 months median overall survival
        • DTIC only (N=50, ~88% Stage IV) 12.2 months survival
    • Based on Phase III subset analysis, DTIC/TMZ median survival estimated at ~12 months
  • Could the treatment arm be living longer (18 month survival original assumption)?
    • 22.5 months possible based on healthier Phase III population
    • Modified RECIST criteria-patients can get more Allovectin
    • Impact of new therapies-potential synergy
    • Conclude that number could be greater than 22.5 months

The Vical assumption that survival in the control arm would be no more than 13 months is eminently reasonable given the current state of the clinical evidence, but two further points warrant elaboration:

  • Is the trial population as a whole living longer because of the beneficial effects of Yervoy and Zelboraf?
    • Yervoy and Zelboraf were only approved in 2011, March and August respectively in the U.S., later in Europe. The last Allovectin trial patients were treated in February 2012, so given the initial lag in adoption of the new therapies, only a small number of patients could have received these new treatments once they had completed the maximal two year Allovectin treatment period.
    • The improvement in median overall survival by Yervoy was 2-4 months and approximately 4 months for Zelboraf. These improvements in overall survival are not sufficient to throw the trial completion estimate off by two years.
    • Recent animal work has suggested that there is a synergistic effect between Allovectin and the mouse equivalent of Yervoy, and if this holds in humans, then any benefit to the trial population should affect the treatment arm at least as much as the control.
    • If treatment arm patients on average live longer than control arm patients (an underlying assumption in the trial), more Allovectin patients than chemotherapy patients should have been alive long enough to access Yervoy and Zelboraf after their respective approvals.
    • In November 2012, Vical released the results of a comprehensive sweep conducted by the company in September 2012 of all active clinical sites in its Phase III trial to eliminate any time lag in death event reporting. The sweep confirmed that the target number of events had not been reached. It also confirmed the steady progress towards that goal. Based on the moving average monthly event rate, the company has revised its projection for reaching the target number of death events to mid-2013. In other words, before there could have been the slightest impact of the new immunotherapies, the Phase III study death rate was already much lower than predicted. Vical is blinded to which groups the deaths are occurring in, as are the assessors.
  • Could the Phase III trial population be living longer because they are healthier than the Phase II trial population?
    • From the patient demographics, it can be seen that, in fact, the Phase III trial population are sicker than the Phase II, with 50% of the population being over 64 years old and 63% stage IV melanoma who not be expected to live as long.
Phase II Phase III
Number of patients 127 390
Stage III/IV percentages 52/48 37/63
ECOG 0/1 percentages 77/23 80/20
Median Age 60 64
  • Patients were sicker in the Yervoy and Zelboraf studies because brain and liver metastases were included in the studies, but these studies post-date the start of the Allovectin Phase III study and were not the basis for the Vical survival model assumptions.

Phase III timeline

Vical completed a secondary sweep of active clinical sites at the end of March this year, and an update is expected at the next earnings call in May; independent assessment of response rates is ongoing. I expect that this may be the event that rekindles investor interest and initiates a run-up before the eventual data release. The size of the study is such that to maximize the statistical power for evaluation of the survival endpoint (treatment impact of immunotherapy may occur late, long tail effect may cause late separation of the Kaplan-Meier curves) there has been no interim look for efficacy, only several for safety. I also suspect that the event rate required to trigger unblinding is also high for the same reason. From the available data, I posit that the profound efficacy of Allovectin in an increased number of responders and a high pre-specified event rate (number of deaths) has led to trial unblinding being prolonged by two years.

What is trial success?

Since 2007 the landscape of melanoma trials has changed, with Yervoy and Zelboraf demonstrating improved survival. The primary endpoint under Vical's Special Protocol Assessment is response rate, with overall survival being a secondary endpoint. This does not necessarily bind the FDA, and so it is probably fair to say that Vical must demonstrate improved overall survival to be absolutely sure of approval. Amgen is in the same position with Talimogene Laherparepvec. The next question is, then, by how much will overall survival need to be improved? We have the range of 2-4 months for Yervoy and Zelboraf (sicker trial populations). Yervoy and Zelboraf are not without significant serious side effects, and Zelboraf is only indicated in the <50% of patients with the BRAF mutation. Allovectin can be used in the outpatient setting and in over 900 patients has been generally well tolerated. I see no issue if survival is improved by 2-3 months, except for the fact that the trial may not be sufficiently powered for statistical significance. It would be not be unreasonable for the FDA to approve Allovectin based on a statistically significant response rate and positive trend towards overall survival, but this will be a matter for discussion. The regulatory risk due to overall survival becoming an attainable endpoint in melanoma studies could be viewed by some as problematic, but I believe the FDA would be cognizant of the limited options available to melanoma patients, the very significant side-effects from current therapies, and that Allovectin would be an option for those patients who did not want, or are unsuitable for, current toxic immunotherapies and chemotherapy.

Allovectin is licensed to AnGes MG in the Asian market, and owned by Vical in the U.S., EU and rest of world. Worldwide sales could reach $1B on this high-margin product. With broadening of the indication to use in earlier stages as well as to those with good health status in more advanced melanoma, and potential in head and neck cancer and other solid tumors, Allovectin has a peak sales potential of roughly $2 billion. Vical had $86m in cash at the end of 2012 and projected cash burn of $18-20m for the first half of 2013. Depending on the level of market and media hype, oncology companies approaching pivotal phase III data with a solid phase II database reach market capitalizations in the $300-500m range and do not often have such a healthy cash position. Vical's current market capitalization is $318m, so excluding the $66m cash available for 2H13 gives a valuation of $242m for Allovectin and the rest of the pipeline. No doubt the long awaited announcement from Vical confirming a more certain trial readout time will draw more investors into the stock and cause a run into the data announcement.

The market is discounting Vical on the basis of underestimating the probability of Allovectin success and is completely overlooking a broad pipeline with shorter developmental timelines. These oversights have created what we consider a cheap call option on Allovectin data given the pipeline strength. In the context of the overall success rate of Phase III oncology studies, the existing data supporting a positive trial outcome for Allovectin is alluring.

Disclosure: I am long VICL. 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.

Additional disclosure: PropThink is a team of editors, analysts, and writers. This article was written by Dee Kotak, MD. We did not receive compensation for this article, and we have no business relationship with any company whose stock is mentioned in this article. Use of PropThink’s research is at your own risk. You should do your own research and due diligence before making any investment decision with respect to securities covered herein. You should assume that as of the publication date of any report or letter, PropThink, LLC and persons or entities with whom it has relationships (collectively referred to as "PropThink") has a position in all stocks (and/or options of the stock) covered herein that is consistent with the position set forth in our research report. Following publication of any report or letter, PropThink intends to continue transacting in the securities covered herein, and we may be long, short, or neutral at any time hereafter regardless of our initial recommendation. To the best of our knowledge and belief, all information contained herein is accurate and reliable, and has been obtained from public sources we believe to be accurate and reliable, and not from company insiders or persons who have a relationship with company insiders. Our full disclaimer is available at www.propthink.com/disclaimer.