Proper Vetting of HSV-2 Vaccines and its Importance

Antibody + HSV-2 plaques 


In an earlier post on June 30, I suggested that there are two steps that could be taken to expedite the rate of HSV-2 vaccine development, and these are:

Step 1.  Appropriate Use of Small Animal Models (i.e., De-Risking HSV-2 Vaccine Development)

Step 2.  Streamlining the Path to Phase I Human Clinical Testing of HSV-2 Vaccines


In a subsequent post on July 17, I considered the latter issue of (Step 2) how we could (and should) streamline the path for HSV-2 vaccine candidates that look promising in animal models to advance to Phase I human clinical testing.   I argued that a far more sane approach to HSV-2 vaccine development would involve therapeutic vaccine testing of any reasonable HSV-2 vaccine in human patients who suffer from chronic outbreaks of HSV-2 genital herpes.  The basic rationale is that of a “Compassionate Use Trial,” where in essence the patient population in question have (1) nothing to lose from receiving an experimental HSV-2 vaccine and (2) everything to gain if the HSV-2 vaccine actually reduces their genital herpes symptoms.  Even in our current CYA-minded society, it is clear that it would be hard for a HSV-2 vaccine to be worse than the prospect of living with a chronic disease driven by wild-type HSV-1 or HSV-2 for which conventional medical drugs and therapies have failed.  If anyone questions how badly such chronic HSV-1 or HSV-2 infections may affect a person’s psyche, then I invite them to peruse the Comments posted by readers of this blog.

Today, after a 2-month hiatus from adding to the blog (i.e., my teaching responsibilities ramped up in August), I wish to finish up this discussion by considering the importance of the “Appropriate Use of Small Animal Models.”

The underlying issue of Phase I Clinical Testing boils down to “putting the ball across the goal line” and getting new HSV-2 vaccines into people where they can do some good.

The underlying issue of Appropriate Use of Small Animal Models boils down to promoting a common-sense understanding of the difference between (1) HSV-2 vaccines that should work well versus (2) sales pitches, which tout “a promising HSV-2 vaccine” but then offer little to no direct evidence to support the claim.


(and the millions to be gained regardless of whether or not it works)

To an expert who has been working with small animal models of HSV-1 and HSV-2 infection for more than 20 years, the difference between an “effective HSV-2 vaccine” and “a good sales pitch” is quite obvious.  Clearly, Herpevac and all of the related-gD subunit vaccines were the product of very good sales pitches.  Of all the HSV-2 vaccines that are currently being discussed in human clinical trials, Sanofi Pasteur’s ACAM-529 vaccine is the only one to have gone through proper pre-clinical animal testing in a manner that included publishing the data for all the world to see.  For this reason, I endorse the ACAM-529 vaccine as a HSV-2 vaccine that merits our support and consideration for testing in human clinical trials.  In contrast, the Genocea, Agenus, Corridon, and the Immunovex HSV-2 vaccines are / were all examples of HSV-2 vaccine approaches that were advanced to human clinical trials with little effort to publish animal-based research studies that are the conventional means by which new vaccine candidates are vetted (i.e., proof that the “new vaccine” is not just a sales pitch).  In particular, I would like to see the evidence for any and all of these approaches that (1) they should elicit protection against HSV-2 genital herpes that is superior to the Herpevac vaccine in animal models, or (2) they should elicit protection against exogenous HSV-2 infection that is on par with the type of protection that follows from a limited infection with wild-type HSV-2.

Now, the next question that should arise is……Why?  Why would a company not complete and publish pre-clinical animal studies (at a cost of $100,000) that could bolster their case for a HSV-2 vaccine rather than rushing straight to human clinical trials that could easily cost $20 million?

I do not have inside knowledge of what goes on inside these companies, but I suspect that money lies at the core of this disconnect in logic.  While I do not claim to be privvy to the specific details that led to the development of these vaccines, the typical sales pitch for a new HSV-2 vaccine might go something like this………..

1.  Important Scientist convinces his colleagues, “Well Herpevac failed despite what the animal models said should happen, so animal models of HSV-2 infection and challenge are useless as screening tools.”  In other words, the claim is made that you can protect mice and guinea pigs against HSV-2 by injecting them with anything including water, so why waste your time screening a new HSV-2 vaccine in animals when animal models tell you that all HSV-2 vaccines work?  Thus, let’s just jump straight into human testing of our new HSV-2 vaccine candidate and get some “good data” that will mean something.

2.  Company recruits, or is formed around, Important Scientist with new HSV-2 vaccine candidate in hand.

3.  This company recruits an elite Board of Scientific Directors, and they are sold to federal funding agencies as the “Rolex Watch” of up-and-coming HSV-2 vaccines that will deliver the long awaited miracle of a successful HSV-2 vaccine.  Yes, snob appeal is used that overtly as a common sales tactic in selling new technologies to other scientists and funding agencies.

4  A small amount of federal backing in terms of federal grant support (Important Scientist has many friends) is leveraged into the opportunity of a lifetime for biomedical investors to back the first successful HSV-2 vaccine.

5.  Company-supported HSV-2 vaccines are based on either the FDA’s favorite approach, the subunit vaccine, or some equally benign approach that poses no perceived risks so that approval for clinical testing will occur.  Biotech Company’s lawyers work on this while Important Scientist and others raise money.

6.  Only after millions of dollars have been raised from investors and governments do Phase I Clinical trials start.  Initial results of clinical trials (however cursory and speculative) are released as evidence that this new HSV-2 vaccine is really far more promising than past HSV-2 vaccines.  This data is not published but is announced as press releases, and is leveraged into (1) more biomedical investment and (2) more federal research funding for a promising new HSV-2 vaccine.

7.  Somewhere in Phase 2 or Phase 3 Clinical Trials, it turns out that the HSV-2 vaccine does not really work as well as the initial data suggested it might 5 to 10 years earlier.  However, during the years it took to “figure this out,” Biotech Company got lots of publicity and millions of dollars in federal research support as well as investments from biomedical investors whose return on investment is contingent upon the success of the vaccine.

Yes, it is the Emperor’s New Clothes.  Biotech companies want enough data to back a sexy idea, but they deliberately delay the collection of data that might cut into the viability of the story so long as federal governments or private investors are willing to keep spending money on yet more testing.  The ugly little secret is that an unscrupulous person can base a Biotech Company on an “invention of the future” that has no real hope of success so long as there exists an audience who wants to buy / support the hypothetical product.  In vaccine circles, the FDA and NIH have both promoted a pro-subunit vaccine culture and keep throwing money at HSV-2 vaccines that conform to their mental model of what an ideal HSV-2 vaccine “should look like.”  The trouble is that these “ivory tower” HSV-2 vaccine approaches keep falling flat on their face.


To an expert who has been working with small animal models of HSV-1 and HSV-2 infection for more than 20 years, the difference between an “effective HSV-2 vaccine” and “a good sales pitch” is quite obvious.

I have heard many of our current scientific leaders espouse the idea that “(1) Herpevac failed despite what the animal models said should happen, and thus (2) we should abandon animal models of HSV-2 infection as this is not a useful screening tool.”  I am here to say that this notion is patently false, and is a misnomer that needs to be critically re-evaluated.  The mindless repetition of this soundbyte at vaccine meetings needs to end.

When a trained expert runs a rigorous HSV-2 vaccine-challenge study in mice with all of the appropriate controls (including a positive control), it turns out that Herpevac is a lousy HSV-2 vaccine in mice (  Likewise, when similar tests are run using guinea pigs, it turns out that Herpevac is a lousy HSV-2 vaccine in guinea pigs (  If Herpevac (a gD-2 based subunit vaccine) is a lousy vaccine in mice and guinea pigs, then why should it be any better in humans?

Oh, that’s right…Herpevac failed in humans too (; it just took 20+ years of testing the idea over and over in humans to eventually arrive at the correct conclusion that we could have deduced in mice had the experiments been performed in a rigorous and appropriately controlled manner the first time when this line of investigation was started in the 1980s.  In fairness to investigators who came before me, I have the benefit of hindsight and hindsight is 20/20.  Nonetheless, we should not pretend that HSV-2 vaccine-challenge models cannot differentiate a lousy HSV-2 vaccine from a good HSV-2 vaccine; they most certainly can.  However, the experiments (1) need to performed rigorously with an overwhelming dose of HSV-2 challenge virus and (2) both negative and positive controls need to be included in the experiment.  None of the failed HSV-2 vaccines were ever compared to a positive control in an animal model.

The simple reality is that human beings (not nature) chose this circuitous path that we have been following for 30 years, which leads in circles around a HSV-2 vaccine but never seems to get us to the destination we keep blathering on about.  Today,  30 years later, what we have to show for our efforts is the simple realization that Herpevac and other gD-subunit vaccine-based approaches are unlikely to represent a final solution in efforts to vaccinate against HSV-2 genital herpes.

Moving forward, I would propose that we could, and should, (1) man up, (2) admit our past mistakes, and (3) move on to some new and better choices that will actually lead to the deployment of a safe and effective HSV-2 vaccine.

Vaccines are just not that complicated, and the reason that a live-attenuated HSV-2 vaccine would work better than all of the “ivory tower HSV-2 vaccines” we keep pinning our hopes on may be summarized, as follows:

(1) a live-attenuated HSV-2 vaccine can expose the vertebrate immune system to ~40,000 amino acids of HSV-2’s proteome (nearly 100% of the viral antigens are there; we don’t have to worry about choosing the “right stretch” of amino acids);

(2) all of these HSV-2 antigens are presented in their natural context to B-cells, CD4+ T-cells, and CD8+ T-cells (i.e., we don’t have to guess whether or not our antigens will be presented in a way that faithfully mimics a real HSV-2 infection); and

(3) all of these HSV-2 antigens may be presented in their proper immunological context over a period of days to weeks; in other words, the duration of time the immune cells are exposed to HSV-2 antigen should be proportional to the resulting degree of protection.


From the reader’s perspective, you may agree that the three ideas I lay out above might give you a better HSV-2 vaccine, but where is the proof?

If animal models are useful tools for screening HSV-2 vaccine potential, what piece of data can I show you that suggests that a live HSV-2 vaccine does a better job engaging the immune system of a mouse (or a guinea pig) relative to a Herpevac-like vaccine?

Vaccine-induced protection is mediated by lymphocytes, which come in three flavors…..B-cells, CD4+ T-cells, and CD8+ T-cells.  An effective HSV-2 vaccine will engage both B-cells and T-cells.  While I cannot yet offer formal evidence from the T-cell side of the equation, my lab has investigated the B-cell response to HSV-2 vaccines for several years now and the evidence is clear.  A live-attenuated HSV-2 vaccine elicits a much broader (more polyclonal) and quantitatively greater B-cell response than a Herpevac-like vaccine.  At the top of this post, I provide one type of visual evidence of this principle.

Each of the pictures contains a photograph of a single HSV-2 plaque.  This is a round cluster of many cells that are in the process of supporting HSV-2 replication in cell culture, and each of these cells is loaded with tens of thousands of HSV-2 proteins.  Surrounding each plaque is a monolayer of cells that have not yet been infected with HSV-2.   A HSV-2 plaque starts as a single, virus-infected cell and the viral infection spread like a ripple in a pond (after you throw in a pebble) with the advancing concentric circle of infection spreading outward from the first HSV-2-infected cell.  Each of the plaques in the photos above were fixed with formaldehyde and methanol at 36 hours post-infection (by which time ~200 cells were virus-infected), and now I am going to use these plaques loaded with HSV-2 proteins to ask how much anti-HSV-2 antibody is present in three groups of mice.

The first group of mice are immunologically naïve (photo on left).  They have never seen HSV-2, and so they possess no antibodies against HSV-2 and likewise have no protection against HSV-2.  The degree of redness bound to the plaque is an index that these naïve mice have no antibody that specifically binds HSV-2 protein.

The second group of mice were immunized twice with Herpevac (photo in middle).  These mice have a ton of anti-gD antibody, but gD is only 1 of 75 proteins found in HSV-2 infected cells.  Thus, you only see a light red color where anti-gD antibody has found the gD protein in a plaque of HSV-2+ cells.  Likewise, I find that this low level of antibody against total HSV-2 protein (pan-HSV-2 IgG antibody) correlates with only very limited protection against HSV-2 in Herpevac-immunized animals.

The third group of mice were immunized twice with a live-attenuated HSV-2 vaccine (photo on right).  There is a lot more red color bound to the plaque, which is an index that live HSV-2 vaccine-immunized mice have a lot more IgG antibody against total HSV-2 protein (pan-HSV-2 IgG) than mice immunized with a Herpevac-like vaccine.  As reported in the following publication,, my lab found that (1) on average a live HSV-2 vaccine elicits about 40 times more pan-HSV-2 IgG antibody than a Herpevac vaccine in mice and guinea pigs, and likewise (2) this heightened antibody / B-cell response correlates with a 40-fold increase in functional protection against HSV-2 challenge.

Again, this should not be complicated to people who study infectious disease.  Like most things in the natural world, the correct answer proves to be the obvious one:  A live HSV-2 vaccine that (1) contains more antigens (2) expressed in their natural context and (3) expressed for longer periods of time works a lot better than a monovalent gD-2 vaccine that elicits an antibody response that only poorly cross-reacts with the biologically relevant target, HSV-2 virions and HSV-2 infected cells (as illustrated in the Figure above).

The underlying issue of Appropriate Use of Small Animal Models boils down to promoting a common-sense understanding of the difference between HSV-2 vaccines that work versus everything else.  Perhaps it is time that we consider (for the first time) placing more faith in HSV-2 vaccines that have been vetted in properly controlled animal experiments as opposed to being duped by HSV-2 vaccine sales pitches that proceed straight to fundraising efforts without a shred of evidence that the putative vaccine may elicit at least 10% of the protection against HSV-2 genital herpes that is possible.

– Bill H.