Testing vaccinations for *deadly* diseases?

[TaintedAndDelish]

This question came in a discussion the other day. How do you test to see if a vaccination for a deadly or untreatable disease works? Do you ultimately have to vaccinate a population and then deliberately expose them to the disease in question? I'm sure there are plenty of intermediate steps to testing, but ultimately, unless you do something like this, how do you really know if the vaccination works or not? I would imagine that if this is how things are done, then it sort of begs the question of how one does this legally?

[Oniya]

Usually, the first experiments on a disease-killing drug are done on cultures of the disease.  Penicillin, for example, was discovered when a bit of wild mold landed in someone's Petri dish and they noticed a clear area where the bacteria culture was killed off.  After that, it usually moves on to animal testing - embryonated chicken eggs, infected mice, etc. - to make sure that the thing that kills the bacteria doesn't have any big-nasty side effects. 

Now, vaccinations are a little different in that they are usually a killed or otherwise non-infective variety of the disease that triggers the body to produce antigens  The first vaccine, for smallpox (Vaccinia), was taken from a non-lethal disease called 'cowpox' when it was noticed that milkmaids seemed to be resistant to smallpox (this is why milkmaids are stereotypically associated with having lovely skin - no pox scars).  They would contract the milder cowpox from the cows, and it would render them immune to the virulent smallpox.  The way to test if it confers immunity is to give the injection and then test the blood for antigens.

[Beguile's Mistress]

Those vaccinated in studies are tracked for years and compared to the general population for incidents of the disease.  Polio was virtually eliminated this way.  Once it was determined the vaccinated individuals had a miniscule percentage of contracting polio the vaccinations were authorized and mass vaccination programs instituted.  My grandparents were some of the first to be vaccinated against polio.

[TaintedAndDelish]

Well, the danger that I see here is not in catching something from the vaccine but in the vaccine not working. Say you have a vaccine for a deadly disease that seems to work in lab experiments, but you don't yet know for a fact if it will work on a human. Do you approach an uninfected person and say, "I want to test this on you. It involves vaccinating you and then exposing you to an incurable disease. It might not work, so you'll have to sign a legal disclaimer just in case."  In this crude example, only ONE test subject is being used... I would think that you would want to test a group and not an individual.

[Oniya]

Nope - basically, you gather a bunch of volunteers and say 'I'm looking for people to volunteer to get vaccinated against X.  We've tested it on _____ and had these results, and now we've gotten clearance to test on humans.'  You give them the vaccination, and they go about their lives with the added bit of coming in every so often to have a physical.  Their normal doctors are informed and the volunteers submit a form that gives the researchers medical access, so that if they do get the disease, the researchers find out about it.

The researchers don't take the volunteers and deliberately infect them - unless you're looking at the syphilis experiments of 1932, where people with the disease were deliberately not treated (a minor difference on the ethical scale - as in still WRONG - but even then, still not deliberately infecting someone).

[TaintedAndDelish]

Yeah, that's kinda what I meant. You can infer that the vaccination is effective if less than the expected percentage of people don't get sick, but the possibility that those people with the vaccination never got infected still exists. I suppose if you test them afterwards to see if any of them managed to fight off the disease, you might get proof that it worked that way?

[Oniya]

Well, with the polio example, once they determined that there were a) no ill effects on the test subjects and b) a lower incidence of the disease than would be expected (this, incidentally, is why you need a test group of a certain minimum size), they started mass vaccinations.  This further confirmed the 'no ill effects' and the 'lower incidence' factors.

By the way, if you look at the labels of germicidal products (like cleaning supplies) with a virucidal claim, you'll still see it listed as to whether they are effective against Poliovirus or not.  Every so often, each cleaning supply formula has to be re-tested to make sure it still kills all the things it says it kills.

[Caela]

Yeah, that's kinda what I meant. You can infer that the vaccination is effective if less than the expected percentage of people don't get sick, but the possibility that those people with the vaccination never got infected still exists. I suppose if you test them afterwards to see if any of them managed to fight off the disease, you might get proof that it worked that way?

I can't confirm but I would think it is also likely they after the test subjects are vaccinated that the scientists take blood samples to test for antigens, and you could expose the patients blood to the disease and see what sort of reaction you get in a lab setting.

[fireisntaghost]

Hey, kinda glad I stumbled upon this I'm actually doing a PhD in immunology at the moment and my research has implications for vaccine development so I tend to keep a close eye on it at the moment this is a very general pathway for how a vaccine is developed and produced.

For simplicity I'm going to call the disease were looking at Disease A.

We decide to make a vaccine for Disease A. First thing we do is study Disease A to figure out what it does, that the cycle of an infection is like. At the same time we find people who have had Disease A and take samples from them. Once you have encountered a disease you will develop something called Immunological Memory. Basically these are cells that have lived through the infection and they "remember" the infection so if you ever encounter it again they react straight away and can clear an infection before you get sick. These are actually the kinda cells that I study.

We test the peoples samples to find what parts of Disease A their cells recognise. These cells can only recognise specific molecules on the Disease A so we figure out what part of the disease the body most easily recognises and then run with that.

We take those parts of the Disease A and smash them up basically, mix then with things called Adjuvents and this is basically cellular debris, they are very important they help kick start and immune response by prompting release of signalling molecules in the body that activate the immune system and tell the cells to react.

We then take these smashed up bits of Disease A and stick them in a test tube with some immune cells from normally a mouse, mouse models are used a lot in immunology because genetically they share around 99% of the same genes as our immune system. If we get good results and the cells are showing hallmarks of developing memory, note these aren't nerve cells like your brain, memory cells in immunology mean something else entirely. Its just what we use to describe cells that have seen the infection before, we then move on to mouse models.

Once we get past that we then inject mice with the vaccine, we measure if they develop memory by taking blood samples and testing the immune responses. Note mice are NOT injected with the virulent disease, we categorically CAN NOT DO THIS UNDER UN LAWS they are injected with a form of the disease which can not cause harmful effects - they normally knock out the gene that cause the disease to be harmful. We asses them for development of immunological memory and if we have good results we move on to clinical trails.

Note human clinical trials can take 10 years to complete! They are long and arduous process.

Phase 0 clinical trails involve looking at the pharmokinetics and pharmacology in the body - so how the vaccine is processed by the body. These are small groups given tiny tiny tiny dosages to make sure that it doesn't kill people.

Phase 1 clinical trials assess safety. Small groups of people <100 are given the vaccine to test for safety, establish side effects etc.

Phase 2 clinical trials involves establishing the efficacy of the vaccine - so does it work? A few thousand normally in the area where the disease if prevalent. The subjects will be tested normally for antibodies and other immunological memory markers.

Phase 3 clinical trials involve the final establishment of safety and efficacy, the trials are conducted again on a larger scale by other independent labs not affiliated with the labs that did the phase 2 trials. These groups are normally large 3K+

Vaccines that make it through phase 3 generally go on to be used as the general public, their safety, efficacy and benefits, optimal use are looked at through sales in the general population.

Generally patients are not exposed to the disease in the lab, we don't infect them with it to test the vaccine. Note though it depends, some diseases are very rare it's not worth creating vaccines as for them to be truly effective we need to vaccinate whole populations. Once we vaccinate a huge number of people in a population we break up the train of transmission of the disease and we reduce its ability to spread. So a few people who are not immunised or can not be (some have very weak immune systems and can not be vaccinated or they are allergic to components of the vaccine) are also protected though everyone else being vaccinated - we term that Herd immunity.

okay so you know...I'm so sorry thats so long but I hope its helpful.

[mia h]

Well, the danger that I see here is not in catching something from the vaccine but in the vaccine not working. Say you have a vaccine for a deadly disease that seems to work in lab experiments, but you don't yet know for a fact if it will work on a human. Do you approach an uninfected person and say, "I want to test this on you. It involves vaccinating you and then exposing you to an incurable disease. It might not work, so you'll have to sign a legal disclaimer just in case."  In this crude example, only ONE test subject is being used... I would think that you would want to test a group and not an individual.

Deliberately exposing anyone to a disease is illegal, I can't remember if breaches the Genneva Convention or if it's a UN either way you're not allowed to do it, with one tiny exception.

Doctors are allowed to experiment on themselves like the 2005 Nobel prize winner who drank a petri dish full of bacteria to prove them were the cause of stomach ulcers http://en.wikipedia.org/wiki/Barry_Marshall

[TaintedAndDelish]

Deliberately exposing anyone to a disease is illegal, I can't remember if breaches the Genneva Convention or if it's a UN either way you're not allowed to do it, with one tiny exception.

I suppose what I was getting at is whether or not vaccines for deadly diseases actually get tested covertly - say in some small country where it might be easier to cover things up... or given the US government's stellar track record for lawfulness, perhaps on military personnel ( who agree to volunteer for testing ) or prisoners. I think this wild suspicion of mine was a bit of a stretch. Someone shed some light on the testing methodology one or two posts ago. :)

[Kichita]

Thanks you fireisntaghost!  I am in public health and immunize people regularly.  Your description was well written and precise and will greatly help me in explaining the process to others!

[DarklingAlice]

fireisntaghost makes some great points but does get a few things wrong.

1) Law actually requires that (at least in the US) two entirely separate animal models are used before moving on to human trials, just to insure that one wasn't a fluke. Normally this is first mice and then non-human primates, though depending on the disease another model may be more appropriate (e.g. ferrets tend to be the go to model for testing influenza vaccines due to a few special features of the respiratory system).

2) We can and do infect animal models with dangerous forms of disease. My university runs a primate center where tuberculosis treatments are tested in monkey models under BSL-3 conditions. Whether or not you take that step in testing is a matter of balancing benefit, animal welfare, and safety. It generally comes down to the question: 'are their suitable alternatives?'. That question is addressed by a board known as an Institutional Animal Care and Use Committee that any institute that performs animal research is required to have by federal law.