Before precising my question, here are some facts that I presume to be true:

  1. A vaccine works by injecting the antigens of a virus into the body to train the immune system to recognize the virus and to be prepared to fight it if it shows up.

  2. The ELISA test indirectly recognize the presence of a virus into a blood sample by testing the presence of the antigens of the virus and the antibodies that fight against it.

  3. One of the reasons there are still no vaccine against HIV is because this virus has a very high mutation rate.

  4. The ELISA test is used to detect the presence of HIV.

I have the feeling these facts contradict themselves and I would like to know where I am misleading.

Indeed, if we have an ELISA test for HIV, we should be able to create a vaccine because we know the antigens of the virus and we can inject a harmless version of them into our body, right? Apparently, the answer to this question is no because the HIV virus has a high mutation rate and the antibodies produced thanks to the vaccine won't be able to fight every mutation of the virus (that's the answer that gave me my former biology teacher in high school). But if the HIV has a high mutation rate, how can we be able to produce an ELISA test for it?


2 Answers 2


I found the question quite interesting and will to provide a concept and a speculation although I could not answer the question to my own satisfaction (edits to this regard are welcome)

Currently (2020), the advanced form of ELISA tests for HIV are designed to simultaneously detect both the HIV P24 antigen (produced by the virus) and anti-HIV igG and igM antibodies (produced by the body in response to the virus). Refs (CDC: What kinds of tests are available, and how do they work?)(HIV Diagnostic Testing: 30 Years of Evolution)

While it is easy to detect known viral antigens like the p24 capsid protein in a diagnostic, it is not straightforward to use the same antigen as a vaccine and expect the host to mount an immune response that neutralizes the viral infection, although attempts have been ongoing. Possible scenarios could be that the binding strength of the antibody to P24 might be good enough for detection, but not enough for immunity or that recognizing and binding to P24 does not inhibit the infection. Coming to the evolution/mutation question, since both the p24 antigen and a few antibodies are being detected, it is unlikely the virus would mutate so as to evade the ELISA for all the targets.

The first point in Charlie's answer might be getting at this but it did not come through clearly.


Yes. What you said was absolute right. It's a complicated question. Here are some key points:

  1. To begin with, it very very important to understand that ELISA is a well-established experimental set-up, it does not make any sense to make strong correlations between ELISA results and human vaccination outcomes. This is applicable to any research: We cannot try to make direct link between lab work and real-world problems.

  2. To answer your question 'But if the HIV has a high mutation rate, how can we be able to produce an ELISA test for it?'. We now know that there is a type of antibodies called broadly-neutralizing antibodies (short name: bnAbs) that can fight against a wide variety of mutated viruses. About 20% of HIV patients have bnAbs, and it would take them months even years to have bnAbs. What we do in ELISA is detecting HIV with bnAbs that we found from HIV-infected individuals. But that does not mean we can induce bnAbs with vaccines easily. So these are two completely different process: ELISA takes bnAbs we found to detect HIV, we have those antibodies made and we use it in experiments. Vaccine try to elicit bnAbs in all patients, we know some of the patients can produce bnABs, most of them don't, scientists are working on why that's the case and how we can tackle that.

  • 2
    $\begingroup$ I'm not sure to understand your second point. If only 20% of HIV patients have bnAbs and ELISA only detects HIV with bnAbs, the test should produce a lot of false negatives, right? $\endgroup$
    – Pierre
    Commented Apr 12, 2020 at 9:45
  • $\begingroup$ @Pierre He is saying HIV has a few conserved epitopes, when your antibodies target them you won't develop AIDS, and once sucessfully isolated + hydroma stuffs such antibodies provide a good way to detect HIV-antigens independently of its immune-escape intra-host acquired mutations. What is strange to me is that since the virus stays for a while into the body shouldn't our B-cells pools eventually always find those conserved epitopes ? $\endgroup$
    – reuns
    Commented Apr 15, 2020 at 12:06
  • $\begingroup$ @reuns It's very difficult to find epitopes on HIV. One of the main reasons is that the virus is covered in glycan (sugars). And these glycans are produced by infected human cells-that is to say, we the virus is covered in glycan, it looks like us. Our immune system do not want to attack ourselves. Even if B cell find these epitopes, not every antibody is effective. In many cases, antibodies against conserved epitopes are not very effective. $\endgroup$
    – Charlie
    Commented Apr 16, 2020 at 12:16

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .