British man 'recovered from HIV'

November 13, 2005 12:02 AM

British man 'recovered from HIV' Doctors are planning further tests on a British man whose body has reportedly cured itself of HIV
posted by bluehermit (66 comments total)

Fred Phelps has not returned our calls.
posted by Dean Keaton at 12:05 AM on November 13, 2005

Weren't there some reports recently about how most russians are quasi-immune to HIV, they have a genetic mutation (or similar) that counters the virus?
posted by slater at 12:15 AM on November 13, 2005

Maybe they messed up the first test.
posted by Jairus at 12:18 AM on November 13, 2005

A possible J.D. Shapely moment.

(The important bit is about halfway through the chapter, and the anchors are only per chapter...)
posted by Samizdata at 12:18 AM on November 13, 2005

slater: this thing?
posted by queen zixi at 12:28 AM on November 13, 2005

And yet whenever someone mentions an alternative view on the HIV/AIDS situation, everyone yells "But we already know everything about how it works!"
posted by nightchrome at 12:28 AM on November 13, 2005

Headline: HIV-Positive Gay British Man Recovering from Bush and Currying to His Oppressors, but Still Suffers from Imperialist 'Democratizing' and 'Barebacking' Fantasies
posted by orthogonality at 12:59 AM on November 13, 2005

Slater, I also recall hearing that some survivors of a certain kind of cancer are also immune to HIV . Or was it that they had a genetic disposition to a certain kind of cancer that also made them immune to HIV? I don't remember.

Anyway, I'm really hoping something really good comes of this. Maybe I'm just getting my hopes up though.
posted by kosher_jenny at 1:00 AM on November 13, 2005

Yay, yet another BBC link that I can't read! Thanks China! Another great leap forward in the fight against internet crime.

I don't think this has anything to do with mutated CD4 and CCR5 genes, that confers resistance or immunity, but I fail to see how it could result in someone being cured of HIV. I'm not an expert by any means though, so perhaps some mechanism does exist related to this phenomenon.
posted by [expletive deleted] at 1:06 AM on November 13, 2005

Maybe the first AIDS test gave a false positive? Statistically, it's not unheard of for medical tests to falsely indicate a problem when there isn't one.
posted by Rothko at 1:09 AM on November 13, 2005

Ugh, this is the second thread in a row I've had to do this in, but would someone be so kind as to email me the text of the link?
posted by [expletive deleted] at 1:12 AM on November 13, 2005

Yeah, what Rothko said. It would be surprising if this didn't happen from time to time. I'd be more convinced if there was some evidence that he'd had multiple tests.
posted by i_am_joe's_spleen at 1:18 AM on November 13, 2005

For [exp. deleted]:


posted by Rothko at 1:20 AM on November 13, 2005

Rothko, I'm pretty sure they double checked the samples when they double checked their tests. Then again, I may be wrong.

The tests were re-checked by the Chelsea and Westminster Healthcare NHS Trust when Stimpson threatened litigation believing there must be a mistake, but the results confirmed all the tests had been accurate.

http://www.timesonline.co.uk/article/0,,2087-1870340,00.html
posted by spacediver at 2:08 AM on November 13, 2005

Thanks Rothko.

While it's too early for any real conclusions, this could be really promising. If his immune system did overcome an HIV infection, it could be a huge shortcut to developing a vaccine.
posted by [expletive deleted] at 2:18 AM on November 13, 2005

I'm very dubious about this - Occam's Razor would dictate that a mix-up with the original result would be a much more obvious explanation. That said, if there is anything to it, then it's a shame that he seems to prefer revelling in his "miracle" status and selling his story to rags like the News of the World and the Mail on Sunday than actually cooperating and undergoing further tests that might shed further light on a global problem. But, then again, these might reveal that he was never positive in the first place...
posted by gene_machine at 2:43 AM on November 13, 2005

Diagnosing HIV is generally done thru two tests. First, an ELISA Test is done to detect HIV antibodies. Sensitivity and specificity for this test are above 95%. If this test is positive, it is repeated. Another positive test is followed by a Western Blot (detects viral proteins). A positive test a this point diagnoses someone with HIV.

Point of the post, probability of 3 (or more) tests being falsely positive is non-zero but unlikely.
posted by reverenddrjice at 2:51 AM on November 13, 2005

Waitaminnit.

When we became aware of Mr Stimpson's HIV negative test results we offered him further tests to help us investigate and find an explanation for the different results.

"So far Mr Stimpson has declined this offer. It is therefore difficult for us to comment any further."

Whaa?? The guy is "declining" to help the doctors investigate what might be a breakthrough new mechanism for curing HIV that could save millions of lives?

Nobody's asking the guy to donate an organ, for goodness' sake. I can understand maybe not wanting to be a lab rat for a little while, but still, this seems a little...I don't know...selfish?

What kind of tests would this require that the fellow would want to "decline"? Taking some blood, maybe some tissue samples? Or is he afraid of something else?

/me is confused...
posted by darkstar at 3:16 AM on November 13, 2005

Especially confused, given the subsequent comment from the article:

He has told the papers he would do anything he could to help find a cure.

What am I missing?
posted by darkstar at 3:18 AM on November 13, 2005

nightchrome writes " And yet whenever someone mentions an alternative view on the HIV/AIDS situation, everyone yells 'But we already know everything about how it works!'"

That's generally because the alternate view is something profound like, "fucking virgins cures AIDS" or "AIDS is caused by AZT" or "you don't catch AIDS from unprotected sex -- they put AIDS in condoms!"

With alternative views like those, I'll stick with conventional wisdom, thanks.
posted by Deathalicious at 3:25 AM on November 13, 2005

gene_machine writes "actually cooperating and undergoing further tests that might shed further light on a global problem."

From the article:

"He has told the papers he would do anything he could to help find a cure."

Would you...undergo another fucking test?
posted by Deathalicious at 3:30 AM on November 13, 2005

Oh.

Sorry darkstar. Guess I have to start reading all the comments now or something.
posted by Deathalicious at 3:32 AM on November 13, 2005

Would you...undergo another fucking test?

Sure, if they paid me enough.

Hell, PharmaGlobalCorp spends billions traipsing through rain forests looking for the cure for hemorrhoids, but they can't throw a couple hundred thousand quid this guy's way to be a pincushion for a year or two?
posted by Civil_Disobedient at 4:37 AM on November 13, 2005

gene_machine : "Occam's Razor would dictate that a mix-up with the original result would be a much more obvious explanation."

Occam's razor says that the case which requires the fewest assumptions should be selected. In this case, the number of assumptions is equal ("He recovered from AIDS" vs. "They mixed up the lab tests"), so Occam's Razor doesn't really help us in either direction.
posted by Bugbread at 6:12 AM on November 13, 2005

So they are going to sequence his DNA and check that there are no proviral copies of HIV integrated, waiting to express again?

Good luck.
posted by meehawl at 6:19 AM on November 13, 2005

Viruses mutate all the time, which is why you need a different flu shot every year. Maybe this guy has/had a strain of HIV that the human body can fight. Interesting that there have been anecdotal reports of this in Africa for years. I'll be interested to see what turns up - if this jerk every agrees to undergo further tests, that is. I hope his friends subject him to SERIOUS social pressure to do the right thing.
posted by selfmedicating at 6:22 AM on November 13, 2005

But did he have Good AIDS or Bad AIDS? The nine above it are worth hunting down as well.
posted by DrDoberman at 6:25 AM on November 13, 2005

Or is he afraid of something else?

I would guess that he's afraid that the negative test is the one that got messed up, and he actually still has HIV, but doesn't want to face that possibility.
posted by Gator at 6:31 AM on November 13, 2005

i'm with Gator, or else it was a false positive in the beginning. Nowhere does it say he had the full battery of tests--he might have just had the quick ELISA test.

This whole thing is fishy--does he want money? why decline to be retested yet say you want to help find a cure?
posted by amberglow at 6:54 AM on November 13, 2005

Jeez, ease up you guys. The fella obviously had a mental health crisis as a result of being found positive. He needs help and support, not a bunch of lip from a bunch of nose-picking spectators. I recommend showering him with kindness and support.

I vaguely remember reading a Norman Spinrad story about a spontaneous HIV cure where a guy spontaneously became an anti-plague carrier, but the only way he could transmit the cure was by having unprotected sex with as many people as possible. I can't remember the title or where I saw it.
posted by warbaby at 7:01 AM on November 13, 2005

probability of 3 (or more) tests being falsely positive is non-zero but unlikely

Yes, but it more or less unlikely than being cured of HIV?
posted by callmejay at 7:08 AM on November 13, 2005

He tried to sue over the original positive test result and was told he had no case. Now he says he would do anything to help find a cure, but will not submit to further testing. I have a feeling, were he to be offered a bit of money in exchange for his cooperation, he would be much more obliging.
posted by amro at 7:31 AM on November 13, 2005

Is anyone else wondering why this is in the news all of sudden today, when the "miraculous" negative test results supposedly happened just over two years ago?

It does appear that the guy in question has recently granted an interview with News of the World, which looks to be a British tabloid of sorts. The article says he actually tested negative three times, but doesn't say where those tests were administered.

Maybe he just wanted to sell a good story to the papers, seeing as he's a "sandwich maker"?
posted by Gator at 7:35 AM on November 13, 2005

Magic Johnson overcame HIV. Big deal. Overcoming AIDS is the big deal.
posted by banished at 7:52 AM on November 13, 2005

I would guess that he's afraid that the negative test is the one that got messed up, and he actually still has HIV, but doesn't want to face that possibility

this is the best explanation i've read. there is alot of denial in the HIV world.
posted by brandz at 8:13 AM on November 13, 2005

banished, did MJ overcome it or is it just in remission? My impression is that he is "living with HIV" and is one of those lucky enough to keep things in check with a med cocktail, but I could be wrong.

I too think that this "sudden [er...two-year-old] revelation" in a tab with a less than stellar reputation is questionable. However, it seems to me the least he could do is try and help others, not cash in. Most people I know in stituations like that (weirdo or difficult-to-treat/cure disease with favorable outcome) WANT to help others avoid going through what they did. Not only that, but some studies DO pay participants, or provide them with bennies of other sorts.

On preview, what brandz said is another good explanation.
posted by MiHail at 8:22 AM on November 13, 2005

Magic Johnson overcame HIV. Big deal. Overcoming AIDS is the big deal.

This is precisely reversed.

I am very skeptical about this. HIV has been proven to "hide" deep in the "resting" white blood cells even when the viral load undetectable. This (cute) kid should not be running to the tabloids until he's had redundant and thorough workups.

I shower him with support for having survived the virus, but for telling his story to the press before it's been thoroughly vetted, he's doing everyone in the same situation as he's in a disservice.
posted by digaman at 8:23 AM on November 13, 2005

HIV has been proven to "hide" deep in the "resting" white blood cells

This is exactly what I mentioned above. The mechanism of a retrovirus like HIV is that it uses reverse transcriptase to construct a DNA template of its RNA code. It then uses its enzymatic integrase to insert this DNA into the host cell's DNA, and the host cell DNA then kindly completes the insertion using various ligase procedures. The provirus DNA then blends into the approximately 50% of the human genome that is composed of mobile element DNA. If indeed the initial positive tests were correct then much of that person's cellular DNA has been infected and rewritten by HIV. It is very, very unlikely that all such cells with rewritten DNA have died off without leaving any progeny, or that all infected cells have had their HIV provirus sequences excised.
posted by meehawl at 8:42 AM on November 13, 2005

Point of the post, probability of 3 (or more) tests being falsely positive is non-zero but unlikely.

Yeah, but given the very large number of people given HIV tests, the probability of at least one person having a few false positives in a row and then a true negative is probably pretty unremarkable.
posted by ROU_Xenophobe at 9:10 AM on November 13, 2005

Yes, meehawl, thanks for that clear and frightening elucidation.
posted by digaman at 9:31 AM on November 13, 2005

slater/queen zixi: What you're likely thinking of is actually this.

To infect a cell, HIV-1 has to bind to two proteins expressed on the cell surface: CD4 and either CCR5 or CXCR4. People with a mutation that changes the structure of CCR5 -- called the CCR5-delta32 mutation -- are either partially or completely immune to HIV-1 infection, depending on if they have one or two copies of the mutation. The paper I linked to looks at the abundance/distribution of CCR5 in human populations and models various explanations for its continued distribution/abundance and concludes that historical smallpox epidemics, not the black death, is the most likely explanation.

As to this case, I concur with others who have questioned the tests used to diagnose HIV-1 infection. It is uncommon but not unheard of that an initial test for HIV-1 antibodies returned a positive but subsequent tests for viral load returned negatives. However, if the fellow is still returning negatives from a more specific test like RT-PCR; well, that's something to look at.

But, yeah, having the News of the World involved doesn't inspire confidence.
posted by docgonzo at 9:32 AM on November 13, 2005

He probably doesn't want another test because he's afraid that his most recent test was a false negative.
posted by interrobang at 10:36 AM on November 13, 2005

Gee, I wish I had said that. ;)

This article adds to the hubbub by saying that, when he first tested positive, the dude in question actually stopped bothering with safe sex with his HIV+ boyfriend. Yeesh.
posted by Gator at 10:57 AM on November 13, 2005

I love the part where he has unprotected sex with his HIV-positive partner after he got diagnosed positive.

People are so smrt!
posted by AaronRaphael at 11:00 AM on November 13, 2005

What's so dumb about that? You have HIV, I have HIV, why use protection?
posted by [expletive deleted] at 11:23 AM on November 13, 2005

Because of all the other crud you can transmit/receive without it?
posted by Gator at 11:26 AM on November 13, 2005

Yeah, I'm sure no one in this thread has condom-free sex with their partner. HIV-aside, just think of all the other crud you can get!

People are so b1g0t3d!
posted by scarabic at 11:34 AM on November 13, 2005

What's so dumb about that? You have HIV, I have HIV, why use protection?

There are many different strains of HIV, some more virulent than others.
posted by interrobang at 11:36 AM on November 13, 2005

[expletive deleted], there is the risk of "superinfection".
posted by amro at 11:38 AM on November 13, 2005

If we are to be consistent with our moral justifications for vivisection, then if it turns out that this man indeed overcame HIV, then one might make a case that he should be forced into whatever testing is necessary.
posted by spacediver at 11:50 AM on November 13, 2005

AIDS testing is very carefully designed to avoid false positives due to obvious issues with such results. Therefore it is important to understand that nobody is told they are positive before (in the USA at least) at least the ELISA and Western Blot are positive. I am ignoring all the rapid HIV tests as those, if positive, just make you move to ELISA and Western anyway.

This is from UptoDate Online, a commonly used evidence based medicine resource used by doctors.

"False positive serologic tests — The frequency of false positive tests (positive EIA and Western blot) in a low prevalence population ranges from 0.0006 to 0.0007 percent [25,26]. The most common cause of false positive tests is vaccination. Analysis of 266 healthy volunteers in HIV vaccine studies showed 68 percent had a positive EIA and 0 to 44 percent had a positive Western blot, depending upon the immunogen and the criteria for interpretation [27].

A single case of a false positive test ascribed to autoantibodies in a patient with lupus erythematosus and end-stage renal disease has been reported [28]. Such patients should have confirmatory tests such as polymerase chain reaction (PCR) and culture for HIV-1. Occasional patients report factitious HIV infection, emphasizing the need to confirm test results [29]."

Saying a lot of people are tested for HIV is a bit of an exaggeration. It is not a standard screen (i.e. you dont get one unless you think you might be at risk for having it) so actually the number is low compared to many other screens. You are right...the more people get tested the higher the probablity of getting that false positive result on both. However, few scientiest/doctors would call it common for a screening test with a false positive of .00006% (THATS PERCENT) to "commonly" see false positives. This is no birthday problem phenomena.

Whats more likely? False positive on test or that he "cured" himself of AIDS? Probably a better explanation than either of the two.
posted by reverenddrjice at 12:15 PM on November 13, 2005

Woops....wrong number at the end. Make that .0006%. Sorry.
posted by reverenddrjice at 12:18 PM on November 13, 2005

scarabic, most of us are not HIV-positive people having unprotected sex with other HIV-positive people.

Please read amro's link about superinfection above (or google!).
posted by AaronRaphael at 1:13 PM on November 13, 2005

The probability of a false positive is about six in a million. But even if HIV tests aren't standard screens, there are still millions done every year (unless the number has dropped since 1991, which defies the imagination). Seeing at least one false positive out of a pool that large is to be expected; if anything, we'd expect in the low tens of false positives.

It's also unclear from your quote whether a false-positive rate of 0.0006% is in the test itself, or in the test as given. That is, does the 0.0006% include laboratory contamination of samples, mislabeling of samples, printing errors that cause someone to give the wrong reference number, etc, or only errors in the test itself if perfectly conducted and perfectly managed?
posted by ROU_Xenophobe at 1:33 PM on November 13, 2005

I don't know about you, but I'm going barebacking with random strangers right now!
posted by fungible at 2:07 PM on November 13, 2005

warbaby,
You're thinking of Journals of the Plague Years.
posted by Sangermaine at 2:32 PM on November 13, 2005

That number is based on two studies in reputable medical Journals (NEJM and Annals of Internal Medicine). Lots of labs can diagnose HIV so of course, you can never discount a lab error...no lab is perfect. These two studies seemed to "check" the work of random labs so that does incorporate the variables you mentioned. The recheck done by the researchers was probably well controlled but error could happen at that point as well, right? So that number is not set in stone but I doubt the standard deviation is significant (it wasn't even mentioned).

Bigger point, HIV tests are extremely accurate (compare it to other medical tests). Your math makes sense but I think the argument lies in semantics...I agree that we expect a certain number of FPs based on FP rate but I still wouldn't call that common. In fact, the false positive rate is FAR lower than what is accepted by the medical community for a good test.

Stats are confusing to people, and even though it is possible that this guy was a result of a false positive, I wanted to make sure people didn't get the false impression that these tests aren't extremely accurate and reliable.

TI - Measurement of the false positive rate in a screening program for human immunodeficiency virus infections.
AU - Burke DS; Brundage JF; Redfield RR; Damato JJ; Schable CA; Putman P; Visintine R; Kim HI
SO - N Engl J Med 1988 Oct 13;319(15):961-4.

In a program screening civilian applicants for U.S. military service for human immunodeficiency virus (HIV) infection, we studied the frequency of false positive diagnoses retrospectively among applicants seropositive for HIV in a subpopulation with a very low prevalence of infection. That subpopulation was defined as consisting of all applicants tested between October 16, 1985, and June 30, 1987, who were young (17 or 18 years of age) and resided in a rural county in a state with a low incidence of reported acquired immunodeficiency syndrome (n = 135,187). Serum specimens from 15 applicants positive for HIV in this low-prevalence subpopulation were retrieved from a serum bank and retested by two Western blot methods, radioimmunoprecipitation, and an immunoassay constructed from a molecularly cloned and expressed viral envelope polypeptide. Fourteen of the 15 samples were unequivocally positive on all retest assays, and 1 was negative. Thus, the measured rate of false positive diagnoses in this program was 1 in 135,187 persons tested. Factors important in achieving a low false positive rate were redundant, multistep testing algorithm, conservative criteria for interpreting Western blots, the requirement that a second, newly drawn serum specimen be tested for verification before a diagnosis of HIV was considered established, and tight quality control of laboratory testing procedures. We conclude that a screening program for HIV infection in a low-prevalence population can have an acceptably low false positive rate.

And this study:

TI - Performance characteristics of serologic tests for human immunodeficiency virus type 1 (HIV-1) antibody among Minnesota blood donors. Public health and clinical implications.
AU - MacDonald KL; Jackson JB; Bowman RJ; Polesky HF; Rhame FS; Balfour HH Jr; Osterholm MT
SO - Ann Intern Med 1989 Apr 15;110(8):617-21.

STUDY OBJECTIVE: To evaluate performance characteristics of sequential enzyme immunoassay (EIA) and Western blot human immunodeficiency virus type 1 (HIV-1) antibody testing in a low-risk population. DESIGN: Three-year prospective study of a selected sample from a community-based population. SETTING: Two blood collection facilities in Minnesota. POPULATION: Minnesota blood donors. RESULTS: During the study period, 630,190 units of blood (donations) from an estimated 290,110 Minnesota-resident donors were screened for HIV-1 antibody. Seventeen Minnesota-resident donors were identified as positive for HIV-1 antibody. Sixteen donors were available for follow-up HIV-1 culture: all were culture positive. The other donor, who was not available for follow-up culture, was likely infected with HIV-1 based on a history of high-risk behavior and positive serologic findings for hepatitis B surface antigen. Using 95% binomial confidence intervals, performance characteristics for sequential EIA and Western blot HIV-1 antibody serology were as follows: false-positive rate by number of donations, 0% to 0.0006%; specificity by number of donations, 99.9994% to 100%; predictive value of a positive test, 81% to 100%. CONCLUSIONS: In this low-risk population, the false-positive rate of serologic tests for HIV-1 antibody, using HIV-1 culture as the definitive standard for infection status, was extremely low and test specificity was extremely high.
posted by reverenddrjice at 2:51 PM on November 13, 2005

Yes, of course HIV tests are very reliable, and the probability of a false positive is very, very, very low. But over millions of cases, you'll see several very-low-probability events. That's all.

It is very unlikely that you will be hit by a meteor. It is, at the same time, wholly unremarkable that at least one person has been struck by a meteor.
posted by ROU_Xenophobe at 5:48 PM on November 13, 2005

Today's Guardian reports in more detail.

What this guy seems to have stumbled upon is the disconnect between Lawyers, for whom 0.0006% is a watertight, case-winning impossibility and Scientists, who are more circumspect.

The Lawyer: "Edwina Azimi, the [NHS Litigation] Authority's case manager, is reported to have denied any false diagnosis, hailing the results as "exceptional and medically remarkable". She said: "You have recovered from a positive antibody result to a negative result."

The Scientist: Prof Jonathan Weber from the division of medicine at Imperial College, London, said: "There have been false reports of this phenomenon before. Very rarely a blood test can be falsely positive."
posted by grahamwell at 3:23 AM on November 14, 2005

The results came back negative and Mr Stimpson began legal action against the trust, convinced there had been a mistake with the original diagnosis. But an extensive investigation, including DNA testing on his blood samples, has confirmed all the results.
posted by warbaby at 7:51 AM on November 14, 2005

But it still says he has declined further testing. Other articles say he has essentially gone into hiding "to escape the media spotlight." Errrrr...

I'm now wondering a couple of things: (1) Unless I've missed it, this story hasn't gotten picked up by any major media outlet in the United States. Why is that? (2) The discussion upthread about false positives was fascinating. Has anyone done any research into false negatives? What are the odds that his negative test results were a mistake and he still has HIV?
posted by Gator at 8:24 AM on November 14, 2005

“I love the part where he has unprotected sex with his HIV-positive partner after he got diagnosed positive.
posted by AaronRaphael at 11:00 AM PST on November 13 [!]”

I have unprotected sex with my partner all the time. Of course, I’m married! To a woman! In the USA!
*smugly basks in glow of society that unfairly tips scales in my favor*
*gets struck by meteor*

Seriously tho, what’d this guy do? Walk it off? Throw dirt on it? Given that the tests aren’t in error of course - how would such a thing correct itself? Wouldn’t he just stay HIV negative?
posted by Smedleyman at 8:34 AM on November 14, 2005

Here's more from Reuters. It's hard to tell what the additional tests mentioned by the Scotsman (and warbaby) involved.

The hospital spokeswoman said subsequent DNA checks proved there had been no mix-up in the identity of the patient and the HIV tests but said she did not know whether there could have been any other error in the original test.
posted by grahamwell at 8:55 AM on November 14, 2005

I could only find the incidence of False Negatives on the ELISA test. Recall, you need a positive ELISA and Western Blot to say someone is HIV positive. Western blots can be positive, negative or indeterminate depending on the number of bands that light up. You would never do a Western with a negative ELISA (EIA) so there are no stats for that.

FALSE NEGATIVE RATES
"False negative serologic tests — The predictive value of the EIA depends upon the seroprevalence rates in the patient population being tested. False negative results in a high prevalence population (injection drug users in Baltimore with a seroprevalence rate of about 30 percent) is about 0.3 percent [9]. In a low prevalence population such as blood donors, the false negative rate is 0.001 percent [10,11]."

COMMON CAUSES OF FALSE NEGATIVES
"The most common cause of a false negative result is that the test is conducted between the time of viral transmission and seroconversion.This period infrequently exceeds six months....Other causes of false negative tests include: Agammaglobulinemia, Infection with strains showing little genetic homology with HIV-1, Immunosuppression due to malignancy or medications, Replacement transfusion, Infection with HIV-1 subtype O or HIV-2."

MORE INFO ON SUBTYPES OF HIV-1
There are multiple "clades" or subtypes of HIV-1 designated A through I as group M viruses, ("M" for major). The group O strain of HIV-1 may cause AIDS and shows 55 to 70 percent homology with M subtypes. Group O is reported primarily from West and Central Africa, especially Cameroon. Only one case of infection involving group O virus had been reported in the United States through July, 1996 [16,17]. The EIA kits for HIV screening in the United States and Europe do not consistently detect group O strains [16,18,19]."

MORE INFO ON HIV-2
"HIV-2 is usually not detected with routine serologic tests unless HIV-2 antigens are included in the reagents. Approximately 80 percent of such patients have a positive HIV-1 EIA; however, the HIV-1 Western blots are usually weakly reactive. As a result, most patients with HIV-2 infection have negative or indeterminant HIV-1 serology [20].

Western blots for HIV-2 are neither well standardized nor FDA-approved [20,21]. An HIV-2 EIA was licensed by the FDA in 1990 and became mandatory for screening blood donors in June 1992. Some commercial laboratories now use combination EIA screening assays to simultaneously detect both HIV-1 and HIV-2, although this is not recommended for routine testing by the CDC [22,23]. HIV-2 infection has been reported in 62 patients in the United States as of June, 1995. Most have been immigrants from West Africa or had a sexual partner from West Africa [21,24]."

There are cases in the literature of people seroconverting from positive to negative but as mentioned, the virus was just "hiding."

TI - Loss of human immunodeficiency virus type 1 (HIV-1) antibodies with evidence of viral infection in asymptomatic homosexual men. A report from the Multicenter AIDS Cohort Study.
AU - Farzadegan H; Polis MA; Wolinsky SM; Rinaldo CR Jr; Sninsky JJ; Kwok S; Griffith RL; Kaslow RA; Phair JP; Polk BF; et al.
SO - Ann Intern Med 1988 Jun;108(6):785-90.

Four asymptomatic homosexual men reverted from positive to negative serologic results for the human immunodeficiency virus, type 1 (HIV-1) over 2.5 years, as shown by enzyme-linked immunosorbent assay (ELISA) and Western blot. Antibody bands in the Western blot from three men were undetectable 6 to 12 months after being positive; gradual fading of the number and intensity of bands was seen in the other man. No HIV-1-p24 antigenemia was detected; cryopreserved peripheral blood mononuclear cells were negative for HIV-1 by standard culture assay. Polymerase chain reaction (gene amplification) assays were done on peripheral blood mononuclear cells and showed the HIV-1 provirus in all subjects 6 to 18 months after the last positive antibody test. Serum specimens from each participant were genetically identical. Polymerase chain reaction showed that peripheral blood mononuclear cells from one subject at different times matched by HLA DNA typing. Clinical and laboratory features of these four men were similar to those of other seronegative subjects. Rare, asymptomatic persons seropositive for HIV-1 may not remain seropositive, but may remain latently infected with HIV-1.

Article:
John G Bartlett, MD. Serologic tests for the diagnosis of HIV infection. In: UpToDate, Rose, BD (Ed 13.3), UpToDate,
Waltham, MA, 2005.
posted by reverenddrjice at 11:44 AM on November 14, 2005

In case anyone's still popping in here, there are now a few reports trickling in that it was a false positive in the beginning, after all. Could just be backpedalling, of course.
posted by Gator at 5:47 PM on November 17, 2005

yup--he never had it to begin with.
posted by amberglow at 10:05 PM on November 19, 2005