Wednesday, 17 February 2010

The Case of the Missing Retrovirus

In October 2009, a team led by Vincent C. Lombardi of the Whittemore Peterson Institute reported the presence of a recently discovered virus, XMRV, in 67% of the blood samples from 101 American patients with chronic fatigue syndrome (CFS). XMRV had previously been linked to some cases of prostate cancer.

This sparked intense interest amongst many people and much discussion. But in January this year, Erlwein et al reported that they did not find any evidence of XMRV in the blood of 186 British CFS patients (my post).

Now, a second British study has appeared, and the results are also negative. The paper is Groom et al's Absence of xenotropic murine leukaemia virus-related virus in UK patients with chronic fatigue syndrome. They found no XMRV in 170 British CFS patients or 395 healthy controls. VirologyBlog has an excellent summary of the latest paper.

In order to help people interested in this topic, I've put together a quick summary of all the data on XMRV infection in humans. If I've left anything out or made any mistakes, let me know in the comments. I'll try to keep this list up to date with every new publication - because there are sure to be plenty more.

Overall, the most striking thing about these results is the national differences. XMRV has been detected in 67% of American CFS patients, in 10-25% of American prostate cancer cases, and in 3-4% of healthy Americans. By contrast, in Germany, Britain and Ireland, it's only been detected in 2 Germans, out of a grand total of 1,300 or so European people who have been tested so far using a variety of methods. The situation elsewhere is unclear; one study claimed to detect XMRV in 1.5% of healthy Japanese blood donors but this is unpublished, and the methodology is unclear.

Other than that, it's not clear what's going on here, and it seems to me that it would be premature to conclude anything about XMRV and CFS (or, indeed, cancer) at this stage.


Last Updated: 06 July 2010
Please let me know if I have omitted any data (published or unpublished)

Published Papers

1. Lombardi et al 2009
  • Patients: "CDC Fukuda Criteria and the 2003 Canadian Consensus Criteria... presenting with severe disability... their diagnosis of CFS is based upon prolonged disabling fatigue ... cognitive deficits and reproducible immunological abnormalities ... impaired exercise performance with extremely low VO2 max measured on stress testing."
  • Origin: USA
  • Method A: PCR of DNA from PBMCs
  • Result: 68 of 101 patients (67%), 8 of 218 controls (3.7%)
  • Method B: PBMC reactivity to anti-MLVp30Gag antibodies
  • Result: 19 of 30 patients (63%), 0 of 16 controls (0%)
  • Method C: Plasma immunoreactivity to SFFV-Env
  • Result: 9 out of 18 patients with XMRV, 0 out of 7 controls
2. Erlwein et al 2010
  • Patients: CDC Fukeda criteria "markedly unwell. Few were working, and 19% were members of patient support groups for CFS/ME... The levels of fatigue in this sample were high ... as were levels of disability"
  • Origin: London, UK
  • Method: PCR of DNA from whole blood
  • Result: 0 out of 186 patients (0%)
3. Groom et al 2010
  • Patients: CDC Fukeda criteria
  • Origin: Bristol, Dorset, London, Birmingham, Norfolk and Epsom, UK
  • Method A: PCR of gDNA from PBMCs
  • Result: 0 of 48 patients (0%)
  • Method B: PCR of gDNA, cDNA, or both from PBMCs
  • Result: 0 out of 142 patients (0%), and 157 controls (0%)
  • Method C: Serum immunoreactivity to XMRV
  • Result: 1 out of 160 patients; 25 out of 395 controls; but positives were not considered specific to XMRV, as they also reacted to and neutralized other viruses.
4. Switzer et al 2010 "CDC study"
  • Patients: Recruited by random population telephone screening and symptom quizzing (unlike other studies). CDC Fukeda 1994 criteria (but also referred to in the abstract as the "revised" 1994 CDC criteria) and symptoms not better accounted for by medical illness or "current psychiatric disorders considered exclusionary for CFS, which included current melancholic depression, current or lifetime bipolar disorder or psychosis, substance abuse within 2 years and eating disorders within 5 years".
  • Origin: Wichita, Kansas, USA and Georgia, USA.
  • Method A: Western blotting serology for anti-XMRV antibodies in plasma
  • Result: 0 of 51 patients (0%) and 0 of 53 controls (0%)
  • Method B: ELISA serology for anti-XMRV gag and pol antibodies in plasma, performed by RKI.
  • Result: 0 of 51 patients (0%) and 0 of 53 controls (0%)
  • Method C: PCR of DNA from PBMCs or whole blood for XMRV gag and pol
  • Result: 0 of 50 patients (0%) and 0 of 97 controls (0%)
  • Method D: PCR of DNA for XMRV gag, performed by BSRI.
  • Result: 0 of 50 patients (0%) and 0 of 56 controls (0%)
5. van Kuppeveld et al 2010 "Dutch study"
  • Patients: Oxford Criteria
  • Origin: Netherlands. Recruited and samples taken 1991-1992
  • Method: PCR of cDNA from PBMCs for XMRV integrase and gag
  • Result: 0 of 32 patients (0%), 0 of 43 controls (0%)
Published Papers - Prostate Cancer

6. Urisman et al 2006
  • Patients: Familial Prostate Cancer
  • Origin: Cleveland, USA
  • Method: PCR on prostate cell DNA
  • Result: 9 of 86 (10.4%); associated with R462Q QQ genotype
7. Schlaberg et al 2009
  • Patients: Prostate Cancer
  • Origin: Columbia University Medical Center, USA
  • Method A: PCR on prostate cell DNA
  • Results: 14/223 prostate cancer patients (6.2%), 2/101 non-cancer prostate controls (2.0%). Not associated with R462Q QQ genotype.
  • Method B: XMRV protein expression (cell reactivity to anti-XMRV serum)
  • Results: XMRV protein expression in 54/223 (23%) cases with prostate cancer and in 4/101 (4%) controls. Not associated with R462Q QQ genotype.
8. Hohn et al 2009
  • Patients: Prostate Cancer
  • Origin: Berlin, Germany
  • Method: PCR on prostate cell DNA
  • Result: 0 out of 589 (0%)
  • Method: Serum immunoreactivity to XMRV proteins (gp70 and Gag)
  • Result: 0 out of 146 patients, 0 out of 5 controls (0%)
9. Fischer et al 2008
  • Patients: Non-familial Prostate Cancer
  • Origin: Hamburg, Germany
  • Method: PCR of prostate cell RNA
  • Results: 1/105 patients (0.95%), 1/70 healthy controls (1.42%).
Unpublished Data

10. D'Arcy et al 2008
  • Patients: Prostrate Cancer
  • Origin: Dublin, Ireland
  • Method A: PCR of prostate cell RNA
  • Results: 0 out of 9 (7 R462Q QQ genotype, 2 others) (0%)
  • Notes: Unpublished data presented at a conference.
11. Furata et al 2009
  • Patients: n/a
  • Origin: Japan
  • Method: "antibodies [to XMRV]"
  • Result: 5/300 controls (healthy blood donors) (1.5%)
  • Note: Unpublished data presented at a conference - I can't access the abstract and am relying on Erlwein et al's summary. In the light of Groom et al, we really need to know whether the antibodies were truly specific to XMRV.
12. "The Harvey Alter NIH/FDA Paper"
  • Patients: CFS?
  • Origin: ?
  • Method: ?
  • Result: "We (FDA & NIH) have independently confirmed the Lombardi group findings." according to a conference presentation by Harvey Alter: see here.
  • Notes: Rumored as of 5th July 2010 to be about to appear in PNAS.
13. Qui et al 2010
  • Patients: n/a
  • Origin: U.S. blood donors
  • Method: Blood immunoreactivity to XMRV env, gag, and p15E.
  • Result: "reactivity to all 3 antigens in a low proportion (~0.1%) of US blood donors." n=?
  • Notes: Conference presentation given at the 17th Conference on Retroviruses and Opportunistic Infections 2010.
ResearchBlogging.orgHarriet Groom, et al. (2010). Absence of xenotropic murine leukaemia virus-related virus in UK patients with chronic fatigue syndrome Retrovirology


Anonymous said...

Great article. Not an area I'm particularly interested in but the clarity of data is superb!

CFS-having Jerk said...

Good point about the two sides of the pond, and we'll certainly see several American papers soon. However, last 9 October we read:

'[Mikovits] is testing a further 500 blood samples gathered from chronic fatigue patients diagnosed in London. "The same percentages are holding up," she said.'

CFS-having Jerk said...


dearieme said...

Just an ordinary case of American Exceptionalism?

Anonymous said...

The WPI study included samples from UK, Ireland, Germany and Australia.

Two groups from UK sent their blood to be tested in the US and about 50% of each group has tested positive. It's been found in Japan.

What's missing is real science in the first UK paper.

What's missing is the proper methods in the second paper. They didn't pay attention to what it takes to cultivate XMVR; they acted as if they were dealing with HIV, but this retrovirus is different.

Sooner or later it will become clear to those who will look at the facts.

Neuroskeptic said...


"The WPI study included samples from UK, Ireland, Germany and Australia."

According to the Supplementary Material to Lombardi et al:

"Samples were selected from several regions of the United States where outbreaks of CFS had been documented."

Neuroskeptic said...

CFS-having Jerk: Wow - I wasn't aware of that claim. If the WPI are really finding rates of 65%-95% in London CFS patients, while two British teams aren't finding any evidence of XMRV in anyone from Britain, methodological differences are likely the cause... especially if the WPI XMRV test comes back positive on the same samples that Erlwein et al or Groom et al tested negative.

In which case we could be looking at an old fashioned, your-methods-suck scientific competency showdown.

CFS-having Grouch said...

I'd have lost most hope by now (not quite all), except for the separate NCI and Cleveland contributions to the Mikovits paper in Science. Even as it is, I think the whole thing only 50% likely to be true.

Kerr and Gow are completely trusted by us; we'd gladly entrust our parakeets and miniature schnauzers to these gentlemen in the midst of a pan-continental joint irruption of famished pythons and malarious vectors. If Gow were transformed into a 16-foot barracuda, I would let him take care of my coho salmon Jimmy, deliver sardines and anchovies unsupervised under my employ, borrow my car repeatedly, and date my kid sister. As outspoken proponents of a biological cause, whose professional and lay status will loft when one is finally found, they should if anything be biased in favor of XMRV. I can't say that I closely read through their oddball serology using pseudotypes, though. Why not keep it simpler?

There has been talk of standardizing assays. Especially since the prostate cancer results were erratic even before the CFS picture became erratic. But nothing has been announced. I'm sure discussions are going on, backchannel.

This limbo could at worst go on for years.

Neuroskeptic said...

Well, the WPI have now replied to the new study, and they make the point that:

"Perhaps the most important issue to focus on is the low level of XMRV in the blood. XMRV is present in such a small percentage of white blood cells that it is highly unlikely that either UK study’s PCR method could detect it using the methods described. Careful reading of the Science paper shows that increasing the amount of the virus by growing the white blood cells is usually required rather than using white blood cells directly purified from the body. When using PCR alone, the Science authors found that four samples needed to be taken at different times from the same patient in order for XMRV to be detected by PCR in freshly isolated white blood cells."

This is a point I hadn't noticed about the Science paper, and it could be relevant.

Although I don't think it could explain the difference in XMRV rates seen in American vs. German prostrate cancer patients...

CFS Grouch said...

> Although I don't think it could explain the difference in XMRV rates seen in American vs. German prostrate cancer patients...

I take it the american prostate group didn't employ culture/incubation prior to DNA extraction, and that's what leads you to this sentiment.

Anyone would admit, it make no little sense for the thing to have many more copies /mL in the prostate than in the blood, if in fact it likes to transmit amorously.

Anonymous said...

Neuroskeptic, Dr Mikovits said in a public presentation that she didn't know that some of the patients in the original study came from UK, Ireland, Germany and Australia until the cohort was unblinded in December. Apparently patients from these countries went to American doctors for treatment. Her presentation is available for viewing on you tube - sorry I'm too sick to look it up for you.

blog friendly said...

CFS Grouch said- "Anyone would admit, it make no little sense for the thing to have many more copies /mL in the prostate than in the blood, if in fact it likes to transmit amorously."

I'm not sure if this is the case; in fact one recently published study showed that something in semen stimulated XMRV replication-

"The xenotropic murine leukemia virus-related virus (XMRV) has recently been detected in prostate cancer tissues and may play a role in tumorigenesis. It is currently unclear how this virus is transmitted and which factors promote its spread in the prostate. We show that amyloidogenic fragments known as semen-derived enhancer of virus infection (SEVI) originating from prostatic acid phosphatase greatly increase XMRV infections of primary prostatic epithelial and stromal cells."
'Fibrils of prostatic acid phosphatase fragments boost infections with XMRV (xenotropic murine leukemia virus-related virus), a human retrovirus associated with prostate cancer'
Hong, S et al.
J Virol. 2009 Jul;83(14):6995-7003.

Plus I think Dr. Mikovits has stated that blood is not the viral resovoir.

David said...

Thanks for sharing this. I'm going to buzz my virology professor 'cuz he brought the XMRV thing up a few months ago in class =)

Angela Kennedy said...

A KEY problem with both the Erlwein and Groome paper relates to the patient cohort and whether they were similar enough to the patient cohort in the Lombardi et al project.

The Lombardi et al research cohort apparently met criteria for 'ME/CFS' as identified by Carruthers et al (in addition to Fukuda et al, which by definition would be the case anyway) which has been demonstrated to select patients with post-exertional malaise and fatigue, sleep dysfunction, pain, clinical neurocognitive, and clinical autonomic/ neuroimmunoendocrine symptoms (Jason et al).

Furthermore, the WPI give this information about their patient cohort in their supporting online material:

"Their diagnosis of CFS is based upon prolonged disabling fatigue and the presence of cognitive deficits and reproducible immunological abnormalities. These included but were not limited to peturbations of the 2-5A synthetase/RNase L antiviral pathway, low natural killer cell cytotoxicity (as measured by standard diagnostic assyas) and elevated cytokines particularly interleukin-6 and interleukin-8. In addition to these immunological abnormalities, the patients characteristically demonstrated impaired exercise performance with extremely low VO2 max measured on stress testing..." (

The Erlwein et al project selected a rather different patient cohort:

""Patients in our CFS cohort had undergone medical screening to exclude detectable organic illness".

The Groome paper used Fukuda alone, and did not establish a 'Canadian' defined cohort.

In any British subsequent research to Lombardi et al, unless the patient cohort selected was similar enough to that in the Lombardi research, it is premature to believe XMRV is not present in the 'CFS' population in Britain.

Ongoing neglect of the importance of establishing a possible ‘CFS’ patient population in Britain, clinically and in research settings, using the Canadian Guidelines, is preventing the development of knowledge that might help extremely ill and disabled people, diagnosed with 'CFS', known to be here in Britain.

The problems I have briefly outlined here do not fully express the range and depths of problems with regard to: the identity of an accurate ‘CFS’ population; the instabilities of ‘CFS’ criteria per se; the faulty concepts of ‘medically unexplained’ or ‘functional‘ and relation to ‘psychogenic‘ explanations for somatic illness; the vagaries of criteria that claim to facilitate a ‘diagnosis of exclusion’; and the psychogenic dismissal of serious organic dysfunction of patients given a ’CFS’ diagnosis, problems that have happened for many years.

CFS Jerk said...

Why would Gow and Kerr study a bunk cohort? I wouldn't altogether put it past Wessely, but our boys are another story. It's far more plausible that they either are right, or simply erred.

Neuroskeptic said...

CFS Jerk: Also, I find it hard to accept the cohort explanation because even if the British "CFS" patients were in fact totally random healthy people, you would still expect 2-4% of them to have XMRV according to the American studies.

To my mind the most likely explanations now are: a) there's no XMRV in Britain or b) there is, but you need to culture PBMCs in order to detect it in blood.

The data I'm most looking forward to seeing is the WPI's results from British samples. Rumour has it Brits have been sending them samples to test, I hope they publish this data at some point...

Angela Kennedy said...

Re the questions about Gow and Kerr studying a 'bunk' cohort, it's possible because there are likely no 'Canadian' criteria- selected research cohorts established in Britain. This was 'quick' research -too quick to establish the kind of cohort that Lombardi did. And failing to replicate that cohort has been a little like looking for the HIV virus in 'AIDS' sufferers with none of the typical presentation of AIDS, none of the medical histories associated with the illness, but who feel a bit tired.

That the cohort discrepancy is a significant problem is plausible when added to the problem of virology detection techniques. It is a key problem in all of 'CFS' research, and adds an extra confounding factor to the possible problems of detection techniques in the case of XMRV. It cannot therefore be dismissed out of hand.

Also - people need to stop both demonising and deifying scientists in the field of CFS. Their scientific method, extrapolations, theoretical problems and biases need to be ascertained and where necessary critiqued, and 'white hat/black hat' constructions are not conducive to rational analysis of the claims made under the auspices of the 'science'. Frankly people (like me) need to be able to critique the claims of those like Gow and Kerr without being made to feel like I've shot a puppy, and be able to critique the claims of those like Wessely, without being falsely accused of having a 'visceral hatred' for him, something that has happened to me.

impatient patient said...

Thanks for keeping this ongoing list, Neuroskeptic. It’s both handy and informative.

I don’t see the Dutch study by Van der Meer et al 2010, “Prevalence of xenotropic murine leukaemia virus-related virus in patients with chronic fatigue syndrome in the Netherlands: retrospective analysis of samples from an established cohort” (

18 Feb 2010 “In which case we could be looking at an old fashioned, your-methods-suck scientific competency showdown.”
Boy, was that a prophetic remark! Scientific smack-down.

In her analysis of the CDC study, Dr Suzanne Vernon, Scientific Director of the CFIDS Association of America, points out “Further, the samples from these three study cohorts were collected using different types of tubes, each of which has a distinct way of being processed. As if this weren’t bad enough, none of the blood tubes used were of the same type used in the Lombardi study. (They used tubes containing sodium heparin that are intended for use with virus isolation). The blood tubes from the 18 Georgia registry patients are designed to collect whole blood and preserve nucleic acid; it is not clear where the plasma came from for these subjects since plasma cannot be obtained using these blood tube types.” I'd like to hear your take on this.

We’re still waiting for a replication study to be published. Somehow you’d think that with something as potentially important as a new infectious human retrovirus, SOMEBODY would do a true replication study, using the same methodology and patient criteria as the original. Wouldn’t you?

Neuroskeptic said...

Oh yeah, good point about the Dutch study. Will put that in asap.

Mr.Kite said...

The retrovirus isn't missing, it's been detected in upward of 95+ percent of CFS patients by WPI and now NIH/FDA.

What's missing are the proper methods for detecting the retrovirus in the studies that failed to find it. Basically all those studies showed is that their authors do not know how to detect XMRV, which they proved with flying colors.

If I tell you that if you dive down a certain distance into a particular area of the Pacific ocean you will see some remarkable and unique sea anemones, and then you tell me that you dove down into Lake Champlain but you didn't see any, all that shows is that you didn't follow the proper procedure for seeing the anemones. It's not evidence that the anemones don't exist in the Pacific.

The negative studies, including the CDC study, are seriously flawed. They are not replication studies, they are studies that prove what we already know: that you can't detect XMRV when you're using the wrong methods. With perceptive and intelligent people working on this - independent scientists who are not subject to government and political pressure - eventually the truth will out. The retrovirus is out of the bag, and CDC can't cover this one up, no matter how many weak and improperly conducted studies they do.

Cort said...

Well done neuroskeptic!

Cort said...

Well done neuroskeptic! Love the well organized look at the different xmrv Studies.