Wednesday, September 1, 2010

FAQ on Lo/Alter Study

The CFIDS Association compiled a series of questions and answers on
XMRV/MLV's with a highly respected group of professionals for this
month's CFIDSLink.
The entire issue can be found here-

FollowUp FAQs to the Study by Lo, Alter et al.
September 2010 CFIDSLink

In the days following the August 23, 2010 publication in the
Proceedings of the National Academy of Sciences (PNAS) of, "Detection
of MLV‐related virus gene sequences in blood of patients with chronic
fatigue syndrome and healthy blood donors" by Shyh‐Ching Lo, Harvey J.
Alter and colleagues at the Food and Drug Administration, National
Institutes of Health and Harvard Medical School, we collected
frequently asked questions (FAQs). We requested answers from a range
of experts familiar with this study, the emerging field of research
into XMRV and other murine leukemia retroviruses, blood safety issues
and the related media coverage. Here are their replies, in their own


Q: Do you consider the PNAS study to have been designed as either a
validation or replication study of the Lombardi et al. study?
A: The PNAS study was stimulated by the Science publication and by the
negative studies that followed. We tested our cohort of CFS patient
samples and controls without bias. That is, we did not have an
expectation in either direction, but simply wanted to know whether or
not we could find XMRV in our samples. As it turned out, we did not
find XMRV, but rather, closely related polytropic murine leukemia
viruses (MLVs). The strong association of the MLVs with CFS identified
in our patient population was similar to the strong association of
XMRV and CFS reported by Lombardi et al. However, the role of MLVs in
the CFS disease process is still not clear. – Harvey J. Alter, MD,
MACP, Chief, Clinical Studies & Associate Director for Research,
Department of Transfusion Medicine, National Institutes of Health and
Shyh‐Ching Lo, MD, PhD, Medical Officer, Division of Cellular and Gene
Therapies, Federal Drug Administration
Q: Were the samples tested under blinded conditions?
A: We did not specifically blind and mix the two sample groups (CFS
and blood donors). However, they were studied in parallel by
polymerase chain reaction (PCR). As shown in the figures of the paper,
those samples with positive amplicons of the predicted sizes were all
sequenced; no sample was considered positive unless sequencing
confirmed PCR reactivity. – H.J. Alter, MD, MACP and S‐C Lo, MD, PhD
Q: The FDA has posted a Q&A on its website, stating that Dr. Lo tested
34 CFS patient samples from CDC and that there were "no XMRV‐positive"
results. Did he test for the broader group of MLVs? Will these results
be published when all 82 samples from the CDC's negative study have
been tested?
A: The CDC samples were tested for the broader group of MLVs. While we
did not find any of the 34 CDC CFS samples to sequence as MLV, there
were three samples that were indeterminate in that they had weak
PCR‐positive signal, but could not be confirmed by sequencing. These
samples are now being further amplified and cloned for resequencing.
CDC and FDA/NIH will continue to work to
understand the disparities, but there is no current plan to publish
the results of this sample exchange. It was not critically designed to
be a study for publication, but larger coded panel testing conducted
by NHLBI in which both CDC and FDA, as well as other labs, are
participating will be published when the data are complete. – H.J.
Alter, MD, MACP and S‐C Lo, MD, PhD
Q: Do the three variants of MLVs found in the CFS patients from your
practice correlate to any particular CFS symptoms, type of onset, age,
illness severity, etc.?
A: Not that we can tell, but the total numbers are small. We need much
larger studies to answer this important question. — Anthony L.
Komaroff, MD, The Simcox/Clifford/Higby Professor of Medicine, Harvard
Medical School; Editor in Chief, Harvard Health Publications, Harvard
Medical School; and Senior Physician, Brigham and Women's Hospital
Q: Eight of the 25 patients from your clinic provided fresh blood
samples earlier this year for testing by Dr. Lo. How were those people
selected? Are you getting samples from the other 17 so that they can
be tested too?
A: These eight patients live in the Boston area and were available to
have the blood drawn on a specific day, with short notice. We are
attempting to get samples from the other individuals as well. – A.L.
Komaroff, MD
Q: You've been at this a long time and have seen evidence that
associates several different infectious agents with CFS. What are your
thoughts on this particular study?
A: Based on my experience talking to the patients, and examining them,
I think that the most likely explanation of their illness in most of
the patients with CFS is that they are suffering from some kind of
chronic infection. I think it is very plausible that the infection is
of a type that cannot be fully cleared by the immune system, although
that has not been proven. I think it is very plausible that the key
symptoms of CFS are caused by the immune system's attack on the
infectious agent(s) that may be involved, although that has not been
proven. Finally, since symptoms are experienced in the brain, I think
it is very plausible that the immune system molecules that may cause
the symptoms are either produced in the brain (because the infection
is there) or reach the brain through the circulation, although that
has not been proven.

This study found strong evidence that a group of retroviruses that
were first discovered to infect mice may also be infecting many
patients with CFS, and also a few healthy blood donors. However, other
studies—both published and unpublished—have not found that. All of the
laboratories that have studied this question need to work together to
try to understand why they have gotten different results. The PCR
techniques that were the basis of most of these studies are very
tricky: they can be falsely positive, and they can be falsely
negative. Dr. Lo's laboratory took great pains to rule out various
types of falsely positive results, as explained in our paper. We also
proposed some reasons why other studies might have obtained falsely
negative results, but that is just speculation.

In summary, our study does not and should not settle the question as
to whether mouse retroviruses may be associated with CFS. It is one
study, one piece of evidence. Scientific conclusions require multiple
studies, and multiple types of evidence. More work needs to be done,
particularly among those laboratories already engaged in the study of
this question, to understand why their results are different. Even if
it is concluded that these viruses are often present in patients with
CFS, that will not prove that the viruses are a cause of CFS. So we
are a long way from the finish line in getting solid answers to these
important questions. – A.L. Komaroff, MD
Q: There isn't any information included in the article about 12
patients whose samples were sent to Dr. Lo in the early 1990s. By
which criteria were they diagnosed? Were there any particular clinical
features about these patients that made them different from other
patients you were seeing at the time or have since in the years since?
A: I do not recall the particulars of such patients as that was almost
20 years ago. However, my clinic then saw only CFS and they are pretty
much the same as what I see today from 26 states and 7 countries. I
have a poster presentation at the International XMRV Workshop on
September 7‐8 regarding XMRV/MLV/HGRV detection in my practice done on
47 consecutive patients from October 2009 to December 2009. Those 47
patients are very well characterized including the percentage of
overlap diagnoses (FM, MCS, Lyme, IBS, mold‐illness), KPS, age, sex
and geographic distribution. The data, I think, are extraordinary, but
I cannot talk about it until after the 8th. – Paul R. Cheney, MD, PhD,
The Cheney Clinic

A: All patients whose samples I sent for study at that time met the
1988 Holmes criteria for CFS. I have always felt that activity
limitation to the imaginary 50% level was crucial to the diagnosis. –
David S. Bell, MD, FAAP, Lyndonville, N.Y.
Q: In a Q&A with PNAS editor‐in‐chief Randy Schekman published in The
Scientist, he says, "It looks like [Alter's] done as good a job as he
could without actually show[ing] that the sequences were contiguous
with human DNA. That I trust will be what he does next." What kind of
evidence is Schekman referring to?
A: He was referring to the third reviewer's request that the team
demonstrate that DNA from a murine leukemia virus has integrated into
the host's (human) DNA. This would provide stronger evidence for
infection with a retrovirus. Different retroviruses target different
integration sites, and evidence of the region of the host's genome
targeted by the virus would also strengthen this work. As the authors
have said, these are early studies and there is still much work to do.
– Susan L. Stramer, PhD, Executive Scientific Officer, Scientific
Support Office, American Red Cross

Q: As presented at the Cold Spring Harbor Laboratory (CSHL) conference
on Retroviruses (May 2010) and reported in the Wall Street Journal's
Health Blog, the Whittemore Peterson Institute (WPI) has found gag
sequences that reflect greater variability amplified from the PBMCs
from CFS patients than was originally reported for XMRV. Do the
variants you have identified match the four MLV variants described in
the PNAS paper, or do they represent even greater diversity within
this family of gammaretroviruses?
A: The variants presented at CSHL were exactly the variants described
in the Lo et al. publication. The abstract was submitted in late
February/ early March by our collaborator Dr. Kathryn S. Jones and she
reported that the majority of patients reported by Lombardi et al. had
both xenotropic and polytropic MLVs. This is consistent with the
antibody and serology data presented in Lombardi et al. which
suggested wider sequence variation as XMRV VP 62‐specific PCR negative
patients shown in Figure 1 [in the original Science paper] indeed
harbored XMRV‐related viruses that could be detected by the antibodies
that recognized all known xenotropic, polytropic and ecotropic

It is unfortunate that no matter how much data we presented and
published (Mikovits et al., Virulence, July 30, 2010) suggesting more
sequence variation as a valid scientific explanation for negative
studies both in CFS and prostate cancer, our data were met not with
critical evaluation but with unsupported allegations of contamination.

We congratulate Dr. Alter and sincerely appreciate his statements
defending our work and stating definitively that his work confirmed
the data in Lombardi et al. and putting to rest any allegations that
there was ever contamination in our work. We had done every possible
control and never was there any evidence of contamination. Hopefully
now that the association of a human gamma retrovirus family with CFS
has been firmly established, the scientific community can work
together to move the research forward to understand the role of these
viruses in disease and develop treatment protocols as soon as
possible. – Judy A. Mikovits, PhD, Director of Research, Whittemore
Peterson Institute
Q: Some of the media reports have referred to murine leukemia viruses
(MLVs) as "cancer‐causing agents." This description refers to some
MLVs being cancer‐causing in mice, but what does it mean for humans?
A: The MLV family of viruses has never been shown to cause cancer in
humans. There are lots of animal viruses that cause tumors in other
species. These viruses are also related to feline leukemia virus and
porcine leukemia virus. Similarly, it is often possible to cure tumors
in mice and rats, but the same agents are ineffective for human
tumors, so it is both inaccurate and misleading to suggest that these
viruses cause human tumors. – Harvey G. Klein, MD, Chief, Department
of Transfusion Medicine, NIH Clinical Center

A: Human T‐lymphotropic virus (HTLV) was the first human virus linked
to cancer, followed by hepatitis B virus (HBV) and hepatocellular
carcinoma. Mouse viruses while having the capacity to infect human
cells, have not been associated with any human cancers. – Susan L.
Stramer, PhD, Executive Scientific Officer, Scientific Support Office,
American Red Cross
Q: Many of the news reports have been accompanied by pictures of field
mice or laboratory mice, like this one from CNN. Does this study (or
any of the other papers) suggest or prove that XMRV and/or MLVs
sequences in humans come through direct contact with mice or other
A: XMRV and viral sequences reported by Lo et al. are closely related
to some murine leukemia viruses making it highly likely that the viral
ancestors emerged from mice prior to infecting humans. However, it is
unknown how or when that occurred, or whether there are intermediate
hosts between mice and humans. There is no suggestion that these
viruses are spread by insects that bite mice, this route of
transmission occurs with other some other viruses, like West Nile
Virus for instance, but not with retroviruses such as XMRV. – Robert
H. Silverman, PhD, Mal and Lea Bank Chair and Professor, Department of
Cancer Biology, Lerner Research Institute, Cleveland Clinic

Q: Will the assays being used in the XMRV Scientific Research Working
Group REDS‐II Panel Study be able to detect XMRV, the MLVs described
by Lo et al., and a wider variety of MLVs that might be associated
with CFS?
A: The assays employed by the labs participating in the REDS‐II Panel
Study employ PCR primers targeting multiple genetic regions of XMRV
and MLV variants. The labs include the Lo lab at FDA and the
Whittemore Peterson Institute lab that have reported high rates of
detection of XMRV and MLV‐related viruses in CFS patients, as well as
labs that have reported absent or low rates of XMRV/MLV virus
detection in clinical populations (Switzer (CDC), Simmons (Blood
Systems Research Institute), and Hewlett (FDA) labs) and one lab that
has not previously evaluated clinical samples for XMRV/MLV but which
has extensive published experience with very sensitive assays for HIV
(National Cancer Institute lab), and is using an assay which has been
shown to detect and distinguish both types of viruses. – Simone Glynn
MD, MSc, MPH, Branch Chief, Transfusion Medicine and Cellular
Therapeutics Branch, National Heart, Lung and Blood Institute and
Jerry A. Holmberg, PhD, Senior Advisor for Blood Policy, Department of
Health and Human Services, both responding on behalf of the DHHS XMRV
Scientific Research Working Group
Q: If geographic restrictions are a source of the discordant data on
XMRV in CFS, will the REDS‐II Phase IV study of donor prevalence using
samples from the Reno/Tahoe region be sufficient to identify MLVs
strains that might be more prevalent in other geographic locations (as
suggested in the PNAS paper)? 5
A: The Phase IV study is currently restricted to donations from the
Reno/Tahoe region so it cannot address the prevalence of XMRV/MLV
variants in other blood donor populations from other geographic
settings. Other studies are in the planning stages to both broaden the
geographic representation of US donors and to investigate archived
donor samples from the past four decades to establish temporal,
geographic and other demographic correlates of infection/exposure to
these viruses in donor populations. – S. Glynn, MD, MSc, MPH and J.A.
Holmberg, PhD responding on behalf of the DHHS XMRV Scientific
Research Working Group
Q: Will the XMRV Scientific Research Working Group REDS‐II study be
adequate on its own to determine whether XMRV and other MLVs pose a
threat to the blood supply?
A: Not on its own, but it will certainly be easier to prove
transmission transfusion than to prove disease causation. If indeed
4‐8% of normal blood donors are infected with MLVs and the rate of
random blood recipients is similar and if infection is a fairly
frequent event after transfusion, then the available repository would
be expected to hold enough positive units to estimate penetration of
the blood supply by MLVs with adequate confidence levels.

A variety of studies and observations will be needed to prove disease
causation, particularly if the disease has a long incubation period.
At the moment, though, we do not know of any significant diseases that
appear to be transfusion transmitted for which there is not an
identified agent. Some forms of post‐transfusion mortality and/or
immune modulation still have not been adequately explained. The first
of these may be associated with "older" blood components; infectious
agents have not generally been suspected in either of these outcomes.

So, additional studies may be needed to prove transfusion transmission
(depending on the extent to which it may occur) and certainly other
studies are needed to prove disease causality. Frozen repositories,
like the National Heart, Lung and Blood Institute (NHLBI) repository,
may be helpful if assays can successfully recover MLVs by nucleic acid
testing from frozen cells and plasma collected from suitable samples.
– Roger Y. Dodd, PhD, Vice President, Research & Development, American
Red Cross; Louis M. Katz, MD, Executive Vice President, Medical
Affairs, Mississippi Valley Regional Blood Center; and Susan L.
Stramer, PhD, Executive Scientific Officer, Scientific Support Office,
American Red Cross
Q: There have been several different tests marketed for XMRV and now
one is available that also tests for MLVs. Are any of these diagnostic
tests approved by the FDA?
A: No; to our knowledge there are currently no FDA‐approved tests for
XMRV for diagnosis or other indications. – Indira Hewlett, PhD, Chief,
Laboratory of Molecular Virology, Division of Emerging and Transfusion
Diseases, Office of Blood Research and Review, Center for Biologics
Evaluation and Research, FDA
Q: Why don't blood collection centers use the tests being marketed to
physicians and patients for XMRV and other MLVs to test the blood
A: In our lab at the Mississippi Valley Blood Centers we need a level
of testing throughput to handle up to 1000 tests in a shift. That's
not possible with the technology being used by labs offering clinical
tests. Also, the tests currently on the market have not yet been
validated for appropriate performance characteristics (sensitivity,
specificity, positive and negative predictive values) for use by blood
collection centers. Obviously, the question also assumes these murine
leukemia viruses both cause human disease and are transfusion
transmitted pathogens, i.e., that we should be testing blood donors,
but that is the more difficult part of this whole discussion. – Louis
M. Katz, MD, Executive Vice President, Medical Affairs, Mississippi
Valley Regional Blood Center
Q: Reports on the Aug. 25 and 26 Wall Street Journal Health Blog
indicate that some scientists have shortened XMRV to "X" and
polytropic MLVs to "P." Another group has suggested renaming the
assortment of viruses found in the Science and PNAS papers to "HGRV
for Human Gamma Retro Virus" and the illnesses caused by these
infections to "HGRAD for Human Gamma Retrovirus Associated Disease."
How do viruses get named?
A: There is an International Committee on the Taxonomy of Viruses
(ICTV) of the Virology Division of the International Union of
Microbiological Societies whose purposes are to develop an
internationally agreed upon taxonomy and nomenclature for viruses; to
maintain an index of virus names; and to communicate the proceedings
of the committee to the international community of virologists. An
update of this work is usually published at three‐year intervals.
However, the most recent published full report was in 2005 and
contains an authoritative database (ICTVdB) containing taxonomic
information for 1,950 virus species. The website database was last
updated on 7/20/2010. Anyone can submit information to the Virology
Subcommittee on taxonomy for consideration in the database and it
doesn't stop informal use of any terms. However, until several items
are resolved, it tends to confuse rather than educate, and clutters up
the literature. – F. Blaine Hollinger, MD, Professor of Medicine,
Molecular Virology & Epidemiology and Director, Eugene B. Casey
Hepatitis Research Center, Baylor College of Medicine.

Q: Some people have been surprised that the PNAS study's finding of
6.8% positives among healthy blood donors has not received more
attention from the press or the public. Could this be due in part to
the fact that there are 68 infectious agents (including XMRV) tracked
by the AABB as being potentially problematic for transfusion safety?
A: The August 2009 Supplement to Transfusion described 68 agents that
were known to be transfusion transmitted, or those with the potential
to be transfusion transmitted, all without an effective intervention
(e.g., donor and/or donation screening or methods of
reduction/inactivation). These agents were selected for inclusion by a
group of experts from the AABB Transfusion Transmitted Disease (TTD)
Committee; this group also prioritized the risk of these agents to
cause harm in transfusion recipients in the U.S. and Canada based on a
specific set of criteria. The selection of agents and prioritization
process may have yielded different results if examined by a different
group of experts. The list of 68 agents was published prior to the
findings of Lombardi et al., and thus XMRV or related mouse‐derived
retroviruses were not included (although an XMRV fact sheet has since
been prepared by the TTD Committee). The agents that received the
highest priority for further consideration of an intervention were all
known transfusion transmitted agents causing diseases that are
difficult to treat or untreatable and have high rates of mortality and
morbidity. These included dengue virus, the parasite Babesia and the
prion variant Creutzfeldt–Jakob disease (vCJD). Agents such as
XMRV/MLV would have been prioritized in a different category, versus
the three highest priority agents, in that agents such as XMRV/MLV are
of public concern, but the science as of yet does not establish their
role in any significant transfusion transmitted disease. Sophisticated
molecular techniques have identified many new viral agents for which,
as of yet, have not yet been identified to cause any human disease
even though some of these agents have high rates of prevalence in the
donor population and have been shown to be transfusion‐transmitted.
While the prevalence of up to 6.8% for a known disease‐causing agent
would be alarming, further work is needed to establish the
incidence/prevalence of XMRV/MLV in the donor population, to determine
if in fact the agent(s) is transfusion‐transmitted and the
consequences in transfused recipients. This will involve the
completion of multiple studies that are in various stages of planning
and execution.

As a further note, as referenced in Voisset, et al. (Microbiology and
Molecular Biology Reviews, March 2008) estimates from the genome
sequencing project suggest that endogenous retroviruses (ERVs) now
comprise some 8% of human DNA, representing around 4,000 proviruses
and thousands more solitary long terminal repeats (LTRs). Therefore,
XMRV/MLVs may not be the only group of agents that emerge as related
to, or associated with disease syndromes that lack a known etiologic
cause. Like XMRV/MLVs, any newly recognized agent(s) may, or may not
have any causal relationship to any known disease state. – Susan L.
Stramer, PhD, Executive Scientific Officer, Scientific Support Office,
American Red Cross and Roger Y. Dodd, PhD, Vice President, Research &
Development, American Red Cross
Q: Many of the headlines for media stories reporting on this study
(and others) used the term "chronic fatigue," even if the writer used
"chronic fatigue syndrome" in the article, as was the case with the
article you wrote for the New York Times about the PNAS study. Who
writes headlines?
A: I understand the unhappiness with the use of "chronic fatigue"
instead of the full name. In U.S. mainstream media, reporters
generally don't write the headlines – the copy‐desk does, often hours
after the story has been edited. Newspapers tend not to like too many
acronyms; since it gets repetitive to use "chronic fatigue syndrome"
over and over again, there's a tendency to use "chronic fatigue" as
the shorthand instead of CFS. Different publications have different
style guides, and these should include the point that "chronic
fatigue" is not an appropriate shorthand for "chronic fatigue
syndrome," but these style guides change slowly. – David Tuller,
contributing writer to the New York Times
Q: How many reporters participated in the Aug. 23, 2010 telebriefing
and how many times was the recording accessed during the week it was
available to the public?
A: We had about 70 mainstream and CFS blog and advocacy publications
participate, as well as private organizations, academic institutions,
disease associations and research and investment groups. We are unable
to obtain information about how many times the line was accessed by
the public and/or press while it was live. – Kelli Carrington, MA,
Office of Communications, Patient Recruitment, and Public Liaison, NIH
Clinical Center
Prepared by the CFIDS Association of America

No comments: