Dr Adrian Philbey nswema05 at
Sun Feb 4 16:37:24 EST 1996

K. Weber on Mon 29 Jan 1996, in a posting to virology at entitled
"quote", wrote:

>"We don't know...if this agent is bona fide HTLV-II or, as some of
>the data suggests, different from HTLV-ii, possibly a new strain of
>HTLV-II, or a more distantly related virus that warrants a new name."
>-Elaine DeFreitas, PhD

I wonder if this quote refers to the paper by DeFreitas et al (1991)
Retroviral sequences related to human T-lymphotropic virus type II
in patients with chronic fatigue immune dysfunction syndrome Proc
Natl Acad Sci USA 88:2922-2926 or to related work by this group. It
prompts me to bring up the subject of how to establish the role of a
virus in a disease or syndrome and, in particular, I would like to
open discussion on the possible role of human T lymphotropic virus
(HTLV) types I and II or related viruses in cutaneous T cell lymphomas
(CTCLs). I choose this subject since I have worked on this problem and
because I know that Elaine DeFreitas' group has also been interested
in the role of HTLV and related viruses in CTCL. Despite many efforts
to resolve this issue, it continues to be controversial. In addition
to their association with chronic fatigue syndrome, HTLV-related
viruses have also been associated with multiple sclerosis by the same
group (Koprowski et al 1985), but the latter association has not been
confirmed in many follow-up studies.

Since HTLV-I was associated aetiologically with adult T cell
leukaemia/lymphoma (ATL), it has been possible to distinguish most
cases of CTCL from ATL on clinicopathological grounds, as well as
the presence or absence of evidence of HTLV-I infection. However, a
number of cases described as classical CTCL have been associated with
HTLV-I. Should these cases be considered HTLV-I-associated CTCL or
should they be considered to be forms of ATL, given that ATL has a
broad range of manifestations? In addition, HTLV-II infection has been
identified in several patients with CTCL.

Several studies have further confused the issue by reporting evidence
of HTLV or HTLV-related viruses in patients with CTCL, including
evidence that incomplete or variant HTLV-I is present in these patients.
Following is a brief and incomplete summary of some of the literature
related to this subject to form a basis for further discussion. I will
not include full references, as most people interested in HTLV and CTCL
will be familiar with most of the literature.

Elaine DeFreitas' group established 23 interleukin 2-dependent T cell
lines from 28 patients with CTCL (Abrams et al 1991). At this stage, I
have not seen any further reports from this group suggesting that
HTLV-I, HTLV-II or related viruses have been identified in these or
similarly cultured cells from patients with CTCL. However, I would be
very interested in an update on any new findings from this group if
such information has been published or publicised recently.

Zucker-Franklin et al (1991) identified HTLV-I or an HTLV-related virus
in a high proportion of HTLV-I/II seronegative patients with CTCL
following analysis of cultured peripheral blood mononuclear cells
(PBMCs) by electron microscopy, immunohistochemistry and polymerase
chain reaction (PCR). In an overlapping study, the presence of
incomplete or variant HTLV-I in 4 of 20 CTCL patients was
demonstrated by PCR and HTLV-II was identified in one patient
(Zucker-Franklin et al 1992). In an extension to this study, Pancake and
Zucker-Franklin (1993) detected HTLV-I/II tax sequences in 31/34 and
HTLV-I pol sequences in 29/39 patients with CTCL by PCR in PBMCs.

Manca et al (1994) detected HTLV-I pol and tax sequences by PCR in 10 of
29 patients with CTCL from Italy, confirming these results in a double
blind trial after a 6 month interval. Chan et al (1993) amplified
HTLV-I gag and tax sequences by PCR from 5 of 11 patients with CTCL in
the USA. HTLV-I pol, env and tax sequences were detected by PCR in 1 of
51 patients with CTCL from France (D'Incan et al 1992). In a study of
30 patients with CTCL from Italy, Carlino et al (1992) amplified HTLV-I
tax and pol sequences by PCR from one. Hall et al (1991) detected
HTLV-I sequences by PCR in 5 patients with CTCL from Sweden. In contrast,
no HTLV sequences were detected by PCR in 21 patients with CTCL from
France and Portugal studied by Capesius et al 1991 or 21 patients with
CTCL studied by Lisby et al 1992. A report of a new HTLV-related virus,
designated HTLV-V (Manzari et al 1987), has not been authenticated.

Incomplete or variant HTLV-related viruses have been identified in
patients with CTCL on the basis of sequencing (Hall et al 1991),
variant restriction enzyme patterns on Southern blot analysis (for
example Kiss et al 1993, Whittaker and Luzzato 1993, Bazarbachi et al
1994) or PCR (Hall et al 1991, Srivastava et al 1992, Pancake and
Zucker-Franklin 1993). The basis for declaration of incomplete or
variant HTLV in samples analysed by PCR has been amplification of HTLV-I
or -II sequences from some regions of the genome, whereas other regions
of the genome were negative for specific sequences by PCR.

I would like to open debate on the need to verify reports of HTLV-
related sequences in patients with CTCL. There are problems with
confirmation of HTLV infection by PCR, since contamination could
lead to false positive results. This issue seems to have been overcome
by the use of a double blind method for confirmation of PCR results by
Manca et al (1994). The use of PCR to indicate the presence of incomplete
HTLV-I or -II provirus is also of questionable value, since a negative
PCR result for some but not other regions of the HTLV genome could be due
to contamination for the positive results or poor sensitivity of
particular primer sets for the negative results. In my opinion, to
verify that a patient with CTCL is infected with HTLV or a related
virus, it is necessary to obtain a substantial length of sequence,
preferably of the entire provirus. Sequencing of a fragment derived
from a PCR reaction is inadequate unless the sequence can be shown
to be substantially different from any known HTLV-I or -II sequences
and different from any positive control DNA used for the same study.
Otherwise, even with the use of positive and negative controls,
contamination cannot be excluded with the stringency required for
declaration of a new virus or variant given the level of controversy
in this field. The lack of adequate sequence data is the major flaw of
all studies so far reported that have attempted to associate HTLV
or related viruses with CTCL.

Reports of detection of retrovirus-like particles and reverse
transcriptase activity in cultured lymphocytes from patients with CTCL
also need to be verified by purification and characterisation of the
putative virus, demonstration of infectivity and establishment of a
stock of type virus. A complete sequence of the virus demonstrating its
uniqueness should then follow. None of the studies so far reported
have been able to meet these criteria.

More information about the Virology mailing list