Recent development in the diagnosis and management of neurofibromatosis

Full text

Archives
of
Disease
in
Childhood,
1989,
64,
745-749
Current
topic
Recent
developments
in
the
diagnosis
and
management
of
neurofibromatosis
S
M
HUSON
Kennedy-Galton
Centre
for
Clinical
Genetics,
Northwick
Park
Hospital,
Harrow
The
last
decade
has
seen
increasing
awareness
of
the
problems
encountered
in
diagnosis
and
management
of
the
different
forms
of
neurofibromatosis.1
2
Modern
techniques
of
clinical
and
laboratory
investigation
have
been
applied
to
the
disease,
culminating
in
1987
with
the
mapping
of
the
genes
for
type
1
4
and
type
25
6
neurofibromatosis
to
chromosomes
17
and
22
respectively.
In
the
short
term
this
means
closely
linked
DNA
markers
can
be
used
for
prenatal
and
presymptomatic
diagnosis
and
in
the
long
term
the
cloning
of
the
actual
genes
will
be
a
major
step
towards
our
understanding
of
the
pathogenesis
of
and
possible
treatments
for
the
two
main
types
of
neurofibromatosis.
Nomenclature
Riccardi
has
suggested
that
there
may
be
seven
or
more
different
forms
of
neurofibromatosis.2
One
of
the
problems
addressed
by
the
1987
National
Institutes
of
Health
(NIH)
Consensus
Development
Conference
on
Neurofibromatosis
was
that
of
nomenclature
and
classification.7
The
consensus
panel
concluded
that
at
the
present
time
there
is
sufficient
evidence
to
clearly
distinguish
neurofibro-
matosis
type
1
(previously
known
as
von
Reckling-
hausen
or
peripheral
neurofibromatosis)
and
type
2
(bilateral
acoustic
or
central
neurofibromatosis).
The
diagnostic
criteria
for
both
type
1
and
2
agreed
by
the
consensus
panel
are
shown
in
table
1.
The
panel
acknowledged
previously
reported
cases
with
atypical
features
that
would
not
have
type
1
or
type
2
by
these
criteria
but
clearly
have
a
form
of
neurofibromatosis.
It
was
felt
that
at
the
present
time
there
was
insufficient
data
available
to
justify
further
subclassification.
Neurofibromatosis
type
1
Neurofibromatosis
type
1
is
the
commonest
form
of
neurofibromatosis
accounting
for
over
90%
of
all
cases.
A
recent
population
survey
in
south
east
Wales
found
a
prevalence
of
1/4950
(20/105)
of
population,8
although
the
actual
birth
incidence
may
be
as
high
as
1/2500
(40/105).9
10
Inheritance
is
autosomal
dominant
with
approximately
50%
of
cases
representing
new
mutations.9
10
Gene
pene-
trance,
in
offspring
of
individuals
with
well
charac-
tensed
type
1,
is
virtually
100%
by
the
age
of
5
years.1
11
DIAGNOSTIC
FEATURES
The
diagnostic
criteria
for
neurofibromatosis
type
1
are
shown
in
table
1.
The
NIH
Consensus
statement
draws
attention
to
distinguishing
patients
with
McCune-Albright
(polyostotic
fibrous
dysplasia,
irregular
skin
pigmentation
and
sexual
precocity)
Table
1
Diagnostic
criteria
for
neurofibromatosis
type
I
and
type
27
Neuroflbromuatosis
type
1
The
diagnostic
criteria
are
met
in
an
individual
if
two
or
more
of
the
following
are
found:
26
Cafe
au
lait
macules
>5
mm
in
greatest
diameter
in
prepubertal
individuals
and
>15
mm
in
greatest
diameter
in
postpubertal
individuals
A2
Neurofibromas
of
any
type
or
one
plexiform
neurofibroma
Freckling
in
the
axillary
or
inguinal
regions
Optic
glioma
¢2
Lisch
nodules
(iris
hamartomas)
A
distinctive
osseous
lesion
such
as
sphenoid
dysplasia
or
thinning
of
long
bone
cortex
with
or
without
pseudoarthrosis
A
first
degree
relative
(parent,
sibling,
or
offspring)
with
type
1
by
the
above
criteria
Neurofibromatosis
type
2
The
criteria
are
met
by
an
individual
who
has:
(1)
Bilateral
eighth
nerve
masses
seen
with
appropriate
imaging
techniques
(for
example,
computed
tomography
or
magnetic
resonance
imaging)
or
(2)
A
first
degree
relative
with
type
2
and
either:
(a)
unilateral
eighth
nerve
mass
or
(b)
two
of
the
following:
neurofibroma,
meningioma,
glioma,
schwannoma,
juvenile
posterior
subcapsular
lenticular
opacity
745

746
Huson
and
Watson
syndrome
(caf6
au
lait
spots,
dull
intelligence,
and
pulmonary
stenosis)
when
applying
these
criteria.
The
major
defining
features
of
the
disease
are
cafe
au
lait
spots,
peripheral
neurofibromas
and
Lisch
nodules.
Cafd
au
lait
spots
are
the
first
disease
feature
to
develop
and
are
obvious
on
naked
eye
examination
by
the
end
of
the
first
year
of
life
in
most
cases.
In
fair
skinned
people
they
may
be
extremely
pale
and
examination
is
aided
by
the
use
of
a
Wood's
lamp.
The
other
form
of
characteristic
skin
pigmentation
is
axillary
freckling,
seen
in
two
thirds
of
affected
people8
and
which
develops
in
middle
childhood.
Peripheral
neurofibromas,
which
may
be
cutaneous
or
subcutaneous,
begin
to
appear
around
the
onset
of
puberty
in
most
cases;
the
youngest
child
the
author
has
seen
with
these
lesions
was
6
years
old.
Although
Lisch
first
described
pigmented
nodules
of
the
iris
in
neurofibromatosis
type
1
sufferers
in
1937,12
their
exact
frequency
and
diag-
nostic
value
has
only
recently
been
appreciated.13
14
Although
occasionally
visible
to
the
naked
eye
they
are
best
seen
by
slit
lamp
examination.
They
have
a
characteristic
dome
shaped
appearance,
are
usually
brown
in
colour,
although
they
can
be
pale
or
almost
white
in
young
children.
Lisch
nodules
are
present
in
over
90%
of
type
1
sufferers
by
the
age
of
5
years.
There
are
two
minor
features
of
neurofibromatosis
type
1,
which
although
not
specific
enough
for
diagnosis,
are
present
in
a
large
proportion
of
affected
individuals.
These
are
macrocephaly
and
short
stature.
In
the
Welsh
study,
45%
of
those
affected
had
a
head
circumference
at
or
above
the
97th
centile
and
34%
had
heights
at
or
below
the
3rd
centile.8
The
standard
deviation
from
normal
for
height
for
affected
patients
was
-1
20
(1-07)
in
affected
individuals
compared
with
-0-12
(1-02)
in
normal
siblings
(p=0-001).
The
diagnosis
of
neurofibromatosis
type
1
is
usually
straightforward;
the
only
problem
arises
in
assessing
young
children
with
multiple
cafe
au
lait
spots
as
their
only
feature
and
whose
parents
are
unaffected.
Although
type
1
is
likely
the
diagnosis
cannot
be
made
until
other
features
develop,
as
families
have
been
reported
with
cafe
au
lait
spots
alone
segregating
as
a
dominant
disorder.2
15
In
this
context
slit
lamp
examination
to
look
for
Lisch
nodules
can
be
helpful,
as
they
are
usually
present
before
axillary
freckling
or
peripheral
neurofibromas.
COMPLICATIONS
The
complications
of
neurofibromatosis
type
1
can
affect
any
of
the
body
systems,
and
as
their
occurrence
cannot
be
predicted
even
within
families
it
is
this
aspect
of the
disease
that
is
most
distressing.
The
most
frequent
complications
of
type
1
in
the
Welsh
study
are
shown
in
table
2.8
The
other
complications
seen
in
the
study
population
were
delayed
puberty
(2%),
hypsarrhythmia
(1%),
meningioangiomatosis
(1%),
congenital
glaucoma
(1%),
and
bony
defect
of
the
lambdoidal
suture
(1%).
Four
previously
recognised
complications
of
type
1
were
not
seen
in
the
study
population
even
though
these
are
part
of
the
range
of
the
disease:
these
are
sphenoid
wing
dysplasia,
pseudoarthrosis
of
the
radius
or
ulna,
or
both,
arterial
disease
other
than
renal
artery
stenosis,
and
atypical
forms
of
childhood
leukaemia.
It
must
be
assumed
that
their
occurrence
is
rare
in
neurofibromatosis
type
1
sufferers
(
1%).
The
age
range
of
presentation
of
type
1
complica-
tions
(unless
obvious)
is
also
shown
in
table
2,16
and
this
can
be
used
in
planning
patient
follow
up,
both
for
reassurance
(for
example,
the
parents
of
a
child
with
no
complications
at
5
years
can
be
reassured
that
the
child
will
not
develop
a
facial
plexiform
neurofibroma
or
pseudoarthrosis)
and
in
monitoring
for
complications
which
may
yet
occur.
At
the
present
time
there
are
insufficient
data
for
many
of
the
complications
to
define
an
exact
age
range
of
presentation.
This
will
only
become
possible
when
large
cohorts of
children
with
neurofibromatosis
type
1
are
followed
up
from
childhood.
From
such
studies
clinical
features
may
emerge
that
predispose
to
some
complications,
such
as
peripheral
nerve
malignancy,
thus
highlighting
a
subset
of
patients
which
require
closer
monitoring.
MANAGEMENT
One
of
the
problems
with
neurofibromatosis
type
1
is
that
because
the
disease
complications
are
so
varied
patients
may
present
to
many
different
specialists
during
their
life
and
yet
find
no
one
doctor
keeping
an
overview
of
their
disease.
In
the
Welsh
study
only
28
out
of
135
(21%)
of
the
type
1
sufferers
were
being
regularly
reviewed
for
their
disease
and
this
was
because
of
complications
in
16
of
the
28.8
The
remaining
12,
eight
of
whom
were
children,
were
being
monitored
regularly
for
the
development
of
complications.
Only
10
individuals
had
received
formal
genetic
counselling.
Many
parents
had
been
told
the
cafe
au
lait
spots
were
'just
birth
marks'
and
were
unaware
that
subsequent
problems
(for
example,
educational
difficulties,
scoliosis)
were
related.
There
is
therefore
a
need
for
improved
patient
care
in
neurofibromatosis
type
1.
Although
many
complications
are
individually
rare,
their
combined
burden
is
appreciable
and
it
is
likely
that
sufferers
are
not
being
diagnosed
sufficiently
early,
nor
receiving
appropriate
follow
up
and
counselling.

Recent
developments
in
the
diagnosis
and
management
of
neurofibromatosis
747
Table
2
Frequency
of
complications
of
neurofibromatosis
type
I
in
a
population
based
study
in
south
east
Wales.8
16
The
age
range
at
which
these
can
occur
is
also
presented
(unless
obvious)
6
Complication
Frequency
Age
range
of
presentation
(%)
(years)
Plexiform
neurofibromas:
All
lesions
26-7
0-18
Large
lesions
of
head
and
neck
1-2
0-1
Intellectual
handicap:
Severe
0-8
Moderate
2-4
Minimal/learning
difficulties
29-8
Epilepsy:
No
known
cause
4-2
Lifelong*
Secondary
to
disease
complications
3-1
Tumours
of
the
central
nervous
system
15-2-2%
Optic
gliomas:
0-20
Other
tumours
(usually
astrocytomas):
lifelong
Spinal
neurofibromas
2-1
Lifelong
Aqueduct
stenosis
2-1
0-30
Malignancy
with
established
disease
association:
Rhabdomyosarcoma
1-5
Lifelongt
Peripheral
nerve
malignancy
1-5
Lifelong
Scoliosis:
Requiring
surgery
5-2
0-18
Less
severe
6-3
Pseudoarthrosis
of
tibia
and
fibula:
Resulting
in
non-union
and
eventual
amputation
2-1
0-5
Less
severe
forms
1
0-5
Gastrointestinal
neurofibromas
2-1
Lifelong
Renal
artery
stenosis
2-1
Lifelong
Endocrine
tumours
(phaeochromocytomas
and/or
duodenal
carcinoid)
3-1
From
10
years
onwards
'Lifelong'
indicates
cases
have
been
reported
presenting
in
all
age
groups.
tAlthough
'lifelong'
risk
given
there
is
strong
evidence
for
a
subset
of
cases
presenting
with
pelvic
rhabdomyosarcomas
in
early
childhood.17
18
The
NIH
Consensus
Development
Conference
addressed
these
problems
and
recommended
that
neurofibromatosis
type
1
sufferers
should
have
an
annual
clinical
assessment
to
monitor
for
complica-
tions.
It
was
felt
that
screening
investigations,
such
as
cranial
computed
tomography,
were
not
justified
unless
there
were
clinical
indications.
As
many
of
the
disease
complications
develop
in
childhood
it
is
this
author's
present
practise
to
recommend
biannual
review
for
children.
Examination
should
be
geared
to
monitoring
for
development
of
complications,
therefore
paying
particular
attention
to
blood
pres-
sure
measurement
and
to
examination
of
the
spine
and
nervous
system
(particularly
for
signs
of
an
optic
glioma).
As
educational
problems
are
so
frequent
a
preschool
psychological
assessment
seems
appro-
priate.
There
is
no
specific
treatment
for
neurofibroma-
tosis
type
1
and
most
complications
are
managed
in
exactly
the
same
way
as
when
they
occur
in
isolation.
Plexiform
neurofibromas,
which
are
rarely
seen
as
isolated
lesions,
are
one
of
the
most
frequent
complications
and
present
a
difficult
management
problem.
They
present
as
diffuse
ill
defined
swellings
often
with
overlying
skin
hypertrophy
or
pigmenta-
tion.
As
their
margins
are
ill
defined
complete
surgical
removal
is
often
impossible.
Therefore
unless
these
lesions
are
causing
severe
cosmetic
problems
or
localised
bony
overgrowth
conservative
management
is
advisable.
Mast
cells
are
a
major
component
of
both
discrete
peripheral
and
plexiform
neurofibromas.
Riccardi
has
postulated
that
they
may
play
an
important
part
in
the
pathogenesis
of
these
lesions.15
In
a
recent
paper,
Riccardi
describes
possible
control
of
the
growth
of
aggressive
plexiform
neurofibromas
and
neurofibroma
associated
pruritus
by
the
use
of
the
drug
ketotifen,
which
blocks
mast
cell
secretions.19
The
trial
had
an
open
label
protocol,
however,
and
involved
a
small
number
of
patients,
therefore
larger
double
blind
studies
are
necessary
before
firm
conclusions
can
be
drawn.
Families
who
are
keen
to
gain
more
information
about
the
disease
and
meet
other
sufferers
may

748
Huson
benefit
from
being
put
in
contact
with
the
British
neurofibromatosis
patients
association,
LINK
(Lets
increase
neurofibromatosis
knowledge).*
The
mapping
of
the
gene
to
chromosome
17
by
family
linkage
studies
has
laid
open
the
way
to
applying
more
complex
molecular
techniques
to
clone
the
gene
itself.
Clinicians
can
help
in
this
research
by
looking
for
abnormalities
of
chromosome
17
in
unusual
cases-for
example,
with
particularly
severe
intellectual
handicap,
dysmorphic
features,
or
more
than
one
genetic
disease.
It
is
also
now
possible
to
examine
whether
the
mechanism
of
tumour
formation
in
neurofibromatosis
type
1
is
similar
to
that
in
type
2.5
It
is
therefore
helpful
if
a
portion
of
any
tumour
removed
from
patients
with
neurofibromatosis
type
1
can
be
preserved
for
DNA
analysis.
GENETIC
COUNSELLING
Genetic
counselling
is
an
integral
part
of
the
care
of
neurofibromatosis
type
1
sufferers
and
their
families.
The
50%
risk
of
transmission
to
offspring
is
straight-
forward.
It
is
more
difficult
to
convey
the
risk
of
complications
in
offspring.
There
is
a
fine
balance
between
providing
adequate
information
without
causing
unnecessary
alarm.
In
counselling
families
the
author
has
found
it
useful
to
divide
the
complications
into
groups
that
focus
on
the
effects
a
given
complication
would
have
on
an
individual's
life.
There
are
(1)
intellectual
handicap:
moderate
to
severe:
2%,
minimal
intellectual
handicap
or
learning
difficulties:
15%;
(2)
complications
that
develop
in
childhood
causing
lifelong
morbidity
(severe
head
and
neck
plexiform
neurofibromas,
orthopaedic
problems):
4%;
(3)
complications
that
can
occur
at
any
time
but
are
'treatable'
(for
example,
epilepsy,
endocrine
tumours):
8%;
and
(4)
central
nervous
system
and
malignant
tumours:
2-3%.
The
frequencies
for
each
group
are
taken
from
the
Welsh
study
and
have
been
halved
to
account
for
dominant
inheritance
and
rounded
to
the
nearest
whole
number.8
Approximately
50%
of
all
cases
of
neurofibroma-
tosis
type
1
are
new
mutations.
Before
parents
of
an
apparently
isolated
care
are
fully
reassured
as
to
their
own
recurrence
risks,
however,
it
is
important
they
have
a
detailed
cutaneous
and
slit
lamp
examination.
There
have
been
several
cases
re-
ported
where
patients
with
limited
disease
features
have
had
affected
children."t
For
example,
Riccardi
and
Lewis
have
reported
a
case
where
a
mother
of
two
affected
children
had
bilateral
Lisch
nodules
as
her
only
feature.11
*LINK,
the
British
Neurofibromatosis
Patients
Association,
Office
B03,
Surrey
House,
34
Eden
Street,
Kingston
KT1
IER.
Since
the
mapping
of
the
gene
for
neurofibroma-
tosis
type
1
to
chromosome
17
in
1987
there
has
been
rapid
progress.
Linkage
studies
of
chromo-
some
17
markers
in
large
numbers
of
families
have
shown
no
evidence
of
non-allelic
genetic
hetero-
genity-that
is,
there
is
no
evidence
for
a
second
type
1
locus
not
on
chromosome
17.20
There
are
now
several
DNA
markers
that
can
be
used
for
prenatal
diagnosis
with
an
accuracy
of
>95%.21
As
their
clinical
application
will
depend
on
a
'gene
tracking'
approach,22
however,
it
will
be
limited
to
families
with
more
than
one
affected
individual
who
are
'informative'
for
the
marker
polymorphisms.
A
more
widely
applicable
DNA
or
biochemical
marker
for
neurofibromatosis
type
1
will
only
become
available
with
the
cloning
of
the
gene
itself.
Neurofibromatosis
type
2
The
diagnostic
criteria
for
neurofibromatosis
type
2
are
shown
in
table
1.
This
form
of
neurofibromatosis
has
been
recently
comprehensively
reviewed
else-
where.23
Type
2
is
much
rarer
than
type
1
with
an
estimated
prevalence
of
1/50
000,
inheritance
is
also
autosomal
dominant.
The
main
features
of
neuro-
fibromatosis
type
2
are
bilateral
acoustic
neuromas,
which
occur
in
over
95%
of
patients.
Other
tumours
of
the
nervous
system,
particularly
schwannomas
and
meningomas
are
often
associated.
Patients
with
neurofibromatosis
type
2
do,
how-
ever,
have
a
few
cafe
au
lait
spots
(but
always
fewer
than
six)
or
peripheral
neurofibromas
more
fre-
quently
than
the
general
population,24
and
in
the
past
the
two
forms
of
the
disease
were
not
clearly
distinguished.
Recent
large
surveys
of
type
1
suf-
ferers
have
shown
no
increased
frequency
of
acoustic
neuromas.8
13
Neurofibromatosis
type
2
does
not
usually
present
in
childhood:
the
mean
(SE)
age
of
onset
of
symptoms
from
acoustic
neuromas
was
20*4
(1.1)
years
in
one
large
series.24
Cases
have
been
reported
with
an
earlier
presentation,
however,
and
the
diagnosis
should
be
considered
in
children
present-
ing
with
meningeal
or
Schwann
cell
tumours
in
any
location.
In
these
cases
the
finding
of
lens
opacities
on
slit
lamp
examination
would
make
the
diagnosis
of
neurofibromatosis
type
2
a
strong
possibility.23
Genetic
counselling
is
also
important
for
type
2
sufferers
and
their
families.
As
most
gene
carriers
develop
acoustic
neuromas,
screening
for
their
development
is
justified23;
surgery
on
small
tumours
at
an
early
stage
in
their
development
is
associated
with
the
preservation
of
hearing
in
some
cases
and
a
noticeable
decrease
in
other
complications.
A
closely
linked
chromosome
22
DNA
marker
has

Recent
developments
in
the
diagnosis
and
management
of
neurofibromatosis
749
been
identified
for
neurofibromatosis
type
2.25
As
the
linkage
studies
have
only
been
done
in
one
large
family,
however,
heterogeneity,
although
unlikely,
remains
a
possibility
and
the
accuracy
of
the
marker
is
not
yet
sufficiently
defined
for
clinical
application.
Neurofibromatosis
clinics
Several
centres
in
the
United
States
have
now
established
multidisciplinary
neurofibromatosis
clinics,'5
with
a
view
to
improved
coordination
of
care
particularly
for
neurofibromatosis
type
1
suf-
ferers.
There
is
a
need
for
other
countries
to
consider
following
this
model
so
that
expertise
can
be
developed
in
treating
some
of
the
more
difficult
complications
that
are
individually
rare
(for
ex-
ample,
facial
plexiform
neurofibromas),
the
assess-
ment
of
unusual
cases,
and
the
application
and
assessment
of
new
developments.
References
Riccardi
VM.
von
Recklinghausen
neurofibromatosis.
N
Engl
J
Med
1981;305:1617-27.
2
Riccardi
VM.
Neurofibromatosis:
clinical
heterogeneity.
Curr
Probl
Cancer
1982;VII:1-34.
3Barker
V,
Wright
E,
Nguyen
K,
et
al.
Gene
for
von
Reckling-
hausen
neurofibromatosis
is
in
the
pericentromeric
region
of
chromosome
17.
Science
1987;236:1100-2.
4Seizinger
BR,
Rouleau
GA,
Ozelius
LG,
et
al.
Gcnetic
linkage
of
von
Recklinghausen
neurofibromatosis
to
the
nerve
growth
factor
receptor
gene.
Cell
1987;49:589-94.
Seizinger
BR,
Martuza
RL,
Gusella
JF.
Loss
of
genes
on
chromosome
22
in
tumorigenesis
of
human
acoustic
neuroma.
Nature
1986;322:644-7.
6
Rouleau
G,
Seizinger
BR,
Ozelius
LG,
et
al.
Genetic
linkage
analysis
of
bilateral
acoustic
neurofibromatosis
to
a
DNA
marker
on
chromosome
22.
Nature
1987;329:246-8.
National
Institutes
of
Health
Consensus
Development
Con-
ference.
Neurofibromatosis
Conference
Statement.
Arch
Neurol
1988;45:575-8.
Huson
SM,
Harper
PS,
Compston
DAS.
von
Recklinghausen
neurofibromatosis:
a
clinical
and
population
study
in
south
east
Wales.
Brain
1988;111:1355-81.
9Crowe
FW,
Schull
WJ,
Neele
JV.
A
clinical,
pathological
and
genetic
study
of
multiple
neurofibromatosis.
Springfield,
Illinois:
Charles
C
Thomas,
1956.
10
Huson
SM,
Compston
DAS,
Harper
PS.
A
genetic
study
of
von
Recklinghausen
neurofibromatosis
(NF-1)
in
south
east
Wales
1:
prevalence,
fitness,
mutation
rate
and
effect
of
parental
transmission
on
severity.
J
Med
Genet
(in
press).
Riccardi
VM,
Lewis
RA.
Penetrance
of
von
Recklinghausen
neurofibromatosis:
a
distinction
between
predecessors
and
descendants.
Am
J
Hum
Genet
1988;42:284-9.
12
Lisch
K.
Uber
beteiligung
der
augen,
insbesondere
das
vorkom-
men
von
irisknotchen
bei
der
neurofibromatose
(Reckling-
hausen).
Zeitschrift
fur
Augenheilkunde
1937;93:137-43.
13
Lewis
RA,
Riccardi
VM.
von
Recklinghausen
neurofibroma-
tosis.
Incidence
of
iris
hamartomata.
Ophthalmology
1981;88:
348-54.
14
Huson
SM,
Jones
D,
Beck
L.
Ophthalmic
manifestations
of
neurofibromatosis.
Br
J
Ophthalmol
1987;71:235-8.
15
Riccardi
VM,
Eichner
JE.
Neurofibromatosis:
phenotype,
natural
history
and
pathogenesis.
Baltimore:
Johns
Hopkins
University
Press,
1986.
16
Huson
SM,
Compston
DAS,
Harper
PS.
A
genetic
study
of
von
Recklinghausen
neurofibromatosis
in
south
east
Wales
II:
guidelines
for
genetic
counselling.
J
Med
Genet
(in
press).
17
Hartley
AL,
Birch
JM,
Marsden
HB,
Harris
M,
Blair
V.
Neurofibromatosis
in
children
with
soft
tissue
sarcoma.
Pediatric
Hematology
and
Oncology
1988;5:7-16.
18
Hope
DG,
Mullvihill
JJ.
Malignancy
in
neurofibromatosis.
Adv
Neurol
1981;29:33-5.
Riccardi
VM.
Mast-cell
stabilization
to
decrease
neurofibroma
growth.
Arch
Dermatol
1987;123:1011-6.
21)
Skolnick
MH,
Ponder
B,
Seizinger
B.
Linkage
of
NFI
to
12
chromosome
17
markers:
a
summary
of
eight
concurrent
reports.
Genomics
1987;1:382-3.
21
Goldgar
DE,
Green
P,
Parry
DM,
Mulvihill
JJ.
Multipoint
linkage
analysis
in
neurofibromatosis
type
1:
an
international
collaboration.
Am
J
Hum
Genet
1989;44:6-12.
22
Pembrey
M.
Impact
of
molecular
biology
on
clinical
genetics.
Br
Med
J
1987;295:71
1-3.
23
Martuza
RL,
Eldridge
R.
Neurofibromatosis
2
(bilateral
acoustic
neurofibromatosis).
N
Engl
J
Med
1988;318:684-8.
24
Kanter
WR,
Eldridge
R,
Fabricant
R,
Allen
JC,
Koerber
T.
Central
neurofibromatosis
with
bilateral
acoustic
neuroma:
genetic,
clinical
and
biochemical
distinctions
from
peripheral
neurofibromatosis.
Neurology
1980;30:85-9.
25
Wertelecki
W,
Rouleau
GA,
Superneau
DW,
et
al.
Neurofibro-
matosis
2:
clinical
and
DNA
linkage
studies
of
a
large
kindred.
N
Engl
J
Med
1988;319:278-83.
Correspondence
to
Dr
SM
Huson,
Kennedy-Galton
Centre,
Clinical
Research
Centre,
Northwick
Park
Hospital,
Watford
Road,
Harrow,
Middlesex
HAl
3UJ.