Læknaneminn - 01.04.2005, Blaðsíða 69

Verkefni 4. árs læknanema
Dánartíðni eftir heilavefsblæðingar
Þóroddur Ingvarsson’, Örn Smári Arnaldsson2, Albert Páll Sigurðsson3.
'Læknadeild Háskóla Islands, ^taugalækningadeild og 2röntgendeild Landspítala
Háskólasjúkrahúss Fossvogi.
Inngangur: Árlega fá um 100 sjúklingar heilavefsblæðingar
(HVB) á íslandi. Erlendar rannsóknir benda til að aukið blæðing-
arrúmmál, blæðing inn í heilahólf og lág stigun á Glasgow Coma
Scale (GCS) spái best fyrir um dauðsföll hjá fólki eftir HVB.
Tilgangur rannsóknarinnar var að meta dánartíðni eftir HVB á
íslandi og I framhaldinu að kanna hvaða þættir úr heilsufarssögu
sjúklinganna spá fyrir um dauðsföll hjá fólki eftir HVB.
Efniviður og aðferðir: Rannsóknin náði til sjúklinga sem
greindust með HVB á Landspítala Háskólasjúkrahúsi Fossvogi
(LSP-Fossvogi) á tímabilinu 1990-1999. Úr sjúkraskýrslum var
safnað upplýsingum um kyn og aldur, tímasetningu einkenna,
lífsmörk við komu, GCS, áhættuþætti, lyf við komu, blóðpróf og
afdrif. Tölvusneiðmyndir af höfði voru skoðaðar með tilliti til
blæðingarstaðs, stærðar og hvort blóð væri í heilahólfum.
Ætlunin er að búa til gagnagrunn úr þessum upplýsingum og
setja saman við rannsókn sem gerð var á LSP-Hringbraut fyrir
sama tímabil.
Niðurstöður: Alls greindust 243 sjúklingar með HVB, 144 karlar
og 99 konur. Aldur sjúklinga var frá 0-99 ára, meðalaldur 65 ár.
Upplýsingaöflun sem og að setja saman upplýsingar úr
sjúkraskýrslum og tölvusneiðmyndum í gagnagrunn og vinna úr
honum endanlegar niðurstöður.
Ályktanir: Bráðabirgða niðurstöður benda til þess að stærð
blæðingar og blóð í heilahólfum geti spáð fyrir um dauða sjúk-
linga með HVB.
Multicolour imaging of basal-like cells
in bimodal breast cancer
(Svipgerðagreining basalfrumu-
krabbameins í brjóstkirtli)
Erna Gudlaugsdóttir, Rene Villadsen, Ole William Petersen
Institute of Medical Anatomy, The Panum Institute, Copenhagen
Introduction: Breast cancer consists of a number of
histological subtypes with different biological behaviours and
clinical outcomes. Recently this classification has been
reappraised in light of a new technology based on molecular
portraits of gene expression data. Notably one of these
subtypes of breast cancer, the basal-like, has a particularly poor
prognosis. This type of breast cancer was in the original
classification also referred to as bimodal, stem cell-like or high
grade, and it is ER-negative. Another characteristic of these
cancers is that they contain cells expressing the markers p63
and keratin 5 and 17. However, whereas each and individually
these markers have been associated with basal-like breast
cancer, it is currently not known whether their cellular
expression is coordinate as part of a more elaborate
differentiation program reminiscent of that seen in the normal
myoepithelial/basal lineage.
Material, methods and discussion: In order to test this, I
immunoperoxidase-stained 26 breast cancers for the estrogen
receptor to define the ER-negative tumors. Next, I
immunoperoxidase-stained these tumors with keratin K5, and
K17 to further narrow down the basal-like subtype. Three out of
26 (12%) tumors were basal-like. These were selected for
multicolor imaging by use of confocal microscopy. Our data
showed that p63, K5 and K17 were indeed coordinately
expressed within on and the same cell type. Since these
markers decorate both cells of the myoepithelial lineage and
metaplastic cells along the epidermal lineage, we further stained
the biopsies with keratin K10 specific of squamous metaplasia.
Clearly, this staining revealed that essentially none of the basal-
like cells or even other parts of the basal-like carcinomas
showed signs of bona fide squamous metaplasia. I therefore
conclude that the basal-like differentiation observed in a subset
of breast cancers is reminiscent of the myoepithelial lineage in
the normal breast of origin.
Since basal-like breast cancer is a very aggressive tumor
subtype with a poor prognosis we wished to analyse a possible
role of basal-like cells in this behaviour. Metastasis of epithelial
cells is believed to occur via epithelial-mesenchymal transition
(EMT) as identified in situ by elongated single cells within the
stroma. We characterized these cells as compared to cells
within tumor cell nests. Accordingly, both were found to express
K17 and K5. However, whereas cells within tumor cell nests
readily stained for p63, this staining was sometimes absent in
the EMT-derived single cells. Since p63 presumably serves to
maintain cells in at a relative low level of differentiation, we
speculated that the EMT-derived cells most likely were
functionally more elaborate than the nest-associated relatives.
Indeed, migration has been considered a highly specialized
property which correlates almost inversely with growth. If this is
correct, and basal-like cells are indeed the source of specialised
migrating cells, then basal-like cells in general should not cycle.
This was born out in a dual labelling experiment including Ki-67
and P63.
Conclusion: I therefore propose that basal-like cells in human
breast cancer are unique cells with a partial myoepithelial
differentiation program and with a candidate function as highly
invasive, non-cycling cells to some extend responsible for the
aggresive behaviour of this particular tumor subtype. Currently,
to substantiate this I will try to define a hierarchy of differentiation
within basal-like breast cancer cell lines and to correlate this
with migratory behaviour in culture.
LÆKNANEMINN
2005
67