I. Jakobsen et al., DETECTION OF NECROSIS IN HUMAN TUMOR XENOGRAFTS BY PROTON MAGNETIC-RESONANCE-IMAGING, British Journal of Cancer, 71(3), 1995, pp. 456-461
Tumours with necrotic regions have an inadequate blood supply and are
expected to differ from well-vascularised tumours in response to treat
ment. The purpose of the present work was to investigate whether proto
n magnetic resonance imaging (MRI) might be used to detect necrotic re
gions in tumours. MR images and histological sections from individual
tumours of three different amelanotic human melanoma xenograft lines (
BEX-t, HUX-t, SAX-t) were analysed in pairs. MRI was performed at 1.5T
using two spin-echo pulse sequences, one with a repetition time (TR)
of 600 ms and echo times (TEs) of 20, 40, 60 and 80 ms and the other w
ith a TR of 2000 ms and TEs of 20, 40, 60 and 80 ms. Spin-lattice rela
xation time (T-1), spin-spin relaxation time (T-2) and proton density
(N-0) were calculated for each volume element corresponding to a pixel
. Synthetic MR images, pure T-1, T-2 and N-0 images and spin-echo imag
es with chosen values for TR and TE were generated from these data. T-
1, T-2 and N-0 distributions of tumour subregions, corresponding to ne
crotic regions and regions of viable tissue as defined by histological
criteria, were also generated. T-1 and T-2 were significantly shorter
in the necrotic regions than in the regions of viable tissue in all t
umours. These differences were sufficiently large to allow the generat
ion of synthetic spin-echo images showing clear contrast between necro
sis and viable tissue. Maximum contrast was achieved with TRs within t
he range 2800-4000 ms and TEs within the range 160-200 ms. Necrotic ti
ssue could also be distinguished from viable tissue in pure T-1 and T-
2 images. Consequently, the possibility exists that MRI might be used
for detection of necrotic regions in tumours and hence for prediction
of tumour treatment response.