Biomedical Engineering Reference
In-Depth Information
numerical analysis to examine the link between the tumor surface structural pat-
tern, measured by its fractal dimensions, and the spatiotemporal expansion ve-
locity. In particular, we found a positive correlation between these two
measures, i.e., higher fracticality of the tumor's surface corresponds to an accel-
erating spatial expansion. Finally, using a truly multiscale model (9), we found
that biopsy specimens containing all available tumor phenotypes (proliferating,
migrating, and quiescent cells) are of more predictive value than separate gene-
expression profiling for each distinct phenotype. Our multiscale model also con-
firmed that it is the invasive phenotype and not the proliferative one that drives
the tumor system's spatial expansion, as indicated by the strong correlations be-
tween the tumor diameter and the gene expression profiles of migrating tumor
cells.
Most importantly, several of these simulation results have already been cor-
roborated by experimental and clinical findings:
&
For example, in (6), we showed that at higher rates of nutrient de-
pletion, tumor cells can not only exhibit maximum spatial expan-
sion velocity under specific circumstances, but actually survive
longer, hence exhibiting a lower apoptosis rate. Interestingly, this
result is supported by recent findings from (12), which reported for
migratory brain tumor cells in vitro not only an increase in expres-
sion of genes implicated in cell motility but also, concomitantly, a
decrease in the expression of apoptosis-related genes.
&
In (7), we reported that introducing a modicum of randomness in
signal processing can actually improve the performance of the tu-
mor. At the same time, however, we also found that an increasing
invasive potential appears to require a higher search precision of
the cells in order to reach maximum velocity. This emergent prop-
erty indicates a more significant role for cell-surface receptor
mechanisms in the invading cell population of more aggressive
neoplasms. This concept has been supported by a recent in situ hy-
bridization (FISH) study investigating EGF-R expression in the in-
filtrative zone of a human glioblastoma specimen (53).
&
In yet another
clinical
study supporting our findings (54), it has
been revealed that there is no apparent relationship between tumor
size and survival of the host patient. Indeed, these authors found
significant statistical correlations between the tumor's volumetric
surface and survival time, thus substantiating their claim that a bet-
ter measure of the tumor's invasive capability is its surface condi-
tions rather than its entire size. Their results thus corroborated our
work in (8), which found no significant correlations between tumor
