inhibiting proliferation and inflammation, and inducing “cancer cell”
differentiate into hepatocytes at the same time. There are several
reasons for recommending this combined therapy to treat HCC.
Firstl, each strategy has been successfully used to cure cancer respec-
tively [77–81]. Proliferation and inflammation-related proteins have
been approved by the US Food and Drug Administration (FDA) to
treat certain human cancer [77, 81]. Liver-specific transcription fac-
tors are overexpressed to induce HCC differentiation has shown an
anti-tumor effect [79, 80]. However, the separate treatment strate-
giesmayresult in dilemma such as drugresistance[78], or just cannot
obtain favorable results in clinical practice [80]. Our predication here
reveals that we should use combined therapies, in particular we need
to induce HCC differentiation. Second, similar strategies have been
successfully used in a subtype of APL to make it from highly fatal to
highly curable [82], our predication suggests that these concepts
should be used in solid tumor. Third, clinical cases have indepen-
dently reported the spontaneous regression of HCC, some sponta-
neous regression even reported without specific treatment and
evidence of recurrence [83]; however, the spontaneous regression is
a rare event. The predicated combined strategies here may make rare
HCC spontaneous regression to a more probable phenomenon.
A second characteristic qualitative property of dynamical systems
is hysteresis, in the present context which means that the genesis and
regression is asymmetric. Expectedly, the model reveals that inducing
a normal liver attractor to a HCC attractor may be achieved by
activating the proliferative related feedback loops, RTKs/Ras, Akt,
ERK,β-catenin/Myc, HIF and inflammation related feedback loops,
TNF-α, IL-1, IL-6/NF-κB (Kuppfer cell). However, we cannot
induce HCC to normal liver by inhibiting the two positive feedback
loops, we still need to activate liver-specific positive feedbacks loop
HNF4α/C/EBPα/Foxa2. The asymmetric genesis and regression
of HCC is also due to the positive feedbacks loops which may be
activated by consistent activation of one agent in the loop; however,
inactivation of one agent in the loops often cannot inactivate the
positive feedback loops. This mechanism may explain why many
tumors will relapse and acquire drug resistance [20, 84].3 Discussion
3.1 Endogenous
Molecular-Cellular
Network Hypothesis
and Genetic Mutation
Hypothesis
The proposed cancer endogenous molecular-cellular network
hypothesis incorporates both genetic information and the bio-
chemical reactions among endogenous molecular-cellular agents
beyond the genomic information [26, 27]. On the genetic infor-
mation level, the present hypothesis incorporates the hypothesis
that cancer is a genetic disease. The gain and loss of functions via
genetic mutations can be simply represented by removing and
adding molecular-cellular agents or interactions in the endogenous236 Gaowei Wang et al.