Apoptosis

Carcinogenesis vol.21 no.3 pp.485–495, 2000

Apoptosis in cancer

Scott W.Lowe1 and Athena W.Lin
Cold Spring Harbor Laboratory, 1 Bungtown Road, PO Box 100,
Cold Spring Harbor, New York, NY 11724, USA
1To

whom correspondence should be addressed
E-mail: lowe@cshl.org

In the last decade, basic cancer research has produced
remarkable advances in our understanding of cancer biology andcancer genetics. Among the most important of
these advances is the realization that apoptosis and the
genes that control it have a profound effect on the malignant
phenotype. For example, it is now clear that some oncogenic
mutations disrupt apoptosis, leading to tumor initiation,
progression or metastasis. Conversely, compelling evidence
indicates that other oncogenic changes promoteapoptosis,
thereby producing selective pressure to override apoptosis
during multistage carcinogenesis. Finally, it is now well
documented that most cytotoxic anticancer agents induce
apoptosis, raising the intriguing possibility that defects in
apoptotic programs contribute to treatment failure.
Because the same mutations that suppress apoptosis during
tumor development also reduce treatmentsensitivity,
apoptosis provides a conceptual framework to link cancer
genetics with cancer therapy. An intense research effort is
uncovering the underlying mechanisms of apoptosis such
that, in the next decade, one envisions that this information
will produce new strategies to exploit apoptosis for therapeutic benefit.
Introduction
Apoptosis was initially described by its morphologicalcharacteristics, including cell shrinkage, membrane blebbing, chromatin condensation and nuclear fragmentation (1–3). The
realization that apoptosis is a gene-directed program has had
profound implications for our understanding of developmental
biology and tissue homeostasis, for it implies that cell numbers
can be regulated by factors that influence cell survival as well
as those that controlproliferation and differentiation. Moreover,
the genetic basis for apoptosis implies that cell death, like any
other metabolic or developmental program, can be disrupted
by mutation. In fact, defects in apoptotic pathways are now
thought to contribute to a number of human diseases, ranging
from neurodegenerative disorders to malignancy (4).
The notion that apoptosis might influence the malignantphenotype goes back to the early 1970s. Kinetic studies of
tumor growth implied that cell loss from tumors could be
massive; indeed, observed tumor growth rates could be 5%
of that predicted from proliferation measurements alone (1,2).
In principle, changes in this ‘cell loss factor’ could have a
major impact on tumor growth or regression. Although most
Abbreviations: IGF, insulin-like growthfactor; NF-κB, nuclear factor κB;
PTP, permeability transition pore; TNF-α, tumor necrosis factor α; TRAF-2,
TNF receptor-associated factor.
© Oxford University Press

of this death was assumed to arise from necrosis—a catastrophic (and easily discernible) type of cell death—Kerr et al.
raised the possibility that a large percentage of cell loss from
tumors was due to apoptosis (1).Subsequent studies revealed
a high frequency of apoptosis in spontaneously regressing
tumors and in tumors treated with cytotoxic anticancer agents
(3). Together, these observations suggested that apoptosis
contributed to the high rate of cell loss in malignant tumors
and, moreover, could promote tumor progression. Nevertheless,
the importance of apoptosis in cancer remained under-appreciated for15 years.
Apoptosis and tumorigenesis
The cloning and characterization of the bcl-2 oncogene established the importance of apoptosis in tumor development. bcl2 was first identified at the chromosomal breakpoint of t(14;
18) in a human leukemia line and was later shown to be a
common event in follicular lymphoma (5,6). At this time,
oncogenes were classified as either ‘transforming’ or...