overview of current breast cancer hormonal etielogy

[ttweed@wildrockies.org (Tony Tweedale)]

Hormone Alterations in Breast Cancer: Examining the Hypotheses 
William H. Hindle, MD

Abstract

Many of the epidemiologic risk factors for breast cancer
offer clinicians little help in anticipating who is likely to be
struck with the disease or how to prevent it. There are only a
handful of clinically significant risk factors for breast
cancer. These include being a woman, growing older, already
having breast cancer in 1 breast, and having a first-degree
relative (mother, sister, or daughter) who has been diagnosed
with breast cancer. Most risk factors have a weak
association with breast cancer, occur too infrequently, or
are physiologic events not amenable to intervention. In
recent years, the search for breast cancer causes has led to
the identification of genetic markers that seem to
predispose some women to breast cancer. For decades,
however, researchers have been assessing and analyzing
hormonal changes in the hope of finding a predictable breast
cancer marker or cause that can easily be manipulated to
prevent or more effectively treat the disease. Nearly a dozen
hormonal hypotheses of breast cancer causes have been
proposed -- among them estrogen excess, low luteal-phase
progestational activity, adrenal androgen deficiency, ovarian
androgen excess, melatonin deficiency, prolactin excess, and
thyroid insufficiency. For most, the data are equivocal and
inconclusive. The androgen deficiency hypotheses, however,
may have some bearing on premenopausal breast cancer, and
the ovarian dysfunction hypothesis may have some bearing on
postmenopausal breast cancer. [Medscape Women's Health
4(1), 1999. © 1998 Medscape, Inc.]


How Significant Are Epidemiologic Risk Factors?

One of the more troubling aspects of the disease is how
little researchers and clinicians really know about what
causes it. Although numerous epidemiologic risk factors for
breast cancer have been reported,[2] most of them are not
clinically meaningful, as they involve low levels of relative
risk (less than 2.0). Such low levels of relative risk are
considered "weak associations" even when statistically
significant and may well reflect simple chance.

The clinically significant risk factors for breast cancer are
(1) being a woman, (2) growing older, (3) already having
breast cancer in 1 breast, and (4) having a first-degree
relative (mother, sister, or daughter) who has been diagnosed
with breast cancer. High-dose exposure to radiation of the
anterior chest, biopsy-proven atypical epithelial hyperplasia
(and possibly sclerosing adenosis), and lobular cancer in situ
are clinically significant risk factors, but they are rarely
encountered in clinical practice.

In addition, there are epidemiologic risk factors for breast
cancer involving the reproductive system that are
physiologic events not amenable to therapeutic
interventions. Among the events/factors that raise the risk
of breast cancer are (1) early menarche (before age 12
years), (2) nulliparity, (3) having a first child after age 30,
(4) menstrual history of more than 30 years, and (5) late
menopause (onset after age 55). Published reports of these
risk factors reflect low levels of relative risk (less than
2.0). Postmenopausal obesity is similarly reported to be a
low-level relative risk factor. There are contradictory
reports on the impact of breast-feeding but, at most, not
breast-feeding is a low-level relative risk factor.


Examining Key Hormone Hypotheses

The various hormonal hypotheses of the origin of breast
cancer are of intellectual and historic interest. But how well
do they hold up to close scrutiny? Endocrinologic data from
studies in the 1960s and 1970s may not be technically valid.
The risk of invalid data is particularly high in studies
attempting to analyze urinary hormone concentration. In
recent years, however, the methodology of hormone analysis
has advanced at a rapid pace.

Furthermore, we have increased our understanding that the
actions of hormones (including hormone metabolites) at the
intra- and extracellular levels, over time, are probably
critical to tumor genesis and progression of malignancy. In
addition, we have learned that animal and tissue culture
endocrinologic data, though of interest, may not be
applicable to the human female.


Adrenal Androgen Insufficiency

The adrenal androgen insufficiency hypothesis is supported
by data demonstrating decreased production and low plasma
levels of dehydroisoandrosterone (DHEA) and
dehydroisoandrostene sulfate (DHEAS) in premenopausal
women with breast cancer.[3,4] Although women in whom
breast cancer develops before menopause generally have been
observed to have abnormally low levels of the adrenal
androgens DHEA and DHEAS, the theory of androgen deficiency
as a cause of breast cancer leaves unexplained the
well-documented rise in the incidence of breast cancer that
occurs after menopause, when DHEA and DHEAS tend to
decrease.

Anovulation-Luteal Inadequacy

The anovulation-luteal inadequacy hypothesis was based on
the observation that women with breast cancer tend to have
a high incidence of infertility consistent with anovulation.
Also, women with an anovulatory syndrome, such as
polycystic ovary disease, have a higher than normal
incidence of postmenopausal, but not premenopausal, breast
cancer. The published data, however, are conflicting.[5,6] In
the 1970s, researchers reported low luteal-phase
progesterone levels in women with advanced breast cancer
and normal levels in women with early cancer,[7] while
others[8,9] reported normal levels. Some suggest that luteal
progesterone inadequacy is a sign of ovarian dysfunction
associated with ovarian androgen excess, which is another
hypothesized cause of breast cancer.[10]

Estriol Action

The estriol hypothesis, as proposed in 1969,[11] postulated
that the ratio of estriol to the total of estrone plus estradiol
produced from puberty to the age of 25 years determined
lifetime breast-cancer risk. It was presumed that estriol
acted as an antiestrogen.[12] After the estrogenic effects of
estriol were documented, the originators of this hypothesis
abandoned it.[13]

Estrogen Excess

The estrogen excess hypothesis is based on the observation
that breast cancer developed in experimental animals given
high doses of estrogen. Researchers then attempted to relate
elevated estrogen levels to the origin and progression of
breast cancer in women.[14] However, after initial
enthusiasm and then reports of conflicting findings in
humans, the investigator of the initial report was unable to
reproduce the prior results.[15]

Estrogen Window

The estrogen window hypothesis focused on the 2 times in a
woman's life when anovulation and luteal progesterone
inadequacy are normal -- the early postmenarchial and
premenopausal periods when estrogen was presumed to be
dominant.[16] The theory is that the breast is particularly
vulnerable to carcinogens during these times. The estrogen
window hypothesis was a variant of the anovulation/luteal
deficiency hypothesis. The data cited in support of the
hypothesis are circumstantial: Included in the "evidence" was
the prediction that women who were 10 to 14 years old at
the time of atomic-boom radiation exposure (ie, early
postmenarchial at the present time) would have a higher
incidence of breast cancer than those who were younger or
older at the time of the bomb. In fact, no such increase in
incidence has been observed. The estrogen window has been
criticized and questioned; there has been no recent support
for the theory.

Estrone

The estrone hypothesis evolved from the erroneous
supposition that estrone, produced in excess from androgenic
precursors, particularly in obese women, was carcinogenic
and led to postmenopausal breast cancer.[17] The theory was
based largely on the belief that estrone could cause cancer,
while estriol and estradiol were not carcinogenic. However,
animal studies have shown that estrone, estradiol, and
estriol have similar carcinogenic effects, which led to the
abandonment of this hypothesis.[18]

Melatonin

The melatonin hypothesis is based on the observation that
calcification of the pineal gland, which normally produces
melatonin, has been associated with an increased incidence
of breast cancer.[19] Also, studies of serum melatonin levels
-- which normally rise more than 3-fold, from 20pg/mL in
the morning to 70pg/mL during sleep -- revealed that women
with estrogen-receptor-positive breast tumors had low
melatonin levels in nocturnal plasma.[20] No change was
noted in women with estrogen receptor-negative breast
cancers. The mechanism of the association with breast
cancer is speculative and the data await confirmation by
other studies.

Ovarian Androgen Excess

The ovarian androgen excess hypothesis began with
observations of a higher breast cancer incidence in women
with ovarian dysfunction evidenced by anovulation and low
luteal-phase progestational activity.[21] The theory
proceeded with studies of ovarian androgens related to
atypical endometrial hyperplasia, endometrial
adenocarcinoma, and breast cancer. Increased urine and
serum levels of testosterone have been reported in 17.5% to
60% of women with breast cancer.[22] Although ovarian
androgen production is increased with chronic anovulation
syndrome,[23] further studies using modern technologic
analysis and investigation are needed to clarify the role of
testosterone and other androgens in breast carcinogenesis in
postmenopausal women. If this interesting hypothesis proves
to be valid, it raises the possibility that administering
antiandrogens or inhibiting ovarian androgen production
could lower breast cancer risk. One also wonders whether an
elevated testosterone level at the time of mastectomy would
be an indication to remove the ovaries prophylactically.[3]

Prolactin Excess

The prolactin excess hypothesis was based on animal data
showing that cancer developed in experimental rodents
subjected to excess prolactin.[24] The hypothesis that excess
prolactin can induce cancer, however, has not been
convincingly confirmed by multiple studies in humans.[25]

Interestingly, some researchers[26,27] have reported that
plasma prolactin levels decrease after a full-term pregnancy
and that each additional child adds to the long-term and
permanent decrease in prolactin. Could this explain the
observation that women who have early and multiple
pregnancies seem to be at lower risk for breast cancer?
Zumoff,[3] who reviewed numerous studies that documented
elevated levels of prolactin in women who have breast
cancer, concluded that low levels of prolactin may be
protective against breast cancer, but excess prolactin does
not necessarily increase the risk.

Thyroid Insufficiency

The thyroid insufficiency hypothesis, which was first
proposed in the early 1950s,[28] sought to relate breast
cancer to thyroid dysfunction, particularly hypothyroidism.
One study reported finding high, though still normal, levels
of plasma T3 and T4 levels in some women with breast
cancer.[29] Another study found that thyroid autoantibodies
were increased in women with breast cancer.[30] A review by
the American Thyroid Association concludes that there is
insufficient evidence of any clinical correlation of thyroid
function and breast cancer.[31]


Current Status of Hormone Hypotheses

With subsequent studies using more precise hormonal and
scientific analyses and recent genetic findings, none of
these hypotheses has proven to be a primary cause of
invasive breast cancer in humans. The androgen deficiency
hypotheses, however, may have some bearing on
premenopausal breast cancer, and the ovarian dysfunction
hypothesis may have some bearing on postmenopausal breast
cancer.

What can explain the paradox of the genetic preponderance of
breast cancer occurring in women (as compared to men) and
the increasing incidence of breast cancer as women grow
older, particularly after menopause, when physiologic
estrogen falls to low levels? The differentiation and
maturation of the functioning female breast at puberty may
hold a key.

At puberty, the coordinated and balanced interactions of
estrogen, progesterone, growth hormone, insulin,
adrenocortical steroids, and prolactin upon the ductal
epithelium are essential for the maturation and physiologic
function of the female breast. Estrogen is essential for
stromal, lobular, and alveolar development, growth of the
ducts, and deposition of fat. However, progesterone is also
instrumental to these physiologic changes. Progesterone, as
well as estrogen, is essential for the development of lobular
growth, alveolar budding, and alveolar secretory changes.

None of these physiologic hormonal changes affects the
rudimentary glandular structures in the normal male breast.
Thus, in the human female, estrogen, in its essential role in
the maturation of the ductal and glandular epithelium during
puberty, may concomitantly sensitize the ductal epithelium
to carcinogens. In genetically predisposed women, estrogen
will further increase the probability of invasive breast
cancer developing. Furthermore, during the years when a
woman is having menstrual cycles, the ductal and glandular
epithelium goes through monthly apoptosis (programmed cell
death) and cellular replication. This cyclic mitotic activity
increases the potential for mutations and chromosome
abnormalities.

For lactation to occur, stimulation by prolactin, oxytocin,
cortisol, parathyroid hormone, and growth hormone is
required. Contrarily, estrogen and progesterone are known to
suppress lactation.

The precise role of these complex hormonal interactions is
unknown. However, the profound hormonal changes affecting
the female breast at puberty appear to "sensitize" women to
the development of breast cancer during their later years of
life. If this basic physiologic maturation of the female
breast is the key to carcinogenesis, it seems unlikely that
temporary or transient hormone alterations such as
pregnancy (ie, term delivery, premature delivery, and
spontaneous or induced abortion), lactation, oral
contraceptive therapy,[32] or hormone (estrogen and
progesterone) replacement therapy could play a significant
role in causing breast cancer. Based on review of the
pertinent data, hormone replacement therapy is now a
consideration for estrogen-deficient women who have been
successfully treated for breast cancer.[33,34] Conversely,
there are no published data demonstrating that estrogen
replacement therapy is "safe" for women who have a history
or current evidence of invasive breast cancer. The issue
remains controversial and the debate continues.


Role of Genes in Tumor Suppression

In 1979, the p53 gene was identified and was later mapped
on the short arm of chromosome 17.[35] The p53 gene carries
the code for a p53 protein, which stabilizes the genome in
the cell and suppresses a tumor by blocking cell cycle
progression and/or by promoting apoptosis to kill a cancer
cell.[36]

The p53 gene is called the guardian of the human genome.
Although the gene is involved in more than 52 types of human
cancer, its altered function is observed in fewer than 40% of
breast cancers. The therapeutic effects of irradiation and
tamoxifen are thought to be mediated through increased p53
activity. Intensive cell biology research continues to
investigate the p53 gene and protein in expectation of
discovering a clinically valid therapeutic intervention.

The landmark breakthrough in breast cancer genetic testing
came with the identification and sequencing of the BRCA
genes; their existence was postulated from familial breast
cancer pedigree studies.[37] BRCA-1 is located on the short
arm of chromosome 17 at 17q12-21q, centered on locus
D17S855.[38] BRCA-2 is located on chromosome 13 at
13q12.13, centered on locus D13S60.[39]

The cellular microbiology action of these genes is not clear
but could involve either tumor suppression or genome
stabilization. Although these autosomal dominant genes
appear to be directly involved in the carcinogenesis of two
thirds of familial breast cancer cases, they do not seem to
have a significant role in the etiology of sporadic breast
cancer, which accounts for more than 80% of cases. BRCA-1
is also linked to ovarian carcinoma risk.[40,41]

Commercial testing for BRCA-1 is available. More than 100
breast cancer-related mutations have been identified. The
complete sequencing of the gene by commercial laboratories
can cost several thousand dollars. However, if a few "key"
mutations are identified in a woman with breast cancer, her
relatives can be tested for those mutations at a reduced
price of several hundred dollars. Pre- and posttest
counseling is advised for a woman contemplating breast
cancer genetic testing.[42-44]


Agents That Influence the Growth and Spread of Cancer

Once malignant transformation of breast cancer has
occurred, the growth and spread of the cancer is actively
influenced by 3 distinct types of agents. One type includes
growth factors and their receptors (eg, epidermal growth
factor receptors [EGFR] and HER-2/neu).[45] A primary action
of these receptors may involve the regulation of apoptosis.

A second type of agent is matrix metalloproteinase, a
stromal degrading enzyme that "dissolves" natural
microbiologic barriers (eg, the basement membrane around
the ducts, the surrounding stroma, and the basement
membrane around blood vessels).[46]

The third type is angiogenic factors (eg, basic fibroblast
growth factor and vascular endothelial growth factor,
produced by the cancer).[47] These angiogenic factors
stimulate growth of new blood vessels, which are necessary
for the malignancy to continue growing beyond 1 to 2mm in
size.


Conclusion: No Single Simplistic Etiology

In summary, the concept of a single simplistic etiology of
breast cancer is being abandoned. In its place is the theory
that a complex cascade of events leads to the malignant
transformation of breast glandular cells with the potential
for regional and distant metastasis. A myriad of genetic
mutations, inherited at birth or occurring sporadically
thereafter, can predispose women to the development of
breast cancer.

The interaction of multiple hormones involved in the
physiologic maturation of the breast glandular tissue at
puberty can sensitize the ductal and glandular cells to the
effects of carcinogens. These multiple random causal
effects, when synchronous, can produce malignant
transformation of the cells in a complex stepwise
progression. Then, infinite specific cellular interactions,
mostly enzymatically mediated, can allow a cancer to grow.
When the host's resistance and tumor suppression is
overcome, distant metastases can occur and progress,
leading eventually to death. Each phase of this process
probably involves a multitude of specific extra- and
intracellular changes.

Much is yet unknown.


References for:
Hormone Alterations in Breast Cancer:
Examining the Hypotheses

[Medscape Women's Health 4(1), 1999]

      1.Parker SL, Tong T, Bolden S, et al: Cancer statistics,
          1997. CA Cancer J Clin 47:5-27, 1997.
      2.Kelsey JL: Breast cancer epidemiology: Summary and
          future directions. Epidemiol Rev 15:256-263, 1993.
      3.Zumoff B: Hormonal profiles in women with breast
          cancer. Obstet Gynecol Clin North Am 21:751-772,
          1994.
      4.Zumoff B, Levin J, Rosenfeld RS, et al: Abnormal
          24-hour mean plasma concentration of
          dehydroisoandrosterone and dehydroisoandrostene
          sulfate in women with primary operable breast
          cancer. Cancer Res 41:3360-3363, 1981.
      5.Sherman BM, Wallace RB, Korenman SG: Corpus
          luteum dysfunction and the epidemiology of breast
          cancer: A reconsideration. Breast Cancer Res Treat
          1:287-296, 1981.
      6.Secreto G, Toniolo P, Berrino Fe, et al: Increased
          androgenic activity and breast cancer risk in
          premenopausal women. Cancer Res 44:5902-5905,
          1984.
      7.Swain MC, Bulbrook RD, Hayward JL: Ovulatory
          failure in a normal population and in patients with
          breast cancer. J Obstet Gynaecol Br Commonw
          81(8):640-643, 1974.
      8.England PC, Skinner LG, Cottrell KM, et al: Sex
          hormones in breast disease. Br J Surg
          62(10):806-809, 1975.
      9.Fishman J, Fukushima D, O'Connor J, et al: Plasma
          hormone profiles of young women at risk for
          familial breast cancer. Cancer Res 38(11 Pt
          2):4006-4011, 1978.
    10.Secreto G, Zumoff B: Paradoxical effects associated
          with supranormal urinary testosterone excretion in
          premenopausal women with breast cancer: Increased
          risk of postmastectomy recurrence and higher
          remission rate after ovariectomy. Cancer
          43(7):3408-3411, 1983.
    11.Cole P, MacMahon B: Oestrogen fractions during
          early reproductive life in the aetiology of breast
          cancer. Lancet 1(7595):604-606, 1969.
    12.Lipsett MB: Estrogen profiles and breast cancer.
          Lancet 2:1378, 1971.
    13.MacMahon B, Trichopoulos D, Brown JB, et al: Age at
          menarche, urine estrogens and breast cancer risk.
          Int J Cancer 30:427-431, 1982.
    14.Siiteri PK, Hammond GL, Nisker LA: Increased
          availability of serum estrogens in breast cancer: A
          new hypothesis, in Pike MC, Siiteri PK, Welsch CW
          (eds): Banbury Report No. 8: Hormones and Cancer.
          Cold Spring Harbor, N.Y., Laboratory Press, 1981.
    15.Siiteri PK, Simberg NH, Murai JT: Estrogen and
          breast cancer, in Angeli A, Bradlow HL, Dogliatti L
          (eds): Proceedings of the First International
          Congress of Endocrine Studies in Breast Cancer. Ann
          N Y Acad Sci 464:100-105, 1986.
    16.Korenman SG: Oestrogen window hypothesis of the
          aetiology of breast cancer. Lancet 1:700-701, 1980.
    17.Siiteri PK, Schwarz BE, MacDonald PC: Estrogen
          receptors and the estrone hypothesis in relation to
          endometrial and breast cancer. Gynecol Oncol
          2:228-238, 1974.
    18.Fishman J, Martucci C: Differential biological
          activity of estradiol metabolites. Pediatrics
          62:1128-1133, 1978.
    19.Cohen M, Lippman M, Chabner B: Role of pineal gland
          in aetiology and treatment of breast cancer. Lancet
          2(8094):814-816, 1978.
    20.Tamarkin L, Danforth D, Lichter A, et al: Decreased
          nocturnal plasma melatonin peak in patients with
          estrogen receptor-positive breast cancer. Science
          216:1003-1005, 1982.
    21.Grattarola R: The premenstrual endometrial pattern
          of women with breast cancer. Cancer
          17:1119-1122, 1964.
    22.Secreto G, Fariselli G, Bandieramonte G: Androgen
          excretion in women with a family history of breast
          cancer or with epithelial hyperplasia or cancer of
          the breast. Eur J Cancer Clin Oncol 19(1):5-10,
          1983.
    23.Coulam C, Annegers JF, Kranz JS: Chronic
          anovulation syndrome and associated neoplasia.
          Obstet Gynecol 61:403-407, 1983.
    24.Chan PC, Didato F, Cohen LA: High dietary fat
          elevation of rat serum prolactin and mammary
          cancer. Proc Soc Exp Biol Med 149:133-135, 1975.
    25.Malarkey WB, Schroeder LL, Stevens VC, et al:
          Disordered nocturnal prolactin regulation in women
          with breast cancer. Cancer Res 37:4650-4654,
          1977.
    26.Bruning PF, Bonfrer JMG, Hart AA, et al: Parity and
          age influence hormonal risk factors of breast
          cancer. Progress in Cancer Research and Therapy
          31:335-342, 1984.
    27.Musey VC, Collins DC, Musey PI, et al: Long-term
          effect of a first pregnancy on the secretion of
          prolactin. N Engl J Med 316(5):229-234, 1987.
    28.Repert RW: Breast carcinoma study: Relation of
          thyroid disease and diabetes. J Mich State Med Soc
          51:1315, 1952.
    29.Lemaire M, Baugnet-Mahieu L: Thyroid function in
          women with breast cancer. Eur J Cancer Clin Oncol
          22(3):301-307, 1986.
    30.Rasmusson B, Feldt-Rasmussen U, Hegedus L, et al:
          Thyroid function in patients with breast cancer. Eur
          J Cancer Clin Oncol 23(5):553-556, 1987.
    31.Gorman CA, Becker DV, Greenspan FS, et al: Breast
          cancer and thyroid therapy: A statement by the
          American Thyroid Association. JAMA
          237:1459-1460, 1977.
    32.Collaborative Group on Hormonal Factors in Breast
          Cancer: Breast cancer and hormonal contraceptives:
          Collaborative reanalysis of individual data on
          53,297 women with breast cancer and 100,239
          women without breast cancer from 54
          epidemiological studies. Lancet 347:1713-1727,
          1996.
    33.Cobleigh MA, for the Breast Cancer Committee of
          the Eastern Cooperative Oncology Group: Estrogen
          replacement therapy in breast cancer survivors:
          Time for change. JAMA 272:540-545, 1994.
    34.DiSaia PJ, Grosen EA, Kurosaki T, et al: Hormone
          replacement therapy in breast cancer survivors: A
          cohort study. Am J Obstet Gynecol 174:1494-1498,
          1996.
    35.Levine AJ, Perry ME, Chang A, et al: The 1993 Walter
          Hubert Lecture: The role of the p53
          tumour-suppressor gene in tumorigenesis. Br J
          Cancer 69:409-416, 1994.
    36.Hoffman B, Liebermann DA: Molecular controls of
          apoptosis: Differentiation/growth arrest primary
          response genes, proto-oncogenes, and tumor
          suppresor genes as positive and negative
          modulators. Oncogene 9:1807-1812, 1994.
    37.Ferrell RE, Anderson DE, Chidambaram A, et al: A
          genetic linkage study of familial breast-ovarian
          cancer. Cancer Genet Cytogenet 38:241-248, 1989.
    38.Miki Y, Swensen J, Shattuck-Eidens D, et al: A
          strong candidate for the breast and ovarian cancer
          susceptibility gene BRCA1. Science 266:66-71,
          1994.
    39.Wooster R, Neuhausen SL, Mangion J, et al:
          Localization of a breast cancer susceptibility gene,
          BRCA2 to chromosome 13q12-13. Science
          265:2088-2090, 1994.
    40.Narod SA, Feunteun J, Lynch HT, et al: Familial
          breast-ovarian cancer locus on chromosome
          17q13-23. Lancet 228:82-83, 1991.
    41.Futreal PA, Liu Q, Shattuck-Eldens D, et al: BRCA1
          mutations in primary breast and ovarian
          carcinomas. Science 266:120-122, 1994.
    42.Hoskins KF, Stopfer JE, Calzone KA, et al:
          Assessment and counseling for women with a
          family history of breast cancer: A guide for
          clinicians. JAMA 273:577-585, 1995.
    43.Shattuck-Eidens D, McClure M, Simard J, et al: A
          collaborative survey of 80 mutations in the BRCA1
          breast and ovarian cancer susceptibility gene:
          Implications for presymptomatic testing and
          screening. JAMA 273:535-541, 1995.
    44.Lerman C, Narod S, Schulman K, et al: BRCA1 testing
          in families with hereditary breast-ovarian cancer.
          JAMA 275:1885-1892, 1996.
    45.Rajkumar T, Gullick WJ: The type I growth factor
          receptors in human breast cancer. Breast Cancer Res
          Treat 29:3-9, 1994.
    46.Matrisian LM: The Matrix-degrading
          metalloproteinases. Bioessays 14:455-463, 1992.
    47.Folkman J: How is blood vessel growth regulated in
          normal and neoplastic tissue? Cancer Res
          46:467-473, 1986.
---