- The breast-cancer marker Her2/neu (c-erbB2), trying to make the most use of its prognostic
value - The gene Her2/neu, also known as c-erbB2, has made news in recent years as an
indicator of response to therapy. When overexpressed, that is when increased amounts of the
protein that it codes for are produced, it generally indicates the likelihood of poor response or
treatment failure, but the data have not been as consistent as one would hope. Three articles in
recent issues of different journals deal with Her2/neu.
In the September, 1996 issue of Clinical Cancer Research, doctors at the Lutheran General
Hospital in Park Ridge, Illinois, examined this gene in 25 patients with high-risk breast cancer
(they each had more than 10 positive nodes). These patients received 6 cycles of chemotherapy
with 5-fluorouracil, doxorubicin, and cyclophosphamide, followed by high-dose chemotherapy
with cyclophosphamide and thiotepa, and transplantation with stem cells (cells capable of
proliferating) derived from the patients' own blood (autologous stem cell transplant). There was
an 80% 3-year relapse-free survival, but of the patients relapsing between 6 and 18 months, all
four had overexpression of Her2/neu, whereas none of the patients with borderline or no
Her2/neu expression relapsed (Bitran, Clin Cancer Res 2:1509, 1996).
A second article by researchers at Yale University School of Medicine, published in the
September, 1996 issue of the British Journal of Cancer, looked at whether the increased amount
of Her2/neu protein was actually working protein. When functional, Her2/neu protein is in a
phosphorylated form, that is a phosphate group is inserted. Among 223 invasive or mixed
invasive and in situ breast cancer specimens, Her2/neu was overexpressed in 9%, but only 35%
of these had the phosphorylated active form. Overexpressed Her2/neu specimens tended to be
less frequently positive for estrogen and progestin receptors, but none of the phosphorylated
Her2/neu samples had progestin receptors. The authors of this study are now undertaking a
larger analysis of samples to correlate this with patients' records to determine which of the
various factors best predicts the outcome (DiGiovanna, Br J Cancer 74:802, 1996).
A third report in the October, 1996 issue of the Journal of Clinical Oncology, claimed that
overexpression of the Her2/neu gene is an indicator of lack of response to adjuvant treatment
with tamoxifen in early-stage breast cancer without axillary lymph node involvement. This
study from Naples University, Italy, found that Her2/neu was overexpressed in about 30% of 145
breast cancer patients; these cases also tended to have increased tumor size and reduced estrogen
receptor expression. Adjuvant tamoxifen did not improve disease-free survival and decreased
overall survival of Her2/neu-positive cases, whereas it significantly prolonged both disease-free
and overall survival over a 12 year follow-up period in Her2/neu negative cases. (Carlomagno,
J Clin Oncol 14:2702, 1996)
- Prostate-specific antigen as an indicator for breast cancer prognosis? You have to be
kidding! - So familiar is prostate-specific antigen (PSA) as a marker for prostate cancer, that it
is totally unexpected to read that it could play a similar role for some kinds of breast cancer. Yet
about 30% of breast cancers make PSA, and researchers from Mount Sinai Medical Center in
New York and Dianon Systems in Stratford, Connecticut, used an enhanced PSA test involving
amplifying a fragment of the PSA molecule with the polymerase chain reaction to look for PSA
in the blood of patients with breast cancer. In a report published in the September, 1996 issue of
the British Journal of Cancer, they detected the PSA fragment in 18 of 78 women with breast
cancer. These included 7 women with localized small node-negative tumors, and 22% of those
who had been diagnosed less than 12 months earlier. These results provide further evidence of
how early in its course breast cancer disseminates. The authors considered that this sensitive
test could be applied to find circulating cancer cells very early on in the course of the disease,
and so identify patients requiring additional adjuvant treatment; this has already been done by
other researchers for metastatic prostate cancer cells. (Lehrer, Br J Cancer 74:871, 1996)
- Cyclin E, another prognostic factor for breast cancer - Cyclin E is a protein existing in
several forms which acts as a controlling factor, rather like a traffic light, to permit cells to enter
the division cycle. The gene which codes for it is overexpressed, that is it produces excess of the
active protein product, in a number of cancers, including breast cancer. In the September, 1996
issue of the British Journal of Cancer, researchers from Umea University in Sweden described
levels of cyclin E which varied 200-fold among tumor specimens. They used cyclin E levels in a
standard cell type (BL-42) as a basis for comparison with a cut-off point of 0.5 (half that of BL-42 cells) to indicate high levels in tumor. The risk of death and relapse from breast cancer was
significantly increased in 27 women with high cyclin E levels among 100 women with Stage I-III
disease. This increased risk was seen even in women without positive lymph nodes. Of those
with high cyclin E, 67% were estrogen-receptor negative and 55% negative for progesterone
receptors, other indicators of poor prognosis, . (Nielsen, Br J Cancer 74:874, 1996)
- Guidelines proposed by ASCO for the use of tumor markers in breast cancer - the American
Society of Clinical Oncology (ASCO), an association of over 10,000 oncologists and cancer
researchers, issues annual guidelines regarding the tests for tumor markers that these doctors
believe should be used to assist diagnosis and management of cancer. In the October, 1996 issue
of the Journal of Clinical Oncology, they recommend for breast cancer that estrogen and
progesterone receptors should be measured in every primary specimen, but that it should be
measured subsequently only if it would lead to changes in disease management. They consider
the data to be insufficient at present to recommend routine use of DNA index, DNA flow
cytometry, CA 15-3, carcinoembryonic antigen (CEA), c-erbB-2, p53 or cathepsin D. However,
they suggest that CA 15-3 and CEA levels are useful as evidence of treatment failure when
disease cannot be measured directly. (J Clin Oncol 14:2843, 1996)
Editor's Comment: - These guidelines are very useful, but will be subject to change as new
information becomes available. As discussed under this month's heading "The breast-cancer
marker Her2/neu (c-erbB2), trying to make the most use of its prognostic indications," c-erbB2 (Her2/neu) is increasingly looking like a valuable marker. However, the question is how
much information does it give that is independent of currently-used markers such as estrogen
receptor; additional cost of adding markers to the laboratory work up has to be weighed against
the added benefits. The benefit side of the equation is beginning to look stronger, however.
- No role for multicentric foci of disease in breast cancer recurrence? - Increased use of
conservative surgery in managing breast cancer focuses attention on an issue that was explored
at some depth more than a decade ago - multicentric foci of disease, areas of disease within the
breast that are separate from the primary tumor. Do these foci pose a threat in women who have
limited surgery? The research of Egan (Cancer 49:1123, 1982) and of Holland (Cancer 56:979,
1985) and their colleagues looked at primary tumor and other areas of disease as distributed in
the whole breast removed at mastectomy. Egan's study found that 69% of breasts had such
disease, and Holland found that 53% of these foci were within 2 cm, and 90% within 4 cm of the
primary tumor edge. The implication was that most were in the same quadrant of the breast as
the primary tumor. This looked good for conservative surgery, and more than ten large studies
have since shown that when the cancer recurs after such surgery, it does so in the same quadrant
that had been operated on in 90% of cases.
A study by a group at Tata Memorial Hospital in Bombay, India, took up the issue again by
studying breasts from 30 women who had selected modified radical mastectomy. Their report
appeared in the September, 1996 issue of the British Journal of Cancer. Multicentric foci were
found in 63%, and 53% of these were within 2 cm, 80% within 4 cm, and 90% within 5 cm of
the tumor edge - figures very close to those from other studies. However from a 3-dimensional
viewpoint that took into consideration breast size and contour, rather than simply expressing
distance from the tumor edge, the results were very different. Primary tumor occurred more
frequently in the upper outer quadrant, whereas cancer foci were scattered throughout the breast,
with 79% of those cases with foci, accounting for half of all breasts examined, having them
beyond the 25% of the breast volume (index quadrant) that included the tumor. In view of the
clinical results of conservative surgery, the authors concluded that multicentric foci cannot be
the source of recurrences, which most likely arise mostly from primary tumor left behind or
from circulating tumor cells lodging in the vascular bed of the operated tumor. This agrees with
the fact that the recurrence rate in the other breast quadrants (1% per year) is the same as for the
unoperated breast . (Vaidya, Br J Cancer 74:820, 1996)
Editor's Comment: - This topic should be explored in other studies. It is a timely in that there is
ongoing reevaluation of other aspects of conservative surgery such as the need for radiation.
Breast cancer, unlike some other cancers, has the potential to become disseminated at a very
early stage, which is probably the reason why survival clearly seems better when radiotherapy is
combined with conservative surgery. However, since a high proportion of women with breast
cancer do survive without recurrent disease, most of the cancer cells that are disseminated from
early tumors must not proliferate, presumably for immunological or other biological reasons.
- What is the risk for recurrence of breast cancer? - A biostatistical study used the extensive
data from 10 trials of the Eastern Cooperative Oncology Group to attempt to answer this
question in a report in the October, 1996 of the Journal of Clinical Oncology. Breast cancer is a
disease that may recur long after treatment, and certainly cannot be defined as "cured" based on
5-year survival; this was clearly evident from the data. Peak hazard for recurrence, at an annual
rate of over 13% (20% or more for large tumors and more than 4 positive nodes), was seen 1-2
years after treatment, and the risk then decreased during the 2- to 5-year period. Between 5 and
12 years there was a very slow fall in the hazard (4.7% for years 5-8, 3.4% for years 8-12, and
1.6% for year 11-12), with an overall 4.3% risk of recurrence, but it did not fall to zero. Time
trends were similar for all subsets of patients. Patients with larger tumors or more positive
nodes had a greater hazard for recurrence, as did estrogen receptor-negative patients, because of
higher hazard in the first 5 years. Women who were postmenopausal at treatment had a slightly
higher hazard irrespective of nodes or estrogen receptor status. (Saphner, J Clin Oncol 14:2738,
1996)
- Tamoxifen, a common process at work behind its anticancer and cancer-causing actions? -
The antiestrogen tamoxifen can cause regression of many breast cancers, especially in
combination with chemotherapy. While this is generally agreed to be through some action at the
estrogen receptors, the exact mechanism is still uncertain. At the same time, there is increasing
evidence that tamoxifen can also increase about 7-fold the risk for cancer of the uterine lining,
the endometrium. An article relevant to these questions by a group at the Leon Berard Center in
Lyon, France, appeared in the September, 1996 issue of Clinical Cancer Research. These
researchers looked at the protein produced by the tumor suppressor gene p53, which has been
discussed frequently in prior issues of the CancerWeb Report. They exposed human breast
cancer cell lines to tamoxifen and found that those cells that were positive for the estrogen
receptor showed dramatic decreases in p53 protein levels. At the same time, tamoxifen
prevented the usual increase in p53 protein levels that occurs in response to DNA damage
inflicted by radiation or chemotherapy drugs.
There are two processes controlled by p53 for responding to DNA damage - prevention of cell
division, allowing time for repair to take place before the DNA is replicated, or initiation of
programmed cell death, the suicide-like process of apoptosis. Either process will prevent the
damaged DNA from being copied into new cells which would be likely to become cancerous.
For tumor cells not programmed for cell suicide, loss of p53 will eliminate the restriction on cell
division, which in the presence of chemotherapy drugs that require such division for their action,
will increase the amount of DNA damage and its resulting effects, thus causing more tumor cell
death. On the other hand, loss of the p53 protein function in a normal tissue such as the
endometrium, may allow cells with carcinogen-damaged DNA to accumulate and proliferate,
giving rise to cancer. (Guillot, Clin Cancer Res 2:1439, 1996)
Editor's Comment: - These results are interesting, but we still do not have the whole story about
tamoxifen's actions, both anticancer and cancer-causing. For example, tamoxifen does have
anticancer action against some breast tumors that lack estrogen receptors. Also it is known to
affect the process of new blood vessel formation (angiogenesis) which is essential for tumors to
grow, and it inhibits several enzyme processes that are involved in the signaling system that
initiates cell division. Finally, using very sensitive techniques with labeled drug, tamoxifen has
been found to combine with the DNA of uterine endometrial cells from women with breast
cancer forming what are known as DNA adducts (Hemminki, Cancer Res 56:4374, October 1,
1996). This work was carried out at the Karolinska Institute in Stockholm, Sweden. Such DNA
adducts with other drugs and chemicals are considered to be potential originators of the process
of carcinogenesis. For a related topic see also A new form of tamoxifen with fewer side-effects? in this section of this month's CancerWeb Report.
- A new form of tamoxifen with fewer side-effects? - Tamoxifen is used not only in the
treatment of breast cancer, but also may play a role in preventing its initial appearance or
recurrence in women at risk for breast cancer. However, it has side-effects, most serious of
which is the potential for increasing the risk of endometrial cancer, but which also include
production of hot flashes, depression and problems with the memory, because it is able to enter
the brain. Some experiments in mice and rats, described in the October 1, 1996 issue of Cancer
Research, suggest that a new form of tamoxifen, tamoxifen methiodide (TMI), may lack some of
these problems while still preserving anticancer activity. These researchers at the Weizmann
Institute in Israel and Pharmos Corporation in Alachua, Florida, found that TMI was very poorly
taken up into the brain and had very much less estrogenic action on the uterus compared with
tamoxifen, while still showing anticancer action against mouse breast cancer and preserving its
positive estrogen effects on bone. Much work needs to be done before TMI is ready for clinical
trial, but these results are very promising. (Biegon, Cancer Res 56:4328, 1996)
- How does a progestin such as megestrol cause breast cancer regression in postmenopausal
women? - Treatment of breast cancer relies heavily on the use of hormones and chemicals that
are hormone antagonists or which reduce hormone production. These hormonal approaches are
targeted at estrogens, since growth of a significant proportion of breast tumors, those having
estrogen receptors, is stimulated by estrogens. Megestrol is a widely used hormone that is not an
estrogen but rather a progestin - how does it work? Norwegian researchers report in an article
published in the September, 1996 issue of Clinical Cancer Research that daily oral doses of 40-160 mg of megestrol produced a dose-dependent across the board reduction in hormone
production in postmenopausal breast cancer patients. It was already known that progestins can
suppress the synthesis of hormones by the adrenal gland, and this was seen in the Norwegian
study also, with more than 90% reduction in plasma levels of cortisol and of the androgens
androstenedione and dehydroepiandrosterone. Levels of testosterone were reduced to a
somewhat lesser degree, perhaps because it is also produced by the ovary. In addition, there
were major reductions in plasma levels of the gonadotropins LH and FSH, and reduced levels
together with increased rates of elimination of estrogens. In postmenopausal women, circulating
androgens are the chief source of estrogens which are formed by metabolism in the tissues. It
appears that megestrol, despite being a progestin, is actually acting indirectly as an antiestrogen,
principally by suppressing the function of the adrenal glands. (Lundgren, Clin Cancer Res
2:1515, 1996)
- Slight increase in risk of leukemia in women treated with adjuvant FAC chemotherapy plus
radiation - Researchers at the M.D. Anderson Cancer Center in Houston studied the treatment
history of 14 women who developed leukemia from among 1,474 patients with Stage II or III
breast cancer given adjuvant treatment between 1974 and 1989. They reported the results in the
October, 1996 issue of the Journal of Clinical Oncology. The estimated 10-year risk for all
patients was 1.5%, while for those receiving FAC chemotherapy (5-fluorouracil, doxorubicin
and cyclophosphamide) plus radiation it was 2.5%, compared with only 0.5% for the group
receiving chemotherapy alone, implicating radiation as the cause of the leukemia. There was
also a trend, not statistically significant however, towards increased risk for those getting high-dose cyclophosphamide. (Diamandidou, J Clin Oncol 14:2722, 1996)
- Risk of thromboembolism from adjuvant tamoxifen and chemotherapy for breast cancer -
Thromboembolism has long been associated with cancer, but more recently it has been found
that there is additional risk resulting from some types of hormone treatment and chemotherapy.
The National Cancer Institute of Canada Clinical Trials Group found that among 353 women
receiving tamoxifen plus CMF chemotherapy (cyclophosphamide, methotrexate and 5-fluorouracil) 13.6% experienced one or more thromboembolisms, compared to only 2.6% of 352
women given only tamoxifen. Their results, presented in the October, 1996 issue of the Journal of Clinical Oncology, also showed that thromboembolisms were generally more severe
(including 6 cases with two events) in the combined treatment group, and most occurred while
the patients were receiving chemotherapy. (Pritchard, J Clin Oncol 14:2731, 1996)
[ CancerWeb Home
| Comments
| CancerWeb Report Index ]