Endocrine and paracrine hormones in the promotion, progression and recurrence of breast cancer.

Both normal and neoplastic breast tissues are stimulated by endocrine and paracrine hormones. Epidemiological studies have demonstrated the significant role that hormones, growth factors and cytokines have in the promotion, progression and recurrence of breast cancer. Significant variations in the hormonal environment occur based on age, the cyclical changes occurring during the menstrual cycle and (mammographically determined) variations in breast composition. These variations have a significant influence on rates of local recurrence of breast cancer and survival. This review analyses data relevant to these issues and suggests means by which operative results may be improved.

Pre-, peri-, and postoperative chemotherapy for breast cancer: is one better than the other?

The purpose of the present study was to determine the relative efficacy of pre-, peri-, and postoperative chemotherapy in the prevention of breast cancer relapse and prolongation of host survival. The studies were performed using an experimental mouse breast cancer model. TA3Ha mouse mammary adenocarcinoma was transplanted into the mammary fat pad of syngeneic mice to obtain tumors in their natural organ. The tumors were surgically excised with a "curative" intent. A single treatment with 10 mg/kg doxorubicin was given intravenously pre-, peri-, or postoperatively. Among 74 mice whose tumors were resected but no doxorubicin was given, local recurrence, axillary metastasis, and lung metastasis were seen in 43%, 37%, and 16% of the mice, respectively. Seventeen (23%) mice had no evidence of disease. Doxorubicin given 4 days preoperatively reduced the rate of growth of primary tumor. Local recurrence was reduced in these mice by 30% and metastasis to the axillary lymph nodes and lung was completely prevented. Disease-free survival was increased to 70% (P < 0.01). Similar beneficial effects were obtained when chemotherapy was administered 2 days prior to surgery. The peri-operative chemotherapy group showed 8% (2/26) local recurrence, 4% axillary metastasis, and 0% lung metastasis. Proportion of mice without any evidence of disease increased to 92% (P < 0.00001). Chemotherapy given 4 days postoperatively resulted in 63% (10/16) local recurrence, 38% axillary metastasis, and 6.3% lung metastasis. Only 38% of the mice were disease-free. Thus in the model studied, perioperative chemotherapy offers the best chance for reduced recurrence and for improved disease-free survival.

Role of cytokines and growth factors in promoting the local recurrence of breast cancer.

The pathogenesis of local recurrence in breast cancer is not well understood. Breast-conserving surgery is particularly prone to local recurrence as it leaves behind breast tissue that may harbour occult cancer, and lends itself to enhanced intraoperative shedding of cancer cells due to narrower resection margins and transection of lymphatic channels. A review of clinical breast cancer studies as well as experimental research strongly suggests that these persisting cancerous cells are unlikely to develop into clinically evident disease if their environment remains unstimulated. However, an inordinately high local recurrence rate occurs at the surgical scar, and such recurrence must be triggered by the release of growth factors and cytokines into the healing wound. These factors can stimulate any available cancer cells which express the proper growth factor receptors. Perioperative strategies to neutralize this tumour cell-growth factor interaction should maximize local control.

Influence of smoking on the development of lung metastases from breast cancer.

BACKGROUND: This study examined the association between cigarette smoking status and the development of lung metastases in a group of 835 women diagnosed with primary malignant unilateral breast cancer. METHOD: Female patients with breast cancer diagnosed between 1982 and 1991 at Roswell Park Cancer Institute (RPCI) in Buffalo, New York, who provided information on their cigarette smoking history at the time of their diagnosis were included. The subsequent disease status of patients was monitored by the RPCI Tumor Registry. The Cox regression model was used to estimate the relationship between smoking status and the development of lung metastases, adjusting for the patient's age, stage of disease at diagnosis, and body weight. RESULTS: Of those patients who developed lung metastases, 8.7% were nonsmokers, 14.1% were former smokers, and 14.3% were current smokers. Tests showed that nonsmokers had significantly fewer lung metastases than either of the two smoking groups (P < 0.01). The estimated relative rates of lung metastases developing adjusting for age, stage, and body weight in women who smoked less than 10,000, between 10,001 and 20,000, and more than 20,000 packs over their lifetimes compared with nonsmokers were 1.06 (95% CI, 0.51-2.20), 3.10 (95% CI, 1.5-6.3), and 3.73 (95% CI, 1.6-8.9) respectively. The Cox regression model showed that every 1000 packs of cigarettes consumed over a lifetime increased a woman's risk of developing lung metastases by about 3% to 7% (P < 0.001). CONCLUSION: This study found a significant association between cigarette smoking history and risk of lung metastases developing in women diagnosed with primary invasive unilateral breast cancer. The risk of lung metastases developing increased as the number of cigarettes smoked in a lifetime increased.

T/Tn antigen vaccine is effective and safe in preventing recurrence of advanced breast carcinoma.

Since 1974, and as of March, 1993, we have used T/Tn antigen vaccine in safe, specific, effective, long-term intradermal vaccination against recurrence of advanced breast carcinoma (CA). Staging is by the pathologic TNM system. Treatment is ad infinitum. Of 19 consecutive breast carcinoma patients vaccinated, six Stage IV, six Stage III, and seven Stage II all survived > 5 years postoperatively. Three Stage III, three Stage IV, and five Stage II patients (i.e., 11) survived > 10 to > 18 years. Five others are alive but have not reached 10 years; three of them have no evidence of disease (NED). Three patients died of CA before reaching 10 years. An additional three breast CA patients are being treated for > 2 years, but, < 5 years postoperatively, they are NED. The vaccination are presented as a delayed-type hypersensitivity reaction with significant inflammation with increase of helper T lymphocytes and decrease of T suppressor/cytotoxic cell ratio.

Growth and metastasis of human breast cancers in athymic nude mice.

To evaluate critically the merit of utilizing a wound model for growing human tumors, a series of increasingly difficult human tumor types were tested for growth at sites of trauma in athymic nude mice. In vitro tumor lines as well as fresh tumors from the breast, colon, rectum, lung, and a metastasis from an unknown primary were intraperitoneally injected into mice subjected to intra-abdominal organ injury. Successful xenografts were obtained from nine of 10 cell lines and 14 of 24 fresh tumors. The latter included five of six (83%) colon cancers, one lung tumor, metastatic tumor of unknown primary, three of four (75%) metastatic breast cancers and four of six (67%) estrogen receptor (ER)-negative breast primary tumors. Six ER-positive breast tumors tested failed to grow in mice without estrogen supplementation. Xenografts from two breast, two colon and the lung cancers formed spontaneous metastases and all xenografts tested were able to yield serial transplants in the surgical wound model. Histologically, all xenografts and their metastases were identical to their respective donor tumors. Transplantability in mice without exogenous estrogen supplementation was linked to the absence of estrogen and progesterone receptors in breast tumors. Transplantability of the cell lines was associated with the expression of cell surface receptors for fibronectin and hyaluronic acid. Receptors for other extracellular matrix components, namely, laminin, vitronectin, collagen, fibrinogen or von Willebrand factor were not associated with transplantability. These results demonstrate that a large proportion of human tumors, including the breast tumors, can be successfully xenografted into athymic mice by providing them with a healing wound environment, and that such xenografts grown at ectopic sites exhibit metastatic ability.

Effect of trauma on implantation of metastatic tumor in bone in mice.

We studied the influence of surgical trauma to the iliac bone on the implantation of I.V. injected tumor cells, which formed tumor in the surgical wounds of 27/84 mice (32%). None of these mice or nonsurgical mice developed tumor in the opposite or uninjured pelvic bone (P < 0.0001). When different numbers (10(5), 5 x 10(5), and 10 x 10(5)) of TA3Ha cells were injected I.V. immediately after surgery, the frequency of tumor formation showed an increase (respectively, 32%, 63%, 71%). As the interval between induction of trauma and tumor cell injection was increased from 0 to 15 days, the frequency of tumor formation declined from 32% to 0%. These results suggest that the healing wound is a privileged site for experimental metastasis, particularly in the early stages. It is likely that the proteins in the blood clotting cascade are involved in local tumor implantation.

T/Tn antigen vaccine is effective and safe in preventing recurrence of advanced human breast carcinoma.

For nearly 20 yrs, we used T/Tn antigen vaccine in safe, specific, effective, long-term intradermal vaccination against recurrence of advanced breast carcinoma. Treatment is ad infinitum. All 18 breast carcinoma patients treated, pTNM Stages IV (6), III (6), and II (6), survived > 5 yrs postoperatively; 10 survived > 10 to > 18 yrs; of the latter, three patients each are Stages III and IV. Five additional 5 yr survivors have not yet reached 10 yrs. The probability that our survival results are due to chance, with NCI "1991 Standard PDQ Data" as control, for all three stages taken together is: 5-yr survival: p < 1 x 10(-8); 10-yr survival: p < 1 x 10(-5). There were no untoward side effects. The vaccination area presented as a delayed-type hypersensitivity reaction, but at variance with the PPD reaction, with significant inflammation, increase of helper T lymphocytes and decrease of the T suppressor/cytotoxic cell ratio.

The role of fibronectin in tumor implantation at surgical sites.

Fibronectins are a family of glycoproteins with modular functional domains. They mediate cell-cell and cell-matrix interactions which are important in embryogenesis, wound healing, metastasis and other processes. We present data on the influence of fibronectin on wound implantation of a murine mammary carcinoma line, TA3Ha. Fibronectin used in these studies was derived from bovine plasma, human serum, human foreskin fibroblasts, and mouse embryo cultures. TA3Ha cells rarely form tumors in the liver of syngeneic mice when injected intravenously but after hepatic wedge resection, 45% (107/240) of the mice develop tumors in the hepatic wound. Wound implantation is markedly reduced when the cells are pre-exposed to 200 micrograms/ml bovine plasma fibronectin (13%, P = 0.007), human serum fibronectin (0%, P = 0.02), human cellular fibronectin (0%, P = 0.02), or mouse cellular fibronectin (0%, P = 0.04). Lung colonization is also reduced by these fibronectins. These effects are not due to a cytotoxic action of fibronectin, since intraperitoneally injected fibronectin-treated cells form ascites tumor as effectively as do control untreated cells. Local application of a solution containing 0.25 mg/ml mouse cellular fibronectin to the hepatic wound reduces the frequency of tumor implantation from 45% to 5% (1/21, P = 0.001). No tumor implantation inhibition is seen when only suspending medium or albumin in suspending medium is used. The mechanism by which topical application of fibronectin reduces hepatic wound implantation of tumor cells is unclear, but this finding raises an exciting possibility of preventing local recurrence of cancer.

Inhibition of tumor implantation at sites of trauma by Arg-Gly-Asp containing proteins and peptides.

We report on the inhibition of wound implantation by TA3Ha mammary carcinoma cells by Arg-Gly-Asp containing proteins and peptides using a hepatic wedge resection model. Intravenously injected TA3Ha cells rarely form tumor in the liver of syngeneic mice, but after hepatic wedge resection, 45% (107/240) of the mice develop tumors in the hepatic wound. Hepatic wound implantation is significantly (P = 0.01) inhibited by pretreating the cells with whole mouse plasma, but not with fibrinogen-depleted plasma or serum. Tumor inhibition is also achieved by pretreatment of cells with fibrinogen (P = 0.05-0.0004), fibronectin (P = 0.007) and laminin, but not by albumin. The active domain appears to be the RGDS sequence since the deca- and tetrapeptides containing RGDS inhibit wound implantation (P less than 0.05). However, the tetrapeptide Arg-Gly-Glu-Ser has no such activity. None of these agents affects ascites tumor formation by the intraperitoneally injected cells, suggesting that anchorage independent growth of cells is not affected. We propose that proteins and peptides containing RGD occupy the binding sites and prevent the cells from interacting with cell adhesion proteins in healing wounds. Proteins and/or peptides containing RGD may be useful for preventing local recurrence in postsurgical cancer patients.

Inhibition of tumor implantation at sites of trauma by plasminogen activators.

The authors report on the influence of plasminogen activators (PA) on implantation of TA3Ha mammary tumor cells in the healing hepatic wounds of syngeneic strain A mice. Intravenously injected TA3Ha cells, although they rarely metastasize to the liver, formed tumors in the hepatic wounds of a significant percent (42%, P less than 0.0001) of mice. The frequency of tumor formation declined as the interval between surgery and tumor cell inoculation was increased. Furthermore, preexposure of cells to fibrinogen, fibronectin, laminin, or peptides containing the arginine-glycine-aspartic acid-serine residues dramatically reduced the frequency of tumor formation in the hepatic wounds. These results indicate that TA3Ha cells interact with fibrinogen-related proteins in the wound to aid their attachment and growth. Because these proteins are susceptible to digestion by plasmin, PA were used in this study to examine whether administration of these drugs to the mice would modulate tumor formation in the liver wounds. Among the PA tested, human plasmin B-chain-streptokinase complex (B-SK) and recombinant tissue plasminogen activator (t-PA) inhibited tumor implantation in a dose-related manner. Administration of 900 units (U) of B-SK or 3300 U of t-PA per mouse reduced the frequency of tumor formation from 42% to 0% (P = 0.02) and 11% (P = 0.02), respectively. The B-SK was complexed with p-nitrophenyl-p-guanidinobenzoate; it did not activate the plasminogen or inhibit tumor formation in the hepatic wounds. Although urokinase activated the plasminogen, it did not inhibit tumor implantation in the hepatic wound. Heparin, an anticoagulant that prevents conversion of fibrinogen to fibrin without being fibrinolytic, had no influence on tumor formation in the hepatic wounds. The PA can generate plasmin that digests the cell attachment proteins in wounds and consequently inhibits tumor cell attachment.

The role of reconstruction in breast cancer.

Lumpectomy and irradiation is an increasingly popular treatment option for women with breast cancer. Currently, modified radical mastectomy is selected about as often. Most women choosing to have a mastectomy desire reconstruction to improve the cosmetic result; this procedure can be done immediately or at a later time. If immediate reconstruction using a plastic implant is selected, an expander can be used if the skin cover is inadequate. A myocutaneous flap can be used either immediately or later and whether sufficient skin is available. The trans rectus abdominal myocutaneous (TRAM) flap is increasingly popular because a large amount of tissue can be transferred and the cosmetic blemish from tissue transfer is preferable on the lower abdomen than elsewhere. Treatment costs are of some importance. The cheapest procedure is the modified radical mastectomy; lumpectomy with iridium implant is the most expensive, and the other treatments fall in between.

An alternative approach to nonpalpable breast biopsies.

Traditionally, when a negative specimen radiograph is obtained during biopsy of a nonpalpable breast lesion, immediate re-excision is performed in an attempt to successfully remove the lesion. Based on a retrospective study of the biopsy results of 792 nonpalpable breast lesions, the authors suggest delaying the re-excision, despite a negative specimen x-ray, until postoperative mammography confirms the persistence of the lesion. Utilization of this approach was associated with a comparably low incidence of missed lesions (3%) and had the added advantages of preserving breast tissue and decreasing operative time.

Sodium-24 studies in postmastectomy lymphedema.

Seven patients were studied with 24Na to determine the relative disappearance time of the isotope from the postmastectomy lymphedematous arm as compared to the normal side. The results tend to confirm previously held convictions that the edema is usually confined to the subcutaneous fat and skin. The disappearance time (T1/2) of the radioactive sodium from the muscle of the edematous side was usually comparable to that of the control side. The data also indicate that the impairment of fluid drainage from other areas, such as the lateral chest wall, that normally drain into the axilla, is impaired similarly to that of the subcutaneous fat of the arm. Operative procedures designed to relieve the edema of the arm by providing an alternate route of drainage should provide a conduit for the fluid to an area that does not normally drain to the axilla of the affected side.