Improved screening for breast cancer associated with a telephone-based risk assessment

BACKGROUND

Our objective was to develop and field-test a telephone-based breast cancer risk assessment and to assess its efficacy in improving screening behavior. The study was performed at a financial institution and a manufacturing corporation with main offices in Boston, Massachusetts, and branch offices in various regions of the United States.

METHODS

A longitudinal study consisting of an initial health risk assessment administered by telephone, with a subsequent follow-up study initiated 8 months later, was performed. Study design was influenced by some of the suggestions made by the benefits departments of the corporate sponsors. A voice-response, telephone system collected risk information from callers and gave real-time risk assessment. These callers could receive a risk assessment over the phone and remain completely anonymous or furnish name and address to receive a more detailed written report. Main outcome measures included the response rate and demographics of the respondents, risk profiles of the callers, and breast cancer screening statuses.

RESULTS

There were 343 participants of whom 189 relinquished anonymity to receive more detailed information by mail and were available for a follow-up study. Sixty-three women (18%) reported a family history of breast cancer, with 34 women (10%) responding that one first-degree blood relative had been diagnosed before the age of 50. A strong positive correlation between the level of familial risk and the decision to remain anonymous existed (P < 0.0001). There was an increase in compliance with breast self-examination from 34% (40/119) at time of use of the system to 62% (74/119) at follow-up, P < 0.0001. Clinical breast exams showed similar improvements, from 82 (98/119) to 92% (110/119), P < 0.0137. Paired and unpaired data of women 40 years of age and older indicate an improvement in mammography compliance from time of system use to follow-up, 76 (22/29) to 93% (27/29), P < 0.0572, and 79 (33/42) to 93% (27/29), P < 0.0129, respectively.

CONCLUSIONS

A population of women with a risk profile higher than that of the U.S. population called the survey. System use is associated with an improvement in breast cancer screening habits. Self-reported, increased genetic risk for breast cancer was strongly correlated with a decision to remain anonymous.

Role of protein kinase A in tax transactivation of the human T-cell leukemia virus type I long terminal repeat

The human T-cell leukemia virus type I (HTLV-I) long terminal repeat (LTR) is inducible both by the retroviral tax gene product and by cyclic AMP in the murine thymoma S49 cell line. The cis-acting sequences that control transcriptional induction by tax and by cyclic AMP are in close proximity within the HTLV-I promoter. By using a protein kinase A (PKA)-deficient S49 mutant cell line, the response of the viral promoter to cyclic AMP was shown to depend on PKA, whereas the response to tax did not require the activity of this enzyme. Transactivation of the HTLV-I LTR by tax, however, decreased in PKA-deficient and adenylate cyclase-deficient cells. The evidence presented supports largely independent mechanisms of promoter induction by cyclic AMP and tax but also suggests a role for PKA-mediated phosphorylation in the regulation of HTLV-I LTR-driven gene expression by tax.

Protein kinase A-dependent binding of a nuclear factor to the 21-base-pair repeat of the human T-cell leukemia virus type I long terminal repeat

The long terminal repeat (LTR) of the human T-cell leukemia virus type I (HTLV-I) contains an imperfect repeat of 21 nucleotides which governs the response to the virus trans-activator protein tax and to cyclic AMP. In a murine thymocyte cell line defective in the catalytic subunit of protein kinase A, the response of the HTLV-I LTR to cyclic AMP is abolished and the response to tax is substantially diminished. This report shows that a factor present in nuclear extracts of wild-type cells binds to the HTLV-I 21-nucleotide sequence and that this binding activity is missing from the extracts of protein kinase A-defective cells. Treatment of nuclear extracts of protein kinase A-defective cells with the bovine protein kinase A catalytic subunit restores the binding activity, whereas treatment of wild-type nuclear extracts with a protein phosphatase destroys the binding activity. The binding factor is referred to as protein kinase A-dependent factor (PKAF). These results indicate that in murine thymocytes the response of the HTLV-I LTR to cyclic AMP depends upon the binding of a phosphorylated protein to the 21-nucleotide repeat sequence and that the response to tax is partially dependent upon binding of the phosphorylated protein. The results suggest a model in which the phosphorylation of a transcription factor by protein kinase A regulates HTLV-I gene expression.

Response of the human T-cell leukemia virus type 1 long terminal repeat to cyclic AMP

The sequences that control transcriptional initiation of the provirus of the human T-cell leukemia virus type 1 (HTLV-1) are shown to be responsive to intracellular levels of cyclic AMP. A heptanucleotide sequence present within the 21-nucleotide repeat sequence that is similar to the cyclic AMP-responsive consensus (CRE) sequence was required for cyclic AMP-mediated increase in gene expression. Although the CRE-like sequences were contained within sequences that were responsive to the virally encoded trans-activator (tax), the evidence presented indicates that the mechanisms of promoter induction by the tax product and cyclic AMP are independent. The implication of cyclic AMP stimulation of HTLV-1 provirus gene expression for long-term persistence of infected T cells and for virus-induced transformation is discussed.

Lymphocyte abnormalities in the BB rat

The BB rat has a marked T cell lymphocytopenia, with a near absence of peripheral "helper" T cells recognized by monoclonal antibody W3/25 (W3/25+ T cells). The lymphocytes of the BB rat's spleen and thymus were examined for the presence of W3/25+ T cells, which were found to be absent in the spleen but present in normal amounts in the thymus. Concanavalin A (Con A) responsiveness was absent in the BB's peripheral blood and spleen but present in the thymus. Thus, in these three lymphoid compartments, Con A responsiveness directly correlated with the presence or absence of W3/25+ T cells. These lymphocyte abnormalities in the BB rat are notably different from lymphocyte changes present in human type I diabetes.