This research trial studies deoxyribonucleic acid (DNA) in tumor tissue from women with node-positive breast cancer to see if genetic factors are related to the patient's response to chemotherapy. DNA analysis of tumor tissue may help doctors predict how well patients will respond to treatment with certain chemotherapy drugs.

PRIMARY OBJECTIVES:

1. To determine if amplification and/or overexpression of erbB-2 is associated with either additional or less benefit from increasing doses of doxorubicin or four cycles of paclitaxel.
2. To determine if abnormalities in p53 (mutations, deletions, protein stabilization) are associated with either additional or less benefit from increasing doses of doxorubicin or four cycles of paclitaxel.
3. To compare methods of analysis of erbB-2 and p53, using the most common means of assaying for these biomarkers, in order to determine whether one method is more predictive of clinical outcome than another, or whether methods are complementary.
4. To identify biomarkers that can be used to individually tailor the use of adjuvant dose-dense therapy.
5. To identify groups of patients who have a poor prognosis, despite adjuvant chemotherapy, who should be prospectively targeted for new approaches to adjuvant treatment.
6. To identify biomarkers that can be used to individually tailor the use of adjuvant paclitaxel therapy.
7. To identify groups of patients who have a poor prognosis despite adjuvant chemotherapy who should be prospectively targeted for new approaches to adjuvant treatment.
8. To determine if the intrinsic subtype, as determined by PAM50 assay, is associated with benefit from paclitaxel (C9344) and from dose dense chemotherapy (C9741) regardless of clinical estrogen receptor (ER) and human epidermal growth factor receptor 2 (HER2) status.
9. To determine if intrinsic subtype is associated with benefit from paclitaxel and from dose dense chemotherapy in the HER2 negative subset, as defined by the tissue microarray (TMA) analysis of HER2 completed in sections B2 and B3 of this protocol.
10. To determine if relapse score, as determined by PAM50 assay, is associated with benefit from paclitaxel and from dose dense chemotherapy regardless of clinical ER and HER2 status.
11. To determine if relapse score is associated with benefit from paclitaxel and from dose dense chemotherapy in the HER2 negative subset.
12. To evaluate the association of a PAM50-defined proliferation score and benefit from paclitaxel and from dose dense chemotherapy.
13. To evaluate whether patients identified as having basal breast cancer by the PAM50 assay benefit from paclitaxel.

SECONDARY OBJECTIVES:

1. To compare the performance of the PAM50-defined intrinsic subtypes with the subtype diagnoses determined by the antibody panel, developed in sections B2 and B3 of this protocol, in predicting benefit from paclitaxel and from dose dense chemotherapy in the HER2 negative subset.
2. To evaluate the prognostic value of the PAM50-defined relapse score in patients receiving no paclitaxel therapy (C9344).
3. To investigate the performance of a multivariate model including proliferation score and risk of relapse in predicting benefit from paclitaxel and from dose dense chemotherapy.

OUTLINE:

Formalin-fixed, paraffin-embedded tissue blocks are analyzed for amplification and/or overexpression of erbB-2 (HER-2/neu) and for mutations or deletions of p53 by fluorescent in situ hybridization (FISH) and immunohistochemistry.

1. Registration to CALGB 9344 or 9741

2. Samples collected, shipped and stored appropriately at the CALGB Pathology Coordinating Office

3. Institutional Review Board (IRB) review and approval at the institution where the laboratory work will be performed is required.

4. Informed consent: The CALGB does not require that a separate consent form be signed for this study.

   • The subject population to be studied in this protocol includes patients selected from one or more of the following CALGB treatment protocols: CALGB 9344 and 9741.
   • All patients have signed a written informed consent document meeting all federal, state and institutional guidelines as part of entry into those trials.
   • All samples to be studied were obtained and stored as part of the patient's respective treatment trial. The data obtained from the patient's record will be used to obtain appropriate clinical information. In no instance will the patient be contacted directly.
   • There should be no physical, psychological, social or legal risks associated with this study. No invasive procedures are recommended or requested.
   • All appropriate and necessary procedures will be utilized to maintain confidentiality. All patients who have had samples submitted for analysis will have their CALGB ID number used to identify specimens.
   • This study does not require direct patient contact and no specific risk or benefits to individuals involved in the trial are anticipated. It is likely, however, tIt is likely, however, that the information gained will substantially help similar patients in the future.

Patients with breast cancer enrolled on Cancer and Leukemia Group B (CALGB) 9344 or 9741.