Abstract BACKGROUND: Controversy exists whether women newly diagnosed with triple negative breast cancer (TNBC) should be referred to genetic counseling as they may be more likely to be BRCA carriers. However, prior studies included small numbers of carriers and their results have had limited influence on practice guidelines. The objective of this study was to determine the association of breast cancer molecular subtype and BRCA status in a large group of medically insured women. METHODS: We examined a cohort of 2,105 women with breast cancer history tested for BRCA mutations in a large California health plan from 1997–2011. BRCA test results were recorded in the health plan's clinical genetics registry. Of the 2,105 breast cancer patients, 249 were BRCA mutation carriers (143 BRCA1 carriers, and 106 BRCA2 carriers). We conducted data linkages of all patients with the health plan's NCI-SEER affiliated tumor registry and identified ER, PR, and HER2. HER2 status was also captured from pathology reports using natural language processing. ER, PR, and HER2 status were assessed by immunohistochemical or FISH techniques. Patients were classified into four main biologic subtypes: triple negative (ER−/PR−/HER2−); luminal A (ER+ and/or PR+/HER2−); luminal B (ER+ and/or PR+/HER2+); and HER2-enriched (HER2+/ER−). We examined the association between molecular subtypes (collapsed into TNBC vs. non TNBC categories) and BRCA1/2 mutation status using contingency table analyses. P-values (two-sided) were estimated using chi-square analysis. Multivariable logistic regression was used to estimated adjusted odds ratios (OR) and 95% confidence intervals. RESULTS: TNBC subtype was strongly associated with BRCA status (P < 0.0001). Women with TNBC tumors were five-fold more likely to be BRCA carriers than women who had non-TNBC breast tumors (OR = 5.6, 95% CI: 4.1–7.5). Specifically, the association of TNBC with BRCA1 was more robust (OR = 12.2, 95% CI: 8.3–17.9). Adjusting for age and stage of breast cancer diagnosis and race/ethnicity did not materially modify the association between TNBC and BRCA1 status. TNBC was not associated with BRCA2 status (OR = 1.6, 95% CI: 0.9–2.7). CONCLUSION: TNBC was strongly associated with BRCA1 status, but not with BRCA2 status. Statistically significant numbers of patients with BRCA mutations have a TNBC profile. These patients should therefore be referred to clinical genetics for further evaluation and possible testing. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-08-06.
Reina Haque, PhD1, Marianne Ulcickas Yood, PhD2, Aruna Kamineni, PhD3, Chantal Avila, MA1, Jiaxiao Shi, PhD1, Rebecca Silliman, MD, PhD4 and Virginia Quinn, PhD1 1Kaiser Permanente Southern California 2Henry Ford Health System 3Group Health Research Institute 4Boston University Medical Center
Abstract Background: HER2 status is important in breast cancer prognosis but has not been well collected in NCI-SEER affiliated cancer registries until recent years. For example, in a cohort of 2,846 women who were diagnosed with breast cancer from 1997 to 2011 in a large health plan, only 49% of these patients had known HER2 status. However, many of these with missing HER2 status were documented in clinical notes. This study used Natural Language Processing (NLP) technology to identify and extract HER2 status. NLP results were validated by comparison with cancer registry data and any discrepancies were manually reviewed. Method: We assembled a cohort of 2,846 breast cancer patients from the membership of a health plan based in southern California, Kaiser Permanente. All free text notes including pathology reports, progress notes, and discharge summaries were extracted from our electronic medical record (EMR) system. A window of 9 months before and after the diagnosis date was applied to restrict the number of notes needed to be processed. Overall, 513,903 clinical notes were processed and indexed, which averages to 180 notes per patient. Separate ontologies were created for the HER2 terminology and HER2 status. HER2 status values included positive, negative, borderline, test performed and test not performed. The NLP system employed additional components such as spelling correction, acronym recognition and negation identification. The output from the NLP system was further processed in three steps: Identification of the HER2 concept, followed by extraction of the most likely HER2 value, and lastly, a decision module to select the most likely HER2 value if there were conflicting values. Results: Use the cancer registry data as the gold standard, for positive and negative HER2 values, the sensitivity and specificity of the NLP algorithm were 94.7% and 93.3%, respectively, including the cases where NLP did not select either positive or negative values. A manual chart review was performed on the discrepant cases. We found that the NLP were correct for many of these cases. For example, out of the 39 NLP positive but registry negative cases, 4 were false positives and 35 were true positives. Compared to the cancer registry data, NLP increased capture of positive and negative HER2 cases from 49% to 73% of the cohort population. Discussion: NLP provides the opportunity to process clinical notes which are added to the EMR after the cancer registry has completed documenting the patients' initial course of cancer treatment. On the other hand, NLP was not able to access some of the HER2 related clinical notes. For example, according to our chart review, some notes were stored as scanned images and not retrievable. In addition, clinical notes with incorrect filing dates so patients without any diagnosis date were not processed. We also identified key challenges in determining HER2 results using NLP in this study. First, non-standard terminology for the HER2 receptor in the clinical notes hampered the NLP effectiveness. Second, clinicians described HER2 results in many different ways. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P1-07-12.
Abstract Abstract #3099 Background: Molecular marker expression analyses have led to the identification of unique breast cancer subtypes that have clinical and prognostic importance. Using data from a large population-based breast cancer cohort, we report results examining combined HER2/neu and estrogen receptor (ER) subtype distribution in relation to patient and tumor characteristics as well as breast cancer specific survival.
 Methods: We conducted a retrospective cohort study of 1,645 female members of a large managed care organization newly diagnosed with invasive breast cancer from 1988 to 1995. We collected patient, tumor, treatment, and outcome information from medical records and electronic cancer registry files. We used immunohistochemistry techniques to evaluate gene expression of specific molecular markers including HER2/neu and estrogen receptor (ER). We examined prevalence of markers as well as odds ratios (OR) and 95% confidence intervals (CI) comparing outcomes by marker expression.
 Results: Women with HER2+/ER- tumors were more likely to have been diagnosed at a younger age (less than age 60) than women with luminal A subtype (HER2-, ER+ and/or PR+) tumors (65.5% versus 47.8%, P<0.0001). A higher proportion of women with HER2+/ER- tumors were other than white race compared with women with luminal A subtype (21.8% versus 11.6%, P=0.007). Women with luminal B (HER2+, ER+ and/or PR+) or basal-like (HER2-, ER-, PR) subtype tumors were similarly more likely to be older at diagnosis and from minority backgrounds than women with luminal A subtype. Family history of breast cancer was not clearly associated with HER2/ER subtype. Women with luminal A tumors were more likely to be diagnosed with stage 1 disease than women with other tumor subtypes (luminal A: 56.3%; luminal B: 38.9%; basal-like: 38.1%; HER2+/ER-: 25.5%; P<0.0001). Residual tumor in surgical margins was not associated with HER2/ER subtype. Compared with women with luminal A tumors, women with HER2+/ER- tumors were much more likely to die of their cancers overall (OR=6.5, CI=3.5-11.9), and by stage (stage 1: OR=11.5, CI=2.6-50.3; stage 2: OR=3.8, CI=1.6-9.2; stage 3/4: OR=9.0, CI=1.1-77.2). Odds of breast cancer death, overall and by stage, was also increased for women with luminal B and basal-like subtype tumors compared with women having luminal A subtype.
 Conclusions: HER2/ER subtype was clearly associated with patient as well as tumor characteristics in this large cohort of breast cancer patients. Specific tumor subtypes (HER2+/ER-, luminal B, and basal-like) were also associated with increased likelihood of dying of breast cancer. Further work is warranted to define treatment strategies by molecular subtypes. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 3099.
Reina Haque1, Syed Ahmed1, Joanne Schottinger1, Marilyn Kwan2, Jiaxiao Shi1, Joanie Chung1, Galina Inzhakova1, Chantal Avila1, Ken Kleinman3 and Suzanne Fletcher1 1Kaiser Permanente Southern California 2Kaiser Permanente Northern California 3Harvard Pilgrim Health Care Institute, Harvard Medical School
