Search

Supplementary Table 1. Estimates of lung cancer risk associated with metformin use in patient subgroups. Supplementary Table 2. Estimates of lung cancer risk associated with metformin use, by gender. Supplementary Table 3. Estimates of lung cancer risk associated with metformin use, by tumor stage

Supplementary Table S1. Descriptive factors for KP study population, broken down by study. Supplementary Table S2. Genome-wide significant SNPs found in our cohort. Supplementary Table S3. Cis-eQTL expression of rs4646284. Supplementary Table S4. Results at the 105 loci previously found to be associated with prostate cancer. Supplementary Table S5. Risk score of and variance explained by the 105 previously reported hits. Supplementary Figure S1. Manhattan and Q-Q plots of each race/ethnicity and meta-analysis. Supplementary Figure S2. Local plots of novel replicated rs4646284 and suggestive rs2659124. Supplementary Figure S3. Cis-eQTL of SLC22A1 and SLC22A3. Supplementary Figure S4. Comparison of ORs of KP to previous reports by race/ethnicity. Supplementary Figure S5. KP AUC estimates.

Methodologic biases may explain why observational studies examining metformin use in relation to lung cancer risk have produced inconsistent results. We conducted a cohort study to further investigate this relationship, accounting for potential biases. For 47,351 patients with diabetes ages ≥40 years, who completed a health-related survey administered between 1994 and 1996, data on prescribed diabetes medications were obtained from electronic pharmacy records. Follow-up for incident lung cancer occurred from January 1, 1997, until June 30, 2012. Using Cox regression, we estimated lung cancer risk associated with new use of metformin, along with total duration, recency, and cumulative dose (all modeled as time-dependent covariates), adjusting for potential confounding factors. During 428,557 person-years of follow-up, 747 patients were diagnosed with lung cancer. No association was found with duration, dose, or recency of metformin use and overall lung cancer risk. Among never smokers, however, ever use was inversely associated with lung cancer risk [HR, 0.57; 95% confidence interval (CI), 0.33–0.99], and risk appeared to decrease monotonically with longer use (≥5 years: HR, 0.48; 95% CI, 0.21–1.09). Among current smokers, corresponding risk estimates were >1.0, although not statistically significant. Consistent with this variation in effect by smoking history, longer use was suggestively associated with lower adenocarcinoma risk (HR, 0.69; 95% CI, 0.40–1.17), but higher small cell carcinoma risk (HR, 1.82; 95% CI, 0.85–3.91). In this population, we found no evidence that metformin use affects overall lung cancer risk. The observed variation in association by smoking history and histology requires further confirmation. Cancer Prev Res; 8(2); 174–9. ©2014 AACR.

Background: Given the limited literature, we conducted a study to examine the epidemiology of herpes zoster (HZ) among newly diagnosed cancer patients.Methods: We identified adult health plan members of Kaiser Permanente Northern California diagnosed with invasive cancer from 2001 to 2005. Electronic health records with inpatient and outpatient diagnoses, laboratory tests, and antiviral medications were used to identify HZ diagnoses from 2001 to 2006. HZ diagnoses and associated complications were confirmed by medical chart review. Treatment with chemotherapy and corticosteroids was used to classify patients by immunosuppression level.Results: Among 14,670 cancer patients, 424 were diagnosed with HZ during follow-up (median 22 months). The incidence of HZ was 31/1,000 person-year (PY) in patients with hematologic malignancies and 12/1,000 PY in patients with solid tumors. The corresponding 2-year cumulative incidence of HZ was approximately 6% and 2%, respectively. Compared with incidence rates of HZ reported in a general US population, the age- and sex-standardized rates of HZ were 4.8 times higher [95% confidence interval (CI), 4.0–5.6] in patients with hematologic malignancies and 1.9 times higher (95% CI, 1.7–2.1) in those with solid tumors. HZ risk increased with increasing level of immunosuppression. Among HZ cases, 19% with hematologic malignancies and 14% with solid tumors had HZ-associated pain for at least 30 days. The corresponding numbers for nonpain-related complications were 30% and 18%, respectively.Conclusions: Cancer patients are at substantially increased risk of HZ and among those with HZ, complications are relatively common.Impact: Better HZ prevention and treatment options for cancer patients are needed. Cancer Epidemiol Biomarkers Prev; 22(1); 82–90. ©2012 AACR.

Background: Body composition may partially explain the U-shaped association between body mass index (BMI) and colorectal cancer survival.Methods: Muscle and adiposity at colorectal cancer diagnosis and survival were examined in a retrospective cohort using Kaplan–Meier curves, multivariable Cox regression, and restricted cubic splines in 3,262 early-stage (I–III) male (50%) and female (50%) patients. Sarcopenia was defined using optimal stratification and sex- and BMI-specific cut points. High adiposity was defined as the highest tertile of sex-specific total adipose tissue (TAT). Primary outcomes were overall mortality and colorectal cancer–specific mortality (CRCsM).Results: Slightly over 42% patients were sarcopenic. During 5.8 years of follow-up, 788 deaths occurred, including 433 from colorectal cancer. Sarcopenic patients had a 27% [HR, 1.27; 95% confidence interval (CI), 1.09–1.48] higher risk of overall mortality than those who were not sarcopenic. Females with both low muscle and high adiposity had a 64% higher risk of overall mortality (HR, 1.64; 95% CI, 1.05–2.57) than females with adequate muscle and lower adiposity. The lowest risk of overall mortality was seen in patients with a BMI between 25 and 2, a range associated with the greatest number of patients (58.6%) who were not at increased risk of overall mortality due to either low muscle or high adiposity.Conclusions: Sarcopenia is prevalent among patients with non-metastatic colorectal cancer, and should, along with adiposity be a standard oncological marker.Impact: Our findings suggest a biologic explanation for the obesity paradox in colorectal cancer and refute the notion that the association between overweight and lower mortality is due solely to methodologic biases. Cancer Epidemiol Biomarkers Prev; 26(7); 1008–15. ©2017 AACR.

Supplemental Tables 1-3

PDF file - 83K, This supplementary document describes methods used to create an algorithm for classifying patients' level of immunosuppression

Supplementary Figure 1 from CYP1A1/2 Haplotypes and Lung Cancer and Assessment of Confounding by Population Stratification

Supplementary Tables 1-4 from CYP1A1/2 Haplotypes and Lung Cancer and Assessment of Confounding by Population Stratification

Prior studies of lung cancer and CYP1A1/2 in African-American and Latino populations have shown inconsistent results and have not yet investigated the haplotype block structure of CYP1A1/2 or addressed potential population stratification. To investigate haplotypes in the CYP1A1/2 region and lung cancer in African-Americans and Latinos, we conducted a case-control study (1998–2003). African-Americans (n = 535) and Latinos (n = 412) were frequency matched on age, sex, and self-reported race/ethnicity. We used a custom genotyping panel containing 50 single nucleotide polymorphisms in the CYP1A1/2 region and 184 ancestry informative markers selected to have large allele frequency differences between Africans, Europeans, and Amerindians. Latinos exhibited significant haplotype main effects in two blocks even after adjusting for admixture [odds ratio (OR), 2.02; 95% confidence interval (95% CI), 1.28–3.19 and OR, 0.55; 95% CI, 0.36–0.83], but no main effects were found among African-Americans. Adjustment for admixture revealed substantial confounding by population stratification among Latinos but not African-Americans. Among Latinos and African-Americans, interactions between smoking level and haplotypes were not statistically significant. Evidence of population stratification among Latinos underscores the importance of adjusting for admixture in lung cancer association studies, particularly in Latino populations. These results suggest that a variant occurring within the CYP1A2 region may be conferring an increased risk of lung cancer in Latinos. [Cancer Res 2009;69(6):2340–8]

Supplementary Figure 3 from CYP1A1/2 Haplotypes and Lung Cancer and Assessment of Confounding by Population Stratification

Supplementary Figure 2 from CYP1A1/2 Haplotypes and Lung Cancer and Assessment of Confounding by Population Stratification

Background: Body composition may partially explain the U-shaped association between body mass index (BMI) and colorectal cancer survival. Methods: Muscle and adiposity at colorectal cancer diagnosis and survival were examined in a retrospective cohort using Kaplan–Meier curves, multivariable Cox regression, and restricted cubic splines in 3,262 early-stage (I–III) male (50%) and female (50%) patients. Sarcopenia was defined using optimal stratification and sex- and BMI-specific cut points. High adiposity was defined as the highest tertile of sex-specific total adipose tissue (TAT). Primary outcomes were overall mortality and colorectal cancer–specific mortality (CRCsM). Results: Slightly over 42% patients were sarcopenic. During 5.8 years of follow-up, 788 deaths occurred, including 433 from colorectal cancer. Sarcopenic patients had a 27% [HR, 1.27; 95% confidence interval (CI), 1.09–1.48] higher risk of overall mortality than those who were not sarcopenic. Females with both low muscle and high adiposity had a 64% higher risk of overall mortality (HR, 1.64; 95% CI, 1.05–2.57) than females with adequate muscle and lower adiposity. The lowest risk of overall mortality was seen in patients with a BMI between 25 and Conclusions: Sarcopenia is prevalent among patients with non-metastatic colorectal cancer, and should, along with adiposity be a standard oncological marker. Impact: Our findings suggest a biologic explanation for the obesity paradox in colorectal cancer and refute the notion that the association between overweight and lower mortality is due solely to methodologic biases. Cancer Epidemiol Biomarkers Prev; 26(7); 1008–15. ©2017 AACR.

Background: Prospective clinical trials using clinical criteria and observational studies using diagnostic codes from electronic health records have reported seemingly contradictory cardiotoxicity risk for adjuvant trastuzumab (T). Accurate estimates of individualized patient specific cardiotoxicity risk are essential for treatment decisions in early HER2+ breast cancer (BC). Methods: 1,109 consecutive non-metastatic HER2+ invasive BC's diagnosed 5/1/2005 to 12/31/2011 at Kaiser Permanente Northern California receiving adjuvant T were reviewed for symptomatic congestive heart failure (SxCHF), baseline and post-T cardiac ejection fraction (EF), anthracycline (A) use, and CHF risk factors (RF) including age, race, hypertension (HTN), diabetes (DM), obesity, smoking. Records of patients with CHF ICD9 codes or an EF drop to Results: Median age of 962 eligible patients was 54 years (range 24-95). 305 (31.7%) were > 60 years old. During a median follow-up of 4.1 years, 4.6% of patients had CHF ICD9 codes, but only 2.5% had SxCHF or cardiac death confirmed by clinical review. At 1 year, cumulative incidence of an EF fall to Predictors of Trastuzumab Cardiotoxicity Cumulative Incidence (%)PredictorsNFall EF to60, 2+ RF(HTN, DM, BMI 30+)10114.95.08.9 Conclusions: Risk of clinically confirmed CHF/cardiac death was substantially lower than risk based on ICD codes alone. Risk was consistent with prior clinical trials and differed substantially by age, baseline EF, use of A, and other CHF risk factors. Greatest risk was with age of 70+, borderline baseline EF, and comorbidities known to increase CHF risk. Quite low risk (1.1% at 5 years) was seen in patients under 60 years old. Citation Format: Fehrenbacher L, Capra A, Krishnaswami A, Quesenberry C, Habel L. Adjuvant trastuzumab +/- anthracycline and cardiotoxicity in a community cohort of 962 HER2+ breast cancers from 2005-2011: Comparison of incidence by risk factors and by diagnostic codes vs clinical chart review. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P1-15-01.

A genome-wide association study (GWAS) of prostate cancer in Kaiser Permanente health plan members (7,783 cases, 38,595 controls; 80.3% non-Hispanic white, 4.9% African-American, 7.0% East Asian, and 7.8% Latino) revealed a new independent risk indel rs4646284 at the previously identified locus 6q25.3 that replicated in PEGASUS (N = 7,539) and the Multiethnic Cohort (N = 4,679) with an overall P = 1.0 × 10−19 (OR, 1.18). Across the 6q25.3 locus, rs4646284 exhibited the strongest association with expression of SLC22A1 (P = 1.3 × 10−23) and SLC22A3 (P = 3.2 × 10−52). At the known 19q13.33 locus, rs2659124 (P = 1.3 × 10−13; OR, 1.18) nominally replicated in PEGASUS. A risk score of 105 known risk SNPs was strongly associated with prostate cancer (P < 1.0 × 10−8). Comparing the highest to lowest risk score deciles, the OR was 6.22 for non-Hispanic whites, 5.82 for Latinos, 3.77 for African-Americans, and 3.38 for East Asians. In non-Hispanic whites, the 105 risk SNPs explained approximately 7.6% of disease heritability. The entire GWAS array explained approximately 33.4% of heritability, with a 4.3-fold enrichment within DNaseI hypersensitivity sites (P = 0.004). Significance: Taken together, our findings of independent risk variants, ethnic variation in existing SNP replication, and remaining unexplained heritability have important implications for further clarifying the genetic risk of prostate cancer. Our findings also suggest that there may be much promise in evaluating understudied variation, such as indels and ethnically diverse populations. Cancer Discov; 5(8); 878–91. ©2015 AACR. This article is highlighted in the In This Issue feature, p. 783

Background: The PAM50, a gene expression assay to categorize breast tumors into intrinsic subtypes, has not been previously used to examine short- and long-term prognostication in a population-based cohort where treatment patterns and time of initial follow-up vary. Methods: In a stratified case–cohort design of 1,691 women from the LACE and Pathways breast cancer survivor cohorts, we used PAM50 to categorize tumors into Luminal A (LumA), Luminal B (LumB), HER2-enriched (HER2-E), Basal-like and Normal-like, and to examine risk of early and late recurrence and mortality by Cox proportional hazards regression. Results: Compared with LumA, cumulative risk of recurrence and breast cancer death was higher for LumB, HER2-E, and Basal-like tumors at 2, 5, and 10 years. However, HR of breast cancer death varied over time [ 5 years (late)] for both Basal-like (HR, 6.23 early vs. HR, 0.63 late) and HER2-E tumors (HR, 2.97 early vs. HR, 0.73 late) but not for LumB tumors where risk was elevated consistently (HR, 2.67 early vs. HR, 1.47 late). The contrast between LumB, HER2-E, and Basal-like compared with LumA on early recurrence was stronger when subtype was defined by PAM50 than by immunohistochemistry (IHC) markers. Conclusions: The PAM50 categorized intrinsic subtypes in a manner that more accurately predicts recurrence and survival, especially for luminal tumors, compared with commonly used methods that rely on traditional IHC clinical markers. Impact: The PAM50 is robust for use in epidemiologic studies and should be considered when archived tumor tissues are available. Cancer Epidemiol Biomarkers Prev; 23(5); 725–34. ©2014 AACR.

Basal-like intrinsic subtype and poor prognosis are characteristic of breast cancers in BRCA1 families, but other genes also confer inherited predisposition to breast cancer, and subtype associations may differ. The objective of this study was to examine whether family history of breast cancer is associated with basal-like subtype and/or poor prognosis in a population-based sample of women with breast cancer. METHODS: The study population was the Life After Cancer Epidemiology (LACE) cohort, diagnosed with breast cancer in 1996-2000, and the Pathways cohort, diagnosed with breast cancer in 2006-2008. History of breast cancer among first-degree relatives was obtained by self-report at study enrollment, and women were followed for breast cancer recurrence and mortality. Primary breast cancer tissue was obtained for a case-cohort sample. Intrinsic subtypes were classified based on RT-PCR assay of 50 genes (PAM50). Sample-weighted subtype distributions were compared by family history. Hazard ratios (HR) for recurrence and mortality were estimated from Cox proportional hazards models. RESULTS: From 4,256 women in the parent cohorts who responded to the family history question, 20.5% reported a first-degree family history of breast cancer. Among women with family history, 80.6% were diagnosed at ages 50 and older, compared with 76.3% of women without family history, p = 0.007. Among cases with family history, 79.2% were non-Hispanic whites, compared with 72.3% of cases without family history, p < 0.001. Within the subcohort with PAM50 results (n = 1,319), cases with and without family history had 8.5% and 10.2% prevalence of basal-like tumors, respectively, p = 0.27. For cases diagnosed before age 50, the prevalences of basal-like subtype for women with and without family history were 18.6% and 15.9%, respectively, p = 0.62. Women with family history had similar recurrence risk to those without, HR 0.82 (95% CI 0.61, 1.11; age- and race-adjusted). Women with family history had a suggestive reduced risk of death from all causes, HR 0.75 (95% 0.55, 1.02). Family history was not significantly associated with recurrence or survival within any intrinsic subtype group. DISCUSSION: Although BRCA1 is known to predispose to basal-like subtype, we found no association between family history of breast cancer and basal-like subtype or worse prognosis in this population-based cohort. Mutation status for BRCA1 and other cancer susceptibility genes are unknown for cohort members. First-degree family history of breast cancer may represent the presence of one of the several inherited breast cancer susceptibility genes, or chance aggregation of sporadic cases. Particularly for a woman diagnosed at an older age, her mother and sisters would also have reached ages when sporadic breast cancer incidence is high. Among women diagnosed before age 50, family history was associated with a small and non-significant excess of basal-like tumors. Implications of this study are that for women from the general population diagnosed with breast cancer, a first-degree family history of breast cancer indicates neither a higher probability of a basal-like tumor nor worse prognosis. Support: NIH CA129059 and CA105274. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P3-07-01.

Background: Weight change after a breast cancer diagnosis has been linked to lower survival. To further understand effects of postdiagnostic weight variation on survival, we examined the relationship by comorbid status and initial body mass index (BMI). Methods: The current analysis included 12,915 patients with breast cancer diagnosed between 1990 and 2006 with stage I–III tumors from four prospective cohorts in the United States and China. HRs and 95% confidence intervals (CI) representing the associations of five weight change categories [within Results: Mean weight change was 1.6 kg. About 14.7% women lost and 34.7% gained weight. Weight stability in the early years postdiagnosis was associated with the lowest overall mortality risk. Weight loss ≥10% was related to a 40% increased risk of death (HR, 1.41; 95% CI, 1.14–1.75) in the United States and over three times the risk of death (HR, 3.25; 95% CI: 2.24, 4.73) in Shanghai. This association varied by prediagnosis BMI, and in the United States, lower survival was seen for women who lost weight and had comorbid conditions. Weight gain ≥10% was associated with a nonsignificant increased risk of death. Conclusions: Prevention of excessive weight gain is a valid public health goal for breast cancer survivors. Although intentionality of weight loss could not be determined, women with comorbid conditions may be particularly at risk of weight loss and mortality. Impact: Weight control strategies for breast cancer survivors should be personalized to the individual's medical history. Cancer Epidemiol Biomarkers Prev; 21(8); 1260–71. ©2012 AACR.

Background: A large body of evidence dating back over 30 years suggests that obese women have poorer survival after a breast cancer (BC) diagnosis compared to non-obese women. Despite most studies supporting an association of elevated risk of overall mortality with obesity, the relationship of obesity with risk of BC recurrence, BC mortality and non-BC mortality remains unclear. Furthermore, reports suggest that the association of BMI with BC outcomes may be U or J shaped, prompting the necessity of examining underweight and more severely obese women as independent groups. We conducted a pooled investigation of pre-diagnosis BMI and BC recurrence and survival using data from the After Breast Cancer Pooling Project (ABCPP). Materials and Methods: The ABCPP includes 14,950 BC survivors from four prospective cohorts (three US and one Shanghai, China) diagnosed from 1990–2006 with invasive primary AJCC Stage I-III BC at ages 20–83 years. A random effects meta-analysis was conducted to assess heterogeneity across studies and poolability of data. Delayed entry Cox proportional hazards models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for the associations of pre-diagnosis BMI (underweight Results: No heterogeneity in effect estimates by study was found. 2104 deaths (1416 BC-related) and 2320 recurrences were observed after a mean (SD) of 7.66 (3.95) years of follow-up. Both underweight and obese women had a statistically significant increased risk of overall death compared to normal-weight women (underweight HR=1.69; 95% CI: 1.25, 2.28 and obese HR=1.22; 95% CI: 1.08, 1.38; p for nonlinear association Discussion: In this large pooling study of nearly 15,000 BC survivors, we found that the association between BMI and BC outcomes, specifically overall death and non-BC death, was U shaped with both underweight and obese women at greatest risk. Morbidly obese women were at even greater risk compared to other obesity groups. Maintaining a healthy weight throughout adult life may be beneficial for BC prognosis and survival. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-08-02.

Background: Hyperglycemia may increase the risk of colorectal neoplasia by serving as an energy source for neoplastic growth. We sought to determine whether glycemic control measured by serial hemoglobin A1c (HbA1c) was associated with the risk of colorectal adenoma. Methods: Among a cohort of patients with type 2 diabetes mellitus who received health care within the Kaiser Permanente Northern California from 1994 to 2005, we conducted 2 case–control analyses. Cases had at least 1 colorectal adenoma identified at either colonoscopy (analysis 1) or sigmoidoscopy (analysis 2). Controls had no colorectal neoplasia identified at the corresponding endoscopic examination. Serial HbA1c levels between the cases and the controls were compared using a longitudinal model. Results: Case–control analysis 1 included 4,248 patients, of whom 1,296 (31%) had at least 1 adenoma. The adjusted mean HbA1c levels among those without any adenomas was 8.20% versus 8.26% among those with at least 1 adenoma, a difference of 0.06% (95% CI = −0.02 to 0.14, P = 0.16). Case–control analysis 2 included 9,813 patients, of whom 951 (10%) had at least 1 distal adenoma. The adjusted mean HbA1c levels among those without any distal adenomas was 8.32% versus 8.37% among those with at least 1 distal adenoma, a difference of 0.05% (95% CI = −00.04 to 0.14, P = 0.25). The results were similar for advanced adenomas. Conclusions: Glycemic control was not associated with the risk of colorectal adenoma among diabetic persons. Impact: These results would suggest that glycemic control is unlikely to confound the reported association between diabetes medications and the risk of colorectal cancer. Cancer Epidemiol Biomarkers Prev; 19(12); 3027–36. ©2010 AACR.

Background: We examined whether mammographic density predicts risk of second breast cancers among patients with ductal carcinoma in situ (DCIS). Methods: The study included DCIS patients diagnosed during 1990 to 1997 and treated with breast-conserving surgery at Kaiser Permanente Northern California. Medical records were reviewed for clinical factors and subsequent breast cancers (DCIS and invasive). Ipsilateral mammograms from the index DCIS were assessed for density without knowledge of subsequent cancer status. Cox regression modeling was used to examine the association between mammographic density and risk of breast cancer events. Results: Of the 935 eligible DCIS patients, 164 (18%) had a subsequent ipsilateral breast cancer, and 59 (6%) had a new primary cancer in the contralateral breast during follow-up (median, 103 mo). Those with the greatest total area of density (upper 20% of values) were at increased risk for invasive disease in either breast [hazard ratio (HR), 2.1; 95% confidence interval (95% CI), 1.2-3.8] or any cancer (DCIS or invasive) in the ipsilateral (HR, 1.7; 95% CI, 1.0-2.9) or contralateral (HR, 3.0; 95% CI, 1.3-6.9) breast compared with those with the smallest area of density (bottom 20%). HRs for these same end points comparing those in the highest with those in the lowest American College of Radiology Breast Imaging Reporting and Data System category were 1.6 (95% CI, 0.7-3.6), 1.3 (95% CI, 0.7-2.6), and 5.0 (95% CI, 1.4-17.9), respectively. There was a suggestion of increasing risk of contralateral, but not ipsilateral, cancer with increasing percent density. Conclusions: Women with mammographically dense breasts may be at higher risk of subsequent breast cancer, especially in the contralateral breast. Impact: Information about mammographic density may help with DCIS treatment decisions. Cancer Epidemiol Biomarkers Prev; 19(10); 2488–95. ©2010 AACR.

Background: Prostate-specific antigen (PSA) screening for prostate cancer results missed diagnoses as well as many unnecessary biopsies and their associated morbidities. Because PSA levels are heritable, traditional PSA screening may be improved by adjustment for constitutive germline genetics that influence PSA independently of prostate cancer. Methods: In a previous genome-wide association study in non-Hispanic white men without prostate cancer in the Kaiser Permanente Research Program on Genes, Environment and Health cohort, we identified 40 single nucleotide polymorphisms (SNP) independently and significantly associated with PSA levels. Among them, 24 were not associated with prostate cancer risk. To calculate genetically normalized PSA (i.e., PSA’), we multiplied each man’s most recent PSA measurement by a factor that adjusted for the amount by which his 24 PSA-specific SNPs may have increased his PSA level. We then compared how men without prostate cancer were classified based on their PSA and PSA’ levels with regard to decisions for diagnostic testing. Results: PSA and PSA’ were highly correlated among non-Hispanic white men without prostate cancer (r2: 0.959; 95% CI: 0.958-0.960). Still, 4.6% were reclassified from above to below a cutpoint of 2.5 ng/ml when using the PSA’ measure, and 2.7% were reclassified from below to above this cutpoint (4.6% - 2.7% = 1.9% net reclassified to below the cutoff). The reclassification was even more pronounced when restricting to controls with a negative prostate biopsy: 5.0% downward and 1.5% upward (3.5% net reclassified to below the cutoff). Conclusions: Normalizing PSA levels using germline PSA variants may reduce the number of subjects without prostate cancer receiving unnecessary biopsies. Citation Format: Rebecca E. Graff, Thomas J. Hoffmann, Michael N. Passarelli, Nima C. Emami, Lori C. Sakoda, Eric Jorgenson, Laurel A. Habel, Jun Shan, Dilrini K. Ranatunga, Charles P. Quesenberry, Chun R. Chao, Nirupa R. Ghai, David Aaronson, Joseph Presti, Tobias Nordström, Zhaoming Wang, Sonja I. Berndt, Stephen J. Chanock, Jonathan D. Mosley, Robert J. Klein, Mridu Middha, Hans Lilja, Olle Melander, Mark N. Kvale, Pui-Yan Kwok, Catherine Schaefer, Neil Risch, Neil Risch, Stephen K. Van Den Eeden, John S. Witte. Genetic reclassification of prostate-specific antigen levels for personalized prostate cancer screening [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1297. doi:10.1158/1538-7445.AM2017-1297

Early detection strategies for lung cancer may be improved by using valid risk prediction models to identify persons at highest risk for the disease. However, external validation of lung cancer risk prediction models has been limited. We sought to externally validate the PLCOM2012 model, which predicts the probability of lung cancer within six years on the basis of age, race, education, body mass index, chronic obstructive pulmonary disease, personal history of cancer, family history of lung cancer, and smoking status, quantity, duration, and quit years, in the Kaiser Permanente Northern California (KPNC) Research Program on Genes, Environment, and Health (RPGEH) cohort. To increase comparability to the populations of smokers used to initially develop and validate the PLCOM2012 model, we restricted our analysis to the 28,757 ever smokers ages 55 to 74 with no history of lung cancer, no history of other non-melanoma skin cancers in the prior five years, and complete data on all model predictors. For each person, the predicted probability of lung cancer risk was estimated with data ascertained from the RPGEH survey on all predictors except quit years, which was ascertained from electronic health records. Using KPNC Cancer Registry data, we identified 672 diagnosed with lung cancer within six years post-survey. Both calibration and discrimination were examined to assess model performance. Calibration was assessed by determining the mean absolute difference in observed and predicted probabilities of lung cancer for each decile of predicted risk. Discrimination was assessed by estimating the area under curve (AUC). The absolute difference in observed and predicted probabilities of lung cancer risk was generally small: Citation Format: Lori C. Sakoda, Laurel A. Habel, Khanh K. Thai, Charles P. Quesenberry, Jr. External validation of a risk prediction model for lung cancer among smokers. [abstract]. In: Proceedings of the AACR Special Conference: Improving Cancer Risk Prediction for Prevention and Early Detection; Nov 16-19, 2016; Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2017;26(5 Suppl):Abstract nr PR13.

Background: Breast cancer-related lymphedema (BCRL) is a serious chronic condition that can occur in about 30% of patients after breast cancer (BC) surgery and treatment. Patient risk factors include younger age, higher BMI, and less physical activity. With few studies to date, it is largely unknown if BCRL risk varies by race/ethnicity. Methods: In a prospective study of 2,953 BC patients, we examined the association of self-reported BCRL status at 12 and 48 months post-BC diagnosis with self-reported race/ethnicity and genetic ancestry. We also assessed associations with other clinical and patient factors. Race/ethnicity (White, African American (AA), Hispanic, Asian, and Other) was asked in the baseline interview at cohort entry. Genetic ancestry was estimated based on a validated panel of 124 ancestry informative markers (AIMs) using the STRUCTURE program. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated by multivariable Cox proportional hazards models, with follow-up until date of BCRL self-report or last patient contact, whichever occurred first. Results: 342 (11.6%) women reported having BCRL at 12 or 48 month follow-up, with 204 who were White, 37 AA, 41 Hispanic, 49 Asian, and 11 Other. Younger age at BC diagnosis, higher BMI at baseline, and less moderate-vigorous physical activity were associated with greater BCRL risk. After adjusting for sociodemographic and clinical factors, AA women had a significant 1.6-fold increased risk of BCRL (HR = 1.57; 95% CI: 1.09, 2.26) and Hispanic women had a borderline 1.4-fold increased risk (HR = 1.37; 95% CI: 0.99, 1.89), compared with White women. Consistent with self-reported race/ethnicity, African ancestry was associated with a 1.8-fold increased risk of BCRL (HR = 1.80; 95% CI: 1.15, 2.81). When the race/ethnicity and ancestry models were further adjusted for BCRL risk factors, i.e., age at BC diagnosis, BMI, and physical activity, associations became attenuated and non-significant among AA (HR = 1.26, 95% CI: 0.87-1.83) and Hispanic women (HR = 1.19, 95% CI: 0.85-1.68), and with African ancestry (HR = 1.38, 95% CI: 0.88-2.19). Results were similar when excluding BCRL events within 6 months of BC diagnosis to rule out transient post-operative swelling, except a suggestive increased risk of BCRL remained among AA women (HR = 1.47, 95% CI: 0.94-2.28) and with African ancestry (HR = 1.70, 95% CI: 0.99-2.91) after adjustment for BCRL risk factors. Further, the elevated BCRL risk seen in AA or Hispanic women, and with African ancestry, appeared to be stronger in non-obese compared with obese women. Discussion: AA women had increased risk of BCRL compared with White women, which is partly attributed to differences in age at BC diagnosis, BMI, and physical activity. This is the first large-scale, prospective study to examine a racial/ethnic disparity of BCRL risk with self-report and genetic ancestry data. Funded by R01 CA105274. Citation Format: Marilyn L. Kwan, Valerie S. Lee, Janise M. Roh, Isaac J. Ergas, Yali Zhang, Susan E. Kutner, Charles P. Quesenberry, Christine B. Ambrosone, Lawrence H. Kushi, Song Yao. Race/ethnicity, genetic ancestry, and breast cancer-related lymphedema. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3724. doi:10.1158/1538-7445.AM2015-3724

Background: Aromatase inhibitors (AIs) have been replacing tamoxifen (TAM) as adjuvant hormonal therapy (HT) for hormone receptor-positive, postmenopausal breast cancer (BC). AIs are known for their adverse effects on bone health, yet bone health history among AI users before BC diagnosis is unknown, which may impact fracture risk after AI therapy. In a cross-sectional analysis of a prospective cohort of BC patients on HT, we describe prior history of osteoporosis and fracture before BC diagnosis among AI users compared with TAM users. Methods: A total of 3280 women diagnosed with invasive BC from 2005-2013 and initially treated with AI or TAM were identified from an ongoing prospective cohort study at Kaiser Permanente Northern California (KPNC). Data on baseline demographic and lifestyle factors came from in-person interviews about two months post-diagnosis. Information on BC diagnosis and treatment, HT, and prior bone health history including bisphosphonate (BP) use was obtained from the KPNC cancer registry and clinical databases. Using age- and race-adjusted logistic regression, we examined the association of lifestyle and clinical factors at BC diagnosis with prior history of 1) osteoporosis, 2) any fracture (excluding head, face, fingers, toes), and 3) any major fracture (hip, spine, humerus, wrist). Results: Compared with initial TAM users (n=1140), initial AI users (n=2140) were more likely to be older (mean 64.5 y vs. 50.5 y) and obese (38% vs. 26%), yet less likely to have had chemotherapy (36% vs. 51%). AI users were more likely to have a prior history of osteoporosis (n=164, 7.7% vs. n=62, 5.4%, age-adjusted pmedian OR=0.4; 95% CI: 0.2, 0.7). Discussion: This descriptive analysis is one of the first to characterize AI users and risk factors for prior bone morbidity before BC diagnosis. In the future, this study will evaluate the impact of AIs on risk of subsequent bone outcomes and examine lifestyle, molecular, and genetic risk factors for AI-induced fractures. Funded by R01 CA166701 and R01 CA105274. Citation Format: Marilyn L. Kwan, Joan C. Lo, Li Tang, Cecile Laurent, Janise M. Roh, Malini Chandra, Theresa E. Hahn, Chi-Chen Hong, Lara Sucheston-Campbell, Dawn L. Hershman, Bette J. Caan, Charles P. Quesenberry, Barbara Sternfeld, Christine B. Ambrosone, Lawrence H. Kushi, Song Yao. Prior bone health history in breast cancer patients on aromatase inhibitors. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2162. doi:10.1158/1538-7445.AM2014-2162

Background: Obesity is associated with breast cancer (BC) in postmenopausal women, but its relationship with intrinsic subtype is unclear. We examined associations of body mass index (BMI) with intrinsic subtype and expression of selected genes among women with invasive BC. Methods: A case-cohort of 1,676 BC patients was sampled from two prospective Kaiser Permanente Northern California cohorts: LACE (AJCC stage I (≥1 cm), II, IIIA diagnosed 1997-2000) and Pathways (invasive stage ≥0.5 cm diagnosed 2006-2008). The PAM50 quantitative reverse transcriptase polymerase chain reaction (RT-qPCR) assay was used to: a) classify tumors into intrinsic subtype (Luminal A, Luminal B, Basal-like, HER2-E, Normal-like) and b) assess gene expression levels for ESR1, PGR, ERBB2, and a composite of 10 proliferation genes. BMI was from self-reported weight and height within one year prior to BC diagnosis and categorized as: normal ( Results: The cohort was comprised of 52% Luminal A, 21% Luminal B, 10% Basal-like, 14% HER2-E, and 3% Normal-like subtypes. Basal-like was more prevalent in the highly obese (19%), compared with the other BMI groups (8%-10%), whereas Luminal A was less common in the highly obese (36%), compared with the other groups (49%-57%). Greater expression of proliferation genes was seen with increasing BMI (p Discussion: Among postmenopausal women with breast cancer, those who were highly obese, but not mildly obese, had increased odds of Basal-like and Luminal B subtypes and had tumors with greater expression of proliferation genes. Tumor subtypes may vary by level of obesity. Citation Format: Marilyn L. Kwan, Candyce H. Kroenke, Carol Sweeney, Philip S. Bernard, Erin Weltzien, Adrienne Castillo, Rachel E. Factor, Kaylynn Shakespear, Inga J. Stijleman, Charles P. Quesenberry, Laurel A. Habel, Lawrence H. Kushi, Bette J. Caan. Association of high obesity with PAM50 breast cancer intrinsic subtypes and gene expression. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3263. doi:10.1158/1538-7445.AM2014-3263

BACKGROUND The PAM50 gene expression signature for identifying intrinsic subtypes of breast cancer has been shown to add significant clinical information beyond standard molecular biomarkers (ER, PR, Her2). Previous studies with the PAM50 have focused primarily on retrospective clinical trials in which the drug regimens and enrollment criteria are relatively homogenous. We assess the prognostic significance of subtyping within the community setting using therapies appropriate at the time of diagnosis. METHODS Two prospective studies of breast cancer survivors, Life After Cancer Epidemiology (LACE) and Pathways, were subtyped by the PAM50 using quantitative RT-PCR. LACE participants were diagnosed from 1997-2000 with tumors ≥1 cm; while Pathways enrolled subjects at time of diagnosis (2006-2008) with tumors ≥0.5 cm. Both studies included AJCC stages I-III. LACE participants were enrolled on average 2 years post-diagnosis, such that early recurrences were excluded. In addition, women in Pathways were diagnosed after 2006 when Herceptin treatment was available. Formalin-fixed, paraffin-embedded tumor blocks were centrally reviewed by a pathologist and 1mm directed punches of invasive cancer were taken for RNA extraction and expression profiling. A published algorithm was used for subtyping and provided a continuous gene expression score for proliferation, ESR1/ER, PGR/PR, and ERBB2/Her2 (Bastien et al, BMC Med Genomics. 2012 Oct 4;5:44). Subtype classification was correlated with disease-free survival, estimated using Kaplan-Meier plots and log-rank testing. Multivariable delayed entry Cox regression analyses determined the prognostic significance of subtypes in the context of standard clinical indicators of survival. RESULTS The PAM50 subtype distribution for the cohort was Luminal A (52%), Luminal B (20%), HER2-E (14%), Basal-like (10%), and Normal-like (4%). Approximately 7% of Luminal tumors (A and B) were called ER negative by immunohistochemistry (IHC). Luminal A tumors were more likely to be PR positive (82%) compared to those classified as Luminal B (61%). Only 2% of Luminal B tumors were low proliferation. The HER2-E subtype contained 66% clinically Her2+ tumors and the Basal-like tumors were 90% triple negative. In univariate analyses, women with non-Luminal A subtypes were all at higher risk of recurrence. Controlling for tumor size and positive nodes, those with Basal-like subtype still had a significantly higher risk of recurrence (HR=2.34, 95% CI 1.37, 3.98). CONCLUSIONS There are proportionally more Luminal A patients within the population-based studies compared to clinical trials that have focused on higher risk patients. The PAM50 subtype remains a significant prognostic indicator of recurrence in the context of standard therapies used in community practice today. Citation Format: Philip Bernard, Erin Weltzien, Inge Stijleman, Candyce H. Kroenke, Carole Davis, Marilyn L. Kwan, Charles Quesenberry, Adrienne Castillo, Laurel Habel, Rachel Factor, Larry Kushi, Carol Sweeney, Bette Caan. PAM50 breast cancer subtyping and risk of recurrence in a population-based cohort. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3650. doi:10.1158/1538-7445.AM2013-3650

Observational studies suggest that cancer incidence is lower in patients with type 2 diabetes treated with metformin, relative to patients treated with other diabetes medications. The few studies to examine metformin use in relation to lung cancer risk, however, have produced inconsistent results. To clarify this relationship, we conducted a retrospective cohort study of 50,722 adults aged 40 years or older included in the Kaiser Permanente Northern California Diabetes Registry, who completed a mailed survey administered in 1994 to 1996 on health-related traits and behaviors. All prescribed diabetes medications were identified for each patient from pharmacy records. Patients were followed for lung cancer from 1997 to 2009, with a median follow-up of 8.1 years. Using Cox regression, with age as the time scale, rate ratios (RR) and 95% confidence intervals (CI) were calculated to estimate risk of lung cancer associated with new use of metformin, adjusted for sex, race/ethnicity, smoking status, income, education level, diabetes duration, baseline measures of hemoglobin A1c and creatinine, and ever use of other types of diabetes medications. Measures of metformin use, along with ever use for each type of diabetes medication, were modeled as time-dependent variables. During 389,887 person-years of follow-up, 685 patients were diagnosed with lung cancer. The adjusted RR for ever use of metformin was 0.95 (95% CI, 0.79-1.14). For duration of metformin use, the adjusted RRs for Citation Format: Lori C. Sakoda, Ninah S. Achacoso, Charles P. Quesenberry, Tiffany Peng, Samantha F. Ehrlich, Assiamira Ferrara, Laurel A. Habel. Metformin use and risk of lung cancer in patients with diabetes. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2274. doi:10.1158/1538-7445.AM2013-2274

Background: Lifestyle and reproductive factors known to influence estrogen levels have been associated with breast cancer (BC) risk and prognosis. However, it is unknown whether associations vary by tumor molecular subtype at diagnosis. Methods: A case-cohort study population of 1,439 women with invasive BC was sampled from two prospective cohort studies: LACE (AJCC stage I (≥1 cm), II, IIIA diagnosed 1997-2000), and Pathways (any invasive stage ≥0.5 cm diagnosed 2006-2008). Molecular subtype (luminal A, luminal B, HER2-enriched, basal-like) was determined by RT-PCR of RNA extracted from samples from formalin-fixed, paraffin-embedded primary tumor blocks and assayed for expression of PAM50 genes. Information on hormone replacement therapy (HRT), oral contraceptive use (OC), reproductive factors (age at menarche, parity, age at first birth, breastfeeding), alcohol intake, and body mass index (BMI) were obtained from baseline participant questionnaires, and an index of lifetime hormonal exposure [low (0-1, reference), medium (2-4), high (5-6)] was created based on their expected influence on circulating estrogen. Specifically, any factor associated with more estrogen exposure was assigned a score of 1 (e.g., HRT yes, OC no, BMI≥25 kg/m2) while less exposure was assigned a score of 0 (e.g., HRT no, OC yes, BMI Results: There were 370 recurrences and 501 total deaths (269 BC-related) over a median 6.6±3.9 years. In the overall study population, a higher hormonal exposure index was associated with an increased risk of breast cancer death (p for trend=0.02). By molecular subtype, the highest tertile of the hormonal index was associated with both recurrence (HR=3.8; 95% CI: 1.0, 14.5) and BC-specific death (HR=7.3; 95% CI: 1.7, 31.7), compared with a low hormonal index, among the luminal A subtype. In contrast, no associations of the hormonal index with recurrence and BC-specific death were observed in the luminal B, HER2-enriched, or basal-like subtypes (p for homogeneity=0.01). A high hormonal index was not associated with all-cause mortality overall and by subtype. Discussion: Traditional hormonal risk factors for BC showed differing associations with prognosis and survival depending on tumor molecular subtype of the primary BC. Subtypes that are treated similar based on conventional prognostic indicators such as ER status (luminal A and luminal B) may have differing associations between risk factors and prognosis. Funded by NCI R01 CA129059. Citation Format: Marilyn L. Kwan, Carol Sweeney, Laurel A. Habel, Erin Weltzien, Adrienne Castillo, Carole Davis, Rachel E. Factor, Candyce H. Kroenke, Kaylynn Shakespear, Charles P. Quesenberry, Inge J. Stijleman, Philip S. Bernard, Lawrence H. Kushi, Bette J. Caan. Hormonal risk factors and breast cancer prognosis and survival by PAM50 molecular subtype. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr LB-29. doi:10.1158/1538-7445.AM2013-LB-29

Background: African-Americans have poorer breast cancer (BC) prognosis, and Asians and Hispanics have better prognosis, compared with Whites. The Black-White difference is unexplained and has been attributed to diagnosis with less treatable tumor subtypes in African-Americans. However, little research has examined whether race-survival differences exist by breast cancer subtype. Therefore, we examined associations between race and BC survival by PAM50 breast cancer subtypes (luminal A, luminal B, basal-like, HER2 enriched) in a prospective cohort of 1,282 breast cancer survivors from the Life After Cancer Epidemiology and Pathways cohorts. Methods: Self-reported race was obtained at study entry from mailed questionnaires. 1 mm punches were obtained from areas of representative tumor in formalin-fixed, paraffin-embedded tumor blocks. Expression of the PAM50 genes for molecular subtype was determined by RT-PCR of extracted RNA. After a median 6.3 years of follow-up (range 0.3-15.5 years), 213 deaths, with 118 from breast cancer, were reported. Delayed entry Cox proportional hazard models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) of BC mortality, controlling for time from diagnosis to enrollment, socioeconomic status, BC severity, BC subtype, treatment, and other known prognostic factors. Logistic regression was used to evaluate associations between race and subtype. Survival analyses stratified by subtype were adjusted for age, time from diagnosis to enrollment, BC severity, and BC treatment. Results: Consistent with previous research, adjusted for stage and breast cancer treatment, BC mortality was significantly higher in African-Americans (HR=2.90, 95% CI: 1.74-4.86) and lower in Latinas and Asians (HR=0.51, 95% CI: 0.26-0.99), compared with Whites. In addition, compared with Whites, African-Americans had a lower likelihood of the luminal A (OR=0.61, 95% CI: 0.38-0.98) and luminal B (OR=0.43, 95% CI: 0.23-0.82) subtypes and a greater likelihood of the less treatable basal subtype (OR=2.93, 95% CI: 1.94-4.44). Asians were less likely to be diagnosed with the basal subtype (OR=0.41, 95% CI: 0.23-0.71) and somewhat more likely to be diagnosed with the HER2-enriched subtype (OR=1.47, 95% CI: 0.97-2.21). Stratified by subtype, African-Americans had poorer prognosis among those with luminal A (HR=2.64, 95% CI: 0.93-7.49), luminal B (HR=2.20, 95% CI: 0.43-11.26), basal-like (HR=1.66, 95% CI: 0.78-3.54), and HER2 enriched (HR=3.25, 95% CI: 1.04-10.15) subtypes than Whites. Conclusion: African-Americans had worse breast cancer survival than other racial/ethnic groups and had less treatable types of breast cancer. However, breast cancer mortality was higher in African-Americans across tumor subtypes, suggesting that the Black-White survival difference may not be attributable to differential diagnosis by subtype. Citation Format: Candyce Kroenke, Marilyn Kwan, Philip Bernard, Adrienne Castillo, Carole Davis, Rachel Factor, Laurel Habel, Larry Kushi, Charles Quesenberry, Kaylynn Shakespear, Inge Stijleman, Carol Sweeney, Erin Weltzien, Bette Caan. Race and breast cancer prognosis by PAM50 subtype in the LACE and Pathways studies. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 131. doi:10.1158/1538-7445.AM2013-131

BackgroundPolysomy 17 is found in breast cancer and may complicate interpretation of HER2 results. HER2 status by FISH and quantitative RT-PCR were determined in a case-control study that quantitated HER2 mRNA levels in polysomy 17 patients and that examined breast cancer survival in tamoxifen (TAM) treated and untreated patients.MethodsA Kaiser Permanente nested case-control study was performed in node-negative breast cancer patients diagnosed from 1985-94 who were not treated with chemotherapy. Cases died of breast cancer prior to 2002; up to three controls were matched for each case, with a total of 568 unique patients. HER2 was measured by quantitative RT-PCR by Oncotype DX; pre-defined cutoffs ≥11.5 expression units (positive), ≥10.7 to 2.2 (positive), 1.8 to 2.2 (equivocal), Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 6004.

BACKGROUND: Most HER2 data have been generated from populations of invasive breast cancer (IBC) patients with selected patient and tumor characteristics (clinical trials, referral centers). The average age of women in US adjuvant trastuzumab trials was 49. Pre 2000, clinical trials of adjuvant chemotherapy included 25-40% HER2+ cases. We used an unselected population of women with IBC to estimate the age-specific incidence rates of HER2+ IBC and compare patient and tumor characteristics of HER2+ and HER2- disease.METHODS: Among female members of Kaiser Permanente Northern California (KPNC) 20 years or older (n=1.22 million), we identified all those diagnosed with a new primary IBC in 2000-2006 (n=16,975). During this period, all IBCs were routinely tested for HER2 by KPNC's regional IHC laboratory (center of excellence for HercepTest, CLIA licensed and CAP certified) and those that were IHC 2+ were sent for FISH testing; IHC 3+ or IHC 2+/FISH+ (ratio ≥ 2.0) were considered HER2+. Demographic and tumor characteristics were derived from the KPNC tumor registry. *Complete HER2 testing available for 94% of IBCs.RESULTS: Incidence of HER2+ IBC increased less dramatically with age and peaked 20 years earlier than incidence of all IBC (Table 1). Among those with IBC, the percent that were HER2+ decreased with age. Of all HER2+ IBCs, 69% were ≥50 years old and 42% were ≥60 years old.Incidence of all IBC and incidence of HER2+ diseaseAge GroupPerson-YrsNo. of IBCsIBC rate/100KHER2+ IBCs*%HER2+HER2+IBC rate/100K20-39288153371024.6416424.75.6940-4917402132680154.0049819.728.6250-5915535284354280.2759914.538.5660-699677624280442.2649412.351.0570-796676013318497.002718.740.5980+3666531633445.381429.638.73ALL817729016975207.59216813.626.51 The percent of tumors that were HER2+ increased with stage (local=11%, regional=18%, distant =25%), tumor differentiation (well =2%, moderate =12%, poor=26%, undifferentiated=30%), and generally with size (5 cm=20%). ER- and PR- tumors were more likely to be HER2+ (of all ER- tumors, 29% were HER2+; of all PR- tumors, 25% were HER2+; of all ER+ tumors, 10% were HER2+; of all PR+ tumors, 9% were HER2+). Among all IBC patients, percent of disease that was HER2+ was highest for Asians (20%) and lowest for Caucasians (12%).CONCLUSION: In a large, unselected, population of invasive breast cancer patients, the incidence of HER2+ disease was highest among women aged 60-69 years. Among breast cancer patients, HER2+ disease was highest in Asians. The percent HER2+ was 13.6% for all breast cancer patients in this population based cohort and highest (24.7%) in the 20-39 year age group. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 3058.

Background: Four studies have reported that a two-gene ratio (HOXB13:IL17BR) may be prognostic and/or predictive of tamoxifen benefit among early stage breast cancer patients. More recently, a 5-gene (BUB1B, CENPA, NEK2, RACGAP1, RRM2) molecular grade index (MGI) was reported to predict clinical outcome. We evaluated the performance of these two gene signatures among an independent population of lymph node-negative breast cancer patients from a community hospital setting.Methods: A case-control study was conducted among 4,964 Kaiser Permanente patients diagnosed with node-negative invasive breast cancer from 1985-94 and not treated with adjuvant chemotherapy (Habel, 2006). Archived tumor tissues were available on 169 cases (breast cancer deaths) and 335 matched controls and, using RT-PCR, were analyzed for expression levels of the above 7 cancer-related genes and 4 reference genes. Logistic regression methods were used to estimate the 10-year risk and the relative risk (RR) of breast cancer death associated with pre-specified risk categories based on cross-classification of HOXB13:IL17BR and MGI scores.Results: Approximately 50% of patients were classified as low risk by the HOXB13:IL17BR+MGI risk classifier. At 10 years, the risks of breast cancer death for ER+, tamoxifen-treated patients were 3.9% (95% confidence interval (CI) 2.2% - 5.5%), 5.6% (95% CI 2.6% - 8.5%), and 15.6% (95% CI 8.7% - 21.9%) for those in the low, intermediate and high risk groups, respectively. They were 5.9% (95% CI 4.2% - 7.5%), 15.2% (95% CI 8.3% - 21.5%), and 16.5% (95% CI 10.4% - 22.2%) for ER+ patients not treated with tamoxifen. After adjusting for tumor size and grade, the RRs of breast cancer death associated with high vs. low HOXB13:IL17BR+MGI risk categories were attenuated but remained elevated in both treatment groups; the RR continued to be statistically significant in tamoxifen treated patients only.Conclusion: In this population of lymph-node negative patients not treated with adjuvant chemotherapy, the HOXB13:IL17BR+MGI risk classifier was predictive of risk of breast cancer death and appeared to provide risk information beyond standard prognostic factors. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 4034.