Rosemary D. Cress

Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: A population-based study from the California Cancer Registry

Tumor markers are becoming increasingly important in breast cancer research because of their impact on prognosis, treatment, and survival, and because of their relation to breast cancer subtypes. The triple‐negative phenotype is important because of its relation to the basal‐like subtype of breast cancer.

Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: a population-based study from the California cancer Registry.

Tumor markers are becoming increasingly important in breast cancer research because of their impact on prognosis, treatment, and survival, and because of their relation to breast cancer subtypes. The triple‐negative phenotype is important because of its relation to the basal‐like subtype of breast cancer.

Incidence of cutaneous melanoma among non-Hispanic Whites, Hispanics, Asians, and Blacks: an analysis of California Cancer Registry data, 1988-93

Cutaneous malignant melanoma occurs less frequently among non-Whitepopulations than among Whites. As a result, little is known about theincidence and epidemiology of melanoma among other race/ethnicity groups.Data from the California Cancer Registry (United States) among 879 Hispanic,126 Asian, and 85 Black men and women diagnosed with melanoma in 1988-93 wereanalyzed and compared with data for 17,765 non-Hispanic White cases. Average,annual, age-adjusted incidence rates per 100,000 population were 17.2 for men(M) and 11.3 for women (W) for non-Hispanic Whites; 2.8 (M), 3.0 (W) forHispanics; 0.9 (M), 0.8 (W) for Asians; and 1.0 (M), 0.7 (W) for non-HispanicBlacks. Among men, melanoma occurred on the lower extremity for 20 percent ofHispanics, 36 percent of Asians, and 50 percent of Blacks compared with ninepercent of non-Hispanic Whites, with similar but less pronounced differencesin site distribution by race/ethnicity for women. Among men, melanoma wasdiagnosed after it had metastasized to a remote site for 15 percent ofHispanics, 13 percent of Asians, and 12 percent of Blacks, compared with sixpercent of non-Hispanic Whites. Among women, seven percent of Hispanics, 21percent of Asians, and 19 percent of Blacks were diagnosed with late-stagemelanoma compared with four percent of non-Hispanic Whites. Althoughhistologic type was not specified for nearly half of the cases, Hispanic,Asian, and Black patients were more likely than non-Hispanic White patientsto have been diagnosed with acral lentiginous melanoma. MelanomaamongHispanics, Asians, and Blacks differs in incidence, site distribution,stage at diagnosis, andhistologic type from melanoma among non-HispanicWhites, and identification of risk factors for melanoma in theserace/ethnicity groups would elucidate further the role of sun and otherfactors in the etiology of melanoma.

Understanding the burden of human papillomavirus‐associated anal cancers in the US

Anal cancer is an uncommon malignancy in the US; up to 93% of anal cancers are associated with human papillomavirus.

Malignant melanoma in pregnancy. A population-based evaluation.

For many years, there has been controversy in the medical community regarding the correlation of female hormonal factors with the outcome of women with malignant melanoma. There have been multiple reports that women with high hormone states, such as pregnancy, had thicker tumors and/or a worse prognosis compared with a group of control women.

Secular changes in colorectal cancer incidence by subsite, stage at diagnosis, and race/ethnicity, 1992-2001.

Cancers of the colon and rectum are the third most common malignancy among males and females in the United States, although incidence and mortality have declined in recent years. We evaluated recent trends in colorectal cancer incidence in the United States by subsite and stage at diagnosis.

Ovarian cancer: Can we make the clinical diagnosis earlier?

Patients with ovarian cancer often report having symptoms for months before diagnosis, but such findings are subject to recall bias. The aim of this study was to provide an objective evaluation of symptoms that precede a diagnosis of ovarian cancer.

Thyroid cancer in pregnancy

Objective: To compare stage at diagnosis, treatment and survival among pregnant women with thyroid cancer to non‐pregnant women with thyroid cancer, and to assess the impact of treatment on maternal and perinatal outcomes. Methods: A database containing maternal and newborn discharge records linked to the California Cancer Registry was queried to obtain information on all thyroid cancers from 1991–1999. Women with thyroid cancer occurring during pregnancy were compared to age‐matched non‐pregnant women with thyroid cancer. Results: 595 cases of thyroid cancers were identified (129 antepartum and 466 postpartum). About 64% of thyroid cancers were diagnosed at stage 2 among pregnant women versus 58% among non‐pregnant controls. The odds of thyroid cancer were 1.5 times higher among Asian/Pacific Islanders than among Non‐Hispanic White women. Pregnancy had no significant effect on mortality after diagnosis of thyroid cancer. Thyroidectomy during pregnancy was not associated with adverse maternal or neonatal outcomes. Conclusions: Thyroid cancer discovered during or after pregnancy does not appear to have a significant impact on the prognosis of the disease.

Evidence of Poorer Survival in Pregnancy-Associated Breast Cancer

OBJECTIVE: To compare stage distribution, tumor characteristics, and survival outcome in pregnancy-associated and non–pregnancy-associated breast cancer, and to evaluate pregnancy as a risk factor for mortality in breast cancer. METHODS: The California Cancer Registry (1991–1999) was linked with the California Patient Discharge Data Set to identify women with breast cancer occurring within 9 months before or 1 year after an obstetric delivery. Age-matched, non–pregnancy-associated breast cancer controls were also identified. Demographics, cancer stage, tumor size, histology, hormone receptor status, type of treatment, and survival were reviewed and compared. Predictive factors for death from breast cancer were identified using proportional hazards modeling. RESULTS: Seven hundred ninety-seven pregnancy-associated breast cancer cases were compared with 4,177 non–pregnancy-associated breast cancer controls. Pregnancy-associated breast cancer cases were significantly more likely to have more advanced stage, larger primary tumor, hormone receptor negative tumor, and mastectomy as a component of their treatment. In survival analysis, pregnancy-associated breast cancer had a higher death rate than non–pregnancy-associated breast cancer (39.2% compared with 33.4%, P=.002). In a multivariable analysis, advancing stage (2.22–10.76 times the risk of death for stages II–IV), race (African Americans had 68% increased risk of death over non-Hispanic whites), hormone receptor–negative tumors (20% increased risk of death over receptor-positive tumors), and pregnancy (14% increased risk of death over nonpregnant women) all were significant predictors of death. CONCLUSION: Pregnancy-associated breast cancer presented with more advanced disease, larger tumors, and increased percentage of hormone receptor–negative tumors. When controlled for stage, race, and hormone receptor status, pregnancy-associated breast cancer cases had a slightly higher risk of death, even when only localized-stage disease was considered. LEVEL OF EVIDENCE: II

Completeness of information on adjuvant therapies for colorectal cancer in population-based cancer registries

Background. Population-based cancer registries represent a potentially valuable tool to evaluate treatment; however, information on the completeness of registry treatment data is sparse. Objective. To evaluate the completeness of registry treatment data for patients with colorectal cancer and to identify predictors of complete reporting. Research Design. We surveyed physicians or reviewed office records of 1956 northern California patients diagnosed with colorectal cancer during 1996 to 1997 to assess the completeness of registry data regarding use of adjuvant chemotherapy and radiation therapy. Results. For patients with a record of receipt of chemotherapy in either the registry or physician survey, information was in the original registry records for 82.0%. In the multivariate analysis, completeness of chemotherapy reporting was lower for patients aged 65 to 74, those with colon cancer, and those treated in teaching hospitals; and higher for patients treated in hospitals that are part of a large health maintenance organization (HMO). For patients with a record of receipt of radiation therapy, information was in the original registry records for 90.2%. In the multivariate analysis, completeness of radiation therapy reporting was higher for patients aged 18 to 54 and those treated in HMO hospitals. Conclusions. Because the completeness of the registry treatment data varied by patient and hospital characteristics, use of registry data without supplementation could bias estimates of the proportion of patients treated, and of the patient and provider characteristics associated with treatment. Enhanced cancer registry data could be a valuable component of population-based cancer data systems for assessing quality of cancer care.