Sharmila Roy-Chowdhury

Quantitating the impact of stage migration on staging accuracy in colorectal cancer.

BACKGROUND The number of nodes retrieved is a powerful predictor of survival in node-negative colorectal cancer (CRC). Whether this is because of improved staging or improved surgery, or both, is unclear. We sought to quantitate the impact of stage migration in node-negative colorectal cancer. STUDY DESIGN Between January 1994 and December 2003, 7,192 patients in Region 5 of the California Cancer Registry, diagnosed with node-negative or node-positive colorectal cancer, were reviewed. The number of nodes examined, node-positive rate, and disease-specific survival (DSS) were analyzed. RESULTS The mean number of nodes examined was 9.3 (range 0 to 89 nodes). The 5-year DSS was 84.5% for N0, 65.2% for N1, and 46.8% for N2 disease. The 5-year DSS difference for those who had more than 12 nodes retrieved was 87.3% (95% CI, 85.2% to 89.3%) and for those with 0 to 3 nodes retrieved, 83.7% (95% CI, 80.6% to 86.82%; p = 0.0009). As the number of retrieved nodes increased, the risk of understaging patients decreased. For 0 to 3 nodes, 78.3% of patients were N0; for 4 to 7 nodes, 67.6%; 8 to 9 nodes, 62.1%; 10 to 12 nodes, 59.5%; and only 57.2% for more than 12 nodes examined. Using more than 12 nodes as the definition of an adequate lymphadenectomy and staging, for apparently N0 patients with 0 to 3 nodes retrieved, DSS survival would be affected by approximately 5%. CONCLUSIONS Differences in outcomes between CRC patients with limited numbers of nodes harvested and those with more than 12 nodes harvested are substantial. Stage migration alone can explain the entire DSS difference between patients with more than 12 nodes retrieved and those with smaller numbers of nodes retrieved.

Disparities in BRCA testing: when insurance coverage is not a barrier

BACKGROUND Strategies to reduce the risk of developing breast and ovarian cancer in carriers of deleterious BRCA 1 and 2 mutations are readily available. However, many people who are at high risk of having these genetic mutations are reluctant to obtain the test. We sought to identify factors associated with choice of testing. METHODS We performed a retrospective cohort review of high-risk patients referred to a multidisciplinary breast health center for BRCA testing between January 2001 and March 2008. Demographic variables were compared by using logistic regression between those who completed genetic testing and those who did not. RESULTS A total of 213 patients were referred for BRCA testing. The mean age was 49.2 years (range, 16-84 y). Five patients were male. The majority of individuals (63.4%) were white, 15% were Hispanic, 6.6% were black, and 4.7% were Asian. Insurance coverage for testing was available in 91.1% of patients, of whom 49.2% had private insurance, 26.7% had managed care insurance, and 24.1% had government-sponsored insurance. A total of 111 patients (52.1%) underwent testing. On multivariate analysis, patients were significantly more likely to complete testing if they had a personal history of breast cancer (73.0% of tested patients) (P = .005) and had at least some college education (61.3%) (P = .03). There were no statistically significant differences in tested versus untested groups by age, race, language, family history, parity, marital status, religion, socioeconomic status, or insurance status. Of patients whose insurance plans offered coverage for genetic testing, 51.4% underwent testing and 48.6% did not (P = not significant [NS]). Of those who had no insurance coverage for testing, 41.2% underwent testing and 58.9% did not (P = NS). CONCLUSIONS Our data show that half of those patients at risk for carrying a BRCA mutation do not undergo testing. Insurance coverage for genetic testing does not influence the decision to test. Developing counseling instruments that explain the benefits of testing to unaffected high-risk individuals or targeted to those with a high school level education may be a strategy to improve testing rates.

The prognostic significance of the ratio of positive to total nodes in colorectal cancer

4045 Background: Patients with node positive colorectal cancer (CRC) have a grave prognosis. Current AJCC staging distinguishes between those with 1–3 positive lymph nodes (N1) and those with 4 or more (N2). We sought to determine whether the ratio of positive to total nodes examined (LNR) was a more powerful predictor of outcome than N staging. Methods: 2,636 patients diagnosed between January 1, 1994 to December 31, 2003 with stage III CRC from Region 5 of the California Cancer Registry were analyzed. LNR was stratified into quintiles of <0.19, 0.2–0.39, 0.4–0.59, 0.6–0.70 or 0.8–1.0. The major endpoint was actuarial 5-year disease specific survival (DSS). Significant differences in DSS between groups were determined by the log-rank test. A multivariate Cox proportional hazard regression was performed for all patients to determine whether N stage, LNR, number of lymph nodes examined, or number of positive lymph nodes were independent prognostic factors. Results: The median age was 71 with women making u...

Cytokines and Tumor Angiogenesis

The term angiogenesis was first coined by John Hunter, a British surgeon, to describe blood vessel formation in reindeer antlers in 1787. The first histologic description of angiogenesis was presented by Arthur Tremain Hertig in 1935, detailing the formation of placental blood vessels. In 1939, based on observations of tumors transplanted to transparent chambers in in vivo animal models, A. G. Ide and colleagues proposed the existence of a tumor-derived factor capable of stimulating blood vessel formation (1). In 1945, G. H. Algire and his co-workers at the National Cancer Institute (NCI) recognized that the blood supply of tumors was derived from the host and was essential for tumor growth (2). The findings of Algire et al. offered the first evidence supporting the association between tumor vasculature and tumor growth. In 1948, I. C. Michaelson postulated that a diffusible angiogenic factor produced by the retina conferred the neovascularization phenotype observed during proliferative diabetic retinopathy and named this soluble factor as “Factor X” (3). Twenty years later, Greenblatt and Shubik (4) and Ehrmann and Knoth (5) demonstrated independently that tumor cells could stimulate vasoproliferation through transfilter diffusion experiments and confirmed the notion of a diffusible angiogenic factor elaborated by tumor cells. In 1971, Judah Folkman further refined the view on the relationship between tumor and its vasculature by hypothesizing that the growth of cancer from a few cell layers thick into a gross tumor requires angiogenic stimuli which are mediated by substances produced by the tumor. He proposed that anti-angiogenic therapy, therefore, might be an effective approach to treating human cancers (6–8). This view on tumor angiogenesis eas provocative and contrary to the prevailing thoughts of the time. Despite not being accepted by the scientific community, Dr. Folkman endeavored along this line of research and initiated the effort that lead to the isolation of the first angiogenic cytokine, basic fibroblast growth factor (bFGF) (9). Findings from his work have come to form the foundation in the molecular understanding of tumor angiogenesis and Dr. Folkman is often regarded as the “father of angiogenesis”.