Excess risk of subsequent primary cancers among colorectal carcinoma survivors, 1975-2001.

BACKGROUND: Studies of persons with colorectal cancer have reported increased risk of subsequent primary cancers. Results have not been consistent, however, and there is little information about such risk in specific races and ethnic populations. METHODS: Using 1975-2001 data from the Surveillance, Epidemiology, and End Results (SEER) Program, we assembled 262,600 index cases of colorectal carcinoma to assess the occurrence of subsequent primary cancers in 13 noncolonic sites. Observed (O) subsequent cancers were compared with those expected (E) based on age-/sex-/race-/year-/site-specific rates in the SEER population. The standardized incidence ratio (SIR) and the absolute excess risk (AER) represent 'O / E' and 'O - E,' respectively. RESULTS: Colorectal carcinoma patients had significantly elevated SIRs for small gut, stomach (males), kidney, and corpus uteri cancers, ranging from 1.13 for stomach cancer in males to 3.45 for small gut cancer in females. Elevated SIRs for additional sites were seen in certain population subgroups: pancreas and ovary in persons aged <50 years, and prostate in black males. The excess burden, as assessed by AER, was notable for prostate cancer in black males and for corpus uteri cancer in females aged <50 years (26.5 and 9.5 cancers per 10,000 person-years, respectively), and it persisted beyond 5 years of follow-up. CONCLUSIONS: Although significantly elevated SIRs were found for several cancers, the excess burden was notable only for cancer of the prostate in black males and of the corpus uteri in females under age 50.

Hierarchical modeling and other spatial analyses in prostate cancer incidence data.

BACKGROUND: State central cancer registries are often asked to respond to questions about the spatial distribution of cancer cases. Spatial analysis methods and technology are evolving rapidly, and can be a considerable challenge to registries that do not have staff with training in this area. The purpose of this article is to describe a general methodological approach that potentially might be a starting point for many cancer registry spatial analyses at the county level. METHODS: Prostate cancer incident cases (N=31,159) from the Louisiana Tumor Registry from 1988 to 1999 were used for illustrative purposes. To explore spatio-temporal patterns, analyses focused on four time periods, each 3 years in length: 1998-1990, 1991-1993, 1994-1996, and 1997-1999. For each time period, race-specific (white and black), direct age-adjusted incidence rates and indirect standardized incidence ratios (SIRs) were calculated, smoothed using Bayesian methods, and assessed for evidence of spatial autocorrelation using global and local Moran's I. Hierarchical generalized linear models (HGLM) were fitted to identify significant covariates. Clusters of elevated and lower rates were identified using a spatial scan statistic (SaTScan). RESULTS: Temporal trends in SIRs in both race groups were consistent with the introduction of prostate specific antigen (PSA) testing in Louisiana during the late 1980s and early 1990s, but possibly with a lag in black males. Clusters of lower than expected values were observed for white males in the central (p=0.001) and southeastern coastal areas (p=0.001), and to a greater extent for black males in the central (p=0.001), southwestern and southeastern coastal parishes (p=0.001). CONCLUSIONS: Mapping disease occurrence by time period is an effective way to explore spatio-temporal patterns. HGLM models and software are available to control for covariates and for unstructured and spatially structured variability that may confound spatial variability patterns.

Subsite-specific colorectal cancer incidence rates and stage distributions among Asians and Pacific Islanders in the United States, 1995 to 1999.

OBJECTIVE: This study examined subsite-specific colorectal cancer incidence rates and stage distributions for Asians and Pacific Islanders (API) and compared the API data with data for Whites and African Americans. METHODS: Data included 336,798 invasive colorectal cancer incident cases for 1995 to 1999 from 23 population-based central cancer registries, representing about two thirds of API population in the United States. Age-adjusted rates, using the 2000 U.S. standard population, and age-specific rates and stage distributions were computed by anatomic subsite, race, and gender. All rates were expressed per 100,000. SEs and rate ratios were calculated for rate comparison. A significance level of 0.05 was used for all analyses. RESULTS: Overall, age-adjusted colorectal cancer incidence rates were significantly lower in API than in Whites and African Americans across anatomic subsites, particularly for proximal colon cancer in which rates were 40% to 50% lower in API males and females. Exception to this pattern was the significantly (10%) higher rectal cancer incidence rate in API males than in African American males. The incidence patterns by anatomic subsite within API differed from those of Whites and African Americans. Among API, the rate of rectal cancer (19.2 per 100,000) was significantly higher than the rates of proximal (15.2 per 100,000) and distal (17.7 per 100,000) colon cancers in males, with little variations in rates across anatomic subsites in females. In contrast, among White and African American males and females, proximal colon cancer rates were over 25% higher than the rates of distal colon and rectal cancers. Increases in age-specific rates with advancing age were more striking for proximal colon cancer than for distal colon and rectal cancers in Whites and African Americans, while age-specific rates were very similar for different subsites in API with parallel increases with advancing age, especially in API males. Similar to Whites and African Americans, in API, proximal colon cancers (32% to 35%) were also less likely to be diagnosed with localized stage compared with distal colon (38% to 42%) and rectal (44% to 52%) cancers. CONCLUSION: The patterns of subsite-specific colorectal cancer incidence in API, especially API males, differ from those of Whites and African Americans. Similar to Whites and African Americans, lower percentage of localized disease in API for proximal colon cancer than for distal colon and rectal cancers was also observed.

A vision for cancer incidence surveillance in the United States.

A comprehensive framework for cancer surveillance should span the entire lifespan and be capable of providing information on risk, burden, disparity, cost, cancer care, survival, and death. Cancer incidence, the point in the continuum when an individual is diagnosed with cancer, has a strong, well-developed system to produce information about newly diagnosed cancer cases. However, in the future, this system must be enhanced and integrated with other cancer surveillance networks and other systems to provide timely information on the burden of newly diagnosed patients with respect to various cross-cutting population characteristics (e.g., social, economic, race/ethnic, urbanicity, or access to care) to define, monitor, and reduce incidence and various disparities noted among population groups. Collaboration in data collection, standard setting, surveillance activities, research, education and training, data use, and advocacy among all registries and national programs will be important to the continued success of the cancer incidence surveillance system. The cancer registry is an integral part of the infrastructure to reduce the burden of cancer, including the numbers of newly diagnosed cases.

Patterns of treatment for ductal carcinoma in situ of the breast in Louisiana, 1988-1999.

Data from numerous studies show that lumpectomy (breast-conserving therapy) plus radiation therapy provides survival equivalent to that following mastectomy (either modified radical or radical mastectomy) for patients with ductal carcinoma in situ (DCIS). According to the data from the National Cancer Data Base and the Surveillance Epidemiology End Results (SEER) Program, use of lumpectomy among female DCIS patients has increased dramatically over the last decade. This study examined population-based trends in treatment for DCIS among Louisiana women and compared the trends with the SEER data. Our data revealed that the percentage of the DCIS patients who received a lumpectomy increased from 34.3% in 1988-1991 to 53.7% in 1996-1999 in Louisiana (p<0.05) while DCIS patients who received a modified radical mastectomy decreased from 51.7% to 26.1% (p<0.05). Increasing use of lumpectomy was seen across all races, age groups, rural/urban areas, and poverty-level areas. Utilization of lumpectomy was about the same for white and African-American women but varied by age group, rural/urban area, and poverty level. Female DCIS patients residing in rural areas or high poverty level areas were less likely to receive a lumpectomy than those residing in urban or affluent areas. Among the patients who had a lumpectomy, 34.4% received post-lumpectomy radiotherapy in the first study period (1988-1991) and 49.7% in the last study period (1996-1999). In Louisiana, utilization of post-lumpectomy radiotherapy decreased with advancing age. Despite the increase in use of lumpectomy, its utilization remained approximately 10% lower than in the SEER areas throughout the study period. A similar deficit was observed for post-lumpectomy radiation therapy.

Cancer incidence in adolescents and young adults in the United States, 1992-1997.

PURPOSE: To examine cancer incidence patterns among adolescents and young adults in the United States. METHODS: Cancer incidence data from 26 population-based central cancer registries for 1992-1997 were used. Individual cancers were grouped into specific diagnostic groups and subgroups using an integrated classification scheme. The integrated scheme was developed for this study and was based on the most commonly used schemes in population-based epidemiologic studies: Surveillance, Epidemiology, and End Results Program's site groups, International Classification of Childhood Cancer, and International Agency for Research on Cancer's Histological Groups for Comparative Studies. Percent distributions and age-specific incidence rates per million population were computed for adolescents (aged 15-19 years) and young adults (aged 20-24 years) by gender. RESULTS: The data for 26,010 cancer cases were examined. Among 15-19-year-olds, the five most common cancers were Hodgkin's disease, leukemia, cancer in the brain and other nervous system, bone cancer, and non-Hodgkin's disease. Among 20-24-year-olds, the five most common cancers were Hodgkin's disease, testicular cancer, thyroid cancer, melanoma of the skin, and leukemia. The proportions and rates of the histologic subtypes for most of the common cancers changed with advancing age. For example, among 15-19-year-olds, acute lymphocytic leukemia accounted for approximately 60% of leukemias in males and 50% in females. Among 20-24-year-olds, however, the corresponding percentages of acute lymphocytic leukemia were 37% in males and 31% in females. For ovarian cancer, the germ cell tumor was the most common subtype (54.6% of all ovarian cancers) among 15-19-year-olds. In contrast, ovarian carcinoma was the predominant subtype (70.4%) among 20-24-year-olds. For both age groups, the incidence rates of nodular Hodgkin's disease, melanoma of the skin, and thyroid cancer were significantly greater in females than in males. CONCLUSIONS: Cancer incidence patterns among adolescents and young adults are distinctive. In these age groups, a transition from predominantly pediatric histologic subtypes to adult subtypes was observed for Hodgkin's disease, leukemia, ovarian cancer, and soft tissue sarcoma. Gender differences were found for Hodgkin's disease, melanoma of the skin, and thyroid cancer.

Childhood cancer in Louisiana 1988-1996.

Utilizing data from the Louisiana Tumor Registry, cancer incidence among children younger than 15 years of age is presented by major cancer type, according to the primarily histology-based International Classification of Childhood Cancer scheme. Cases include those diagnosed and/or treated at any hospitals and medical facilities in Louisiana, St. Jude Children's Research Hospital in Memphis, M.D. Anderson in Houston, and from neighboring states. Rates were age-adjusted, presented as rates per million, and were compared to the combined rates of the Surveillance, Epidemiology, and End Results (SEER) Program. The significance of rate differences were assessed at 0.05 level. From 1988-1996, about 125 children were diagnosed with cancer each year. In general, rates are higher in younger than older children, males than females, and white children than African-American children. The five most common childhood cancers are: leukemias (28% of total cases), central nervous system malignancies (22%), lymphomas (13%), renal tumors (8.4%), and soft tissue sarcomas (7.6%). Major findings of these cancers and their associated risk factors are presented.