Gestational trophoblastic neoplasia in American Indians.

OBJECTIVE: To analyze gestational trophoblastic neoplasia (GTN) trends among American Indians (AI) using population-based data. STUDY DESIGN: GTN incidence, by race and age, was calculated using data collected by the New Mexico Tumor Registry over 29 years (1973-2001). Live birth, pregnancy and women at risk were tabulated using data derived from the state's vital record annual reports and from the registry. Statistical methods included trends analysis and Poisson regression. There is no national registry in the United States for all GTN. Therefore, the Surveillance, Epidemiology and End Results (SEER) database was used to identify choriocarcinoma cases in American Indians between 1973 and 1999. RESULTS: Within New Mexico, 1,082 cases of GTN were identified among 752,374 live births and 904,831 pregnancies, with ratios of 1:695 and 1:836, respectively, affecting 234 AI, 355 non-Hispanic whites (NHW), 463 Hispanic whites (HW) and 30 other nonwhites. Ratios per live births (pregnancy), respectively, were significantly higher in AI (AI 1:439 [1:487], NHW 1:739 [1:949], HW 1:783 [1:903]), as was age-adjusted incidence per 100,000 woman-years (AI 10.62, NHW 3.53, HW 5.15; all P<.0001). Using Poisson models with live birth and woman-year denominators, AI were found to be at increased risk for all GTN histologic subsets (complete, partial and invasive hydatidiform mole and choriocarcinoma). Of 524 total gestational choriocarcinoma cases identified within SEER, 8 (1.8%) affected American Indians; of them, 7 were from New Mexico. CONCLUSION: In New Mexico, AI continue to be at higher risk of GTN than are other groups. Given the rarity of choriocarcinoma within SEER, especially among AI, the New Mexico dataset provides the best available estimate of trends in U.S. AI GTN risk.

Ethnic differences at risk for gestational trophoblastic disease in New Mexico: A 25-year population-based study.

OBJECTIVE: The purpose of this study was to compare gestational trophoblastic disease incidence rates with the use of population-based data. STUDY DESIGN: All incident cases between 1973 and 1997 and live birth, pregnancy, and women at risk were tabulated with the use of data that were derived from the New Mexico Tumor Registry and Vital Records and Health Statistics Annual Reports. Statistical methods included trends analyses, odds ratios, and Poisson regression. RESULTS: Of 939 total cases, 312 non-Hispanic white women, 399 Hispanic white women, 201 American Indian women, and 27 other women were affected. Age-adjusted incidence rates were significantly higher for American Indian women (11.16%) compared with non-Hispanic (3.57%) or Hispanic white women (5.32%); the probability value was <.001. When live birth (1:438 women) and pregnancy (1:486 women) denominators were considered, American Indian women alone were at increased risk, and the ratio increased by 56% over 25 years. American Indian women were also at increased risk for partial mole (relative risk, 4.03; 95% CI, 2.57-6.31), invasive mole (relative risk, 26.7; 95% CI, 7.81-93.14), and choriocarcinoma (relative risk, 6.29; 95% CI, 1.81-22.66) variants. CONCLUSION: American Indians are at increased risk relative to the other predominant ethnic groups in New Mexico. Age-adjusted standardization provided a reproducible measurement that may be applicable across other registries.