Population Study of Uterine Leiomyosarcoma Incidence, Mortality and Survival
Alice Lytwyn, Aisling O'Meara, Laurie Elit, Noori Akhtar-Danesh. McMaster University, Hamilton, ON, Canada
Background: Uterine leiomyosarcoma (U-LMS) is the most common uterine sarcoma. Overall crude survival is 15-30%. The current 2003 WHO classification is derived from the 1994 Stanford study that correlated histologic features with patient outcomes. Most survival estimates for U-LMS are based on single institution studies, with few population-based studies. Stage predicts for survival; other factors have been inconsistent prognosticators. We estimated trends in incidence and mortality indices for patients diagnosed with U-LMS in Canada from 1992 to 2005.
Design: We searched the Canadian Cancer Registry with ICD-10 codes C54 (uterine corpus) and C55 (uterus), and then included only cases with ICD-O-3 codes 8890/3 (U-LMS), 8891/3 (epithelioid LMS) and 8896/3 (myxoid LMS). A flexible parametric model was used to estimate relative survival (RS) (observed survival rate among cancer patients divided by expected survival rate in the general population), by age group (<50, 50-59, 60-69, 70 years and older) in women diagnosed with LMS between 1992 and 2005. Crude probability of death was modeled to identify proportionate contributions of death due to LMS and death due to other causes.
Results: There were 682 women with LMS, with median age at diagnosis of 52.0 years, and 382 deaths. The number of cases of LMS decreased 1992 to 2005. Excess mortality rate showed a sharp peak in the first year after diagnosis for all age groups. RS decreased significantly with increasing age. RS progressively worsened with year of diagnosis, in that women with LMS diagnosed in more recent years had greater risk of mortality than women diagnosed in 1992. Crude probability of death showed that while most deaths were due to LMS in all age groups, a greater proportion was due to other causes as age increased.
Conclusions: This is the first study that reports incidence rates and compares mortality indices for U-LMS across Canada based on patient age and year of diagnosis. The peak in excess mortality in the first year after diagnosis is likely due to deaths in patients who present in advanced stage. Decreased RS with age warrants further investigation to determine if factors such as presentation in more advanced stage or less access to supportive care in the elderly may be factors. Refinement in histologic criteria are likely responsible for the apparent decrease in U-LMS through the years which may leave tumors with worse prognosis in the U-LMS category and explain the apparent worsening of RS in recent years.
Category: Gynecologic & Obstetrics
Tuesday, March 5, 2013 1:00 PM
Poster Session IV # 242, Tuesday Afternoon