Hermann Brenner1, Lina Jansen2. 1. Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), INF 581, D-69120 Heidelberg, Germany; Division of Preventive Oncology, German Cancer Research Center (DKFZ) and National Center for Tumor Diseases (NCT), INF 460, D-69120 Heidelberg, Germany; German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ), INF 280, D-69120 Heidelberg, Germany. Electronic address: h.brenner@dkfz.de. 2. Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), INF 581, D-69120 Heidelberg, Germany.
Abstract
OBJECTIVE: Monitoring cancer survival is a key task of cancer registries, but timely disclosure of progress in long-term survival remains a challenge. We introduce and evaluate a novel method, denoted "boomerang method," for deriving more up-to-date estimates of long-term survival. STUDY DESIGN AND SETTING: We applied three established methods (cohort, complete, and period analysis) and the boomerang method to derive up-to-date 10-year relative survival of patients diagnosed with common solid cancers and hematological malignancies in the United States. Using the Surveillance, Epidemiology and End Results 9 database, we compared the most up-to-date age-specific estimates that might have been obtained with the database including patients diagnosed up to 2001 with 10-year survival later observed for patients diagnosed in 1997-2001. RESULTS: For cancers with little or no increase in survival over time, the various estimates of 10-year relative survival potentially available by the end of 2001 were generally rather similar. For malignancies with strongly increasing survival over time, including breast and prostate cancer and all hematological malignancies, the boomerang method provided estimates that were closest to later observed 10-year relative survival in 23 of the 34 groups assessed. CONCLUSION: The boomerang method can substantially improve up-to-dateness of long-term cancer survival estimates in times of ongoing improvement in prognosis.
OBJECTIVE: Monitoring cancer survival is a key task of cancer registries, but timely disclosure of progress in long-term survival remains a challenge. We introduce and evaluate a novel method, denoted "boomerang method," for deriving more up-to-date estimates of long-term survival. STUDY DESIGN AND SETTING: We applied three established methods (cohort, complete, and period analysis) and the boomerang method to derive up-to-date 10-year relative survival of patients diagnosed with common solid cancers and hematological malignancies in the United States. Using the Surveillance, Epidemiology and End Results 9 database, we compared the most up-to-date age-specific estimates that might have been obtained with the database including patients diagnosed up to 2001 with 10-year survival later observed for patients diagnosed in 1997-2001. RESULTS: For cancers with little or no increase in survival over time, the various estimates of 10-year relative survival potentially available by the end of 2001 were generally rather similar. For malignancies with strongly increasing survival over time, including breast and prostate cancer and all hematological malignancies, the boomerang method provided estimates that were closest to later observed 10-year relative survival in 23 of the 34 groups assessed. CONCLUSION: The boomerang method can substantially improve up-to-dateness of long-term cancer survival estimates in times of ongoing improvement in prognosis.