Open AccessA population‐based comparison of second primary cancers in G ermany and S weden between 1997 and 2006: clinical implications and etiologic aspects
Author(s) -
Liu Hao,
Hemminki Kari,
Sundquist Jan,
Holleczek Bernd,
Katalinic Alexander,
Emrich Katharina,
Jansen Lina,
Brenner Hermann
Publication year - 2013
Publication title -
cancer medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.403
H-Index - 53
ISSN - 2045-7634
DOI - 10.1002/cam4.116
Subject(s) - spearman's rank correlation coefficient , rank correlation , cancer , ranking (information retrieval) , german , medicine , population , correlation , prostate cancer , rank (graph theory) , demography , oncology , statistics , mathematics , environmental health , combinatorics , computer science , geography , information retrieval , sociology , geometry , archaeology
Second primary cancer ( SPC ) has become an increasing concern in cancer survivorship. Patterns of SPC s in different populations may offer clinical implications and research priorities into SPC s. This study is devoted to compare the occurrences and rank correlations of SPC s between G ermany and S weden. Patients diagnosed with 10 common first primaries between 1997 and 2006 from the S wedish Family‐Cancer Database and 10 G erman cancer registries were included in this population‐based study. Spearman's rank correlation coefficients were used to evaluate the strength of the relationship of SPC s between the G erman and S wedish datasets. Spearman's rank correlation coefficients suggested a strong positive correlation between the G erman and S wedish datasets based on the ranks of thirty possible SPC s after all selected first cancers. This was also true when we compared the rankings and proportions of the five most common SPC s after site‐specific first primaries between the two populations. For kidney cancer, non‐Hodgkin's lymphoma, and leukemia the components of the five most common SPC s was exactly the same. Also, the ranking and the proportions for the three most common SPC s (i.e., colorectal, bladder, and lung cancers) after prostate cancer were identical in the two populations, as were those after most other primary cancers. The strikingly consistent patterns of SPC s in the two populations provide excellent opportunities for joint studies and they also suggest that many underlying reasons for SPC may have universal and tangible causes that await mechanistic dissection.