NPS-2143 br Funding This work was supported by the
Funding This work was supported by the National Natural Science Foundation of China (Grant numbers:81502475) and the Science and Technology Project of Henan Province (Grant numbers: 172102310067).
Conflict of interest statement
Introduction The ageing population and improvements in diagnostics and treatment have increased the number of breast cancer (BC) survivors over the last decades [, , ]. The incidence of BC was 4682 in Denmark in 2015, which corresponds to an incidence rate of 123.2 per 100,000 (age-standardised European rate) . International studies report recurrence rates of 12–49% in patients diagnosed with BC, depending on tumour stage, adjuvant therapy and follow-up time [, , , , ]. Insight into when, where and how cancer recurrence presents is essential to provide optimal care for cancer survivors [3,10,11]. Recurrence is not routinely registered in most patient registers, and identification of patients with cancer recurrence remains a challenge. BC recurrence is registered in the clinical cancer database by the Danish Breast Cancer Group (DBCG). Yet, these registrations are incomplete, and follow-up ends after 10 years . Three studies have reported on the identification of BC recurrence through administrative data. A recent Danish study yielded a sensitivity (SEN) of 88% and a positive predictive value (PPV) of 73% . A UK study identified 92% of all recurrences , and a US study yielded a SEN of 89% and a specificity (SPE) of 99% . Other studies have developed algorithms to identify recurrence of two or more cancers, including BC [, , , ]. These studies have been criticised for moderate performance, infrequent recurrence, small sample size and sampling of non-representative populations from academic centres and single institutions . Two recent Danish studies have reported promising results on register-based algorithms to identify patients with recurrence of colorectal cancer  and NPS-2143 cancer . This indicates a possibility to improve the performance of an algorithm to identify women with BC recurrence.
Material and methods We conducted a cohort study based on data from Danish national health registers. The unique personal registration number assigned to all Danish residents was used to link data at the individual level .
Results A total of 487 patients fulfilled the inclusion criteria (Fig. 2), and 471 patients constituted the final study population; 149 (32%) of these had cancer recurrence according to the gold standard. The characteristics of the population stratified by cancer recurrence status according to the gold standard are presented in Table 1. The median follow-up time for the gold standard population was 7.5 years (inter quartile range (IQR): 5–9) since the primary BC surgery.
Funding This work was supported by the Danish Cancer Society and Aarhus University. The funders had no role in the study.
Declarations of interest
Introduction Data from the Unites States and Sweden have earlier suggested that prostate cancer is a disease that men live with rather than die of, such that the majority of men with prostate cancer will die from causes other than their prostate cancer [1,2]. Prostate cancer mortality rates have been reported to be higher in Denmark than in other Scandinavian countries and the United States with an age-standardized (World Standard Population) mortality rate of 16.5 per 100,000 persons per year for the period 2011–2015 [, , ]. In the same period from 2011 to 2015, approximately 4500 new cases of prostate cancer and approximately 1200 deaths were attributed to prostate cancer each year according to Nordic cancer statistics . There are several plausible explanations for the reported excess of prostate cancer deaths in Denmark, including differences in national PSA screening strategies contributing to lead-time bias and underlying comorbidity in the Danish male population . Misattribution in death certificates may also contribute to an over-reporting of prostate cancer deaths. Studies from other countries with similar demographics have quantified the magnitude of misattribution in both population-based and clinical trials settings with differing results [, , , ].