P. Karihtala (Helsinki, Finland)
Helsinki University Hospital Comprehensive Cancer Centre and University of HelsinkiAuthor Of 1 Presentation
140P - How breast cancer recurrences are found? - a real-world, prospective cohort study
Abstract
Background
There is a very limited amount of data available how most breast cancer relapses, either distant metastases (DM) or locoregional recurrences (LRR), are actually found in routine clinical practice.
Methods
From a prospective cohort of 621 women diagnosed and treated for early invasive breast cancer in Oulu University Hospital between the years 2003-2013, we analysed the patients who were later diagnosed with DM (n=61) and the patients who had LRR (n=34). The patients had routine annual control visits up to 10 years from diagnosis with annual clinical visits, mammographies, blood counts and liver function tests. We evaluated 1) how the relapse was found; 2) what were the first signs or symptoms of the relapse and 3) if these factors affected prognosis.
Results
The median follow-up time was 98 months. Most of DMs (n=38, 62.3%) were found when a patient contacted the health care because of a new symptom. Twelve (19.7%) metastatic diseases were coincidental findings during other medical examinations. Only ten (16.4%) DMs were detected at the pre-planned control visits. The most common first signs or symptoms of metastases were pain (n=23, 37.7%), abnormalities in imaging, laboratory tests or clinical examination (n=18, 29.5%) and palpable or visible lesion detected by the patient (n=7, 11.5%). Abnormalities in routine laboratory test did not lead to the detection of any relapse. Patients with pain as the first indicator of metastasis had worse survival in metastatic disease (hazard ratio 4.40; 95% confidence interval 1.77-10.94; p=0.001). LRRs were mostly found at the pre-planned control visits (n=14, 41.2%). Rest of the LRRs were detected after patient contacted health care because of a relapse-related symptom (n=13, 38.2%) or as coincidental findings or in screening mammographies (n=7, 20.6%).
Conclusions
In this prospective, contemporary, real-world study, the vast majority of both DMs and LRRs were detected outside the pre-planned control visits. Pain was the most common first indicator of DM and its presence also predicted dismal survival. These results highlight the importance of finding the ways to lower the threshold to contact the surveillance unit, instead of annual routine controls.
Legal entity responsible for the study
The authors.
Funding
Suomen Lääketieteen Säätiö, Pohjois-Suomen Terveydenhuollon Tukisäätiö.
Disclosure
All authors have declared no conflicts of interest.