Abstract 180P
Background
The one-step nucleic acid amplification (OSNATM) assay (Sysmex, Kobe, Japan) is an intraoperative diagnostic procedure performed for the detection of lymph node metastases using molecular techniques to amplify and detect cytokeratin 19 (CK19) mRNA. This multicenter study aimed to construct stratification of prognosis by CK19 mRNA copy number and development of a prognosis prediction model in breast cancer patients.
Methods
A total of 4,757 breast cancer patients who underwent sentinel lymph node (SLN) biopsy using the OSNA assay between 2008 and 2012 were enrolled from 12 institutions in Japan. The patients were randomly divided at a ratio of 2:1 into the training (n = 3,171) and validation (n = 1,586) groups. First, the cutoff value of total tumour load (TTL: sum of CK19 mRNA copy number of each positive SLN) for distant disease-free survival (DDFS) of the training group was determined using the Youden index. Subsequently, we constructed a prognosis prediction model that predicts distant recurrence in the training group using a multi-logistic regression model. Finally, the prognosis prediction model was validated using the validation group.
Results
The median follow-up period of both the training and validation groups was 5.5 years. The cutoff value of TTL to classify the DDFS was 1,100 copies per microliter. Multivariate analysis of the training group revealed that TTL, pathological tumour size, tumour grade, progesterone receptor status, postoperative chemotherapy, and postoperative anti-HER2 therapy were significantly associated with DDFS. A prognosis prediction model was constructed using these parameters, and the area under the receiver operating characteristic curve of the training group was 0.82. The sensitivity, specificity, and accuracy of the prognosis prediction model were 70.1%, 79.9%, and 79.5% using the training group and 61.9%, 80.0%, and 79.5% using the validation group.
Conclusions
We constructed a prognosis prediction model using TTL, which is an independent prognostic factor, for the prediction of distant recurrence. This model can accurately predict the prognosis of breast cancer patients using the OSNA assay.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Sysmex Kobe Japan.
Disclosure
T. iIshikawa: Honoraria (self): Pfizer; Honoraria (self): Eisai; Honoraria (self): Taiho; Honoraria (self): Novartis; Honoraria (self): Chugai; Honoraria (self): AstraZeneca; Honoraria (self): Lily; Honoraria (self): Kyowa Kirin. S. Noguchi: Honoraria (self), Honoraria (institution), Research grant/Funding (institution): Sysmex. All other authors have declared no conflicts of interest.