1137PD - Carcinoma of unknown primary: features and outcomes of patients managed in a large UK centre

Date 29 September 2014
Event ESMO 2014
Session Neuroendocrine & endocrine tumours and CUP
Topics Carcinoma of Unknown Primary Site (CUP)
Presenter Rebecca Lee
Citation Annals of Oncology (2014) 25 (suppl_4): iv394-iv405. 10.1093/annonc/mdu345
Authors R. Lee1, S. Baxter2, S.J. Ayers1, C.L. Mitchell1, J. Hasan1
  • 1Medical Oncology, The Christie NHS Foundation Trust, M20 4BX - Manchester/GB
  • 2Medical Oncology, The Christie NHS Foundation Trust, Manchester/GB




Carcinoma of unknown primary (CUP) is a metastatic epithelial or neuroendocrine malignancy identified on final histology with no primary site detected despite investigations and specialist review (National Institute of Clinical Excellence (NICE), UK). 3-5% of malignancies are CUP with associated poor prognosis. In this retrospective case series we reviewed the pathology and prognostic factors of patients (pts) with CUP and analysed responses to chemotherapy (CT).


A retrospective review was performed on pts presenting between 2005 and 2011 with confirmed CUP (as per NICE definition). Data from medical records, histological, laboratory and radiological investigations were analysed. Univariate and multivariate analysis was carried out to evaluate the effects of patient and disease characteristics overall survival (OS).


249 pts with suspected CUP were referred with 134 pts having histological confirmation. Median age at diagnosis was 64.5 years (range 19 to 85 years); 54.5% male and 45.5% female. Median OS for pts with confirmed CUP was 23.9 weeks (range 0.14-441 wks). On univariate analysis overall predictors of decreased OS were raised lactate dehydrogenase (P<0.001), performance status (PS) > = 2 (P<0.001), adenocarcinoma histology (P<0.001), >1 organ affected (P < 0.001), presence of liver metastases (P = 0.002). Predictors of better OS were squamous histology (P < 0.001), predominantly lymph node involvement (P < 0.001), treatment with chemotherapy (P = 0.015) or surgery (P<0.001). On multivariate analysis PS > = 2 (P < 0.001) was significant for decreased OS with a trend for pts with adenocarcinoma (P = 0.07). Pts managed with surgery had superior outcome compared to best supportive care (BSC) (median OS 68 vs 4 wks P<0.001); chemotherapy (median OS 30 wks P < 0.001) or radiotherapy (median OS 18 wks) also did significantly better. 60 pts received systemic chemotherapy with a variety of regimens. Complete response was achieved in 4/60 pts, partial response in 9/60, stable disease in 13/60 and progressive disease in 18/60. 6 pts received second line treatment. Response to chemotherapy significantly improved median OS. Multivariate analysis showed pts with PS> = 2 (P < 0.001), >1 organ affected (P = 0.017) and liver metastases (P = 0.005) had decreased OS. No significant difference in OS seen in pts receiving platinum or non platinum chemotherapy.


CUP remains a difficult to treat entity and there are sub groups such as adenocarcinoma and poor PS with very poor prognosis. It is important to have careful discussions with pts regarding benefits of chemotherapy when their disease has adverse features.


All authors have declared no conflicts of interest.