Aim: Screening for ovarian cancer in women aged 50-84, using serial serum CA125 assay in a risk of ovarian cancer algorithm (ROCA), may reduce mortality from this disease. Other conditions such as non-Hodgkin lymphoma (NHL) are known to express CA125 > 35U/ml in approximately 40% of cases at the time of clinical diagnosis and could also be intercepted during a pre-clinical phase.
Methods: A retrospective pilot study was performed in one centre of the United Kingdom Collaborative Trial for Ovarian Cancer Screening (UKCTOCS). 4775 women aged 50-74 were recruited to the multimodal arm of the study in Portsmouth and underwent annual CA125 testing, supported by repeat serum assay and imaging examination when indicated. During ten years, 32,642 annual screening episodes were conducted. 273 individual women with persistent abnormalities were identified by protocol defined criteria and seen by one gynaecological oncologist.
Results: Four of these 273 women were consequently diagnosed with high risk NHL; 2 were entirely asymptomatic. In all 4 cases a trebling of the CA 125 from the individual’s original baseline level was observed prior to definitive biopsy. This factor of change discriminated the NHL women with 100% sensitivity and specificity from 25 aged matched controls referred to the screening clinic who turned out to have benign causes of their ROCA anomaly. The histology dictated that all 4 women were treated with chemotherapy. Eight years after her NHL diagnosis, 1 of the 4 was diagnosed with biopsy-proven primary squamous cell lung carcinoma, and died 8 months later. The other 3 women are relapse free more than 5, 10 and 11 years from NHL diagnosis.
Conclusions: It is appropriate to undertake a CT scan of the chest and abdomen as well as evaluating the pelvis following ROCA index abnormalities, as non-gynaecological causes of rising CA125 levels may be identified fortuitously.