SerUm and Plasma MicroRNAs in Malignant Ovarian gERm Cell Tumours

Case population (MOGCTs):

1. Serum aFP (alpha fetoprotein), hCG (human chorionic gonadotrophin), LDH (lactate dehydrogenase) and Ca125 (cancer antigen 125) tumour markers measured at time of diagnosis
2. Additional serum and plasma sample to be obtained at the same time as initial bloods
3. Subsequent serum and plasma samples to be obtained as per study arms 1-3 based on cancer staging
4. Tissue samples will be obtained at the time of surgery

Control population:

Participants acting as controls will be identified from those attending for TV-USS where either a benign teratoma or no pathology is identified (e.g. from a gynaecology clinic). A total of 26ml of blood will be taken. Serum aFP, hCG, LDH and Ca125 will be measured as per routine clinical practice in patients with a benign OGCT.

Patients in the control population (benign germ cell tumours or patients with no known gynaecological pathology) will have a single blood test for miRNA analysis.

If participants require an operation for the benign GCT, tissue samples will be obtained at the time of surgery.

miRNA analysis will be performed using NanoString for the discovery panel. Further targeted sequencing will be performed in the larger prospective study using specific miRNA panels. This may be done using NanoString or a custom MiRNA assay primer.

Endpoint The pilot will be completed after the final patient has completed 12 months of follow up. The pilot data analysis will take place at that point.

We anticipate that the study will end 12 months after the final participant has completed 24 months of follow up, which gives 12 months for analysis of the samples. The number of participants to take part will be based on a power calculation performed at the time of analysis of the pilot data.

DATA COLLECTION AND MANAGEMENT Patients will enter study arm 1-3 based on cancer staging, or will belong to the control group.

Each participant will be followed up for a maximum of 2 years after they receive their definitive treatment (surgery and/or (neo)adjuvant chemotherapy). We anticipate this will be a maximum of 30 months total time in the study.

An interim analysis of data will be performed at 12 months, with a view to:

1. Identify key miRNAs for prospective validation
2. Allow for appropriate power calculation of the prospective validation study

• Blood test (Diagnostic test): Serum and plasma microRNA quantification
• Pathology specimen miRNA expression (Diagnostic test): Pathology specimen mirNA expression - ALL MOGCT; OPTIONAL for BOGCT

Primary outcomes

• Difference in microRNA expression (plasma) between benign and malignant masses [24 months]: Comparison of heatmaps based on mean expression of clusters across samples
• Difference in microRNA expression (serum) between benign and malignant masses [24 months]: Comparison of heatmaps based on mean expression of clusters across samples
• microRNA expression (plasma) [24 months]: Differential expression of microRNAs (assessed using a moderated t statistic and P values adjusted for multiple testing)
• microRNA expression (serum) [24 months]: Differential expression of microRNAs (assessed using a moderated t statistic and P values adjusted for multiple testing)

Secondary outcomes

• Quantitative measure of circulating miRNA before treatment [24 months]: miRNA expression in plasma samples of women with a diagnosis of a MOGCT compared to those of women with BOGCTs using nCounter miRNA Expression panels
• Quantitative measure of circulating miRNA before treatment [24 months]: miRNA expression in serum samples of women with a diagnosis of a MOGCT compared to those of women with BOGCTs using nCounter miRNA Expression panels
• Quantitative measure of circulating miRNA after treatment [24 months]: miRNA expression in plasma samples of women with a diagnosis of a MOGCT compared to those of women with BOGCTs using nCounter miRNA Expression panels
• Quantitative measure of circulating miRNA after treatment [24 months]: miRNA expression in serum samples of women with a diagnosis of a MOGCT compared to those of women with BOGCTs using nCounter miRNA Expression panels
• Performance of segmentation model on ultrasound images [24 months]: Correlation between identification of region of interest (ROI) to examiner segmentation. Dice surface coefficient (DICE).
• Performance of segmentation model on MRI images [24 months]: Correlation between identification of region of interest (ROI) to examiner segmentation. Dice surface coefficient (DICE)
• Performance of classification model on ultrasound images [24 months]: Correlation between benign or malignant diagnosis by model vs ultrasound subjective assessment or histopathological diagnosis (area under the ROC curve (AUC), F1-score)
• Performance of classification model on MRI images [24 months]: Correlation between benign or malignant diagnosis by model vs ultrasound subjective assessment or histopathological diagnosis (area under the ROC curve (AUC), F1-score)

Sex: Female, ages: 16 Years to any

Inclusion Criteria:

* All patients with a new diagnosis of a malignant ovarian germ cell tumour.
* The control population will include all patients with a new diagnosis of a benign ovarian germ cell tumour or no known gynaecological pathology.

Exclusion Criteria:

* Previous or ongoing chemotherapy for MOGCT
* Previous surgery for MOGCT
* Pregnancy - this will be verbally communicated for those not having surgery or chemotherapy, for those having surgery or chemotherapy a urine pregnancy test should be negative and documented in the clinical notes.
* Fetal circulating DNA is known to be present in maternal blood and therefore pregnant women should not be included in this study
* Denial of informed consent
* Age <16 years
* History of any other cancer

• Srdjan Saso, PhD MRCOG MRCS, Principal investigator, Imperial College Healthcare NHS Trust