Short summary

Treatment of platinum resistant ovarian cancer is a major challenge. Despite numerous chemotherapeutic agents to offer to these patients only a minor subgroup will benefit from the treatment and there is an obvious need for new markers to select such patients. 

This randomized study examines if the application of circulating HOXA9 methylated DNA for treatment monitoring of recurrent ovarian cancer can reduce the number of treatment cycles without affecting overall survival. 

Rationale

Recurrent epithelial ovarian cancer represents a major problem in gynecologic oncology. Despite extensive surgery and combination chemotherapy, treatment is often only life-prolonging and not curative. Initially, the disease is often chemotherapy sensitive but over time it becomes resistant to treatment. The subgroup of patients resistant or refractory to platinum-based chemotherapy in particular have a dismal prognosis and it may be discussed whether they should be treated with chemotherapy if not in a protocol. 

The evaluation of treatment effect is another major challenge. Imaging by CT scan is useful in the evaluation of measurable lesions although with considerable uncertainty. Many patients do not have clearly measurable disease since it is often not bulky but rather spread throughout the peritoneal cavity in the form of carcinosis. 

CA125 is a blood-based biomarker approved for general use in ovarian cancer diagnostics and treatment monitoring, but it has several drawbacks. Most importantly, not all tumors shed CA125 into the circulation. The marker is unspecific and its level may increase for a number of reasons, including any affection of the coelomic epithelium. A changed level is therefore an inaccurate evaluation of the tumor size and disease status. 

There is an obvious need for new markers, which alone or in combination can be used for treatment monitoring with the perspective of sparing patients ineffective chemotherapy. 

Most malignant tumors shed tumor specific DNA into the circulation (ctDNA), which has shown to hold prognostic information in ovarian cancer. The technological development over the last decade has provided substantial improvement of the analytical performance and ctDNA can now be measured with high sensitivity and at low analytical variation. The current literature focuses on measurement of somatic mutations in circulating tumor DNA, but more recently, tumor specific methylation has drawn attention. 

The HOXA9 gene is the homeobox gene of importance to morphogenesis and differentiation. Methylated HOXA9 ctDNA (HOXA9 meth-ctDNA) seems to hold prognostic information in epithelial ovarian cancer. As the gene is tumor specific and not found in healthy individuals, it has potential to be more exact than CA125. 

A recent study from our group suggests that elevated HOXA9 meth-ctDNA after the first treatment cycle implies a grave prognosis. The results were confirmed in a new study, which indicated that the level after three cycles held considerable prognostic information. Thus, consecutive analysis may be applied for treatment monitoring and modification of treatment.

Objective:

To compare CA125 and HOXA9 meth-ctDNA as response evaluation tools in recurrent ovarian cancer patients estimated by the median number of treatment cycles.

Endpoints: 

• Number of treatment cycles (primary)

• Overall survival (secondary)

• Response rate (secondary)

Design:

The study is a two-arm, non-blind, block randomized trial based on treatment monitoring by CA125 and HOXA9 meth-ctDNA in the blood. After randomization, patients will be evaluated according to CA125 (GCIG criteria) + CT scan (RECIST 1.1) or according to HOXA9 meth-ctDNA + CT scan (RECIST 1.1). 

Description of the cohort

The study is a two arm, non-blind, block randomized trial based on treatment monitoring by CA125 and HOXA9 meth-ctDNA in the blood from recurrent ovarian cancer patients during palliative treatment at the Department of Oncology, Vejle Hospital. 

The patients are offered three different chemotherapeutic treatment options (Liposomal Doxorubicin, Topotecan or Paclitaxel) prior to randomization. 

Data and biological material

Blood. 

A blood sample will be drawn at baseline/randomization, before start of chemotherapy. Every three weeks after a cycle of chemotherapy a blood sample will be drawn for evaluation of response (CA125 or methylated HOXA9 ctDNA) until progression or stop of treatment for other reasons. 

Collaborating researchers and departments

Department of Oncology, Vejle Hospital

• Professor Karina Dahl Steffensen, PhD, MD

Department of Biochemistry and Immunology, Vejle Hospital

• Molecular biologist Rikke Fredslund Andersen, PhD