New test for ovarian cancer mutation heads into clinical trials

Intermediate magnification micrograph of a low malignant potential (LMP) mucinous ovarian tumour. Image credit: Michael Bonert CC BY-SA 3.0

A recent study led by Professor Chris Lord and Dr Rachael Natrajan at The Institute of Cancer Research, London, has uncovered an accurate predictor of whether patients have an error in a specific gene, ARID1A, which is found in a subset of gynaecological cancers – a method which could be used in a potential upcoming international clinical trial.

Over 230,000 women around the world are diagnosed annually with ovarian cancer, 7,300 of whom are from the UK, according to Cancer Research UK – making it the 6^th most common cancer in women. Mutations in ARID1A have been in 57% of ovarian clear cell carcinomas (OCCC), 40% of uterine endometrioid carcinomas and between 20 and 36% of uterine carcinosarcomas.

These mutations in this gene cause the loss of protein function, and because the ARID1A protein is involved in the process of DNA replication and repair, its lack of function leads to uncontrolled cell division, and subsequently tumour formation. Due to the strong link between mutations in ARID1A and gynaecological cancers, the ability to identify whether a patient has this error is important for deciding which treatment is most suitable for them.

Choosing the optimal antibody

This research – which was part-funded by the Gynaecological Cancer Fund through The Royal Marsden Cancer Charity and carried out in collaboration with Dr Susana Banerjee, Consultant Medical Oncologist at The Royal Marsden NHS Foundation Trust – examined the use of three commercially-available antibodies to show if the protein is present or not using a technique called immunohistochemistry (IHC). By staining tumour cells with these three anti-ARID1A antibodies, EPR13501, D2A8U and HPA005456, they were able to compare the differences between ARID1A-mutated tumour cells or tumour cells and the healthy surrounding cells with normal ARID1A expression.

Dr Saira Khalique, clinician and first author of the study, explained:

“The percentage of cells stained positively for the ARID1A mutation, as well as the intensity of the staining, was then used to score the sample using an ‘immunoreactive scoring system’. The scoring was carried out separately by three pathologists who did not know whether the samples contained the mutated version or the non-mutated version of the gene. Combining these two approaches, sequencing and IHC, makes the results of this study very reliable.”

The results then allowed the scientists to determine the immunoreactive scores which indicated that the cells had the mutated gene. The scientists found that EPR13501 was the optimal antibody for ARID1A mutation detection, showing the least variation between the three scorers and was the most cost-effective. 

Personalising ovarian cancer treatment

The scientists are hoping to take this optimised test forward to be used in a clinical trial for patients with ARID1A-mutated gynaecological cancers, to be led by Dr Susana Banerjee, which could provide potentially life-changing options for these women. Once the ARID1A mutation was detected, it could then be matched to the appropriate treatment – adding a level of personalisation to individual patient’s treatment.

Unfortunately for women with these cancers, the prognosis is poor and deterioration in their condition can occur very suddenly. As a clinician, Dr Khalique emphasised the need for a reliable test with a quick turn-around to enable her to make timely decisions about her patient’s treatment.

Dr Natrajan added:

“Although it is still early stages, we hope to see this test used widely in the next decade. It is really important to be able to make decisions about patient treatment quickly to best benefit them. It is a cost-effective, straight-forward test that would enable patients to get accurate results much more quickly than gene sequencing directing them to treatment.”

As the clinical trial approaches, the two researchers look to the future. As ARID1A is also common in gastric and pancreatic cancers, and is seen in metastatic breast cancer, this test could also be adapted to improve treatment pathways for all these patients.