781P - Chemotherapy sensitivity score based on ex vivo 3D tumour testing to predict clinical response for ovarian cancer patients

Background

Precision medicine for cancer patients has brought effective novel therapy options over the past decades. However, treatment for high-risk ovarian cancer is still based on platinum containing chemotherapy. Around 30% of primary disease patients do not respond, and 40% of complete responders relapse within a year. Strong predictive markers are lacking. This study reports a predictive chemotherapy sensitivity score to support patient-centred and better informed treatment decisions.

Methods

89 ovarian cancer patients eligible for chemotherapy were included in the trial between 2019 and 2022 in the Netherlands (IRB P18.032). Tumour clusters were isolated from ascites and solid tumour, seeded into a 384-well plate, and exposed to carboplatin, paclitaxel, three second-line chemotherapies, PARPi, and six immunotherapies. Next, ex vivo drug sensitivity was quantified using high-throughput 3D imaging and extraction of morphological features. The chemotherapy sensitivity score was generated by correlation of the assay results with the patients' response to carboplatin-paclitaxel. In parallel, patients’ relative sensitivity to other chemo-, targeted- and immunotherapies was assessed.

Results

The platform enabled accurate prediction of carboplatin-paclitaxel with a classification accuracy of 80%. Differential ex vivo patient response profiles were observed across second-line therapies, and 11% of the samples demonstrated a strong response to at least one of the immune compounds, indicating that our test method could provide alternative treatment options. For doxorubicin, the ex vivo results match clinical response (n=3), providing further evidence for the predictivity of the assay. Importantly, the assay has a success rate of 89%.

Conclusions

The prediction of clinical response to first-line therapy and identification of patient-specific sensitivity to ten second-line therapies in parallel, allows the support of informed treatment decisions for multiple treatment classes. Its two-week timeline makes the assay applicable in a clinical environment. A prospective trial for patients with suboptimal response and recurrent disease is ongoing.

Clinical trial identification

CME LUMC-P18.032.

Editorial acknowledgement

Legal entity responsible for the study

Leiden University Medical Center Radboud University Medical Center.

Funding

VitroScan B.V.

Disclosure

C.D. de Kroon: Other, Personal, Local PI: LUMC. A. van Altena: Other, Personal, Local PI: Radboud UMC. D. van der Meer, L. Ceton, J. van der Valk, T. Sijsenaar: Financial Interests, Personal, Full or part-time Employment: VitroScan. F. Grillet, M. Garcia Montero: Non-Financial Interests, Personal, Non-financial benefits: VitroScan. W. Vader: Financial Interests, Personal, Stocks or ownership: VitroScan; Financial Interests, Personal, Full or part-time Employment: VitroScan. P.B. Ottevanger: Non-Financial Interests, Personal, Local PI: Radboud UMC. J.R. Kroep: Non-Financial Interests, Personal, Local PI: LUMC.