Skip to main content


Biomarker-Guided First-Line Maintenance Therapy in Advanced Ovarian Cancer: Real-World Outcomes and Cost

April 27, 2021


Daniel Simmons, PharmD, MS, AstraZeneca, discusses results from a study comparing progression-free survival (PFS) and costs per month of PFS between a biomarker-guided approach and non-biomarker guided approach to first-line maintenance therapy for advanced ovarian cancer.

These results were presented at the 2021 Society of Gynecologic Oncology (SGO) Annual Meeting on Women’s Cancer.


Hi my name is Dan Simmons and I am Associate Director of Health Economics and Outcomes Research at AstraZeneca. I am excited to share a poster that my colleagues from AstraZeneca and the Johns Hopkins University School of Medicine presented at the 2021 Society of Gynecologic Oncology meeting. The title of our poster is “Clinical outcomes and cost associated with HRD biomarker guided first line maintenance therapy in advanced ovarian cancer”.

Despite first line surgery and chemotherapy, most women with advanced ovarian cancer tend to recur. Several first line maintenance therapies have been developed to help delay or potentially prevent recurrence or progression. Olaparib is a Poly ADP Ribose Polymerase or PARP inhibitor indicated for first line maintenance therapy in patients with a mutation to the breast cancer or BRCA gene based on the SOLO-1 trial. It is also indicated in combination with bevacizumab for patients with a Homologous Recombination Deficiency or HRD based on the PAOLA-1 trial.

HRD is an emerging biomarker in about 50% of patients that includes BRCA mutations, other mutations in the homologous recombination repair pathway, and other causes of altered gene expression that may be prognostic for a response to PARP inhibitors. Niraprib is another PARP inhibitor indicated for maintenance regardless of biomarker status. While it is approved without a biomarker test, the group with a positive HRD test in the PRIMA trial had a larger magnitude of benefit than those who were HRD negative. Niraparib with bevacizumab has not been tested in a multi-arm comparative trial and limited data is available. Lastly, bevacizumab monotherapy is also approved without a biomarker test.  

Considering the variety of available maintenance options, the objective of our research was to model the costs and clinical outcomes associated with using an HRD biomarker guided versus a nonbiomarker guided approach to maintenance. An HRD guided approach prescribes PARP plus bevacizumab to all HRD positive maintenance eligible patients and uses bevacizumab or routine surveillance for all HRD negative patients. A nonbiomarker guided approach prescribes a PARP inhibitor for all maintenance eligible patients. Because germline BRCA testing is a standard of care we treated PARP cost as a weighted average of niraparib and olaparib to focus on the impact of HRD testing.

We utilized a parametric Markov model for 100 newly diagnosed stage 3 or 4 advanced ovarian cancer patients prescribed first line maintenance according to an HRD biomarker guided strategy or a non- biomarker guided strategy. The model takes a US commercial payer perspective and a five-year horizon during which patients can pass through up to 4 additional lines of treatment before progressing to discontinuation or death. The model includes testing, drug acquisition, administration, and adverse event management costs.

We estimated PFS duration by applying exponential curves to median PFS from adjusted trial data. There were substantial differences in the risk of recurrence or progression between populations in the PRIMA, SOLO-1 and PAOLA-1 trials. To account for this heterogeneity we utilized two methods of adjusting the trial data. Method one utilized population adjusted data which adjusted the median PFS estimates to a common patient population. Method two adjusted the trial data using clinician input.

Costs and other inputs such as adverse event rates were based on published literature where available. For drug dosing we utilized the FDA approved doses, trial data, and real-world data from the Optum research database. Drug costs were based on wholesale acquisition costs and testing costs were based on the list price for the Myriad MyChoice test which is the approved companion diagnostic for HRD testing for olaparib in first line maintenance.

In both arms, 65% of patients received and responded to platinum chemotherapy and were eligible to receive maintenance therapy. In the biomarker guided arm, 31 received PARP plus bevacizumab, 10 received bevacizumab, and 24 received routine surveillance. In the non-biomarker guided arm these 65 received PARP. The 35 patients in each arm not eligible for maintenance were assumed to be the same in both groups and were not included in the model.

Our model estimated, using both methods of adjustment, that the biomarker guided approach produces better clinical outcomes and substantially lower costs. The biomarker guided approach leads to between 4 and 213 additional months of first line progression free survival for the cohort. This range was determined by using the two methods of adjusting trial results for heterogeneity. 

The five-year cost for the biomarker guided approach was between $25.8 and $26.5 million compared to $28.3 to $31.1 million for the nonbiomarker guided approach. This is equivalent to a cost savings of $1,002 to $6,598 per month of PFS across the 100-patient population. Across all tested scenarios and methods of estimating outcomes, the biomarker guided approach was cost saving and resulted in better clinical outcomes for the population.

There are some key limitations and areas for future research that I would like to discuss. PFS was modeled by fitting an exponential curve to median PFS data. Future research should explore other extrapolation techniques. Real-world PFS data, once available, could also better inform estimates for modeling. This model does not measure quality adjusted life years or overall survival and focuses on progression free survival. Future research could evaluate these outcomes. The model also assumes that patients in the biomarker guided arm all receive combination therapy if they are HRD positive. Cost-savings may be larger if some patients received monotherapy PARP, especially those with a BRCA mutation. Reflex testing may also impact costs.

Overall, our research demonstrates the importance of personalized medicine in advanced ovarian cancer maintenance. The more you are able to personalize treatment for a patient, the better the outcomes and the more cost-effective the treatment. Our model highlights that HRD testing, despite its initial costs can be viewed as an investment. Over five years, using a biomarker guided approach could lead to better population level outcomes for patients and cost-savings to a payer.

Bradley W, et al. Clinical Outcomes and Cost Associated with HRD Biomarker Guided First Line Maintence Therapy in Advanced Ovarian Cancer. Presented at: the virtual 2021 SGO Annual Meeting on Women’s Cancer; March 19-25, 2021. Poster 11035.

Dr Simmons reports employment by AstraZeneca.

Back to Top