Cost-effectiveness of management strategies for patients with recurrent ovarian cancer and inoperable malignant bowel obstruction (078)

Abstract

Objectives: Patients with recurrent platinum-resistant ovarian cancer often present with inoperable malignant bowel obstruction from large burden abdominal disease. End-of-life interventions such as total parenteral nutrition (TPN) and chemotherapy may modestly improve survival but may also result in a worsened quality of life (QOL) and high cost. We aimed to describe the relative cost-effectiveness of these interventions to inform clinical decision-making. Methods: Four strategies, including home hospice, TPN alone, chemotherapy alone, and TPN+chemotherapy, for management of platinum-resistant recurrent ovarian cancer with inoperable malignant bowel obstruction and gastrostomy tube, were compared using a simple decision model. Chemotherapy regimens modeled included standard monotherapies (liposomal doxorubicin, pacli-taxel, gemcitabine, or topotecan). An alternative scenario assumed a higher cost novel therapy (pembrolizumab). Costs and rates of blood transfusion ($451,15%) and granulocyte colony-stimulating factor ($3079, 9.6%) were incorporated into standard chemotherapy costs. Median survival and rates of hospitalization were obtained from the detailed literature review: hospice (survival 38 days, 6% hospitalization), chemotherapy (42 days, 24%), TPN (55 days, 24%), and TPN+chemotherapy (74 days, 43%). End-of-life QOL-related utilities (0.08 for end-stage with bowel obstruction; 0.07 for additional hospitalization) were incorporated. Outcomes included the average cost of strategy and incremental cost-effectiveness ratios (ICERs) in US dollars per quality-adjusted life-year (QALY) gained. Multiple one-way sensitivity analyses performed on utilities, costs, probability of hospitalization, and expected survival time (Figure 1). Results: TPN+chemotherapy was the most costly strategy ($33,099) in the base case scenario and was associated with the highest QALYs (0.016). The lowest cost strategy was standard chemotherapy alone (cost $7,078, QALYs 0.009), which dominated home hospice ($8,666, QALYs 0.008) due to lower cost and slightly higher QALYs. The TPN alone strategy (cost $21,025, QALYs 0.012) was dominated by a combination of other strategies. The ICER of TPN+chemotherapy was $3.8 million/QALY compared to chemotherapy alone. In the novel therapy scenario, hospice was the least costly strategy, novel therapy alone was dominated, and TPN+novel therapy had an ICER of $11 million/QALY compared to the next less costly strategy. The model’s results were stable to multiple one-way sensitivity analyses. Objectives: Patients with recurrent platinum-resistant ovarian cancer often present with inoperable malignant bowel obstruction from large burden abdominal disease. End-of-life interventions such as total parenteral nutrition (TPN) and chemotherapy may modestly improve survival but may also result in a worsened quality of life (QOL) and high cost. We aimed to describe the relative cost-effectiveness of these interventions to inform clinical decision-making. Methods: Four strategies, including home hospice, TPN alone, chemotherapy alone, and TPN+chemotherapy, for management of platinum-resistant recurrent ovarian cancer with inoperable malignant bowel obstruction and gastrostomy tube, were compared using a simple decision model. Chemotherapy regimens modeled included standard monotherapies (liposomal doxorubicin, pacli-taxel, gemcitabine, or topotecan). An alternative scenario assumed a higher cost novel therapy (pembrolizumab). Costs and rates of blood transfusion ($451,15%) and granulocyte colony-stimulating factor ($3079, 9.6%) were incorporated into standard chemotherapy costs. Median survival and rates of hospitalization were obtained from the detailed literature review: hospice (survival 38 days, 6% hospitalization), chemotherapy (42 days, 24%), TPN (55 days, 24%), and TPN+chemotherapy (74 days, 43%). End-of-life QOL-related utilities (0.08 for end-stage with bowel obstruction; 0.07 for additional hospitalization) were incorporated. Outcomes included the average cost of strategy and incremental cost-effectiveness ratios (ICERs) in US dollars per quality-adjusted life-year (QALY) gained. Multiple one-way sensitivity analyses performed on utilities, costs, probability of hospitalization, and expected survival time (Figure 1). Results: TPN+chemotherapy was the most costly strategy ($33,099) in the base case scenario and was associated with the highest QALYs (0.016). The lowest cost strategy was standard chemotherapy alone (cost $7,078, QALYs 0.009), which dominated home hospice ($8,666, QALYs 0.008) due to lower cost and slightly higher QALYs. The TPN alone strategy (cost $21,025, QALYs 0.012) was dominated by a combination of other strategies. The ICER of TPN+chemotherapy was $3.8 million/QALY compared to chemotherapy alone. In the novel therapy scenario, hospice was the least costly strategy, novel therapy alone was dominated, and TPN+novel therapy had an ICER of $11 million/QALY compared to the next less costly strategy. The model’s results were stable to multiple one-way sensitivity analyses.