Palliation Proves Cost Effective

June 1, 1997

ASCO--In a randomized trial of patients with symptomatic refractory prostate cancer, chemotherapy plus prednisone provided significantly better pain control than prednisone alone; now, an economic analysis suggests that the combination was less expensive overall due to fewer hospital admissions.

ASCO--In a randomized trial of patients with symptomatic refractoryprostate cancer, chemotherapy plus prednisone provided significantly betterpain control than prednisone alone; now, an economic analysis suggeststhat the combination was less expensive overall due to fewer hospital admissions.

Speaking at the ASCO meeting, David J. Bloomfield, MD, of Princess MargaretHospital, Toronto, said that the retrospective review of 114 patients showeda treatment cost, from randomization to death, of $29,000 Canadian withmitox-antrone (Novantrone) plus prednisone versus $27,300 for prednisonealone.

The dominant cost category was inpatient admissions, accounting fornearly 60% of total costs. Thus, the cost of the mitoxantrone was morethan offset by an increase in inpatient cost in the predni-sone only arm.

Although there was no difference in survival between the two groups,a cost utility analysis showed that patients receiving chemotherapy hadmore quality-adjusted weeks of life (39.5 versus 27.5 for those on prednisonealone).

Since the confidence interval was plus or minus $6,000, he said, "atthe bottom end of our confidence estimates, the saving could be as muchas $9,200; at the upper end, there is an increased cost of $5,800."This works out to a cost of $22,400 per quality-adjusted life-year formitoxantrone/prednisone, well within the range of other accepted interventions.

Dr. Bloomfield noted that more than half of the patients randomizedto predni-sone alone crossed over to the combination during the trial,"making the trial one effectively of initial versus delayed mitoxantrone,"and creating a methodological problem for the cost analysis.

"It is usual to have two equivalent time lines to compare costs,"he said, "and in this trial there is no right time at which to comparethe two strategies."

To solve this problem, the researchers plotted all resources used ina time line by date for each patient. A plot of the cumulative mean costsfor all patients showed that mitoxantrone was "consistently cheaperall along the time line."