Quality of Life, at What Cost?

Along with the many scientific advances in cancer therapeutics, the field of oncology has advanced by focusing on ways to improve the symptomatic effects this therapy has on our patients. We are learning how to better quantify the observation that some medications make our patients feel worse than others. With improved statistical methods, we now routinely incorporate quality of life measurements in phase III clinical trials. From the patients' perspective, it is critical that we continue this practice, especially in the evaluation of therapy for incurable diseases, where the quality of life is as important as the quantity.

In the study by Kiebert, Jonas, and Middleton[[1]] in this issue of Cancer Investigation, health-related quality of life (HRQL) was measured in patients with metastatic melanoma who were randomized to treatment with temozolomide or dacarbazine. Patients completed the EORTC Quality of Life Questionnaire after each cycle, and the HRQL data from the two arms were compared at 12 and 24 weeks. Because there were not enough patients and HRQL data at 24 weeks for accurate data analysis, we were left with evaluating the information at 12 weeks. At this time period, patients taking temozolomide had a small but statistically significant improvement in the physical functioning scale and individual symptom scales for fatigue and sleep disturbance compared to those taking dacarbazine. The other four functioning scales and seven symptom scales showed no difference between the two groups. The overall global quality of life scale was also not different statistically. If melanoma patients taking temozolomide have a better HRQL than those taking dacarbazine, the gain is small.

This study, and the original trial[[2]] comparing these medications, both claim a statistical advantage for temozolomide in progression-free survival (PFS). The PFS for temozolomide and dacarbazine was 1.9 months and 1.5 months, respectively. In the trial, patients were assessed for response after every two cycles, but the cycle length was different for temozolomide (28 days) and dacarbazine (21 days). Therefore, the first radiological evaluation for temozolomide was at 56 days (about 1.9 months) and for dacarbazine was at 42 days (about 1.5 months). The median time to progression effectively equaled the time of first radiological evaluation for each drug.

Finally, to complete a comparison of the two medications, we must examine the issue of drug cost. As of September 2002, the average wholesale price (AWP) in the United States for 2600 mg of dacarbazine, enough for one cycle in a patient with a body surface area of 2.0, is $360.45. The AWP for a cycle of temozolomide in the same patient would be $2895.50. Using the same 12-week evaluation period, the cost of the medications would be $1441.80 for dacarbazine and $8686.50 for temozolomide. The difference in cost appears to be substantially greater than the difference in quality of life and efficacy.