Abstract
Multiple myeloma (MM) is an incurable disease with poor survival outcomes. Recent trials have suggested improved overall survival with newer agents like bortezomib (VELCADE). In the process of developing a life-time cost-effectiveness analysis of therapies for relapsed or refractory MM, we noted the lack of long-term survival data.
Objective: We sought to extrapolate 1 to 3 year overall survival data to 10 years (most patients die by then) in order to estimate the number of life years expected with several MM therapies. The therapies included were bortezomib, high dose dex (HDD), thalidomide regimens and standard care. Standard care was a basket of treatments weighted by frequency of use reported in a Canadian survey of physicians treating MM patients. It included HDD, thalidomide regimens, MP, VAD, cyclophosphamide, bortezomib, and repeat stem cell transplant.
Method: The APEX trial results were used to inform the 3 year clinical benefits of bortezomib and HDD for relapsed MM (Richardson, 2005). Three-year survival for bortezomib was available from the APEX study. However due to positive interim results HDD patients were allowed to switch to bortezomib, thus limiting the HDD data available to one year. One to three year survival with HDD, thalidomide and standard care therapies were estimated from published trials and observational studies. The natural history of relapsed MM patients from Kumar (2004) was used to extrapolate survival to 10 years. Kumar studied the clinical course of 578 relapsed MM patients at the Mayo clinic from 1985 to 1998. From this study, the conditional survival, S(t|t-1) was calculated as the ratio of survival at the end of year (S(t)) to the survival at the end of the year before (S(t-1)). We assumed the rate of death in years 4 to 5, 5 to 6, etc. was the same for all therapies.
Results: Figure 1 illustrates the estimated 10 year overall survival by therapy. From these survival curves a method of estimating the area-under-the curve is used to obtain the average life-years for patients on each therapy. Using bortezomib as an example, at year one 80 of 100 patients would be alive - resulting in 80 life years. At year two, 57 of the patients would be alive. Thus after two years, we accumulate 138 (80+58) years of life. Continuing to 10 years, we accumulate a total of 284 life years, or 2.84 years per patient on bortezomib. Using the same method for the other treatments, patients would live on average 1.81 years with HDD, 2.39 years with thalidomide regimens and 2.25 years with standard care. Bortezomib provides the largest mean overall survival over a 10-year time horizon. In fact, bortezomib can provide up to 1.03 years of added life compared to HDD.
Conclusion: This is a conservative estimate of the overall survival advantage of bortezomib. It was assumed the mortality rate in years 4 to 5, 5 to 6, etc. was the same for all therapies, thus not giving bortezomib any clinical advantage from years 4 to 10. Long-term extrapolated analyses are needed in order to capture the full benefit of therapies for which only short-term trial data is available.
10-Year Survival By Therapy 10-Year Survival By Therapy
Using Citronella to Protect Bees (honeybee Apis mellifera L.) from certain Insecticides and Their Nano Formulations
