Big databases: outcomes research begins to yield results.

R esearchers recently unveiled the first comparative-effectiveness study in oncology that used an outcomes database. The presentation took place at the American Society of Hematology meeting in New Orleans in December, where a research group using the National Comprehensive Cancer Network's (NCCN) database for non-Hodgkin lymphoma reported that patients with mantle cell lym-phoma, a rare but fatal form of the disease, did just as well when receiving autologous stem cell transplantation after standard che-motherapy as patients who received a highly toxic regimen of five drugs. The five-drug regimen's developers, mostly from the University of Texas M. D. Anderson Cancer Center in Houston, claimed that their method was superior because of a 75% progression-free survival rate at 3 years. After identifying 229 patients younger than 65 years with MCL, whose medical records had been entered into the NCCN database between August 2000 and February 2009, researchers conducted a retrospective cohort study and found that overall survival between the two treatment approaches was similar. The median progression-free survival rate at 3 years was also the same. One major difference existed, though: The median number of days in the hospital was lower in the transplant group, according to the abstract presented at the meeting. " This [finding] validated that it is reasonable to do the auto transplant because it has the same outcome plus less toxicity, " said Ann LaCasce, M.D., of Dana–Farber Cancer Institute in Boston, who led the 10-person research team from major cancer centers that contribute data to the NCCN project. They conducted the study because MCL treatment is driven largely by physician preference and expert opinion. No randomized clinical trials or prospective observational studies have ever been done comparing the two approaches, and none are anticipated. " People [with