High-dose chemotherapy and autologous stem-cell transplantation for ovarian cancer: An autologous blood and marrow transplant registry report

Background: Autologous transplantation is increasingly used to treat epithe lial ovarian cancer. However, it is not clear which patients may benefit. Objective: To determine overall and progression-free survival and factors a ssociated with favorable outcome after autotransplantation for ovarian canc er. Design: Observational cohort study. Setting: 57 centers reporting to the Autologous Blood and Marrow Transplant Registry (ABMTR). Patients: 421 women who received transplants between 1989 and 1996. Interventions: High-dose chemotherapy using diverse regimens with hematopoi etic stem-cell rescue. Measurements: Primary outcomes were progression-free survival and overall s urvival. Multivariate analyses using Cox proportional hazards regression co nsidered the following factors: age, Kamofsky performance score, initial st age, histologic characteristics, previous therapy, remission status, extent of disease, graft source, transplant regimen, and year of transplantation. Results: Most patients had extensive previous chemotherapy. Forty-one perce nt had platinum-resistant tumors, and 38% had tumors at least 1 cm in diame ter. Only 34 patients (8%) received transplants as part of initial therapy. The probability of death within 100 days was 11% (95% CI, 8% to 14%). Two- year progression-free survival was 12% (CI, 9% to 16%), and 2-year overall survival was 35% (CI, 30% to 41%). Younger age, Karnofsky performance score of at least 90%, non-clear-cell disease, remission at transplantation, and platinum sensitivity were associated with better outcomes. Progression-fre e and overall survival were 22% (CI, 12% to 33%) and 55% (CI, 42% to 66%), respectively, for women with a high Karnofsky performance score and non-cle ar-cell, platinum-sensitive tumors. Conclusions: Some subgroups of patients with ovarian cancer seem to have go od outcomes after autotransplantation, although several biases may have aff ected these observations. Phase III trials are needed to compare such outco mes with outcomes of conventional chemotherapy.