Risk Factors and Outcomes for Relapse after Autologous Stem Cell Transplantation for Hodgkin Lymphoma.

Despite successful outcomes for many patients (pts) with relapsed or refractory Hodgkin lymphoma, disease recurrence leads to a significant number of treatment failures after high-dose chemotherapy with autologous stem cell transplantation (ASCT). Using a prospectively maintained database, a retrospective analysis was conducted in 245 consecutive pts who underwent ASCT for Hodgkin lymphoma from 1985 through 2005 at the Cleveland Clinic Foundation. Objectives were to identify risk factors and outcomes associated with early relapse (within 1 year post ASCT) versus late relapse (over 1 year post ASCT). Patient characteristics: male gender in 64%; median age, 34 years (range, 18–70); median time from diagnosis to transplant, 22 months (4–327); Karnofsky score > 80 in 89%; median 2 prior regimens (1–4); prior XRT in 52%; bulky disease > 10 cm in 20%; stage III-IV in 62%; disease status of CR1/PR1 in 8%, CR2/PR2 in 69%, refractory in 11%, other in 12%; preparative regimen of BuCyVP in 65%, CBV in 26%, TBI/others in 9%. Median follow-up after ASCT for all pts was 35 months (0–223); for surviving pts, 74 months (3–223). Relapse occurred in 115 pts; 5-year estimated relapse-free survival was 36%, and 5-year estimated relapse rate was 44%. Early relapse occurred in 72 pts and was associated with poor survival (median 13.8 months), with only 5 of these pts surviving at least 5 years. Late relapse occurred in 43 pts and was characterized by markedly better median survival of 57.9 months. Among these pts, 20 survived for 5 or more years post ASCT, and 2 survived more than 10 years.

In univariate analyses, early relapse was associated with gender (P=0.023), number of prior chemotherapy regimens (P=0.01), bulky disease (P=0.045), and disease status at transplant (P=0.06). Multivariate analysis confirmed that early relapse was associated with gender (male/female: HR 2.01, 95% CI 1.04–3.90, P=0.038), disease status at transplant (CR or PR vs. other: HR 0.33, 95% CI 0.17–0.64, P=0.001), and number of prior chemotherapy regimens (4 or more vs. 1–3: HR 6.27, 95% CI 2.21–17.8, P=0.001). In contrast, late relapse was associated with time from diagnosis to ASCT (P=0.09) and slower platelet engraftment (P=0.047) in univariate analyses. Time from diagnosis to ASCT was the factor associated most strongly with late relapse in multivariate analysis (per 6 month increase: HR 0.92, 95% CI 0.86–1.00, P=0.049). In conclusion, a substantial proportion of pts who relapse more than 1 year post ASCT for Hodgkin lymphoma may survive for 5 years or longer, but relapse within 1 year after ASCT is associated with a poor prognosis. The association of late relapse after ASCT with a longer period from initial diagnosis to ASCT signifies a subset of pts whose lymphoma exhibits relatively indolent behavior throughout its clinical course. Risk factors for early relapse after ASCT differ from those associated with late relapse, suggesting that novel interventions could be directed toward improving outcomes for pts at high risk for early relapse.