Validation of the predictive power of the hematopoietic cell transplantation co-morbidity index, performance status for non-relapse mortality and long-term survival after autologous transplantation in patients with hematological malignancies

Abstract

The hematopoietic cell transplantation comobrbidity index (HCTCI) was developed as a sensitive tool to capture pretransplant comorbidities among transplant recipients which will have infl uence on non relapse mortality (NRM) and overall posttransplant survival (OS). HCT-CI has not been widely validated among autologous recepients. We retrospectivelly evaluated if HCT-CI and karnofsky performance status (PS) and other readily available pretransplant variables concerning pretransplant mobilization strategies can predict the outcome of autologous recipients in our transplant center. We stratifi ed outcomes among 120 consecutive adult autologous recipients (47 AML in fi rst remission, 24 HD, 27 MM, 16 NHL, 4ALL). HCT-CI risk was low in 10 (12%), intermediate in 22 (27%) high in 45 (55%) and undetermined in 5 (6%). Two year OS was 45% (95%CI: 24–64%), 55% (95%CI: 40–68%) and 42% (95%CI: 24–64%) in the low, intermediate and high-risk HCT-CI groups respectively. Two year NRM was 36% (95% CI: 17–36%), 26% (95% CI: 15–39%) and 30% (95% CI: 22–39%) in the low, intermediate and high-risk HCT-CI groups respectively. The multivariate analysis revealed that HCT-CI failed in prediction of OS and NRM but KPS (< 90%) was a strong predictor of NRM as an independent predictor. The variables concerning mobilization of stem cells (chemotherapy with G-CSF versus G-CSF alone and the dose of infused CD34 + > 4,0 × 106/kg and < 4.0 × 106/kg in the three risk HCT-CI groups revealed that patients with HCTCI score > 3 and intermediate and high risk disease that received < 4.0 × 106/kg had 2 year NRM <30% and OS<45%, as well the patients mobilized with chemotherapy + G-CSF showed lower NRM in the HCT-CI > 3 (intermediate and high risk disease). To determine the validity of HCT-CI, KPS and weather to include the the independent variables concerning the mobilization strategy and stem cell dose that we analyzed, a multi-center collaboration is necessary to produce an adequately powered validation study for risk stratifi cation of autologous recipients.