We thank Tey et al for their analysis of our prognostic model. We are pleased that they were able to easily determine the prognostic factors and apply our prognostic model to their patient population. This has satisfied our intent to provide an easy-to-use but accurate prognostic model. However, we discourage using the model in allogeneic BMT patients who receive nonmyeloablative transplants (NMT), since the effect of NMT on graft-versus-leukemia effect and treatment-related mortality has not been fully studied and may render the prognostic model inaccurate in these patients. All of the data presented in our publication are from patients who received myeloablative conditioning regimens; we have not treated enough Hodgkin's disease patients with an NMT to determine if the model is applicable to this patient population. Only half (n=6) of the Royal Brisbane Hospital (RBH) patients received myeloablative conditioning regimens and represent too small a cohort to perform subgroup comparisons.
We simulated the data from the 12 patients treated at RBH, which supports the applicability of our prognostic model to allogeneic BMT patients. We reported EFS at 2 years in our data; however, all RBH patients had an event (relapse or death) before 13 months post-BMT, making a comparison between centers difficult. We re-ran our analysis and now report EFS at day 100 in order to compare our data with that from RBH. As shown in the table below, both the day 100 EFS estimates and median EFS from the RBH data fall well within our 95% confidence intervals and are comparable to our estimates. Although the RBH EFS curves did not yield a statistically significant difference, this could be due to either low statistical power or to the inclusion of NMT patients.
This is a preview of subscription content, access via your institution