Outcomes of Hematopoietic Stem Cell Transplantation in Patients Aged 70 Years and Older

Introduction

Many older adults are precluded from consideration for hematopoietic stem cell transplant (HCT) based on chronologic age despite this being potentially curative for hematologic malignancies. Within the past two decades, the development of reduced intensity HCT conditioning regimens and improvements in supportive care measures have extended the application of HCT to include more older adults. However, there remains limited data on HCT outcomes in patients who are aged 70 years or greater in this recent period. Here we describe outcomes of patients age ≥70 who received autologous (autoHCT) or allogeneic (alloHCT) at two institutions.

Methods

This retrospective study included patients age ≥70 who received HCT 1/1/2011-12/31/2020 at Vanderbilt University Medical Center or Tennessee Valley Healthcare System. The primary endpoint was overall survival (OS), characterized by Kaplan-Meier survival curve. Secondary endpoints included disease progression, cause of death and rates of Graft vs. Host Diseased (GVHD) in patients who underwent alloHCT. Pre-transplant characteristics included hematopoietic cell transplantation-specific comorbidity index (HCT-CI), number of previous treatments and disease type for which Cox regression analysis was performed. Multivariable procedures were calculated by applying lasso penalized regression with the penalty parameter selected by k-fold cross-validation.

Results

For patients receiving alloHCT (n=58), the median age was 71.9 (range: 70-76.2; IQR 70.8-73.4). The majority of patients were male (n= 50, 86%) and white (n=53, 91.4%). Most patients had myeloid neoplasms, specifically AML (n=26, 45%) and MDS (n= 23, 40%). All patients received non-myeloablative conditioning regimens. Most common regimens were fludarabine, busulfan and ATG (Flu-Bu-ATG; n=32, 55.2%) and fludarabine, cyclophosphamide and total body irradiation (Flu-Cy-TBI; n=15, 25.9%). All patients received peripheral blood stem cell graft. Median OS was 632 days (IQR: 282, 1560) and the primary cause of death was progression/relapse of disease (n=19, 48.7%). HCT-CI was not significantly associated with OS (r=0.011, p=0.89) or progression (r=0.064, p=0.71). Patients with more prior treatments were associated with decreased OS however this was not significant (r=0.057, p=0.78). Cumulative incidence of acute GVHD II-III was 29.3%. Higher grade GHVD was associated with decreased OS, but the effect was not significant (r=0.19, p=0.58).

For patients receiving autoHCT (N=201), the median age was 71.6 (range: 70-77.8; IQR 70.6-73.0). Patients were primarily male (n= 170, 85%) and white (n=162, 81%) and most patients (n=155, 77.1%) had multiple myeloma. The most common conditioning regimen was Melphalan (n= 168, 83.6%). Median OS was 1512 days (IQR: 882, 2099) and the primary cause of death was progression/relapse of disease (n=42, 49.4%). Higher HCT-CI was significantly associated with decreased OS (r=0.15, p=0.009). Patients with more prior treatments were associated with decreased OS (r=0.20; p=0.11) and increased progression (r=0.21; p=0.20) however these were not significant.

Conclusions

Autologous and allogenic HCT are safe in septuagenarians, but survival is still limited by disease progression/relapse. Patients should not be excluded from these intensive therapies based on chronologic age alone. HCT-CI in this population predicts for decreased OS with autoHCT. Improved understanding of clinical outcomes in this group of older adults will inform efforts to more accurately risk stratify patients and potentially broaden the number of older adults who have acceptable transplant risks, as well as identify potential interventions prior to transplant to remove possible barriers to curative intent treatment.