27
Geriatric Assessment (GA) to Predict Survival in Older Allogeneic Hematopoietic Cell Transplantation (HCT) Recipients
BACKGROUND
GA predicts for mortality in older solid tumor cancer patients but its prognostic value in allogeneic HCT has not previously been reported.
METHODS
Patients >=50 years were eligible. GA was performed within 1 month prior to HCT, and evaluated function and disability, frailty, comorbidity, mental health, nutrition, and inflammation with the following measures: ECOG performance status (PS), Activities of Daily Living, Instrumental Activities of Daily Living (IADL), SF-36 Physical Component Summary, Fried Frailty Index (incorporates walk speed and grip strength), HCT- Comorbidity Index (HCT-CI), Cumulative Illness Rating Scale- Geriatrics, SF-36 Mental Component Summary (SF36-MCS), serum albumin, self-reported weight loss, and serum C-reactive protein (CRP). Survival curves were generated using Kaplan-Meier method; log-rank test compared groups. Multivariate models for 2-year overall survival (OS) incorporated each GA variable significant at P <0.10 in univariate analysis, adjusting for standard HCT variables (age, HCT-CI, conditioning regimen intensity, disease risk).
RESULTS
203 adults >=50 years completed GA and underwent HCT. Mean age was 59 years (range 50-73); 45% had high disease risk, 76% received reduced intensity conditioning, and 14% underwent cord blood HCT. With median follow-up of 36 months, IADL limitations (P < 0.0001), slow walk speed (P = 0.01), low SF36-MCS (P = 0.01), and high CRP (P <0.001) were significantly associated with inferior OS, independent of standard HCT variables. The prognostic effect of these GA variables was greater in older recipients (Table 1). We then created a simple risk score with 1 point for the most prognostic functional measure (IADL impairment) and 1 point for comorbidity (HCT-CI >=3). This significantly stratified outcomes, particularly in those ³60 years, such that 2-year OS was 63%, 29%, and 0% for 0, 1, and 2 points, respectively.
CONCLUSION
GA measures confer independent prognostic utility in older HCT recipients, especially in those >=60 years. Implementation of GA prior to HCT may aid in appropriate selection of older adults for HCT.
Table 1. Multivariate OS Analysis
| Total Cohort HR 95% CI P
| 50-59 Years HR 95% CI P
| 60-73 Years HR 95% CI P
| ||||||
STANDARD VARIABLES |
|
|
|
|
|
|
|
|
|
Age >60
| 1.83
| 1.26-2.65
| .001
|
|
|
|
|
|
|
HCT-CI >=3
| 1.56
| 1.07-2.28
| .02
| 1.50
| .88-2.53
| .13
| 1.72
| .99-2.98
| .05
|
Active Disease
| 1.31
| .90-1.90
| .16
| 1.54
| .92-2.58
| .10
| 1.27
| .71-2.27
| .42
|
Ablative Regimen
| 1.54
| 1.02-2.31
| .04
| 2.14
| 1.24-3.69
| .01
| 1.07
| .54-2.10
| .85
|
GA VARIABLES |
|
|
|
|
|
|
|
|
|
IADL Impairment
| 2.38
| 1.59-3.56
| <.001
| 1.86
| 1.07-3.24
| .03
| 3.25
| 1.75-6.05
| <.001
|
Slow Walk Speed
| 1.80
| 1.14-2.83
| .01
| 1.16
| .60-2.28
| .66
| 3.27
| 1.68-6.39
| .001
|
Reduced Mental Health
| 1.67
| 1.13-2.48
| .01
| 1.55
| .92-2.62
| .10
| 1.87
| 1.01-3.49
| .04
|
Low Albumin
| 1.52
| .94-2.46
| .09
| 1.23
| .57-2.63
| .60
| 2.62
| 1.26-5.47
| .01
|
High CRP
| 2.51
| 1.54-4.09
| <.001
| 1.89
| .94-3.79
| .07
| 3.13
| 1.52-6.46
| .002
|
*Each GA measure analyzed separately, adjusting for standard variables
| |||||||||
See more of: BMT Tandem "Scientific" Meeting