Comparison of the Impact of Pre-Transplant Co-Morbidity Scores on Allogeneic Hematopoietic Stem Cell Transplant Outcomes
Bibliographic record
Abstract
Abstract Introduction: Allogeneic hematopoietic cell transplantation (allo-HCT) is potentially curative for the treatment of various hematological diseases, in part due to the effect of conditioning chemotherapy, and in part due to graft-versus-malignancy effect. However, alloHCT is associated with significant morbidity and mortality. Multiple co-morbidity indices have been published in the literature for the purpose of pre-transplant risk assessment. The purpose of the presented study is to assess a number of these pre-transplant scores on a single-center transplant population and to determine the score with improved risk stratification ability using concordance statistics. Methods: We investigated the impact of the prospectively collected Hematopoietic Cell Transplantation-Comorbidity Index (HCT-CI) on post-transplant outcomes for 243 recipients of allo-HCT performed between August 2014 and October 2016 at the Princess Margaret Cancer Center (Toronto, Canada), and compared this score to other pre-transplant scores including the age-adjusted HCT-CI, PAM score (Pre-transplant Assessment of Mortality Score) and the Disease Risk Index (DRI). Partitioning of the HCT-CI, HCT-CI/age and PAM scores into three groups was performed based on maximum significant differences on univariate analysis for overall survival (OS). Concordance statistics were used to compare the stratification power of the scores. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Inc, Cary, NC). Results: The median age at transplant is 56 years, patients were transplanted for AML (53%), ALL (7.5%), MDS (13.5%), MPN (14%), NHL/CLL (8.5%) and (3.5%) AA. Donors were matched related in 37%, unrelated in 59% and haploidentical in 3% of the patients. Reduced intensity conditioning chemotherapy was used in 132 patients (54%), 153 patients (63%) received in-vivo T-cell depletion by using Campath or ATG, both donor and recipient were CMV negative in 48 (20%) of the patients. DRI was high in 67 (29%), intermediate in 145 (62%) and low in 22 (9%) of patients. HCT-CI was 0 in 90 (37%), 1 in 49(21%) and ≥2 in 103 (43%) of patients. HCT-CI/age was 0 in 22 (10%), 1 in 72 (30%) and ≥2 in 148 (62%). PAM score was 1-17 in 157(68%), 18-24 in 70 (30%) and 25-27 in 7 (3%) of patients. Median follow up of survivors was 28 months (range 17-44 months). OS of the entire cohort was 51% and 43% at 2 and 5 years post-transplant respectively. Cumulative incidence of relapse (CIR) was 19% at 2 years. For OS, as grouped above, the DRI did not demonstrate a significant difference between groups (p=0.77). For HCT-CI, p=0.034 (Figure 1), for HCT-CI/age p=0.02 and for the PAM score p=0.38. For OS, for the DRI, the C-statistic was 0.51 (se=0.03, 95%CI 0.45-0.57). For the PAM score, C-statistic was 0.51 (se=0.02,95%CI 0.45-0.56). For the HCT-CI age, C-statistic was 0.56 (se=0.024, 95%CI 0.51-0.61). For the HCT-CI, C-statistic was 0.56 (se 0.02, 95% CI 0.50-0.61). For CIR, the PAM score demonstrated a superior C-statistic of 0.56 (se=0.06, 95%CI 0.44-0.67) compared to the other scores. For NRM, the HCT-CI score (Figure 2, p=0.039) is superior with C-statistic 0.56 (se=0.04, 95%CI=0.49-0.63). Conclusion: Based on the above described analysis, the original HCT-CI score as described by Sorror et aldemonstrates superior prognostic stratification ability for OS and NRM in our patient cohort compared to other scores. Further investigation for the development of an optimal risk scoring system for allogeneic HCT is required. Figure 1. Figure 1. Disclosures Kim: Paladin: Consultancy; Pfizer: Consultancy; Novartis: Consultancy, Honoraria, Research Funding; BMS: Consultancy, Honoraria, Research Funding. Lipton:Novartis: Consultancy, Honoraria, Research Funding; Pfizer: Consultancy, Honoraria, Research Funding; BMS: Consultancy, Honoraria, Research Funding; Takeda: Consultancy, Honoraria, Research Funding.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".