Data quality in trauma transfusion studies and the impact of missing data on predicting massive transfusion

Maximizing data quality may be especially difficult in trauma-related clinical research. Strategies are needed to improve data quality and assess the impact of data quality on clinical predictive models. This study had two objectives. The first was to compare missing data between two multi-center trauma transfusion studies: a retrospective study (RS) using medical chart data with minimal data quality review and the PRospective Observational Multi-center Major Trauma Transfusion (PROMMTT) study with standardized quality assurance. The second objective was to assess the impact of missing data on clinical prediction algorithms by evaluating blood transfusion prediction models using PROMMTT data. RS (2005-06) and PROMMTT (2009-10) investigated trauma patients receiving ≥ 1 unit of red blood cells (RBC) from ten Level I trauma centers. Missing data were compared for 33 variables collected in both studies using mixed effects logistic regression (including random intercepts for study site). Massive transfusion (MT) patients received ≥ 10 RBC units within 24h of admission. Correct classification percentages for three MT prediction models were evaluated using complete case analysis and multiple imputation based on the multivariate normal distribution. A sensitivity analysis for missing data was conducted to estimate the upper and lower bounds of correct classification using assumptions about missing data under best and worst case scenarios. Most variables (17/33=52%) had <1% missing data in RS and PROMMTT. Of the remaining variables, 50% demonstrated less missingness in PROMMTT, 25% had less missingness in RS, and 25% were similar between studies. Missing percentages for MT prediction variables in PROMMTT ranged from 2.2% (heart rate) to 45% (respiratory rate). For variables missing >1%, study site was associated with missingness (all p≤0.021). Survival time predicted missingness for 50% of RS and 60% of PROMMTT variables. MT models complete case proportions ranged from 41% to 88%. Complete case analysis and multiple imputation demonstrated similar correct classification results. Sensitivity analysis upper-lower bound ranges for the three MT models were 59-63%, 36-46%, and 46-58%. Prospective collection of ten-fold more variables with data quality assurance reduced overall missing data. Study site and patient survival were associated with missingness, suggesting that data were not missing completely at random, and complete case analysis may lead to biased results. Evaluating clinical prediction model accuracy may be misleading in the presence of missing data, especially with many predictor variables. The proposed sensitivity analysis estimating correct classification under upper (best case scenario)/lower (worst case scenario) bounds may be more informative than multiple imputation, which provided results similar to complete case analysis.^

Inducible caspase 9 suicide gene to improve the safety of allodepleted T cells after haploidentical stem cell transplantation.

Addback of donor T cells following T cell-depleted stem cell transplantation (SCT) can accelerate immune reconstitution and be effective against relapsed malignancy. After haploidentical SCT, a high risk of graft-versus-host disease (GVHD) essentially precludes this option, unless the T cells are first depleted of alloreactive precursor cells. Even then, the risks of severe GVHD remain significant. To increase the safety of the approach and thereby permit administration of larger T cell doses, we used a suicide gene, inducible caspase 9 (iCasp9), to transduce allodepleted T cells, permitting their destruction should administration have adverse effects. We made a retroviral vector encoding iCasp9 and a selectable marker (truncated CD19). Even after allodepletion (using anti-CD25 immunotoxin), donor T cells could be efficiently transduced, expanded, and subsequently enriched by CD19 immunomagnetic selection to >90% purity. These engineered cells retained antiviral specificity and functionality, and contained a subset with regulatory phenotype and function. Activating iCasp9 with a small-molecule dimerizer rapidly produced >90% apoptosis. Although transgene expression was downregulated in quiescent T cells, iCasp9 remained an efficient suicide gene, as expression was rapidly upregulated in activated (alloreactive) T cells. We have demonstrated the clinical feasibility of this approach after haploidentical transplantation by scaling up production using clinical grade materials.

Treatment of solid organ transplant recipients with autologous Epstein Barr virus-specific cytotoxic T lymphocytes (CTLs).

We have investigated the in vivo safety, efficacy, and persistence of autologous Epstein Barr virus (EBV)-specific cytotoxic T lymphocytes (CTLs) for the treatment of solid organ transplant (SOT) recipients at high risk for EBV-associated posttransplantation lymphoproliferative disease (PTLD). EBV-CTLs generated from 35 patients expanded with normal kinetics contained both CD8 and CD4 lymphocytes and produced significant specific killing of autologous EBV-transformed B lymphoblastoid cell lines (LCLs). Twelve SOT recipients at high risk for PTLD, or with active disease, received autologous CTL infusions without toxicity. Real-time polymerase chain reaction (PCR) monitoring of EBV-DNA showed a transient increase in plasma EBV-DNA suggestive of lysis of EBV-infected cells, although there was no consistent decrease in virus load in peripheral-blood mononuclear cells. Interferon-gamma enzyme-linked immunospot (ELISPOT) assay and tetramer analysis showed an increase in the frequency of EBV-responsive T cells, which returned to preinfusion levels after 2 to 6 months. None of the treated patients developed PTLD. One patient with liver PTLD showed a complete response, and one with ocular disease has had a partial response stable for over one year. These data are consistent with an expansion and persistence of adoptively transferred EBV-CTLs that is limited in the presence of continued immunosuppression but that nonetheless produces clinically useful antiviral activity.

A case-control study of hepatitis C virus and its interaction with hepatitis B virus on the development of hepatocellular carcinoma in the USA

Objective. The aim of this study was to assess the independent risk of hepatitis C virus (HCV) infection in the development of hepatocellular carcinoma (HCC). The independent risk of hepatitis B virus (HBV), its interaction with hepatitis C virus and the association with other risk factors were examined.^ Methods. A hospital-based case-control study was conducted between January 1994 and December 1995. We enrolled 115 pathologically confirmed HCC patients and 230 nonliver cancer controls, who were matched by age ($\pm$5 years), gender, and year of diagnosis. Both cases and controls were recruited from The University of Texas M. D. Anderson Cancer Center at Houston. The risk factors were collected through personal interviews and blood samples were tested for HCV and HBV markers. Univariate and multivariate analyses were performed through conditional logistic regression.^ The prevalence of anti-HCV positive is 25.2% in HCC cases compared to 3.0% in controls. The univariate analysis showed that anti-HCV, HBsAg, alcohol drinking and cigarette smoking were significantly associated with HCC, however, family history of cancer, occupational chemical exposure, and use of oral contraceptive were not. Multivariate analysis revealed a matched odds ratio (OR) of 10.1 (95% CI 3.7-27.4) for anti-HCV, and an OR of 11.9 (95% CI 2.5-57.5) for HBsAg. However, dual infection of HCV and HBV had only a thirteen times increase in the risk of HCC, OR = 13.9 (95% CI 1.3-150.6). The estimated population attributable risk percent was 23.4% for HCV, 12.6% for HBV, and 5.3% for both viruses. Ever alcohol drinkers was positively associated with HCC, especially among daily drinkers, matched OR was 5.7 (95% CI 2.1-15.6). However, there was no significant increase in the risk of HCC among smokers as compared to nonsmokers. The mean age of HCC patients was significantly younger among the HBV(+) group and among the HCV(+)/HBV(+) group, when compared to the group of HCC patients with no viral markers. The association between past histories of blood transfusion, acupuncture, tattoo and IVDU was highly significant among the HCV(+) group and the HBV(+)/HCV(+) group, as compared to HCC patients with no viral markers. Forty percent of the HCC patients were pathologically or clinically diagnosed with liver cirrhosis. Anti-HCV(+) (OR = 3.6 95% CI 1.5-8.9) and alcohol drinking (OR = 2.7 95% CI 1.1-6.7), but not HBsAg, are the major risk factors for liver cirrhosis in HCC patients.^ Conclusion. Both hepatitis B virus and hepatitis C virus were independent risk factors for HCC. There was not enough evidence to determine the interaction between both viruses. Only daily alcoholic drinkers showed increasing risk for HCC development, as compared to nondrinkers. ^