Biomedical subjects

Mei-Jie Zhang

Publications and source records attributed to Mei-Jie Zhang.

At least 19 recordsLinked to original sources

PubMed2007

Asymptotic theory for the Cox semi-Markov illness-death model.

Irreversible illness-death models are used to model disease processes and in cancer studies to model disease recovery. In most applications, a Markov model is assumed for the multistate model. When there are covariates, a Cox (1972, J Roy Stat Soc Ser B 34:187-220) model is used to model the effect of covariates on each transition intensity. Andersen et al. (2000, Stat Med 19:587-599) proposed a Cox semi-Markov model for this problem. In this paper, we study the large sample theory for that model and provide the asymptotic variances of various probabilities of interest. A Monte Carlo study is conducted to investigate the robustness and efficiency of Markov/Semi-Markov estimators. A real data example from the PROVA (1991, Hepatology 14:1016-1024) trial is used to illustrate the theory.

Biometry↗

PubMed2006

Unrelated donor bone marrow transplantation for the treatment of Fanconi anemia.

Bone marrow transplantation (BMT) is the only known cure for the hematologic manifestations of Fanconi anemia (FA). Potential benefits of unrelated donor BMT for FA, however, have been severely limited by graft rejection and treatment-related mortality with resultant poor survival. Therefore, we evaluated the impact of potential prognostic factors on hematopoietic recovery, graft-versus-host disease (GVHD), and mortality in 98 recipients of unrelated donor BMT who received transplants between 1990 and 2003. Probabilities of neutrophil (89% vs 69%; P = .02) and platelet (74% vs 23%; P < .001) recovery were higher after fludarabine-containing regimens than nonfludarabine-containing regimens. Risks of acute GVHD (relative risk [RR], 4.29; P < .001) were higher with non-T-cell-depleted grafts. The day-100 mortality rate was significantly higher after nonfludarabine-containing regimens than fludarabine-containing regimens (65% vs 24%, respectively; P < .001). Corresponding 3-year adjusted overall survival rates were 13% versus 52% (P < .001). In addition, mortality was higher in recipients who were older (> 10 years), who were cytomegalovirus (CMV) seropositive, and who received more than 20 blood product transfusions before BMT. Based on these results, significant practice changes are suggested: use of a fludarabine-containing conditioning regimen in the context of T-cell-depleted marrow allografts, and earlier referral for transplantation prior to excessive transfusions in patients with marrow failure.

Acute Disease↗

PubMed2006

Long-term outcome of patients given transplants of mobilized blood or bone marrow: A report from the International Bone Marrow Transplant Registry and the European Group for Blood and Marrow Transplantation.

We previously compared outcomes after allogeneic peripheral-blood stem cell (PBSC) and bone marrow (BM) transplantation in 706 patients with leukemia. We obtained long-term follow up on 413 of 491 patients who were alive at the time of the initial report: 141 PBSC and 272 BM recipients. Chronic graft-versus-host disease (GVHD) was more frequent after PBSC compared to BM transplantation (RR 1.65, P < .001) yet relapse rates were similar in both groups. Leukemia-free survival rates were higher after PBSC than BM transplantation for patients with advanced chronic myeloid leukemia (33% versus 25%) but lower for those in first chronic phase (41% versus 61%) due to higher rates of late transplant-related mortality. Leukemia-free survival was similar after PBSC and BM transplantation for acute leukemia. These data represent the early experience with PBSC grafts. Long-term outcomes in recipients of more recent transplants are required to better evaluate the role of PBSC grafts relative to BM transplantation.

Bone Marrow Transplantation↗

PubMed2006

Outcomes after HLA-matched sibling transplantation or chemotherapy in children with B-precursor acute lymphoblastic leukemia in a second remission: a collaborative study of the Children's Oncology Group and the Center for International Blood and Marrow Transplant Research.

The best treatment approach for children with B-precursor acute lymphoblastic leukemia (ALL) in second clinical remission (CR) after a marrow relapse is controversial. To address this question, we compared outcomes in 188 patients enrolled in chemotherapy trials and 186 HLA-matched sibling transplants, treated between 1991 and 1997. Groups were similar except that chemotherapy recipients were younger (median age, 5 versus 8 years) and less likely to have combined marrow and extramedullary relapse (19% versus 30%). To adjust for time-to-transplant bias, treatment outcomes were compared using left-truncated Cox regression models. The relative efficacy of chemotherapy and transplantation depended on time from diagnosis to first relapse and the transplant conditioning regimen used. For children with early first relapse (< 36 months), risk of a second relapse was significantly lower after total body irradiation (TBI)-containing transplant regimens (relative risk [RR], 0.49; 95% confidence interval [CI] 0.33-0.71, P < .001) than chemotherapy regimens. In contrast, for children with a late first relapse (> or = 36 months), risks of second relapse were similar after TBI-containing regimens and chemotherapy (RR, 0.92; 95% CI, 0.49-1.70, P = .78). These data support HLA-matched sibling donor transplantation using a TBI-containing regimen in second CR for children with ALL and early relapse.

Adolescent↗

PubMed2006

Comparable long-term survival after unrelated and HLA-matched sibling donor hematopoietic stem cell transplantations for acute leukemia in children younger than 18 months.

PURPOSE: To describe outcomes after unrelated donor stem cell transplantation (HCT) in children (< 18 months at diagnosis) with acute leukemia and compare these with outcomes after human leukocyte antigen (HLA)-matched sibling donor HCT. PATIENTS AND METHODS: We compared the results of unrelated donor HCT with bone marrow (n = 85) or cord blood grafts (n = 81) and HLA-matched sibling donor HCT with bone marrow grafts (n = 101) for acute myeloid or acute lymphoblastic leukemia using Cox proportional hazards models. Unrelated donor HCT recipients were younger, more likely to have MLL gene rearrangement, to have advanced leukemia, and to receive irradiation before HCT. RESULTS: Treatment-related mortality rates were 6%, 15%, and 31% after matched sibling, unrelated donor bone marrow, and cord blood HCT, respectively. Risks of relapse, overall and leukemia-free survival were significantly associated with disease status at transplantation. Though leukemia recurrence was lowest after unrelated donor HCT in first clinical remission (CR), overall survival, and leukemia-free survival rates were similar after matched sibling and unrelated donor HCT, after adjustment for disease status. Relapse, overall and leukemia-free survival did not differ by graft type (bone marrow v cord blood) or type of leukemia. Three-year probabilities of leukemia-free survival were 49% and 54% after HLA-matched sibling and unrelated donor transplantation in first CR, respectively. Corresponding rates for those with advanced leukemia were 20% and 30%. CONCLUSION: Unrelated donor HCT should be considered for infants with acute leukemia in first CR using the same eligibility criteria as are currently used for those with HLA matched sibling donors.

Female↗

PubMed2006

Regression models for hazard rates versus cumulative incidence probabilities in hematopoietic cell transplantation data.

In this article, we consider methods of regression modeling in the competing risks setting commonly encountered in analyzing stem cell transplantation data. We clarify the distinction between modeling the cause-specific hazard rate and modeling the cumulative incidence probability or function, and we review regression techniques for both types of quantities. We apply them to 2 examples: 1 comparing engraftment and 1 examining relapse. These examples illustrate that different conclusions may result depending on the type of regression model used for comparing treatments. Finally, we show how these discrepancies occur because 2 different characteristics of the time-to-event distribution are being modeled.

Graft Rejection↗

PubMed2006

Pretransplantation consolidation chemotherapy decreases leukemia relapse after autologous blood and bone marrow transplants for acute myelogenous leukemia in first remission.

Controversy exists over whether pretransplantation consolidation chemotherapy affects the outcome of subsequent autotransplantation for acute myelogenous leukemia (AML). The current study was undertaken to determine the association between previous consolidation and outcome of autotransplantation for AML in first remission. Posttransplantation outcomes of 146 patients receiving no consolidation were compared with those of 244 patients receiving standard-dose (<1 gm/m(2)) and 249 patients receiving high-dose (1-3 gm/m(2)) cytarabine, using proportional hazards regression to adjust for differences in prognostic variables. One-year transplantation-related mortality was similar among the cohorts. Five-year relapse rates were 49% (95% confidence interval CI} = 39%-58%) with no consolidation, 35% (95% CI = 29%-42%) with standard-dose cytarabine, and 40% (95% CI = 33%-48%) with high-dose cytarabine (P = .07). Five-year leukemia-free survival rates were 39% (95% CI = 30%-47%) with no consolidation, 53% (95% CI = 46%-60%) with standard-dose cytarabine, and 48% (95% CI = 40%-56%) with high-dose cytarabine (P = .03). Similarly, 5-year overall survival was better in those patients receiving consolidation: 42% (95% CI = 34%-51%) with no consolidation, 59% (95% CI = 52%-65%) with standard-dose cytarabine, and 54% (95% CI = 46%-61%) with high-dose cytarabine (P = .01). Although most patients received 1 or 2 cycles of consolidation, the number of courses had no detectable effect on transplantation outcome. In multivariate analysis, risks of relapse and treatment failure were lower in the patients receiving consolidation, especially among those patients receiving blood cell grafts. Outcomes with standard-dose and high-dose cytarabine were similar. Based on our findings, we recommend that patients with AML in first remission receive consolidation before undergoing autotransplantation.

Adolescent↗

PubMed2006

A comparison of cyclophosphamide and total body irradiation with etoposide and total body irradiation as conditioning regimens for patients undergoing sibling allografting for acute lymphoblastic leukemia in first or second complete remission.

We compared the outcomes of 298 patients with acute lymphoblastic leukemia in first or second complete remission (CR1 or CR2) receiving HLA-matched sibling allografts after cyclophosphamide and total body irradiation (Cy-TBI) conditioning with 204 patients receiving etoposide and TBI. Consequently, 4 groups were compared: Cy-TBI <13 Gy (n = 217), Cy-TBI > or =13 Gy (n = 81), etoposide-TBI <13 Gy (n = 53), and etoposide-TBI > or =13 Gy (n = 151). Analyses of relapse, leukemia-free survival (LFS), and survival were performed separately for CR1 and CR2 transplantations. Transplant-related mortality did not differ by conditioning regimen. In CR1, there were also no significant differences in relapse, LFS, or survival by conditioning regimen. In CR2, these outcomes differed among conditioning groups. In comparison with Cy-TBI <13 Gy, the risks of relapse, treatment failure (inverse of LFS), and mortality tended to be lower with etoposide (regardless of TBI dose) or with TBI doses > or =13 Gy. For both CR1 and CR2 transplantations, causes of death were similar among the groups; disease recurrence accounted for 47% of deaths. We conclude that for HLA-identical sibling allografts for acute lymphoblastic leukemia in CR2, there is an advantage in substituting etoposide for Cy or, when Cy is used, in increasing the TBI dose to > or =13 Gy.

Adolescent↗

PubMed2006

The impact of human herpesvirus 6B reactivation on early complications following allogeneic hematopoietic stem cell transplantation.

Although human herpesvirus 6 (HHV-6) has been considered an important opportunistic and potentially fatal pathogen for allogeneic hematopoietic stem cell transplantation (HSCT), the clinical significance of HHV-6 reactivation remains controversial. In this study, we monitored HHV-6 DNAemia in 72 consecutive allogeneic HSCT recipients by real-time quantitative polymerase chain reaction. A total of 680 peripheral blood specimens were collected from the recipients before HSCT or at weekly intervals after HSCT. As the predominant variant, HHV-6B was detectable at least once in 47.2% (34/72) of HSCT recipients on the median day 21 (range, 7-84 days); HHV-6A reactivation occurred in only 1 recipient (1.4%). Detectable HHV-6B reactivation was associated with increased probability of skin rash by day 30 after HSCT (hazard ratio [HR], 3.68; 95% confidence interval [CI], 1.24-10.92; P = .019), cytomegalovirus (CMV) antigenemia (HR, 2.35; 95%CI, 1.32-4.19; P = .004), and hemorrhagic cystitis (HC) (HR, 2.59; 95%CI, 0.96-6.98; P = .061) by day 100 after HSCT. Neutrophil and platelet engraftment, mortality for 100 days after HSCT were not affected by HHV-6B reactivation. In conclusion, HHV-6 reactivation is a common event, and this study demonstrates a correlation between HHV-6B infection and CMV reactivation, early rash, and possibly increased incidence of HC after transplantation.

Adolescent↗

PubMed2006

Autotransplantation versus HLA-matched unrelated donor transplantation for acute myeloid leukaemia: a retrospective analysis from the Center for International Blood and Marrow Transplant Research.

Most acute myeloid leukaemia (AML) patients lack human leucocyte antigen-identical sibling donors for transplantation. Autotransplants and unrelated donor (URD) transplants are therapeutic options. To compare autologous versus URD transplantation for AML in first (CR1) or second complete remission (CR2), we studied the outcomes of 668 autotransplants were compared with 476 URD transplants reported to the Center for International Blood and Marrow Transplant Research. Proportional hazards regression adjusted for differences in prognostic variables. In multivariate analyses transplant-related mortality (TRM) was significantly higher and relapse lower with URD transplantation. Adjusted 3-year survival probabilities were: in CR1 57 (53-61)% with autotransplants and 44 (37-51)% URD (P = 0.002), in CR2 46 (39-53)% and 33 (28-38)% respectively (P = 0.006). Adjusted 3-year leukaemia-free survival (LFS) probabilities were: CR1 53 (48-57)% with autotransplants and 43 (36-50)% with URD (P = 0.021), CR2 39 (32-46)% and 33 (27-38)% respectively (P = 0.169). Both autologous and URD transplantation produced prolonged LFS. High TRM offsets the superior antileukaemia effect of URD transplantation. This retrospective, observational database study showed that autotransplantation, in general, offered higher 3-year survival for AML patients in CR1 and CR2. Cytogenetics, however, were known in only two-thirds of patients and treatment bias cannot be eliminated.

Acute Disease↗

PubMed2005

Conditioning including antithymocyte globulin followed by unmanipulated HLA-mismatched/haploidentical blood and marrow transplantation can achieve comparable outcomes with HLA-identical sibling transplantation.

The outcomes of 293 patients with leukemia undergoing HLA-identical sibling (n = 158) or related HLA-mismatched (n = 135) hematopoietic cell transplantation (HCT) performed during the same time period were compared. Patients received BUCY2 in HLA-identical sibling HCT or BUCY2 + ATG in mismatched HCT as conditioning regimens, followed by unmanipulated marrow and/or peripheral blood (PB) transplantation. All patients achieved full engraftment. The cumulative incidences of grades II to IV acute graft-versus-host disease (aGVHD) in the matched and mismatched cohorts were 32% (CI, 25%-39%) versus 40% (CI, 32%-48%, P = .13), respectively, with the relative risk (RR) = 0.64 (95% CI, 0.43-0.94), P = .02. The incidence of chronic GVHD did not differ significantly between the cohorts (P = .97). Two-year incidences of treatment-related mortality and relapse for matched versus mismatched were 14% (range, 9%-20%) versus 22% (range, 15%-29%) with P = .10 and 13% (range, 8%-19%) versus 18% (range, 10%-27%) with P = .40, respectively. Two-year adjusted leukemia-free survival (LFS) and overall survival were 71% (range, 63%-78%) versus 64% (range, 54%-73%) with P = .27 and 72% (range, 64%-79%) versus 71% (range, 62%-77%) with P = .72, respectively. Multivariate analyses showed that only advanced disease stage and a diagnosis of acute leukemia had increased risk of relapse, treatment failure, and overall mortality. In summary, HCT performed with related HLA-mismatched donors is a feasible approach with acceptable outcomes.

Acute Disease↗

PubMed2005

Pairwise multiple comparison adjustment in survival analysis.

Many clinical studies have as their endpoint the time until some event (such as death) occurs. Often in such studies researchers are interested in comparing several treatment or prognostic groups with one another in terms of their survival curves. When many such pairwise group comparisons are done, the chance of finding a false significance among all of the comparisons is inflated above the usual desired significance level. This paper investigates methods of adjusting the survival analysis for the number of comparisons being made. These methods are applied to a retrospective study conducted by the International Bone Marrow Transplant Registry and compared in a simulation study in terms of the power to detect actual differences in the survival curves between the groups.

Bone Marrow Transplantation↗

PubMed2005

A class of goodness of fit tests for a copula based on bivariate right-censored data.

The copula of a bivariate distribution, constructed by making marginal transformations of each component, captures all the information in the bivariate distribution about the dependence between two variables. For frailty models for bivariate data the choice of a family of distributions for the random frailty corresponds to the choice of a parametric family for the copula. A class of tests of the hypothesis that the copula is in a given parametric family, with unspecified association parameter, based on bivariate right censored data is proposed. These tests are based on first making marginal Kaplan-Meier transformations of the data and then comparing a non-parametric estimate of the copula to an estimate based on the assumed family of models. A number of options are available for choosing the scale and the distance measure for this comparison. Significance levels of the test are found by a modified bootstrap procedure. The procedure is used to check the appropriateness of a gamma or a positive stable frailty model in a set of survival data on Danish twins.

Algorithms↗

PubMed2004

Association of transplant center and physician factors on mortality after hematopoietic stem cell transplantation in the United States.

The effect of the organization and delivery of health care at medical centers, referred to as "center effects," with clinical outcomes after hematopoietic stem cell transplantation (HSCT) is not clear. We examined the association between center and treatment provider factors and mortality after HSCT. We surveyed 163 (87% response rate) United States transplantation centers that performed HLA-identical sibling HSCT for leukemia or autologous HSCT for lymphoma between 1998 and 2000 among patients at least 18 years old. One hundred thirteen (69%) centers performed HLA-identical sibling transplantations, whereas 162 (99%) performed autologous transplantations. Factors associated with decreased 100-day mortality in the allogeneic setting include a higher patient-per-physician ratio (P = .003) and centers where physicians answer calls after office hours (P = .03). Medical school affiliation was not associated with increased 100-day mortality except in centers where students/residents are present without fellows (P = .02). Center effects were weaker in autologous HSCT at 1 year. Differences in 100-day mortality in patients receiving transplants in centers with favorable versus unfavorable factors were greater in allogeneic than autologous HSCT. Greater physician involvement in patient care is important in producing favorable outcomes after HSCT. To more clearly establish the role of the factors we identified, further studies are recommended.

Academic Medical Centers↗

PubMed2004

Autologous hematopoietic stem-cell transplantation for children with acute myeloid leukemia in first or second complete remission: a prognostic factor analysis.

PURPOSE: To determine prognostic factors correlated with outcomes after autologous hematopoietic stem-cell transplantation (HSCT) in children with acute myeloid leukemia (AML). PATIENTS AND METHODS: We studied 219 children who received autologous HSCT for AML in first complete remission (CR) and 73 children in second CR and who were reported to the Autologous Blood and Marrow Transplant Registry. Among 29 of 73 patients who underwent transplantation in second CR, duration of first CR was > or = 12 months. RESULTS: Three-year cumulative incidences of relapse were 37% (95% CI, 31% to 44%), 60% (95% CI, 41% to 74%), and 36% (95% CI, 20% to 53%) for children in first CR, second CR after a short (< 12 months) first CR, and second CR after a long (> or = 12 months) first CR, respectively. Corresponding 3-year probabilities of leukemia-free survival were 54% (95% CI, 47% to 60%), 23% (95% CI, 10% to 39%), and 60% (95% CI, 42% to 75%). In multivariate analyses, risks of relapse, mortality, and treatment failure (relapse or death, inverse of leukemia-free survival) were higher for patients in second CR after a short first CR than for the other two groups. Transplant-related mortality, treatment failure, and overall mortality rates were higher in older (> 10 years) children. CONCLUSION: Duration of first CR seems to be the most important determinant of outcome. Results in children who experience treatment failure with conventional chemotherapy support the use of autologous transplantation as salvage therapy if such patients achieve a subsequent CR.

Acute Disease↗

PubMed2004

Myeloablative allogeneic hematopoietic stem cell transplantation in patients who experience relapse after autologous stem cell transplantation for lymphoma: a report of the International Bone Marrow Transplant Registry.

Myeloablative allogeneic hematopoietic stem cell transplantation (allo-HSCT) is increasingly used in patients with lymphoma who experience disease relapse after autologous hematopoietic stem cell transplantation (auto-HSCT) because the allograft is tumor free and may induce a graft-versus-tumor effect. We analyzed 114 patients treated with this approach from 1990 to 1999 to assess disease progression, progression-free survival (PFS), and overall survival (OS). Cumulative incidence of disease progression at 3 years was 52%, whereas treatment-related mortality was 22%, lower than previously reported. Three-year probabilities of OS and PFS were 33% and 25%, respectively. With prolonged follow-up, however, nearly all patients experienced disease progression, and 5-year probabilities were 24% and 5%, respectively. Complete remission at the time of allo-HSCT and use of total body irradiation (TBI) in patients with non-Hodgkin lymphoma (NHL) were associated with lower rates of disease progression and higher rates of OS. In summary, allo-HSCT is feasible for patients with lymphoma who have relapses after auto-HSCT and can result in prolonged survival for some, but it is usually not curative. Most likely to benefit are patients who have HLA-matched sibling donors, are in remission, and have good performance status.

Adolescent↗

PubMed2003

Decreased treatment failure in recipients of HLA-identical bone marrow or peripheral blood stem cell transplants with high CD34 cell doses.

We studied the association between CD34 cell dose and transplant outcomes in 359 bone marrow (BM) and 511 peripheral blood stem cell (PBSC) transplant recipients from human leucocyte antigen (HLA)-identical siblings, reported to the International Bone Marrow Transplant Registry (IBMTR). Transplants for leukaemia were performed between 1995 and 1998. Patients were divided into those receiving below or above the median CD34+ dose, for BM (3 x 106/kg) and PBSC (6 x 106/kg) grafts respectively. Cox proportional hazards regression was used to adjust for baseline patient-, disease- and transplant-related characteristics. Analysis of the BM recipients showed that high CD34 cell dose was associated with lower transplant-related mortality [relative risk (RR) = 0.60, P = 0.033] and treatment failure (inverse of leukaemia-free survival, RR = 0.69, P = 0.032). Among PBSC recipients, high CD34 dose was associated with faster recovery of neutrophils to > 0.5 x 109/l (RR = 1.38, P < 0.001) and platelets to > 20 x 109/l (RR = 1.34, P = 0.003), lower risk of relapse (RR = 0.62, P = 0.029) and treatment failure (RR = 0.74, P = 0.03). We conclude that higher CD34 cell doses decrease treatment failure in recipients of HLA-identical sibling BM and PBSC transplants.

Acute Disease↗

PubMed2003

Extensions and applications of the Cox-Aalen survival model.

Cox's regression model is the standard regression tool for survival analysis in most applications. Often, however, the model only provides a rough summary of the effect of some covariates. Therefore, if the aim is to give a detailed description of covariate effects and to consequently calculate predicted probabilities, more flexible models are needed. In another article, Scheike and Zhang (2002, Scandinavian Journal of Statistics 29, 75-88), we suggested a flexible extension of Cox's regression model, which aimed at extending the Cox model only for those covariates where additional flexibility are needed. One important advantage of the suggested approach is that even though covariates are allowed a nonparametric effect, the hassle and difficulty of finding smoothing parameters are not needed. We show how the extended model also leads to simple formulae for predicted probabilities and their standard errors, for example, in the competing risk framework.

Biometry↗