Chronic graft-versus-host disease: A prospective cohort study☆

Article Outline

• Abstract
• Introduction
• Methods
  • Eligibility criteria
  • Treatment plan
    • Therapy for symptomatic CGVHD
    • Measurement of response
• Statistical analysis
  • Response to therapy
    • Predictors of response
  • Survival
    • Predictors of mortality
  • Complications after transplantation
    • Infections following BMT
• Results
  • Response to therapy
  • Predictors of response
  • Survival
  • Predictors of mortality
  • Prevalence of CGVHD
  • Complications of CGVHD
  • Infections
  • Duration of treatment and outcome
• Discussion
• References
• Copyright

Abstract 

Chronic graft-versus-host disease (CGVHD) is a major cause of morbidity and mortality following allogeneic bone marrow transplantation (BMT). We studied 159 patients with CGVHD longitudinally to characterize the natural history of CGVHD and identify reliable predictors of response and long-term mortality. Rates of response to treatment were 61%, 53%, and 50% at 6 months, 1 year, and 2 years, respectively. A high incidence of infections (7 of 1000 patient-days at 0 to 6 months, 2.5 of 1000 patient-days at 6 months to 1 year, and 0.6 of 1000 patient-days at 1 to 2 years) was observed. After a median follow-up of 8.4 years, an overall survival rate of 40% was observed. The overall survival rate was 63% (95% confidence interval [CI], 56%-71%) at 1 year, 51% (95% CI, 43%-59%) at 2 years, and 39% (95% CI, 31%-47%) at 10 years. In multivariate analysis, age older than 20 years (RR = 1.5; 95% CI, 0.9%-2.5%; P = .09), progressive onset of CGVHD (RR = 1.6; 95% CI, 1.0%-2.4%; P = .04), platelet count of <100,000/μL (RR = 2.1; 95% CI, 1.3%-3.4%; P = .001), and GI involvement (RR = 1.5; 95% CI, 1.0%-2.4%; P = .05) were associated with increased mortality. Among patients surviving more than 6 months, no response (RR = 4.5; 95% CI, 1.9%-10.5%; P = .0006) and partial response (RR = 2.5; 95% CI, 1.1%-6.1%; P = .04) to treatment at 6 months also were significant predictors of mortality. The prevalence of active CGVHD was 33% at 2 years. However, the cumulative incidence of successful discontinuation of therapy was only 13% at 2 years. Among patients with clinical resolution of CGVHD, only 18% were off immunosuppressive therapy by 2 years, and 89% by 4 years. Despite high initial response rates, a large majority of patients had active disease requiring prolonged immunosuppression. This requires improved infection prevention for a longer time. Recognition of a high-risk group should facilitate assignment of more intensified regimens. Better treatment regimens need to be identified to improve survival and limit toxicity of prolonged immunosuppression. © 2003 American Society for Blood and Marrow Transplantation

Biology of Blood and Marrow Transplantation 9:38-45 (2003)

Keywords:  Chronic graft-versus-host disease (CGVHD), Cohort study, Bone marrow transplantation

Back to Article Outline

Introduction 

Chronic graft-versus-host disease (CGVHD) is a late complication of bone marrow transplantation (BMT) characterized by a connective tissue–like disease usually, but not always, occurring more than 100 days after blood stem cell or bone marrow transplantation. Clinically, CGVHD may be manifest by involvement of the skin, oral mucosa, gastrointestinal (GI) tract, eyes, liver, lungs, and joints ^[1]. CGVHD develops in 30% to 60% of transplantation survivors ², ³ and remains a major cause of morbidity and mortality after allogeneic BMT. The major cause of death in patients with CGVHD is infection due to continuing immunodeficiency ¹, ⁴. Poor prognostic factors that have been identified include thrombocytopenia (platelet count < 100,000/μL), increased age, lichenoid skin pathology, liver involvement, and progressive presentation of CGVHD ⁵, ⁶ (CGVHD developing before resolution of acute GVHD).

The survival rate in patients with CGVHD remains poor. We conducted a prospective cohort study of patients with CGVHD to characterize the clinical presentation and course of the disease and identify groups at high risk for development of complications (infections, poor response to treatment, or mortality). We report their long-term follow-up, prevalence of CGVHD, and prolonged duration of immunosuppressive therapy.

Back to Article Outline

Methods 

Eligibility criteria 

Patients who underwent allogeneic BMT at the University of Minnesota between November 1987 and August 1993 and who developed CGVHD were eligible for inclusion in the study. The cohort includes 159 patients with symptomatic CGVHD. CGVHD was classified as single or advanced with multi-organ disease. The diagnosis and organ involvement with CGVHD was based on standard clinical and/or histological criteria. GI and liver involvement with CGVHD were defined based on symptomatology and positive laboratory tests or positive biopsy results (positive malabsorption tests or increased liver enzymes). In all cases, obstruction, infections, and drugs were ruled out as possible causes.

Treatment plan 

Limited skin involvement was treated with topical steroids (0.1% triamcinolone cream to body, 1% hydrocortisone cream to face) 3 times a day. Patients with only hepatic involvement (bilirubin < 3 mg/dL) did not receive therapy but were closely observed for disease progression.

Therapy for symptomatic CGVHD 

All patients (n = 159) with more extensive CGVHD resulting in clinical symptoms received initial therapy with high-dose methylprednisone given at a dosage of 15 mg/kg as an intravenous (IV) injection weekly for 8 weeks.

Patients enrolled in the initial cohort between November 1987 and November 1988 (n = 22) also received treatment with prednisone at a dosage of 0.5 mg/kg orally every other day and azathioprine at a dosage of 1 mg/kg orally each day. If patients developed an absolute neutrophil count (ANC) of 1000 to 1500/μL and/or a platelet count of 50,000 to 80,000/μL the dosage of azathioprine was reduced to 0.5 mg/kg/d whereas azathioprine was withheld for ANC of <1000/μL and platelets of <50,000/μL.

Patients enrolled in the cohort from November 1988 to August 1993 (n = 137) were to receive cyclosporine (CSA) in addition to azathioprine and prednisone as above. CSA was started at 6.25 mg/kg orally twice daily (or 1.5 mg/kg IV twice daily) with the dose modified to maintain trough levels > 200 ng/mL. However, due to tolerance and clinical toxicities, a total of 55 patients received prednisone and azathioprine, 57 patients received prednisone alone, 16 patients received CSA in addition to prednisone and azathioprine, 25 patients were treated with prednisone and CSA, 3 patients received prednisone and thalidomide, and 3 patients received other agents. In patients with no response to the initial 8 weeks of therapy or progression after initial stabilization or response, methylprednisone was given at a dosage of 15 mg/kg/d IV for 5 days, and oral prednisone was increased to 1 mg/kg/d for 6 weeks, then tapered over a 3-month period to 0.5 mg/kg/d on alternate days. Therapy was continued until 9 months following the last clinical evidence of active CGVHD followed by a taper over 2 to 3 months.

Measurement of response 

Patients returned for evaluation at the University of Minnesota at 6 months, 1 year, and 2 years after enrollment into the study and approximately yearly thereafter. Because patients did not always return to the transplantation center on the scheduled dates, window periods were used to maximize the completeness of evaluation. For the 6-month visit, the evaluation window period was between 3 and 9 months from enrollment. For the 1- and 2-year visits, window periods of 9 to 15 months and 21 to 27 months, respectively, were used. All patients included in the study were enrolled at least 6 months prior to study termination. The median follow-up of surviving patients was 8.4 years (range, 0.7-13.4 years).

Response to therapy was graded as complete response (CR) defined as resolution of all signs and symptoms of CGVHD. Partial response (PR) was defined as improvement in 1 or more organs of involvement and no evidence of worsening in any organ. Flare included patients with PR or CR followed by worsening of CGVHD that was less severe than at the baseline evaluation. No response was defined as either progression of CGVHD to worse than at baseline evaluation, or no improvement in CGVHD after 6 months of therapy. Prevalence of CGVHD was defined as proportion of all patients with active CGVHD among surviving patients. Improvement or worsening of disease were determined through both subjective and objective criteria. Subjective criteria were symptomatic change in cough, dyspnea, anorexia, nausea, vomiting, diarrhea, arthralgia, or dry eyes. Objective criteria included physical examination of skin, oral mucosa, weight change and/or liver function tests, pulmonary function tests, Schirmer's test, biopsies, and radiological studies. All patients who died of complications related to CGVHD (infections or multi-organ dysfunction) were considered at subsequent time points as non-responders.

Back to Article Outline

Statistical analysis 

The study was designed as a prospective cohort study; 159 patients with CGVHD were enrolled and followed longitudinally. A database was created to include all patients with CGVHD and variables such as baseline demographic data, type of onset of CGVHD, dates of diagnosis and treatment, organ involvement at diagnosis, complications, and causes of death/relapse. Data collection and entry was carried out by trained personnel. This was supplemented by chart reviews for specific organ involvement, complications, and continuation of therapy at 6 months, 1 year, and 2 years from diagnosis. Grading of response was performed without knowledge of the subsequent survival status. Statistical analysis of response was performed before analysis of survival to maintain blinding.

Response to therapy 

Response to therapy was assessed at 6 months, 1 year, and 2 years.

Predictors of response 

Eleven potential predictors were evaluated. These included recipient age at transplantation, gender of recipient and donor, cytomegalovirus (CMV) serological status of recipient and donor, type of transplant (allogeneic sibling versus unrelated donor), HLA mismatch, prior presence and clinical grade of acute GVHD, onset of CGVHD (de novo, progressive, or quiescent), organ involvement with CGVHD (eyes, mouth, skin, lungs, GI, and liver), platelet count, serum bilirubin level, and alanine aminotransferase (ALT) level at baseline. Pearson's chi-square test was used in the univariate analyses to compare the proportion of subjects with response to therapy within each category of potential predictors. Multivariate logistic regression was used to evaluate the independent effect of study variables on treatment response. A stepwise regression with forward selection was used. A significance level of .1 was used to enter into the model and .15 to stay in the model.

Survival 

Patient survival was determined using the Kaplan-Meier ^[7] estimation with 95% confidence intervals (CIs) derived from standard errors. Patients were censored at the date of last contact. Survival was estimated from onset of CGVHD.

Predictors of mortality 

Potential factors associated with mortality were studied. The Kaplan-Meier product limit method ^[7] and log-rank testing was used to compare survival in the subsets and the Cox regression model ^[8] was used to assess the independent effect of the predictors on survival as well as any potential confounding on the effect of treatment. A stepwise regression with forward selection was used. A significance level of .1 was used to enter into the model and .15 to stay in the model. Acute GVHD (grade III and IV) was excluded from the model because of colinearity between progressive onset and acute GVHD. Cumulative incidence rates were used to estimate the probability of discontinuation of all immunosuppressive treatment ^[9] over the complete period of follow-up.

Complications after transplantation 

Complications studied included hypertension, hyperglycemia requiring treatment, seizures, thrombotic thrombocytopenic purpura, avascular necrosis, and infections.

Infections following BMT 

Clinically significant infections with microbiological confirmation or requiring systemic antibiotic therapy or hospitalization were tabulated during the first 2 years of therapy. To account for multiple events, density incidence was used to describe the total rate of infections. It was defined as the total number of infections per 1000 patient-days of observation.

Back to Article Outline

Results 

During the 4.9 years of patient enrollment, the overall incidence of CGVHD among patients undergoing allogeneic BMT was 44%. Table 1 lists the characteristics of the patients with CGVHD.

Table 1. Clinical characteristics at study entry

ATG indicates antithymocyte globulin.

The median age at transplantation was 30 years. Of the recipients, 33% received a transplant from an unrelated donor. Progressive onset of disease was seen in 45% of patients. The median platelet count at study entry was 72,000/μL and 30% of patients had liver involvement at diagnosis.

Response to therapy 

As shown in Tables 2 and 3, the response to therapy was evaluated at 6 months, 1 year, and 2 years from diagnosis.

Table 2. Patient evaluability over time

NE indicates not evaluable (secondary to relapsed disease or insufficient data within defined time windows for evaluation).

Note. All patients who died of causes other than relapse were evaluable for response.

Table 3. Frequency of response

NR indicates no response.

Note. Patients who died of causes other than relapse were considered nonresponders.

Of the 159 patients enrolled, the number of patients evaluable for response was 155, 148, and 141 at 6 months, 1 year, and 2 years, respectively. Twelve patients (8%) relapsed and died within 2 years and did not have response assessed. Of these, 3 patients had relapsed by 6 months, 8 patients had relapsed by 1 year, and 12 patients had relapsed by 2 years. One, 3, and 6 patients had evaluations performed only outside the defined window periods for evaluation at 6 months, 1 year, and 2 years, respectively. Complete plus partial responses were observed in 61%, 53%, and 50% of patients at 6 months, 1 year, and 2 years, respectively. Complete response was seen in 30% of patients by 1 year and 39% of patients by 2 years. Only 12% of responders at 6 months experienced a flare of the disease by 1 year.

Predictors of response 

We analyzed 11 clinical factors as potential predictors of response at 2 years. In univariate analysis, responses were more frequent in patients with de novo/quiescent onset of CGVHD (60% versus 39%; P = .01) and platelet count of ≥100,000/μL (64% versus 41%; P = .01). In the multivariate analysis, de novo/quiescent onset of disease (odds ratio [OR] 2.3; P = .03) and platelet count of ≥100,000/μL (OR 2; P = .06) remained significant predictors of more frequent response. In addition, age of 20 years or older (OR 2.1; P = .08) and absence of GI involvement with CGVHD (OR 1.9; P = .08) also were independent predictors of more frequent response with marginal significance (Table 4).

Table 4. Predictors of response: Multivariate analysis

Note. Logistic regression showing the independent OR and 95% CI favoring CR or PR at 2 years.

Recipients of transplants from unrelated donors versus related donors had similar responses.

Survival 

After a median follow-up of 8.4 years (range, 0.7-13.4 years), an overall survival rate of 40% was observed. The survival rate was 63% (95% CI, 56%-71%) at 1 year, 51% (95% CI, 43%-59%) at 2 years, 40% (95% CI, 32%-48%) at 8 years, and 39% (95% CI, 31%-47%) at 10 years (Figure 1).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Overall and relapse-free survival after onset of CGVHD estimated using Kaplan-Meier method.

Overall, 95 patients died. Most deaths (86%) were observed within 2 years of onset, but deaths continued to occur up to 8.6 years after diagnosis. Nearly all (86%) patients died of complications related to CGVHD (Table 5).

Table 5. Causes of non-relapse mortality

Note. Other complications included causes of death unrelated to transplant or CGVHD. Infection includes infectious etiology as a cause of death in patients without active CGVHD who were off all immunosuppressive therapy.

Infections accounted for 63% of deaths in patients with active CGVHD (50% due to bacterial infections, 35% due to viral infections, and 15% due to fungal infections) and 9% of deaths in patients without active disease. (Patients with clinical resolution of CGVHD who were off all immunosuppression.) Two patients died of a second malignancy and 2 patients died of complications unrelated to transplantation. Seventeen patients relapsed and died (13 within 2 years of CGVHD). Thirteen late deaths were observed (beyond 2 years from diagnosis). Of these, 11 patients died of complications related to CGVHD. Two of these 11 patients died of a second malignancy.

Predictors of mortality 

In univariate analysis, several factors were associated with poor survival. Progressive onset of CGVHD (P = .01), platelet count of <100,000/μL (P < .01), involvement of GI tract with CGVHD (P < .01), bilirubin of ≥3 mg/dL, and no CR or PR at 6 months (P < .01) were each significant predictors of poor survival. Patients with any combination of 3 or more of these high-risk factors had a 10-year survival rate of only 11%, whereas patients with 0 or 1 risk factor had a 10-year survival rate of 72% and 52%, respectively (Figure 2).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 2. 

  Predictors of survival: univariate analysis. Analysis by number of risk factors at onset of CGVHD is shown in panel F. Risk factors considered include those shown in the other panels, A-E, as statistically significant.

In the multivariate analysis, platelets < 100,000/μL (RR 2.1; P = .001), progressive onset of disease (RR 1.6; P = .04), and involvement of GI tract with CGVHD (RR 1.5; P = .05) were independently significant predictors of poor survival (Table 6).

Table 6. Predictors of poor survival: Multivariate analysis

Note. The relative risk (95% CI) shown is of mortality following therapy for CGVHD derived from Cox multiple regression analysis.

Older age (20 years or older) (RR 1.5; P = .09) was an important predictor with marginal statistical significance. Patients with progressive onset of disease had more frequent involvement of the GI tract (45.8 versus 31%; P = .06). Even after stratification of the analysis by onset of disease (progressive versus de novo/quiescent), GI involvement remained an independent predictor of poor survival (RR 1.5; P = .06). In a separate regression analysis (Table 7) excluding patients who died within 6 months of study entry, no response (RR 4.5; P = .0006) and partial response (RR 2.5; P = .04) at 6 months also were significant independent predictors of poor survival.

Table 7. Predictors of poor survival: Multivariate analysis (patients surviving ≥ 6 months)

Abbreviations: NR, PR, no response or partial responses at 6 month evaluation.

Note. Shown is the relative risk (95% CI) of mortality following therapy for CGVHD derived from Cox multiple regression analysis.

Prevalence of CGVHD 

The prevalence of CGVHD (persistence of active CGVHD among surviving patients) was evaluated over time. The assessment of prevalence incorporates both flares and persistence of CGVHD and as the prevalence decreases it reflects a clinical cure of CGVHD. The prevalence was 56% at 6 months, 60% at 1 year, and 33% at 2 years.

Complications of CGVHD 

Hypertension (19%) and hyperglycemia (13%) were the most frequent noninfectious complications observed among patients treated for CGVHD with immunosuppressive therapy (159 patients). Seizures, thrombotic thrombocytopenic purpura, and avascular necrosis were observed at a frequency of 8%, 9%, and 6%, respectively (Table 8).

Table 8. Complications of therapy

Abbreviations: TTP/HUS, thrombotic thrombocytopenic purpura/hemolytic uremic syndrome.

Note. Proportion of patients experiencing complications among all patients receiving immunosuppressive therapy (n = 159).

Infections 

In all patients, a high incidence density of infections (7 of 1000 patient-days at 6 months, 2.4 of 1000 patient-days at 1 year, and 0.6 of 1000 patient-days at 2 years) was observed. Bacterial infections were most frequent across all time periods (3.4 of 1000 patient-days at 6 months, 1.2 of 1000 patient-days at 1 year, and 0.3 of 1000 patient-days at 2 years) (Figure 3).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 3. 

  Infections complicating CGVHD. These panels show the incidence density of bacterial, viral, and fungal infections expressed as infections per 1000 patient-days (p-d) of follow-up alive. In panel A, the assessments of infections over the 2-year observation period are shown.

Duration of treatment and outcome 

Seventy-seven patients survived 2 years or more after diagnosis of CGVHD. At 2 years after diagnosis, 64 patients were still receiving immunosuppression. Of these, 36 patients could be tapered off medications by 3 years. Twenty-eight patients required immunosuppression for 4 or more years. The cumulative incidence of successful discontinuation of immunosuppressive treatment was 13% at 2 years, 47% at 3 years, and 75% at 4 years. From 4 to 10 years, 25% of patients were still on treatment (Figure 4).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 4. 

  Graphs showing the cumulative incidence of discontinuation of immunosuppressive therapy for CGVHD in patients surviving more than 2 years. Panels A and B show the cumulative incidence of discontinuation of therapy for all patients and for patients with CR, respectively. Immunosuppression was routinely administered for 9 months beyond the clinical resolution of CGVHD.

By 2 years, 55 patients had complete clinical resolution of CGVHD. Of these, only 10 patients (18%) were able to discontinue immunosuppression by 2 years. Forty patients (73%) discontinued therapy by 3 years and 51 patients (93%) discontinued therapy by 4 years.

Back to Article Outline

Discussion 

CGVHD remains the major cause of late morbidity and mortality following allogeneic transplantation. We postulated that examination of clinical presentation of CGVHD, response to treatment, medical complications, and long-term survival would allow clarification of the natural history of CGVHD and lead to identification of reliable predictors of response and mortality.

In this high-risk cohort, we observed good initial response rates of 61% at 6 months. Our overall response rates of 53% and 50% at 1 year and 2 years, respectively, are similar to those observed in previous studies. We observed CRs in 30% and 39% at 1 and 2 years, respectively. Overall responses of 56% and 71% were reported in high-risk (platelet count < 100,000/μL) and refractory (to prednisone with or without azathioprine) disease, respectively, by Sullivan et al. in patients receiving alternating day CSA and prednisone ^[10]. CRs were reported in 33% of high-risk patients and 19% of initially resistant patients. Sullivan et al. had reported an overall response rate of 55% and a CR rate of 30% in patients treated with prednisone and azathioprine or prednisone alone for treatment of CGVHD ^[5]. A higher response rate of 80% was observed in our randomized clinical trial of thalidomide ^[11]. The difference may reflect differences in the patient populations as well as different time periods of study. More frequent high-risk features were seen in the current study (progressive onset of CGVHD was seen in 45% of patients in the current study versus 17% in our randomized thalidomide trial; platelet count < 100,000/μL in 59% versus 37% of patients, respectively).

We evaluated responses that persisted at or beyond 2 years, to evaluate the frequency of sustained responses during tapering of immunosuppression. At 2 years we observed a high rate of sustained response of 50%. Initial treatment responses were continuing and durable in the majority of responding patients with CGVHD; flares were uncommon beyond 2 years.

Few patients died beyond the first 2 years and with a median follow-up of 8.4 years 40% survive, similar to estimates reported in other studies. Sullivan et al. reported survival estimates of 61%, 47%, and 26% after transplantation in patients receiving prednisone and placebo, prednisone and azathioprine, and prednisone alone after a median follow up of 5.6, 6, and 5.5 years, respectively ^[5]. Treatment with CSA and prednisone led to survival rates of 58% after primary therapy and 70% for salvage therapy after a median follow-up of 41 and 52 months, respectively ^[10], from the time of transplantation rather than from the onset of CGVHD. Similar to the current report, Akpek et al. ^[12] reported a survival rate of 51% at 10 years after onset of CGVHD and Wingard et al. ^[6] reported a survival rate of 52% with a median follow-up of 3.7 years after onset of CGVHD with a projected actuarial survival at 10 years of 42%.

We evaluated potential prognostic factors associated with improved response or survival. Among these, progressive onset, thrombocytopenia, and age 20 years or older have been cited as potential predictors of poor survival in other studies ⁵, ⁶, ¹¹, ¹². Akpek et al. ^[12] reported extensive skin involvement to be a significant predictor of poor survival. However, in our study, the extent of skin involvement was not evaluable and overall skin involvement was not a significant predictor of survival. Skin histology also was not evaluated as a potential predictor of survival.

In addition, we found that involvement of the GI tract was important in predicting survival. Diagnosis of GI involvement was based on clinical and/or pathological criteria. Not all patients underwent biopsies. However, other causes of similar symptomatology (drugs, infection, or mechanical) were ruled out in all patients. GI involvement was associated with progressive onset of disease (45.8% versus 31%; P = .06), a strong predictor of poor survival. However, both progressive onset and GI involvement were independent significant predictors of survival in the multivariate analysis. Even after stratification of the analysis by onset of disease, GI involvement remained an independent predictor of poor survival, suggesting that GI involvement is a powerful predictor of poor survival. Early response to therapy also was important in predicting survival. Patients with no response or PR at 6 months had a markedly worse survival than patients with early CR. If confirmed in other series, this suggests that incomplete initial response can identify a high-risk group that might benefit from intensified treatment.

A very high density incidence of infections was observed in this cohort, with most infections occurring in the first 6 months. Infections were the leading cause of mortality from CGVHD ⁴, ¹³. Beyond the choice of immunosuppression, effective prophylaxis against infection is a critical component of CGVHD management. Vogelsang suggested that in spite of completing immunosuppressive therapy, poor splenic function persists, and, for this reason, recommended long-term penicillin prophylaxis ^[14].

Although immunosuppression was routinely administered for 9 months beyond the resolution of GVHD, 87% of patients required treatment for more than 2 years with a substantial number of patients having active and continuing CGVHD beyond 4 years of treatment. Among patients with clinical resolution of CGVHD, only 18% were able to discontinue therapy by 2 years and 25% required 4 or more years of immunosuppressive therapy. These data suggest that, despite good initial response rates, a substantial number of patients had active disease requiring prolonged immunosuppression. This underscores the requirement for effective infection prophylaxis for all patients and especially this group needing long-term therapy. The noninfectious morbidity of chronic immunosuppression (impaired bone density, joint necrosis, nephrotoxicity, and second cancers) also must be accounted for in evaluating the cumulative toxicity of ongoing CGVHD.

Recognition of a high-risk group (age 20 years or older, progressive onset of disease, platelets < 100,000/μL, GI involvement, and those without a complete response after 6 months) should facilitate assignment of more intensified regimens to these patients. Better treatment regimens need to be identified to improve survival in patients with CGVHD. Newer treatment strategies should focus on a high-risk group and on those requiring prolonged immunosuppression.

Back to Article Outline

References 

1. Shulman HM, Sullivan KM, Weiden PL, et al.  Chronic graft-versus-host syndrome in man. A long term clinicopathologic study of 20 Seattle patients. Am J Med. 1980;69:204–217
   • View In Article
   • Abstract
   • Full-Text PDF (4599 KB)
   • CrossRef
2. Cole CH, Rogers PCJ, Pritchard S, Philips G, Chan KW. Thalidomide in the management of chronic graft-versus-host disease in children following bone marrow transplantation. Bone Marrow Transplant. 1994;14:937–942
   • View In Article
   • MEDLINE
3. Schiller G, Gale RP. Is there an effective therapy for chronic graft-versus-host disease. Bone Marrow Transplant. 1993;11:189–192
   • View In Article
   • MEDLINE
4. Ochs L, Shu XO, Miller J, et al.  Late infections after allogeneic bone marrow transplantation: comparison of incidence in related and unrelated donor transplant recipients. Blood. 1995;86:3979–3986
   • View In Article
   • MEDLINE
5. Sullivan KM, Witherspoon RP, Storb R, et al.  Prednisone and azathioprine compared with prednisone and placebo for treatment of chronic graft-versus-host disease: prognostic influence of prolonged thrombocytopenia after allogeneic marrow transplantation. Blood. 1988;72:546–554
   • View In Article
   • MEDLINE
6. Wingard JR, Piantadosi S, Vogelsang GB, et al.  Predictors of death from chronic graft-versus-host-disease after bone marrow transplantation. Blood. 1989;74:1428–1435
   • View In Article
   • MEDLINE
7. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:457
   • View In Article
8. Cox DR. Regression models and life-tables. J R Stat Soc (B). 1972;34:187
   • View In Article
9. Lin DY. Non parametric inference for cumulative incidence functions in competing risk studies. Stat Med. 1997;16:901–910
   • View In Article
   • MEDLINE
   • CrossRef
10. Sullivan KM, Witherspon RP, Storb R, et al.  Alternating day cyclosporin and prednisone for treatment of high-risk chronic graft-versus-host disease. Blood. 1988;72:555–561
    • View In Article
    • MEDLINE
11. Arora M, Wagner JE, Davies SM, et al.  Randomized clinical trial of thalidomide, cyclosporine and prednisone versus cyclosporine and prednisone as initial therapy for chronic graft-versus-host-disease. Biol Blood Marrow Transplant. 2001;7:265–273
    • View In Article
    • Abstract
    • Full-Text PDF (182 KB)
    • CrossRef
12. Akpek G, Zahurak ML, Piantadosi S, et al.  Development of a prognostic model for grading chronic graft-versus-host disease. Blood. 2001;97:1219–1226
    • View In Article
    • MEDLINE
    • CrossRef
13. Atkinson K, Storb R, Prentice RL, et al.  Analysis of late infections in 89 long-term survivors of bone marrow transplantation. Blood. 1979;53:720
    • View In Article
    • MEDLINE
14. Vogelsang GB. How I treat chronic graft-versus-host disease. Blood. 2001;97:1196–1201
    • View In Article
    • MEDLINE
    • CrossRef