Graft-Versus-Host Disease

New clinical grading system for chronic GVHD predicts duration of systemic immunosuppressive treatment and GVHD-specific and overall survival


We investigated outcomes according to a new clinical grading system for chronic graft-versus-host disease (chronic GVHD) in 38 patients who developed chronic GVHD after an allogeneic hematopoietic stem cell transplantation. We categorized the patients into three grade groups, namely, grade I, grade II and grade III, according to the presence of three risk factors: extensive skin involvement, thrombocytopenia (TP) and progressive type of onset. Sixteen patients were classified into grade 1, 19 into grade II and three into grade III. The probability of withdrawal of systemic immunosuppression (IST) at 1, 2 and 3 years was 61, 76 and 87%, respectively. Patients with grades 2 or 3 chronic GVHD had prolonged duration of systemic IST compared to grade 1 (P=0.043). The probability of GVHD-specific survival (GSS) at 5 years was 52%. Twenty-two of 38 patients with chronic GVHD were still alive and the estimated 3-year overall survival (OS) rate was 60%, whereas that for the group with chronic GVHD grade I and grade II+III was 64 and 48% (P<0.05). Multivariate analysis showed that prior occurrence of acute GVHD, chronic GVHD grade, serum bilirubin over 1.5 mg/dl, date of diagnosis of chronic GVHD (<day 150 versus >day 150) and transplantation-risk factor were independent prognostic factors for GSS and OS.


Chronic graft-versus-host disease (chronic GVHD) represents one of the major complications of allogeneic hematopoietic cell transplantation (HCT).1, 2 Despite enormous efforts to decrease the incidence and severity of acute GVHD, chronic GVHD has emerged as an increasingly frequent complication of allogeneic transplantation (BMT). Patients with chronic GVHD have decreased performance status, impaired quality of life and increased risk of mortality.3, 4 Chronic GVHD may compromise a patient's quality of life (QOL) due to chronic immunosuppressive therapy (IST) and the symptoms of GVHD itself.5 Chronic GVHD appears to be one of the major determinants in the survival and QOL of patients after allogeneic HCT. According to the study by Akpek et al.,6 extensive skin involvement (ESI), thrombocytopenia (TP)7 (<100 × 109/l) and progressive type onset8 (PTO) were identified as significant predictors of poor prognosis in patients with chronic GVHD. Moreover, their clinical grading system has the potential to improve clinical management by adjusting therapy, improving the grouping of patients for clinical trials. Akpek et al.6 developed a prognostic model for predicting chronic GVHD-specific survival (GSS), in which deaths solely due to relapse were not considered an event. Considering that a proportion of the patients with chronic GVHD die of their underlying disease, GSS appears to be more appropriate as an end point than is overall survival (OS) for assessing the impact of chronic GVHD on survival. Many patients with chronic GVHD receive more than one regimen of systemic IST for a prolonged period, and systemic IST usually causes discomfort and has side effects. Therefore, the duration of systemic IST may have an important influence upon the patient's QOL, but there have been few reports regarding this matter.

We retrospectively investigated the duration of systemic IST, GSS and OS in 38 patients who underwent allogeneic hematopoietic stem cell transplantation to validate Akpek's clinical grading system for chronic GVHD.

Patients and methods


Between April 1996 and June 2003, 133 patients underwent allogeneic HCT at the Samsung Medical Center, Seoul, Korea. Thirty-four out of 125 patients who underwent allogeneic BMT and four out of eight patients who underwent allogeneic peripheral blood transplantation suffered chronic GVHD. We analyzed the medical data of those patients. Human leukocyte antigen (HLA)-matching for donor selection was based on serological typing for HLA-A, B and C antigens, and on DNA typing for HLA-DRB1 antigen. Clinical characteristics and details of HCT are given in Table 1. Median age (25 males and 13 females) was 36 years. The underlying diseases were acute myelogenous leukemia in 12 patients, acute lymphoblastic leukemia (ALL) in three patients, acute biphenotype leukemia in three patients, severe aplastic anemia in four patients, chronic myelogenous leukemia (CML) in 14 patients and myelodysplastic syndrome (MDS) in two patients. At the time of HCT, nine patients had high-risk disease; seven acute leukemia and two MDS. Nineteen patients received allogeneic hematopoietic cell grafts from HLA-identical siblings, two from HLA-mismatched siblings, 14 from HLA-phenotypically identical unrelated donors (one; umbilical cord blood HCT) and three from HLA-mismatched unrelated donors. Thirteen of 38 patients (34.2%) had acute GVHD before the onset of chronic GVHD.

Table 1 Patients and transplant characteristics before onset of chronic GVHD

Transplant procedure

The source of stem cells was bone marrow in 33 patients, umbilical cord blood in one and peripheral blood in four patients. Peripheral blood stem cells were collected from HLA-matched donors (siblings, three; unrelated, one) mobilized with granulocyte-colony-stimulating factor (G-CSF). All patients received myeloablative HCT (Table 1). Sixteen patients were conditioned with busulfan 3.2 mg/kg/day intravenously on days −9 to −6 or on days −7 to −4 and cyclophosphamide (CY) 60 mg/kg/day intravenously on days −5 to −2 or days −2 to −1), nine patients with CY 60 mg/kg/day intravenously on days −7 to −6 and total body irradiation (TBI) 300 cGy/day on days −5 to −2), five patients with a voluntary counseling and testing for HIV (etoposide (VP-16) 1.8gm/m2 intravenously on days −6 to −3, CY 60 mg/kg/day intravenously on days −6 to −4 and TBI 300 cGy on days −3 to −1), four patients with a CY/ATG (anti-thymocyte globulin) (CY 50 mg/kg/day intravenously on days −5 to −2 and ATG 30 mg/kg/day intravenously on days −5 to −3) and others.

None of the hematopoietic cell grafts were T-cell depleted. All patients received prophylaxis for GVHD with methotrexate (MTX)+cyclosporin A (CsA)+steroid (n=18), MTX+CsA (n=16) and others (n=4). MTX was given intravenously at a dose of 15 mg/m2 on day 1 and 10 mg/m2 on days 3, 6 and 11. The last dose of MTX was omitted for patients with severe mucositis or impaired renal function. Cyclosporine (CSA) 1.5 mg/m2 was administered intravenously every 12 h starting on day −1 and switched to an oral dose when oral intake became tolerable. CSA levels were monitored weekly, with reference blood levels of 200–300 ng/ml. The CSA dose was reduced by 50% when the serum creatinine level exceeded twice the baseline. CSA was discontinued if the serum creatinine level exceeded three times the baseline. All patients received a fixed dose of G-CSF 300 μg intravenously once daily starting 24 h after the last MTX dose (day 11) and given until the peripheral blood absolute neutrophil count exceeded 5000/μl.

Diagnosis and grading of chronic GVHD

If the mode of presentation for chronic GVHD was progressive, the date of onset was defined as day 100. The diagnosis of chronic GVHD was confirmed on the basis of clinical symptoms, laboratory tests and, whenever possible, histopathological findings of skin, oral mucosa and gastrointestinal tract.9 If clinical manifestations, laboratory and radiologic findings were sufficient to allow a clinical diagnosis of chronic GVHD in specific organs, the diagnosis was made without histological confirmation. The grading of chronic GVHD was made according to the new chronic GVHD grading system of Akpek et al. A prognostic score (PS) for all patients was calculated based on three risk factors, including ESI, TP and PTO of chronic GVHD.10, 11, 12 The formula for the calculation of PS is PS=(1.949 × (skin extent))+(1.293 × (platelet))+(0.514 × (type of onset)). The following conditions apply: (1) if the extent of skin involvement is more than 50% of BSA, allocate 1, otherwise 0; (2) if the platelet count is less than 100 000 cells/μl, allocate 1, otherwise 0; and (3) if the chronic GVHD is PTO, allocate 1, otherwise 0. We categorized the patients into three grades: grade 1 (0 RF), grade 2 (1 risk factor (RF) or TP+PTO) and grade 3 (2–3 RFs).

Treatment of chronic GVHD

Patients with chronic GVHD were treated according to the new chronic GVHD grading system. All the patients were treated with systemic IST. Patients were treated with glucocorticoid based immunosuppressive regimens according to the grade of chronic GVHD. Nineteen patients were treated with CSA plus glucocorticoid and six were treated with CSA alone. Eight were treated with tacrolimus (FK506) plus glucocorticoid, one received mycophenolate mofetil (MMF) in addition to FK506 plus glucocorticoid and two were treated with FK506 alone. One patient received ATG in addition to FK506. One received azathioprine and ATG with glucocorticoid. All patients received prophylactic trimethoprim-sulfamethoxazole for pneumocystis carinii pneumonia.

Prognostic factors and statistical analysis

We analyzed prognostic factors for survival including age (<40 years versus >40 years), sex, underlying diagnosis, time interval between diagnosis of underlying disease and HCT (<120 versus >120 days), disease status of HCT (standard risk versus high risk), donor–recipient sex pair (female to male versus others), prior acute GVHD, grade of chronic GVHD, date of diagnosis of chronic GVHD (<day 150 versus >day 150), serum alkaline phosphatase (<120 versus >120 IU/l), serum total bilirubin (<1.5 versus >1.5 mg/dl) and conditioning regimens. Standard-risk diseases were defined as hematological malignancies including acute leukemia or aggressive lymphoma of first complete remission (CR) or chronic phase CML, whereas high-risk diseases were defined as acute leukemia in more than first CR, Philadelphia+ALL, advanced phase CML, primary refractory or multiply relapsed malignancies.

We defined GSS as the time from date of diagnosis of chronic GVHD to death directly related to chronic GVHD. Deaths unrelated to chronic GVHD were censored in GSS. Duration of systemic IST was time from the start to withdrawal of systemic IST. OS was defined as duration from the date of transplantation to death.

We used a multivariate analysis to identify significant prognostic factors for the duration of systemic IST, GSS and OS (Table 2). A cutoff P-value of 0.05 was adopted for all statistical analyses. Statistical data were obtained using an SPSS software package (SPSS 11.0).

Table 2 Multivariable analysis of factors related to the duration of IST, GSS and OS


Chronic GVHD

Median time from diagnosis to HCT was 155 days (range, 22–2557). The median onset of chronic GVHD was 186.5 days (range, 100–635). In approximately two-thirds of patients, chronic GVHD developed de novo without prior acute GVHD. Grade I chronic GVHD appeared in 16 patients, grade II in 19 and grade III in three. Extensive skin involvement was observed in eight patients, TP (<100 000/μl) in 26 and progressive type of onset in six. Skin and oral mucosa were the organs most commonly involved. Baseline characteristics at the onset of chronic GVHD before initiating systemic IST are given in Table 3.

Table 3 Patients characteristics at initial diagnosis of chronic GVHD

Immunosuppressive treatment

At the time of analysis, 25 (65.7%) of the 38 patients had discontinued systemic IST. The median duration of systemic IST for chronic GVHD was 270 days (range 16–1714). Probability of withdrawing systemic IST at 1, 2 and 3 years was 61, 76 and 87%, respectively (Figure 1). Analysis based on a multivariate model showed that chronic GVHD grade (grade I versus grades II and III) was the only independent predictive factor for the duration of IST (P=0.043).

Figure 1

Probability of withdrawal from systemic IST.

GSS and OS

Sixteen patients died. Fifteen patients died of chronic GVHD-associated causes including infectious complications and multiple organ failure. Of those 15 chronic GVHD-associated deaths, eight occurred within the first 100 days after the onset of chronic GVHD. The cause of death in the remaining one patient was relapse of underlying disease. The probability of GSS at 5 years was 52% (Figure 2). Analysis based on a multivariate model showed that prior occurrence of acute GVHD (P<0.001), chronic GVHD grade (P=0.004), serum bilirubin over 1.5 mg/dl (P=0.049), date of diagnosis of chronic GVHD (<day 150 versus >day 150; P=0.019) and transplantation risk factor (P=0.003) were independent prognostic factors for GSS. Twenty-two of 38 patients with chronic GVHD were still alive and the estimated 3-year OS rate for the whole chronic GVHD patients was 60%, whereas that for the group with chronic GVHD grade I and grade II+III was 64 and 48%, respectively (Figure 3). Analysis based on a multivariate model showed that new chronic GVHD grade (P=0.005), prior occurrence of acute GVHD (P<0.001), transplantation risk factor (P=0.024), serum bilirubin over 1.5 mg/dl (P=0.018) and date of diagnosis of chronic GVHD (<day 150 versus >day 150; P=0.004) were independent prognostic factors in OS (Table 2).

Figure 2

Chronic GVHD-specific survival.

Figure 3

Overall survival.


Chronic GVHD may compromise QOL due to chronic IST and the symptoms of GVHD itself.1, 2 Intensity and duration of IST depend on the severity of chronic GVHD.13 An exact evaluation of the severity of chronic GVHD is essential to ensure the best treatment of chronic GVHD. Akpek et al.6 developed a new prognostic scoring system based on three risk factors (RF), including ESI, TP,7 and onset type of chronic GVHD,8 which allows clinicians to identify diversity in outcome within extensive-stage chronic GVHD.10, 11, 12 According to the new grading system, compared with the classical Seattle system9 in terms of defining the severity of chronic GVHD in our study, 24 patients in the group with limited GVHD were classified as grade I (n=10), grade II (n=11) and grade III (n=3). Fourteen patients with extensive GVHD were classified as grade I (n=6), grade II (n=8) and grade III (n=0). There was no significant correlation between the classical Seattle system9 and the new grading system of the severity of chronic GVHD12 (P=0.354). In the multivariable model of duration of IST, chronic GSS and OS, the classical Seattle system was not a significant prognostic factor contrary to the new grading system. Chronic GVHD has been reported to be associated with poor survival despite its graft-versus-leukemic effect. Leukemia relapse rate is lower in patients with chronic GVHD, regardless of its severity, compared to patients without chronic GVHD.4, 7, 8, 11 Treatment-related mortality, however, is higher in patients with severe chronic GVHD. It is essential to stratify patients with chronic GVHD into different risk groups for GVHD-related mortality. From this point of view, it is appropriate to use the new grading system in evaluating the severity of chronic GVHD.

Our study analyzed risk factors associated with the duration of IST, GSS and OS among patients with chronic GVHD. Duration of IST was only significantly associated with the difference between grade I or II and III. The new chronic GVHD grade, a prior history of acute GVHD,14 risk of transplantation, serum bilirubin level more than 1.5 mg/dl at diagnosis and date of diagnosis of chronic GVHD (<day 150 versus >day 150)15, 16, 17 were significantly associated with both GSS and OS.

Several reports have shown that the use of peripheral blood as a source of hematopoietic stem cells increases the incidence of chronic GVHD.18 However, stem cell source did not affect the duration of systemic IST, GSS and OS in this study. Alloimmunization of the donor increases the risk of developing chronic GVHD according to donor parity.19 In general, male donors are superior to female donors, because female to male transplantation has been associated with decreased survival in some studies.15, 16, 17 However, our study showed that donor parity was not associated with the duration of systemic IST, GSS and OS. Underlying disease, degree of HLA mismatch, nature of the conditioning regimen and whether the donor was sibling or not were not related to the duration of systemic IST, GSS and OS.

Our results may inform physicians and patients of clinically relevant factors about chronic GVHD affecting the duration of IST, GSS and OS. We did not assess the effect of the type of IST appropriately because the various ISTs used in our study were too diverse. The diversity of IST therapy related to the retrospective nature and lack of randomization in this study. We propose using the new grading system to evaluate the severity of chronic GVHD and risk factors according to the nature of immunosuppression used.

Recently, the National Institute of Health (NIH) presented the criteria for diagnosing and staging chronic GVHD. These criteria are similar to the grading system used in our study, but are more standardized and objective. Patients with progressive chronic GVHD in our study have persistent acute GVHD or overlap syndrome in the NIH categorization.

In conclusion, this study suggests that the new grading system for chronic GVHD based on ESI, TP and PTO could be useful in predicting duration of IST, GSS and OS. These results have possibly been influenced by the retrospective nature of the study, the limited number of patients and lack of randomization. A multi-hospital prospective study is warranted.


  1. 1

    Atkinson K . Chronic graft-versus-host disease. Bone Marrow Transplant 1990; 5: 69–82.

  2. 2

    Vogelsang GB . How I treat chronic graft-versus-host disease. Blood 2001; 97: 1196–1201.

  3. 3

    Duell T, van Lint MT, Ljungman P, Tichelli A, Socie G, Apperley JF et al. Health and functional status of long-term survivors of bone marrow transplantation. EBMT Working Party on Late Effects and EULEP Study Group on Late Effects. European Group for Blood and Marrow Transplantation. Ann Intern Med 1997; 126: 184–192.

  4. 4

    Socie G, Stone JV, Wingard JR, Weisdorf D, Henslee-Downey PJ, Bredeson C et al. Long-term survival and late deaths after allogeneic bone marrow transplantation. Late Effects Working Committee of the International Bone Marrow Transplant Registry. N Engl J Med 1999; 341: 14–21.

  5. 5

    Sutherland HJ, Fyles GM, Adams G, Hao Y, Lipton JH, Minden MD et al. Quality of life following bone marrow transplantation: a comparison of patient reports with population norms. Bone Marrow Transplant 1997; 19: 1129–1136.

  6. 6

    Akpek G, Zahurak ML, Piantadosi S, Margolis J, Doherty J, Davidson R et al. Development of a prognostic model for grading chronic graft-versus-host disease. Blood 2001; 97: 1219–1226.

  7. 7

    Sullivan KM, Witherspoon RP, Storb R, Deeg HJ, Dahlberg S, Sanders JE et al. Alternating-day cyclosporine and prednisone for treatment of high-risk chronic graft-v-host disease. Blood 1988; 72: 555–561.

  8. 8

    Wingard JR, Piantadosi S, Vogelsang GB, Farmer ER, Jabs DA, Levin LS et al. Predictors of death from chronic graft-versus-host disease after bone marrow transplantation. Blood 1989; 74: 1428–1435.

  9. 9

    Shulman HM, Sullivan KM, Weiden PL, McDonald GB, Striker GE, Sale GE 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.

  10. 10

    Arora M, Burns LJ, Davies SM, Macmillan ML, Defor TE, Miller WJ et al. Chronic graft-versus-host disease: a prospective cohort study. Biol Blood Marrow Transplant 2003; 9: 38–45.

  11. 11

    Lee SJ, Klein JP, Barrett AJ, Ringden O, Antin JH, Cahn JY et al. Severity of chronic graft-versus-host disease: association with treatment-related mortality and relapse. Blood 2002; 100: 406–414.

  12. 12

    Akpek G, Lee SJ, Flowers ME, Pavletic SZ, Arora M, Lee S et al. Performance of a new clinical grading system for chronic graft-versus-host disease: a multicenter study. Blood 2003; 102: 802–809.

  13. 13

    Stewart BL, Storer B, Storek J, Deeg HJ, Storb R, Hansen JA et al. Duration of immunosuppressive treatment for chronic graft-versus-host disease. Blood 2004; 104: 3501–3506.

  14. 14

    Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J et al. 1994 Consensus Conference on Acute GVHD Grading. Bone Marrow Transplant 1995; 15: 825–828.

  15. 15

    Kondo M, Kojima S, Horibe K, Kato K, Matsuyama T . Risk factors for chronic graft-versus-host disease after allogeneic stem cell transplantation in children. Bone Marrow Transplant 2001; 27: 727–730.

  16. 16

    Ochs LA, Miller WJ, Filipovich AH, Haake RJ, McGlave PB, Blazar BR et al. Predictive factors for chronic graft-versus-host disease after histocompatible sibling donor bone marrow transplantation. Bone Marrow Transplant 1994; 13: 455–460.

  17. 17

    Randolph SS, Gooley TA, Warren EH, Appelbaum FR, Riddell SR . Female donors contribute to a selective graft-versus-leukemia effect in male recipients of HLA-matched, related hematopoietic stem cell transplants. Blood 2004; 103: 347–352.

  18. 18

    Cutler C, Giri S, Jeyapalan S, Paniagua D, Viswanathan A, Antin JH . Acute and chronic graft-versus-host disease after allogeneic peripheral-blood stem-cell and bone marrow transplantation: a meta-analysis. J Clin Oncol 2001; 19: 3685–3691.

  19. 19

    Flowers ME, Parker PM, Johnston LJ, Matos AV, Storer B, Bensinger WI et al. Comparison of chronic graft-versus-host disease after transplantation of peripheral blood stem cells versus bone marrow in allogeneic recipients: long-term follow-up of a randomized trial. Blood 2002; 100: 415–419.

Download references

Author information

Correspondence to C W Jung.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kim, S., Jung, C., Lee, J. et al. New clinical grading system for chronic GVHD predicts duration of systemic immunosuppressive treatment and GVHD-specific and overall survival. Bone Marrow Transplant 39, 711–716 (2007) doi:10.1038/sj.bmt.1705661

Download citation


  • chronic GVHD
  • survival
  • immunosuppressive treatment
  • clinical grading

Further reading