Purpose: Malignancy relapse remains a major obstacle for successful allogeneic hematopoietic cell transplantation (HCT). Chronic graft-versus-host disease (cGVHD) is associated with fewer relapses. However, when studying effects of cGVHD on relapse, it is difficult to separate from acute GVHD effects as most cases of cGVHD occur within the first year after transplant at the time when acute GVHD is still active.

Experimental Design: This study based on CIBMTR registry data investigated cGVHD and its association with the incidence of late relapse and survival in 7,489 patients with acute myelogenous leukemia (AML), acute lymphoblastic leukemia (ALL), chronic myelogenous leukemia (CML), and myelodysplastic syndromes (MDS), who were leukemia free at 12 months after myeloablative allogeneic HCT.

Results: Forty-seven percent of the study population was diagnosed with cGVHD at 12 months after transplant. The protective effect of cGVHD on late relapse was present only in patients with CML [RR, 0.47; 95% confidence interval (CI), 0.37–0.59; P < 0.0001). cGVHD was significantly associated with higher risk of treatment-related mortality (TRM; RR, 2.43; 95% CI, 2.09–2.82; P < 0.0001) and inferior overall survival (RR, 1.56; 95% CI, 1.41–1.73; P < 0.0001) for all diseases. In patients with CML, all organ sites and presentation types of cGVHD were equally associated with lower risk of late relapse.

Conclusions: These results indicate that clinically relevant antileukemia effects of cGVHD on late relapses are present only in CML but not in AML, ALL, or MDS. Chronic GVHD in patients who are 1-year survivors after myeloablative allogeneic HCT is primarily associated with higher TRM and inferior survival. Clin Cancer Res; 21(9); 2020–8. ©2014 AACR.

See related commentary by Gill, p. 1981

Translational Relevance

Clinical evidence implicates graft-versus-host disease (GVHD) in enhancing malignancy control. We focused here on assessing chronic GVHD effects in 7,489 patients who were alive and free of disease at 1 year after allo-HCT. This study demonstrates the protective effect of chronic GVHD on late relapse only in patients with CML. A protective effect of chronic GVHD against late relapse was not seen in AML, ALL, or MDS. The presence of chronic GVHD was associated with significantly higher treatment-related mortality and worse overall survival across all diseases studied. As it is important to identify the setting in which chronic GVHD is most beneficial for leukemia control, these data provide support for focusing on developing better chronic GVHD therapies and prevention to improve survival of patients with leukemia after myeloablative allogeneic hematopoietic cell transplantation.

Allogeneic hematopoietic cell transplantation (HCT) following myeloablative conditioning (MAC) regimen is a standard and curative treatment option for hematologic malignancies (1). The antileukemic activity of MAC and allogeneic HCT relies not only on the effects of high-dose chemotherapy or irradiation given during the conditioning regimen, but also on the immune-mediated graft-versus-leukemia (GVL) effect (2–4). The immune reactivity between donor T cells that is responsible for the GVL effect and the recipient is also associated with the major complications of allogeneic HCT, namely acute (aGVHD) and chronic graft-versus-host disease (cGVHD).

Chronic GVHD is a serious complication and is an important cause of morbidity and nonrelapse mortality (NRM) in patients who survive 12 months after transplantation (5). Chronic GVHD is associated with a lower risk of relapse (2, 4, 6). Despite the protective effect of cGVHD, adult patients with cGVHD experience leukemia late relapse (7). The reduction in relapse risk may be secondary to the immune-mediated graft-versus-tumor (GVT) effect associated with GVHD. This effect is most prominent during the first year after allogeneic HCT, when the peak incidence of relapse occurs (7, 8). However, onsets of acute and chronic GVHD chronologically overlap and it is difficult to decipher their relative contributions to antileukemia effects because at the time both acute and chronic GVHD are likely active (9). Given this close time overlap between aGVHD and cGVHD, this study sought to investigate antileukemic effects of cGVHD isolated from acute GVHD by assessing relapses and survival in allogeneic HCT recipients who were alive and relapse free at 1 year following MAC transplant. In addition, we evaluated whether any specific cGVHD clinical characteristics or organ manifestations are more predominantly associated with the GVL effects of cGVHD.

Data source

The Center for International Blood and Marrow Transplant Research (CIBMTR) is a research affiliation of the International Bone Marrow Transplant Registry (IBMTR), Autologous Blood and Marrow Transplant Registry (ABMTR), and the National Marrow Donor Program (NMDP), which was established in 2004 and comprises a voluntary working group of more than 450 transplantation centers worldwide that contribute detailed data on consecutive allogeneic and autologous HCT to a Statistical Center at the Medical College of Wisconsin in Milwaukee and the NMDP Coordinating Center in Minneapolis. Participating centers are required to report all transplants consecutively; compliance is monitored by on-site audits. Patients are followed longitudinally. Computerized checks for discrepancies, physicians' review of submitted data, and on-site audits of participating centers ensure data quality. Observational studies conducted by the CIBMTR are performed in compliance with all applicable federal regulations pertaining to the protection of human research participants. Protected Health Information used in the performance of such research is collected and maintained in CIBMTR's capacity as a Public Health Authority under the HIPAA Privacy Rule.

Patient selection

Between 1995 and 2004, 19,861 patients with acute myelogenous leukemia (AML), acute lymphoblastic leukemia (ALL), chronic myelogenous leukemia (CML), and myelodysplastic syndromes (MDS) received allogeneic HCT. Patients who received reduced intensity conditioning regimen (n = 2,620), died or relapsed (n = 9,162) within the first year after allogeneic HCT were excluded from the analysis. Patients who received second allergenic HCT (n = 146) within the first posttransplant year or syngeneic HCT were excluded as were patients who had relevant data missing from the CIBMTR database (n = 444). The final study population consisted of a total of 7,489 patients that received MAC regimen who were alive and free of disease at 1 year after transplantation.

Study definitions and endpoints

Patients were considered to have early disease whether they were in first remission (acute leukemia) or first chronic phase (CML) or MDS with refractory anemia or refractory anemia with ringed sideroblasts (RA, RARS); intermediate disease: second or later complete remission (acute leukemia), second or later chronic phase/accelerated phase (CML); advanced disease: relapse or primary induction failure (acute leukemia) or blast crisis (CML) or MDS with refractory anemia with excess blasts or excess blasts in transformation (RAEB, RAEB-t). The NMDP classification of HLA matching status that allows adequate adjustment for donor-recipient HLA compatibility while accounting for best available resolution of typing was used to categorize HLA-matching status as well-matched, partially matched, or mismatched (10). Acute GVHD was grouped as none versus grade 1 versus grade 2–4 and was graded according to IBMTR criteria based on the pattern of severity of abnormalities in skin, gastrointestinal tract, and liver (11). cGVHD was diagnosed according to standard CIBMTR criteria, which included all patients with clinical criteria of cGVHD with or without positive histology, irrespective of time of onset of symptoms (9, 12). The NIH consensus criteria were not available at the time of data collection for this analysis (13).

The primary endpoint was leukemia relapse and was defined as time to onset of disease recurrence (by hematologic or cytogenetic criteria) with treatment-related mortality (TRM) as a competing risk. Secondary endpoints were overall survival (OS), TRM, and disease-free survival (DFS). Outcomes were assessed up to 5 years after HCT. TRM was defined as death in continuous remission, and DFS was defined as either death or relapse with the time calculated from date of transplant. OS was calculated from date of transplant. Death from any cause was treated as an event and surviving patients were censored at the date of last contact.

Statistical analysis

Variables related to patient, disease, and transplant characteristics were summarized using descriptive statistics. Cumulative incidence for relapse was calculated treating TRM as a competing risk (14). Patient-related, disease-related, and treatment-related variables were included in the multivariate analyses using a stepwise forward selection technique and P ≤ 0.01 was the criterion for inclusion in final models. Patient- and transplant-related variables tested in the models included recipient age, sex, donor type, graft type (bone marrow or peripheral blood), donor-recipient gender mismatch, donor parity, donor-recipient cytomegalovirus (CMV) serology, use of total body irradiation (TBI), use of ATG or alemtuzumab, GVHD prophylaxis, presence and grade of prior aGVHD, and year of transplantation. Disease-related variables were diagnosis and disease status pretransplant. Chronic GVHD-specific variables included platelet count at diagnosis (<100 × 109/L and ≥100 × 109/L), serum bilirubin at diagnosis (<1, 1–2, and >2 mg/dL), type of onset of cGVHD (progressive, interrupted, de novo), Karnofsky Performance Scale/Lansky score (KPS/L) at diagnosis, severity of cGVHD at 1 year after transplant (mild vs. moderate vs. severe), and organ involvement of cGVHD at 1 year after transplant.

In the multivariate analysis, all the endpoints were analyzed using the Cox proportional hazards model and all variables were tested for affirmation of the proportional hazards assumption. Variables that did not satisfy the proportional hazards assumption were adjusted for by stratification. A stepwise model-building procedure was used to develop models for each outcome with a threshold of 0.05 for both entry and retention in the model. Interaction between the main variable and adjusted variables were tested, and no significant interactions were identified at P < 0.01.

Patients

The study included 7,489 patients who were alive and disease-free 1 year after HCT. Patient and donor characteristics are summarized in Table 1. The median age at HCT was 32 years for AML, 17 years for ALL, 36 years for CML, and 37 years for MDS. At 1 year after HCT 44%, 41%, 54%, and 54% of the AML, ALL, CML, MDS patients, respectively, had developed cGVHD. Chronic GVHD characteristics are summarized in Table 1.

Table 1.

Patient, donor, and GVHD characteristics

AMLALLCMLMDS
Patient and donor characteristics 
 Characteristics N (%) N (%) N (%) N (%) 
 Number of patients (n = 7,489) 2,541 1,798 2,498 652 
 Median recipient age at HCT, y (range) 32 (<1–74) 17 (<1–59) 36 (1–66) 37 (<1–65) 
 Age of recipient, y 
  <2 58 (2) 58 (3) 1 (<1) 30 (5) 
  2–17 541 (21) 886 (49) 233 (9) 122 (19) 
  18–29 561 (22) 431 (24) 551 (22) 96 (15) 
  30–39 512 (20) 230 (13) 768 (31) 121 (19) 
  40–49 565 (22) 136 (8) 673 (27) 144 (22) 
  50–59 281 (11) 57 (3) 259 (10) 129 (20) 
  60+ 23 (<1) 13 (<1) 10 (2) 
 Gender 
  Male 1,346 (53) 1,121 (62) 1,469 (59) 359 (55) 
  Female 1,195(47) 677(38) 1,029(41) 293(45) 
 Disease status at HCT 
  Early 1,518 (60) 804 (45) 2,110 (84) 229 (35) 
  Intermediate 612 (24) 862 (48) 349 (14) 6 (1) 
  Advanced 411 (16) 132 (7) 39 (2) 417 (64) 
 Graft source 
  Bone marrow 1,611 (63) 1,383 (77) 1,928 (77) 434 (67) 
  Peripheral blood 930 (37) 415 (23) 570 (23) 218 (33) 
 HLA match 
  HLA-identical sibling 1,503 (59) 810 (45) 1,284 (51) 264 (40) 
  Well-matched unrelated 445 (18) 352 (20) 434 (17) 185 (28) 
  Partially matched unrelated 368 (14) 385 (21) 486 (19) 132 (20) 
  Mismatched unrelated 225 (9) 251 (14) 294 (12) 71 (11) 
 Donor-recipient gender match 
  Male → male 796 (31) 675 (38) 947 (38) 246 (38) 
  Male → female 640 (25) 375 (21) 551 (22) 155 (24) 
  Female → male 550 (22) 446 (25) 522 (21) 113 (17) 
  Female → female 555 (22) 302 (17) 478 (19) 138 (21) 
 Donor-recipient CMV status 
  Negative → negative 727 (29) 673 (37) 791 (32) 219 (34) 
  Negative → positive 526 (21) 316 (18) 410 (16) 139 (21) 
  Positive → negative 313 (12) 267 (15) 328 (13) 81 (12) 
  Positive → positive 872 (34) 483 (27) 869 (35) 189 (29) 
  Unknown 103 (4) 59 (3) 100 (4) 24 (4) 
 Median donor age, years (range) 33 (<1–72) 30 (<1–73) 36 (<1–73) 36 (<1–71) 
 HLA-identical sibling donor age, years, median (range) 32 (<1–72) 18 (<1–73) 36 (<1–73) 36 (<1–71) 
 HLA-identical sibling donor age, y 
  <10 135 (9) 187 (23) 32 (2) 28 (11) 
  10–17 181 (12) 206 (25) 86 (7) 20 (8) 
  18–29 360 (24) 203 (25) 269 (21) 53 (20) 
  30–39 330 (22) 108 (13) 388 (30) 52 (20) 
  40–49 297 (20) 66 (8) 326 (25) 65 (25) 
  50–59 140 (9) 29 (4) 124 (10) 33 (13) 
  60+ 41 (3) 4 (<1) 31 (2) 8 (3) 
  Missing 19 (1) 7 (<1) 28 (2) 5 (2) 
 URD age, years, median (range) 35 (19–60) 35 (19–59) 35 (18–65) 35 (19–60) 
 URD age, y 
  18–29 315 (30) 284 (29) 365 (30) 104 (27) 
  30–39 404 (39) 367 (37) 434 (36) 151 (39) 
  40–49 233 (22) 234 (24) 287 (24) 97 (25) 
  50–59 52 (5) 57 (6) 78 (6) 20 (5) 
  60+ 2 (<1) 2 (<1) 1 (<1) 
  Missing 32 (3) 46 (5) 48 (4) 15 (4) 
 Donor parity before transplant 
  Male donor 1,436 (57) 1,050 (58) 1498 (60) 401 (62) 
  Female, no pregnancy 420 (17) 380 (21) 324 (13) 78 (12) 
  1 or more pregnancies 473 (19) 266 (15) 483 (19) 128 (20) 
  Missing 212 (8) 102 (6) 193 (8) 45 (7) 
 GVHD prophylaxis 
  Ex vivo T-cell depletion 246 (10) 218 (12) 204 (8) 81 (12) 
  Cyclosporine ± methotrexate ± other 1,900 (75) 1,362 (76) 2,008 (80) 472 (72) 
  Tacrolimus ± methotrexate ± other 395 (16) 218 (12) 286 (11) 99 (15) 
 ATG and/or alemtuzumab used 
  None 2,131 (84) 1,399 (78) 2,088 (84) 519 (80) 
  Yes 312 (12) 332 (18) 318 (13) 113 (17) 
  Missing 98 (4) 67 (4) 92 (4) 20 (3) 
 Total body irradiation use in the conditioning regimen 
  No 1,314 (52) 228 (13) 1,268 (51) 379 (58) 
  Yes 1,227 (48) 1,570 (87) 1,230 (49) 273 (42) 
 Year of transplant 
  1995–1999 1,332 (52) 1,011 (56) 1,724 (69) 351 (54) 
  2000–2004 1,209 (48) 787 (44) 774 (31) 301 (46) 
 Prior aGVHD grade 
  0–1 1,537 (60) 864 (48) 1,334 (53) 317 (49) 
  2–4 1,004 (40) 934 (52) 1,164 (47) 335 (51) 
 Median follow-up of survivors, months (range) 80 (12–179) 80 (12–175) 93 (12–172) 80 (12–168) 
Chronic GVHD characteristics 
 cGVHD present at one-year after HCT 1,117/2,541 (44%) 742/1,798 (41%) 1,344/2,498 (54%) 350/652 (54%) 
 Time from HCT to cGVHD months (range) 5 (1–11) 5 (1–11) 5 (1–11) 5 (1–11) 
 Severity of cGVHD 
  Mild 485 (43) 265 (36) 425 (32) 136 (39) 
  Moderate 243 (22) 159 (21) 307 (23) 67 (19) 
  Severe 70 (6) 48 (6) 77 (6) 14 (4) 
  Missing 319 (29) 270 (36) 535 (40) 133 (38) 
 Onset of cGVHD 
  Progressive 366 (33) 306 (41) 509 (38) 130 (37) 
  Interrupted 277 (25) 204 (27) 339 (25) 96 (27) 
  De novo 392 (35) 179 (24) 422 (31) 91 (26) 
  Missing 82 (7) 53 (7) 74 (6) 33 (9) 
 Organ involvement 
  Skin 597 (53) 445 (60) 733 (55) 189 (54) 
  Eyes 331 (30) 170 (23) 401 (30) 105 (30) 
  Mouth 560 (50) 331 (45) 685 (51) 173 (49) 
  Lung 93 (8) 53 (7) 109 (8) 34 (10) 
  Gastrointestinal 217 (19) 177 (24) 273 (20) 89 (25) 
  GU 33 (3) 12 (2) 29 (2) 13 (4) 
  Liver 425 (38) 213 (29) 517 (38) 122 (35) 
  Musculoskeletal 59 (5) 54 (7) 110 (8) 21 (6) 
  Hematologic 109 (10) 95 (13) 138 (10) 45 (13) 
 Number of organs involved in cGVHD 
  1 259 (23) 195 (26) 304 (23) 82 (23) 
  2 258 (23) 177 (24) 342 (25) 74 (21) 
  3 212 (19) 127 (17) 251 (19) 59 (17) 
  4 138 (12) 76 (10) 149 (11) 49 (14) 
  5 35 (3) 21 (3) 46 (3) 15 (4) 
  6 5 (<1) 2 (<1) 9 (1) 2 (1) 
  Missing 210 (19) 144 (19) 243 (18) 69 (20) 
 KPS at diagnosis of cGVHD, n (%) 
  <80 215 (19) 154 (21) 367 (27) 90 (26) 
  80–100 704 (63) 453 (61) 789 (59) 173 (49) 
  Missing 198 (18) 135 (18) 188 (14) 87 (25) 
 Platelet count at diagnosis cGVHD × 109/L, n (%) 
  <100 294 (26) 214 (29) 450 (33) 95 (27) 
  ≥100 639 (57) 409 (55) 682 (51) 183 (52) 
  Missing 184 (16) 119 (16) 212 (16) 72 (21) 
 Serum bilirubin at diagnosis of cGVHD mg/dL, n (%) 
  <1 665 (60) 445 (60) 638 (47) 191 (55) 
  1–2 152 (14) 86 (12) 311 (23) 66 (19) 
  >2 159 (14) 117 (16) 220 (16) 37 (11) 
  Missing 141 (13) 94 (13) 175 (13) 56 (16) 
AMLALLCMLMDS
Patient and donor characteristics 
 Characteristics N (%) N (%) N (%) N (%) 
 Number of patients (n = 7,489) 2,541 1,798 2,498 652 
 Median recipient age at HCT, y (range) 32 (<1–74) 17 (<1–59) 36 (1–66) 37 (<1–65) 
 Age of recipient, y 
  <2 58 (2) 58 (3) 1 (<1) 30 (5) 
  2–17 541 (21) 886 (49) 233 (9) 122 (19) 
  18–29 561 (22) 431 (24) 551 (22) 96 (15) 
  30–39 512 (20) 230 (13) 768 (31) 121 (19) 
  40–49 565 (22) 136 (8) 673 (27) 144 (22) 
  50–59 281 (11) 57 (3) 259 (10) 129 (20) 
  60+ 23 (<1) 13 (<1) 10 (2) 
 Gender 
  Male 1,346 (53) 1,121 (62) 1,469 (59) 359 (55) 
  Female 1,195(47) 677(38) 1,029(41) 293(45) 
 Disease status at HCT 
  Early 1,518 (60) 804 (45) 2,110 (84) 229 (35) 
  Intermediate 612 (24) 862 (48) 349 (14) 6 (1) 
  Advanced 411 (16) 132 (7) 39 (2) 417 (64) 
 Graft source 
  Bone marrow 1,611 (63) 1,383 (77) 1,928 (77) 434 (67) 
  Peripheral blood 930 (37) 415 (23) 570 (23) 218 (33) 
 HLA match 
  HLA-identical sibling 1,503 (59) 810 (45) 1,284 (51) 264 (40) 
  Well-matched unrelated 445 (18) 352 (20) 434 (17) 185 (28) 
  Partially matched unrelated 368 (14) 385 (21) 486 (19) 132 (20) 
  Mismatched unrelated 225 (9) 251 (14) 294 (12) 71 (11) 
 Donor-recipient gender match 
  Male → male 796 (31) 675 (38) 947 (38) 246 (38) 
  Male → female 640 (25) 375 (21) 551 (22) 155 (24) 
  Female → male 550 (22) 446 (25) 522 (21) 113 (17) 
  Female → female 555 (22) 302 (17) 478 (19) 138 (21) 
 Donor-recipient CMV status 
  Negative → negative 727 (29) 673 (37) 791 (32) 219 (34) 
  Negative → positive 526 (21) 316 (18) 410 (16) 139 (21) 
  Positive → negative 313 (12) 267 (15) 328 (13) 81 (12) 
  Positive → positive 872 (34) 483 (27) 869 (35) 189 (29) 
  Unknown 103 (4) 59 (3) 100 (4) 24 (4) 
 Median donor age, years (range) 33 (<1–72) 30 (<1–73) 36 (<1–73) 36 (<1–71) 
 HLA-identical sibling donor age, years, median (range) 32 (<1–72) 18 (<1–73) 36 (<1–73) 36 (<1–71) 
 HLA-identical sibling donor age, y 
  <10 135 (9) 187 (23) 32 (2) 28 (11) 
  10–17 181 (12) 206 (25) 86 (7) 20 (8) 
  18–29 360 (24) 203 (25) 269 (21) 53 (20) 
  30–39 330 (22) 108 (13) 388 (30) 52 (20) 
  40–49 297 (20) 66 (8) 326 (25) 65 (25) 
  50–59 140 (9) 29 (4) 124 (10) 33 (13) 
  60+ 41 (3) 4 (<1) 31 (2) 8 (3) 
  Missing 19 (1) 7 (<1) 28 (2) 5 (2) 
 URD age, years, median (range) 35 (19–60) 35 (19–59) 35 (18–65) 35 (19–60) 
 URD age, y 
  18–29 315 (30) 284 (29) 365 (30) 104 (27) 
  30–39 404 (39) 367 (37) 434 (36) 151 (39) 
  40–49 233 (22) 234 (24) 287 (24) 97 (25) 
  50–59 52 (5) 57 (6) 78 (6) 20 (5) 
  60+ 2 (<1) 2 (<1) 1 (<1) 
  Missing 32 (3) 46 (5) 48 (4) 15 (4) 
 Donor parity before transplant 
  Male donor 1,436 (57) 1,050 (58) 1498 (60) 401 (62) 
  Female, no pregnancy 420 (17) 380 (21) 324 (13) 78 (12) 
  1 or more pregnancies 473 (19) 266 (15) 483 (19) 128 (20) 
  Missing 212 (8) 102 (6) 193 (8) 45 (7) 
 GVHD prophylaxis 
  Ex vivo T-cell depletion 246 (10) 218 (12) 204 (8) 81 (12) 
  Cyclosporine ± methotrexate ± other 1,900 (75) 1,362 (76) 2,008 (80) 472 (72) 
  Tacrolimus ± methotrexate ± other 395 (16) 218 (12) 286 (11) 99 (15) 
 ATG and/or alemtuzumab used 
  None 2,131 (84) 1,399 (78) 2,088 (84) 519 (80) 
  Yes 312 (12) 332 (18) 318 (13) 113 (17) 
  Missing 98 (4) 67 (4) 92 (4) 20 (3) 
 Total body irradiation use in the conditioning regimen 
  No 1,314 (52) 228 (13) 1,268 (51) 379 (58) 
  Yes 1,227 (48) 1,570 (87) 1,230 (49) 273 (42) 
 Year of transplant 
  1995–1999 1,332 (52) 1,011 (56) 1,724 (69) 351 (54) 
  2000–2004 1,209 (48) 787 (44) 774 (31) 301 (46) 
 Prior aGVHD grade 
  0–1 1,537 (60) 864 (48) 1,334 (53) 317 (49) 
  2–4 1,004 (40) 934 (52) 1,164 (47) 335 (51) 
 Median follow-up of survivors, months (range) 80 (12–179) 80 (12–175) 93 (12–172) 80 (12–168) 
Chronic GVHD characteristics 
 cGVHD present at one-year after HCT 1,117/2,541 (44%) 742/1,798 (41%) 1,344/2,498 (54%) 350/652 (54%) 
 Time from HCT to cGVHD months (range) 5 (1–11) 5 (1–11) 5 (1–11) 5 (1–11) 
 Severity of cGVHD 
  Mild 485 (43) 265 (36) 425 (32) 136 (39) 
  Moderate 243 (22) 159 (21) 307 (23) 67 (19) 
  Severe 70 (6) 48 (6) 77 (6) 14 (4) 
  Missing 319 (29) 270 (36) 535 (40) 133 (38) 
 Onset of cGVHD 
  Progressive 366 (33) 306 (41) 509 (38) 130 (37) 
  Interrupted 277 (25) 204 (27) 339 (25) 96 (27) 
  De novo 392 (35) 179 (24) 422 (31) 91 (26) 
  Missing 82 (7) 53 (7) 74 (6) 33 (9) 
 Organ involvement 
  Skin 597 (53) 445 (60) 733 (55) 189 (54) 
  Eyes 331 (30) 170 (23) 401 (30) 105 (30) 
  Mouth 560 (50) 331 (45) 685 (51) 173 (49) 
  Lung 93 (8) 53 (7) 109 (8) 34 (10) 
  Gastrointestinal 217 (19) 177 (24) 273 (20) 89 (25) 
  GU 33 (3) 12 (2) 29 (2) 13 (4) 
  Liver 425 (38) 213 (29) 517 (38) 122 (35) 
  Musculoskeletal 59 (5) 54 (7) 110 (8) 21 (6) 
  Hematologic 109 (10) 95 (13) 138 (10) 45 (13) 
 Number of organs involved in cGVHD 
  1 259 (23) 195 (26) 304 (23) 82 (23) 
  2 258 (23) 177 (24) 342 (25) 74 (21) 
  3 212 (19) 127 (17) 251 (19) 59 (17) 
  4 138 (12) 76 (10) 149 (11) 49 (14) 
  5 35 (3) 21 (3) 46 (3) 15 (4) 
  6 5 (<1) 2 (<1) 9 (1) 2 (1) 
  Missing 210 (19) 144 (19) 243 (18) 69 (20) 
 KPS at diagnosis of cGVHD, n (%) 
  <80 215 (19) 154 (21) 367 (27) 90 (26) 
  80–100 704 (63) 453 (61) 789 (59) 173 (49) 
  Missing 198 (18) 135 (18) 188 (14) 87 (25) 
 Platelet count at diagnosis cGVHD × 109/L, n (%) 
  <100 294 (26) 214 (29) 450 (33) 95 (27) 
  ≥100 639 (57) 409 (55) 682 (51) 183 (52) 
  Missing 184 (16) 119 (16) 212 (16) 72 (21) 
 Serum bilirubin at diagnosis of cGVHD mg/dL, n (%) 
  <1 665 (60) 445 (60) 638 (47) 191 (55) 
  1–2 152 (14) 86 (12) 311 (23) 66 (19) 
  >2 159 (14) 117 (16) 220 (16) 37 (11) 
  Missing 141 (13) 94 (13) 175 (13) 56 (16) 

Relapse

The cumulative incidence of relapse after the first post-HCT year for each disease is shown in Table 2. In multivariate analysis, a protective effect of cGVHD was seen only in patients with CML [RR, 0.47; 95% confidence interval (CI), 0.37–0.59; P <0.0001; Fig. 1A and B]. Other factors protective against late relapse were aGVHD in the first 29 days after HCT, GVHD prophylaxis other than T-cell depletion, well-matched or partially matched unrelated donor (URD; as compared with HLA-identical sibling), and early disease status at transplant (Table 3, Supplementary Table S1).

Figure 1.

A and B, cumulative incidence of late relapse according to prior history of chronic GVHD among patients who are disease free at 1 year after HCT. C and D, cumulative incidence of TRM and OS according to prior history of chronic GVHD among patients who are disease free at 1 year after HCT.

Figure 1.

A and B, cumulative incidence of late relapse according to prior history of chronic GVHD among patients who are disease free at 1 year after HCT. C and D, cumulative incidence of TRM and OS according to prior history of chronic GVHD among patients who are disease free at 1 year after HCT.

Close modal
Table 2.

Incidence of late relapse and death

Cumulative incidence of late relapse (up to 5 years post HCT) according to diagnosis among patients who were disease free at 1 year after HCT
AMLALLCMLMDSTotal
No relapse 2,170 (85.4%) 1,472 (81.87%) 2,176 (87.11%) 582 (89.26%) 6,400 (85%) 
Relapse 371 (14.6%) 326 (18.13%) 322 (12.89%) 70 (10.74%) 1,089 (15%) 
Total 2,541 1,798 2,498 652 7,489 
Cumulative incidence of late relapse according to diagnosis among patients who were disease free and had no cGVHD at 1 year after HCT 
 Probability (95% CI) at different time points after HCT  
Time after HCT AML n = 1,424 ALL n = 1,056 CML n = 1,154 MDS n = 302  
2 years 9 (7–10) 12 (10–14) 8 (6–10) 5 (3–8)  
3 years 11 (10–13) 15 (13–18) 12 (11–15) 8 (5–12)  
5 years 15 (13–17) 18 (16–21) 17 (14–19) 10 (7–14)  
Cumulative incidence of late relapse according to diagnosis among patients who were disease free and had cGVHD at 1 year after HCT 
 Probability (95% CI) at different time points after HCT  
Time after HCT AML n = 1,117 ALL n = 742 CML n = 1,344 MDS n = 350  
2 years 7 (6–9) 13 (11–15) 4 (3–5) 5 (3–8)  
3 years 11 (9–13) 15 (13–18) 6 (5–8) 7 (5–11)  
5 years 13 (11–15) 18 (16–21) 8 (6–9) 10 (7–13)  
Cumulative incidence of death (up to 5 years after HCT) according to diagnosis among patients who were disease free at 1 year after HCT 
 AML ALL CML MDS Total 
Alive 1,922 (75.6%) 1,354 (75.31%) 2,018 (80.78%) 474 (72.7%) 5,768 (77%) 
Dead 619 (24.3%) 444 (24.6%) 480 (19.2%) 178 (27.3%) 1,721 (23%) 
Total 2,541 1,798 2,498 652 7,489 
Cumulative incidence of late relapse (up to 5 years post HCT) according to diagnosis among patients who were disease free at 1 year after HCT
AMLALLCMLMDSTotal
No relapse 2,170 (85.4%) 1,472 (81.87%) 2,176 (87.11%) 582 (89.26%) 6,400 (85%) 
Relapse 371 (14.6%) 326 (18.13%) 322 (12.89%) 70 (10.74%) 1,089 (15%) 
Total 2,541 1,798 2,498 652 7,489 
Cumulative incidence of late relapse according to diagnosis among patients who were disease free and had no cGVHD at 1 year after HCT 
 Probability (95% CI) at different time points after HCT  
Time after HCT AML n = 1,424 ALL n = 1,056 CML n = 1,154 MDS n = 302  
2 years 9 (7–10) 12 (10–14) 8 (6–10) 5 (3–8)  
3 years 11 (10–13) 15 (13–18) 12 (11–15) 8 (5–12)  
5 years 15 (13–17) 18 (16–21) 17 (14–19) 10 (7–14)  
Cumulative incidence of late relapse according to diagnosis among patients who were disease free and had cGVHD at 1 year after HCT 
 Probability (95% CI) at different time points after HCT  
Time after HCT AML n = 1,117 ALL n = 742 CML n = 1,344 MDS n = 350  
2 years 7 (6–9) 13 (11–15) 4 (3–5) 5 (3–8)  
3 years 11 (9–13) 15 (13–18) 6 (5–8) 7 (5–11)  
5 years 13 (11–15) 18 (16–21) 8 (6–9) 10 (7–13)  
Cumulative incidence of death (up to 5 years after HCT) according to diagnosis among patients who were disease free at 1 year after HCT 
 AML ALL CML MDS Total 
Alive 1,922 (75.6%) 1,354 (75.31%) 2,018 (80.78%) 474 (72.7%) 5,768 (77%) 
Dead 619 (24.3%) 444 (24.6%) 480 (19.2%) 178 (27.3%) 1,721 (23%) 
Total 2,541 1,798 2,498 652 7,489 
Table 3.

Multivariate analysis of late relapses up to 5 years after HCT

NumberRR (95% CI)P
AML patients 
 No cGVHD 1,424 1.00  
 cGVHD 1,117 0.93 (0.75–1.14) 0.4809 
ALL patients 
 No cGVHD 1,056 1.00  
 cGVHD 742 1.09 (0.87–1.37) 0.4423 
CML patients 
 No cGVHD 1,154 1.00  
 cGVHD 1,344 0.47 (0.37–0.59) <0.0001 
MDS patients 
 No cGVHD 302 1.00  
 cGVHD 350 1.12 (0.70–1.79) 0.6391 
Acute GVHD   <0.0001 
 none 4,051 1.00  
 aGVHD in first 29 days after HCT 2,446 0.74 (0.63–0.85) <0.0001 
 aGVHD after day 29 after HCT 986 1.15 (0.96–1.37) 0.1223 
Disease status at transplant   <.0001 
 Early 4,661 1.00  
 Intermediate 1,823 1.57 (1.35–1.83) <0.0001 
 Advanced 999 2.57 (2.13–3.10) <0.0001 
HLA status   0.0200 
 Sibling donor 1,861 1.00  
 Well-matched unrelated 1,412 0.83 (0.69–0.99) 0.0370 
 Partially matched unrelated 1,369 0.78 (0.65–0.93) 0.0049 
 Poorly matched unrelated 841 0.85 (0.69–1.04) 0.1143 
GVHD prophylaxis   0.0043 
 T-cell depletion 748 1.00  
 Cyclosporine ± methotrexate ± other 5,738 0.75 (0.62–0.91) 0.0032 
 Tacrolimus ± methotrexate ± other 997 0.68 (0.53–0.88) 0.0026 
NumberRR (95% CI)P
AML patients 
 No cGVHD 1,424 1.00  
 cGVHD 1,117 0.93 (0.75–1.14) 0.4809 
ALL patients 
 No cGVHD 1,056 1.00  
 cGVHD 742 1.09 (0.87–1.37) 0.4423 
CML patients 
 No cGVHD 1,154 1.00  
 cGVHD 1,344 0.47 (0.37–0.59) <0.0001 
MDS patients 
 No cGVHD 302 1.00  
 cGVHD 350 1.12 (0.70–1.79) 0.6391 
Acute GVHD   <0.0001 
 none 4,051 1.00  
 aGVHD in first 29 days after HCT 2,446 0.74 (0.63–0.85) <0.0001 
 aGVHD after day 29 after HCT 986 1.15 (0.96–1.37) 0.1223 
Disease status at transplant   <.0001 
 Early 4,661 1.00  
 Intermediate 1,823 1.57 (1.35–1.83) <0.0001 
 Advanced 999 2.57 (2.13–3.10) <0.0001 
HLA status   0.0200 
 Sibling donor 1,861 1.00  
 Well-matched unrelated 1,412 0.83 (0.69–0.99) 0.0370 
 Partially matched unrelated 1,369 0.78 (0.65–0.93) 0.0049 
 Poorly matched unrelated 841 0.85 (0.69–1.04) 0.1143 
GVHD prophylaxis   0.0043 
 T-cell depletion 748 1.00  
 Cyclosporine ± methotrexate ± other 5,738 0.75 (0.62–0.91) 0.0032 
 Tacrolimus ± methotrexate ± other 997 0.68 (0.53–0.88) 0.0026 

NOTE: Factors tested, but not listed in table: Recipient age, sex, graft type, donor-recipient gender mismatch, donor parity, donor-recipient CMV serology, use of TBI, use of ATG, aGVHD grade, year of transplant, platelet count at cGVHD diagnosis, serum bilirubin at cGVHD diagnosis, type of cGVHD onset, KPS/L at cGVHD diagnosis, severity of cGVHD at 1 year after transplant, organ involvement of cGVHD at 1 year after transplant.

Transplant-related mortality and OS

Multivariate analysis of risk factors associated with TRM and OS are outlined in Tables 4 and 5, respectively. The presence of cGVHD was associated with a higher risk of TRM for all diseases (RR, 2.43; 95% CI, 2.09–2.82; P <0.0001; Fig. 1C, Table 4). Other factors associated with a higher risk of TRM included aGVHD, advanced disease status at HCT, well-matched, partially matched, or poorly matched URD (as compared with HLA-identical sibling donor), use of TBI in conditioning, female donor to male recipient (vs. male donor for male recipient), and use of peripheral blood stem cells (PBSC; vs. bone marrow) as the graft source (Supplementary Table S1).

Table 4.

Multivariate analyses of risk factors for late transplant-related mortality

NumberRR (95% CI)P
Prior chronic GVHD 
 No prior diagnosis of cGVHD 3,936 1.00  
 Prior cGVHD 3,547 2.43 (2.09–2.82) <0.0001 
Acute GVHD history   <0.0001 
 None 4,051 1.00  
 aGVHD in first 29 days after HCT 2,446 1.40 (1.21–1.63) <0.0001 
 aGVHD after day 29 after HCT 986 1.58 (1.32–1.91) <0.0001 
Disease Status at HCT   <0.0001 
 Early 4,661 1.00  
 Intermediate 1,823 0.93 (0.78–1.10) 0.3908 
 Late 999 1.43 (1.20–1.69) <0.0001 
HLA match   <0.0001 
 Sibling donor 1,861 1.00  
 Well-matched unrelated 1,412 1.39 (1.15–1.67) 0.0006 
 Partially matched unrelated 1,369 1.36 (1.13–1.63) 0.0013 
 Poorly matched unrelated 841 1.56 (1.25–1.93) <0.0001 
TBI in the conditioning regimen 
 No 3,186 1.00  
 Yes 4,292 1.47 (1.27–1.70) <0.0001 
Donor-recipient gender match   <0.0001 
 Male → male 2,662 1.00  
 Male → female 1,720 0.63 (0.77–1.11) 0.4146 
 Female → male 1,629 1.55 (1.32–1.83) <0.0001 
 Female → female 1,472 1.11 (0.93–1.33) 0.2623 
Graft source 
 Bone marrow 5,354 1.00  
 Mobilized blood stem cells 2,129 1.48 (1.26–1.73) <0.0001 
NumberRR (95% CI)P
Prior chronic GVHD 
 No prior diagnosis of cGVHD 3,936 1.00  
 Prior cGVHD 3,547 2.43 (2.09–2.82) <0.0001 
Acute GVHD history   <0.0001 
 None 4,051 1.00  
 aGVHD in first 29 days after HCT 2,446 1.40 (1.21–1.63) <0.0001 
 aGVHD after day 29 after HCT 986 1.58 (1.32–1.91) <0.0001 
Disease Status at HCT   <0.0001 
 Early 4,661 1.00  
 Intermediate 1,823 0.93 (0.78–1.10) 0.3908 
 Late 999 1.43 (1.20–1.69) <0.0001 
HLA match   <0.0001 
 Sibling donor 1,861 1.00  
 Well-matched unrelated 1,412 1.39 (1.15–1.67) 0.0006 
 Partially matched unrelated 1,369 1.36 (1.13–1.63) 0.0013 
 Poorly matched unrelated 841 1.56 (1.25–1.93) <0.0001 
TBI in the conditioning regimen 
 No 3,186 1.00  
 Yes 4,292 1.47 (1.27–1.70) <0.0001 
Donor-recipient gender match   <0.0001 
 Male → male 2,662 1.00  
 Male → female 1,720 0.63 (0.77–1.11) 0.4146 
 Female → male 1,629 1.55 (1.32–1.83) <0.0001 
 Female → female 1,472 1.11 (0.93–1.33) 0.2623 
Graft source 
 Bone marrow 5,354 1.00  
 Mobilized blood stem cells 2,129 1.48 (1.26–1.73) <0.0001 

NOTE: Factors tested, but not listed in table: Recipient age, sex, donor parity, donor-recipient CMV serology, use of ATG, GVHD prophylaxis, aGVHD grade, year of transplant, disease, platelet count at cGVHD diagnosis, serum bilirubin at cGVHD diagnosis, type of cGVHD onset, KPS/L at cGVHD diagnosis, severity of cGVHD at 1 year after transplant, organ involvement of cGVHD at 1 year after transplant.

Table 5.

Multivariate analyses of risk factors for overall mortality among patients who were alive and disease free at 1 year after HCT

NumberRR (95% CI)P
Presence of cGVHD 
 No cGVHD 3,936 1.00  
 cGVHD 3,547 1.56 (1.41–1.73) <0.0001 
Acute GVHD history   <0.0001 
 None 4,051 1.00  
 aGVHD in first 29 days 2,446 1.11 (0.93–1.24) 0.0716 
 aGVHD after day 29 986 1.46 (1.28–1.67) <0.0001 
Disease   <0.0001 
 AML 2,541 1.00  
 ALL 1,798 1.26 (1.10–1.45) 0.0008 
 CML 2,498 0.73 (0.64–0.83) <0.0001 
 MDS 652 0.84 (0.70–1.00) 0.0521 
Disease status at HCT   <0.0001 
 Early 4,661 1.00  
 Intermediate 1,823 1.25 (1.10–1.41) <0.0001 
 Advanced 999 1.83 (1.59–2.11) 0.0005 
HLA match   0.0191 
 Sibling donor 1,861 1.00  
 Well-matched unrelated 1,412 1.18 (1.03–1.36) 0.0199 
 Partially matched unrelated 1,369 1.12 (0.92–1.29) 0.1148 
 Poorly matched unrelated 841 1.26 (1.07–1.49) 0.0049 
TBI in the conditioning regimen 
 No 3,186 1.00  
 Yes 4,292 1.20 (1.07–1.34) 0.0015 
Sex match   <0.0001 
 Donor male/recipient male 2,662 1.00  
 Donor male/recipient female 1,720 0.94 (0.83–1.08) 0.3982 
 Donor female/recipient male 1,629 1.43 (1.27–1.62) <0.0001 
 Donor female/recipient female 1,472 1.03 (0.90–1.18) 0.7114 
Graft source 
 Bone marrow 5,354 1.00  
 Mobilized blood stem cells 2,129 1.24 (1.10–1.40) 0.0004 
NumberRR (95% CI)P
Presence of cGVHD 
 No cGVHD 3,936 1.00  
 cGVHD 3,547 1.56 (1.41–1.73) <0.0001 
Acute GVHD history   <0.0001 
 None 4,051 1.00  
 aGVHD in first 29 days 2,446 1.11 (0.93–1.24) 0.0716 
 aGVHD after day 29 986 1.46 (1.28–1.67) <0.0001 
Disease   <0.0001 
 AML 2,541 1.00  
 ALL 1,798 1.26 (1.10–1.45) 0.0008 
 CML 2,498 0.73 (0.64–0.83) <0.0001 
 MDS 652 0.84 (0.70–1.00) 0.0521 
Disease status at HCT   <0.0001 
 Early 4,661 1.00  
 Intermediate 1,823 1.25 (1.10–1.41) <0.0001 
 Advanced 999 1.83 (1.59–2.11) 0.0005 
HLA match   0.0191 
 Sibling donor 1,861 1.00  
 Well-matched unrelated 1,412 1.18 (1.03–1.36) 0.0199 
 Partially matched unrelated 1,369 1.12 (0.92–1.29) 0.1148 
 Poorly matched unrelated 841 1.26 (1.07–1.49) 0.0049 
TBI in the conditioning regimen 
 No 3,186 1.00  
 Yes 4,292 1.20 (1.07–1.34) 0.0015 
Sex match   <0.0001 
 Donor male/recipient male 2,662 1.00  
 Donor male/recipient female 1,720 0.94 (0.83–1.08) 0.3982 
 Donor female/recipient male 1,629 1.43 (1.27–1.62) <0.0001 
 Donor female/recipient female 1,472 1.03 (0.90–1.18) 0.7114 
Graft source 
 Bone marrow 5,354 1.00  
 Mobilized blood stem cells 2,129 1.24 (1.10–1.40) 0.0004 

NOTE: Factors tested, but not listed in table: Recipient age, sex, donor parity, donor-recipient CMV serology, use of ATG, GVHD prophylaxis, aGVHD grade, year of transplant, platelet count at cGVHD diagnosis, serum bilirubin at cGVHD diagnosis, type of cGVHD onset, KPS/L at cGVHD diagnosis, severity of cGVHD at 1 year after transplant, organ involvement of cGVHD at 1 year after transplant

The presence of cGVHD was associated with a higher risk of overall mortality for all diseases (RR, 1.56; 95% CI, 1.41–1.73; P < 0.0001; Fig. 1D, Table 5). Patients undergoing HCT for CML and MDS had a lower risk of overall mortality as compared with AML, whereas patients undergoing HCT for ALL had the highest risk of overall mortality. Other factors associated with a higher overall mortality risk were aGVHD after the first 29 days of HCT, intermediate, and advanced disease status at HCT, transplant from a well-matched URD or a poorly matched URD (as compared with HLA-identical sibling donor), use of TBI, transplant from a female donor to a male recipient (as compared with male donor to a male recipient), and use of PBSC as the graft type (Supplementary Table S1).

Compared with patients without cGVHD, the presence of cGVHD was associated with worse DFS in patients with AML (RR, 1.40; 95% CI, 1.20–1.63; P < 0.0001), ALL (RR, 1.46; 95% CI, 1.22–1.75; P < 0.0001), and MDS (RR, 1.58; 95% CI, 1.16–2.14; P < 0.0001) whereas the presence of cGVHD in CML did not significantly affect DFS (P = 0.4).

Chronic GVHD characteristics and impact on relapse, TRM, and survival

Because the presence of cGVHD reduced the risk of relapse only in patients with CML compared with those without cGVHD, we evaluated cGVHD-related variables associated with protection against relapse in CML. All types of cGVHD (i.e., mild, moderate, or severe; progressive, interrupted, or de novo), and any site of cGVHD involvement (i.e., skin or liver) provided protection against late relapse when compared with no cGVHD (Supplementary Table S2, CML patients with and without cGVHD). Pairwise comparisons performed to evaluate the impact of each subtype of cGVHD against each other (for example, the impact of presence of skin vs. no skin cGVHD, or severe cGVHD vs. moderate cGVHD or mild cGVHD) showed that none provided better protection. This finding was similar when the analysis on cGVHD-specific variables was limited only to patients with cGVHD; in patients with CML, any site or type of cGVHD provided protection against relapse, whereas patients with in AML, ALL, and MDS, the presence of cGVHD at any site or type did not provide protection against relapse.

In patients with CML, cGVHD characteristics associated with higher TRM and lower OS were the presence of moderate or severe cGVHD, skin involvement, lower platelet count, and KPS at cGVHD diagnosis. Moderate or severe cGVHD, lower platelet count, and KPS were associated with lower DFS (Supplementary Table S3).

In patients with AML, ALL, and MDS, moderate or severe cGVHD, low KPS at cGVHD diagnosis, and liver or gastrointestinal or hematologic involvement were associated with higher TRM. Higher cGVHD severity, and gastrointestinal, liver and hematologic involvement were also associated with lower OS. Moderate or severe cGVHD, gastrointestinal or genitourinary (GU) or liver or hematologic involvement were associated with lower DFS (Supplementary Table S4).

Allogeneic HCT is an effective immunotherapy for hematologic malignancies, which is mediated by GVL effect. Clinical evidence implicates acute and chronic GVHD in enhancing malignancy control and is best demonstrated in patients with acute leukemia and CML after MAC (2, 4, 6, 7, 15). GVHD is also a major complication of allogeneic HCT substantially contributing to overall TRM but its prevention or treatment with systemic immunosuppression may have harmful effects on GVL. Currently, there are no established clinical strategies to guide the intensity of systemic immunosuppressive therapy for GVHD based on presumed leukemia relapse risk. It would be useful to know more precisely in which setting GVHD is most beneficial for leukemia control versus those which predominantly increase TRM. Because occurrence and effects of acute and chronic GVHD overlap during the first year after transplant, it is difficult to study antitumor effects attributed specifically to cGVHD in the proximity of acute GVHD and therapy-related systemic immunosuppression. We focused here on assessing cGVHD effects isolated from acute GVHD by studying only patients who were alive and free of disease at 1 year after HCT, when acute GVHD effects were sufficiently distant.

This study demonstrates the protective effect of cGVHD on late relapse only in patients with CML. A protective effect of cGVHD against late relapse was not seen in AML, ALL, or MDS. Data in this study also confirm that the presence of cGVHD is associated with significantly higher TRM and worse OS across all diseases studied (9).

Several retrospective studies have demonstrated that cGVHD decreases relapses in AML, ALL, and especially in CML (2, 4, 6, 16). These studies evaluated antileukemia effects from time of transplant to relapse and did not account for possible confounding effects by acute GVHD. The lack of impact of cGVHD on late relapse in acute leukemia and MDS in this study may be due to differences in biology and susceptibility to immune responses in each disease.

Relapse of leukemia following allo-HCT could be due to failure of GVL or immune clonal escape of the leukemia cells, including downregulation of MHC class I and II antigens, associated with a decreased ability to stimulate allo-geneic–proliferative T-cell responses, and decreased susceptibility to lysis by cytotoxic T lymphocytes or natural killer cells. Because GVL requires time to become fully established after immune reconstitution, the degree to which GVL can control leukemia after allo-HCT may depend on the growth kinetics of the leukemia. Thus, patients with CML compared with patients with acute leukemia may be more susceptible to a durable GVL effect because of their slower pace of proliferation.

The majority of leukemia relapses occur within the first year after HCT. Similar risk of relapse has been observed using URD transplant or HLA-identical sibling donors in AML, ALL, and CML (15). In this study, late relapses occurred in all leukemia types and a beneficial cGVHD effect was detected only in those with CML. Differential susceptibility to GVL effects has been observed after administration of donor leukocyte infusion, which is more effective in relapsed CML than other hematologic malignancies (17–20).

Despite the expectation of protective effects of cGVHD in all diseases, the 5-year incidence of relapse was only lower in CML. For this reason, we analyzed whether any of the cGVHD-specific clinical characteristics or organ manifestations, showed a more dominant association with relapse in patients with CML, as such findings could lead toward a better understanding of GVL mechanisms dependent on cGVHD. We found that all sites and types of cGVHD involvement equally affected relapse, suggesting that patients with CML with minimal cGVHD clinical manifestation may benefit from its antitumor effect.

This study has several limitations. Chronic GVHD was not classified in the study per NIH consensus criteria and the data analysis used retrospective design. Specific treatment information for cGVHD is not ascertained and variability in treatment modalities for acute and chronic GVHD during the first year after HCT may have affected the incidence of late relapse. However, any such effects should be balanced out by the very large number of patients in this study cohort. This study was selected to evaluate a specific question: impact of cGVHD on late relapse (after 1 year after HCT). To classify patients accurately and determine cutoff points for analysis, we evaluated time to onset of cGVHD. Most patients (>95%) developed cGVHD within 1 year of HCT. Hence, we chose this time point for evaluation. Patients who developed cGVHD after 1 year and were eligible to be included in the dataset were classified as patients with non-cGVHD. This may also have affected the analysis. In addition, patients who developed cGVHD during the first year after allogeneic HCT and were successfully treated for their cGVHD may have been reported to the registry as not having active cGVHD. This would be quite unlikely, considering the usually slowly receding nature of cGVHD, and the standard procedure of 12 months registry data collection, which requires reporting events during the whole previous observation period. Data on the use of tyrosine kinase inhibitors (TKI) before allogeneic HCT were not available in patients with CML; however, during this study period of 1995 to 2004, only a minority of patients could have had prior TKI exposure.

This report findings only applies to cGVHD effects after MAC. The more recently expanded use of reduced intensity conditioning (RIC) or truly non-MAC regimens has shifted some of the burden of tumor cell kill after allogeneic HCT from the conditioning regimen to the immune-mediated GVL effects (18). Weisdorf and colleagues (21) recently investigated the effects of acute and chronic GVHD on late relapse after RIC and MAC conditionings regimens in patients with AML and MDS. Similar to this study, in patients with AML and MDS following MAC they found the risk of late relapse not significantly affected by cGVHD. However, following RIC regimens, in patients who had both acute and cGVHD late relapse rates were significantly lower. Baron and colleagues (22) also evaluated the GVL effects of cGVHD in patients with AML that underwent RIC allo-HCT. In a landmark analysis of patients who were leukemia free at 18 months after HCT, patients with cGVHD before the landmark day had a lower relapse rate than those without cGVHD. These data combined, demonstrate a differential effect of cGVHD on late relapse based on the type of conditioning used in allo-HCT.

In conclusion, this study suggest that cGVHD impact on late relapse of leukemia after MAC HCT is not clinically relevant in AML, ALL, and MDS and the beneficial effects on late relapse are confined to patients with CML. The potentially positive impact of the GVL effects on survival after MAC HCT are blunted by a higher cGVHD-related mortality resulting with higher TRM and lower OS for all studied diseases. These data may have practical clinical implications, as developing more aggressive strategies to prevent and treat cGVHD may be justified for hematologic malignancy patients after HCT with MAC regimens. Future studies aiming to advance the understanding of the cGVHD role in controlling hematologic malignancy should be done prospectively in contemporary patient cohorts and incorporating high level of detail on cGVHD clinical data collection.

D. Weisdorf reports receiving commercial research grants from Alexion and Tobira. No potential conflicts of interest were disclosed by the other authors.

The views expressed in this article do not reflect the official policy or position of the NIH, the Department of the Navy, the Department of Defense, or any other agency of the U.S. Government.

Conception and design: M. Boyiadzis, M. Arora, A. Urbano-Ispizua, B.J. Bolwell, J.-Y.Y. Cahn, C.S. Cutler, M.E. Flowers, R.P. Gale, L.M. Isola, D.A. Jacobsohn, M.M. Horowitz, S.Z. Pavletic

Development of methodology: M. Boyiadzis, M. Arora, B.J. Bolwell, L.M. Isola, T.R. Klumpp, S.Z. Pavletic

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): M. Arora, R. Herzig, L.M. Isola, H.C. Schouten, L.F. Verdonck, J.R. Wingard, M.M. Horowitz, S.Z. Pavletic

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M. Boyiadzis, M. Arora, J.P. Klein, A. Hassebroek, M. Hemmer, J.H. Antin, B.J. Bolwell, J.-Y.Y. Cahn, C.S. Cutler, M.E. Flowers, R.P. Gale, R. Herzig, L.M. Isola, M.H. Jagasia, T.R. Klumpp, S.R. Spellman, H.C. Schouten, L.F. Verdonck, D.J. Weisdorf, M.M. Horowitz, S.Z. Pavletic

Writing, review, and/or revision of the manuscript: M. Boyiadzis, M. Arora, J.P. Klein, A. Urbano-Ispizua, J.H. Antin, B.J. Bolwell, J.-Y.Y. Cahn, M.S. Cairo, C.S. Cutler, M.E. Flowers, R.P. Gale, R. Herzig, L.M. Isola, D.A. Jacobsohn, M.H. Jagasia, T.R. Klumpp, S.J. Lee, E.W. Petersdorf, S. Santarone, S.R. Spellman, H.C. Schouten, L.F. Verdonck, J.R. Wingard, D.J. Weisdorf, M.M. Horowitz, S.Z. Pavletic

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): M. Arora, A. Hassebroek, M. Hemmer, S.Z. Pavletic

Study supervision: M. Boyiadzis, M. Arora, A. Urbano-Ispizua, C.S. Cutler, S.Z. Pavletic

The CIBMTR is supported by Public Health Service grant/cooperative agreement U24-CA76518 from the National Cancer Institute (NCI), the National Heart, Lung, and Blood Institute (NHLBI) and the National Institute of Allergy and Infectious Diseases (NIAID); a grant/cooperative agreement 5U01HL069294 from NHLBI and NCI; a contract HHSH234200637015C with Health Resources and Services Administration (HRSA/DHHS); two grants N00014-06-1-0704 and N00014-08-1-0058 from the Office of Naval Research; and grants from Allos, Inc.; Amgen, Inc.; Angioblast; Anonymous donation to the Medical College of Wisconsin; Ariad; Be the Match Foundation; Blue Cross and Blue Shield Association; Buchanan Family Foundation; CaridianBCT; Celgene Corporation; CellGenix, GmbH; Children's Leukemia Research Association; Fresenius-Biotech North America, Inc.; Gamida Cell Teva Joint Venture Ltd.; Genentech, Inc.; Genzyme Corporation; GlaxoSmithKline; HistoGenetics, Inc.; Kiadis Pharma; The Leukemia & Lymphoma Society; The Medical College of Wisconsin; Merck & Co, Inc.; Millennium: The Takeda Oncology Co.; Milliman USA, Inc.; Miltenyi Biotec, Inc.; National Marrow Donor Program; Optum Healthcare Solutions, Inc.; Osiris Therapeutics, Inc.; Otsuka America Pharmaceutical, Inc.; RemedyMD; Sanofi; Seattle Genetics; Sigma-Tau Pharmaceuticals; Soligenix, Inc.; StemCyte, A Global Cord Blood Therapeutics Co.; Stemsoft Software, Inc.; Swedish Orphan Biovitrum; Tarix Pharmaceuticals; Teva Neuroscience, Inc.; THERAKOS, Inc.; and Wellpoint, Inc.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Passweg
JR
,
Baldomero
H
,
Gratwohl
A
,
Bregni
M
,
Cesaro
S
,
Dreger
P
, et al
The EBMT activity survey: 1990–2010
.
Bone Marrow Transplant
2012
;
47
:
906
23
.
2.
Weiden
PL
,
Sullivan
KM
,
Flournoy
N
,
Storb
R
,
Thomas
ED
. 
Antileukemic effect of chronic graft-versus-host disease: contribution to improved survival after allogeneic marrow transplantation
.
N Engl J Med
1981
;
304
:
1529
33
.
3.
Weiden
PL
,
Flournoy
N
,
Thomas
ED
,
Prentice
R
,
Fefer
A
,
Buckner
CD
, et al
Antileukemic effect of graft-versus-host disease in human recipients of allogeneic-marrow grafts
.
N Engl J Med
1979
;
300
:
1068
73
.
4.
Horowitz
MM
,
Gale
RP
,
Sondel
PM
,
Goldman
JM
,
Kersey
J
,
Kolb
HJ
, et al
Graft-versus-leukemia reactions after bone marrow transplantation
.
Blood
1990
;
75
:
555
62
.
5.
Ferrara
JL
,
Levine
JE
,
Reddy
P
,
Holler
E
. 
Graft-versus-host disease
.
Lancet
2009
;
373
:
1550
61
.
6.
Sullivan
KM
,
Weiden
PL
,
Storb
R
,
Witherspoon
RP
,
Fefer
A
,
Fisher
L
, et al
Influence of acute and chronic graft-versus-host disease on relapse and survival after bone marrow transplantation from HLA-identical siblings as treatment of acute and chronic leukemia
.
Blood
1989
;
73
:
1720
8
.
7.
Pavletic
SZ
,
Kumar
S
,
Mohty
M
,
de Lima
M
,
Foran
JM
,
Pasquini
M
, et al
NCI First International Workshop on the biology, prevention, and treatment of relapse after allogeneic hematopoietic stem cell transplantation: report from the Committee on the epidemiology and natural history of relapse following allogeneic cell transplantation
.
Biol Blood Marrow Transplant
2010
;
16
:
871
90
.
8.
Bierman
PJ
,
Sweetenham
JW
,
Loberiza
FR
 Jr
,
Taghipour
G
,
Lazarus
HM
,
Rizzo
JD
, et al
Syngeneic hematopoietic stem-cell transplantation for non-Hodgkin's lymphoma: a comparison with allogeneic and autologous transplantation–The Lymphoma Working Committee of the International Bone Marrow Transplant Registry and the European Group for Blood and Marrow Transplantation
.
J Clin Oncol
2003
;
21
:
3744
53
.
9.
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
14
.
10.
Weisdorf
D
,
Spellman
S
,
Haagenson
M
,
Horowitz
M
,
Lee
S
,
Anasetti
C
, et al
Classification of HLA-matching for retrospective analysis of unrelated donor transplantation: revised definitions to predict survival
.
Biol Blood Marrow Transplant
2008
;
14
:
748
58
.
11.
Bacigalupo
A
,
Lamparelli
T
,
Barisione
G
,
Bruzzi
P
,
Guidi
S
,
Alessandrino
PE
, et al
Thymoglobulin prevents chronic graft-versus-host disease, chronic lung dysfunction, and late transplant-related mortality: long-term follow-up of a randomized trial in patients undergoing unrelated donor transplantation
.
Biol Blood Marrow Transplant
2006
;
12
:
560
5
.
12.
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
17
.
13.
Filipovich
AH
,
Weisdorf
D
,
Pavletic
S
,
Socie
G
,
Wingard
JR
,
Lee
SJ
, et al
National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report
.
Biol Blood Marrow Transplant
2005
;
11
:
945
56
.
14.
Gooley
TA
,
Leisenring
W
,
Crowley
J
,
Storer
BE
. 
Estimation of failure probabilities in the presence of competing risks: new representations of old estimators
.
Stat Med
1999
;
18
:
695
706
.
15.
Ringden
O
,
Pavletic
SZ
,
Anasetti
C
,
Barrett
AJ
,
Wang
T
,
Wang
D
, et al
The graft-versus-leukemia effect using matched unrelated donors is not superior to HLA-identical siblings for hematopoietic stem cell transplantation
.
Blood
2009
;
113
:
3110
8
.
16.
Kolb
HJ
,
Mittermuller
J
,
Clemm
C
,
Holler
E
,
Ledderose
G
,
Brehm
G
, et al
Donor leukocyte transfusions for treatment of recurrent chronic myelogenous leukemia in marrow transplant patients
.
Blood
1990
;
76
:
2462
5
.
17.
Collins
RH
 Jr
,
Shpilberg
O
,
Drobyski
WR
,
Porter
DL
,
Giralt
S
,
Champlin
R
, et al
Donor leukocyte infusions in 140 patients with relapsed malignancy after allogeneic bone marrow transplantation
.
J Clin Oncol
1997
;
15
:
433
44
.
18.
Porter
DL
. 
Allogeneic immunotherapy to optimize the graft-versus-tumor effect: concepts and controversies
.
Hematol Am Soc Hematol Educ Program
2011
;
2011
:
292
8
.
19.
Porter
DL
,
Collins
RH
 Jr
,
Hardy
C
,
Kernan
NA
,
Drobyski
WR
,
Giralt
S
, et al
Treatment of relapsed leukemia after unrelated donor marrow transplantation with unrelated donor leukocyte infusions
.
Blood
2000
;
95
:
1214
21
.
20.
Simula
MP
,
Marktel
S
,
Fozza
C
,
Kaeda
J
,
Szydlo
RM
,
Nadal
E
, et al
Response to donor lymphocyte infusions for chronic myeloid leukemia is dose-dependent: the importance of escalating the cell dose to maximize therapeutic efficacy
.
Leukemia
2007
;
21
:
943
8
.
21.
Weisdorf
D
,
Zhang
MJ
,
Arora
M
,
Horowitz
MM
,
Rizzo
JD
,
Eapen
M
. 
Graft-versus-host disease induced graft-versus-leukemia effect: greater impact on relapse and disease-free survival after reduced intensity conditioning
.
Biol Blood Marrow Transplant
2012
;
18
:
1727
33
.
22.
Baron
F
,
Labopin
M
,
Niederwieser
D
,
Vigouroux
S
,
Cornelissen
JJ
,
Malm
C
, et al
Impact of graft-versus-host disease after reduced-intensity conditioning allogeneic stem cell transplantation for acute myeloid leukemia: a report from the Acute Leukemia Working Party of the European group for blood and marrow transplantation
.
Leukemia
2012
;
26
:
2462
8
.