- Email: [email protected]
Current Graft-versus-Host Disease–Free, Relapse-Free Survival: A Dynamic Endpoint to Better Define Efficacy after Allogenic Transplant
Accepted Manuscript Title: Current GVHD-Free, Relapse-Free Survival – a Dynamic Endpoint to Better Define Efficacy Following Allogenic Transplant Author: Scott R. Solomon, Connie Sizemore, Xu Zhang, Michelle Ridgeway, Melhem Solh, Lawrence E. Morris, H. Kent Holland, Asad Bashey PII: DOI: Reference:
S1083-8791(17)30365-8 http://dx.doi.org/doi: 10.1016/j.bbmt.2017.02.022 YBBMT 54624
To appear in:
Biology of Blood and Marrow Transplantation
Received date: Accepted date:
2-2-2017 22-2-2017
Please cite this article as: Scott R. Solomon, Connie Sizemore, Xu Zhang, Michelle Ridgeway, Melhem Solh, Lawrence E. Morris, H. Kent Holland, Asad Bashey, Current GVHD-Free, Relapse-Free Survival – a Dynamic Endpoint to Better Define Efficacy Following Allogenic Transplant, Biology of Blood and Marrow Transplantation (2017), http://dx.doi.org/doi: 10.1016/j.bbmt.2017.02.022. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Current GVHD-Free, Relapse-Free Survival – a Dynamic Endpoint to Better Define Efficacy Following Allogenic Transplant Scott R. Solomon, MD1, Connie Sizemore, Pharm.D1*, Xu Zhang, PhD2*, Michelle Ridgeway1*, Melhem Solh, MD1, Lawrence E. Morris, MD1, H. Kent Holland, MD1 and Asad Bashey, MD, PhD1 1
The Blood and Marrow Transplant Program at Northside Hospital, Atlanta, GA; 2Center for Biostatistics, University of Mississippi Medical Center, Jackson, MS Corresponding author:
Scott R Solomon 5670 Peachtree Dunwoody Road NE Suite 1000 10th floor Atlanta, GA 30342 Phone# 404-255-1930 Fax# 404-459-8510 Email: [email protected]
Short Title: CGRFS defines efficacy following allogeneic transplant Conflict of interest statement: There are no competing financial interests in relation to the work described. Text word count: 2638 Abstract word count: 253 Number of Figures: 4 Number of Tables: 2 Number of References: 28
1 Page 1 of 19
HIGHLIGHTS
CGRFS is a novel composite endpoint which attempts to incorporate the most important elements of transplant success. It acknowledges the important impact that clinically-significant ongoing GVHD has on the quality of life of long-term survivors
ABSTRACT An accurate measure of allogeneic transplant efficacy should take into account quality-of-life issues associated with graft-versus-host disease (GVHD). However unlike death and relapse, GVHD morbidity is temporary in many patients, and this fact must be reflected in such an outcome measure. Therefore, we have defined a new composite endpoint, called current GVHD-free, relapse-free survival (CGRFS), which is the probability, at any time post-transplant, of being alive, in remission, and without clinically significant chronic GVHD, defined as moderate-severe by the NIH consensus criteria. Chronic GVHD is considered a dynamic event, which can resolve once manifestations are quiescent and systemic immunosuppression discontinued. CGRFS is achieved through linear combination of relevant KaplanMeier estimates. We evaluated 422 consecutive patients receiving an allogeneic transplant at a single institution between January 2010 and July 2015. With a median follow-up of 36 months, estimated 3-yr overall and disease-free survival was 60% and 54%, respectively. Conventionally defined GVHD-free, relapse-free survival (GRFS) at 1, 2, 3 and 4 years was 33%, 26%, 23%, and 22% respectively. In contrast, the corresponding rates of CGRFS was 45%, 46%, 47% and 49%, respectively. Patients living with active moderate-severe chronic GVHD decreased over time, quantitated at 23%, 14%, 7% and 4% respectively at 1, 2, 3 and 4 years post-transplant. Whereas only approximately a quarter of patients achieve transplant success as defined by conventional GRFS, nearly half of patients, by CGRFS, are considered cured without the morbidity of ongoing GVHD. We propose that CGRFS may represent a more dynamic and accurate estimate of long-term transplant effectiveness. Keywords: Graft-versus-host disease GVHD GRFS Allogeneic Hematopoietic stem cell transplantation
2 Page 2 of 19
INTRODUCTION Allogeneic hematopoietic cell transplantation (alloHCT) remains the best curative option to treat many patients with high-risk hematological malignancies.1 This is due to the combined effects of the intensive preparative regimen and the potent graft-versus-malignancy effect of allogeneic immune cells transferred in the allograft. However, disease recurrence and non-relapse mortality (NRM) remain major obstacles to alloHCT success. Even in patients that survive transplant without relapse, long-term complications, particularly chronic graft-versus-host disease (GVHD), may impose significant impacts on quality-of-life. Chronic GVHD is the most important cause of late morbidity and mortality after allogeneic stem cell transplantation, occurring in up to 60% to 70% of long-term survivors.2-5 With the increasing number of allogeneic transplants using PBSCs, mismatched and unrelated donors and the increasing age of transplant recipients, chronic GVHD will continue to be a serious challenge following allogeneic transplantation. Chronic GVHD requires therapy for many months and often years,6,7 and is the cause of death in up to one third of all long-term survivors after transplantation for leukemia.8 In addition to the effects on mortality, chronic GVHD can markedly reduce quality of life (QOL).9-12 The most commonly used endpoint to assess transplant success is disease-free survival (DFS), which reflects survival without disease relapse or progression. However, the main criticism of this endpoint is that it ignores the potential morbidity and QOL implications of chronic GVHD. Therefore, a composite endpoint that combines DFS and chronic GVHD would be desirable to more accurately reflect transplant success. Recently, a composite endpoint incorporating the occurrence of GVHD as an event has been developed, called GVHD-free, relapse-free survival (GRFS). 13,14 However, in all current iterations of GRFS, patients with resolved GVHD continue to be treated as a transplant failure. However, unlike death and usually disease recurrence, GVHD is a dynamic process which may resolve completely or partially after treatment. We have therefore developed a new method to generate GRFS, termed current GRFS (CGRFS), which we define as survival without relapse, and without active moderateto-severe chronic GVHD at the time of most recent assessment. We evaluated this endpoint on large subset of allogeneic transplants performed at a single center to help better define the meaning of transplant success. PATIENTS AND METHODS Objective and Definitions The objective of this single-institution retrospective analysis was to assess the success of alloHCT using a composite endpoint of CGRFS, which reflects the 3 Page 3 of 19
probability of being alive, disease-free, and without clinically significant chronic GVHD at any time point post-transplant. CGRFS events included moderate-to-severe chronic GVHD (2005 NIH consensus criteria)15 at the time of the most recent assessment, relapse/progression or death. Additional outcomes analyzed were overall survival (OS), DFS (survival without evidence of relapse/progression of the underlying malignancy after transplantation), and conventional GRFS13,14 (survival without evidence of grades III to IV acute GVHD, chronic GVHD requiring systemic treatment, relapse, or death). Maximum cumulative incidence of acute GVHD was assessed at 6 months after transplantation and classified as clinically significant (grades 2 to 4) or severe (grades 3 to 4).16 Chronic GVHD was classified as mild, moderate, or severe by 2005 National Institutes of Health consensus criteria.15 Acute and chronic GVHD was prospectively evaluated, graded, and documented by a single practitioner within the program. Study Population The institutional review board of Northside Hospital approved this study. This retrospective study includes four hundred and twenty two patients who underwent a first alloHCT for hematologic malignancy using an HLA-identical sibling donor (MRD, n = 125); ≥7/8 HLA-A, -B, -C, and -DRB1 allele matched volunteer unrelated donor (MUD, n = 165), or T cell–replete haploidentical graft using post-transplant cyclophosphamide (HID, n = 132) at our center were included in this analysis. The transplants were performed consecutively between January 2010 and June 2015. This time frame was chosen to allow a minimum of 12 months of post-transplant follow-up for surviving patients. Patients underwent conditioning using a variety of preparative regimens classified as myeloablative, reduced-intensity (RIC), and nonmyeloablative conditioning according to accepted criteria.17 Covariates Patient-, disease-, and transplant-related variables were prospectively documented and obtained for this analysis from our comprehensive institutional database. Clinical factors examined included age (<55 years, ≥55 years), sex, CMV status, year of transplant (2010 to 2012 or 2013 to 2015), diagnosis (acute myelogenous leukemia (AML), myelodysplastic syndrome (MDS)/myeloproliferative neoplasms (MPN)/chronic myelogenous leukemia (CML), acute lymphoblastic leukemia (ALL), non-Hodgkin lymphoma (NHL)/Hodgkin lymphoma (HL)/chronic lymphocytic leukemia (CLL), or other malignancy), conditioning intensity (nonmyeloablative, RIC, myeloablative), donor type (MRD, MUD, HID), graft source (bone marrow or PBSC), hematopoietic cell transplant comorbidity index (HCT-CI),18 and disease risk index (DRI)19. Statistical Analysis OS, DFS, and GRFS were estimated by the Kaplan-Meier method.20 Generation of the CGRFS curve was accomplished by linear combination of five Kaplan-Meier 4 Page 4 of 19
estimates to accommodate two episodes of chronic GVHD and their resolution.21 The model is based on nine health states that a patient may be in after transplant (Figure 1). The states are as follows: 0. Alive and in first remission. 1. Dead or relapsed/progressed before first chronic GVHD. 2. Alive, in remission, and with first episode of moderate-to-severe chronic GVHD. 3. Dead or relapsed/progressed in first chronic GVHD. 4. Alive, in remission, after resolution of first chronic GVHD and before second chronic GVHD (symptoms quiescent and systemic immunosuppression discontinued). 5. Dead or relapsed/progressed after resolution of first chronic GVHD and before second chronic GVHD. 6. Alive, in remission, and with recurrence of moderate-to-severe chronic GVHD (second episode) 7. Dead or relapsed/progressed in second episode of chronic GVHD. 8. Alive, in remission, after resolution of second chronic GVHD (symptoms quiescent and systemic immunosuppression discontinued). 9. Dead or relapsed/progressed after resolution of second GVHD. CGRFS is probability that one stays in state 0, state 4 or state 8; represented by the formula: C(t) = S1(t) + [S2(t) ─ S3(t)] + [S4(t) ─ S5(t)]. For S1(t), the event is death, relapse/progression or initial chronic GVHD. This is the usual chronic GVHD and relapse free survival function and is the chance that the patient is in state 0. For S2(t), the event is second chronic GVHD or any death, relapse/progression before second chronic GVHD. For S3(t), the event is resolution of first chronic GVHD or any death, relapse/progression before resolution of first chronic GVHD. The difference between S2(t) and S3(t), [S2(t) ─ S3(t)], is the probability of being in state 4. S4(t) is DFS with any death, relapse/progression as the events. For S5(t), the event is resolution of second chronic GVHD or any death, relapse/progression before resolution of second chronic GVHD. The difference, [S4(t) – S5(t)], is the probability of being in state 8. CGRFS is estimated by linear combination of five Kaplan-Meier estimates. A simulation-based supremum test in the time interval [0, 4] (years) was conducted to compare CGRFS between two independent samples.22 At year 4 about 20% of patients remained in the study. Testing in [0, 4] (years) could avoid large variability at the tail and generate stabilized result. In each test, 1000 realizations of process of difference in CGRFS were generated to approximate its distribution. The supremum was found for the absolute value of each realized process in [0, 4] (years). The p value for a two-sample test was obtained by finding the proportion of 1000 supremum values greater than the sample supremum. RESULTS Patient Characteristics 5 Page 5 of 19
In total, 422 adult patients were included in this analysis with a median age at transplant of 54 years (range, 18 to 77) (Table 1). Patients were transplanted for AML (38%), MDS/MPN/CML (27%), NHL/HL/CLL (20%), and ALL (13%), and 30% of patients were high-/very high-risk by DRI. Donor type was MRD, MUD, and HID in 30%, 39%, and 31% respectively. Regimen intensity was myeloablative in 50% of patients, and PBSC was the stem cell source in 78%. The median time of followup was 36 months (range, 12 to 78). OS, DFS, and conventional GRFS Estimated one-, two-, three- and four-year OS was 78%, 69%, 60% and 59% respectively; corresponding rates of DFS were 68%, 60% 54% and 52%. The major determinants of DFS were DRI and age. Estimated DFS was improved in patients with low (63%) or intermediate (55%) DRI vs. high (43%) or very high (24%) DRI (low/intermediate vs. high/very high, p<0.001), and in patients younger than 55 years of age (60%) vs. older patients (44%, p<0.001). Estimated one-, two-, threeand four-year conventional GRFS was 33%, 26%, 23% and 22% respectively. Current GVHD-free, relapse-free survival (CGRFS) The estimated one-, two-, three- and four-year CGRFS in this cohort was 45%, 46%, 47% and 49%, respectively; this corresponds to an absolute increase of 12%, 20%, 24% and 27% respectively when compared to conventional GRFS from the same time points (Figure 2a). This improvement in outcome is due a steady recovery from chronic GVHD over time (Figure 2b), as well the omission of grade 3-4 acute GVHD as an endpoint. The component events that make up the CGRFS endpoint (death, relapse, and modsevere chronic GVHD) are graphically depicted in Figure 3. At one year posttransplant, the components of death, relapse and chronic GVHD proportionally accounted for 40%, 18% and 42% respectively of patients defined as treatment failure by CGRFS. The proportion of patients meeting a CGRFS “event” because of chronic GVHD decreased successively in the first four years post-transplant, accounting for 42%, 26%, 13% and 8% respectively. This was due to the fact that a decreasing percentage of patients were living with active moderate-to-severe chronic GVHD over time, quantitated at 23%, 14%, 7% and 4% respectively at 1, 2, 3 and 4 years post-transplant. A summary of survival and incidence estimates can be found in Table 2. Effect of Patient-, Disease- and Transplant-Related Factors on CGRFS In univariate analysis, CGRFS was significantly correlated only with DRI (low/intermediate vs. high/very high, p<0.001) (figure 3), but not transplant type (MRD, MUD, haplo), age (<55, ≥55 years), HCT-CI (0-2, ≥3), graft source (BM, PBSC), regimen intensity (MA, RIC/NMA), donor/recipient sex mismatch, donor or recipient CMV status or year of transplant (Figure 4). In patients with low-, 6 Page 6 of 19
intermediate-, high- and very high-risk DRI, the estimated 3-year CGRFS was 56%, 50%, 36% and 33% respectively. DISCUSSION We evaluated 422 consecutive patients receiving a MRD, MUD, or HID alloHCT at a single institution in an attempt to develop a novel endpoint, which we have called CGRFS. This composite endpoint attempts to incorporate the most important elements of transplant success. It acknowledges the important impact that clinically-significant ongoing GVHD has on the QOL of long-term survivors of alloHCT. CGRFS is defined as survival without relapse, and without active moderateto-severe chronic GVHD, by NIH consensus criteria, at the time of most recent assessment. This differs significantly from the conventional definition of GRFS, which, in addition to death and relapse, treats grade 3-4 acute GVHD and immunosuppression-requiring chronic GVHD as “fixed events” defining treatment failure. The conventional definition of GRFS has significant shortcomings, most notably that GVHD is treated as a fixed event, similar to death or relapse. However, treating chronic GVHD as a fixed complication does not properly reflect the reality experienced by the majority of patients following alloHCT. In our series, patients living with active moderate-to-severe chronic GVHD decreased continuously following transplant, accounting for 23%, 14%, 7% and 4% of patients in the first four years post-transplant. This is critically important since once chronic GVHD resolves, a patient’s health status does not appear to differ significantly from those patients who never had chronic GVHD.11 Secondly, we believe that treating grade 34 acute GVHD as an “event” is unnecessary, as this event almost universally will either resolve with time or convert to either death or chronic GVHD, which are both captured as part of the CGRFS endpoint. Although severe acute GVHD can be associated with a significant mortality risk, it is unlikely to affect the long-term QOL of patients that recover from it and do not develop clinically significant chronic GVHD.23 Furthermore, it is now well recognized that chronic GVHD is extremely heterogeneous in its presentation. Some patients present with mild symptoms restricted to a few organs, while others can have extensive involvement lasting many years and causing severe medical, social, and QOL issues. Chronic GVHD has been much better characterized since the first NIH Consensus classification was developed in 2005 and prospectively validated.15,24-26 It is now better appreciated that only moderate-to-severe chronic GVHD, by the NIH criteria, is associated with inferior survival and worsened functional status; whereas neither is seen in patients with mild severity chronic GVHD.25,26 As a result, the CGRFS endpoint includes only patients with moderate-to-severe NIH severity chronic GVHD, rather than the requirement for systemic immunosuppression, as utilized by conventional GRFS. We hypothesize that CGRFS composite endpoint is a more accurate surrogate for 7 Page 7 of 19
transplant success, when compared with more traditional endpoints such as DFS or conventional GRFS. At one year post-transplant, estimated DFS, CGRFS, and conventional GRFS is 68%, 45%, and 33% respectively, whereas by four years posttransplant, these figures are 52%, 49%, and 22% respectively. Several lessons are apparent from these survival numbers. In the early years post-transplant, (1) DFS appears to be an inadequate reflection of transplant success as it fails to reflect the significant morbidity and QOL impact related to moderate-to-severe chronic GVHD; and (2) conventional GRFS appears to dramatically underestimate “transplant success.” Although the latter is apparent already in the first year post-transplant (12% differential between CGRFS and GRFS), it becomes a significantly greater issue with time (27% differential between CGRFS and GRFS by four years posttransplant). It’s also important to note that by four years post-transplant, CGRFS and DFS are nearly identical since only 4% of patients are living with active moderate-to-severe chronic GVHD at that time. If one accepts that CGRFS is a useful indicator of transplant success post alloHCT, it is noteworthy that this endpoint does not appear to be significantly impacted by a majority of recipient- or transplant-related factors, such as patient age, HCT-CI, donor type (MRD, MUD, HID), regimen intensity, or stem cell source. Only disease risk, as measured by DRI, had a significant impact on CGRFS. The fact that transplant success, as measured by CGRFS, appears largely independent of either donor type, stem cell source, or regimen intensity, suggests that these factors should perhaps not play a major role in the decision to offer alloHCT to a patient. This fact may be especially relevant for racial or ethnic minorities who are significantly less likely to find a suitable matched related or unrelated donor.27 A major limitation of this study was the restriction of patients to a single center. Further validation of the CGRFS endpoint will be required in patients transplanted from other institutions. In this regards, it is of great interest that the Center for International Blood and Marrow Transplant Research (CIBMTR) has recently revised their data collection forms so that significant chronic GVHD data points will be recorded such as maximum GVHD grade and the dates of systemic immunosuppression initiation and discontinuation. An additional limitation of this study is the lack of health-related QOL measurements as well as analyses of the economic burdens of GVHD management, which should be incorporated in future studies analyzing composite endpoints such as CGRFS. Finally, it should be stipulated that survivors of alloHCT have an increased incidence of chronic health problems28 that can occur even in the absence of GVHD, and such morbidity will not be reflected in the CGRFS endpoint. In spite of the above limitations, we suggest that the composite endpoint of CGRFS represents a more accurate estimate of transplant success following alloHCT. It represents survival without disease relapse/progression and without continuing morbidity related to ongoing moderate-to-severe chronic GVHD. This endpoint yields significantly more information that the simpler measurements of DFS or OS. Furthermore, the treatment of chronic GVHD as a dynamic rather than static event 8 Page 8 of 19
creates a more accurate picture of the experience of patients post-transplant. We hypothesize that CGRFS will be a more useful measurement of transplant success when assessing and comparing various alloHCT strategies. ACKNOWLEDGEMENTS The authors thank Stacey Brown and Katelin Jackson for excellent assistance in performing data query. AUTHORSHIP CONTRIBUTIONS S.R.S. and X.Z. designed the research; S.R.S., M.S., L.E.M., H.K.H, and A.B. treated patients; C.S. and M.R. collected the data; S.R.S. analyzed and interpreted data; X.Z. performed the statistical analysis; S.R.S. wrote the manuscript; and all authors revised and approved the final version of the manuscript. CONFLICT OF INTEREST DISCLOSURE There are no competing financial interests in relation to the work described.
9 Page 9 of 19
REFERENCES 1. Thomas ED BK, Forman SJ, Appelbaum FR: Thomas' Hematopoietic Cell Transplantation (ed 4), John Wiley & Sons, 2004 2. Abou-Mourad YR, Lau BC, Barnett MJ, et al: Long-term outcome after allo-SCT: close follow-up on a large cohort treated with myeloablative regimens. Bone Marrow Transplantation 45:295-302, 2010 3. Bhatia S, Francisco L, Carter A, et al: Late mortality after allogeneic hematopoietic cell transplantation and functional status of long-term survivors: report from the Bone Marrow Transplant Survivor Study. Blood 110:3784-3792, 2007 4. Schmitz N, Eapen M, Horowitz MM, et al: Long-term outcome of patients given transplants of mobilized blood or bone marrow: a report from the International Bone Marrow Transplant Registry and the European Group for Blood and Marrow Transplantation. Blood 108:4288-4290, 2006 5. Higman MA, Vogelsang GB: Chronic graft versus host disease. Br J Haematol 125:435-54, 2004 6. Koc S, Leisenring W, Flowers ME, et al: Therapy for chronic graftversus-host disease: a randomized trial comparing cyclosporine plus prednisone versus prednisone alone. Blood 100:48-51, 2002 7. Vogelsang GB: How I treat chronic graft-versus-host disease. Blood 97:1196-201, 2001 8. Ratanatharathorn V, Ayash L, Lazarus HM, et al: Chronic graft-versushost disease: clinical manifestation and therapy. Bone Marrow Transplant 28:121-9, 2001 9. Sutherland HJ, Fyles GM, Adams G, et al: Quality of life following bone marrow transplantation: a comparison of patient reports with population norms. Bone Marrow Transplant 19:1129-36, 1997 10. Chiodi S, Spinelli S, Ravera G, et al: Quality of life in 244 recipients of allogeneic bone marrow transplantation. British Journal of Haematology 110:614619, 2000 11. Fraser CJ, Bhatia S, Ness K, et al: Impact of chronic graft-versus-host disease on the health status of hernatopoietic cell transplantation survivors: a report from the Bone Marrow Transplant Survivor Study. Blood 108:2867-2873, 2006 12. Kiss TL, Abdolell M, Jamal N, et al: Long-term medical outcomes and quality-of-life assessment of patients with chronic myeloid leukemia followed at least 10 years after allogeneic bone marrow transplantation. Journal of Clinical Oncology 20:2334-2343, 2002 13. Holtan SG, DeFor TE, Lazaryan A, et al: Composite end point of graftversus-host disease-free, relapse-free survival after allogeneic hematopoietic cell transplantation. Blood 125:1333-8, 2015 14. Solh M, Zhang X, Connor K, et al: Factors Predicting Graft-versus-Host Disease-Free, Relapse-Free Survival after Allogeneic Hematopoietic Cell 10 Page 10 of 19
Transplantation: Multivariable Analysis from a Single Center. Biol Blood Marrow Transplant 22:1403-9, 2016 15. Filipovich AH, Weisdorf D, Pavletic S, et al: National Institutes of Health consensus development project on criteria for clinical trials in chronic graftversus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant 11:945-56, 2005 16. Gratwohl A, Hermans J, Apperley J, et al: Acute graft-versus-host disease: grade and outcome in patients with chronic myelogenous leukemia. Working Party Chronic Leukemia of the European Group for Blood and Marrow Transplantation. Blood 86:813-8, 1995 17. Giralt S, Ballen K, Rizzo D, et al: Reduced-intensity conditioning regimen workshop: defining the dose spectrum. Report of a workshop convened by the center for international blood and marrow transplant research. Biol Blood Marrow Transplant 15:367-9, 2009 18. Sorror ML, Maris MB, Storb R, et al: Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood 106:2912-9, 2005 19. Armand P, Kim HT, Logan BR, et al: Validation and refinement of the Disease Risk Index for allogeneic stem cell transplantation. Blood 123:3664-71, 2014 20. Kaplan E, Meier P: Nonparametric estimation from incomplete observations. . Journal of the American Statistical Association 84:360-372, 1958 21. Klein JP, Keiding N, Shu Y, et al: Summary curves for patients transplanted for chronic myeloid leukaemia salvaged by a donor lymphocyte infusion: the current leukaemia-free survival curve. Br J Haematol 109:148-52, 2000 22. Parzen MI, Wei LJ, Ying Z: Simultaneous Confidence Intervals for the Difference of Two Survival Functions. Scandinavian Journal of Statistics 24:309-314, 1997 23. Lee SJ, Kim HT, Ho VT, et al: Quality of life associated with acute and chronic graft-versus-host disease. Bone Marrow Transplant 38:305-10, 2006 24. Jagasia MH, Greinix HT, Arora M, et al: National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graftversus-Host Disease: I. The 2014 Diagnosis and Staging Working Group Report. Biology of Blood and Marrow Transplantation 21:389-401.e1, 2015 25. Moon JH, Sohn SK, Lambie A, et al: Validation of National Institutes of Health Global Scoring System for Chronic Graft-Versus-Host Disease (GVHD) According to Overall and GVHD-Specific Survival. Biology of Blood and Marrow Transplantation 20:556-563, 2014 26. Pidala J, Kurland B, Chai X, et al: Patient-reported quality of life is associated with severity of chronic graft-versus-host disease as measured by NIH criteria: report on baseline data from the Chronic GVHD Consortium. Blood 117:4651-4657, 2011 27. Gragert L, Eapen M, Williams E, et al: HLA match likelihoods for hematopoietic stem-cell grafts in the U.S. registry. N Engl J Med 371:339-48, 2014 28. Sun C-L, Kersey JH, Francisco L, et al: Burden of Morbidity in 10+ Year Survivors of Hematopoietic Cell Transplantation: Report from the Bone Marrow 11 Page 11 of 19
Transplantation Survivor Study. Biology of Blood and Marrow Transplantation 19:1073-1080
12 Page 12 of 19
Fig 1. Possible transitions in multistate model for CGRFS. Fig 2. (A) Comparison of CGRFS to Conventional GRFS, DFS, and OS, (B) Gain in CGRFS Due to the Treatment of chronic GVHD as a Dynamic Event. Fig 3. Component Events of the CGRFS endpoint. Fig 4. CGRFS according to (A) DRI, (B) Age, (C) HCT-CI, (D) Donor Type, (E) Conditioning Intensity, and (F) Stem Cell Source.
13 Page 13 of 19
Table 1. Patient, diagnosis, and transplant characteristics ________________________________________________________________________ Patients (n=422) - N (%) ________________________________________________________________________ Age (years) Donor Type MRD 125 (30%) Median (range) 54 (18-77) MUD 165 (39%) Gender Male 235 (56%) HID 132 (31%) Female 187 (44%) Regimen Intensity F → M 67 (16%) RIC/NMA 209 (50%) HCT-CI MA 213 (50%) 0-2 228 (54%) Stem Cell Source ≥3 194 (46%) BM 93 (22%) Diagnosis PBSC 329 (78%) ALL 55 (13%) Year of Transplant 2010 65 (16%) AML 162 (38%) MDS/MPN/CML 112 (27%) 2011 68 (16%) 2012 69 (16%) NHL/HL/CLL 83 (20%) Other 10 (2%) 2013 77 (18%) 2014 77 (18%) DRI Low 59 (14%) 2015 66 (16%) Intermediate 230 (55%) CMV High 107 (25%) Recipient Pos 284 (67%) Very High 22 (5%) Recipient Neg 138 (33%)
Abbreviations : F→M, female donor-to-male recipient; HCT-CI, hematopoietic cell transplantation-comorbidity index; ALL, acute lymphoblas tic leukemia; AML, acute myeloid leukemia; MDS, myelodysplas tic syndrome; MPN, myeloproliferative neoplasm; CML, chronic myelogenous leukemia; NHL, non-Hodgkins lymphoma; HL, Hodgkins lymphoma; CLL, chronic lymphocytic leukemia; DRI, disease ris k index; MRD, matched related donor; MUD, matched unrelated donor; HID, haploidentical donor; RIC, reduced intens ity conditioning; NMA, nonmyeloablative; MA, myeloablative; BM, bone marrow; PBSC, peripheral blood stem cells ; CMV, cytomegalovirus seros tatus .
14 Page 14 of 19
Table 2. Summary of Survival and Incidence Estimates 1
Post-Transplant Year 2 3
4
OS (%)
78
69
60
59
DFS (%)
68
60
54
52
CGRFS (%)
45
46
47
49
GRFS (%)
33
26
23
22
CGRFS-GRFS (%)
12
20
24
27
Moderate-severe cGVHD, relapse-free survival (%)
43
34
31
30
Recovery from cGVHD (%)
2
12
16
19
Active moderate-severe cGVHD (%)
23
14
7
4
15 Page 15 of 19
Figure 1
16 Page 16 of 19
Figure 2
A
B
17 Page 17 of 19
Figure 3
18 Page 18 of 19
Figure 4
A
B
C
D
E
F
19 Page 19 of 19
S1083-8791(17)30365-8 http://dx.doi.org/doi: 10.1016/j.bbmt.2017.02.022 YBBMT 54624
To appear in:
Biology of Blood and Marrow Transplantation
Received date: Accepted date:
2-2-2017 22-2-2017
Please cite this article as: Scott R. Solomon, Connie Sizemore, Xu Zhang, Michelle Ridgeway, Melhem Solh, Lawrence E. Morris, H. Kent Holland, Asad Bashey, Current GVHD-Free, Relapse-Free Survival – a Dynamic Endpoint to Better Define Efficacy Following Allogenic Transplant, Biology of Blood and Marrow Transplantation (2017), http://dx.doi.org/doi: 10.1016/j.bbmt.2017.02.022. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Current GVHD-Free, Relapse-Free Survival – a Dynamic Endpoint to Better Define Efficacy Following Allogenic Transplant Scott R. Solomon, MD1, Connie Sizemore, Pharm.D1*, Xu Zhang, PhD2*, Michelle Ridgeway1*, Melhem Solh, MD1, Lawrence E. Morris, MD1, H. Kent Holland, MD1 and Asad Bashey, MD, PhD1 1
The Blood and Marrow Transplant Program at Northside Hospital, Atlanta, GA; 2Center for Biostatistics, University of Mississippi Medical Center, Jackson, MS Corresponding author:
Scott R Solomon 5670 Peachtree Dunwoody Road NE Suite 1000 10th floor Atlanta, GA 30342 Phone# 404-255-1930 Fax# 404-459-8510 Email: [email protected]
Short Title: CGRFS defines efficacy following allogeneic transplant Conflict of interest statement: There are no competing financial interests in relation to the work described. Text word count: 2638 Abstract word count: 253 Number of Figures: 4 Number of Tables: 2 Number of References: 28
1 Page 1 of 19
HIGHLIGHTS
CGRFS is a novel composite endpoint which attempts to incorporate the most important elements of transplant success. It acknowledges the important impact that clinically-significant ongoing GVHD has on the quality of life of long-term survivors
ABSTRACT An accurate measure of allogeneic transplant efficacy should take into account quality-of-life issues associated with graft-versus-host disease (GVHD). However unlike death and relapse, GVHD morbidity is temporary in many patients, and this fact must be reflected in such an outcome measure. Therefore, we have defined a new composite endpoint, called current GVHD-free, relapse-free survival (CGRFS), which is the probability, at any time post-transplant, of being alive, in remission, and without clinically significant chronic GVHD, defined as moderate-severe by the NIH consensus criteria. Chronic GVHD is considered a dynamic event, which can resolve once manifestations are quiescent and systemic immunosuppression discontinued. CGRFS is achieved through linear combination of relevant KaplanMeier estimates. We evaluated 422 consecutive patients receiving an allogeneic transplant at a single institution between January 2010 and July 2015. With a median follow-up of 36 months, estimated 3-yr overall and disease-free survival was 60% and 54%, respectively. Conventionally defined GVHD-free, relapse-free survival (GRFS) at 1, 2, 3 and 4 years was 33%, 26%, 23%, and 22% respectively. In contrast, the corresponding rates of CGRFS was 45%, 46%, 47% and 49%, respectively. Patients living with active moderate-severe chronic GVHD decreased over time, quantitated at 23%, 14%, 7% and 4% respectively at 1, 2, 3 and 4 years post-transplant. Whereas only approximately a quarter of patients achieve transplant success as defined by conventional GRFS, nearly half of patients, by CGRFS, are considered cured without the morbidity of ongoing GVHD. We propose that CGRFS may represent a more dynamic and accurate estimate of long-term transplant effectiveness. Keywords: Graft-versus-host disease GVHD GRFS Allogeneic Hematopoietic stem cell transplantation
2 Page 2 of 19
INTRODUCTION Allogeneic hematopoietic cell transplantation (alloHCT) remains the best curative option to treat many patients with high-risk hematological malignancies.1 This is due to the combined effects of the intensive preparative regimen and the potent graft-versus-malignancy effect of allogeneic immune cells transferred in the allograft. However, disease recurrence and non-relapse mortality (NRM) remain major obstacles to alloHCT success. Even in patients that survive transplant without relapse, long-term complications, particularly chronic graft-versus-host disease (GVHD), may impose significant impacts on quality-of-life. Chronic GVHD is the most important cause of late morbidity and mortality after allogeneic stem cell transplantation, occurring in up to 60% to 70% of long-term survivors.2-5 With the increasing number of allogeneic transplants using PBSCs, mismatched and unrelated donors and the increasing age of transplant recipients, chronic GVHD will continue to be a serious challenge following allogeneic transplantation. Chronic GVHD requires therapy for many months and often years,6,7 and is the cause of death in up to one third of all long-term survivors after transplantation for leukemia.8 In addition to the effects on mortality, chronic GVHD can markedly reduce quality of life (QOL).9-12 The most commonly used endpoint to assess transplant success is disease-free survival (DFS), which reflects survival without disease relapse or progression. However, the main criticism of this endpoint is that it ignores the potential morbidity and QOL implications of chronic GVHD. Therefore, a composite endpoint that combines DFS and chronic GVHD would be desirable to more accurately reflect transplant success. Recently, a composite endpoint incorporating the occurrence of GVHD as an event has been developed, called GVHD-free, relapse-free survival (GRFS). 13,14 However, in all current iterations of GRFS, patients with resolved GVHD continue to be treated as a transplant failure. However, unlike death and usually disease recurrence, GVHD is a dynamic process which may resolve completely or partially after treatment. We have therefore developed a new method to generate GRFS, termed current GRFS (CGRFS), which we define as survival without relapse, and without active moderateto-severe chronic GVHD at the time of most recent assessment. We evaluated this endpoint on large subset of allogeneic transplants performed at a single center to help better define the meaning of transplant success. PATIENTS AND METHODS Objective and Definitions The objective of this single-institution retrospective analysis was to assess the success of alloHCT using a composite endpoint of CGRFS, which reflects the 3 Page 3 of 19
probability of being alive, disease-free, and without clinically significant chronic GVHD at any time point post-transplant. CGRFS events included moderate-to-severe chronic GVHD (2005 NIH consensus criteria)15 at the time of the most recent assessment, relapse/progression or death. Additional outcomes analyzed were overall survival (OS), DFS (survival without evidence of relapse/progression of the underlying malignancy after transplantation), and conventional GRFS13,14 (survival without evidence of grades III to IV acute GVHD, chronic GVHD requiring systemic treatment, relapse, or death). Maximum cumulative incidence of acute GVHD was assessed at 6 months after transplantation and classified as clinically significant (grades 2 to 4) or severe (grades 3 to 4).16 Chronic GVHD was classified as mild, moderate, or severe by 2005 National Institutes of Health consensus criteria.15 Acute and chronic GVHD was prospectively evaluated, graded, and documented by a single practitioner within the program. Study Population The institutional review board of Northside Hospital approved this study. This retrospective study includes four hundred and twenty two patients who underwent a first alloHCT for hematologic malignancy using an HLA-identical sibling donor (MRD, n = 125); ≥7/8 HLA-A, -B, -C, and -DRB1 allele matched volunteer unrelated donor (MUD, n = 165), or T cell–replete haploidentical graft using post-transplant cyclophosphamide (HID, n = 132) at our center were included in this analysis. The transplants were performed consecutively between January 2010 and June 2015. This time frame was chosen to allow a minimum of 12 months of post-transplant follow-up for surviving patients. Patients underwent conditioning using a variety of preparative regimens classified as myeloablative, reduced-intensity (RIC), and nonmyeloablative conditioning according to accepted criteria.17 Covariates Patient-, disease-, and transplant-related variables were prospectively documented and obtained for this analysis from our comprehensive institutional database. Clinical factors examined included age (<55 years, ≥55 years), sex, CMV status, year of transplant (2010 to 2012 or 2013 to 2015), diagnosis (acute myelogenous leukemia (AML), myelodysplastic syndrome (MDS)/myeloproliferative neoplasms (MPN)/chronic myelogenous leukemia (CML), acute lymphoblastic leukemia (ALL), non-Hodgkin lymphoma (NHL)/Hodgkin lymphoma (HL)/chronic lymphocytic leukemia (CLL), or other malignancy), conditioning intensity (nonmyeloablative, RIC, myeloablative), donor type (MRD, MUD, HID), graft source (bone marrow or PBSC), hematopoietic cell transplant comorbidity index (HCT-CI),18 and disease risk index (DRI)19. Statistical Analysis OS, DFS, and GRFS were estimated by the Kaplan-Meier method.20 Generation of the CGRFS curve was accomplished by linear combination of five Kaplan-Meier 4 Page 4 of 19
estimates to accommodate two episodes of chronic GVHD and their resolution.21 The model is based on nine health states that a patient may be in after transplant (Figure 1). The states are as follows: 0. Alive and in first remission. 1. Dead or relapsed/progressed before first chronic GVHD. 2. Alive, in remission, and with first episode of moderate-to-severe chronic GVHD. 3. Dead or relapsed/progressed in first chronic GVHD. 4. Alive, in remission, after resolution of first chronic GVHD and before second chronic GVHD (symptoms quiescent and systemic immunosuppression discontinued). 5. Dead or relapsed/progressed after resolution of first chronic GVHD and before second chronic GVHD. 6. Alive, in remission, and with recurrence of moderate-to-severe chronic GVHD (second episode) 7. Dead or relapsed/progressed in second episode of chronic GVHD. 8. Alive, in remission, after resolution of second chronic GVHD (symptoms quiescent and systemic immunosuppression discontinued). 9. Dead or relapsed/progressed after resolution of second GVHD. CGRFS is probability that one stays in state 0, state 4 or state 8; represented by the formula: C(t) = S1(t) + [S2(t) ─ S3(t)] + [S4(t) ─ S5(t)]. For S1(t), the event is death, relapse/progression or initial chronic GVHD. This is the usual chronic GVHD and relapse free survival function and is the chance that the patient is in state 0. For S2(t), the event is second chronic GVHD or any death, relapse/progression before second chronic GVHD. For S3(t), the event is resolution of first chronic GVHD or any death, relapse/progression before resolution of first chronic GVHD. The difference between S2(t) and S3(t), [S2(t) ─ S3(t)], is the probability of being in state 4. S4(t) is DFS with any death, relapse/progression as the events. For S5(t), the event is resolution of second chronic GVHD or any death, relapse/progression before resolution of second chronic GVHD. The difference, [S4(t) – S5(t)], is the probability of being in state 8. CGRFS is estimated by linear combination of five Kaplan-Meier estimates. A simulation-based supremum test in the time interval [0, 4] (years) was conducted to compare CGRFS between two independent samples.22 At year 4 about 20% of patients remained in the study. Testing in [0, 4] (years) could avoid large variability at the tail and generate stabilized result. In each test, 1000 realizations of process of difference in CGRFS were generated to approximate its distribution. The supremum was found for the absolute value of each realized process in [0, 4] (years). The p value for a two-sample test was obtained by finding the proportion of 1000 supremum values greater than the sample supremum. RESULTS Patient Characteristics 5 Page 5 of 19
In total, 422 adult patients were included in this analysis with a median age at transplant of 54 years (range, 18 to 77) (Table 1). Patients were transplanted for AML (38%), MDS/MPN/CML (27%), NHL/HL/CLL (20%), and ALL (13%), and 30% of patients were high-/very high-risk by DRI. Donor type was MRD, MUD, and HID in 30%, 39%, and 31% respectively. Regimen intensity was myeloablative in 50% of patients, and PBSC was the stem cell source in 78%. The median time of followup was 36 months (range, 12 to 78). OS, DFS, and conventional GRFS Estimated one-, two-, three- and four-year OS was 78%, 69%, 60% and 59% respectively; corresponding rates of DFS were 68%, 60% 54% and 52%. The major determinants of DFS were DRI and age. Estimated DFS was improved in patients with low (63%) or intermediate (55%) DRI vs. high (43%) or very high (24%) DRI (low/intermediate vs. high/very high, p<0.001), and in patients younger than 55 years of age (60%) vs. older patients (44%, p<0.001). Estimated one-, two-, threeand four-year conventional GRFS was 33%, 26%, 23% and 22% respectively. Current GVHD-free, relapse-free survival (CGRFS) The estimated one-, two-, three- and four-year CGRFS in this cohort was 45%, 46%, 47% and 49%, respectively; this corresponds to an absolute increase of 12%, 20%, 24% and 27% respectively when compared to conventional GRFS from the same time points (Figure 2a). This improvement in outcome is due a steady recovery from chronic GVHD over time (Figure 2b), as well the omission of grade 3-4 acute GVHD as an endpoint. The component events that make up the CGRFS endpoint (death, relapse, and modsevere chronic GVHD) are graphically depicted in Figure 3. At one year posttransplant, the components of death, relapse and chronic GVHD proportionally accounted for 40%, 18% and 42% respectively of patients defined as treatment failure by CGRFS. The proportion of patients meeting a CGRFS “event” because of chronic GVHD decreased successively in the first four years post-transplant, accounting for 42%, 26%, 13% and 8% respectively. This was due to the fact that a decreasing percentage of patients were living with active moderate-to-severe chronic GVHD over time, quantitated at 23%, 14%, 7% and 4% respectively at 1, 2, 3 and 4 years post-transplant. A summary of survival and incidence estimates can be found in Table 2. Effect of Patient-, Disease- and Transplant-Related Factors on CGRFS In univariate analysis, CGRFS was significantly correlated only with DRI (low/intermediate vs. high/very high, p<0.001) (figure 3), but not transplant type (MRD, MUD, haplo), age (<55, ≥55 years), HCT-CI (0-2, ≥3), graft source (BM, PBSC), regimen intensity (MA, RIC/NMA), donor/recipient sex mismatch, donor or recipient CMV status or year of transplant (Figure 4). In patients with low-, 6 Page 6 of 19
intermediate-, high- and very high-risk DRI, the estimated 3-year CGRFS was 56%, 50%, 36% and 33% respectively. DISCUSSION We evaluated 422 consecutive patients receiving a MRD, MUD, or HID alloHCT at a single institution in an attempt to develop a novel endpoint, which we have called CGRFS. This composite endpoint attempts to incorporate the most important elements of transplant success. It acknowledges the important impact that clinically-significant ongoing GVHD has on the QOL of long-term survivors of alloHCT. CGRFS is defined as survival without relapse, and without active moderateto-severe chronic GVHD, by NIH consensus criteria, at the time of most recent assessment. This differs significantly from the conventional definition of GRFS, which, in addition to death and relapse, treats grade 3-4 acute GVHD and immunosuppression-requiring chronic GVHD as “fixed events” defining treatment failure. The conventional definition of GRFS has significant shortcomings, most notably that GVHD is treated as a fixed event, similar to death or relapse. However, treating chronic GVHD as a fixed complication does not properly reflect the reality experienced by the majority of patients following alloHCT. In our series, patients living with active moderate-to-severe chronic GVHD decreased continuously following transplant, accounting for 23%, 14%, 7% and 4% of patients in the first four years post-transplant. This is critically important since once chronic GVHD resolves, a patient’s health status does not appear to differ significantly from those patients who never had chronic GVHD.11 Secondly, we believe that treating grade 34 acute GVHD as an “event” is unnecessary, as this event almost universally will either resolve with time or convert to either death or chronic GVHD, which are both captured as part of the CGRFS endpoint. Although severe acute GVHD can be associated with a significant mortality risk, it is unlikely to affect the long-term QOL of patients that recover from it and do not develop clinically significant chronic GVHD.23 Furthermore, it is now well recognized that chronic GVHD is extremely heterogeneous in its presentation. Some patients present with mild symptoms restricted to a few organs, while others can have extensive involvement lasting many years and causing severe medical, social, and QOL issues. Chronic GVHD has been much better characterized since the first NIH Consensus classification was developed in 2005 and prospectively validated.15,24-26 It is now better appreciated that only moderate-to-severe chronic GVHD, by the NIH criteria, is associated with inferior survival and worsened functional status; whereas neither is seen in patients with mild severity chronic GVHD.25,26 As a result, the CGRFS endpoint includes only patients with moderate-to-severe NIH severity chronic GVHD, rather than the requirement for systemic immunosuppression, as utilized by conventional GRFS. We hypothesize that CGRFS composite endpoint is a more accurate surrogate for 7 Page 7 of 19
transplant success, when compared with more traditional endpoints such as DFS or conventional GRFS. At one year post-transplant, estimated DFS, CGRFS, and conventional GRFS is 68%, 45%, and 33% respectively, whereas by four years posttransplant, these figures are 52%, 49%, and 22% respectively. Several lessons are apparent from these survival numbers. In the early years post-transplant, (1) DFS appears to be an inadequate reflection of transplant success as it fails to reflect the significant morbidity and QOL impact related to moderate-to-severe chronic GVHD; and (2) conventional GRFS appears to dramatically underestimate “transplant success.” Although the latter is apparent already in the first year post-transplant (12% differential between CGRFS and GRFS), it becomes a significantly greater issue with time (27% differential between CGRFS and GRFS by four years posttransplant). It’s also important to note that by four years post-transplant, CGRFS and DFS are nearly identical since only 4% of patients are living with active moderate-to-severe chronic GVHD at that time. If one accepts that CGRFS is a useful indicator of transplant success post alloHCT, it is noteworthy that this endpoint does not appear to be significantly impacted by a majority of recipient- or transplant-related factors, such as patient age, HCT-CI, donor type (MRD, MUD, HID), regimen intensity, or stem cell source. Only disease risk, as measured by DRI, had a significant impact on CGRFS. The fact that transplant success, as measured by CGRFS, appears largely independent of either donor type, stem cell source, or regimen intensity, suggests that these factors should perhaps not play a major role in the decision to offer alloHCT to a patient. This fact may be especially relevant for racial or ethnic minorities who are significantly less likely to find a suitable matched related or unrelated donor.27 A major limitation of this study was the restriction of patients to a single center. Further validation of the CGRFS endpoint will be required in patients transplanted from other institutions. In this regards, it is of great interest that the Center for International Blood and Marrow Transplant Research (CIBMTR) has recently revised their data collection forms so that significant chronic GVHD data points will be recorded such as maximum GVHD grade and the dates of systemic immunosuppression initiation and discontinuation. An additional limitation of this study is the lack of health-related QOL measurements as well as analyses of the economic burdens of GVHD management, which should be incorporated in future studies analyzing composite endpoints such as CGRFS. Finally, it should be stipulated that survivors of alloHCT have an increased incidence of chronic health problems28 that can occur even in the absence of GVHD, and such morbidity will not be reflected in the CGRFS endpoint. In spite of the above limitations, we suggest that the composite endpoint of CGRFS represents a more accurate estimate of transplant success following alloHCT. It represents survival without disease relapse/progression and without continuing morbidity related to ongoing moderate-to-severe chronic GVHD. This endpoint yields significantly more information that the simpler measurements of DFS or OS. Furthermore, the treatment of chronic GVHD as a dynamic rather than static event 8 Page 8 of 19
creates a more accurate picture of the experience of patients post-transplant. We hypothesize that CGRFS will be a more useful measurement of transplant success when assessing and comparing various alloHCT strategies. ACKNOWLEDGEMENTS The authors thank Stacey Brown and Katelin Jackson for excellent assistance in performing data query. AUTHORSHIP CONTRIBUTIONS S.R.S. and X.Z. designed the research; S.R.S., M.S., L.E.M., H.K.H, and A.B. treated patients; C.S. and M.R. collected the data; S.R.S. analyzed and interpreted data; X.Z. performed the statistical analysis; S.R.S. wrote the manuscript; and all authors revised and approved the final version of the manuscript. CONFLICT OF INTEREST DISCLOSURE There are no competing financial interests in relation to the work described.
9 Page 9 of 19
REFERENCES 1. Thomas ED BK, Forman SJ, Appelbaum FR: Thomas' Hematopoietic Cell Transplantation (ed 4), John Wiley & Sons, 2004 2. Abou-Mourad YR, Lau BC, Barnett MJ, et al: Long-term outcome after allo-SCT: close follow-up on a large cohort treated with myeloablative regimens. Bone Marrow Transplantation 45:295-302, 2010 3. Bhatia S, Francisco L, Carter A, et al: Late mortality after allogeneic hematopoietic cell transplantation and functional status of long-term survivors: report from the Bone Marrow Transplant Survivor Study. Blood 110:3784-3792, 2007 4. Schmitz N, Eapen M, Horowitz MM, et al: Long-term outcome of patients given transplants of mobilized blood or bone marrow: a report from the International Bone Marrow Transplant Registry and the European Group for Blood and Marrow Transplantation. Blood 108:4288-4290, 2006 5. Higman MA, Vogelsang GB: Chronic graft versus host disease. Br J Haematol 125:435-54, 2004 6. Koc S, Leisenring W, Flowers ME, et al: Therapy for chronic graftversus-host disease: a randomized trial comparing cyclosporine plus prednisone versus prednisone alone. Blood 100:48-51, 2002 7. Vogelsang GB: How I treat chronic graft-versus-host disease. Blood 97:1196-201, 2001 8. Ratanatharathorn V, Ayash L, Lazarus HM, et al: Chronic graft-versushost disease: clinical manifestation and therapy. Bone Marrow Transplant 28:121-9, 2001 9. Sutherland HJ, Fyles GM, Adams G, et al: Quality of life following bone marrow transplantation: a comparison of patient reports with population norms. Bone Marrow Transplant 19:1129-36, 1997 10. Chiodi S, Spinelli S, Ravera G, et al: Quality of life in 244 recipients of allogeneic bone marrow transplantation. British Journal of Haematology 110:614619, 2000 11. Fraser CJ, Bhatia S, Ness K, et al: Impact of chronic graft-versus-host disease on the health status of hernatopoietic cell transplantation survivors: a report from the Bone Marrow Transplant Survivor Study. Blood 108:2867-2873, 2006 12. Kiss TL, Abdolell M, Jamal N, et al: Long-term medical outcomes and quality-of-life assessment of patients with chronic myeloid leukemia followed at least 10 years after allogeneic bone marrow transplantation. Journal of Clinical Oncology 20:2334-2343, 2002 13. Holtan SG, DeFor TE, Lazaryan A, et al: Composite end point of graftversus-host disease-free, relapse-free survival after allogeneic hematopoietic cell transplantation. Blood 125:1333-8, 2015 14. Solh M, Zhang X, Connor K, et al: Factors Predicting Graft-versus-Host Disease-Free, Relapse-Free Survival after Allogeneic Hematopoietic Cell 10 Page 10 of 19
Transplantation: Multivariable Analysis from a Single Center. Biol Blood Marrow Transplant 22:1403-9, 2016 15. Filipovich AH, Weisdorf D, Pavletic S, et al: National Institutes of Health consensus development project on criteria for clinical trials in chronic graftversus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant 11:945-56, 2005 16. Gratwohl A, Hermans J, Apperley J, et al: Acute graft-versus-host disease: grade and outcome in patients with chronic myelogenous leukemia. Working Party Chronic Leukemia of the European Group for Blood and Marrow Transplantation. Blood 86:813-8, 1995 17. Giralt S, Ballen K, Rizzo D, et al: Reduced-intensity conditioning regimen workshop: defining the dose spectrum. Report of a workshop convened by the center for international blood and marrow transplant research. Biol Blood Marrow Transplant 15:367-9, 2009 18. Sorror ML, Maris MB, Storb R, et al: Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood 106:2912-9, 2005 19. Armand P, Kim HT, Logan BR, et al: Validation and refinement of the Disease Risk Index for allogeneic stem cell transplantation. Blood 123:3664-71, 2014 20. Kaplan E, Meier P: Nonparametric estimation from incomplete observations. . Journal of the American Statistical Association 84:360-372, 1958 21. Klein JP, Keiding N, Shu Y, et al: Summary curves for patients transplanted for chronic myeloid leukaemia salvaged by a donor lymphocyte infusion: the current leukaemia-free survival curve. Br J Haematol 109:148-52, 2000 22. Parzen MI, Wei LJ, Ying Z: Simultaneous Confidence Intervals for the Difference of Two Survival Functions. Scandinavian Journal of Statistics 24:309-314, 1997 23. Lee SJ, Kim HT, Ho VT, et al: Quality of life associated with acute and chronic graft-versus-host disease. Bone Marrow Transplant 38:305-10, 2006 24. Jagasia MH, Greinix HT, Arora M, et al: National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graftversus-Host Disease: I. The 2014 Diagnosis and Staging Working Group Report. Biology of Blood and Marrow Transplantation 21:389-401.e1, 2015 25. Moon JH, Sohn SK, Lambie A, et al: Validation of National Institutes of Health Global Scoring System for Chronic Graft-Versus-Host Disease (GVHD) According to Overall and GVHD-Specific Survival. Biology of Blood and Marrow Transplantation 20:556-563, 2014 26. Pidala J, Kurland B, Chai X, et al: Patient-reported quality of life is associated with severity of chronic graft-versus-host disease as measured by NIH criteria: report on baseline data from the Chronic GVHD Consortium. Blood 117:4651-4657, 2011 27. Gragert L, Eapen M, Williams E, et al: HLA match likelihoods for hematopoietic stem-cell grafts in the U.S. registry. N Engl J Med 371:339-48, 2014 28. Sun C-L, Kersey JH, Francisco L, et al: Burden of Morbidity in 10+ Year Survivors of Hematopoietic Cell Transplantation: Report from the Bone Marrow 11 Page 11 of 19
Transplantation Survivor Study. Biology of Blood and Marrow Transplantation 19:1073-1080
12 Page 12 of 19
Fig 1. Possible transitions in multistate model for CGRFS. Fig 2. (A) Comparison of CGRFS to Conventional GRFS, DFS, and OS, (B) Gain in CGRFS Due to the Treatment of chronic GVHD as a Dynamic Event. Fig 3. Component Events of the CGRFS endpoint. Fig 4. CGRFS according to (A) DRI, (B) Age, (C) HCT-CI, (D) Donor Type, (E) Conditioning Intensity, and (F) Stem Cell Source.
13 Page 13 of 19
Table 1. Patient, diagnosis, and transplant characteristics ________________________________________________________________________ Patients (n=422) - N (%) ________________________________________________________________________ Age (years) Donor Type MRD 125 (30%) Median (range) 54 (18-77) MUD 165 (39%) Gender Male 235 (56%) HID 132 (31%) Female 187 (44%) Regimen Intensity F → M 67 (16%) RIC/NMA 209 (50%) HCT-CI MA 213 (50%) 0-2 228 (54%) Stem Cell Source ≥3 194 (46%) BM 93 (22%) Diagnosis PBSC 329 (78%) ALL 55 (13%) Year of Transplant 2010 65 (16%) AML 162 (38%) MDS/MPN/CML 112 (27%) 2011 68 (16%) 2012 69 (16%) NHL/HL/CLL 83 (20%) Other 10 (2%) 2013 77 (18%) 2014 77 (18%) DRI Low 59 (14%) 2015 66 (16%) Intermediate 230 (55%) CMV High 107 (25%) Recipient Pos 284 (67%) Very High 22 (5%) Recipient Neg 138 (33%)
Abbreviations : F→M, female donor-to-male recipient; HCT-CI, hematopoietic cell transplantation-comorbidity index; ALL, acute lymphoblas tic leukemia; AML, acute myeloid leukemia; MDS, myelodysplas tic syndrome; MPN, myeloproliferative neoplasm; CML, chronic myelogenous leukemia; NHL, non-Hodgkins lymphoma; HL, Hodgkins lymphoma; CLL, chronic lymphocytic leukemia; DRI, disease ris k index; MRD, matched related donor; MUD, matched unrelated donor; HID, haploidentical donor; RIC, reduced intens ity conditioning; NMA, nonmyeloablative; MA, myeloablative; BM, bone marrow; PBSC, peripheral blood stem cells ; CMV, cytomegalovirus seros tatus .
14 Page 14 of 19
Table 2. Summary of Survival and Incidence Estimates 1
Post-Transplant Year 2 3
4
OS (%)
78
69
60
59
DFS (%)
68
60
54
52
CGRFS (%)
45
46
47
49
GRFS (%)
33
26
23
22
CGRFS-GRFS (%)
12
20
24
27
Moderate-severe cGVHD, relapse-free survival (%)
43
34
31
30
Recovery from cGVHD (%)
2
12
16
19
Active moderate-severe cGVHD (%)
23
14
7
4
15 Page 15 of 19
Figure 1
16 Page 16 of 19
Figure 2
A
B
17 Page 17 of 19
Figure 3
18 Page 18 of 19
Figure 4
A
B
C
D
E
F
19 Page 19 of 19