The aim of this study was to evaluate extramedullary (EM) relapses and its features in an allogeneic hematopoietic stem cell transplantation (alloHSCT) cohort, which consisted of patients with acute leukemia and advanced-phase chronic myeloid leukemia. One hundred and twenty-eight alloHSCT patients transplanted between the years 2001 and 2014 were analyzed. EM relapses observed in acute lymphoblastic leukemia (ALL) were more frequent than that of in acute myeloid leukemia (AML) and CML, although calculation of cumulative risk incidence, BM relapse, EM relapse, and non-relapse mortality were considered as competing risks of each other. At the 60th month, estimated CBMR and CEMR incidences were, respectively, 14.3 (5.1)% and 25.9 (6.6)% in ALL, 25.8 (5.9)% and 15.5 (4.8)% in AML, and 61.5 (16.5)% and 17.9 (13.4)% in CML. Among multiple parameters, the only type of conditioning regimen (p:0.046), EM involvement at diagnosis (p:0.009), and the presence of GVHD were found to be associated with EM relapse risk independently (p:0.045). Chronic GVHD and TBI-based regimens significantly decreased the EM relapse risk, whereas it was higher with Mel/Flu and its variants. In conclusion, EM relapse is not uncommon after alloHSCT. GVHD and TBI-based regimens may prevent this complication.
Allogeneic hematopoietic stem cell transplantation (alloHSCT) is considered to be the only curative treatment for the patients with acute leukemia (AL)—chronic myeloid leukemia (CML). Transplantation-related mortality and morbidity have been reduced with better supportive care and reduced intensity conditioning (RIC) regimens combined with better graft versus host disease (GVHD) prophylaxis. However, leukemia relapse after alloHSCT still remains a devastating complication as the main reason for treatment failure with about 40% rate [1,2,3,4,5,6]. Between those relapsed patients, the patients with extramedullary (EM) relapse have worse prognosis compared with the remaining cases . In some recent studies, higher EM relapse incidence was reported in alloHSCT . Although the bone marrow (BM) relapse has been well studied, there are limited systematic data on the EM relapse in the literature.
The aim of this study is to assess EM relapse and its patterns, cumulative incidence, and risk factors in an AL and advanced-phase CML alloHSCT cohort.
Patients and methods
Records of all patients with AL and accelerated/blastic-phase CML, who underwent alloHSCT in our center between June 2001 and September 2014, were analyzed retrospectively. These data were recorded prospectively during patients’ follow-up.
The ethical considerations were handled in accordance with the Helsinki Declaration. As a standard of action at our hospital, informed consent has been taken from all of the study participants at the time of hospitalization.
The study cohort consisted of the patients with acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), and the advanced-phase CML. Nine of those patients had EM involvement at diagnosis. They have all received chemotherapy regimens in order to reach remission before alloHSCT. The chemotherapy regimens, which patients had received, varied according to the underlying leukemic process. The patients with ALL were treated with the standart Berlin Frankfurt-Munster (BFM) 95 regimens or Cancer and Leukemia Group B Study protocol (CALGB regimen) [7, 8]. The patients with AML were received the anthracycline-based chemotherapy regimens (idarubicin 12 mg/m2/day intravenously (i.v.) for the first 3 days and cytosine arabinoside 100 mg/m2/day i.v. from the first to seventh day) as a remission induction therapy and continued with the chemotherapy protocol consisting of high-dose cytosine arabinoside (3000 mg/m2/day i.v. for 3 days) as a consolidation. Antracycline-based regimens and tyrosine kinase inhibitors were given to the patients with CML. AlloHSCT was performed when the patients had reached the remission.
Nine of the patients included in the study had EM involvement at diagnosis. Eight of them were the patients with ALL and one of them with AML. No CML patient had EM involvement at diagnosis. Involved EM sites of those patients were the central nervous system (CNS) (n: 5), mediastinum (n: 2), bone (n: 1), and gastrointestinal tract (n: 1). Those patients received systemic chemotherapeutic protocols such as BFM 95, CALGB, or anthracycline-based regimens according to underlying leukemia. In addition to systemic therapy, intrathecal therapies consisting of dexamethasone 4 mg, methotrexate (MTX) 15 mg, and cytosine arabinoside 30 mg were given twice per week. They had not received any local therapy.
Of all transplanted patients, donors were human leukocyte antigen full-matched siblings and the stem cell source was peripheral blood. The transplantation types were RIC and myeloablative conditioning regimens in 102 and 26 patients, respectively. The conditioning regimens were busulfan/fludarabine/anti-thymoglobulin (BU/Flu/ATG) in 63 [BU: 0.8 mg/kg/day i.v., Flu: 50 mg/m2/day i.v., ATG: 5 mg/kg/day i.v.], melphalan/fludarabine (Mel/Flu) and its variants in 23 (Mel: 140 mg/m2/day i.v., Flu: 50 mg/m2/day i.v.), total body irradiation-based (TBI-based) regimens in 16, busulfan/cyclophosphamide (BU/Cy) (BU: 0.8 mg/kg/day i.v., Cy: 50 mg/kg/day) and its variants in 15, and BU/Flu (BU: 0.8 mg/kg/day i.v., Flu: 30 mg/m2 i.v.) and its variants in 11 patients. Cyclosporine A (CsA) (3 mg/kg/day i.v.) and MTX (10 mg/kg/day i.v. on days 1, 3, 5, and 11) were started for GVHD prophylaxis. CsA was started on the day of the transplantation with the target blood level of 200–300 ng/mL and it was started to taper 2 months after transplantation, according to GVHD status of the patient.
The term EM relapse included isolated EM relapse and EM relapse together with BM relapse. The relapse in an EM compartment was detected with imaging techniques such as computerized tomography and/or magnetic resonance imaging, and confirmed histologically by performing biopsy. The CNS relapse was diagnosed with a biopsy and/or in the most case cerebrospinal fluid (CSF) examination, if leukemic cells were identified > 10/ml of CSF. CML relapse to chronic phase was not considered as an event. Hyperleukocytosis was defined as a peripheral white blood cell count > 50 × 109/L in AML, > 30 × 109/L in B lymphoblastic leukemia, and > 100 × 10/L in T lymphoblastic leukemia at the time of diagnosis [9, 10]. The term high-risk cytogenetics was based on the European LeukemiaNet criteria [4, 11].
Primary endpoints of the study were cumulative EM relapse incidence (CEMRI) and cumulative BM relapse incidence (CBMRI). These durations were calculated from the time of transplant to the day of determination of relapse. Overall survival (OS) was calculated from time of transplant to the date of mortality of any reason. The patients who did not die and those who did not relapse at last follow-up were censored, although calculations of cumulative risk incidence, BM relapse, EM relapse, and non-relapse mortality were considered as competing risks. Chi-square (or Fisher’s exact test) and independent-samples t-test were used to compare the categorical and continuous data, respectively. Kaplan–Meier method and log-rank test was used for survival analyses and comparison of survival rates.
In order to identify the risk factors for EM relapses in our cohort, univariate and multivariate analysis were performed among the variables including age, sex, disease type, disease status, EM involvement, hyperleukocytosis at diagnosis, cytogenetic risk, type of transplantation (myeloablative vs. RIC), conditioning regimen that was given, and GVHD status (acute GVHD (aGVHD), chronic GVHD (cGVHD), no GVHD).
Univariate comparisons with a P-value < 0.2 were included in multivariate analyses in which statistical significance threshold was accepted as P < 0.05. Cox regression analysis was used to study simultaneous effect of selected variables on EM relapse. CEMRI, CBMRI, and cumulative non-relapse mortality incidence were calculated according to the Gray’s test [12, 13]. Cumulative incidences were calculated by statistical software environment R Version 3.2.2 (The R Foundation for Statistical Computing, Vienna, Austria) . Statistical Packages for the Social Sciences v20.0.0 (SPSS, Inc., Chicago, IL, USA) software was used for other statistical analyses.
Different regimens were given in different disease subtypes. The distribution of the conditioning regimen among the disease subtypes were given in Table 1. Fifty-one of those patients were ALL (45 in the first remission, 3 in the subsequent remissions, 3 active disease), 64 were AML (63 in the first remission, 1 active disease), and 13 were advanced-phase CML (9 blastic phase including 4 with active disease, 5 accelerated phase, 4 other advanced disease). Important descriptive data of the patients are summarized in Table 2.
Relapse was observed in 52 (40.6%) patients. Seventy-seven of the 128 patients were still surviving (65 without relapse) and 51 died (12 without relapse) at last follow-up. Median follow-up duration of the surviving patients was 51.1 (3.1–133) months.
Types of the relapses were BM in 29, BM and EM in 18, and isolated EM in 5 patients. In the patients with EM relapse, the median time to relapse (10.73 months) was longer than one in the BM relapse (5.8 months). However, this difference was not statistically significant (p: 0.210). In the subgroup of isolated EM relapse, median time to relapse was clearly longer than the others. That was 19.2 months (5.73–22.23) for isolated EM relapse and 6.63 months (0.73–38.37) for EM and BM relapse, and BM relapse only. The involvement sites of EM relapses included the CNS, breast, bone, kidney, testicle, skin, and mediastinum. In all the three subtypes of leukemia, the most common involved site was CNS and four patients had developed EM relapse in more than one site besides the CNS involvement (Table 3).
There were five patients who experienced isolated EM relapse without BM involvement. The involved sites of those patients were the bone (two), CNS, skin, and testicle.
Different relapse types were observed among the different disease subtypes. We found that overall relapse rate had been higher in the patients with CML than the ones in the patients with ALL and AML (p: 0.086). However, EM relapses were more frequent in the patients with ALL (7 only BM, 11 BM and EM, 1 only EM) than in patients with AML (15 only BM, 7 BM and EM, 2 only EM) and with CML (7 only BM, 2 only EM as a myeloid sarcoma) (χ2, p: 0.034).
Estimated CBMRI and CEMRIs at the 60th month [ratio (SE)] were 14.3 (5.1)% and 25.9 (6.6)% in ALL, 25.8 (5.9)% and 15.5 (4.8)% in AML, and 61.5 (16.5)% and 17.9 (13.4)% in CML (competing risk analyzes by Gray’s test, p: 0.007 and p: 0.357 for CBMRI and CEMRI, respectively) (Fig. 1).
In univariate analysis among multiple disease and patient-related parameters, sex (p: 0.032), EM involvement at diagnosis (p: 0.009), type of conditioning regimen (p: 0.025) and GVHD status (p: 0.045) were found to be associated with EM relapse risk in the study cohort. EM relapse was more frequent in female patients than in male patients and the patients with EM involvement at diagnosis develop EM relapse more than the others. TBI-based regimens significantly decreased the EM relapse risk, whereas it was higher with Mel/Flu and its variants.
In analysis on the subtypes of the leukemia, it was found that, sex, GVHD status, and conditioning regimens of the patients had been related to the EM relapse risk (p: 0.022, 0.05, and 0.172, respectively), whereas hyperleukocytosis did not have any effect on this (p: 0.536) in patients with ALL. EM relapse was more frequent in female patients than in male patients and those who received Mel/Flu and its variants as a conditioning regimen had higher EM relapse rate than others. Presence of GVHD had a protective effect for EM relapse. For the patients with AML, cytogenetic risk and type of transplantation gained importance in the EM relapse risk. The patient with high cytogenetic risk and the patient who received RIC had higher relapse rate than the others (p: 0.040 and p: 0.047, respectively). Due to the small number, we did not perform statistical analysis in the group of the patients with CML.
In the whole study group, 17 patients developed aGVHD and 31 had cGVHD (Table 4). We found that aGVHD did not have any effect on EM relapse risk, whereas patients with cGVHD had significantly lower EM relapse risk than the others. cGVHD, EM involvement at diagnosis, and type of conditioning regimen were independent factors effecting EM relapse risk in Cox regression analysis, (p-values 0.035, 0.039, and 0.046, respectively) (Table 5).
Amongst the relapsed patients, there was no significant difference in OS according to presence of EM relapse (χ2, p: 0.848). Median estimated survival of the patients with BM relapse was 6.24 months, whereas that of the patients with EM relapse was 6.14 months (Fig. 2).
Among alloHSCT patients, a significant proportion develops EM relapse. However, there are limited data about characteristics and related factors. The incidence of EM relapse after alloHSCT shows variability within the range of 5–20% in the literature [4, 11, 15, 16]. Those different rates may be due to limitations of the studies such as being retrospective analysis and including limited number of patients. We observed 23 (18%) EM relapse patients (5 of them isolated) within a cohort of 128 AL and advanced-phase CML patients. The incidence of EM relapse was higher in patients with ALL (29.3%) than the patients with AML (10.8%) and CML (10.4%). This finding was compatible with the reports on the literature . It was reported previously that time to EM relapse after alloHSCT had been longer than the time to BM relapse after HSCT [16, 17]. We did not find any statistically significant difference between the median relapse times of isolated BM and combined BM and EM relapses. However, only the patients with isolated EM relapse had longer relapse time compared with the others.
In the previous studies, age ( > 18 years at diagnosis), high-risk cytogenetics, and EM involvement at diagnosis were reported as risk factors for EM relapses [1, 16, 18]. In univariate analyses, we found that sex, EM involvement at the time of diagnosis, conditioning regimen that was used, and cGVHD affected the EM relapse rate. Multivariate analysis showed that those parameters apart from the sex were independent risk factors for development of EM relapse. Unlike previous studies, hyperleukocytosis and male sex were not determined as risk factors in this study . We found that EM relapse was more frequent in female patients than in male patients. However, it was not an independent risk factor for EM relapse in multivariate analysis. The results about use of TBI-based regimen were similar to former studies. In our study, use of TBI -based regimen lowered the risk of EM relapse, as compatible with the literature [5, 18, 19]. There is a conflict about the role of GVHD status in development of EM relapse in the literature. In some studies, a protective effect of GVHD was reported for EM relapse, whereas other studies reported no effect [15, 16]. In this study, cGVHD was determined as a protective factor for EM relapse, whereas aGVHD had no effect on EM relapse. To explain this protective effect of cGVHD, it may be speculated that graft vs. leukemia effect might protect the patients from EM relapse. In some recent studies, it was reported that EM involvement before transplantation had had no impact on the relapse rate [20, 21]. Unlike this finding our data showed that having EM involvement at diagnosis had created a tendency to develop EM relapse after alloHSCT too. However, because of small number of EM involvement in the patients AML (n: 1) and CML (n: 0), this finding is not suitable to be generalized to whole study group. Some previous studies reported breast and testicles to be the most common localization of EM tumors [1, 2, 11, 22]. In our study, the most common EM involvement site was the CNS including all three types of leukemia, unlike the previous reports in the literature. A possible explanation for this observation is that CNS may provide a sanctuary site for leukemic cells after alloHSCT. In some studies, it was reported that isolated EM relapse had a better prognosis than BM relapse [17, 23, 24]. Unlike this finding, OS of these two groups were similar in our study. However, it should be noted that there were only five isolated EM relapse cases in our cohort.
The major limitations of our study are its retrospective nature and including relatively small numbers of patients.
In conclusion, EM relapse is quite frequent after alloHSCT, especially in ALL. Besides unchangeable factors such as female sex, use of TBI-based regimen for conditioning, and the presence of cGVHD are also related to EM relapse risk.
Ge L, Ye F, Mao X, Chen J, Sun A, Zhu X, et al. Extramedullary relapse of acute leukemia after allogeneic hematopoietic stem cell transplantation: different characteristics between acute myelogenous leukemia and acute lymphoblastic leukemia. Biol Blood Marrow Transplant. 2014;20:1040–7. https://doi.org/10.1016/j.bbmt.2014.03.030. e-pub ahead of print 2014/04/08
Huang Q, Reddi D, Chu P, Snyder DS, Weisenburger DD. Clinical and pathologic analysis of extramedullary tumors after hematopoietic stem cell transplantation. Hum Pathol. 2014;45:2404–10. https://doi.org/10.1016/j.humpath.2014.07.022. e-pub ahead of print 2014/10/09
Shimizu H, Saitoh T, Hatsumi N, Takada S, Handa H, Jimbo T, et al. Prevalence of extramedullary relapses is higher after allogeneic stem cell transplantation than after chemotherapy in adult patients with acute myeloid leukemia. Leuk Res. 2013;37:1477–81. https://doi.org/10.1016/j.leukres.2013.08.017. e-pub ahead of print 2013/09/24
Dohner H, Estey EH, Amadori S, Appelbaum FR, Buchner T, Burnett AK, et al. Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the European LeukemiaNet. Blood. 2010;115:453–74. https://doi.org/10.1182/blood-2009-07-235358. e-pub ahead of print 2009/11/03
Ruiz-Arguelles GJ, Gomez-Almaguer D, Vela-Ojeda J, Morales-Toquero A, Gomez-Rangel JD, Garcia-Ruiz-Esparza MA, et al. Extramedullary leukemic relapses following hematopoietic stem cell transplantation with nonmyeloablative conditioning. Int J Hematol. 2005;82:262–5. https://doi.org/10.1532/ijh97.04195. e-pub ahead of print 2005/10/07
Au WY, Kwong YL, Lie AK, Ma SK, Liang R. Extra-medullary relapse of leukemia following allogeneic bone marrow transplantation. Hematol Oncol. 1999;17:45–52. e-pub ahead of print 1999/10/16
Larson RA, Dodge RK, Burns CP, Lee EJ, Stone RM, Schulman P, et al. A five-drug remission induction regimen with intensive consolidation for adults with acute lymphoblastic leukemia: cancer and leukemia group B study 8811. Blood. 1995;85:2025–37. e-pub ahead of print 1995/04/15
Moricke A, Reiter A, Zimmermann M, Gadner H, Stanulla M, Dordelmann M, et al. Risk-adjusted therapy of acute lymphoblastic leukemia can decrease treatment burden and improve survival: treatment results of 2169 unselected pediatric and adolescent patients enrolled in the trial ALL-BFM 95. Blood. 2008;111:4477–89. https://doi.org/10.1182/blood-2007-09-112920. e-pub ahead of print 2008/02/21
Greenwood MJ, Seftel MD, Richardson C, Barbaric D, Barnett MJ, Bruyere H, et al. Leukocyte count as a predictor of death during remission induction in acute myeloid leukemia. Leuk & Lymphoma. 2006;47:1245–52. https://doi.org/10.1080/10428190600572673. e-pub ahead of print 2006/08/23
Gokbuget N, Hoelzer D. Treatment of adult acute lymphoblastic leukemia. Semin Hematol. 2009;46:64–75. https://doi.org/10.1053/j.seminhematol.2008.09.003. e-pub ahead of print 2008/12/23
Baccarani M, Saglio G, Goldman J, Hochhaus A, Simonsson B, Appelbaum F, et al. Evolving concepts in the management of chronic myeloid leukemia: recommendations from an expert panel on behalf of the European LeukemiaNet. Blood. 2006;108:1809–20. https://doi.org/10.1182/blood-2006-02-005686. e-pub ahead of print 2006/05/20
Gray RJ. A Class of K-Sample Tests for Comparing the Cumulative Incidence of a Competing Risk. Annal Stat, 1988; 16: 1141-1154.
Scrucca L, Santucci A, Aversa F. Competing risk analysis using R: an easy guide for clinicians. Bone Marrow Transplant. 2007;40:381–7. https://doi.org/10.1038/sj.bmt.1705727. e-pub ahead of print 2007/06/15
Team RC R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. pp. ISBN 3–900051–900007–900050, The R project for statistical computing website. Available: http://www.R-project.org/. Accessed 30 December 2013 (2012).
Shi JM, Meng XJ, Luo Y, Tan YM, Zhu XL, Zheng GF, et al. Clinical characteristics and outcome of isolated extramedullary relapse in acute leukemia after allogeneic stem cell transplantation: a single-center analysis. Leuk Res. 2013;37:372–7. https://doi.org/10.1016/j.leukres.2012.12.002. e-pub ahead of print 2013/01/26
Harris AC, Kitko CL, Couriel DR, Braun TM, Choi SW, Magenau J, et al. Extramedullary relapse of acute myeloid leukemia following allogeneic hematopoietic stem cell transplantation: incidence, risk factors and outcomes. Haematologica. 2013;98:179–84. https://doi.org/10.3324/haematol.2012.073189. e-pub ahead of print 2012/10/16
Lee KH, Lee JH, Choi SJ, Lee JH, Kim S, Seol M, et al. Bone marrow vs extramedullary relapse of acute leukemia after allogeneic hematopoietic cell transplantation: risk factors and clinical course. Bone Marrow Transplant. 2003;32:835–42. https://doi.org/10.1038/sj.bmt.1704223. e-pub ahead of print 2003/10/02
Kogut N, Tsai NC, Thomas SH, Palmer J, Paris T, Murata-Collins J, et al. Extramedullary relapse following reduced intensity allogeneic hematopoietic cell transplant for adult acute myelogenous leukemia. Leuk Lymphoma. 2013;54:665–8. https://doi.org/10.3109/10428194.2012.720375. e-pub ahead of print 2012/08/22
Kim JH, Stein A, Tsai N, Schultheiss TE, Palmer J, Liu A, et al. Extramedullary relapse following total marrow and lymphoid irradiation in patients undergoing allogeneic hematopoietic cell transplantation. Int J Radiat Oncol Biol Phys. 2014;89:75–81. https://doi.org/10.1016/j.ijrobp.2014.01.036. e-pub ahead of print 2014/04/15
Goyal SD, Zhang MJ, Wang HL, Akpek G, Copelan EA, Freytes C, et al. Allogeneic hematopoietic cell transplant for AML: no impact of pre-transplant extramedullary disease on outcome. Bone Marrow Transplant. 2015;50:1057–62. https://doi.org/10.1038/bmt.2015.82. e-pub ahead of print 2015/04/29
Bourlon C, Lipton JH, Deotare U, Gupta V, Kim DD, Kuruvilla J, et al. Extramedullary disease at diagnosis of AML does not influence outcome of patients undergoing allogeneic hematopoietic cell transplant in CR1. Eur J Haematol. 2017;99:234–9. https://doi.org/10.1111/ejh.12909. e-pub ahead of print 2017/05/31
Oshima K, Kanda Y, Yamashita T, Takahashi S, Mori T, Nakaseko C, et al. Central nervous system relapse of leukemia after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2008;14:1100–7. https://doi.org/10.1016/j.bbmt.2008.07.002. e-pub ahead of print 2008/09/23
Solh M, DeFor TE, Weisdorf DJ, Kaufman DS. Extramedullary relapse of acute myelogenous leukemia after allogeneic hematopoietic stem cell transplantation: better prognosis than systemic relapse. Biol Blood Marrow Transplant. 2012;18:106–12. https://doi.org/10.1016/j.bbmt.2011.05.023. e-pub ahead of print 2011/06/28
Yoshihara S, Ando T, Ogawa H. Extramedullary relapse of acute myeloid leukemia after allogeneic hematopoietic stem cell transplantation: an easily overlooked but significant pattern of relapse. Biol Blood Marrow Transplant. 2012;18:1800–7. https://doi.org/10.1016/j.bbmt.2012.05.010. e-pub ahead of print 2012/05/29
Conflict of interest
The authors declare that they have no conflict of interest.
About this article
Cite this article
Gunes, G., Goker, H., Demiroglu, H. et al. Extramedullary relapses of acute leukemias after allogeneic hematopoietic stem cell transplantation: clinical features, cumulative incidence, and risk factors. Bone Marrow Transplant 54, 595–600 (2019). https://doi.org/10.1038/s41409-018-0303-5
Prognosis and risk factors for central nervous system relapse after allogeneic hematopoietic stem cell transplantation in acute myeloid leukemia
Annals of Hematology (2021)
Better disease control before allogeneic stem cell transplantation is crucial to improve the outcomes of transplantation for acute myeloid leukemia patients with extramedullary disease
Bone Marrow Transplantation (2020)
Successful chimeric antigen receptor T cells therapy in extramedullary relapses of acute lymphoblastic leukemia after allogeneic hematopoietic stem cell transplantation
Bone Marrow Transplantation (2020)
Incidence, risk factors and outcome of extramedullary relapse after allogeneic hematopoietic stem cell transplantation in patients with adult acute lymphoblastic leukemia
Annals of Hematology (2020)