Hematological Oncology Hematol Oncol 2015 Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/hon.2201
Original Research Article
Allogeneic stem cell transplantation using lymphoablative rather than myeloablative conditioning regimen for relapsed or refractory lymphomas Jae-Ho Yoon1, Young-Woo Jeon1, Sung-Eun Lee1, Byung-Sik Cho1, Ki-Seong Eom1, Yoo-Jin Kim1, Seok Lee1, Hee-Je Kim1, Chang-Ki Min1, Jong-Wook Lee1, Woo-Sung Min1 and Seok-Goo Cho1,2* 1
Department of Hematology, Catholic Blood and Marrow Transplantation Center, Seoul St. Mary’s Hospital, The Catholic University of Korea, Seoul, Korea 2 Institute for Translational Research and Molecular Imaging, College of Medicine, The Catholic University of Korea, Seoul, Korea *Correspondence to: Professor Seok-Goo Cho, M.D., Ph.D., Department of Hematology, Catholic Blood and Marrow Transplantation Center, Seoul St. Mary’s Hospital, The Catholic University of Korea, 222, Banpo-daero, Seocho-gu 137701, Seoul, Korea. E-mail: [email protected]
Received 5 January 2015 Revised 4 February 2015 Accepted 18 February 2015
Abstract In relapsed or refractory non-Hodgkin lymphoma (NHL), allogeneic hematopoietic stem cell transplantation (allo-HSCT) provides graft-versus-lymphoma activity resulting in fewer incidences of relapse. However, therapy-related mortality (TRM) remains an important challenge. We attempted to introduce our reduced-intensity conditioning (RIC) regimen. From 2007 to 2013, we treated 28 relapsed or refractory NHLs with allo-HSCT. All were pre-conditioned with fludarabine [FLU, 180 mg/body surface area (BSA)/6 days] and melphalan (MEL, 70 mg/BSA/1 day); 25 (all but 3) were additionally treated with total body irradiation (TBI, 800 cGy/4Fx/2 days). Peripheral blood stem cells were collected from matched siblings (n = 10) or suitably matched unrelated (n = 18) donors. There were eight diffuse large B-cell lymphomas, seven peripheral T-cell lymphoma not otherwise specified, give lymphoblastic lymphomas, two mantle cell lymphomas, and six various other lymphomas. Of these patients, 10 relapsed after auto-HSCT, 5 relapsed after chemotherapy, and 13 were refractory lymphomas. After allo-HSCT, complete remission was achieved in 22 (78.5%) patients. After a median follow-up of 24.8 months, 3-year overall survival and disease-free survival were 62.4 and 59.2% and the 3-year TRM and relapse incidence were 14.9 and 28.6% respectively. Acute and chronic graft-versus-host diseases (GVHDs) were identified in 17 (≥Grade II in 12 patients) and 18 patients respectively, and the group with chronic GVHD showed favourable survival outcomes. In relapsed or refractory NHL, RIC-allo-HSCT using FLU + MEL + 800 cGy TBI showed favourable survival outcomes with acceptable TRM and relapse incidence. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: non-Hodgkin lymphoma; allogeneic; hematopoietic stem cell transplantation; refractory; relapse; reduced intensity conditioning regimen
Introduction Patients with non-Hodgkin lymphoma (NHL) who have relapsed, are refractory to salvage chemotherapy or have adverse risk factors such as advanced stage, high lactate dehydrogenase (LDH) level, or extramedullary organ or bone marrow (BM) involvement, may not achieve longterm disease-free survival (DFS) even after autologous hematopoietic stem cell transplantation (auto-HSCT), and a historical series suggested that the median survival of patients who relapse after auto-HSCT is only 3 months [1]. Recent registry data analysis from the European Group for Blood and Marrow Transplantation showed that highly selected patients with relapsed DLBCL after auto-HSCT
Copyright © 2015 John Wiley & Sons, Ltd.
might derive survival benefits from allogeneic (allo)-HSCT [2]. Although numerous improved salvage therapies and concomitant novel agents have been introduced, long-term survival or curability is still challenging. Given this situation, allo-HSCT for the treatment of aggressive relapsed or refractory NHL is appealing for the prospect of providing both a tumour-free graft and the graft-versuslymphoma (GVL) effect. However, there have been only limited experiences concerning allo-HSCT because of the higher regimen-related toxicity and mortality of myeloablative conditioning (MAC) regimens [3–6]. Radich et al. reported clinical outcomes of MAC–alloHSCT in patients relapsed after auto-HSCT that showed a higher rate of treatment-related mortality (TRM) up to
J-H Yoon et al.
78% [6]. However, several studies have reported on reduced intensity conditioning (RIC)-allo-HSCT regimens that showed the possibility of acceptable TRM with lower relapse incidence [7–9]. However, the long-term benefits of these procedures have yet to be elucidated. In relapsed diffuse large B-cell lymphomas (DLBCL), overall survival (OS) was reported equivalent between auto-HSCT and allo-HSCT, which was attributed to the risks regarding relapse and TRM [10,11]. A representative study by Rodriquez et al. showed that patients who received an RIC regimen had significantly higher relapse rates, but there were no differences in OS and progression-free survival (PFS) compared with those receiving an MAC regimen [12]. In indolent NHL-like follicular lymphoma, RIC-allo-HSCT appears to be an acceptable alternative approach. However, there are insufficient data on relapsed or refractory NHL. We have been treating these highly aggressive patients with NHL with allo-HSCT using both MAC and RIC regimens; however, we changed our strategy to RIC regimen after 2007. We report our experiences of acceptable survival outcomes, lower regimen-related toxicity, and lower TRM rate of an RIC pre-conditioning regimen consisting of fludarabine (FLU), melphalan (MEL), and total body irradiation (TBI), followed by allo-HSCT for relapsed or refractory lymphoma.
Materials and methods Patients Adult patients with relapsed or refractory lymphomas presenting from January 2007 to October 2013 at our institute were screened, and subjects who received allo-HSCT with an RIC regimen were selected for analysis. Follow-up was extended to March 2014 to ensure a minimum follow-up duration of 6 months. This retrospective study was approved by the institutional review board of Seoul St. Mary’s Hospital of The Catholic University of Korea in accord with the Declaration of Helsinki. We treated 28 relapsed or refractory patients with RIC-allo-HSCT after various salvage chemotherapies. Pathologic diagnosis was made by experienced lymphoma pathologists according to the World Health Organization 2008 formulations [13]. The individual patients were reviewed for demographic information; international prognostic index (IPI) [14]; response to firstline treatment and salvage treatments, including autoHSCT; disease status before allo-HSCT; and outcome.
Pre-transplant treatment strategy and response evaluation Pre-HSCT treatment was performed according to the institution’s guidelines for lymphoma. The first-line Copyright © 2015 John Wiley & Sons, Ltd.
chemotherapy for B-cell lymphomas was CHOP-base therapy [15] or alternatively applied hyper–cyclophosphamide, vincristine, doxorubicin, dexamethasone and high-dose methotrexate plus cytarabine [16] with additional rituximab for CD20-positive lymphomas. For T-cell lymphomas, we used a proMACE (prednisolone, cyclophosphamide, doxorubicin, etoposide)-cytaBOM (cytarabine, bleomycin, vincristine, MTX) regimen [15] as a first-line chemotherapy. For salvage chemotherapy, ifosfamide, carboplatin, etoposide and dexamethasone, doxorubicin, cytarabine, cisplatin regimens were alternatively used. After every three cycles of chemotherapy, response was evaluated with both computed tomography (CT) and fluorine 18-fluorodeoxyglucose positron-emission tomography (FDG-PET). The definition of complete remission (CR) was regression of the entire palpable mass to normal size on CT with negative FDG-PET. FDG-PET was evaluated according to the criteria of standardized uptake value [17]. Partial remission (PR) required a reduction of at least 50% in the sum of the products of the dimensions on CT with positive FDG-PET without newly developed lesion. Disease response and progression were retrospectively evaluated according to the revised version of the International Working Group response criteria for malignant lymphoma [18]. Chemosensitivity at the time of allo-HSCT was defined as achievement of CR or PR to the time of the most recent chemotherapy. Patients were considered to have chemorefractoriness if the response was less than a PR; if the disease continuously progressed during any chemotherapy, the state was defined as primarily refractory.
Preparative regimen, transplantation, and supportive care All patients were pre-conditioned with FLU 30 mg/m2 for six consecutive days (total 180 mg/m2) plus MEL 70 mg/m2 for 1 day. Twenty-five patients were additionally treated with 800 cGy TBI in four fractionated doses for 2 days. Stem cells were collected from fully matched siblings (n = 10) or suitably matched (450 IU/L) 22 (78.6%) Extranodal involvement 0~1 11 (39.3%) >1 17 (60.7%) BM involvement 15 (53.6%) ECOG 0~1 2 HCT-CI 0–1 >1 Response after 1st line chemotherapy CR PR Refractory (SD + PD) History of prior auto-HSCT Frontline auto-HSCT Salvage auto-HSCT Disease status before allo-HSCT Chemosensitive (CR + PR) Chemorefractory (SD + PD)
19 (67.9%) 9 (32.1%) 9 (32.1%) 19 (67.9% 12 (43.0%) 8 (28.5%) 8 (28.5%) 4 (14.3%)† 8 (28.5%)† 9 (32.1%, 4 + 5) 19 (67.9%)
†
Among 12 patients who were treated with auto-HSCT, 10 patients achieved CR, but all 12 patients progressed after auto-HSCT. Abbreviation: IPI, international prognostic index; LDH, lactate dehydrogenase; BM, bone marrow; ECOG, East Oncology Group; HCT-CI, Hematopoietic cell transplantation-specific comorbidity index; HSCT, hematopoietic stem cell transplantation; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease.
Hematol Oncol 2015 DOI: 10.1002/hon
J-H Yoon et al.
92.8%) with BM involvement (n = 15, 53.6%) and elevated LDH level (n = 22, 78.6%) at diagnosis. There were 17 (60.7%) patients who presented two or more extranodal involvements, and most of the patients had an East Oncology Group (ECOG) performance status 500/μl) and 13.4 days for platelet count (>50 000/μl for three consecutive days without transfusion). All 28 of the patients’ peripheral blood samples were examined for donor chimerism, and all samples showed chimerism greater than 97% at D + 30. After allo-HSCT, CR was identified in 22 (78.5%) patients. Among the 20 chemorefractory patients before
transplantation, 15 (75.0%) patients achieved CR after a median of 6.5 months (range of 3.1–11.5). The pre-HSCT disease status of six patients, who failed to achieve CR after allo-HSCT, was primary chemorefractoriness in 2 patients, relapsed state after auto-HSCT in 3 patients, and relapsed state after chemotherapy with chemorefractoriness in 1 patient. Of them, 5 patients died of disease progression, leaving 1 patient with DLBCL who is alive with stable disease status after treatment with 6 cycles of bendamustine plus rituximab chemotherapy. After a median follow-up of 24.8 months (7.0–92.1) in the surviving patients, the 3-year OS and DFS were 62.4 and 59.2% respectively, and the CIP and TRM at 3 years were 28.6 and 14.9% respectively (Figure 1). TRM occurred in 4 patients; the causes of death were pneumonia (n = 2) at 2.9 and 6.3 months, organ failure followed by hemorrhagic cystitis (n = 1) at 4.6 months, and chronic lung GVHD (n = 1) at 18.3 months. Figure 2 presents the survival outcomes according to pathologic subtype. T-cell lymphomas showed relatively favourable 3-year OS and DFS at 73.1 and 72.0% respectively, compared with 48.6 and 41.7% for B-cell lymphomas (Figure 2A). PTCLNOS showed favourable 3-year OS and DFS at 71.4 and 71.4% respectively, compared with 33.3 and 25.0% for DLBCL (Figure 2B). The relapse rate was 18.8% for T-cell lymphomas (14.3% for PTCL NOS) and 41.7% for B-cell lymphoma (62.5% for DLBCL), which was not statistically different (p = 0.183). Next, we also calculated OS between chemosensitive versus chemorefractory group and according
Figure 1. Clinical outcomes of relapsed or refractory NHL patients treated with allo-HSCT Copyright © 2015 John Wiley & Sons, Ltd.
Hematol Oncol 2015 DOI: 10.1002/hon
Allogeneic stem cell transplantation for relapsed or refractory lymphoma
Figure 2. (A) OS and DFS results according to lymphoma subtype. (B) Results of PTCL NOS and DLBCL
to hematopoietic cell transplantation comorbidity index (HCT-CI). Chemosensitive group and lower HCT-CI group showed trend towards favourable OS, but the results were not statistically significant (Figure 3A and 3B).
Regimen-related toxicity and GVHD Regimen-related toxicity and GVHD profiles are presented in Table 2. Our regimen—FLU + MEL + TBI 800 cGy— showed a tolerable toxicity profile with less than 20% for most of the complications except febrile neutropenia. There was no grade 4 oral mucositis or other gastrointestinal
symptoms. Within 1 month after transplantation, there was only one case of Clostridium difficile-associated diarrhoea and catheter-related infection. Among the 4 patients who were identified to have bacteremia, 3 had Gram-negative bacilli (i.e. Escherichia coli or Klebsiella pneumoniae) and 1 had Gram-positive cocci (i.e. Streptococcus viridans). Acute GVHD ≥ grade II and GVHD ≥ grade III were identified in 12 (42.8%) and 5 (17.9%) patients respectively, and chronic GVHD was identified in 18 (64.3%) patients. Among the 18 patients, classic de novo chronic GVHD was identified in 7 patients, while the other 11 patients presented interrupted chronic GVHD after treatment of acute GVHD, and there was no case of overlap syndrome. There
Figure 3. (A) OS result according to the chemosensitivity. (B) OS result according to HCT-CI Copyright © 2015 John Wiley & Sons, Ltd.
Hematol Oncol 2015 DOI: 10.1002/hon
J-H Yoon et al.
Table 2. Toxicity profile† (≥Grade 2) at peri-transplantation period and GVHD after stem cell infusion Total 28 patients
N
Gastrointestinal Oral mucositis Nausea Vomiting Diarrhoea
5 (17.8%) 4 (14.2%) 3 (10.7%) 4 (14.2%)
Febrile neutropenia Infection Pneumonia Clostridium difficile associated diarrhoea Bacteremia Catheter-related infection Organ dysfunction Toxic hepatitis
14 (50.0%) 0 (0.0%) 1 (3.5%) 4 (14.2%) 1 (3.5%) 2 (7.1%)
Acute GVHD (≥Grade 2) Gut GVHD Skin GVHD Gut and skin GVHD Hepatic GVHD
12 (42.8%) 3 (10.7%) 2 (7.1%) 6 (21.4%) 1 (3.6%)
Chronic GVHD Mild Moderate Severe
18 (64.3%) 6 (21.4%) 10 (35.7%) 2 (7.1%)
†
According to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 4.0.
were two cases of severe chronic GVHD that originated from scores 2 to 3 lung GVHD. Figure 4 shows that 18 patients with chronic GVHD had significantly superior OS and DFS rates compared with patients without chronic GVHD.
Discussion Relapsed or refractory lymphomas after salvage chemotherapy or high-dose chemotherapy followed by auto-HSCT is common [23,24]. In such cases, the prognosis is very poor and there are no established optimal treatment strategies. Nevertheless, allo-HSCT offers the potential of curative
treatment on the basis of the GVL effect [5,25,26]. In addition, given the evidence that prior rituximab therapy and early relapse less than 12 months after diagnosis were significant factors predicting poorer OS and PFS of relapsed DLBCL after salvage auto-HSCT [27], early allo-HSCT rather than salvage auto-HSCT might be considered for those high-risk relapsed patients. Thus, an early decision of allo-HSCT may have benefit for selected high-risk patients who no longer have options for curative treatment. Our regimen consisted of a maximal dose of fractionated TBI based on the criteria for RIC regimen with attenuated MEL and intensified FLU [28,29], from which we expected higher lymphoablative action with lower regimenrelated toxicity. The data showed acceptable clinical outcomes for relapsed or refractory lymphoma, and the overall CR rate was 78.5% for entire patients. CR rate was 75% even among the patients with chemorefractory lymphoma. In addition, we did indeed observe that the therapy was well tolerated by most of the patients during transplantation, and they experienced low-regimen-related toxicity. Although many previous studies showed that an MAC regimen results in a higher TRM rate, with the major causes of death being infection, interstitial pneumonitis, veno-occlusive disease, and GVHD [4,5], recent prospective study reported acceptable OS (66% of 3-year OS in patients with matched donor receiving ATG) with 34% of 1year non-relapse mortality of MAC regimen [30]. Besides, many reports also showed that RIC regimen found acceptable OS and PFS with lower TRM rate, although they showed higher relapse rates [7–9]. As a result, the optimal conditioning regimen remains undetermined, given that the results of long-term OS and PFS are similar between RIC and MAC regimens in the recent registry data [2,31,32] and the fact that more phase III studies should address this question. In addition, the role of TBI has yet to be clearly elucidated in both auto-HSCT and allo-HSCT, although one previous report on MCL showed a trend for longer OS in the TBI group [33]. The previous study had small number of patients to arrive at a definite conclusion, and another recent study showed no advantages of TBI for the same lymphoma subtype [34]. Although it is still
Figure 4. Patients with chronic GVHD showed favourable OS and DFS Copyright © 2015 John Wiley & Sons, Ltd.
Hematol Oncol 2015 DOI: 10.1002/hon
Allogeneic stem cell transplantation for relapsed or refractory lymphoma
debatable, lymphoma is generally considered to be sensitive to irradiation without an increase in TRM [4]. Freytes et al. showed TBI-containing regimen improved PFS in relapsed NHL patients after auto-HSCT [35]. In conclusion, the European Group for Blood and Marrow Transplantation cautiously states that TBI might be useful for patients in non-CR before auto-HSCT with an expectation of lower relapse rate compared with non-TBI-based regimen. Based on the fact that allo-HSCT is generally performed in relapsed or refractory lymphoma, we suggest that four fractionated 800 cGy TBI followed by alkylating agents is reasonable without significant regimen-related toxicity. Collectively, the 28 relapsed or refractory patients in our study had 3-year OS and DFS of 62.4 and 59.2% respectively, and 1-year and 3-year TRMs of 10.7 and 14.9% respectively. We suggest that these results are acceptable and superior compared with previous Western data dealing with RIC regimens, which showed OS and PFS
Original Research Article
Allogeneic stem cell transplantation using lymphoablative rather than myeloablative conditioning regimen for relapsed or refractory lymphomas Jae-Ho Yoon1, Young-Woo Jeon1, Sung-Eun Lee1, Byung-Sik Cho1, Ki-Seong Eom1, Yoo-Jin Kim1, Seok Lee1, Hee-Je Kim1, Chang-Ki Min1, Jong-Wook Lee1, Woo-Sung Min1 and Seok-Goo Cho1,2* 1
Department of Hematology, Catholic Blood and Marrow Transplantation Center, Seoul St. Mary’s Hospital, The Catholic University of Korea, Seoul, Korea 2 Institute for Translational Research and Molecular Imaging, College of Medicine, The Catholic University of Korea, Seoul, Korea *Correspondence to: Professor Seok-Goo Cho, M.D., Ph.D., Department of Hematology, Catholic Blood and Marrow Transplantation Center, Seoul St. Mary’s Hospital, The Catholic University of Korea, 222, Banpo-daero, Seocho-gu 137701, Seoul, Korea. E-mail: [email protected]
Received 5 January 2015 Revised 4 February 2015 Accepted 18 February 2015
Abstract In relapsed or refractory non-Hodgkin lymphoma (NHL), allogeneic hematopoietic stem cell transplantation (allo-HSCT) provides graft-versus-lymphoma activity resulting in fewer incidences of relapse. However, therapy-related mortality (TRM) remains an important challenge. We attempted to introduce our reduced-intensity conditioning (RIC) regimen. From 2007 to 2013, we treated 28 relapsed or refractory NHLs with allo-HSCT. All were pre-conditioned with fludarabine [FLU, 180 mg/body surface area (BSA)/6 days] and melphalan (MEL, 70 mg/BSA/1 day); 25 (all but 3) were additionally treated with total body irradiation (TBI, 800 cGy/4Fx/2 days). Peripheral blood stem cells were collected from matched siblings (n = 10) or suitably matched unrelated (n = 18) donors. There were eight diffuse large B-cell lymphomas, seven peripheral T-cell lymphoma not otherwise specified, give lymphoblastic lymphomas, two mantle cell lymphomas, and six various other lymphomas. Of these patients, 10 relapsed after auto-HSCT, 5 relapsed after chemotherapy, and 13 were refractory lymphomas. After allo-HSCT, complete remission was achieved in 22 (78.5%) patients. After a median follow-up of 24.8 months, 3-year overall survival and disease-free survival were 62.4 and 59.2% and the 3-year TRM and relapse incidence were 14.9 and 28.6% respectively. Acute and chronic graft-versus-host diseases (GVHDs) were identified in 17 (≥Grade II in 12 patients) and 18 patients respectively, and the group with chronic GVHD showed favourable survival outcomes. In relapsed or refractory NHL, RIC-allo-HSCT using FLU + MEL + 800 cGy TBI showed favourable survival outcomes with acceptable TRM and relapse incidence. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: non-Hodgkin lymphoma; allogeneic; hematopoietic stem cell transplantation; refractory; relapse; reduced intensity conditioning regimen
Introduction Patients with non-Hodgkin lymphoma (NHL) who have relapsed, are refractory to salvage chemotherapy or have adverse risk factors such as advanced stage, high lactate dehydrogenase (LDH) level, or extramedullary organ or bone marrow (BM) involvement, may not achieve longterm disease-free survival (DFS) even after autologous hematopoietic stem cell transplantation (auto-HSCT), and a historical series suggested that the median survival of patients who relapse after auto-HSCT is only 3 months [1]. Recent registry data analysis from the European Group for Blood and Marrow Transplantation showed that highly selected patients with relapsed DLBCL after auto-HSCT
Copyright © 2015 John Wiley & Sons, Ltd.
might derive survival benefits from allogeneic (allo)-HSCT [2]. Although numerous improved salvage therapies and concomitant novel agents have been introduced, long-term survival or curability is still challenging. Given this situation, allo-HSCT for the treatment of aggressive relapsed or refractory NHL is appealing for the prospect of providing both a tumour-free graft and the graft-versuslymphoma (GVL) effect. However, there have been only limited experiences concerning allo-HSCT because of the higher regimen-related toxicity and mortality of myeloablative conditioning (MAC) regimens [3–6]. Radich et al. reported clinical outcomes of MAC–alloHSCT in patients relapsed after auto-HSCT that showed a higher rate of treatment-related mortality (TRM) up to
J-H Yoon et al.
78% [6]. However, several studies have reported on reduced intensity conditioning (RIC)-allo-HSCT regimens that showed the possibility of acceptable TRM with lower relapse incidence [7–9]. However, the long-term benefits of these procedures have yet to be elucidated. In relapsed diffuse large B-cell lymphomas (DLBCL), overall survival (OS) was reported equivalent between auto-HSCT and allo-HSCT, which was attributed to the risks regarding relapse and TRM [10,11]. A representative study by Rodriquez et al. showed that patients who received an RIC regimen had significantly higher relapse rates, but there were no differences in OS and progression-free survival (PFS) compared with those receiving an MAC regimen [12]. In indolent NHL-like follicular lymphoma, RIC-allo-HSCT appears to be an acceptable alternative approach. However, there are insufficient data on relapsed or refractory NHL. We have been treating these highly aggressive patients with NHL with allo-HSCT using both MAC and RIC regimens; however, we changed our strategy to RIC regimen after 2007. We report our experiences of acceptable survival outcomes, lower regimen-related toxicity, and lower TRM rate of an RIC pre-conditioning regimen consisting of fludarabine (FLU), melphalan (MEL), and total body irradiation (TBI), followed by allo-HSCT for relapsed or refractory lymphoma.
Materials and methods Patients Adult patients with relapsed or refractory lymphomas presenting from January 2007 to October 2013 at our institute were screened, and subjects who received allo-HSCT with an RIC regimen were selected for analysis. Follow-up was extended to March 2014 to ensure a minimum follow-up duration of 6 months. This retrospective study was approved by the institutional review board of Seoul St. Mary’s Hospital of The Catholic University of Korea in accord with the Declaration of Helsinki. We treated 28 relapsed or refractory patients with RIC-allo-HSCT after various salvage chemotherapies. Pathologic diagnosis was made by experienced lymphoma pathologists according to the World Health Organization 2008 formulations [13]. The individual patients were reviewed for demographic information; international prognostic index (IPI) [14]; response to firstline treatment and salvage treatments, including autoHSCT; disease status before allo-HSCT; and outcome.
Pre-transplant treatment strategy and response evaluation Pre-HSCT treatment was performed according to the institution’s guidelines for lymphoma. The first-line Copyright © 2015 John Wiley & Sons, Ltd.
chemotherapy for B-cell lymphomas was CHOP-base therapy [15] or alternatively applied hyper–cyclophosphamide, vincristine, doxorubicin, dexamethasone and high-dose methotrexate plus cytarabine [16] with additional rituximab for CD20-positive lymphomas. For T-cell lymphomas, we used a proMACE (prednisolone, cyclophosphamide, doxorubicin, etoposide)-cytaBOM (cytarabine, bleomycin, vincristine, MTX) regimen [15] as a first-line chemotherapy. For salvage chemotherapy, ifosfamide, carboplatin, etoposide and dexamethasone, doxorubicin, cytarabine, cisplatin regimens were alternatively used. After every three cycles of chemotherapy, response was evaluated with both computed tomography (CT) and fluorine 18-fluorodeoxyglucose positron-emission tomography (FDG-PET). The definition of complete remission (CR) was regression of the entire palpable mass to normal size on CT with negative FDG-PET. FDG-PET was evaluated according to the criteria of standardized uptake value [17]. Partial remission (PR) required a reduction of at least 50% in the sum of the products of the dimensions on CT with positive FDG-PET without newly developed lesion. Disease response and progression were retrospectively evaluated according to the revised version of the International Working Group response criteria for malignant lymphoma [18]. Chemosensitivity at the time of allo-HSCT was defined as achievement of CR or PR to the time of the most recent chemotherapy. Patients were considered to have chemorefractoriness if the response was less than a PR; if the disease continuously progressed during any chemotherapy, the state was defined as primarily refractory.
Preparative regimen, transplantation, and supportive care All patients were pre-conditioned with FLU 30 mg/m2 for six consecutive days (total 180 mg/m2) plus MEL 70 mg/m2 for 1 day. Twenty-five patients were additionally treated with 800 cGy TBI in four fractionated doses for 2 days. Stem cells were collected from fully matched siblings (n = 10) or suitably matched (450 IU/L) 22 (78.6%) Extranodal involvement 0~1 11 (39.3%) >1 17 (60.7%) BM involvement 15 (53.6%) ECOG 0~1 2 HCT-CI 0–1 >1 Response after 1st line chemotherapy CR PR Refractory (SD + PD) History of prior auto-HSCT Frontline auto-HSCT Salvage auto-HSCT Disease status before allo-HSCT Chemosensitive (CR + PR) Chemorefractory (SD + PD)
19 (67.9%) 9 (32.1%) 9 (32.1%) 19 (67.9% 12 (43.0%) 8 (28.5%) 8 (28.5%) 4 (14.3%)† 8 (28.5%)† 9 (32.1%, 4 + 5) 19 (67.9%)
†
Among 12 patients who were treated with auto-HSCT, 10 patients achieved CR, but all 12 patients progressed after auto-HSCT. Abbreviation: IPI, international prognostic index; LDH, lactate dehydrogenase; BM, bone marrow; ECOG, East Oncology Group; HCT-CI, Hematopoietic cell transplantation-specific comorbidity index; HSCT, hematopoietic stem cell transplantation; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease.
Hematol Oncol 2015 DOI: 10.1002/hon
J-H Yoon et al.
92.8%) with BM involvement (n = 15, 53.6%) and elevated LDH level (n = 22, 78.6%) at diagnosis. There were 17 (60.7%) patients who presented two or more extranodal involvements, and most of the patients had an East Oncology Group (ECOG) performance status 500/μl) and 13.4 days for platelet count (>50 000/μl for three consecutive days without transfusion). All 28 of the patients’ peripheral blood samples were examined for donor chimerism, and all samples showed chimerism greater than 97% at D + 30. After allo-HSCT, CR was identified in 22 (78.5%) patients. Among the 20 chemorefractory patients before
transplantation, 15 (75.0%) patients achieved CR after a median of 6.5 months (range of 3.1–11.5). The pre-HSCT disease status of six patients, who failed to achieve CR after allo-HSCT, was primary chemorefractoriness in 2 patients, relapsed state after auto-HSCT in 3 patients, and relapsed state after chemotherapy with chemorefractoriness in 1 patient. Of them, 5 patients died of disease progression, leaving 1 patient with DLBCL who is alive with stable disease status after treatment with 6 cycles of bendamustine plus rituximab chemotherapy. After a median follow-up of 24.8 months (7.0–92.1) in the surviving patients, the 3-year OS and DFS were 62.4 and 59.2% respectively, and the CIP and TRM at 3 years were 28.6 and 14.9% respectively (Figure 1). TRM occurred in 4 patients; the causes of death were pneumonia (n = 2) at 2.9 and 6.3 months, organ failure followed by hemorrhagic cystitis (n = 1) at 4.6 months, and chronic lung GVHD (n = 1) at 18.3 months. Figure 2 presents the survival outcomes according to pathologic subtype. T-cell lymphomas showed relatively favourable 3-year OS and DFS at 73.1 and 72.0% respectively, compared with 48.6 and 41.7% for B-cell lymphomas (Figure 2A). PTCLNOS showed favourable 3-year OS and DFS at 71.4 and 71.4% respectively, compared with 33.3 and 25.0% for DLBCL (Figure 2B). The relapse rate was 18.8% for T-cell lymphomas (14.3% for PTCL NOS) and 41.7% for B-cell lymphoma (62.5% for DLBCL), which was not statistically different (p = 0.183). Next, we also calculated OS between chemosensitive versus chemorefractory group and according
Figure 1. Clinical outcomes of relapsed or refractory NHL patients treated with allo-HSCT Copyright © 2015 John Wiley & Sons, Ltd.
Hematol Oncol 2015 DOI: 10.1002/hon
Allogeneic stem cell transplantation for relapsed or refractory lymphoma
Figure 2. (A) OS and DFS results according to lymphoma subtype. (B) Results of PTCL NOS and DLBCL
to hematopoietic cell transplantation comorbidity index (HCT-CI). Chemosensitive group and lower HCT-CI group showed trend towards favourable OS, but the results were not statistically significant (Figure 3A and 3B).
Regimen-related toxicity and GVHD Regimen-related toxicity and GVHD profiles are presented in Table 2. Our regimen—FLU + MEL + TBI 800 cGy— showed a tolerable toxicity profile with less than 20% for most of the complications except febrile neutropenia. There was no grade 4 oral mucositis or other gastrointestinal
symptoms. Within 1 month after transplantation, there was only one case of Clostridium difficile-associated diarrhoea and catheter-related infection. Among the 4 patients who were identified to have bacteremia, 3 had Gram-negative bacilli (i.e. Escherichia coli or Klebsiella pneumoniae) and 1 had Gram-positive cocci (i.e. Streptococcus viridans). Acute GVHD ≥ grade II and GVHD ≥ grade III were identified in 12 (42.8%) and 5 (17.9%) patients respectively, and chronic GVHD was identified in 18 (64.3%) patients. Among the 18 patients, classic de novo chronic GVHD was identified in 7 patients, while the other 11 patients presented interrupted chronic GVHD after treatment of acute GVHD, and there was no case of overlap syndrome. There
Figure 3. (A) OS result according to the chemosensitivity. (B) OS result according to HCT-CI Copyright © 2015 John Wiley & Sons, Ltd.
Hematol Oncol 2015 DOI: 10.1002/hon
J-H Yoon et al.
Table 2. Toxicity profile† (≥Grade 2) at peri-transplantation period and GVHD after stem cell infusion Total 28 patients
N
Gastrointestinal Oral mucositis Nausea Vomiting Diarrhoea
5 (17.8%) 4 (14.2%) 3 (10.7%) 4 (14.2%)
Febrile neutropenia Infection Pneumonia Clostridium difficile associated diarrhoea Bacteremia Catheter-related infection Organ dysfunction Toxic hepatitis
14 (50.0%) 0 (0.0%) 1 (3.5%) 4 (14.2%) 1 (3.5%) 2 (7.1%)
Acute GVHD (≥Grade 2) Gut GVHD Skin GVHD Gut and skin GVHD Hepatic GVHD
12 (42.8%) 3 (10.7%) 2 (7.1%) 6 (21.4%) 1 (3.6%)
Chronic GVHD Mild Moderate Severe
18 (64.3%) 6 (21.4%) 10 (35.7%) 2 (7.1%)
†
According to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 4.0.
were two cases of severe chronic GVHD that originated from scores 2 to 3 lung GVHD. Figure 4 shows that 18 patients with chronic GVHD had significantly superior OS and DFS rates compared with patients without chronic GVHD.
Discussion Relapsed or refractory lymphomas after salvage chemotherapy or high-dose chemotherapy followed by auto-HSCT is common [23,24]. In such cases, the prognosis is very poor and there are no established optimal treatment strategies. Nevertheless, allo-HSCT offers the potential of curative
treatment on the basis of the GVL effect [5,25,26]. In addition, given the evidence that prior rituximab therapy and early relapse less than 12 months after diagnosis were significant factors predicting poorer OS and PFS of relapsed DLBCL after salvage auto-HSCT [27], early allo-HSCT rather than salvage auto-HSCT might be considered for those high-risk relapsed patients. Thus, an early decision of allo-HSCT may have benefit for selected high-risk patients who no longer have options for curative treatment. Our regimen consisted of a maximal dose of fractionated TBI based on the criteria for RIC regimen with attenuated MEL and intensified FLU [28,29], from which we expected higher lymphoablative action with lower regimenrelated toxicity. The data showed acceptable clinical outcomes for relapsed or refractory lymphoma, and the overall CR rate was 78.5% for entire patients. CR rate was 75% even among the patients with chemorefractory lymphoma. In addition, we did indeed observe that the therapy was well tolerated by most of the patients during transplantation, and they experienced low-regimen-related toxicity. Although many previous studies showed that an MAC regimen results in a higher TRM rate, with the major causes of death being infection, interstitial pneumonitis, veno-occlusive disease, and GVHD [4,5], recent prospective study reported acceptable OS (66% of 3-year OS in patients with matched donor receiving ATG) with 34% of 1year non-relapse mortality of MAC regimen [30]. Besides, many reports also showed that RIC regimen found acceptable OS and PFS with lower TRM rate, although they showed higher relapse rates [7–9]. As a result, the optimal conditioning regimen remains undetermined, given that the results of long-term OS and PFS are similar between RIC and MAC regimens in the recent registry data [2,31,32] and the fact that more phase III studies should address this question. In addition, the role of TBI has yet to be clearly elucidated in both auto-HSCT and allo-HSCT, although one previous report on MCL showed a trend for longer OS in the TBI group [33]. The previous study had small number of patients to arrive at a definite conclusion, and another recent study showed no advantages of TBI for the same lymphoma subtype [34]. Although it is still
Figure 4. Patients with chronic GVHD showed favourable OS and DFS Copyright © 2015 John Wiley & Sons, Ltd.
Hematol Oncol 2015 DOI: 10.1002/hon
Allogeneic stem cell transplantation for relapsed or refractory lymphoma
debatable, lymphoma is generally considered to be sensitive to irradiation without an increase in TRM [4]. Freytes et al. showed TBI-containing regimen improved PFS in relapsed NHL patients after auto-HSCT [35]. In conclusion, the European Group for Blood and Marrow Transplantation cautiously states that TBI might be useful for patients in non-CR before auto-HSCT with an expectation of lower relapse rate compared with non-TBI-based regimen. Based on the fact that allo-HSCT is generally performed in relapsed or refractory lymphoma, we suggest that four fractionated 800 cGy TBI followed by alkylating agents is reasonable without significant regimen-related toxicity. Collectively, the 28 relapsed or refractory patients in our study had 3-year OS and DFS of 62.4 and 59.2% respectively, and 1-year and 3-year TRMs of 10.7 and 14.9% respectively. We suggest that these results are acceptable and superior compared with previous Western data dealing with RIC regimens, which showed OS and PFS
