- Email: [email protected]
Is there a “gold” standard treatment for patients with isolated myeloid sarcoma?
Biomedicine & Pharmacotherapy 67 (2013) 72–77
Available online at
www.sciencedirect.com
Original article
Is there a ‘‘gold’’ standard treatment for patients with isolated myeloid sarcoma? Darko Antic a,*,b, Ivo Elezovic a,b, Natasa Milic b,c, Nada Suvajdzic a,b, Ana Vidovic a,b, Maja Perunicic a, Irena Djunic a, Mirjana Mitrovic a, Dragica Tomin a,b a
Clinic for hematology, Clinical Center Serbia, Koste Todorovica 2, 11000 Belgrade, Serbia Faculty of Medicine, University of Belgrade, Belgrade, Serbia c Institute for medical statistics and informatics, Belgrade, Serbia b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 1 September 2012 Accepted 22 October 2012
Isolated myeloid sarcoma is an extramedullary tumor of immature myeloid cells defined by the absence of leukemia history, myelodisplastic syndrome, or myeloproliferative neoplasma with a negative bone marrow biopsy. Myeloid sarcoma is a very rare condition, and few cases have been reported. We reviewed data of 12 patients with isolated myeloid sarcoma managed at a single center to determine the possible prognostic factors affecting patient survival, such as age, sex, type, localization, and treatment options. Patients were mostly men (n = 8), with a median age of 39 years. Patients were initially treated with chemotherapy (n = 7) or surgery (n = 5). In three patients, hematopoietic stem cell transplantation was performed. During the follow-up period, nine patients died. The median overall survival was 13 months, while event-free survival was 8 months. Regarding initial treatment strategy, no significant difference in overall survival was observed. Both chemotherapy and hematopoietic stem cell transplantation independently improved event-free survival. In addition, patients who received chemotherapy combined with hematopoietic stem cell transplantation had significantly longer eventfree survival than those treated with chemotherapy alone. Age < 40 years together with chemotherapy/ hematopoietic stem cell transplantation significant affected event-free survival. Based on our results, the treatment of myeloid sarcoma requires a systemic rather than a localized approach with surgery or radiotherapy. While prospective evaluations are needed, chemotherapy with allogenic hematopoietic stem cell transplantation should be considered as the optimal therapy for isolated myeloid sarcoma. ß 2012 Elsevier Masson SAS. All rights reserved.
Keywords: Myeloid sarcoma Chemotherapy Stem cell transplantation
1. Introduction Myeloid sarcoma (MS), also known as granulocytic/monoblastic sarcoma, extramedullary myeloid tumors, or chloroma, is a tumor mass comprising blasts or immature cells of the myeloid series, occurring at an anatomical site other than the bone marrow [1]. The condition was first described in 1811 by Burns [2] and later further described by King as tumors displaying a predominantly green color, due to the presence of myeloperoxidase (MPO) [3]. In 1966, Rappaport proposed the term ‘‘granulocytic sarcoma’’ [4], and finally, in 2002, the term myeloid sarcoma was accepted by World Health Organization (WHO) [1]. MS can occur in patients with active AML and in patients with chronic myeloproliferative disease, where it may occur as the first manifestation of blast transformation, the first manifestation of AML relapse in previously treated patients, or isolated MS in patients without bone marrow infiltration [1]. In almost 90% of untreated patients who initially had no hematological disorder, if untreated, AML develops within 10.5 to 11 months. The incidence of MS is
* Corresponding author. Tel.: +38111 30 65 112; fax: +31811 30 65 112. E-mail address: [email protected] (D. Antic). 0753-3322/$ – see front matter ß 2012 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.biopha.2012.10.014
unknown and believed to be underestimated. Isolated MS occurs in 2/ 1,000,000 adults, and in 0.7/1,000,000 children. Thus far, the involvement of the skin, periosteum, bowels, lymph nodes, genital system, central nervous system, heart, and lungs has been reported [5–12]. Several markers have been employed for identification of the myeloid origin of tumor infiltration in MS. Immunostaining with antibodies to myeloperoxidase (MPO), lysosomes, and chloroacetate esterase are important for identification of MS, bearing in mind that the myeloblasts in MS have an antigen profile similar to that present in the blasts and precursor cells of acute myeloid leukemia (AML) [10,11]. Herein, we analyze data from a single center in 12 patients with isolated MS to identify putative prognostic factors with a focus on accurate and prompt diagnosis and timely appropriate treatment to reduce the risk of progression to AML. 2. Materials and methods 2.1. Patient selection This study included 12 patients with isolated MS, diagnosed at the Clinic for Hematology of Clinical Center Serbia during the
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77
period of 2002 to 2012. This group included patients who had pathohistologically proved MS based on WHO criteria [1], and who, when diagnosed with MS, did not have any signs of AML, chronic myeloproliferative disease, or myelodysplastic syndromes (MDS) in the bone marrow. All data were analyzed in order to present clinico-pathological characteristics, therapy, and outcomes. Patient evaluation included analysis of history, physical examination, complete blood count, bio-humoral status, bone marrow analysis, and available radiological examinations (radiography, ultrasonography, and computed tomography). Chemotherapy (CT) had been used in accordance with standard protocols applied for the treatment of AML at the Institute for Hematology. 2.2. Pathohistological analysis Morphologic examination of biopsy specimens and immunohistochemical analysis were used to determine the myeloid origin of tumors and to rule out lymphoproliferative diseases and mesenchymal malignancies. The following monoclonal antibodies were used: CD13, CD33, CD34, MPO, Ly, CD68, CD43, HLA-DR, and CD117 (myeloid series antigens); antibodies to T and B cell markers (CD3, CD5, CD20, and CD79a); and CD15 and CD30 that were used to rule out non-Hodgkins lymphoma (NHL) and Hodgkins lymphoma (HL). Desmin, S-100, and cytokeratin abundance was determined in several patients to rule out sarcoma or malignant melanoma. 2.3. Statistical data analysis In this study, we used methods of descriptive and analytical statistics. Descriptive statistics for continuous variables are presented as median values with ranges or 95% confidence intervals for medians. Categorical data are presented by absolute numbers with percentages. Overall survival (OS), relapse-free survival (RFS), leukemia-free survival (LFS), and event-free survival (EFS), as well as cumulative incidences (CI) of these events were estimated using the Kaplan–Meier method. Definitions used to measure time to survival categories are presented in Table 1. Differences between the Kaplan–Meier survival curves of applied therapeutic approaches were evaluated by the log rank test. A P value of < 0.05 was considered statistically significant.
73
3. Results The 12 patients included in this study comprised mostly men (n = 8), with a median age of 39 years (range: 19–67 years). The MS-affected sites included the spine, tonsils, testis, heart, lymph nodes, mediastinum (two patients), skull base, uterus (two patients), spleen, and multiple gastrointestinal localizations (the stomach, duodenum, and bowel). MPO and HLA-DR were positive in ten and nine patients, respectively, and presented the most commonly expressed markers. In two patients, MPO staining was not performed for technical reasons. One of these two patients had clearly positive monoblastic series markers, while myeloid origin in the other patient was confirmed based on positive results for HLA-DR, CD117, and CD34. Five patients were CD68 positive and were therefore classified as having MS. Analysis of Ki67 expression was performed in four patients. Mediastinal MS exhibited the highest proliferative activity of 80%. Initially, three patients had been previously misdiagnosed, as Ewing sarcoma (spinal MS), eosinophilic granuloma (skull base MS), and mucosa-associated lymphoid tissue (MALT) lymphoma (the stomach, duodenum, and bowel). At the time of MS diagnosis, examination of the bone marrow showed normal features in all patients. The initial therapeutic approach was CT in seven patients and surgical resection in five patients. Within the group of patients treated with CT, two patients initially underwent partial surgical tumor resection based on vital indications – progressive neurological failure in a patient with spinal MS, and superior vena cava syndrome in a patient with cardiac MS. After one and two CT cycles, respectively, local radiotherapy was applied in both patients. Moreover, in two patients, surgical tumor resection (tonsillectomy and splenectomy, respectively) was performed for diagnostic purposes. All patients treated with CT received the standard induction therapy of 3 days with an anthracycline and 7 days with cytarabine (3 + 7). In three patients with complete remission (CR), we continued high-dose therapy followed by autologous (n = 1) and allogeneic (n = 2) hematopoietic stem cell transplantation (auto- and allo-HSCT, respectively) from related donors. Therapeutic approaches and outcomes are shown in Table 2. One of the patients treated with allo-HSCT developed relapse (isolated breast MS) and died. Two patients are currently in CR.
Table 1 Outcome measures in isolated myeloid sarcoma (MS). Category
Definition
Overall survival (OS)
Defined for all patients of a study; measured from the date of entry into a study to the date of death from any cause; patients not known to have died at last follow-up are censored on the date they were last known to be alive
Relaps free survival (RFS)
Defined only for patients achieving CR or CRI; measured from the date of achievement of a remission until the date of relapse or death from any cause; patients not known to have relapsed or died at last follow-up are censored on the date they were last examined
Cumulative incidence of relaps (CI of relaps)
Defined for all patients achieving CR or CRI; measured from the date of achievement of a remission until the date of relapse; patients not known to have relapsed are censored on the date they were last examined; patients who died without relapse are counted as a competing cause of failure
Leukemia-free survival (LFS)
Defined for all patients of a study; measured from the date of entry into a study to the date of occurrence of systemic leukemia or death from any cause; patients not known to have systemic leukemia or died at last follow-up are censored on the date they were last examined
Cumulative incidence of leukemia (CI of leukemia)
Defined for all patients of a study; measured from the date of entry into a study to the date of occurrence of systemic leukemia; patients not known to have systemic leukemia are censored on the date they were last examined; patients who died without systemic leukemia are counted as a competing cause of failure
Event-free survival (EFS)
Defined for all patients of a study; measured from the date of entry into a study to the date of extramedullary relapse, occurrence of systemic leukemia or death from any cause; patients not known to have any of these events are censored on the date they were last examined
CR: complete remission; CRI: complete remission with incomplete recovery.
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77
74
Table 2 Disease localization, treatment strategy, presence of extramedullary relapse or AML development and survival time. Patient/gender/age
MS site
Therapy
EM relapse/months/localisation
AML/months
Follow-up/months
1/m/24 2/m/39 3/m/58 4/m/38 5/f/41 6/f/67 7/m/26 8/f/19 9/m/31 10/f/28 11/m/40 12/m/57
Spine Tonsil Testis Heart Uterus Uterus Limph nodes Mediastinum Skull base GIT Spleen Mediastinum
CHT + autoPBSCT CHT + alloPBSCT Surgery CHT Surgery Surgery CHT CHT + alloPBSCT Surgery Surgery CHT CHT
No No Yes/6/skin No No No No Yes/18/breast Yes/4/skull base No No No
No No No No Yes/2 Yes/13 Yes/8 No No Yes/3 No No
Live/62 Live/50 Death/35 Death/6 Live/79 Death/14 Death/13 Death/28 Death/10 Death/3 Death/4 Death/8
M: male; f: female; CHT: chemotherapy; HSCT: hematopoietic stem cell transplantation.
Two patients treated with CT died during the course of treatment from sepsis, as a result of febrile neutropenia in one case, and multiple organ failure caused by disseminated disease in the other. Eight months after diagnosis, bone marrow aplastic anemia followed by evolution to AML occurred in a patient with lymph node MS. Surgical resection as the initial therapeutic method was performed in five patients, and all of them developed AML or extramedullary relapse. Three patients with MS developed AML within the postsurgical period of 13 months. One of them has exhibited CR and is alive at the time of study, 79 months after the application of systemic therapy. In the two other patients who underwent surgical resection, extramedullary relapse occurred after 4 and 6 months, and both patients did not survive. The incidence of AML and extramedullary relapse was significantly higher in patients initially treated by surgical resection only (P = 0.028). During the follow-up period, nine patients died. The survival outcomes and causes of death are presented in Table 3. The Kaplan–Meier estimate of median OS was 13 months (95% CI: 6–19 months) (Fig. 1), and the shortest OS was observed in a patient with multiple gastrointestinal localization (3 months). The patients undergoing CT had an OS of 13 months (95% CI: 1–25 months), whereas patients treated with surgery alone had an OS of 14 months (95% CI: 5–22 months). No significant difference in OS was observed between these two groups of patients (P > 0.05) (Fig. 2). Median survival time without the occurrence of systemic leukemia or extramedullary relapse was 8 months (95% CI: 3–12
months) and 14 months (95% CI: 1–30 months), respectively. The Kaplan–Meier estimate of median EFS was 6 months (95% CI: 3–9 months) (Fig. 3). Patients who underwent CT had a significantly longer EFS time compared to those who did not undergo CT, 8 vs. 4 months, respectively (log rank = 3.988, P = 0.046) (Fig. 4). A statistically significant difference in EFS was observed between patients undergoing HSCT and patients without HSCT (Log Rank = 7.608, P = 0.006). A statistically significant difference was found in EFS between patients treated with CT only (without HSCT) (median 6 months; 95% CI 3–8 months) and patients undergoing CT with HSCT (median not reached) (Log rank = 5.300; P = 0.021) (Fig. 5). Other variables that were analyzed, such as sex, adjuvant radiotherapy, underlying disorders, and MS, did not affect OS or EFS, whereas age < 40 had a significant influence on EFS, in combination with CT and HSCT as applied therapeutic options.
4. Discussion This study monitored the treatment of 12 patients with MS over a 10-year period at a single center. Thus far, most published literature on MS comprises case reports; only a few published
Table 3 Survival outcomes of patients with isolated myeloid sarcoma. Survival outcomes 5-year OS % Median
25 12% 13 months
5-year LFS % Median CI of leukemia
17 10% 8 months 44 18%
5-year RFS % Median CI of relaps
37 17% 14 months 53 20%
5-year EFS % Median
17 11% 6 months
Deaths
9
OS; overall survival; LFS; leukemia-free survival; CI: cumulative incidence; RFS: relaps free survival; EFS: event-free survival.
Fig. 1. Kaplan–Meier curve of median overall survival.
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77
75
Fig. 2. Kaplan–Meier curve of overall survival in patient treated with chemotherapy and surgery.
Fig. 4. Kaplan–Meier curve of event-free survival in patient treated with chemotherapy and surgery.
studies are based on a larger series of patients. A large proportion of these studies discuss patients with leukemic MS, whereas, in our study, we investigated only patients with isolated MS.
Diagnosis of MS in patients poses a huge challenge, and it is therefore necessary to analyze the expression of more than few markers that confirm the myeloid origin of neoplastic cells. Earlier, in the absence of hematological disorders, within a single series of patients, as many as 75% were initially diagnosed with a condition other than MS, NHL in most cases [13]. In another large series, 46% of patients were not initially diagnosed as MS, whereas in our group, three patients (25%) were initially misdiagnosed [14]. A number of studies have been published on the immunoprofile of MS. MPO, CD68, CD43, and lysozyme are frequently expressed in MS, but none of these are specific markers. To establish a definitive diagnosis, we recommend that the minimum panel of immunohistochemical markers should include the previously mentioned markers in addition to markers such as CD33, CD34, and CD117. Considering that cytogenetic abnormalities are detected in approximately 55% of tested cases with MS, the identification of an AML-associated abnormality (i.e., trisomy 8, t[8;21][q22;q22] or inv[16][p13.1q22]/t[16;16][p13.1;q22]) may help to arrive at the correct diagnosis (4,11). Several studies have summarized the outcomes of patients with MS. In 1981, Neiman et al. [15] published a study of 50 patients with MS, 15 of which had isolated MS. The majority of patients who did not have AML developed it within a period of 1 to 49 months (10 months on average). This result is in agreement with the results of the study by Meis et al., who reported that the average time for development of AML is 10.5 months [13]. In this series, prognostic indicators suggestive of causes influencing the increased risk for development of AML were not identified. No statistical correlation was found between the type of therapy administered and survival. In our group, 4 patients developed AML, with a median time of 24 months (95% CI 1–50 months), similar to the results of previous studies.
Fig. 3. Kaplan–Meier curve of median event-free survival.
76
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77
Fig. 5. Kaplan–Meier curve of event-free survival in patient treated with chemotherapy with or without hematopoietic stem cell transplantation.
Yamauchi and Yasuda investigated the association between therapeutic regimens and disease-free survival (DFS) in a group of 74 patients with MS. DFS was significantly longer in patients treated with systemic chemotherapy (median 12 months) than in those treated with surgery (3 months) or local irradiation (6 months) [16]. Tsimberidou et al. reported data regarding the incidence, therapy, and outcome of MS in 1520 patients with AML, and in 402 patients with high-risk MDS. MS occurred in 20 patients, 80% of whom received AML-type chemotherapy. Thirteen patients (65%) achieved CR, median OS was 20 months, and median DFS was 12 months [17]. Pileri et al. [18] reported the outcome of 67 patients with MS, of which 25 (27%) presented with non-leukemic MS. Among them, 47 received chemotherapy (70.1%), six underwent allogeneic bone marrow transplant (BMT) (allo-BMT) (9.0%), and 4 underwent autologous BMT (auto-BMT) (6.0%). At a median follow-up period of 150 months, only seven patients, six of whom underwent alloBMT, were alive (10.5%) and in CR. Overall, patients treated with auto- or allo-BMT exhibited longer survival than those who received conventional therapy (OS at 48 months: 76 vs. 0%). Paydas et al. [6] analyzed 32 patients with MS, seven of whom were initially misdiagnosed with NHL. Systemic chemotherapy had been administered to all the patients. The median survival was 9.5 months; seven patients survived for > 1 year, while 1 patient lived > 8 years. Breccia et al. [10] analyzed 12 patients with MS. In three patients with isolated MS, median survival to development of AML was 5 months, median survival was 7 months, and one patient was alive 49 months after BMT. A recent study published by Lan et al. evaluated sex, age, location, MS antedating leukemia, underlying disorders, treatment type, and stem cell transplantation as prognostic factors deter-
mining survival in 24 patients with MS (seven with isolated sarcoma, others with AML, chronic myeloid leukemia, or MDS). The 5-year survival rate for patients with MS was 21%. Patients receiving chemotherapy had a significantly longer OS period compared to those who did not (P = 0.0009). In our group, the 5-year OS rate for patients with isolated MS was 25%, with the median OS period of 13 months, whereas the 5-year EFS rate was 17% with median OS period of 6 months. Treatment strategy did not affect OS, but EFS was significantly longer in patients treated with chemotherapy and HSCT (P = 0.046 and P = 0.006, respectively) [19]. The optimal timing and treatment of isolated MS are not clear, but delayed or inadequately systemically treated isolated MS will almost always progress to AML [20]. In our practice, we use remission–induction chemotherapy similar to that used for AML. Postremission chemotherapy has not been adequately studied in isolated MS; therefore, one of the most important questions in the treatment of MS is the role of HSCT. Patients in the current study treated with HSCT had a longer EFS. Chevallier et al. [21] confirmed this approach by assessing the outcome of 30 patients with isolated MS who underwent allo-HSCT. Patients with isolated MS had a 5year OS and LFS of 45% and 33%, respectively, with relatively acceptable toxicity (overall 5-year non-relapse moratlity 17%). Considering the entire cohort, in univariate analysis, factors associated with higher LFS and lower incidence of relapse were CR status at transplant and cytogenetics. The only factor associated with high non-relapse mortality was patient sex (male: 27% 6% vs. female: 10% 5%, P = 0.04). In multivariate analysis, high-risk cytogenetics remained significantly associated with a poorer LFS and an increased incidence of relapse. However, the latter should be considered with caution as the cytogenetic data were missing in most cases of isolated MS. According to the results in our series of patients, as well as the published data, it is clear that in cases of suspected MS, employment of a wide spectrum of antibodies during immunohistochemical work-up of tumor tissue is mandatory. Timely diagnosis has a huge influence on treatment results, i.e., longer survival as a consequence of adequate therapy. While prospective evaluations are needed, chemotherapy with allo-HSCT could be considered as the optimal therapy for isolated MS. Moreover, the role of new molecular prognostic markers such as NPM1 or FLT3ITD mutations in MS should be investigated. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. References [1] Pileri SA, Orayi A, Falini B. Myeloid sarcoma. In: Swerdlow S, Campo E, Harris NL., Jaffe ES., Pileri SA., Stein H, Thiele J, Vardiman JW., editors. WHO classification of tumours of haematopoetic and lymphoid tissues. Lyon: IARC; 2008. p. 140–1. [2] Burns A. Observation of surgical anatomy, head and neck. Edinburgh: Thomas Royce and co; 1811. p. 364–6. [3] King A. A case of chloroma. Monthly J med 1853;17:97. [4] Rappaport H. Tumors of the hematopoetic system. Armed Forces Inst Pathol 1966;241–3. [5] Hutchinson RE, Kurec AS, Davey FR. Granulocytic sarcoma. Clin Lab Med 1990;10:889–901. [6] Paydas S, Zorludemir S, Ergin M. Granulocytic sarcoma: 32 cases and review of the literature. Leuk Lymphoma 2006;47(12):2527–41. [7] Antic D, Verstovsek S, Elezovic I, Grujicic D, Gotic M, Bila J, et al. Spinal epidural granulocytic sarcoma in non-leukemic patient. Int J Hematol 2009;89(1):95–7. [8] Antic D, Vuckovic M, Elezovic I. Right atrial myeloid sarcoma causing superior vena cava syndrome. Br J Haematol 2008;141(2):134. [9] Audouin J, Comperat E, Le Tourneau A, Camilleri-Broe¨t S, Adida C, Molina T, et al. Myeloid sarcoma: clinical and morphologic criteria useful for diagnosis. Int J Surg Pathol 2003;11(4):271–82. [10] Alexiev BA, Wang W, Ning Y, Chumsri S, Gojo I, Rodgers WH, et al. Myeloid sarcomas: a histologic, immunohistochemical, and cytogenetic study. Diagn Pathol 2007;2:42.
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77 [11] Klco JM, Welch JS, Nguyen TT, Hurley MY, Kreisel FH, Hassan A, et al. State of the art in myeloid sarcoma. Int J Lab Hematol 2011;33(6):555–65. [12] Breccia M, Mandelli F, Petti MC, D’Andrea M, Pescarmona E, Pileri SA, et al. Clinico-pathological characteristics of myeloid sarcoma at diagnosis and during follow-up: report of 12 cases from a single institution. Leuk Res 2004;28(11):1165–9. [13] Meis JM, Butler JJ, Osborne BM, Manning JT. Granulocytic sarcoma in nonleukemic patients. Cancer 1986;58(12):2697–709. [14] Byrd JC, Edenfield WJ, Shields DJ, Dawson NA. Extramedullary myeloid cell tumors in acute nonlymphocytic leukemia: a clinical review. J Clin Oncol 1995;13(7):1800–16. [15] Neiman RS, Barcos M, Berard C, Bonner H, Mann R, Rydell RE, et al. Granulocytic sarcoma: a clinicopathologic study of 61 biopsed cases. Cancer 1981;48:1426–37. [16] Yamauchi K, Yasuda M. Comparison in treatments of nonleukemic granulocytic sarcoma: report of two cases and a review of 72 cases in the literature. Cancer 2002;94(6):1739–46.
77
[17] Tsimberidou AM, Kantarjian HM, Wen S, Keating MJ, O’Brien S, Brandt M, et al. Myeloid sarcoma is associated with superior event-free survival and overall survival compared with acute myeloid leukemia. Cancer 2008;113(6):1370–8. [18] Pileri SA, Ascani S, Cox MC, Campidelli C, Bacci F, Piccioli M, et al. Myeloid sarcoma: clinico pathologic, phenotypic and cytogenetic analysis of 92 adult patients. Leukemia 2007;21(2):340–50. [19] Lan TY, Lin DT, Tien HF, Yang RS, Chen CY, Wu K. Prognostic factors of treatment outcomes in patients with granulocytic sarcoma. Acta Haematol 2009;122(4):238–46. [20] Bakst RL, Tallman MS, Douer D, Yahalom J. How I treat extramedullary acute myeloid leukemia. Blood 2011;118(14):3785–93. [21] Chevallier P, Labopin M, Cornelissen J, Socie´ G, Rocha V, Mohty M. ALWP of EBMT Allogeneic hematopoietic stem cell transplantation for isolated and leukemic myeloid sarcoma in adults: a report from the Acute Leukemia Working Party of the European group for blood and marrow transplantation. Haematologica 2011;96(9):1391–4.
Available online at
www.sciencedirect.com
Original article
Is there a ‘‘gold’’ standard treatment for patients with isolated myeloid sarcoma? Darko Antic a,*,b, Ivo Elezovic a,b, Natasa Milic b,c, Nada Suvajdzic a,b, Ana Vidovic a,b, Maja Perunicic a, Irena Djunic a, Mirjana Mitrovic a, Dragica Tomin a,b a
Clinic for hematology, Clinical Center Serbia, Koste Todorovica 2, 11000 Belgrade, Serbia Faculty of Medicine, University of Belgrade, Belgrade, Serbia c Institute for medical statistics and informatics, Belgrade, Serbia b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 1 September 2012 Accepted 22 October 2012
Isolated myeloid sarcoma is an extramedullary tumor of immature myeloid cells defined by the absence of leukemia history, myelodisplastic syndrome, or myeloproliferative neoplasma with a negative bone marrow biopsy. Myeloid sarcoma is a very rare condition, and few cases have been reported. We reviewed data of 12 patients with isolated myeloid sarcoma managed at a single center to determine the possible prognostic factors affecting patient survival, such as age, sex, type, localization, and treatment options. Patients were mostly men (n = 8), with a median age of 39 years. Patients were initially treated with chemotherapy (n = 7) or surgery (n = 5). In three patients, hematopoietic stem cell transplantation was performed. During the follow-up period, nine patients died. The median overall survival was 13 months, while event-free survival was 8 months. Regarding initial treatment strategy, no significant difference in overall survival was observed. Both chemotherapy and hematopoietic stem cell transplantation independently improved event-free survival. In addition, patients who received chemotherapy combined with hematopoietic stem cell transplantation had significantly longer eventfree survival than those treated with chemotherapy alone. Age < 40 years together with chemotherapy/ hematopoietic stem cell transplantation significant affected event-free survival. Based on our results, the treatment of myeloid sarcoma requires a systemic rather than a localized approach with surgery or radiotherapy. While prospective evaluations are needed, chemotherapy with allogenic hematopoietic stem cell transplantation should be considered as the optimal therapy for isolated myeloid sarcoma. ß 2012 Elsevier Masson SAS. All rights reserved.
Keywords: Myeloid sarcoma Chemotherapy Stem cell transplantation
1. Introduction Myeloid sarcoma (MS), also known as granulocytic/monoblastic sarcoma, extramedullary myeloid tumors, or chloroma, is a tumor mass comprising blasts or immature cells of the myeloid series, occurring at an anatomical site other than the bone marrow [1]. The condition was first described in 1811 by Burns [2] and later further described by King as tumors displaying a predominantly green color, due to the presence of myeloperoxidase (MPO) [3]. In 1966, Rappaport proposed the term ‘‘granulocytic sarcoma’’ [4], and finally, in 2002, the term myeloid sarcoma was accepted by World Health Organization (WHO) [1]. MS can occur in patients with active AML and in patients with chronic myeloproliferative disease, where it may occur as the first manifestation of blast transformation, the first manifestation of AML relapse in previously treated patients, or isolated MS in patients without bone marrow infiltration [1]. In almost 90% of untreated patients who initially had no hematological disorder, if untreated, AML develops within 10.5 to 11 months. The incidence of MS is
* Corresponding author. Tel.: +38111 30 65 112; fax: +31811 30 65 112. E-mail address: [email protected] (D. Antic). 0753-3322/$ – see front matter ß 2012 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.biopha.2012.10.014
unknown and believed to be underestimated. Isolated MS occurs in 2/ 1,000,000 adults, and in 0.7/1,000,000 children. Thus far, the involvement of the skin, periosteum, bowels, lymph nodes, genital system, central nervous system, heart, and lungs has been reported [5–12]. Several markers have been employed for identification of the myeloid origin of tumor infiltration in MS. Immunostaining with antibodies to myeloperoxidase (MPO), lysosomes, and chloroacetate esterase are important for identification of MS, bearing in mind that the myeloblasts in MS have an antigen profile similar to that present in the blasts and precursor cells of acute myeloid leukemia (AML) [10,11]. Herein, we analyze data from a single center in 12 patients with isolated MS to identify putative prognostic factors with a focus on accurate and prompt diagnosis and timely appropriate treatment to reduce the risk of progression to AML. 2. Materials and methods 2.1. Patient selection This study included 12 patients with isolated MS, diagnosed at the Clinic for Hematology of Clinical Center Serbia during the
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77
period of 2002 to 2012. This group included patients who had pathohistologically proved MS based on WHO criteria [1], and who, when diagnosed with MS, did not have any signs of AML, chronic myeloproliferative disease, or myelodysplastic syndromes (MDS) in the bone marrow. All data were analyzed in order to present clinico-pathological characteristics, therapy, and outcomes. Patient evaluation included analysis of history, physical examination, complete blood count, bio-humoral status, bone marrow analysis, and available radiological examinations (radiography, ultrasonography, and computed tomography). Chemotherapy (CT) had been used in accordance with standard protocols applied for the treatment of AML at the Institute for Hematology. 2.2. Pathohistological analysis Morphologic examination of biopsy specimens and immunohistochemical analysis were used to determine the myeloid origin of tumors and to rule out lymphoproliferative diseases and mesenchymal malignancies. The following monoclonal antibodies were used: CD13, CD33, CD34, MPO, Ly, CD68, CD43, HLA-DR, and CD117 (myeloid series antigens); antibodies to T and B cell markers (CD3, CD5, CD20, and CD79a); and CD15 and CD30 that were used to rule out non-Hodgkins lymphoma (NHL) and Hodgkins lymphoma (HL). Desmin, S-100, and cytokeratin abundance was determined in several patients to rule out sarcoma or malignant melanoma. 2.3. Statistical data analysis In this study, we used methods of descriptive and analytical statistics. Descriptive statistics for continuous variables are presented as median values with ranges or 95% confidence intervals for medians. Categorical data are presented by absolute numbers with percentages. Overall survival (OS), relapse-free survival (RFS), leukemia-free survival (LFS), and event-free survival (EFS), as well as cumulative incidences (CI) of these events were estimated using the Kaplan–Meier method. Definitions used to measure time to survival categories are presented in Table 1. Differences between the Kaplan–Meier survival curves of applied therapeutic approaches were evaluated by the log rank test. A P value of < 0.05 was considered statistically significant.
73
3. Results The 12 patients included in this study comprised mostly men (n = 8), with a median age of 39 years (range: 19–67 years). The MS-affected sites included the spine, tonsils, testis, heart, lymph nodes, mediastinum (two patients), skull base, uterus (two patients), spleen, and multiple gastrointestinal localizations (the stomach, duodenum, and bowel). MPO and HLA-DR were positive in ten and nine patients, respectively, and presented the most commonly expressed markers. In two patients, MPO staining was not performed for technical reasons. One of these two patients had clearly positive monoblastic series markers, while myeloid origin in the other patient was confirmed based on positive results for HLA-DR, CD117, and CD34. Five patients were CD68 positive and were therefore classified as having MS. Analysis of Ki67 expression was performed in four patients. Mediastinal MS exhibited the highest proliferative activity of 80%. Initially, three patients had been previously misdiagnosed, as Ewing sarcoma (spinal MS), eosinophilic granuloma (skull base MS), and mucosa-associated lymphoid tissue (MALT) lymphoma (the stomach, duodenum, and bowel). At the time of MS diagnosis, examination of the bone marrow showed normal features in all patients. The initial therapeutic approach was CT in seven patients and surgical resection in five patients. Within the group of patients treated with CT, two patients initially underwent partial surgical tumor resection based on vital indications – progressive neurological failure in a patient with spinal MS, and superior vena cava syndrome in a patient with cardiac MS. After one and two CT cycles, respectively, local radiotherapy was applied in both patients. Moreover, in two patients, surgical tumor resection (tonsillectomy and splenectomy, respectively) was performed for diagnostic purposes. All patients treated with CT received the standard induction therapy of 3 days with an anthracycline and 7 days with cytarabine (3 + 7). In three patients with complete remission (CR), we continued high-dose therapy followed by autologous (n = 1) and allogeneic (n = 2) hematopoietic stem cell transplantation (auto- and allo-HSCT, respectively) from related donors. Therapeutic approaches and outcomes are shown in Table 2. One of the patients treated with allo-HSCT developed relapse (isolated breast MS) and died. Two patients are currently in CR.
Table 1 Outcome measures in isolated myeloid sarcoma (MS). Category
Definition
Overall survival (OS)
Defined for all patients of a study; measured from the date of entry into a study to the date of death from any cause; patients not known to have died at last follow-up are censored on the date they were last known to be alive
Relaps free survival (RFS)
Defined only for patients achieving CR or CRI; measured from the date of achievement of a remission until the date of relapse or death from any cause; patients not known to have relapsed or died at last follow-up are censored on the date they were last examined
Cumulative incidence of relaps (CI of relaps)
Defined for all patients achieving CR or CRI; measured from the date of achievement of a remission until the date of relapse; patients not known to have relapsed are censored on the date they were last examined; patients who died without relapse are counted as a competing cause of failure
Leukemia-free survival (LFS)
Defined for all patients of a study; measured from the date of entry into a study to the date of occurrence of systemic leukemia or death from any cause; patients not known to have systemic leukemia or died at last follow-up are censored on the date they were last examined
Cumulative incidence of leukemia (CI of leukemia)
Defined for all patients of a study; measured from the date of entry into a study to the date of occurrence of systemic leukemia; patients not known to have systemic leukemia are censored on the date they were last examined; patients who died without systemic leukemia are counted as a competing cause of failure
Event-free survival (EFS)
Defined for all patients of a study; measured from the date of entry into a study to the date of extramedullary relapse, occurrence of systemic leukemia or death from any cause; patients not known to have any of these events are censored on the date they were last examined
CR: complete remission; CRI: complete remission with incomplete recovery.
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77
74
Table 2 Disease localization, treatment strategy, presence of extramedullary relapse or AML development and survival time. Patient/gender/age
MS site
Therapy
EM relapse/months/localisation
AML/months
Follow-up/months
1/m/24 2/m/39 3/m/58 4/m/38 5/f/41 6/f/67 7/m/26 8/f/19 9/m/31 10/f/28 11/m/40 12/m/57
Spine Tonsil Testis Heart Uterus Uterus Limph nodes Mediastinum Skull base GIT Spleen Mediastinum
CHT + autoPBSCT CHT + alloPBSCT Surgery CHT Surgery Surgery CHT CHT + alloPBSCT Surgery Surgery CHT CHT
No No Yes/6/skin No No No No Yes/18/breast Yes/4/skull base No No No
No No No No Yes/2 Yes/13 Yes/8 No No Yes/3 No No
Live/62 Live/50 Death/35 Death/6 Live/79 Death/14 Death/13 Death/28 Death/10 Death/3 Death/4 Death/8
M: male; f: female; CHT: chemotherapy; HSCT: hematopoietic stem cell transplantation.
Two patients treated with CT died during the course of treatment from sepsis, as a result of febrile neutropenia in one case, and multiple organ failure caused by disseminated disease in the other. Eight months after diagnosis, bone marrow aplastic anemia followed by evolution to AML occurred in a patient with lymph node MS. Surgical resection as the initial therapeutic method was performed in five patients, and all of them developed AML or extramedullary relapse. Three patients with MS developed AML within the postsurgical period of 13 months. One of them has exhibited CR and is alive at the time of study, 79 months after the application of systemic therapy. In the two other patients who underwent surgical resection, extramedullary relapse occurred after 4 and 6 months, and both patients did not survive. The incidence of AML and extramedullary relapse was significantly higher in patients initially treated by surgical resection only (P = 0.028). During the follow-up period, nine patients died. The survival outcomes and causes of death are presented in Table 3. The Kaplan–Meier estimate of median OS was 13 months (95% CI: 6–19 months) (Fig. 1), and the shortest OS was observed in a patient with multiple gastrointestinal localization (3 months). The patients undergoing CT had an OS of 13 months (95% CI: 1–25 months), whereas patients treated with surgery alone had an OS of 14 months (95% CI: 5–22 months). No significant difference in OS was observed between these two groups of patients (P > 0.05) (Fig. 2). Median survival time without the occurrence of systemic leukemia or extramedullary relapse was 8 months (95% CI: 3–12
months) and 14 months (95% CI: 1–30 months), respectively. The Kaplan–Meier estimate of median EFS was 6 months (95% CI: 3–9 months) (Fig. 3). Patients who underwent CT had a significantly longer EFS time compared to those who did not undergo CT, 8 vs. 4 months, respectively (log rank = 3.988, P = 0.046) (Fig. 4). A statistically significant difference in EFS was observed between patients undergoing HSCT and patients without HSCT (Log Rank = 7.608, P = 0.006). A statistically significant difference was found in EFS between patients treated with CT only (without HSCT) (median 6 months; 95% CI 3–8 months) and patients undergoing CT with HSCT (median not reached) (Log rank = 5.300; P = 0.021) (Fig. 5). Other variables that were analyzed, such as sex, adjuvant radiotherapy, underlying disorders, and MS, did not affect OS or EFS, whereas age < 40 had a significant influence on EFS, in combination with CT and HSCT as applied therapeutic options.
4. Discussion This study monitored the treatment of 12 patients with MS over a 10-year period at a single center. Thus far, most published literature on MS comprises case reports; only a few published
Table 3 Survival outcomes of patients with isolated myeloid sarcoma. Survival outcomes 5-year OS % Median
25 12% 13 months
5-year LFS % Median CI of leukemia
17 10% 8 months 44 18%
5-year RFS % Median CI of relaps
37 17% 14 months 53 20%
5-year EFS % Median
17 11% 6 months
Deaths
9
OS; overall survival; LFS; leukemia-free survival; CI: cumulative incidence; RFS: relaps free survival; EFS: event-free survival.
Fig. 1. Kaplan–Meier curve of median overall survival.
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77
75
Fig. 2. Kaplan–Meier curve of overall survival in patient treated with chemotherapy and surgery.
Fig. 4. Kaplan–Meier curve of event-free survival in patient treated with chemotherapy and surgery.
studies are based on a larger series of patients. A large proportion of these studies discuss patients with leukemic MS, whereas, in our study, we investigated only patients with isolated MS.
Diagnosis of MS in patients poses a huge challenge, and it is therefore necessary to analyze the expression of more than few markers that confirm the myeloid origin of neoplastic cells. Earlier, in the absence of hematological disorders, within a single series of patients, as many as 75% were initially diagnosed with a condition other than MS, NHL in most cases [13]. In another large series, 46% of patients were not initially diagnosed as MS, whereas in our group, three patients (25%) were initially misdiagnosed [14]. A number of studies have been published on the immunoprofile of MS. MPO, CD68, CD43, and lysozyme are frequently expressed in MS, but none of these are specific markers. To establish a definitive diagnosis, we recommend that the minimum panel of immunohistochemical markers should include the previously mentioned markers in addition to markers such as CD33, CD34, and CD117. Considering that cytogenetic abnormalities are detected in approximately 55% of tested cases with MS, the identification of an AML-associated abnormality (i.e., trisomy 8, t[8;21][q22;q22] or inv[16][p13.1q22]/t[16;16][p13.1;q22]) may help to arrive at the correct diagnosis (4,11). Several studies have summarized the outcomes of patients with MS. In 1981, Neiman et al. [15] published a study of 50 patients with MS, 15 of which had isolated MS. The majority of patients who did not have AML developed it within a period of 1 to 49 months (10 months on average). This result is in agreement with the results of the study by Meis et al., who reported that the average time for development of AML is 10.5 months [13]. In this series, prognostic indicators suggestive of causes influencing the increased risk for development of AML were not identified. No statistical correlation was found between the type of therapy administered and survival. In our group, 4 patients developed AML, with a median time of 24 months (95% CI 1–50 months), similar to the results of previous studies.
Fig. 3. Kaplan–Meier curve of median event-free survival.
76
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77
Fig. 5. Kaplan–Meier curve of event-free survival in patient treated with chemotherapy with or without hematopoietic stem cell transplantation.
Yamauchi and Yasuda investigated the association between therapeutic regimens and disease-free survival (DFS) in a group of 74 patients with MS. DFS was significantly longer in patients treated with systemic chemotherapy (median 12 months) than in those treated with surgery (3 months) or local irradiation (6 months) [16]. Tsimberidou et al. reported data regarding the incidence, therapy, and outcome of MS in 1520 patients with AML, and in 402 patients with high-risk MDS. MS occurred in 20 patients, 80% of whom received AML-type chemotherapy. Thirteen patients (65%) achieved CR, median OS was 20 months, and median DFS was 12 months [17]. Pileri et al. [18] reported the outcome of 67 patients with MS, of which 25 (27%) presented with non-leukemic MS. Among them, 47 received chemotherapy (70.1%), six underwent allogeneic bone marrow transplant (BMT) (allo-BMT) (9.0%), and 4 underwent autologous BMT (auto-BMT) (6.0%). At a median follow-up period of 150 months, only seven patients, six of whom underwent alloBMT, were alive (10.5%) and in CR. Overall, patients treated with auto- or allo-BMT exhibited longer survival than those who received conventional therapy (OS at 48 months: 76 vs. 0%). Paydas et al. [6] analyzed 32 patients with MS, seven of whom were initially misdiagnosed with NHL. Systemic chemotherapy had been administered to all the patients. The median survival was 9.5 months; seven patients survived for > 1 year, while 1 patient lived > 8 years. Breccia et al. [10] analyzed 12 patients with MS. In three patients with isolated MS, median survival to development of AML was 5 months, median survival was 7 months, and one patient was alive 49 months after BMT. A recent study published by Lan et al. evaluated sex, age, location, MS antedating leukemia, underlying disorders, treatment type, and stem cell transplantation as prognostic factors deter-
mining survival in 24 patients with MS (seven with isolated sarcoma, others with AML, chronic myeloid leukemia, or MDS). The 5-year survival rate for patients with MS was 21%. Patients receiving chemotherapy had a significantly longer OS period compared to those who did not (P = 0.0009). In our group, the 5-year OS rate for patients with isolated MS was 25%, with the median OS period of 13 months, whereas the 5-year EFS rate was 17% with median OS period of 6 months. Treatment strategy did not affect OS, but EFS was significantly longer in patients treated with chemotherapy and HSCT (P = 0.046 and P = 0.006, respectively) [19]. The optimal timing and treatment of isolated MS are not clear, but delayed or inadequately systemically treated isolated MS will almost always progress to AML [20]. In our practice, we use remission–induction chemotherapy similar to that used for AML. Postremission chemotherapy has not been adequately studied in isolated MS; therefore, one of the most important questions in the treatment of MS is the role of HSCT. Patients in the current study treated with HSCT had a longer EFS. Chevallier et al. [21] confirmed this approach by assessing the outcome of 30 patients with isolated MS who underwent allo-HSCT. Patients with isolated MS had a 5year OS and LFS of 45% and 33%, respectively, with relatively acceptable toxicity (overall 5-year non-relapse moratlity 17%). Considering the entire cohort, in univariate analysis, factors associated with higher LFS and lower incidence of relapse were CR status at transplant and cytogenetics. The only factor associated with high non-relapse mortality was patient sex (male: 27% 6% vs. female: 10% 5%, P = 0.04). In multivariate analysis, high-risk cytogenetics remained significantly associated with a poorer LFS and an increased incidence of relapse. However, the latter should be considered with caution as the cytogenetic data were missing in most cases of isolated MS. According to the results in our series of patients, as well as the published data, it is clear that in cases of suspected MS, employment of a wide spectrum of antibodies during immunohistochemical work-up of tumor tissue is mandatory. Timely diagnosis has a huge influence on treatment results, i.e., longer survival as a consequence of adequate therapy. While prospective evaluations are needed, chemotherapy with allo-HSCT could be considered as the optimal therapy for isolated MS. Moreover, the role of new molecular prognostic markers such as NPM1 or FLT3ITD mutations in MS should be investigated. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. References [1] Pileri SA, Orayi A, Falini B. Myeloid sarcoma. In: Swerdlow S, Campo E, Harris NL., Jaffe ES., Pileri SA., Stein H, Thiele J, Vardiman JW., editors. WHO classification of tumours of haematopoetic and lymphoid tissues. Lyon: IARC; 2008. p. 140–1. [2] Burns A. Observation of surgical anatomy, head and neck. Edinburgh: Thomas Royce and co; 1811. p. 364–6. [3] King A. A case of chloroma. Monthly J med 1853;17:97. [4] Rappaport H. Tumors of the hematopoetic system. Armed Forces Inst Pathol 1966;241–3. [5] Hutchinson RE, Kurec AS, Davey FR. Granulocytic sarcoma. Clin Lab Med 1990;10:889–901. [6] Paydas S, Zorludemir S, Ergin M. Granulocytic sarcoma: 32 cases and review of the literature. Leuk Lymphoma 2006;47(12):2527–41. [7] Antic D, Verstovsek S, Elezovic I, Grujicic D, Gotic M, Bila J, et al. Spinal epidural granulocytic sarcoma in non-leukemic patient. Int J Hematol 2009;89(1):95–7. [8] Antic D, Vuckovic M, Elezovic I. Right atrial myeloid sarcoma causing superior vena cava syndrome. Br J Haematol 2008;141(2):134. [9] Audouin J, Comperat E, Le Tourneau A, Camilleri-Broe¨t S, Adida C, Molina T, et al. Myeloid sarcoma: clinical and morphologic criteria useful for diagnosis. Int J Surg Pathol 2003;11(4):271–82. [10] Alexiev BA, Wang W, Ning Y, Chumsri S, Gojo I, Rodgers WH, et al. Myeloid sarcomas: a histologic, immunohistochemical, and cytogenetic study. Diagn Pathol 2007;2:42.
D. Antic et al. / Biomedicine & Pharmacotherapy 67 (2013) 72–77 [11] Klco JM, Welch JS, Nguyen TT, Hurley MY, Kreisel FH, Hassan A, et al. State of the art in myeloid sarcoma. Int J Lab Hematol 2011;33(6):555–65. [12] Breccia M, Mandelli F, Petti MC, D’Andrea M, Pescarmona E, Pileri SA, et al. Clinico-pathological characteristics of myeloid sarcoma at diagnosis and during follow-up: report of 12 cases from a single institution. Leuk Res 2004;28(11):1165–9. [13] Meis JM, Butler JJ, Osborne BM, Manning JT. Granulocytic sarcoma in nonleukemic patients. Cancer 1986;58(12):2697–709. [14] Byrd JC, Edenfield WJ, Shields DJ, Dawson NA. Extramedullary myeloid cell tumors in acute nonlymphocytic leukemia: a clinical review. J Clin Oncol 1995;13(7):1800–16. [15] Neiman RS, Barcos M, Berard C, Bonner H, Mann R, Rydell RE, et al. Granulocytic sarcoma: a clinicopathologic study of 61 biopsed cases. Cancer 1981;48:1426–37. [16] Yamauchi K, Yasuda M. Comparison in treatments of nonleukemic granulocytic sarcoma: report of two cases and a review of 72 cases in the literature. Cancer 2002;94(6):1739–46.
77
[17] Tsimberidou AM, Kantarjian HM, Wen S, Keating MJ, O’Brien S, Brandt M, et al. Myeloid sarcoma is associated with superior event-free survival and overall survival compared with acute myeloid leukemia. Cancer 2008;113(6):1370–8. [18] Pileri SA, Ascani S, Cox MC, Campidelli C, Bacci F, Piccioli M, et al. Myeloid sarcoma: clinico pathologic, phenotypic and cytogenetic analysis of 92 adult patients. Leukemia 2007;21(2):340–50. [19] Lan TY, Lin DT, Tien HF, Yang RS, Chen CY, Wu K. Prognostic factors of treatment outcomes in patients with granulocytic sarcoma. Acta Haematol 2009;122(4):238–46. [20] Bakst RL, Tallman MS, Douer D, Yahalom J. How I treat extramedullary acute myeloid leukemia. Blood 2011;118(14):3785–93. [21] Chevallier P, Labopin M, Cornelissen J, Socie´ G, Rocha V, Mohty M. ALWP of EBMT Allogeneic hematopoietic stem cell transplantation for isolated and leukemic myeloid sarcoma in adults: a report from the Acute Leukemia Working Party of the European group for blood and marrow transplantation. Haematologica 2011;96(9):1391–4.