OPEN FORUM Pros and Cons of Splenectomy in Patients

Pros and cons of splenectomy in patients with myeloﬁbrosis undergoing stem cell transplantation Z Li and HJ Deeg

Fred Hutchinson Cancer Research Center and the University of Washington, Seattle, WA, USA

During fetal development, the spleen is a major hemopoietic the possible immunological consequences of splenectomy, organ. In the adult human, this task is relinquished to the bone and discuss factors that may influence the decision on marrow. However, under the stress of certain pathologic con- ditions, extramedullary hemopoiesis may again occur in the splenectomy in the management of myelofibrosis. spleen. This is especially true for diseases of the marrow, in particular, myeloproliferative disorders such as agnogenic myeloid metaplasia, which is associated with severe fibrosis of Clinical controversies the marrow space. At the same time, the spleen sequesters blood cells and contributes to peripheral blood cytopenias, Surgical splenectomy and splenic irradiation have been part which may improve following splenectomy. However, success is unpredictable, and the operative mortality of splenectomy is of the management of AMM for two reasons: firstly, to relieve on the order of 10%. As a growing number of patients undergo symptoms associated with splenomegaly, and secondly, to hemopoietic stem cell transplantation as definitive therapy for possibly slow disease progression, although this notion is not myelofibrosis, the decision on splenectomy has additional well supported by data. In fact, some reports suggest that in ramifications since the spleen plays an important role in the patients with AMM, splenectomy is associated with an kinetics of engraftment of donor cells and in immune reconsti- increased risk of blast transformation.4 Tefferi and colleagues2 tution. We conclude from our analysis of available information that the benefit of splenectomy is difficult to predict, although recently reported the Mayo Clinic experience with splen- after transplantation splenectomized patients have faster hem- ectomy in 223 patients with AMM. The indications for surgery opoietic recovery. It appears that the most important indication were anemia, symptoms related to splenomegaly, portal for splenectomy in these patients is the relief of symptoms hypertension or severe thrombocytopenia. Pre-splenectomy from massive spleen enlargement. Leukemia (2001) 15, 465– thrombocytopenia and the absence of a cellular marrow were 467. risk factors for decreased survival. No durable remissions of Keywords: myelofibrosis; splenectomy; hemopoietic stem cell transplantation thrombocytopenia were achieved. The rate of blast transformation was 16%; the rate was higher in patients with thrombocytopenia and increased spleen mass, but blast transformation did not affect survival. The authors surmised that Introduction thrombocytopenia was a surrogate for advanced disease. Overall, the data suggested that splenectomy was beneficial Agnogenic myeloid metaplasia (AMM) and other disorders, for the treatment of severe constitutional symptoms and portal such as the spent phase of polycythemia vera (P vera) or hypertension, and the improved quality of life may justify the essential thrombocythemia (ET) are typically associated with procedure despite an operative mortality of approximately myelofibrosis (MF) and splenomegaly as a result of extramed- 10%. With regards to the management of anemia, patients ullary hematopoiesis.1 Severe splenomegaly may be compli- with a hypercellular marrow may require fewer transfusions cated by splenic infarcts, blood cell sequestration, and hemo- post-splenectomy and may enjoy prolonged survival. lysis and may cause considerable discomfort to the patient. Importantly, however, other than being associated with The management of this situation has remained contro- increased blast transformation in thrombocytopenic patients, versial.2,3 As a growing number of patients with myeloprolifer- splenectomy neither retarded nor hastened the progression of ative disorder and splenomegaly are undergoing marrow or AMM. In an earlier study from the same institutions, investi- peripheral blood stem cell transplantation, this question has gators had shown that among 23 patients treated with been further accentuated. It is uncertain to what extent the 277.5 cGy of irradiation to the spleen, significant cytopenias spleen is involved in post-transplant immune reconstitution occurred in 10 (43.5%), including six patients (26%) with and whether splenectomy is associated with improved clinical prolonged life-threatening pancytopenia.5 outcome. Prospective randomized trials are not available, and While these data overall are in agreement with earlier in any event, would be difficult to perform due to the hetero- reports,6,7 others advise cautious judgment because of uncer- geneity of the underlying diseases, and the operative mor- tain benefits and considerable morbidity and mortality asso- bidity and mortality associated with splenectomy. Firm rec- ciated with splenectomy.3 Conclusions from these and other ommendations to guide clinicians in regards to a decision on reports are summarized in Table 1. splenectomy are lacking. Here, we review this controversial As more patients with myelofibrosis are being treated by issue in the current literature, highlight our own experience hemopoietic stem cell transplantation, the question arises as at the Fred Hutchinson Cancer Research Center, consider to whether pre-transplant splenectomy is beneficial or detri- mental in regard to post-transplant outcome. Only limited data have been reported (Table 1). Guardiola et al8 recently Correspondence: HJ Deeg, Clinical Research Division, Fred Hutchin- son Cancer Research Center, 1100 Fairview Avenue North, D1-100, presented results in 55 patients with AMM who were trans- Seattle, WA 98109, USA; Fax: 206 667 6124 planted from HLA-matched related (n = 49) or alternative (n Received 12 October 2000; accepted 31 October 2000 = 6) donors. The 5-year probability of survival was 47% for Splenectomy and transplantation for myelofibrosis Z Li and HJ Deeg 466 Table 1 Controversies surrounding splenectomy in the manage- Theoretical considerations ment of myelofibrosis Splenomegaly in AMM or other hemopoietic disorders may Hemopoietic stem cell transplantation be a reflection of an expansion of the underlying malignant 8 No improvement in survival Guardiola et al, 1999 clone. Splenectomy should, therefore, not be expected to Faster hemopoietic recovery Li et al, 20009 Martino et al, 199410 change the natural history of a hemopoietic stem cell disorder. Von Bueltzingsloewen et al, However, such a view is difficult to test experimentally. 199411 Therefore, it may be useful to consider various possible Ringden and Nilsson, 198514 consequences of splenectomy. Michallet et al, 199115 The spleen is the largest lymphoid organ in the body, 18 Tollemar et al, 1989 accounting for about 25% of the body’s lymphocytes.12 It is Bostro¨m et al, 199016 Increased risk of GVHD Ringden and Nilsson, 198514 a reservoir of both effector and memory T and B lymphocytes. Michallet et al, 199115 It is one of the sites in which B and T cells initially encounter Bostro¨m et al, 199016 blood-borne pathogens. The spleen plays an important role in Higher risk of deep fungal Tollemar et al, 198918 the clearance of microorganisms, the synthesis of antibodies, infection and the provision of opsonins required for antibody-depen- 18 Increased risk of late Tollemar et al, 1989 dent cytotoxicity. Furthermore, the spleen, specifically the bacterial infection splenic outer periarteriolar lymphoid sheath (PALS), is a criti- Conventional management cal site where B cells undergo antigen-driven selection, acti- No benefit in patients Tefferi et al, 20002 vation and deletion, and where long-lived memory B cells are receiving conventional selected.13 In response to T cell dependent antigens, the proli- therapy ferating B cells undergo plasma cell differentiation either 6 Decreased transfusion Brenner et al, 1988 within the outer PALS, or in the follicles after migration. In requirement Tefferi et al, 20002 Jarvinen et al, 198219 the absence of T cell help, B cells involved in T cell dependent Increased risk of blast Barosi et al, 19984 responses die, but B cells involved in T cell independent anti- transformation Tefferi et al, 20002 gen responses differentiate into plasma cells in the outer PALS. The outer PALS is also the site of inactivation and elimination of autoreactive B cells, although the mechanism remains incompletely defined. Taken together, the spleen plays a piv- otal role in the initiation of B cell responses, the generation of long-term memory B cells, and the elimination of autoreactive B cell clones. It is unclear what role the spleen has the overall group, and 54% for patients receiving an unmani- in the reconstitution of the immune system post stem cell pulated HLA-matched related transplant. Splenectomy prior to transplantation. It is safe to state, however, that the increased transplant was associated with significantly faster recovery of incidence of fatal bacteremia associated with asplenism is not neutrophil and platelet counts. There was no correlation with simply the consequence of diminished opsonization, but is acute or chronic GVHD, and there was no difference between due to significant dysregulation of the immune response. In patients with (n = 27) and without (n = 28) splenectomy prior view of these considerations, it is not surprising that a role of to transplant with regard to 1-year transplant-related mortality the spleen in GVHD and immune reconstitution post- (30% vs 25%) or 5-year overall survival (48% vs 46%). transplant has been considered.14–18 We recently analyzed the impact of pre-transplant splenectomy on post-transplant outcome in 26 patients transplanted at the Fred Hutchinson Cancer Research Center for AMM (n Practical approach = 17) or fibrosis associated with the spent phase of P vera or Given the lack of convincing data that would favor splen- = 9 ET (n 9). Patients were 18.2 to 55.3 (median 37) years old ectomy in the management of AMM and the important role at the time of transplant. They had been followed for 1.5 to of the spleen in humoral immunity, splenectomy cannot be 10 (median 2.6) years. Eleven of these patients (44%) had recommended as a routine procedure. Splenomegaly by itself undergone splenectomy prior to transplant, and 15 (56%) had is not necessarily a sign of disease progression and cannot an intact spleen. Patients with splenectomy were older serve as a prognostic marker. The clinical decision about (median 44 vs 36 years) and had longer disease duration splenectomy should be made on an individual basis. If trans- before transplant, albeit not significantly so, than patients plantation is not an option for such a patient, the decision on without splenectomy (median 37.5 vs 14 months). Overall splenectomy should probably be made primarily on the basis post-transplant mortality was comparable in splenectomized of patient symptomatology as recently discussed by Tefferi et (25%) and non-splenectomized patients (33%). As suggested al2 and in our own study.9 Symptoms from splenomegaly may by previous reports,10,11 post-transplant granulocyte recovery also justify splenectomy in patients who proceed to hemopo- was faster among splenectomized patients (14–51 (median 18) ietic stem cell transplantation. Splenectomized patients tend days to reach 0.5 × 109/l, compared to 19–85 (median 23) to have faster hemopoietic recovery and require fewer trans- days among patients with intact spleen) (P = 0.04), and the fusions. However, the risk of overwhelming infections and need for both red blood cell and platelet transfusions was impaired immune recovery has to be considered. greater among patients who had their spleens intact (mean 9.8 vs 6.1 units for red blood cells, and 21.7 vs 16 units for platelets). The 3-year probability of disease-free survival was Conclusions 73% for splenectomized patients and 64% for patients without 9 Despite considerable clinical interest, it is still controversial splenectomy (not significant). whether and when splenectomy should be performed in

Leukemia Splenectomy and transplantation for myelofibrosis Z Li and HJ Deeg 467 patients with myelofibrosis. This controversy is mainly due to 8 Guardiola P, Anderson JE, Bandini G, Cervantes F, Runde V, a complete lack of prospective randomized studies and our Arcese W, Bacigalupo A, Przepiorka D, O’Donnell MR, Polchi P, incomplete knowledge of the role of the spleen in the patho- Buzyn A, Sutton L, Cazals-Hatem D, Sale G, De Witte T, Deeg HJ, Gluckman E, for the International Collaboration for Transplan- genesis of myelofibrosis and myeloid metaplasia. Retrospec- tation in Agnogenic Myeloid Metaplasia. Allogeneic stem cell tive studies have shown little, if any, advantage of splen- transplantation for agnogenic myeloid metaplasia: a European ectomy for clinical outcome. This fact, coupled with the Group for Blood and Marrow Transplantation, Socie´te´ Française importance of the spleen in the body’s immune response, cau- de Greffe de Moelle, Gruppo Italiano per il Trapianto del Midollo tion against routine splenectomy in patients with AMM or Osseo, and Fred Hutchinson Cancer Research Center collabor- myelofibrosis of different etiologies. ative study. Blood 1999; 93: 2831–2838. 9 Li Z, Gooley T, Appelbaum FR, Deeg HJ. Splenectomy and hemopoietic stem cell transplantation for myelofibrosis (letter). Blood (in press). Acknowledgements 10 Martino R, Altes A, Muniz-Diaz E, Brunet S, Sureda A, Domingo- Albos A, Madoz P. Reduced transfusion requirements in a splenec- This study was supported in part by PHS grants HL36444, tomized patient undergoing bone marrow transplantation. Acta Haematol 1994; 92: 167–168. CA18029, and CA87948. HJD is also supported by a grant 11 von Bueltzingsloewen A, Bordigoni P, Dorvaux Y, Witz F, Schmitt from the Gabrielle Rich Leukemia Fund. We thank H Craw- C, Chastagner P, Sommelet D. Splenectomy may reverse pancyto- ford and B Larson for help with manuscript preparation. The penia occurring after allogeneic bone marrow transplantation current address of ZL is Center for Immunotherapy, MC 1601, (letter). Bone Marrow Transplant 1994; 14: 339–340. University of Connecticut Health Center, Farmington, CT 12 Brown AR. Immunological functions of splenic B-lymphocytes 06030-1601. (review). Crit Rev Immunol 1992; 11: 395–417. 13 Liu YJ. Sites of B lymphocyte selection, activation, and tolerance in spleen (review). J Exp Med 1997; 186: 625–629. 14 Ringden O, Nilsson B. Death by graft-versus-host disease associa- References ted with HLA mismatch, high recipient age, low marrow cell dose, and splenectomy. Transplantation 1985; 40: 39–44. 1 Tefferi A. Myelofibrosis with myeloid metaplasia (review). N Engl 15 Michallet M, Corront B, Bosson JL, Molina L, Peissel B, Maranin- J Med 2000; 342: 1255–1265. chi D, Reiffers J, Chabannon C, Gaspard MH, Stoppa AM, Blaise 2 Tefferi A, Messa RA, Nagorney DM, Schroeder G, Silverstein MN. D, Marit G, Hollard D, Carcassone Y, Broustet A, Demongeot J. Splenectomy in myelofibrosis with myeloid metaplasia: a single- Role of splenectomy in incidence and severity of acute graft-ver- institution experience with 223 patients. Blood 2000; 95: 2226– sus-host disease: a multicenter study of 157 patients. Bone Marrow 2233. Transplant 1991; 8: 13–17. 3 Benbassat J. Myelofibrosis with myeloid metaplasia. N Engl J Med 16 Bostro¨m L, Ringdeń O, Jacobsen N, Zwaan F, Nilsson B. A Euro- 2000; 343: 659. pean multicenter study of chronic graft-versus-host disease. The 4 Barosi G, Ambrosetti A, Centra A, Falcone A, Finelli C, Foa P, role of cytomegalovirus serology in recipients and donors, acute Grossi A, Guarnone R, Rupoli S, Luciano L, Petti MC, Pogliani E, graft-versus-host disease, and splenectomy. Transplantation 1990; Russo D, Ruggeri M, Quaglini S. Splenectomy and risk of blast 49: 1100–1105. transformation in myelofibrosis with myeloid metaplasia. Blood 17 Kalhs P, Schwarzinger I, Anderson G, Mori M, Clift RA, Storb R, 1998; 91: 3630–3636. Buckner CD, Appelbaum FR, Hansen JA, Sullivan KM. A retro- 5 Elliott MA, Chen MG, Silverstein MN, Tefferi A. Splenic irradiation spective analysis of the long-term effect of splenectomy on late for symptomatic splenomegaly associated with myelofibrosis with infections, graft-versus-host disease, relapse, and survival after myeloid metaplasia. Br J Haematol 1998; 103: 505–511. allogeneic marrow transplantation for chronic myelogenous leu- 6 Brenner B, Nagler A, Tatarsky I, Hashmonai M. Splenectomy in kemia. Blood 1995; 86: 2028–2032. agnogenic myeloid metaplasia and postpolycythemic myeloid 18 Tollemar J, Ringden O, Bostrom L, Nilsson B, Sundberg B. Vari- metaplasia. A study of 34 cases. Arch Intern Med 1988; 148: ables predicting deep fungal infections in bone marrow transplant 2501–2505. recipients. Bone Marrow Transplant 1989; 4: 635–641. 7 Coon WW, Liepman MK. Splenectomy for agnogenic myeloid 19 Jarvinen H, Kivilaakso E, Ikkala E, Vuopio P, Hastbacka J. Splen- metaplasia. Surg Gynecol Obstet 1982; 154: 561–563. ectomy for myelofibrosis. Ann Clin Res 1982; 14: 66–71.

Leukemia