2583 Thrombotic Microangiopathy Associated with Interferon Therapy for Patients with Chronic Myelogenous Leukemia Coincidence or True Side Effect? Farhad Ravandi-Kashani, M.B., Jorge Cortes, M.D.1 Moshe Talpaz, M.D.2 Hagop M. Kantarjian, M.D.1 1 B.S. 1 Department of Leukemia, M. D. Anderson Cancer Center, Houston, Texas. 2 Department of Bioimmunotherapy, M. D. Anderson Cancer Center, Houston, Texas. BACKGROUND. Interferon-a (rIFN-a) is an established therapy for patients with myeloproliferative disorders. Unusual immune-mediated side effects have been associated with rIFN-a therapy. The association of rIFN-a therapy with hemolytic uremic syndrome (HUS) and thrombotic thrombocytopenic purpura (TTP) has been reported infrequently. METHODS. Two patients with chronic myelogenous leukemia (CML) treated with rIFN-a– based regimens at the University of Texas M. D. Anderson Cancer Center developed thrombotic microangiopathy (HUS/TTP). The course of their disease is described. A third patient who developed renal failure while receiving rIFN-a therapy and had no other causative factor for his renal failure is also described. RESULTS. The patients were ages 24, 49, and 36 years, and they had received rIFN-a therapy for 37, 67, and 92 months, respectively, prior to the development of the disorder. One patient had discontinued rIFN-a 1 month before the event because of presumed rIFN-a–related cardiomyopathy. Two patients received hydroxyurea and cytarabine as part of their therapy. No patient was receiving any medication known to be associated with HUS/TTP. None had a history of diarrheal illness, but Escherichia coli OH157.H7 was grown from the stool of one patient. Two patients responded to plasmapheresis with normalization of counts and other indices, but both developed renal failure and became dependent on dialysis. One patient had evidence of disease progression and died of multiorgan failure. The third patient required dialysis for 18 months but is currently off dialysis; this patient has some residual renal impairment. CONCLUSIONS. Although no definitive association between rIFN-a therapy and thrombotic microangiopathies can be concluded from these data, these and other previously reported cases suggest that HUS/TTP is a rare side effect of rIFN-a therapy that should be managed in the standard fashion. Hypotheses regarding the mechanism underlying this association are discussed in this article. Cancer 1999; 85:2583– 8. © 1999 American Cancer Society. KEYWORDS: chronic myelogenous leukemia, interferon-a, thrombotic microangiopathy, renal failure. Published in abstract form in Proceedings of ASH, San Diego, California, 1998, and in Blood 1988, Volume 92, Number 10, Supplement 1. Address for reprints: Hagop M. Kantarjian, M.D., Department of Leukemia, Box 61, M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030. © 1999 American Cancer Society F ollowing the initial studies in the 1980s that demonstrated the efficacy of interferon-alpha (rIFN-a) therapy in inducing hematologic and cytogenetic responses in patients with chronic myelogenous leukemia (CML),1– 4 recent trials have established this agent as the current standard in the treatment of patients with this disorder who are not eligible for allogeneic bone marrow transplantation.5 The efficacy of rIFN-a in achieving hematologic and cytogenetic re- 2584 CANCER June 15, 1999 / Volume 85 / Number 12 sponses is improved when combined with other agents, such as cytosine arabinoside (ara-C) or homoharringtonine.6 – 8 Interferon therapy has been associated with several well-documented toxicities.5,9 These include a flulike illness with fever, malaise, and myalgia, the severity of which may be associated with higher initial white blood cell count (WBC) and therefore can be partly alleviated by the initial use of hydroxyurea to reduce the WBC count.9 Other reported side effects include neuropsychiatric symptoms with fatigue, depression, and insomnia.5 Uncommon and immunemediated complications of interferon therapy have been described.10 These include cutaneous vasculitis, immune hemolytic anemia, hypothyroidism, immune-mediated thrombocytopenia, nephrotoxicity, pemphigus foliaceous, rheumatoid arthritis, systemic lupus erythematosus, and other connective tissue disorders, including polymyositis and Raynaud syndrome.11–20 Rare cases of retinopathy and cardiac toxicity have been reported and may be immunemediated.21,22 Occasional reports have described hemolytic uremic syndrome (HUS) developing in patients receiving rIFN-a therapy for CML.23–26 Shammas et al. also reported a case of thrombotic microangiopathy in a patient with CML on hydroxyurea therapy.27 HUS has also been reported in a patient with hairy cell leukemia who received rIFN-a therapy.28 Therefore, it remains unclear whether this association is a true side effect of rIFN-a therapy or a mere coincidence. Thrombotic microangiopathy is a term that encompasses HUS as well as thrombotic thrombocytopenic purpura (TTP). Its association with malignant disorders has been well established.29,30 The association of HUS/TTP with chemotherapeutic agents for cancer patients was first reported by Liu et al.31 They described mitomycin C–induced renal toxicity associated with microangiopathic hemolytic anemia and thrombocytopenia. Other agents reported to cause HUS/TTP include methyl-carmustine, bleomycin, daunomycin, gemcitabine, and deoxycoformycin.32–34 We describe two cases of thrombotic microangiopathic disorders and a third case of renal failure that was probably secondary to undiscovered HUS/TTP, in patients who received rIFN-a therapy for CML, and consider the possible underlying mechanisms. CASE REPORTS Case 1 A male age 49 years was diagnosed with Philadelphiachromosome (Ph) negative, BCR-ABL positive CML in August 1991, when he presented with fatigue, weight loss, and a WBC count of 226 3 109/L. He was initially treated with hydroxyurea 3 g/day and rIFN-a 10 MU/ day. Following a hematologic response, he was maintained on rIFN-a at the same dose and hydroxyurea at 0.5 g/day, which resulted in a fall in the BCR-ABL positive cells to 3% after 32 months of therapy. Five years after the initial diagnosis, he developed thrombocytosis with a platelet (PLT) count of 636 3 109/L. Ara-C 10 mg/day was added to the regimen and a search for a bone marrow transplantation (BMT) donor was initiated. He was found to be 96% Ph-positive by fluorescent in situ hybridization (FISH) analysis. Sixty-seven months after initiation of rIFN-a therapy, in April 1997, he developed severe, refractory anemia with a hemoglobin level of 6.0 g/dL and was commenced on steroids for a presumptive diagnosis of IFN-induced immune hemolysis. He had the following results: PLT 32 3 109/L, total bilirubin 2.1 with indirect bilirubin 2.0 mg/dL, lactate dehydrogenase (LDH) 6957 IU/L, normal liver enzymes, blood urea nitrogen (BUN) 35 mg/dL, and creatinine 1.6 mg/dL. The following tests were negative or normal: rapid plasma reagent (RPR), antinuclear antibody (ANA), hepatitis B surface antigen (HBsAg), hepatitis B core antibody (HBcAb), human immunodeficiency virus antibody (HIV), human T-lymphotrophic virus antibody (HTLV)-1/2, cytomegalovirus antibodies (CMV), prothrombin time (PT), partial thromboplastin time (PTT), serum fibrinogen, fibrin split products (FSP), and serum haptoglobin. A peripheral blood smear revealed moderate schistocytosis. Bone marrow examination was negative for increased number of blast cells. Stool cultures was positive for Escherichia coli OH157.H7, but all other cultures were negative. A diagnosis of HUS was made, and the patient was started on daily plasmapheresis with cryo-poor plasma. He was not taking any other medications associated with HUS development. On April 26, 1997, he developed vertigo and slurred speech, and a CT scan of brain revealed a small area of hypodensity in the pons and midbrain suggestive of infarction/hemorrhage. Lumbar puncture was performed and cerebrospinal fluid (CSF) was negative for an inflammatory process with 1 WBC/high power field (hpf), 27 RBC/hpf, normal protein and glucose, and negative bacterial, fungal, and mycobacterial cultures. Four days later, he developed respiratory distress that required artificial ventilation with bronchoscopic findings, and imaging studies were suggestive of bronchoalveolar hemorrhage. The patient continued to receive plasmapheresis and later developed nonoliguric renal failure, with rise of BUN and creatinine to 106 mg/dL and 3.7 mg/dL, respectively. He was started on daily hemodialysis. On April 26, 1997, after 6 episodes of plasmapheresis, his blood counts normalized, but in May 1997 he had evidence Thrombotic Microangiopathy with rIFN-a/Ravandi-Kashani et al. 2585 of CML acceleration, with rising WBC to 48 3 109/L, 7% peripheral blasts, and 18% bone marrow blasts. Despite supportive measures, the patient developed multiorgan failure and died of the complications. A request for an autopsy was declined by the patient’s relatives. 109/L, disappearance of schistocytes, and normalization of other indices, such as LDH (519 mg/dL) and bilirubin (0.8 mg/dL), in June 1998. However, she has continued to remain hemodialysis dependent with end-stage renal disease after over 3 months of follow-up. Case 2 Case 3 A white female age 24 years with Ph-positive CML diagnosed in November 1994 was initially treated with hydroxyurea at doses of 1.5– 4.0 g/day for 1 month. Her therapy was changed to a combination of rIFN-a 9 MU/day and ara-C 10 mg/day, and hydroxyurea was discontinued following achievement of complete hematologic response (CHR). After 2 years of therapy, she did not achieve a cytogenetic response; Ph chromosome remained at 100% on bone marrow cytogenetics, and a search for a BMT donor was initiated. In February 1998, 36 months after initiation of therapy with rIFN-a, she developed exertional dyspnea with radiographic evidence of cardiomegaly. Echocardiography demonstrated reduced left ventricular ejection fraction of 35%. Interferon-related cardiomyopathy was suspected and the patient was commenced on steroids, digoxin, and furosemide; she had a good response, and her cardiac status improved. rIFN-a therapy was discontinued and hydroxyurea was increased to maintain normal blood cell counts. One month after discontinuation of rIFN-a, in March 1998, she presented with severe hypertension (BP 230/150) and was noted to have the following laboratory results: PLT 20 3 109/L, hemoglobin 11.9 g/dL, WBC 10 3 109/L, BUN 48 mg/dL, creatinine 1.8 mg/dL, fibrinogen 266 mg/dL, FSP 80 mg/dL, D-Dimer 0.25– 0.5, PT 12s, PTT 25s, LDH 4300, total bilirubin 1.9 mg/dL, direct and indirect Coomb test negative, peripheral blood smear positive for schistocytes, and haptoglobin ,6 (N30 –226). The following tests were negative or normal: HBsAg, HBcAb, HIV-1 and -2, HTLV-1 and -2, smooth muscle antibody (SMA), ANA, c- and p-antineutrophil cytoplasmic antibody (ANCA), complement C4 and C3, and stool for E. coli OH157.H7. Urinalysis showed 50 –100 RBCS with no casts. Urinary collection every 24 hours revealed 4.5 g/day urinary protein loss. Renal ultrasound revealed normal echogenicity of the kidneys without hydronephrosis. The diagnosis of renal failure secondary to HUS with malignant hypertension was made, leading to the initiation of antihypertensive therapy, daily plasmapheresis, and later, in April 1998, with worsening renal function, (BUN 114 mg/dL, creatinine 3.9 mg/dL) hemodialysis. The patient was not taking any other medications at the time of presentation. She responded to therapy with normalization of PLT count to 376 3 A male age 36 years was diagnosed with Ph-positive CML in September 1985 and treated with hydroxyurea. In August 1986, he was started on therapy with rIFN-a combined with IFN-gamma (rIFN-g) at doses of 2 MU/m2/day and 0.01 MU/m2/day, respectively. He was maintained on rIFN-a 5MU/m2/day and IFN-g 0.025 MU/m2/day and achieved a CHR. In October 1989, rIFN-g was held due to excess toxicity with fatigue and neuropsychiatric symptoms. By October 1992 he had achieved a major cytogenetic response with 5% Ph-positive cells on bone marrow analysis. rIFN-g was reintroduced due to improvement in symptoms. In April 1994 he presented to his local hospital with increased fatigue and hypertension, where he was noted to have renal failure (BUN 83 mg/dL, creatinine 6.2 mg/dL), marked anemia with hematocrit of 20, and congestive heart failure with a left ventricular ejection fraction of 35% on echocardiography. Results of other investigations were as follows: WBC 10.9 3 109/L; PLT 282 3 109/L; red blood cell poikilocytosis and anisocytosis; LDH 259 U/L (N90 –200); haptoglobin 144 mg/dL; PT 14S; PTT 23s; urinalysis positive for protein and occasional hyaline and granular casts; 24-hour urinary protein 1.8 g; normal renal ultrasound results; renal biopsy-hyperplastic arteriolosclerosis with negative immunofluorescent studies, consistent with malignant hypertension; myocardial biopsy–focal individual hypertrophy with no appreciable inflammatory cell infiltrate; coronary arteries normal on angiogram with severe, diffuse left ventricular dysfunction; renal arteriogram negative for renal artery stenosis; and esophagogastroduodenoscopy negative for any bleeding source. The patient was not taking any other medications. IFN therapy was discontinued and the patient was started on hemodialysis and antihypertensive therapy. He received peritoneal dialysis for approximately 18 months with stabilization of his renal function (BUN 31 mg/dL, creatinine 2.5 mg/dL), allowing discontinuation of dialysis. He continued to have a major cytogenetic response, with 15% Ph-positive cells on bone marrow examination in January 1996 and 30% Ph-positive cells in July 1996. He began to receive ara-C 10 mg/day in October 1997 when his bone marrow analysis revealed 80% Ph-positive cells, with FISH analysis showing 60% Ph-positive cells. In January 1998 he achieved a com- 2586 CANCER June 15, 1999 / Volume 85 / Number 12 TABLE 1 Characteristics of Patients with CML Developing HUS/TTP While on Therapy with rIFN-aa Patients (Reference no.) Characteristics Case 1 Case 2 Case 3 Case 4 (32) Case 5 (33) Case 6 (34) Age (yrs) Gender CML phase Time on rIFN-a Other CML therapy 49 Male CP-AP 67 HU, ara-C 24 Female CP 37 HU, ara-C 43 Male CP 80 HU, ara-C 46 Male CP 30 HU 34 Male AP-BP 16 HU, 6-MP Infectious agent Outcome E. coli Died/RF — HD/CRF — CRF — CRF — Died/CRF 46 Male CP 30 HU, BMT(busulfan, cyclophosphamide, melphalan) — CRF CML: chronic myelogenous leukemia; HUS: hemolytic uremic syndrome; TTP: thrombotic thrombocytopenic purpura; rIFN-a: interferon-a; CP: chronic phase; AP: accelerated phase; BP: blastic phase; HU: hydroxyurea; BMT: bone marrow transplantation; RF: renal failure; HD: hemodialysis; CRF: chronic renal failure. a All patients had Ph-chromosome positive CML. plete cytogenetic response, with 0% Ph-positive cells on karyotyping and FISH analysis. He was still in complete cytogenetic remission and off dialysis in April 1998. DISCUSSION The association of malignant disorders and cancer chemotherapeutic agents with HUS and TTP has been previously described. Most cases occur in patients with adenocarcinomas, gastric adenocarcinoma being the most common,29,32 but also with small cell lung carcinoma, squamous carcinomas, and Hodgkin disease.32 The relative contributions of chemotherapeutic agents or the malignancy are difficult to assess. CML has been uncommonly associated with microangiopathic disorders.23–27 The possible role of rIFN-a in inducing this toxic effect is increasingly suspected as more case reports are described. The clinical features of our patients and those described in the literature are summarized in Table 1. The underlying mechanisms initiating the injury causing HUS and TTP have been studied in detail. TTP develops from release into the circulation of plateletaggregating agents, including the von Willebrand factor multimers (vWF). vWF multimers are normally produced by platelets and endothelial cells and stored within the alpha granules of platelets and the Weibel– Palade bodies of the endothelial cells (peroxidase positive protein aggregates identified by electron microscope).35 Unusually large vWF multimers (ULvWF) are also released from the damaged endothelial cells and are more effective in binding the platelet membrane glycoproteins in conditions of elevated fluid shear stress (relative parallel motion between fluid planes during flow).36 This results in platelet aggregation, producing the characteristic intraluminar platelet thrombi seen in virtually all organs in patients with TTP. In HUS, renal endothelial cell injury, with endothelial cell swelling and the resulting narrowing of the glomerular capillary lumen, is believed to be the initiating event. HUS is commonly preceded by bloody diarrhea caused by infectious organisms, such as Shigella dysenteriae or E. coli.37 The toxins elaborated by these organisms, including the Shiga toxin and the Shiga-like toxins, are capable of binding the predominant membrane glycophospholipid receptor for the Shiga toxins, globotriosyl ceramide (Gb3).38,39 Gb3 is expressed on the membrane of renal endothelial cells and other endothelial cells, and Shiga toxins have been demonstrated in vitro to be directly cytotoxic to endothelial cells.39 Other agents can up-regulate Gb3 expression on the endothelial cells and therefore potentiate the effects of the toxins.40 These include the cytokines interleukin-1(IL-1)-a or -b and tumor necrosis factor (TNF)-a or -b.39,40 The interval between exposure to the chemotherapeutic agent and the development of HUS/TTP varies, ranging from 1 day to 7 months.32 The mechanism of endothelial cell injury is poorly understood; it may involve drug metabolites (free oxygen radicals) or formation of antiendothelial antibodies, similar to quinine-induced HUS. In patients with HUS associated with quinine intake, antibodies to platelet membrane glycoproteins may interact with endothelial cell receptors directly or may induce neutrophils to interact with endothelial cells and damage them, possibly via release of free radicals.41 Circulating cytotoxic antiendothelial antibodies have been demonstrated in patients with HUS.42 Thrombotic Microangiopathy with rIFN-a/Ravandi-Kashani et al. The mechanisms by which rIFN-a therapy may initiate HUS/TTP are not known, but several mechanisms are possible. Perez et al. reported detectable interferon (IFN) levels in the initial 10-day period from the diagnosis of HUS in 16 of 35 children with typical HUS, as compared with none of 10 controls.43 Three of 17 patients had detectable IFN levels after Day 10. They suggested that activated leukocytes and/or their products, such as TNF, IFN, IL-1, and free radicals, can participate in tissue injury and endothelial cell damage with the resulting deleterious effects. Indeed, in an experimental model of HUS, a central role for leukocytes or leukocyte products had been suggested, and in vivo neutralization of TNF or IFN with a specific antiserum protected mice from an HUS-like reaction.44 Therefore, IFN-induced free radical production by activated phagocytes may result in the pathogenesis of HUS. In another report, IFN production in two patients with HUS associated with adenovirus infection was also demonstrated, and a possible nephrotoxic role from elevated IFN levels was speculated.45 Possible nephrotoxic effects of IFN have been previously reported.46,47 Exogenous interferons could cause similar effects. IFNs, together with other interacting cytokines, are known to exert complex immunomodulatory effects on endothelial cells with differential modulatory effects on various endothelial cell surface markers, including the MHC antigens and intracellular adhesion molecules (ICAM).48 Exogenous IFN may up-regulate endothelial cell Gb3 levels and potentiate the deleterious effects of bacterial toxins in receptive individuals. rIFN-g has also been shown to modulate the fibrinolytic response in cultured human endothelial cells, suggesting another mechanism by which these agents may turn the fibrinolytic potential of the endothelium in a prothrombotic way.49 rIFN-a has been shown to increase leukocyte adherence to vascular endothelium, and this, by way of the mechanisms described above, may initiate endothelial cell damage and subsequent release of ULvWF multimers, causing endothelial cell swelling, platelet aggregation, and intraluminal microthrombi formation. In summary, HUS/TTP is a rare side effect of IFN therapy in patients with CML. The mechanisms contributing to its pathogenesis are poorly understood, but IFNs can, through modulatory effects on endothelial receptors, make these cells more susceptible to the effects of bacterial toxins. They may also mediate free radical production by activating phagocytic cells, and as a result cause endothelial cell injury. The release of platelet-aggregating agents from the damaged endothelial cells is probably the final event, resulting in intraluminal thrombus formation and organ damage, 2587 as seen in HUS/TTP. This rare but definite complication of rIFN-a therapy should be recognized early when clinical and laboratory findings are suggestive, so that rIFN-a therapy can be discontinued immediately and appropriate therapeutic measures can be initiated. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Talpaz M, McCredie KB, Trujillo J, Mavligit GM, Gutterman JU. Leukocyte interferon induced cytoreduction in chronic myelogenous leukemia. Blood 1983;62:689 –92. Talpaz M, Kantarjian HM, McCredie KB, Keating MJ, Trujillo J, Gutterman J. Clinical investigation of human alfa interferon in chronic myelogenous leukemia. Blood 1987;69: 1280 – 8. Talpaz M, Kantarjian HM, McCredie KB, Trujillo J, Keating MJ, Gutterman J. Hematologic remission and cytogenetic improvement induced by recombinant human interferon alpha, in chronic myelogenous leukemia. N Engl J Med 1986;314:1065–9. Niederle N, Kloke O, May D, Becher R, Osieka R, Schmidt CG. Treatment of chronic myelogenous leukemia with recombinant interferon alpha-2b. Invest New Drugs 1987;5:19. Kantarjian HM, O’Brien S, Anderlini P, Talpaz M. Treatment of chronic myelogenous leukemia: current status and investigation options. Blood 1996;87:3069 – 81. Kantarjian HM, Keating M, Estey E, O’Brien S, Pierce S, Beran M, et al. Treatment of advanced stages of Philadelphia chromosome-positive chronic myelogenous leukemia with interferon alpha and low dose cytarabine. J Clin Oncol 1992; 10:772. Kantarjian H, O’Brien S, Keating M, Talpaz M. Interferon alpha and low-dose cytosine arabinoside therapy in Philadelphia chromosome-positive chronic myelogenous leukemia [abstract]. Proc ASCO 1997;16:13a. Gilhout F, Chastang C, Michallet M, Guerci A, Tanzer J, Harousseau JL, et al. Interferon alfa-2b combined with cytarabine versus interferon alone in chronic myelogenous leukemia. N Engl J Med 1997;337:223–9. O’Brien S, Kantarjian HM, Talpaz M. Practical guidelines for the management of chronic myelogenous leukemia with interferon alpha. Leuk Lymphoma 1996;23:247–52. Sacchi S, Kantarjian H, O’Brien S, Cohen PR, Pierce S, Talpaz M. Immune-mediated and unusual complications during interferon alfa therapy in chronic myelogenous leukemia. J Clin Oncol 1995;13:2401–7. Kantarjian HM, Smith T, O’Brien S, Beran M, Pierce S, Talpaz M. Prolonged survival in chronic myelogenous leukemia after cytogenetic response to interferon-alfa. Ann Intern Med 1995;122:254 – 61. Akard LP, Hoffman R, Elias L, Saiers JH. Alpha-interferon and immune hemolytic anemia. Ann Intern Med 1986;105: 306. Abdi EA, Venner PM. Immune thrombocytopenia after alpha-interferon therapy in patients with cancer. JAMA 1986; 255:1878 –9. Ronblom LE, Alm GV, Oberg KE. Autoimmunity after alphainterferon therapy for malignant carcinoid tumors. Ann Intern Med 1991;115:178 – 83. Conlon KC, Urba WJ, Smith JW II, Steis RG, Longo DL, Clark JW. Exacerbation of symptoms of autoimmune disease in patients receiving alpha-interferon therapy. Cancer 1990;65: 2237– 41. 2588 CANCER June 15, 1999 / Volume 85 / Number 12 16. Wandl UB, Nagel-Hiemke M, May D, Kreuzfelder E, Kloke O, Kranzhoff M, et al. Lupus-like auto-immune disease induced by interferon therapy for myeloproliferative disorders. Clin Immunopathol 1992;65:70 – 4. 17. Kurschel E, Metz-Kurschel UM, Niederle N, Aulbert E. Investigations on the subclinical and clinical nephrotoxicity of interferonn-a-2b in patients with myeloproliferative syndromes. Renal Failure 1991;13:87–93. 18. Gisslinger H, Gilly B, Woloszczuk W. Thyroid autoimmunity and hypothyroidism during long-term treatment with recombinant interferon-alpha. J Exp Immunol 1992;90:363–7. 19. Sangster G, Kaye SB, Calman KC. Cutaneous vasculitis associated with interferon. Eur J Cancer Clin Oncol 1983;19: 1647–9. 20. Niizeki H, Inamoto N, Nakamura K, Tsuchimoto K, Hashimoto T. A case of pemphigus foliaceus after interferon alpha-2a therapy. Dermatology 1994;189:195–9. 21. Guyer DR, Tiedeman J, Yannuzzi LA, Slakter JS, Parke D, Kelley J, et al. Interferon-associated retinopathy. Arch Ophthalmol 1993;111:350 – 6. 22. Sonnenblick M, Rosin A. Cardiotoxicity of interferon: a review of 44 cases. Chest 1991;99:557– 61. 23. Harvey M, Rosenfeld D, Davies D, Hall BM. Recombinant interferon alpha and hemolytic uremic syndrome: cause or coincidence? [letter] Am J Hematol 1994;46:152–3. 24. Schlaifer D, Dumazer P, Spenatto N, Mignon-Conte M, Brousset P, Lumbroso C, et al. Hemolytic uremic syndrome in a patient with chronic myelogenous leukemia treated with interferon alpha [letter]. Am J Hematol 1994;47:254 –5. 25. Lipton JH, Minden M. CML may not be part of HUS [letter; comment]. Am J Hematol 1995;49:100 –1. 26. Kobayashi H, Utsunomiya Y, Miyazaki Y, Tokutome G, Kawamura T, Hashimoto T, et al. Case of chronic myelogenous leukemia with hemolytic uremic syndrome–like kidney during long-term interferon-alpha therapy [in Japanese]. J Jpn Int Med 1997;86:1259 – 61. 27. Shammas F, Meyer P, Heikkila R, Apeland T, Goransson L, Berland J, et al. Thrombotic microangiopathy in a patient with chronic myelogenous leukemia on hydroxyurea. Acta Hematol 1997;97:184 – 6. 28. Stratta P, Canavese C, Dogliani M, Thea A, Degani G, Mairone L, et al. Hemolytic uremic syndrome during recombinant alpha-interferon treatment for hairy cell leukemia. Renal Failure 1993;15:559 – 61. 29. Antman K, Skarin A, Mayer R. Microangiopathic hemolytic anemia and cancer: a review. Medicine 1979;58:377– 84. 30. Murgo A. Thrombotic microangiopathy in the cancer patient including those induced by chemotherapeutic agents. Semin Hematol 1987;24:161–77. 31. Liu K, Mittleman A, Sproue E, Elias EG. Renal toxicity in man treated with mitomycin C. Cancer 1971;28:1314 –20. 32. Gordon LI, Kwaan HC. Cancer- and drug-associated thrombotic thrombocytopenic purpura and hemolytic uremic syndrome. Semin Hematol 1997;34:140 –7. 33. Brodowicz T, Breiteneder S, Wiltschke C, Zielinski CC. Gemcitabine-induced hemolytic uremic syndrome: a case report. J Natl Cancer Inst 1997;89:1895– 6. 34. Sakai C, Takagi T, Wakatsuki S, Matsuzaki O. Hemolytic uremic syndrome due to deoxycoformycin: a report of the second case. Int Med 1995;34:593– 6. 35. Ruggeri ZM, Zimmerman TS. von Willebrand factor and von Willebrand disease. Blood 1987;70:895–904. 36. Moake JL, McPherson PD. Abnormalities of von Willebrand factor multimers in thrombotic thrombocytopenic purpura and hemolytic uremic syndrome. Am J Med 1989;87:3–9N. 37. Ashkenazi S. Role of bacterial cytotoxins in hemolytic uremic syndrome and thrombotic thrombocytopenic purpura. Annu Rev Med 1993;44:11. 38. Lindberg AA, Brown JE, Stromberg N, Westling-Ryd M, Schultz JE, Karlsson KA. Identification of the carbohydrate receptor for Shiga toxin produced by Shigella dysenteriae type I. J Biol Chem 1987;262:1779. 39. Obrig TG. Pathogenesis of Shiga toxin (verotoxin)–induced endothelial cell injury. In: Kaplan BS, Trompeter RS, Moake JL. Hemolytic uremic syndrome and thrombotic thrombocytopenic purpura. New York: Marcel Dekker, 1992:405. 40. van de Kar NCAJ, Monnens LAH, Karmali MA, van Hinsberg VWM. Tumor necrosis factor and interleukin-1 induce expression of the verocytotoxin receptor globotriaosylceramide on human endothelial cells: implication for the pathogenesis of hemolytic uremic syndrome. Blood 1992;80:2755. 41. Stroncek DF, Vercellotti GM, Hammerschmidt DE, Christie DJ, Shanker RA, Jacob HS. Characterization of multiple quinine-dependent antibodies in a patient with episodic hemolytic uremic syndrome and immune agranulocytosis. Blood 1992;80:241. 42. Leung DYM, Moake JL, Havens PL, Kim M, Pober JS. Lytic anti-endothelial cell antibodies in hemolytic-uremic syndrome. Lancet 1988;2:183– 6. 43. Perez N, Lasarte P, Bibiloni N, Dejean CB. Alpha/beta interferon in serum from children with the hemolytic uremic syndrome. Acta Paediatr Scand 1989;78:637– 8. 44. Butler T, Rahman H, Ali-Mahmud K, Islam M, Bardhan P, Kabir I, et al. An animal model of hemolytic uremic syndrome in shigellosis: liposaccharides of Shigella dysenteriae I and Shigella flexneri produce leukocyte-mediated renal cortical necrosis in rabbits. Br J Exp Pathol 1985;66:7–15. 45. Blachar Y, Leibovitz E, Levin S. The interferon system in two patients with hemolytic uremic syndrome associated with adenovirus infection. Acta Paediatr Scand 1990;79:108 –9. 46. Gresser I, Maury C, Tovey M, Morel-Maroger L, Pontillon F. Progressive glomerulonephritis in mice treated with interferon preparations at birth. Nature 1976;263:420 –2. 47. Selby P, Kohn J, Raymond J, Judson I, McElwain T. Nephrotic syndrome during treatment with interferon. Br Med J 1985;290:1180. 48. Douglas MS, Rix DA, Dark JH, Talbot D, Kirby JA. Examination of the mechanism by which heparin antagonizes activation of a model endothelium by interferon-gamma (IFNgamma). Clin Exp Immunol 1997;107:578 – 84. 49. Arnman V, Stemme S, Rymo L, Risberg B. Interferon-gamma modulates the fibrinolytic response in cultured human endothelial cells. Thromb Res 1995;77:431– 40.