Primary Nonresponse to Anti‐Cd20 Therapy in Gastrointestinal Posttransplant Lymphoproliferative Disorder

Abstract

INTRODUCTION Posttransplant lymphoproliferative disorder (PTLD) is a life-threatening complication of solid organ transplantation. Pediatric recipients of liver allografts constitute a major risk group for PTLD. PTLD is mostly of B-cell origin and commonly associated with Epstein-Barr virus (EBV) infection (1-3). Etiology of liver failure has been proposed as one of the factors determining the risk of PTLD. Children receiving transplants for biliary atresia, fulminant hepatic failure, progressive familial intrahepatic cholestasis, hepatoblastoma, and particularly for Langerhans cell histiocytosis appear to be at increased risk (4). Other known risk factors include lack of previous EBV infection (i.e., EBV seronegativity), young age, and treatment with antilymphocyte antibodies (3,4). It remains difficult to identify optimal treatment for PTLD patients. Although there is general agreement that reduction of immunosuppression is the mainstay of therapy, indications for auxiliary therapies including antivirals, immunotherapy, and chemotherapy are still debated (2,5). Recently, promising results have been reported by several investigators using anti-CD20 antibodies in pediatric liver transplant recipients (6,7). We report a 4-year-old boy with gastrointestinal (GI) PTLD presenting with perforation who did not respond to reduction of immunosuppression and treatment with anti-CD20 antibody followed by chemotherapy. CASE REPORT A 4-year-old boy who had undergone orthotopic liver transplantation because of fulminant hepatic failure 4 months previous presented to our clinic with fever, anorexia, acute diarrhea, and weight loss. He was receiving tacrolimus and low-dose methylprednisolone (0.3 mg/kg/day); he had not experienced any rejection episode in the posttransplant period. His physical examination was unremarkable other than moderate dehydration. There was mild anemia on his complete blood count; white blood cell count was normal with polymorphonuclear leukocyte dominance on blood film. Blood chemistry revealed an increased blood urea nitrogen with normal serum creatinine levels. Mean trough tacrolimus level was 7 ng/mL. Supportive treatment with intravenous fluids was started. Stool and blood cultures were negative, excluding the possibility of infections. Analysis of the stool for the presence of white blood cells and occult blood was also negative. Abdominal ultrasound did not show any specific findings. He was seropositive for EBV before transplantation and at presentation; however, we were not able to perform EBV viral load determination. During his clinical follow-up, diarrhea, anorexia, and fever did not regress. On the 10th day of admission, he developed an acute abdomen and underwent urgent laparotomy. There were perforations in the distal ileum 30 cm proximal to caecum, thickening of the bowel wall near the perforation, and multiple mesenteric lymph nodes smaller than 1.5 cm in diameter. The perforated segment was resected together with the macroscopically affected intestine. In the pathologic specimen, uniform lymphoid cell infiltration was present (Fig. 1). According to the World health Organization (WHO) classification, histopathologic diagnosis was consistent with monoclonal CD20-positive diffuse large cell lymphoma with kappa light chain expression (WHO-ICD-O-9680) (Fig. 2), although there was no monoclonal gammopathy in serum and urine immune electrophoresis. Immunohistochemical EBV staining with latent membrane protein 1 (LMP1) in the formalin-fixed, paraffin-embedded tissue was positive, suggesting EBV associated PTLD (monoclonal mouse anti-EBV, latent membrane protein, DakoCytomation, Glostrup, Denmark).FIG. 1: (A) Diffuse lymphoid cell infiltration involving all layers of small bowel wall (hematoxylin-eosin [H&E], magnification ×100). (B) Immunoblastic, plasmoblastic, and plasmocytic cell differentiation within neoplastic infiltration (H&E, magnification ×1,000).FIG. 2: Neoplastic cells with diffuse and strong CD20 positivity in the intestinal specimen (anti-CD20 monoclonal antibody "L26," 3-amino-9-ethylcarbazole chromogen, magnification ×400).Further investigations were performed to determine the extent of the disease. On computed tomography (CT) of the abdomen, there was a solid lesion, 3 cm in diameter, near the hilum of the liver. A biopsy sample was taken percutaneously from the lesion, and its histologic and immunohistochemical properties were found to be identical with those of the lesion in the ileum. CT of the head and chest, total body scan, bone marrow aspiration, and lumbar puncture were all negative. Tacrolimus was stopped. Three daily doses of intravenous immunoglobulin (IVIG) (0.4 g/kg/dose) and parenteral ganciclovir (5mg/kg/day) for 10 days were given. Lymphocyte subtype analysis results, obtained at that time, showed increased CD8 (82%) together with decreased CD4, CD19, and CD56 levels (3%, 1.45%, and 9%, respectively), confirming a state of natural killer cell depletion together with a reversed CD4:CD8 ratio. Anti-CD20 treatment (rituximab, Roche, Basel, Switzerland 375 mg/m2 per week) was started on the fifth day of diagnosis, when it became available. During the anti-CD20 treatment, fever was partially controlled, but diarrhea continued and pancytopenia developed. After the third dose of anti-CD20, GI bleeding with massive melena was observed. Endoscopic examination showed a necrotic 1 cm mass in the second portion of the duodenum (Fig 3). Because of uncontrolled lower GI bleeding he was reoperated on. Multiple perforations and necrotic lesions were present throughout the small bowel with diffuse tumoral infiltration and multiple lymphadenopathy in the mesentery. Histology yielded the same characteristics as the primary tumor. The disease was thought to be progressive and chemotherapy (cyclophosphamide, doxorubicin, vincristine, prednisone regimen) was started. Despite chemotherapy, the patient died because of uncontrolled PTLD. His family refused autopsy.FIG. 3: Endoscopic appearance of the necrotic tumoral lesion in the second portion of duodenum.DISCUSSION Here, we report a case of PTLD that progressed despite immunosuppression reduction, antiviral therapy, IVIG, and anti-CD20 monoclonal antibody treatment and subsequently resulted in death. PTLD is a unique form of non-Hodgkin lymphoma evolving after organ transplantation. Incidence of non-Hodgkin lymphoma is increased 20 to 120 times in patients receiving solid organ transplants (5). The degree and duration of immunosuppression are important determining factors in the development of de novo malignancies. The overall incidence of PTLD in liver transplant patients is estimated to be 2% to 4%, but the reported incidence in pediatric patients varies between 4.1% and 14%, depending on the age of the patient (1,3-5). PTLD can occur in different forms: benign self-limiting primary infectious mononucleosis-like syndrome, polyclonic polymorphic B-cell hyperplasia, or aggressive monoclonal immunoblastic non-Hodgkin lymphoma (1,5,8). It is commonly associated with EBV, although it can also occur in the absence of EBV infection (4,6). Several factors predictive of PTLD have been reported. The most widely accepted ones are the recipient's age and EBV viral load. Stable monoclonal gammopathy also appears to be highly predictive of PTLD (5). The immunosuppressive regimen has an important role in PTLD (5,8). The risk factors for our patient were younger age and transplantation for fulminant liver failure. He was seropositive for EBV before transplantation, excluding primary EBV infection as a risk factor for PTLD in the posttransplant period. However, although we were not able to measure EBV viral load, presence of LMP1 in the tumor tissue confirms the diagnosis of EBV-associated PTLD, suggesting reactivation of the virus in the posttransplant period. No monoclonal gammopathy was observed in the immune electrophoresis of serum and urine in our patient. However, lymphoid proliferation is best characterized by using a combination of histopathology, immunophenotyping, and molecular studies of frozen tissue, and the diagnosis of PTLD should be based on tumor biopsy with molecular characterization of clonality (5). EBV-associated monoclonal CD20-positive diffuse large-cell lymphoma with kappa light chain expression was diagnosed by histopathology and immunohistochemistry in our patient. Clinical presentation of PTLD may occur in different forms. Patterns of GI involvement and endoscopic diagnosis in pediatric PTLD have been extensively defined (9-11). GI bleeding is the most striking feature of PTLD with GI involvement, whereas unexplained protracted diarrhea appears to be less specific (9,10). The presenting symptom was acute diarrhea in our patient, and intestinal perforation occurred before any signs of GI bleeding. On the basis of their experience with primary EBV infection in pediatric liver transplant recipients, Smets et al. (12) have suggested that digestive symptoms and digestive PTLDs are late complications of primary EBV infection, and systematic screening for acquisition of EBV infection might help to reduce the risk of digestive PTLD by enabling adjustment of immunosuppression. However, our patient was seropositive for EBV before transplantation. Taking his pretransplant EBV status and normal serum immune electrophoresis into consideration, we did not have a high index of suspicion for PTLD in our patient before the perforation occurred; so, unfortunately, endoscopy was delayed until after definitive diagnosis on the 10th day of admission. There is no general agreement about the optimal treatment for PTLD. Prognostic factors for determining the choice of treatment in different patients have not been assigned (13,14). Reduction or possibly discontinuation of immunosuppression is a major step to be taken. Although successful outcomes using immunosuppression reduction together with antiviral therapy have been reported, decreased immunosuppression per se has shown to be insufficient in a majority of PTLD patients (1,3,4,8,14,15). Nevertheless, on the basis of single-center reports, four major areas of treatment should be considered: reduction of immunosuppression, chemo- and biologic therapy; anti-B-cell monoclonal antibody therapy, and cell-based therapies. Surgical extirpation and localized radiotherapy in limited disease and surgical excision for rapidly growing obstructive tumors of the gut and adenoids have also been proposed (5,16-18). Promising results with anti-B-cell monoclonal antibodies (mAb) reported in recent years have encouraged us to use anti-CD20 together with immunosuppression reduction, antiviral therapy, and IVIG (6-8). Although EBV-negative PTLD might be expected to be less responsive to primary treatment modalities compared with EBV-related disease, a poor treatment response was observed in association with EBV-positive PTLD in our patient (19). LeVasseur et al. (20) have shown that natural killer cell depletion with a reversed CD4:CD8 ratio might predispose an immunosuppressed patient to recalcitrant PTLD, and they reported successful resolution of PTLD after anti-CD20 therapy in these cases. Unfortunately, lymphoproliferative disease was highly progressive in our patient, spreading throughout the small intestine with multiple areas of necrosis and perforation despite anti-CD20 therapy, in contrast with the previous reports of successful results with anti-CD20 even in recalcitrant PTLD and other forms of non-Hodgkin lymphoma (20-22). Intestinal perforation caused by massive necrosis during effective cytotoxic therapy is well-known, and it has been reported even with successful use of anti-CD20 (23). However, the first perforation occurred at the time of diagnosis in our patient before any treatment causing cell lysis had been started. Benkerrou et al. (17) have shown that multivisceral disease, central nervous system involvement and late-onset PTLD are risk factors for partial or no response to treatment with anti-B-cell mAb. None of these existed in our patient. In a recent case-control study, Allen et al. (24) stressed a tendency for unfavorable outcome in PTLD patients who were EBV-seropositive pretransplantation. The clinical course and nonresponse to primary treatment modalities in our patient was in accord with their finding. Moreover, when we reevaluated the initial presentation and clinical course of our patient, we considered that intestinal involvement with necrosis and perforation at diagnosis, suggesting a high turnover of the tumor cells, might be predictive of aggressive course and possibly nonresponse to anti-B-cell mAb. Chemotherapy protocol was started late in the clinical course after 3 weeks of anti-CD20 treatment and unfortunately was also unsuccessful in controlling the progression of the disease. The delay in starting chemotherapy might have contributed to the ineffectiveness of chemotherapy in our patient. Although indications for chemotherapy in the treatment of PTLD are still controversial, excellent results have been reported with selected regimens in particular patients with EBV-related PTLD (25). Leung et al. (26) have also stressed the success of early intensive chemotherapy started at the time of diagnosis in selected patients. Other investigators have also reported early chemotherapy to be effective and safe in GI PTLD without increasing the risk of rejection (27). However, late occurring lymphomas, which are probably a distinct entity, do not appear to be responsive to current chemotherapy protocols (28). Although our patient had early EBV-associated PTLD, his treatment response was poor, contradicting the results reported in previous studies (25,26). However, a recent survey of 107 adult solid organ transplant recipients with PTLD did not show significant difference between early and late PTLDs in terms of survival, despite their different characteristics (29). In conclusion, PTLD is a worrisome complication of solid organ transplantation. Despite advances in its management, there is no uniform and effective treatment. However, there is no doubt that early diagnosis is of paramount importance for successful cure. Although chemotherapy is not proposed as a first-line therapy by many and is reserved for aggressive types of PTLD, there might be some initial findings predictive of an aggressive course and nonresponse to other treatment modalities including anti-B-cell mAb. Intestinal involvement with necrosis and perforation at diagnosis may have predictive value for an aggressive course of early chemotherapy, even in early and EBV-associated PTLD.