At presentation to the National Cancer Institute, the patient reported fatigue, back and pelvic pain, and night sweats. Physical examination showed that the patient had a cushingoid appearance, but peripheral adenopathy and splenomegaly were not seen. To determine the presence of the CD22 antigen on the tumor cells, the patient underwent percutaneous biopsy of the para-aortic adenopathy. The biopsy specimen showed a malignant lymphoma with extensive necrosis (Figure 1, panel A), which was categorized as a large B-cell lymphoma according to the REAL classification [5]. A portion of the biopsy specimen was stained with a panel of monoclonal antibodies by an immunoperoxidase technique [7]. The cells were positive for B-cell antigens, including CD19, CD20, CD21, and CD22, and were negative for T-cell antigens. In situ hybridization was performed for Epstein-Barr virus using the Epstein-Barr encoded RNA (EBER1) probe [8]; U6 (an abundant RNA polymerase III transcript of cellular origin) was used to control for preservation of RNA. All of the large lymphoid cells showed a positive signal (Figure 1, panel B). These data clearly indicate the relation between the occurrence of this tumor and Epstein-Barr virus infection. In paraffin-embedded sections of the biopsy specimens (not shown), the nuclei showed positive staining for p53 tumor suppressor protein. This finding strongly suggests the presence of a mutated form of the p53 tumor suppressor gene, which has been seen in patients with lymphomas who had poor prognosis with chemotherapy [9].

The patient began treatment with the immunotoxin at the previously determined maximal tolerated dose, 19.2 mg/m² body surface area every 192 hours by continuous infusion [10]. The immunotoxin was given in three courses at monthly intervals until July 1994. Since January 1994, the patient had continued to receive the same dose of acyclovir and prednisone (5 mg alternating with 10 mg daily) to prevent renal graft rejection and symptoms of adrenal insufficiency. No further immune manipulation was done before or during administration of the immunotoxin. The first course of immunotoxin therapy was complicated by mild to moderate vascular leak syndrome (grade II) [10]; on the third day of the first course, increasing peripheral edema, moderate weight gain, resting tachycardia, and mildly decreased urine output were seen. The patient continued to receive immunotoxin therapy and was also given furosemide and 1 unit of packed red blood cells; the signs and symptoms then resolved. Azotemia did not occur. After the first course of immunotoxin therapy, the patient was considered to have had a partial response in both para-aortic and pelvic adenopathy, measured by computed tomography of the abdomen and pelvis. No evidence suggested that human antimouse or human antiricin antibodies had developed. With the second and third courses of immunotoxin therapy, the only toxic effect was a slightly decreased serum albumin level.

After the third course of immunotoxin therapy, para-aortic adenopathy markedly improved but the pelvic mass did not decrease further. Computed tomography done 12 months after immunotoxin therapy began showed no evidence of disease (compare panel C with panel D of the Figure 1). A negative gallium scan also indicated that the patient had a stable, complete response, although the avidity of the tumor for gallium had not been assessed before treatment.

In October 1995, biopsy of the transplanted kidney was performed because the serum creatinine level had progressively increased from 0.9 mg/dL (the level in May 1995) to 1.8 mg/dL. The biopsy specimen showed well-preserved glomeruli, mild to moderate interstitial fibrosis, and a rich cellular infiltrate that was consistent with rejection. No evidence suggested recurrent glomerulonephritis or a lymphoma. In situ hybridization studies of the biopsy specimen showed no expression of EBER-1 RNA. The patient was treated with a high, tapering dose of prednisone for 3 weeks, and her creatinine level stabilized at 1.5 mg/dL. A specimen obtained from repeated renal biopsy in December 1995 showed resolution of the cellular infiltrate. Thirty-two months after immunotoxin treatment began, the patient remains clinically well; she is in complete remission, and the transplanted kidney has been retained.

Pharmacology

Using methods described elsewhere [11], we measured the immunotoxin concentration during treatment. The peak concentration achieved during the first course of immunotoxin therapy (1833 ng/mL) was greater than that seen during the second (1130 ng/mL) and third (964 ng/mL) courses. These findings correlated with evidence of less toxicity (grade I vascular leak syndrome during the second and third courses compared with grade II during the first course). This result is consistent with our previous observation of a clear relation between the maximal serum concentration of RFB4-dgA and the degree of vascular leak syndrome [10]. The basis for this nonuniform pharmacologic behavior during each course is not clear but is consistent with the interpatient and intrapatient variability encountered with this immunotoxin [10, 11]. The terminal half-lives observed (19.6 to 37.3 hours) were similar to those seen previously with this immunotoxin in patients with non-Hodgkin lymphoma who did not have circulating tumor cells. Prolonged terminal half-lives of approximately this magnitude have also correlated with development of clinical response to this immunotoxin [11].

Discussion

Top Glossary Case Report Discussion Author & Article Info References

We have documented that the anti-CD22 immunotoxin RFB4-dgA caused a complete, sustained response in a patient who developed an Epstein-Barr virus-associated large B-cell lymphoma after renal transplantation. This patient had clearly been refractory to decreased immunosuppression, specific antiviral therapy, and combination chemotherapy. The treatment was tolerated without serious toxicity or loss of graft function.

The subclassification of post-transplantation lymphoproliferative disorders has been difficult to correlate with clinical outcome. A recent study [3] indicated that histologic progression to monomorphic-cell lymphoma correlated with secondary oncogenic mutational events and poor response to discontinuation of immunosuppression. The most common secondary event in large-cell lymphoma is p53 mutation, seen in two of three cases studied. Notably, our patient's tumor showed strong nuclear staining for p53, a result that correlates closely with the existence of p53 mutation [9]. Thus, our result is notable because it occurred in a patient who was expected to have a poor outcome.

The optimal treatment for non-Hodgkin lymphoma that develops after transplantation, is associated with Epstein-Barr virus, and is refractory to diminished immunosuppression and antiviral treatment must still be defined. Conventional chemo-therapy regimens are not effective and may be associated with an increased risk for serious infections in immunosuppressed patients. Because RFB4-dgA does not target granulocytes, treatment with this immunotoxin may avoid this complication. In a recent series [1], only three of nine "conventionally" treated patients responded to chemotherapy; 75% of patients died after a median of 14 months.

Recent efforts have sought to develop therapies that are based on immunologic modulation of tumor function. Thus, Papadopoulos and colleagues [12] reported complete responses in five of five patients who had a lymphoma associated with Epstein-Barr virus and had undergone allogeneic bone marrow transplantation (in which the tumors are usually of donor origin) after infusions of donor lymphocytes. In this study, however, two deaths resulted from respiratory failure of unknown cause in addition to the expected incidence of graft-versus-host disease. In another study [13], anti-B-cell monoclonal antibodies directed against CD21 and CD24 were ineffective in patients with monoclonal B-cell proliferation but induced complete response in 16 of 22 patients with oligoclonal disease. Complete responses in some patients with monoclonal disease were also documented in a more recent study that used monoclonal B-cell antibodies [14].

The immunotoxin described in our report was recently studied in two clinical trials of chemotherapy-refractory non-Hodgkin lymphoma that was not related to transplantation [10, 11]. When the immunotoxin was given by either intermittent bolus or continuous infusion, approximately 25% of patients had a clinical response. In corroboration of these findings, the pharmacology we observed supports the idea that approximately 1000 ng of immunotoxin per mL (the serum concentration previously defined as safe) is well tolerated; in our patient, this concentration was associated with a notable clinical response.

Our report has some limitations. First, this case represents an isolated although successful use of immunotoxin therapy in a patient with a chemotherapy-refractory Epstein-Barr virus-related large B-cell lymphoma. Second, although RFB4-dgA recognized the malignant B cells in this patient's tumor, other B-cell antigen-directed immunotoxins could also have a role in the treatment of post-transplantation lymphoproliferative disorders. Indeed, it is unclear whether the toxin portion of the molecule is necessary; in mice, however, this portion is required to induce response in human lymphoma xenografts not associated with Epstein-Barr virus [15]. Our result calls for a more systematic study of immunotoxins for this admittedly small but clinically challenging population of patients. Immunotoxins are particularly attractive for study in post-transplantation lymphoproliferative disorders because they may avoid the toxicities associated with conventional chemotherapy; preserve graft function; and, as seen in our patient, ultimately preserve the quality of life.

Drs. Vitetta, Uhr, and Ghetie: Cancer Immunobiology Center, University of Texas Southwestern Medical Center, North Campus, 6000 Harry Hines Boulevard, Dallas, TX 75235.

Drs. Figg and Lush: Medicine Branch, Division of Clinical Science, National Cancer Institute, National Institutes of Health, Building 10, Room 5A01, 9000 Rockville Pike, Bethesda, MD 20892.

Dr. Stetler-Stevenson: Laboratory of Pathology, Division of Clinical Sciences, National Institutes of Health, Building 10, Room 2N108, 9000 Rockville Pike, Bethesda, MD 20892.

Dr. Kershaw: Division of Renal Medicine, University of Massachusetts Medical Center, 55 Lake Avenue North, Worcester, MA 01655.

Dr. Kingma: Laboratory of Pathology, Division of Clinical Sciences, National Institutes of Health, Building 10, Room 2A33, 9000 Rockville Pike, Bethesda, MD 20892.

Dr. Jaffe: Laboratory of Pathology, Division of Clinical Sciences, National Institutes of Health, Building 10, Room 2N202, 9000 Rockville Pike, Bethesda, MD 20892.

Dr. Sausville: Developmental Therapeutics Program, National Cancer Institute, Executive Plaza North, Suite 843, 6130 Executive Boulevard, Rockville, MD 20852.