|Home | About | Journals | Submit | Contact Us | Français|
Current models of neoplasia indicate a stepwise series of events as the usual mechanism by which normal cells become neoplastic cells.1 However, cancer clinically presents as a fully malignant lesion capable of causing the death of the host. For this reason, it is presently not possible to trace the entire process that led to this condition.
The occurrence of lymphoproliferative disorders in immunosuppressed transplant recipients holds significant promise that a model of the early stages of neoplasia might exist in humans. These lesions, originally interpreted as non-Hodgkin’s lymphomas, occur in association with Epstein-Barr virus infection in most cases.2 Both polyclonal and monoclonal forms have been described.3–5
We previously observed that a significant number of posttransplant lymphoproliferative disorders (PTLDs) seen in association with cyclosporine A(CsA)-steroid immunosuppression were capable of regression after a reduction of immunosuppression.4,6
Our continuing studies are designed to explore the limits of this phenomenon and define the host:tumor factors that are operative within this system. The present report describes a retrospective clonal analysis of tumor tissues by using recombinant DNA probes specific for the immunoglobulin heavy-chain region.
Forty-four patients in the Pittsburgh-Denver transplant series received diagnoses of PTLDs between June 1980 and March 1987. Clonal designation was based originally on the results of immunoperoxidase staining of cytoplasmic immunoglobulins in paraffin-embedded tissues. The avidin-biotin method of Hsu et al7 was combined with pronase digestion. Antibodies of k and λ immunoglobulin light chains were obtained from Dako Corp (Santa Barbara, CA).
Twenty-three patients whose lesions were studied by this technique also had sufficient tissue available for clonal analysis of B cells by using recombinant DNA methodology.
DNA purification and restriction endonuclease and Southern blot analyses were performed in accordance with protocols previously described from our laboratory.8 Frozen tissue specimens were pulverized in dry ice, freeze-dried, and solubilized in standard lysing solution before routine purification.9 High-molecular weight DNA was digested with BamHI, which optimally displayed heavy-chain gene rearrangements. The recombinant DNA probe for the joining region of the immunoglobulin heavy-chain gene (Jh) was kindly supplied by Dr Philip Leder, Harvard University.10
The group of 44 patients was composed of 30 males and 14 females (2.1:1). Ages at transplant ranged from 1 to 62 years, with a mean age of 23 years. There were 24 liver, 12 kidney, 5 heart, and 3 heart/lung recipients in the population who accounted for 2.2%, 1.0%, 1.8%, and 4.6% of the respective organ allograft subpopulations. Two patients initially received non-CsA–containing immunosuppressive regimens. Tumors in these patients occurred at 68 and 162 months posttransplant. In the remainder of the population CsA was part of the original immunosuppressive regimen. Tumors in this subgroup of 42 patients occurred at time intervals ranging from less than 1 month up to 26 months posttransplant. The mean time of tumor onset in this latter group was 6.2 months.
As previously reported, the majority of patients had evidence of either primary or reactivation Epstein-Barr virus infection.11
The presentation of PTLD was localized to the head and neck region in 14 patients. This was most often manifested as a mononucleosislike syndrome with marked tonsillar hypertrophy and/or a cervical mass. Eleven patients had disease predominant in or localized to the gastrointestinal tract. A third category of patients had solid organ involvement. In many cases of solid organ presentation, the allograft organ was involved by the process. A subcategory of patients in this group presented with fulminant multisystem disease. Eight other patients presented with a variety of findings ranging from vague constitutional disorders to Pneumocystis pneumonia.
PTLDs may exist predominantly as tumorous or infiltrative processes. In typical cases the predominant cell has features of a B lymphocyte. All histological stages of lymphocyte maturation may be seen, a condition referred to as polymorphous or pleomorphic. Conversely, a monomorphous appearance in which all cells appear identical to each other may be seen. The lesions often show extensive areas of necrosis and may contain large atypical-appearing lymphoid cells.
By using a combination of immunocytochemical and recombinant DNA methodologies we have concluded that tumors were monoclonal in 12 patients, polyclonal in 15 Patients, and both monoclonal and polyclonal (noncontiguous) in four patients. One patient had a polyclonal tumor that later evolved into a probable monoclonal tumor. Clonal status has not yet been assigned in 12 cases.
In 23 cases both immunoperoxidase studies and immunoglobulin heavy-chain gene rearrangement studies were performed. A representative autoradiogram is shown in Fig 1.
Five cases were categorized as monoclonal by immunocytochemical stain. All five were also shown to be monoclonal by DNA analysis. In contrast, ten cases demonstrated a polyclonal phenotype by immunoperoxidase. By using DNA analysis two of these cases were shown to contain a monoclonal element. Six other cases were inconclusive on the basis of immunoglobulin light-chain stains. DNA rearrangement patterns revealed five of these cases to be polyclonal and one to be monoclonal.
On occasion, multiple tumors were analyzed by one technique but not by another. Two cases categorized as polyclonal by phenotype had separate tumors analyzed by DNA probes. In both cases monoclonal tumors were uncovered. The converse situation was obtained in two cases in which additional tissue was available for immunoglobulin staining. Two cases categorized as either monoclonal or polyclonal on the basis of heavy-chain gene rearrangement analysis were each shown to have had separate monoclonal and polyclonal tumors when separate specimens were stained for immunoglobulin light chains.
A full description of the histological aspects of CsA-associated PTLDs is in preparation. In brief, a pleomorphic histology was associated with polyclonality in most but not all cases. Occasional pleomorphic cases contained a monoclonal component. In contrast, a monomorphous histology invariably implied a monoclonal tumor. In some cases, minimal pleomorphism was seen in association with prominent plasmacytoid features. These lesions were frequently associated with monoclonality.
Seven of 12 patients with monoclonal disease are alive at intervals ranging from 7 to 58 months after diagnosis. Only one of the seven received chemotherapy in the initial therapeutic regimen. There have been two cases of recurrence in these seven patients. One case of persistently recurrent disease was subsequently treated with chemotherapy. Another case resolved after reduction of immunosuppression. Five of 12 patients with monoclonal tumor died, one despite the use of chemotherapy.
Resolution of polyclonal lesions was seen in nine of 11 cases. One patient required subsequent radiotherapy for control of a lesion morphologically consistent with Hodgkin’s disease. The remainder of the nine patients responded to a reduction of immunosuppression. Two of the nine later died of other causes with no evidence of tumor. One patient had recurrent polyclonal disease that again resolved. One patient developed unresponsive, widespread, presumably monoclonal disease and died of aspergillosis in association with chemotherapy.
The 5-year survival rate for the entire group of 44 patients is 63%. No difference is seen when these figures are segregated according to clonality.
The increased incidence of lymphoproliferative disorders in transplant patients was first recognized by Starzl in 1968.12 Purtilo first suggested that Epstein-Barr virus might contribute to this disorder.13 This was found to be the case in studies of transplant patients at the University of Minnesota.2
These prior studies were based on patients receiving conventional immunosuppression. Differences in tumor localization, tumor onset time, and reversibility of tumors are seen when CsA-containing immunosuppressive regimens are used.14
The present studies indicate that both monoclonal and polyclonal tumors occur in association with CsA immunosuppression. Analysis of immunoglobulin genes is shown to be a more sensitive indicator of clonal status than is analysis of cytoplasmic immunoglobulin expression.
It is important to emphasize that clonal status alone does not constitute the dividing line between benign and malignant lesions in these cases. In our experience modulation of immunosuppression results in resolution of most polyclonal lesions and some monoclonal lesions. The lack of response in some cases may reflect the fact that therapy was started too late in the course of the disease. Indeed, patients were as likely to die from overwhelming polyclonal disease as from overwhelming monoclonal disease (Starzl et al, unpublished data).
In a distinct minority of cases it is likely that the tumor had progressed to a stage that placed it beyond the reach of host control mechanisms. Prospective studies are under way to identify additional factors that contribute to the fully developed malignant phenotype in these tumors.
The clinician caring for a transplant patient with this syndrome should be aware that the majority of cases will respond to a careful reduction of immunosuppression with control of local complications. Aggressive chemotherapy in these immunosuppressed patients should only be used in those rare cases in which tumor cannot be controlled by other means.
The ultimate lessons of these lesions have not yet been realized. It is likely that continued studies of PTLDs will reveal important underlying mechanisms of immunosurveillance and tumor progression in the human. Thus the importance of PTLDs may be far in excess of their sporadic occurrence in this highly select group of patients.
Supported in part by American Cancer Society Institutional Grant IN-58X (M.A.N.).