Dr Stephan Oertel, Humboldt Universität Berlin, Universitätsklinikum Charite- Campus Virchow, Medizinische Klinik mit Schwerpunkt Hämatologie und Onkologie, Augustenburger Platz 1, 13353 Berlin, Germany. E-mail: email@example.com
Summary. This pilot study assessed the feasibility and efficacy of salvage chemotherapy (carboplatin and etoposide; CE) supported by granulocyte colony-stimulating factor (GCSF) in patients with refractory or relapsed post-transplant lymphoproliferative disorder (PTLD) following solid organ transplantation. Intensified salvage regimens were not feasible for these patients, due to their immunosuppressive conditions and potential organ (especially kidney and bone marrow) malfunctions. Salvage chemotherapy consisted of carboplatin [area under the curve (AUC) 4], on day 1, etoposide (120 mg/m2), on days 1–3 and GCSF (5 μg/kg) starting on day 5. This therapeutic regimen was planned to be repeated every 21 d. Nine patients (seven with refractory, two with relapsed disease) were enrolled. Five patients were heart transplant recipients, three liver transplant recipients and one patient had been a double lung transplant recipient. Five patients achieved a complete remission (CR), with follow-up at 92, 39, 55+, 17 and 9+ months. One patient showed stable disease after two cycles of CE and one patient had progressive disease. Two patients experienced early deaths, after the first and third cycles of chemotherapy respectively. One died of septic complications and one because of a perforated intestine, which had been infiltrated by lymphoma. In respect of the difficulties experienced in treating patients with refractory or relapsed PTLD after solid organ transplantation, the combination of carboplatin and etopoide with GCSF support (filgrastim) proved to be an effective regimen.
Post-transplant lymphoproliferative disorder (PTLD) comprises a spectrum of lymphatic diseases that have accumulated because of the progress in transplantation medicine and the simultaneous development of highly potent immunosuppressive drugs for prophylaxis and treatment of rejection-periods (Penn et al, 1969). The Epstein–Barr virus (EBV) is believed to be the main causative factor in the pathogenesis of PTLD, as it can be detected within these lymphoma cells in up to 90% of patients with PTLD (Young et al, 1989).
The incidence of PTLD is estimated to be 0·5–1%, 1·5–3%, 1·5–3% and 5–10%, after renal, liver, heart and lung/small bowel transplantation respectively. EBV-negative transplant recipients are considered to have an even higher risk (approximately 25%) of developing PTLD.
Histological categories of PTLD are early lesions, polymorphic PTLD, monomorphic PTLD (diffuse large B-cell lymphoma, Burkitt/Burkitt-like lymphoma, plasma cell lymphoma and plasmacytoma-like lesions), T-cell lymphomas and Hodgkin's lymphoma. PTLDs tend to involve extranodal sites, including the allograft and the central nervous system (CNS) (Harris et al, 2001).
Treatment strategy is still controversial. If technically feasible, a total surgical resection can result in complete remission (CR). Immunosuppressive drugs should be reduced, but it is beneficial only in rare cases (Starzl et al, 1984). In patients with an acceptable performance status and EBV association, an antiviral therapy (e.g. with foscarnet or arginine butyrate, in combination with ganciclovir) should be considered and, according to early reports, may result in disease regression (Mentzer et al, 1998; Oertel et al, 1999, 2002). Administration of interferon may result in remission in patients with low tumour burden, but rejection occurs in approximately 30% (Davis, 2001). Further therapeutic options that are under current investigation include treatment with monoclonal antibodies, such as rituximab, which is directed against the CD20 antigen or adoptive T-cell strategies (Milpied et al, 2000; Oertel et al, 2000). A recent paper reported a novel monoclonal antibody directed against interleukin-6, which showed response in polymorphic PTLD (Haddad et al, 2001). Nevertheless, there is still a need for cytotoxic chemotherapy in at least 50% of patients with PTLD. Most centres use adriamycin-based drug combinations, such as cyclophosphamide, hydroxydaunomycin, oncovin, prednisone (CHOP) with granulocyte colony-stimulating factor (GCSF) support (Garrett et al, 1993).
To our knowledge, no reports are available concerning salvage chemotherapy in patients with refractory or relapsed PTLD. In non-organ transplanted patients with high grade B-cell non-Hodgkin's lymphoma (B-NHL) refractory to, or relapsing after initial therapy, different chemotherapeutic combinations may induce a new response (Cortelazzo et al, 2001). However, these responses are rather short-lived and long-term survival is rarely seen. For patients with chemosensitive relapse, salvage therapy followed by high-dose therapy with stem cell support is recommended.
The therapeutic concept for non-organ transplanted patients with refractory or relapsed high grade B-NHL (salvage regimens with or without stem cell support) is not transferable to patients with refractory or relapsed PTLD. In transplant recipients with PTLD, cytotoxic chemotherapy induces an enhanced haematotoxicity, because of the side-effects of immunosuppressive drugs on kidney and bone marrow function. Furthermore, these patients are highly susceptible to infectious, even septic, complications because of an impaired immune system.
Here, we present the results of the first nine patients enrolled in this pilot study, investigating the feasibility and efficacy of carboplatin and etoposide (CE) + GCSF in patients with refractory or relapsed PTLD.
Patients and methods
Between October 1992 and January 2002, all consecutive patients presenting with refractory or relapsed PTLD at Humboldt-Universität Berlin, Universitätsklinikum Charite- Campus Virchow Klinikum received CE and GCSF (filgrastim; Amgen GmbH, Munich, Germany). All patients had previously received CHOP chemotherapy. All patients gave written informed consent.
Diagnosis was based on examination of histological material, obtained by open biopsy or by core needle biopsy. Lesions were classified according to the World Health Organization (WHO) Classification of Tumours (Harris et al, 2001).
The extent of disease was determined by a complete patient history, physical examination, blood tests [white blood cell (WBC) count, biochemical tests, liver tests and lactate dehydrogenase assay], bone marrow biopsy and computed tomography (CT) scans of the chest, abdomen and pelvis. Other studies were performed as clinically indicated.
The CE regimen consisted of carboplatin [area under the curve (AUC) 4 i.v.] on day 1 and etoposide (120 mg/m2 i.v.) on days 1–3 of each cycle. Using the AUC to calculate the carboplatin dose, the renal function of the individual patient is taken into account. The absolute dose (mg) was calculated by the formula of Calvert (Duffull & Robinson, 1997). Treatment cycles were repeated every 21 d. Patient 1 received carboplatin (AUC 4 i.v.) on day 1 of each cycle and etoposide (25 mg/m2 p.o.) on days 1–10. Patient 5 had cytology-proven CNS involvement and was treated additionally with cranial radiation and intrathecal methrotrexate. Patient 7 received an amended treatment dose because of haemodialysis and patient 8 required a dose reduction because of pancytopenia WHO grade 3. Patient 9 received only two cycles of CE. The patient with an abdominal bulk was in partial remission (PR) after three cycles of CHOP, but staging after six cycles showed progressive disease (PD). After two cycles of CE the original bulk was irradiated. CNS prophylaxis was not given routinely. GCSF of 5 μg/kg/d was started on day 5 of every CE cycle until recovery of WBC ≥10 × 109/μl or ≥5 × 109/l on three consecutive days.
Evaluation of response.
Response was assessed using follow-up CT imaging. Restaging was performed following at least two and four cycles. CR was defined in accordance with WHO criteria, as the disappearance of all clinical evidence of active tumour for at least 4 weeks. PR was defined as a minimum 50% reduction in the tumour for at least 4 weeks (no recent metastases, no tumour progression in any location). Stable disease (SD) was defined as no objective change in the lesions. PD was defined as an unequivocal increase in tumour size or the appearance of new lesions.
Patient characteristics are summarized in Table I. Results were analysed on an intention-to-treat basis. Median age was 54 years (range 38–65 years). Five patients were heart recipients (n = 5), three liver recipients (n = 3) and one a double lung recipient (n = 1). The initial diagnosis of PTLD was made between July 1992 and January 2002. Initially, two patients had localized stage II PTLD and seven had disseminated stage IV PTLD. The histology identified diffuse large B-cell lymphomas in seven cases, Mantle cell lymphoma in one case and atypical Burkitt lymphoma in the remaining case. The proliferation index Ki67 was extremely high (see Table I). An EBV association was confirmed in six cases, by immunohistological staining of latent membrane protein 1 and Epstein–Barr nuclear antigen 2 and by in situ hybridization of EBER transcripts.
Table I. Characteristics of patients with refractory or relapsed post-transplant lymphoproliferative disorder.
Pt, patient number; f, female; m, male; TX, organ transplanted; H, heart; DL, double lung; L, liver; R, renal; DLCL, diffuse large B-cell lymphoma; MCL, mantle cell lymphoma; ln, lymphnodes; CNS, central nervous system; EBV, associated with Epstein–Barr virus; LDH, lactate dehydrogenase; nod, nodules.
Both lungs (multiple nod), liver (multiple nod), paraaortal ln
Right lung (multiple nod), supraclavicular ln, stomach, intestine, abdominal bulk
Both lungs (multiple nod)
Liver, lung, cervical ln
Blood, bone marrow, CNS
Stomach, mesenteric ln
Both tonsils, cervical ln, peripheral blood
Bone marrow and renal function at time of diagnoses of PTLD
The median values were: WBC 4·3 × 109/l (range 3·1 × 109–6·8 × 109/l), for haemoglobin 11·2 g/dl (range 8·9–13·6 g/dl) and for platelets 143·6 × 109/l (range 79 × 109–222 × 109/l). Serum creatinine was 213 μmol/l (range 70·7–848·6 μmol/l) and urea at the time of diagnosis of PTLD was 11·9 mmol/l (range 6·2–29·8 mmol/l).
Seven patients with refractory PTLD entered the CE + GCSF trial directly. Two patients with relapsed PTLD had experienced a CR of 11 and 12 months, respectively, after first-line chemotherapy. Three patients received the CE regimen at reduced doses because of moderate organ malfunction, with a median dose of 60% of the intended dose (range 40–60%). Another two patients, one with end-stage cyclosporine nephrotoxicity and haemodialysis, and the other with immunosuppression-associated insufficiency of the bone marrow (WHO grade 3), received the CE regimen at reduced doses (median dose, 30%).
Response to treatment
Treatment and survival of the nine individual patients are summarized in Table II.
Table II. Treatment and results of patients with refractory or relapsed PTLD.
A total of 25 cycles were applied according to the CE-based regimen. Four of nine patients completed all four cycles. In these four patients all manifestations of PTLD disappeared, with a follow-up of 92, 55+, 39 and 17 months. Patient 8 relapsed after having attained CR for 39 months. Patient 9 only received two cycles with all manifestations of PTLD disappearing and was in CR at the last follow-up of 9+ months. One patient achieved SD and one patient showed PD. In two further patients who experienced treatment-related deaths, CR and PR were in evidence at autopsy.
Toxicity of CE
Toxicity in CE and GCSF (filgrastim)-treated patients are summarized in Table III. Grade 4 haematotoxicity occurred in patients after 11 of 25 cycles (44%) and three patients experienced episodes of neutropenic fever (two of which were moderate). The other patient who experienced episodic neutropenic fever died because of infectious complications associated with sepsis and multiorgan failure (MOF).
Table III. Haematotoxicity due to CE cycles graded according to WHO criteria (Miller et al, 1981).
The cut-off date for this analysis was 30 April 2003. Median follow-up time from entry onto the study was 16·1 months and median survival time was 17·0 months. The probability of overall survival at 3 years was 42·0% (Fig 1).
Three patients were alive at the last follow-up (9–55 months). Two patients died 17 and 92 months after having achieved CR without any evidence of PTLD in autopsy. The first patient died because of septic complications of peripheral arterial occlusive disease. The other patient died because of chronic organ failure. Two patients died during therapy (patients 2 and 4). After the first cycle of CE patient 2 died because of intestinal perforation and subsequent peritonitis caused by therapy-induced necrosis through intestinal infiltration by PTLD. After the third cycle of CE, patient 4 died because of neutropenic sepsis with MOF. Autopsies were performed on both patients and revealed PR and no residual manifestation of PTLD respectively. Patients 7 and 8 died because of progressive PTLD.
The aim of this study was to investigate the feasibility and efficacy of CE + GCSF (filgrastim) in patients with refractory or relapsed PTLD after solid organ transplantation. The administration of chemotherapy in organ transplant recipients with PTLD is frequently complicated, because of organ dysfunction (e.g. kidney, liver, or bone marrow insufficiency). Cyclosporin A and tacrolimus, which are primary immunosuppressants after organ transplantation, are responsible for immunosuppressive-induced nephrotoxicity, particularly chronic progressive tubulointerstitial fibrosis, which has led to end-stage renal disease and dialysis (Olyaei et al, 1999). Further side-effects of cyclosporin A and tacrolimus include mild myelotoxcity (Scott & Higenbottam, 1988). The antimetabolites azathioprine and mycophenolate mofetil cause distinct myelotoxicity. In particular, liver or heart transplant recipients suffer from hepatic insufficiencies because of viral reinfections or haemodynamic damage to liver parenchyma. Furthermore, treatment with immunosuppressants, especially glucocorticoids, makes the organ recipient susceptible to infectious complications.
Cisplatin and carboplatin are effective drugs against lymphoma cell lines, but carboplatin does not produce the significant dose-limiting neurotoxicity and nephrotoxicity that is experienced with cisplatin. However, its use is associated with dose-limiting bone marrow suppression, particularly thrombocytopenia. The elimination half-life varies with renal function and is typically between 2 and 6 h in patients with a normal glomerular filtration and may be as long as 18 h in patients with impaired renal function. Relationships between systemic exposure to carboplatin, described as the AUC, and toxicity, have been described. In adults the most common method is that of Calvert, who describes the relationship between dose and AUC (Duffull & Robinson, 1997). Etoposide is a further effective drug against lymphoma cell lines (Drewinko & Barlogie, 1976; Kushner et al, 1987), but which causes an increased haematoxicity in patients with renal impairment and albumin levels <35 g/l. Increased haematological toxicity after etoposide in patients with abnormal organ function is mediated by an increase in free etoposide. Joel et al (1996) identified that a dose reduction is clearly indicated in such patients. The combination of carboplatin and etoposide demonstrated additive effects in lymphoma cell lines. Synergistic and antagonistic effects were not observed (Bishop et al, 1987). CE is a safe combination of cytotoxic drugs that are widely used in conventional approaches for treatment of small cell and advanced non-small cell lung cancers, ovarian cancer and cancers of unknown primary location (Klastersky, 1988; Eiermann et al, 1991; Akutsu et al, 1996).
In our series, CE + GCSF administered in nine patients with refractory or relapsed PTLD after initial CHOP chemotherapy, yielded five CRs. The haematological toxicity observed made the use of GCSF necessary. The higher susceptibility for haematological toxicities and infectious complications in organ recipients with PTLD was explained by the toxic side-effects (especially on kidney and bone marrow functions) of the immunosuppressive drugs and the impaired immune response to infections.
In two patients the intended chemotherapy dose could not be administered because of cyclosporine-associated organ dysfunction (renal end-stage failure, dialysis and pancytopenia, without involvement of bone marrow by PTLD).
The salvage chemotherapy of CE induced long-lasting remissions in our patients, not only in the two patients whose PTLD was chemosensitive to the initial CHOP chemotherapy but, most notably, in four further patients who were refractory to initial CHOP chemotherapy.
In conclusion, this pilot study shows that salvage chemotherapy with CE + GCSF is feasible and effective in patients with refractory or relapsed PTLD after solid organ transplantation. However, it should be added that there is only a very small therapeutic index because of potential organ dysfunctions and infectious complications.
The authors are most grateful to Kristin Zeidler for the collection and management of clinical data, and to the physicians, physician assistants, nurses, and support staff, who participated in the care of patients with PTLD that were included to this study.