Bone Marrow Transplant Service, Department of Clinical Haematology and Medical Oncology, Royal Melbourne Hospital, Parkville, Australia

Correspondence to: A Grigg, Department of Clinical Haematology and Medical Oncology, Royal Melbourne Hospital, Parkville 3050, Victoria, Australia

Relapse of the primary disease remains the predominant cause of death following bone marrow transplantation for high-risk haematological malignancies. Improved supportive care and patient selection have resulted significant improvements in toxicity with standard conditioning regimens. Further dose intensification to reduce the risk of relapse may therefore be feasible. We determined the maximal tolerated dose (MTD) of a 5-day continuous infusion (CI) of etoposide when added to oral busulphan 16 mg/kg and intravenous cyclophosphamide 120 mg/kg (Bu/Cy) as conditioning in 44 autograft and 18 allograft recipients at high risk of relapse. The major toxicity of escalating doses of etoposide was oral and gastro-intestinal mucositis, reflected by a statistically significant increase in the requirement for total parenteral nutrition in both autografts and allograft recipients. Time to neutrophil and platelet recovery, opiate analgesia requirements, and duration of hospitalization were not affected by etoposide dose escalation. The MTD in autograft recipients was 300 mg/m²/day (1500 mg/m² total dose), and 100 mg/m²/day (500 mg/m² total dose) for allograft recipients. Mucositis and hepatotoxicity were more frequent in allograft recipients, suggesting that methotrexate may have contributed to the lower tolerable dose in these patients. As a consequence, further dose escalation may not be possible in heavily pre-treated patients undergoing allogeneic transplantation. Conversely, high dose CI etoposide can be added with relative safety to Bu/Cy in autograft recipients.

Bone Marrow Transplantation (2002) 30, 645-650. doi:10.1038/sj.bmt.1703698

etoposide; VP-16; continuous infusion; bone marrow transplantation

The combination of busulphan and cyclophosphamide (Bu/Cy) is a widely used and generally well tolerated myeloablative conditioning regimen prior to allogeneic^1,2 and autologous stem cell transplantation (SCT).³ Despite intensive conditioning, however, relapse of the primary disease remains the predominant cause of death following transplantation for high-risk haematological malignancies. In an attempt to reduce the incidence of relapse, conditioning regimens have been intensified by the addition to Bu/Cy of agents such as etoposide,^4,5,6,7 cytarabine^8,9 and idarubicin.¹⁰

Etoposide is a topoisomerase-II inhibitor with a steep dose-response against a wide range of malignancies. Previously reported regimens of etoposide in combination with Bu/Cy have identified a wide maximal tolerated dose (MTD) range between 30 and 60 mg/kg when given as a bolus infusion over 4 hours.^5,6,7 Dose-limiting toxicities include mucositis, hepatotoxicity and pneumonitis.^5,6,7,11 The risk of pneumonitis is significantly enhanced by previous lung field irradiation.⁶ The variation in MTD reported in these studies in part reflects variations in type of stem cell source (autologous vs allogeneic), and the patient groups studied (acute leukaemia in first remission¹¹ vs advanced disease⁴). The development of severe mucosal and hepatic toxicity following single dose etoposide added to Bu/Cy led some centres to investigate a divided total dose etoposide of 30 mg/kg¹² or 48 mg/kg¹³ given as a daily bolus dose over 3 days. Despite this divided dose approach, mucosal and hepatic toxicity, including venocclusive disease, have remained dose-limiting at total doses greater than 40 mg/kg in heavily pre-treated patients undergoing autologous BMT.¹⁴

While pharmacokinetic data of etoposide as a single agent delivered as a 34-h continuous infusion (CI) show no difference in the area under the curve (AUC), half-life or toxicity as compared to a 6-h infusion,¹⁵ studies examining the use of etoposide as a CI in combination with cyclophosphamide 200 mg/kg, have shown an MTD of 4.2 g/m², nearly double that achieved with intermittent dose schedules.¹⁶ Moreover, CI of etoposide has been demonstrated to produce an improved anti-tumour effect in a range of malignancies as a result of prolonged drug exposure.¹⁷

These observations provided the rationale for examining CI etoposide as part of myeloablative conditioning. The MTD of etoposide when added as a continuous infusion to Bu/Cy for allograft conditioning has not been previously determined. Our unit has previously observed excellent tolerability of Bu/Cy as conditioning prior to autograft.³ Given this experience and the rationale described, we performed a study of the toxicity of escalating doses of etoposide delivered by CI in addition to Bu/Cy as conditioning prior to allogeneic or autologous BMT for high-risk haematological malignancies.

Patients and methods

Eligibility

Patients with lymphoma or leukaemia who were scheduled for either an autograft or allograft were eligible for entry according to the criteria detailed in Table 1. In general, patients with recurrent disease were transplanted in untreated relapse or chemosensitive relapse; multiply relapsed chemorefractory patients were not eligible. Prior mediastinal radiotherapy was added as an exclusion criterion in autograft recipients following reports of increased pulmonary toxicity in that patient group.⁶ At the time of modification of the exclusion criteria, six patients receiving autologous SCT had been enrolled, none of whom had received mediastinal radiotherapy, or developed pulmonary toxicity. The study was approved by the institutional ethics review committee of the Royal Melbourne Hospital and all patients gave informed written consent.

Study protocol and conditioning regimen

The initial etoposide dose was set at 50 mg/m²/day as a CI for 5 consecutive days from day -8 to day -3 inclusive. In addition, each patient received oral busulphan 1 mg/kg per day on day -7 to day -4 inclusive followed by intravenous (i.v.) cyclophosphamide 60 mg/kg/day on day -3 and day -2. Drug doses were calculated using ideal body weight. On the day of transplant, autograft recipients received either cryopreserved peripheral blood stem cells, marrow or both. All allograft recipients received unmanipulated marrow from HLA-identical sibling donors. GVHD prophylaxis consisted of intravenous cyclosporin (CsA) 3 mg/kg/day and methotrexate (MTX) 15 mg/m² on day 1, and 10 mg/m² on days 3, 6 and 11 with folinic acid rescue (FAR) after each methotrexate dose.

The study protocol was designed to allow for a minimum of three allograft recipients and three autograft recipients to be enrolled at each dose level of etoposide. If no grade 3 or 4 non-haematological toxicity occurred in the initial three patients enrolled at each level within 60 days of BMT, subsequent patients were enrolled at the next highest etoposide dose level. Each etoposide dose escalation was in increments of 50 mg/m²/day. Where fewer than three patients at any dose level had been followed for 60 days, and no grade 3 or 4 toxicities were observed, newly eligible patients were enrolled at the current dose level. If grade 3 or 4 non-haematological toxicity occurred in one of the three initial patients, an additional three patients were enrolled at that same dose. If a further patient developed grade 3 or 4 toxicity, or if more than one of the initial three patients developed grade 3 or 4 toxicity, no further patients were enrolled at that dose level. Three additional patients were then enrolled at the preceding dose level. The MTD was defined as the level at which no more than one of at least six patients experienced grade 3 or 4 non-haematological toxicity.

Supportive care

All patients were nursed in reverse barrier isolation. Prophylactic intravenous antifungal and antiviral therapy was given with fluconazole 200 mg daily and aciclovir 200 mg 8 hourly, respectively. Prophylactic antibacterial therapy was not given. Broad spectrum antibiotic therapy was initiated at the onset of fever. All patients received prophylactic anticonvulsant therapy with clonazepam and uro-epithelial prophylaxis with intravenous hyperhydration and mesna. Filgrastim or other growth factor was not administered routinely. Allograft recipients received intravenous immunoglobulin 400 mg/kg per week from day -1 to day +84. Where the etoposide dose was 200 mg/m²/day or greater, propantheline 30 mg orally every 6 h for six doses was given in an attempt to limit mucositis by reducing secretion of drug in the saliva.¹⁸ Ganciclovir 5 mg/kg three times a week was given from the time of engraftment until day 84, as cytomegalovirus (CMV) prophylaxis to CMV-positive allograft recipients or those receiving grafts from CMV-positive donors. Co-trimoxazole was used from the time of engraftment as prophylaxis against Pneumocystis carinii infection.

Toxicity grading

Toxicity of the conditioning regimen was graded using the scoring system devised by Bearman.¹⁹ Regimen-related toxicity (RRT) was regarded as any organ toxicity occurring within the first 100 days post-BMT. However, the development of bacteremia or fungemia were not regarded as a RRT unless secondary to another organ toxicity such as mucositis or severe colitis. Similarly, GVHD was not regarded as a RRT.

Statistical analysis

Spearman rank correlation was performed for each dose level of etoposide vs days of TPN administration, days of opiate analgesia, days to neutrophil recovery to >0.5 ´ 10⁹/l and days to platelet recovery to >20 ´ 10⁹/l. A P value of <0.05 was considered statistically significant.

Results

A total of 62 patients was enrolled, with 44 undergoing autologous and 18 allogeneic SCT. Patient demographics and diseases treated are shown in Table 2. One allograft patient with AML in first CR was erroneously entered on to the study and received etoposide at the 200 mg/m²/day level.

Autografts

Haematologic toxicity: Forty-four patients received autologous SCT. Peripheral blood stem cells were used alone in 32 patients, unmanipulated marrow in three and both in nine. All recipients achieved rapid neutrophil and platelet engraftment, except for two patients with delayed platelet recovery (Table 3). One patient, who received 50 mg/m²/day of etoposide, had no identifiable cause for prolonged thrombocytopenia, eventually had platelet recovery and remains in remission. The other patient received 250 mg/m²/day, developed persistent thrombocytopenia related to progressive marrow NHL, and died at day 78. Overall there was no identifiable impact of etoposide dose escalation on either platelet (P = 0.6) or neutrophil recovery (P = 0.8).

Grade 2-3 non-haematologic toxicity: The most common side-effect was oral mucositis, with 34 of 44 patients experiencing grade 2 and only one experiencing grade 3 toxicity. The incidence and severity of gastro-intestinal toxicity as measured by the duration of TPN, was directly related to increasing etoposide dose (P = 0.03). Consistent with this association, the median duration of opiate analgesia for mucositis was 8 days following etoposide doses of 250 mg/m² and above, vs 3 days at etoposide doses below 250 mg/m². However, time to discharge was not increased with increasing etoposide dose. Other grade 2-3 toxicities included hepatic (5), bladder (1), cardiac (2), pulmonary (1) dysfunction, with no clear association between these toxicities and etoposide dose.

Grade 4 non-haematologic toxicity: Etoposide doses of 200 mg/m²/day or less did not result in grade 4 toxicity. One patient who received 250 mg/m²/day died from respiratory failure at day 11 secondary to Legionella pneumonia. A further patient who received 350 mg/m²/day, died from intracranial haemorrhage developing 4 days after infusion of cryopreserved stem cells. The infusion was complicated by an acute confusional state thought to represent acute neurological toxicity of dimethlysulfoxide.²⁰

Maximum tolerated dose: Doses of etoposide below 350 mg/m²/day did not result in sufficient toxicity to prevent further dose escalation. Furthermore, as none of the initial four patients enrolled at 350 mg/m²/day developed grade 3-4 toxicity, the etoposide dose was then escalated to 400 mg/m²/day. One of the initial three patients enrolled at this level developed grade 3 gastrointestinal toxicity. Following this, a further three newly eligible patients were enrolled at the 350 mg/m²/day level, and one patient developed grade 4 toxicity (intracranial haemorrhage). Consequently, new patients were enrolled at the 350 mg/m²/day level, and a further four patients were treated at this level. Of these one developed grade 3 central nervous system toxicity (intracerebral bleed) and accrual at the 350 mg/m²/day dose was discontinued. Three additional patients were then enrolled at the 300 mg/m²/day level. No episodes of grade 3 or 4 toxicity were observed in a total of eight patients treated at that level. This determined the MTD as 300 mg/m²/day (1500 mg/m² total) in autograft recipients.

Allografts

Haematologic toxicity: The results for allograft recipients are detailed in Table 4. As expected, neutrophil and platelet recovery were both longer than that seen in autograft patients. There was no statistically significant association for either neutrophil or platelet recovery with escalating etoposide dose (P = 0.15 and P = 0.18 respectively). Failure of both neutrophil and platelet recovery occurred in one patient at each of the 100 mg/m²/day and 150 mg/m²/day treatment levels. These patients died of fungal sepsis and interstitial pneumonitis at day 17 and day 23, respectively. Delayed haematologic recovery was seen in the context of severe GVHD in one further patient who received 100 mg/m²/day. Neutrophil recovery was eventually seen at day 42; however, the patient remained platelet transfusion dependent until his death at 150 days from grade 4 GVHD.

Grade 2-3 non-haematologic toxicity: One patient developed grade 3 mucositis, with most (16/18) experiencing grade 2 mucositis. As with the autograft patients there was a statistically significant trend towards longer duration of TPN (P = 0.004) with escalating etoposide doses. Opiate analgesia use with escalating dose was not statistically significant (P = 0.12). Compared to autograft recipients, allograft recipients had a longer time to discharge with increasing etoposide dose and a longer median duration of opiate analgesia (12 vs 5 days). Additional grade 2-3 toxicity included hepatic (9), bladder (1), renal (4), pulmonary (1), gastro-intestinal (2). No relationship between etoposide dose and hepatic toxicity could be discerned , with grade 2 toxicity seen at etoposide doses of 50 mg/m²/day (two of three patients), 100 mg/m²/day (two of six), 150 mg/m²/day (one of six), 200 mg/m²/day (one of two). Grade 3 hepatic toxicity occurred in one patient at each of the 50 mg/m²/day, 150 mg/m²/day, and 200 mg/m²/day levels.

Grade 4 toxicity: A single treatment-related death was observed at each of the 50 mg/m²/day, 100 mg/m²/day and 150 mg/m²/day etoposide dose levels. Of these, two patients died early from interstitial pneumonitis on days 36 (50 mg/m²/day level) and 17 (150 mg/m²/day level), and one died from fungal sepsis complicating neutropenic colitis at day 23 (100 mg/m²/day level).

Maximal tolerated dose

Allogeneic BMT: Both patients enrolled at the 200 mg/m²/day dose level experienced significant toxicity with grade 3 gastro-intestinal toxicity in one, and grade 3 hepatic, bladder and gastro-intestinal toxicity in the other. Therefore, no further patients were enrolled at this level. At the 150 mg/ m²/day level, two of the six patients enrolled developed significant toxicity: grade 3 hepatic and pulmonary toxicity in one, and in the other, fatal interstitial pneumonitis. The four other patients treated at the 150 mg/m²/day level experienced grade 2 mucositis as the only significant toxicity. At the 100 mg/m²/day level one patient died of fungal sepsis related to neutropenic colitis. The remaining five patients treated at the 100 mg/m²/day dose developed grade 2 mucositis (4), grade 2 hepatic toxicity (1), and grade 2 gastro-intestinal toxicity (1). The maximal tolerated dose was therefore determined to be 100 mg/m²/day (500 mg/m² total) for patients undergoing allogeneic BMT.

Discussion

In this study, we wished to determine the MTD and examine the toxicity of escalating doses of etoposide added to standard-dose Bu/Cy as conditioning prior to either autologous or allogeneic SCT. Given that toxicity has limited the dose of etoposide when delivered as a short infusion on 1 or 3 days, we designed this study to examine if etoposide given as a CI over 5 days would allow more intensive dose escalation.

In autologous SCT recipients the addition of etoposide to Bu/Cy was well tolerated at doses below 250 mg/m²/day with no grade 3 or 4 toxicity observed in 21 patients. At doses at or above 250 mg/m²/day, however, four episodes of grade 3 toxicity and two treatment-related deaths were observed in 23 patients. The development of grade 3 toxicity at the 400 mg/m²/day level and two cases of treatment-related mortality at the 350 mg/m²/day level, indicated a MTD of 300 mg/m²/day (1500 mg/m² total).

In allograft recipients, similar patterns of toxicity were seen, albeit at lower etoposide doses. Grade 2 or above toxicity was seen in all patients, with mucositis most commonly observed. Grade 3 toxicity was seen in half of all patients and three died of treatment-related toxicity at etoposide doses of 50, 100 and 150 mg/m²/day. Grade 3-4 hepatic and pulmonary toxicity at doses of 150-200 mg/m²/day determined an MTD for etoposide of 100 mg/m²/day (500 mg/m² total).

There was no statistically significant effect of etoposide dose on the engraftment of neutrophils or platelets in either the autologous or allogeneic setting. This observation is of particular importance in the light of previous pharmacokinetic data, which have suggested a proportion of patients have detectable etoposide at time of BMT.¹⁶ In five patients included in this study, etoposide plasma concentrations were assessed by high performance liquid chromatography; in no cases was the drug detectable at the time of marrow or stem cell infusion. Due to technical reasons, it was not possible to perform busulphan levels.

Hepatotoxicity has previously been reported as a dose-limiting toxicity of etoposide.¹¹ Despite significant dose escalation, no grade 3-4, and only five (11%) episodes of grade 2 hepatic toxicity were seen in autograft recipients. By contrast, grade 3 hepatic toxicity was seen in 3/18 (16%) allograft recipients. These occurred at various dose levels, suggesting that individual patient variables, such as the amount of prior chemotherapy, were important influences on the development of this complication. The higher incidence of hepatic toxicity in allograft recipients suggests that the use of MTX for GVHD prophylaxis may have also contributed, as has been described by others.²¹

While opiate analgesia requirement was not prolonged with increasing doses of etoposide, a statistically significant dose-related increase in TPN use was observed in both patient groups. This observation reflects the increased severity of mucositis when etoposide doses were escalated. As expected given the concomitant use of MTX, the duration of mucositis was longer in allograft recipients, despite the lower doses of etoposide used in these patients. Antifolate agents, such as MTX, have been shown to increase the number of DNA breaks induced by topoisomerase-II inhibitors in vitro and in vivo²² and therefore this drug combination may have contributed to limiting the MTD in the allograft setting.

The occurrence of fatal interstitial pneumonitis in two allograft recipients contributed to a relatively low MTD of 100 mg/m²/day. The aetiology of interstitial pneumonitis in allograft patients remains problematic. This complication was not observed in autograft recipients despite the use of higher doses of etoposide, and our observations are consistent with the hypothesis that interstitial pneumonitis often represents a GVHD-related phenomenon, rather than a direct drug toxicity per se.²³

By standardising the previously reported MTD of bolus administered etoposide to a 70 kg, 180 cm tall patient (body surface area = 1.9 m²), comparison between dosing schedules may be made. In our experience, the use of CI etoposide has allowed dose escalation in autograft recipients to similar levels to those achieved by a 3-day divided dose (2850 mg vs 2800 mg total dose). In 'standard' allograft recipients, the MTD of 100 mg/m²/day when given as a CI, delivers a total dose of 950 mg. This compares to 2100 mg using a bolus schedule as reported by Kroger et al.¹¹ It is important to note, however, that in the latter report 62% of patients were considered good risk (AML in first complete remission) and the regimen related mortality at 30 mg/kg was 17%, identical to that seen in our higher risk patient group. A higher MTD may therefore be possible in allograft recipients with CI etoposide if used earlier in the disease course.

These findings indicate that the combination of Bu/Cy and CI etoposide may be used with acceptable toxicity in patients undergoing SCT/BMT for poor-risk haematologic malignancies, although doses above 100 mg/m²/day should not be used in heavily pre-treated allograft recipients. In order to achieve further dose escalation, particularly in heavily pretreated patients, real-time monitoring of etoposide levels may be required to minimize the impact of wide inter-patient variability of etoposide pharmakokinetics.^24,25

The major practical conclusion of this study is that high-dose CI etoposide can be added with relative safety to Bu/Cy in autograft recipients. The impact of this regimen on relapse rates remains to be determined. The data presented here allow the design of a prospective trial comparing disease-free survival after autologous SCT with either Bu/Cy or Bu/Cy plus CI etoposide 300 mg/m²/day for 5 days as conditioning.

1 Tutschka PJ. Copelan EA. Klein JP. Bone marrow transplantation for leukemia following a new busulfan and cyclophosphamide regimen. Blood 1987; 70: 1382-1388. MEDLINE

2 Bandini G, Belardinelli A, Rosti G et al. Toxicity of high-dose busulphan and cyclophophamide as conditioning therapy for allogeneic bone marrow transplantation in adults with haematological malignancies. Bone Marrow Transplant 1994; 13: 577-581. MEDLINE

3 Rosenthal MA, Grigg AP, Sheridan WP. High dose busulphan/cyclophosphamide for autologous bone marrow transplantation is associated with minimal non-hemopoietic toxicity. Leuk Lymph 1994; 14: 279-283.

4 Vaughan WP, Dennison JD, Reed EC et al. Improved results of allogeneic bone marrow transplantation for advanced hematologic malignancy using busulfan, cyclophosphamide and etoposide as cytoreductive and immunosuppressive therapy. Bone Marrow Transplant 1991; 8: 489-495. MEDLINE

5 Zander AR, Berger C, Kroger N et al. High dose chemotherapy with busulphan, cyclophosphamide, and etoposide as conditioning regimen for allogeneic bone marrow transplantation for patients with acute myeloid leukemia in first complete remission. Clin Cancer Res 1997; 3: 2671-2675. MEDLINE

6 Crilley P, Topolsky D, Styler MJ et al. Extramedullary toxicity of a conditioning regimen containing busulfan, cyclophosphamide and etoposide in 84 patients undergoing autologous and allogeneic bone marrow transplantation. Bone Marrow Transplant 1995; 15: 361-365. MEDLINE

7 Spitzer TR, Cottler-Fox M, Torrisi J et al. Escalating doses of etoposide with cyclphosphamide and fractionated total body irradiation or busulfan as conditioning for bone marrow transplantation. Bone Marrow Transplant 1989; 4: 559-565. MEDLINE

8 Ratanatharathorn V, Karanes C, Lum LG et al. Allogeneic bone marrow transplantation in high-risk myeloid disorders using busulfan, cytosine arabinoside and cyclophosphamide (BAC). Bone Marrow Transplant 1992; 9: 49-55.

9 Geller RB, Myers S, Devine S et al. Phase I study of busulfan, cyclophosphamide, and timed sequential escalating doses of cytarabine followed by bone marrow transplantation. Bone Marrow Transplant 1992; 9: 41-47. MEDLINE

10 Jerjis S, Roovers E, Muus P et al. Idarubicin to intensify the conditioning of autologous bone marrow transplantation for patients with acute myeloid leukemia in first complete remission. Bone Marrow Transplant 1998; 22: 13-19. Article MEDLINE

11 Kroger N, Zabelina T, Sonnenberg S et al. Dose-dependent effect of etoposide in combination with busulfan plus cyclophosphamide as conditioning for stem cell transplantation in patients with acute myeloid leukemia. Bone Marrow Transplant 2000; 26: 711-716.

12 Kanda Y, Akiyama H, Tanikawa S et al. Etoposide with/without G-CSF with busulfan and cyclophosphamide as conditioning for bone marrow transplantation. The BMT Team. Am J Hematol 1996; 51: 265-268.

13 Jones RJ, Santos GW. New conditioning regimens for high risk marrow transplants. Bone Marrow Transplant 1989; 4 (Suppl. 4): 15-17.

14 Rosenfeld CS, Przepiorka D, Schwinghammer TL et al. Autologous bone marrow transplantation following high-dose busulfan and VP-16 for advanced non-Hodgkin's lymphoma and Hodgkin's disease. Exp Hematol 1991; 19: 317-321.

15 Mross K, Bewermeier P, Reifke J et al. Pharmacokinetics of high dose VP-16: 6-hour infusion versus 34-hour infusion. Bone Marrow Transplant 1994; 13: 423-430.

16 Herzig RH. High-dose etoposide and marrow transplantation. Cancer 1990; 67 (Suppl.): 292-298.

17 Thompson DS, Hainsworth JD, Hande KR et al. Prolonged administration of low-dose, infusional etoposide in patients with etoposide-sensitive neoplasms: a phase I/II study. J Clin Oncol 1993; 11: 1322-1328.

18 Ahmed T, Engelking C, Szalyga J et al. Propantheline prevention of mucositis from etoposide. Bone Marrow Transplant 1993; 12: 131-132.

19 Bearman SI, Appelbaum FR, Back A et al. Regimen-related toxicity and early posttransplant survival in patients undergoing marrow transplantation for lymphoma. J Clin Oncol 1989; 7: 1288-1294. MEDLINE

20 Hoyt R, Szer J, Grigg A. Neurological events associated with the infusion of cryopreserved bone marrow and/or peripheral blood progenitor cells. Bone Marrow Transplant 2000; 25: 1285-1287.

21 Essell JH, Thompson JM, Harman GS et al. Marked increase in veno-occlusive disease of the liver associated with methotrexate use for graft-versus-host disease prophylaxis in patients receiving busulphan/cyclophosphamide. Blood 1992; 79: 2784-2788. MEDLINE

22 Holden SA, Teicher BA, Robinson MF et al. Antifolates can potentiate topoisomerase II inhibitors in vitro and in vivo. Cancer Chemother Pharmacol 1995; 36: 165-171.

23 Workman DL, Clancy J Jr. Phenotypic analysis of pulmonary perivascular mononuclear infiltrates that occur as a direct result of acute lethal graft-versus-host disease describes the onset of interstitial pneumonitis. Am J Pathol 1995; 147: 1350-1360.

24 Porter D, Boddy A, Thomas H et al. Etoposide phosphate infusion with therapeutic drug monitoring in combination with carboplatin dosed by area under the curve: a cancer research campaign phase I/II committee study. Semin Oncol 1996; 23: 34-44.

25 Soni N, Meropol NJ, Pendyala L et al. Phase I and pharmacokinetic study of etoposide phosphate by protracted venous infusion in patients with advanced cancer. J Clin Oncol 1997; 15: 766-772.

Table 1 Inclusion criteria

Table 2 Patient demographics and underlying disease

Table 3 Toxicity in patients following autologous SCT

Table 4 Toxicity in patients following allogeneic BMT