Pediatr Nephrol (2002) 17:825–829 DOI 10.1007/s00467-002-0946-7
O R I G I N A L A RT I C L E
Jeffrey M Saland · Patrick J. Leavey Robert O. Bash · Eleonora Hansch Gerald S. Arbus · Raymond Quigley
Effective removal of methotrexate by high-flux hemodialysis Received: 27 August 2001 / Revised: 27 March 2002 / Accepted: 4 April 2002 / Published online: 9 August 2002 © IPNA 2002
Abstract The purpose of the present study was to examine the clearance of methotrexate (MTX) by high-flux hemodialysis (HD) in pediatric oncology patients. We present three patients who experienced nephrotoxicity and prolonged exposure to toxic MTX concentrations following high-dose infusions during treatment for osteogenic sarcomas. Each patient was successfully treated with highflux HD, followed by carboxypeptidase G2 (CPDG2) in two cases. Minimal systemic toxicity occurred. We review the literature and discuss guidelines for early and aggressive treatment for this complication of high-dose MTX therapy. Clinically important removal of MTX depends upon prompt initiation of HD after detection of nephrotoxicity and delayed clearance of MTX. Therapy is indicated in cases where compassionate use of CPDG2 may not be available, or while awaiting its delivery. Keywords Antineoplastic-antimetabolites · Toxicology · Hemodialysis · Bone neoplasms
Introduction Methotrexate (MTX) is an antimetabolite commonly used in the treatment of several malignant and non-malignant diseases. The utility of the agent is based upon its inhibition of dihydrofolate reductase (DHFR) and the ability to achieve cellular rescue with leucovorin (folinic J.M. Saland Department of Pediatrics, The Mount Sinai School of Medicine, One Gustave L. Levy Place Box 1664, NY 10029–6574, USA P.J. Leavey · R.O. Bash · R. Quigley (✉) Department of Pediatrics, The University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390–9063, USA e-mail:
[email protected] Tel.: +1-214-6483438, Fax: +1-214-6482034 E. Hansch · G.S. Arbus Department of Pediatrics, University of Toronto, The Hospital for Sick Children, Room 1436D, 555 University Avenue, Toronto, Ontario, Canada, M5G 1X8
acid) [1, 2, 3]. High-dose MTX (12 g/m2 per dose) is used in many treatment regimens for osteogenic sarcoma. Because MTX is excreted predominantly by the kidney, nephrotoxicity occurring shortly after MTX infusion leads to persistent elevation of the MTX concentration and may result in life-threatening gastrointestinal, hematological, hepatic, and dermatological complications [1, 2, 3, 4]. The major activity of MTX occurs intracellularly, where serial glutamation of the agent prolongs the intracellular half-life and further enhances its antimetabolic effect [1]. Following intravenous dosages greater than 5 g/m2, the majority (from 50% [3] to more than 90% [2]) of MTX is excreted unchanged in the urine. In addition to glomerular filtration, renal processing includes both tubular secretion and reabsorption by organic anion transporters [2]. Hepatic metabolism leads to formation of 7-hydroxymethotrexate (7-OH-MTX), which retains only 1% of the inhibitory effect upon DHFR as compared with MTX [1]. Because its renal clearance and solubility is less than MTX, the serum concentration of 7-OH-MTX may equal or exceed that of MTX at later times after administration [2] and it may contribute to the renal toxicity sometimes experienced after administration of high-dose MTX [1, 2, 3]. Given its potential toxicity, methods of MTX removal are of great interest. Carboxypeptidase G2 (CPDG2) enzymatically degrades MTX and may be requested from the National Cancer Institute of the National Institutes of Health for use in patients who experience prolonged MTX concentrations in a toxic range. However, the delay introduced before acquisition and administration of CPDG2 may be significant, up to 24–36 h depending upon delivery time and other logistical factors. This report details effective MTX clearance via highflux hemodialysis (HD) in three patients with acute nephrotoxicity and very elevated MTX concentrations that occurred during treatment for osteogenic sarcoma. Two of the patients were subsequently treated with CPDG2. The recent literature is reviewed and guidelines for use of HD are discussed.
826 Fig. 1 Effect of hemodialysis (HD) on serum methotrexate (MTX) concentrations in patient 1. The middle section represents concentrations obtained during dialysis. Post-dialyzer concentrations were lower than pre-dialyzer concentrations
Patients and methods We present three patients who experienced nephrotoxicity and prolonged exposure to toxic MTX concentrations following highdose infusions during treatment for osteogenic sarcomas. A toxic concentration of MTX is defined as greater than 100 µM 24 h after infusion (“delayed clearance”) of the MTX dose. Acute renal failure (ARF) is defined as an elevation of the serum creatinine concentration more than 0.5 mg/dl over baseline. Three patients are included in this retrospective case series and represent all of the cases of delayed clearance of high-dose MTX associated with ARF managed by the authors. MTX (12 g/m2) was administered as a single intravenous (IV) infusion over 4 h with vigorous IV hydration, leucovorin rescue, and urine alkalinization until MTX concentrations were less than 0.2 µM. Upon detection of delayed clearance, routine leucovorin doses (10 mg/m2 every 6 h) were first increased to 100 mg/m2 every 3 h then changed to 30 mg/m2 every hour during HD. Leucovorin was held 2 h before and after CPDG2 administration. MTX therapy was omitted from the subsequent chemotherapy received by each patient. Serum MTX concentrations were analyzed by automated fluorescence polarization immunoassay (FPI) prior to the administration of CPDG2 [5]. After CPDG2 administration, in addition to the FPI, high-performance liquid chromatography (HPLC) at The National Cancer Institute was used to measure MTX concentrations because a product of the CPDG2-mediated degradation cross-reacts with MTX in the FPI. This product is most likely 2,4-diamino-N10-methlypteroic acid (DAMPA). However, the results of the HPLC analysis were not known for several days, so the patients were managed with the FPI results since these were obtained within hours. Fresenius F-80 dialysis membranes were used in patients 1 and 2 and a Fresenius F-40 was used in patient 3 (Fresenius USA, Lexington, Mass., USA). Double-lumen, 12-Fr, subclavian catheters were percutaneously placed. Circuit parameters included a blood flow rate of 370–400 ml/min (patients 1 and 2) and 250 ml/min (patient 3) and a dialysate potassium concentration of 4 mEq/dl with an otherwise standard solution. Ultrafiltration rates were determined by the clinical need for fluid removal. Dialysate flow rates were 600 ml/min. Most patients required infusion of phosphate to prevent severe hypophosphatemia due to the aggressive dialysis treatment. Half-life was defined as the time required for the MTX concentration to be reduced by 50% and was calculated as log(2) divided by the slope of the linear regression of log[MTX] versus time in hours. The HD extraction ratio was calculated as: Extrac-
tion ratio=(CPRE–CPOST)/(CPRE) where CPRE and CPOST were the measured pre- and post-dialyzer MTX concentrations [6]. For each interval of time between MTX measurements, HD MTX clearance was calculated according to: in which QB was the blood flow through the dialysis circuit [6]. Patient 1 A 13-year-old male (body surface area 1.7 m2) received his third course of MTX per protocol. His clinical status was satisfactory and he had tolerated the first two infusions without complications, with a cumulative dosage of cisplatin of 360 mg/m2 prior to treatment. The serum MTX concentration 27 h after the third infusion was 190.2 µM (toxic >5–10 µM) [1]. At that time, the serum creatinine concentration had increased from 0.8 mg/dl pre treatment to 1.5 mg/dl. Urine output remained appropriately high with brisk IV hydration and urinary pH remained between 7 and 8. The first of three sessions of HD was started 40 h after the start of the MTX infusion. Serum MTX concentrations and post-dialysis membrane MTX concentrations for the first session of HD are shown in Fig. 1. The extraction ratio ranged from 0.42 to 0.83. MTX dialysis clearances during the three sessions of HD were 198, 212, and 258 ml/min per 1.73 m2 respectively; the effect of dialysis on the serum MTX half-life is shown in Table 1. The first session of HD was complicated by hypokalemia (2.7 mg/dl) and severe hypophosphatemia (less than 0.5 mg/dl). During subsequent sessions the concentration of potassium in the dialysate was increased to 4 mEq/dl and IV sodium phosphate infusions were administered in anticipation of brisk phosphorus losses. No complications occurred during the second or third sessions of HD. The patient never developed oliguria and the serum creatinine concentration stabilized at 0.7 mg/dl after HD. The patient did not experience mucositis, severe bone marrow suppression, or neurological disturbances, but a brief elevation in the liver transaminases occurred. Patient 2 A 14-year-old female (body surface area 1.6 m2) with osteogenic sarcoma received her sixth treatment of MTX per protocol. She
827 Fig. 2 Effect of HD on serum MTX concentrations in patient 2 prior to administration of carboxypeptidase G2 (CPDG2). The second segment represents concentrations obtained during dialysis. Post-dialyzer concentrations were lower than pre-dialyzer concentrations. All concentrations in this figure were measured by fluorescence polarization immunoassay (FPI)
Table 1 Plasma half-life of methotrexate (MTX) before and during hemodialysis (HD) Patient number and dialysis session
Time after MTX administration (hours)
MTX half-life before HD (h)
MTX half-life during HD (h)
Patient 1: session 1 Patient 1: session 2 Patient 1: session 3 Patient 2: session 1
40 87 113 38
5.3 25.3 30.8 Not applicable: MTX concentration not declining 13.1
2.9 13.4 16.9 2.3
Patient 3: session 1
37.5
had tolerated previous courses without complication and her clinical status was satisfactory; her cumulative cisplatin dosage was 480 mg/m2. Shortly after the MTX infusion she developed severe nausea, vomiting, and oliguria despite administration of large volumes of IV fluid. At 28 h after the start of the MTX infusion, the serum creatinine had risen from a baseline of 0.7 to 4.6 mg/dl and the serum MTX concentration was 1,323 µM. The concentration remained unchanged over the next few hours as CPDG2 was requested; high-flux HD was initiated 38 h after the start of the MTX infusion. Serial serum MTX concentrations and post-dialyzer MTX concentrations during the initial HD session are shown in Fig. 2; the serum MTX half-life during this session was 2.3 h (Table 1) and after 7 h of HD the serum MTX concentration was 146 µM. CPDG2 arrived and was administered 1 h after dialysis was stopped. Within 15 min of CPDG2 administration, the serum MTX concentration decreased from 164 to 2.8 µM (measured by HPLC) and stabilized at concentrations between 3.0 and 5.2 µM (Fig. 3). Because MTX concentrations measured by HPLC were unavailable for several days and the MTX concentration measured by FPI remained considerably elevated, it was not immediately possible to confirm the effect of CPDG2. Therefore, HD was restarted approximately 36 h after CPDG2 administration. Interestingly, the product of the MTX-CPDG2 reaction that was cross-reactive with the MTX FPI, probably DAMPA or a further metabolite of DAMPA, was rapidly cleared by HD, as shown in Fig. 3. The MTX FPI was subsequently used to define the upper limit of the true MTX concentration and HD was intermittently continued over a total of 18 days until the MTX concentration (by FPI) was less than 0.2 µM. The patient recovered from oliguric renal failure over approximately 3 weeks and her glomerular filtration rate measured by
3.4
125I-sodium iothalamate clearance 1 month after her last dialysis session was 119 ml/min per 1.73 m2. The serum concentrations of hepatic transaminases were transiently elevated, but no bone marrow suppression or mucositis occurred.
Patient 3 A 13-year-old boy (body surface area 1.2 m2), with a previous history of left buttock rhabdomyosarcoma at age 2.5 years, received his third course of MTX per protocol. The rhabdomyosarcoma had been treated by resection and chemotherapy, including vincristine, Adriamycin, cyclophosphamide, and ifosfamide. Twelve days before the MTX infusion he had been admitted to the hospital to receive IV fluid and nutritional support for diarrheal dehydration. After recovery (serum creatinine concentration 0.4 mg/dl), he received MTX. His cumulative cisplatin dosage was 140 mg/m2. The serum MTX concentration 24 h post infusion was 600 µM and the serum creatinine was 2.4 mg/dl. Urine output remained appropriately high, with IV hydration increased to 2.5 times usual infusion rates. In addition to leucovorin rescue, IV furosemide and nasogastric charcoal was administered. By 33 h post infusion, the serum MTX concentration was 370 µM. High-flux HD was started 37.5 h after the MTX infusion, but was discontinued after 5.4 h because the filter clotted. The MTX concentration after HD had declined to 90 µM. Thereafter, MTX concentrations continued to slowly decrease until CPDG2 was administered 61 h after the MTX infusion. As in patients 1 and 2, renal function normalized, with a stable serum creatinine of 0.6 mg/dl. There was no mucositis, neurological dysfunction, or prolonged bone marrow suppression.
828 Fig. 3 Effect of HD on serum MTX concentrations in patient 2 following administration of CPDG2. Degradation products of MTX that are falsely measured by FPI were rapidly cleared by HD. Subsequent MTX concentrations measured by FPI approximated those measured by HPLC
Discussion Osteogenic sarcoma is an uncommon tumor of adolescence with an annual incidence of 4.8 per million children less than 20 years of age or approximately 400 new cases in the United States per year [7]. In a 1975 series, of 41 patients who underwent 542 treatments with highdose MTX treatment for this malignancy, Jaffe and Traggis [4] reported an incidence of nephrotoxicity of 1.1% per treatment and 14.6% per patient, with fatal results in two of six cases. Modern protocols and institution of therapeutic drug monitoring have decreased the likelihood of these events, but in such cases, the management of delayed MTX clearance is of great importance. Previous reports have documented MTX clearance with various levels of success using HD [8, 9, 10, 11], HD with charcoal hemoperfusion [12, 13, 14], HD with serum exchange [15], HD with plasma perfusion [16], and hemofiltration with plasma perfusion over an anion exchange resin [17]. In our experience, HD was superior to HD with charcoal hemoperfusion [18]. In a porcine model, hemoperfusion over an exchange resin slightly outperformed HD via a Fresenius F6 (not high-flux) membrane [19]. Wall et al. [11] prospectively administered MTX in patients with chronic dialysis-dependent renal insufficiency and achieved adequate clearance using high-flux HD. Relling et al. [14] found that leucovorin concentrations were well maintained during dialysis with repeated dosages. Although the heterogeneous methods and outcomes of these reports make comparisons difficult, it is evident that minimizing any delay between MTX infusion and treatment is fundamental. In the current patients, clinically important, rapid clearance of MTX by HD was possible. Serial HD sessions were progressively limited by post-dialysis “rebound” of the serum MTX concentration, as noted in prior reports
[12, 15, 16]. This rebound supports the concept of redistribution of MTX from body pools inaccessible to plasmabased clearance methods such as HD. Interestingly, there are few data concerning MTX efflux from normal cells. However, mono and di-glutamated MTX is lost from circulating human erythrocytes [20], and one type of MTX resistance in a human tumor cell line results from upregulation of cellular efflux [21]. Abnormal fluid compartments such as edema, effusion, or ascites also delay excretion due to sequestration in the avascular fluid [2], but none of the patients described in this report had such problems. In these patients, therefore, the MTX reservoir appeared to be localized to the interstitial or intracellular compartment, from which serum MTX concentrations were actively or passively maintained between dialysis sessions. Prompt initiation of HD after detection of delayed clearance could help prevent saturation of these compartments and to maximize MTX removal. CPDG2 is an enzyme produced by recombinant technology that rapidly reduces high serum concentrations of MTX in humans [22, 23, 24]. The enzyme cleaves the C-terminal glutamate from MTX yielding DAMPA, an inactive compound [22, 23, 24]. Although not confirmed, DAMPA or a further metabolite of DAMPA, is likely the non-MTX compound “falsely” detected by the MTX FPI measurements and cleared rapidly by HD. Effective in minutes, as in this report, this enzyme appears to be the best treatment for delayed MTX clearance. Like HD, however, CPDG2 activity is limited to the plasma compartment, since it crosses cell membranes and the blood brain barrier poorly; thus, prevention of severe toxicity continues to depend upon cellular rescue with leucovorin. Rebound of MTX levels within 48 h of administration of CPDG2 has also been reported [23]. The National Cancer Institute of the National Institutes of Health (Bethesda, Md., USA) has supplied CPDG2 for
829
compassionate use. Even with expedited acquisition, however, most patients will likely experience a clinically significant period of exposure to high MTX concentrations while awaiting treatment. We propose guideline therapy for patients who experience unexpected delayed clearance of MTX with nephrotoxicity. MTX concentrations clearly in the toxic range should be detected by at least 24–30 h after the infusion, and CPDG2 should be requested. Residual renal function and urinary alkalinization should be optimized with vigorous IV hydration with bicarbonate, and high-dose leucovorin (30 mg/m2) should be administered every hour while the patient undergoes HD. A high-flux membrane such as the F-80 is essential, and vascular access should allow high blood flow rates, ideally 300–400 ml/min. Ultrafiltration may be used to carefully correct fluid overload while preserving adequate renal perfusion. The dialysate potassium concentration should be 3–4 mEq/dl and IV phosphorus should be administered during each session to prevent severe depletion of these electrolytes. Initial dialysis sessions should be prolonged and repeated until the serum MTX concentration is less than 0.1 µM or until CPDG2 is administered. After use of CPDG2, intermittent sessions should be continued until it is possible to accurately confirm a non-toxic pre-dialysis MTX concentration. High-flux HD effectively clears MTX and is more rapidly available than CPDG2. In conjunction with leucovorin therapy, optimization of residual renal function, and request for CPDG2, HD should be initiated without delay in order to prevent potentially fatal systemic complications in patients who experience nephrotoxicity in conjunction with an MTX concentration in the toxic range. Following CPDG2 administration, HD should be continued until the MTX concentration is confirmed to be non-toxic. Acknowledgements We thank Kathleen Peng, RN, Elizabeth Piva, RN, and the remainder of the nursing staff of the dialysis units at Children's Medical Center of Dallas and The Hospital For Sick Children in Toronto. This work was presented at the American Society of Pediatric Hematology/Oncology (ASPH/O) meeting in Minneapolis, Minn., USA, 14–16 September 2000.
References 1. Bertino JR, Kamen B, Romanini A (1997) Folate antagonists. In: Holland JF, Frei E, Bast RC, Kufe DW, Morton DC, Weichselbaum RR (eds) Cancer medicine. Williams and Wilkins, Baltimore, pp 907–922 2. Crom WR (1998) Methotrexate. In: Grochow LB, Ames MM (eds) A clinician's guide to chemotherapy pharmacokinetics and pharmacodynamics. Williams and Wilkins, Baltimore, pp 311–330 3. Kamen B (1997) Folate and antifolate pharmacology. Semin Oncol 5 [Suppl 18]:S30–S39 4. Jaffe N, Traggis D (1975) Toxicity of high-dose methotrexate (NSC-740) and citrovorum factor (NSC-3590) in osteogenic sarcoma. Cancer Chemother Rep 6:31–36 5. TDx/TDxFLx Methorexate II Product Insert 7A12, 66–5230/R2 (1996) USA, Abbot Laboratories Diagnostics Division 6. Garella S, Lorch JA (1994) Hemodialysis and hemoperfusion for poisoning. In: Henrich WL (ed) Principles and practice of dialysis. Williams and Wilkins, Baltimore, pp 375–392
7. Gurney JG, Swenson AR, Bulterys M (1999) Malignant bone tumors. In: Ries LAG, Smith MA, Gurney JG, Linet M, Tamra T, Young JL, Bunin GR (eds) Cancer incidence and survival among children and adolescents: United States SEER Program 1975–1995, National Cancer Institute, SEER Program. NIH Publishing no. 99–4649. Bethesda, Md. pp 99–110 8. Collett P, Raghavan D, Salasoo S, Tattersall M, Tiller D (1977) Successful treatment of methotrexate-induced renal failure by haemodialysis (letter). Aust N Z J Med 7:441 9. Hande KR, Balow JE, Drake JC, Rosenberg SA, Chabner BA (1987) Methotrexate and hemodialysis (letter). Ann Intern Med 495–496 10. Thomson AH, Daly M, Knepil J, Harden P, Symonds P (1996) Methotrexate removal during haemodialysis in a patient with advanced laryngeal carcinoma. Cancer Chemother Pharmacol 38:566–570 11. Wall SM, Johansen MJ, Molony DA, DuBose TD, Jr, Jaffe N, Madden T (1996) Effective clearance of methotrexate using high-flux hemodialysis membranes. Am J Kidney Dis 28:846– 854 12. Grimes DJ, Bowles MR, Buttsworth JA, Thomson DB, Ravenscroft PJ, Nixon PF, Whiting RF, Pond SM (1990) Survival after unexpected high serum methotrexate concentration in a patient with osteogenic sarcoma. Drug Saf 5:447–454 13. Molina R, Fabian C, Cowley B, Jr (1987) Use of charcoal hemoperfusion with sequential hemodialysis to reduce serum methotrexate levels in a patient with acute renal insufficiency. Am J Med 82:350–352 14. Relling MV, Stapleton FB, Ochs J, Jones DP, Meyer W, Wainer IW, Crom WR, McKay CP, Evans WE (1988) Removal of methotrexate, leucovorin, and their metabolites by combined hemodialysis and hemoperfusion. Cancer 62:884–888 15. Thierry FX, Vernier I, Dueymes JM, Roche H, Canal P, Meeus F, Pourrat JP, Conte JJ (1989) Acute renal failure after high-dose methotrexate therapy. Role of hemodialysis and plasma exchange in methotrexate removal. Nephron 51:416–417 16. Kawabata K, Makino H, Nagake Y, Tokioka H, Matsumi M, Morita Y, Ota K, Shikata K, Ota Z (1995) A case of methotrexate-induced acute renal failure successfully treated with plasma perfusion and sequential hemodialysis. Nephron 71:233–234 17. Greil J, Wyss PA, Ludwig K, Bonakdar S, Scharf J, Beck JD, Ruder H (1997) Continuous plasma resin perfusion for detoxification of methotrexate. Eur J Pediatr 156:533–536 18. Leavey PJ, Nelson S, Sanders S, Quigley R (1999) Clearance of methotrexate using high flux hemodialysis filters (abstract). J Am Soc Pediatr Hematol Oncol 21:320 19. Giardino R, Fini M, Giavaresi G, Spighi M, Faenza S, Florio ML, Orlandi M, Stefoni S (1995) Comparison of hemodialysis versus hemoperfusion in the clearance of high-dose methotrexate in pigs. Artif Organs 19:362–365 20. Schroder H, Fogh K (1988) Methotrexate and its polyglutamate derivatives in erythrocytes during and after weekly lowdose oral methotrexate therapy of children with acute lymphoblastic leukemia. Cancer Chemother Pharmacol 21:145–149 21. Hooijberg JH, Broxterman HJ, Kool M, Assaraf YG, Peters GJ, Noordhuis P, Scheper RJ, Borst P, Pinedo HM, Jansen G (1999) Antifolate resistance mediated by the multidrug resistance proteins MRP1 and MRP2. Cancer Res 59:2532–2535 22. DeAngelis LM, Tong WP, Lin S, Fleisher M, Bertino JR (1996) Carboxypeptidase G2 rescue after high-dose methotrexate. J Clin Oncol 14:2145–2149 23. Widemann BC, Balis FM, Murphy RF, Sorensen JM, Montello MJ, O'Brien M, Adamson PC (1997) Carboxypeptidase-G2, thymidine, and leucovorin rescue in cancer patients with methotrexate-induced renal dysfunction. J Clin Oncol 15:2125–2134 24. Widemann BC, Hetherington ML, Murphy RF, Balis FM, Adamson PC (1995) Carboxypeptidase-G2 rescue in a patient with high dose methotrexate-induced nephrotoxicity. Cancer 76:521–526
