CASE REPORTS

Pharmacokinetics of Ganciclovir in a Patient Undergoing Hemodialysis Suzanne K. Swan, MD, Myrna Y. Munar, PharmD, Mark A. Wigger, MD, and William M. Bennett, MD • The pharmacokinetics and effect of hemodialysis on the clearance of ganciclovir were evaluated in a patient with cytomegalovirus (CMV) retinitis and pneumonitis requiring dialytic support. A dose of 300 mg ganciclovir (5 mg/kg) was administered by intravenous infusion over a 60-minute period. Blood samples were obtained over the next 10 hours and used to assess plasma ganciclovir concentrations. The patient underwent hemodialysis the following day during which paired arterial and venous blood samples were obtained to determine dialyzer clearance of this antiviral agent. High-performance liquid chromatography was used to quantify ganciclovir plasma concentrations. Ganciclovir levels declined in a monoexponential manner following infusion and prior to dialysis. The patient's peak ganciclovir concentration was markedly elevated (20 ,.g/mL) compared with previously reported peak concentrations in patients with normal renal function. Similarly, the elimination half-life (t 1/2 ) was increased (6.3 hours) in this patient compared with values reported in patients with normal renal function. The volume of distribution (0.21 L/kg) and total body clearance prior to hemodialysis (35.5 mL/min) were diminished in this patient. Hemodialysis reduced ganciclovir levels by approximately 62% with an extraction coefficient of 0.29 resulting in a dialyzer clearance of 48.3 mL/min. This supports supplementation of ganciclovir in patients receiving this antiviral agent when they are undergoing hemodialysis. Additionally, close monitoring of ganciclovir concentrations in patients with abnormal renal function is necessary in order to make appropriate dosage adjustments. © 1991 by the National Kidney Foundation, Inc. INDEX WORDS: Ganciclovir; hemodialysis; pharmacokinetics.

C

YTOMEGALOVIRUS (CMV) is a leading cause of infectious complications in renal transplantation recipients. CMV disease contributes significantly to morbidity and mortality in this, and other, immunocompromised patient populations. Ganciclovir (9-[1,3-dihydroxy-2-propoxymethyl] guanine) , an acyclic nucleoside analogue of guanine has been studied extensively as an antiviral agent in CMV infections. Recent studies show ganciclovir's therapeutic efficacy in patients with CMV disease. 1.2 Yet, detailed information regarding its pharmacokinetics, total body and dialysis clearance, and therapeutic versus toxic plasma concentrations in patients with renal dysfunction is limited. We report the pharmacokinetics of an intravenously administered dose of ganciclovir in a renal allograft recipient with CMV retinitis and pneumonitis undergoing hemodialysis. CASE REPORT In 1965, a 28-year-old white woman with end-stage renal disease secondary to chronic pyelonephritis underwent bilateral nephrectomy and splenectomy. She received a haploidentical renal allograft that was intraperitoneally placed and she was maintained on azathioprine (150 mg) and prednisone (10 mg/ d). In April 1989, she noted marked deterioration in vision and mild dyspnea on exertion. Ophthalmologic examination showed bilateral temporal chorioretinal scarring and infiltrate compatible with CMV retinitis. Her chest x-ray and echocardiographic examination were normal.

Over the next 6 months her immunosuppressive therapy was reduced from 150 mg to 50 mg azathioprine per day. Serial chest films showed mild diffuse interstitial infiltrates. Her CMV retinitis progressed to near total blindness, and renal function deteriorated with an elevation of serum creatinine from 150 to 212 ,.mollL (1.7 to 2.4 mg/dL). A 24-hour urine collection showed a creatinine clearance of 0.36 mL/ s (22 mLi min) and bronchoalveolar lavage cultures were positive for CMY. She was admitted to the hospital for intravenous ganciclovir. The patient tolerated ganciclovir therapy without obvious complications. Her vision improved with resolution of CMV retinal infiltrates. Serial chest x-rays remained stable and her dyspnea improved. Further renal dysfunction occurred, presumably due to insufficient immunosuppressive therapy and potential ganciclovir nephrotoxicity. Her serum creatinine level increased to 460 ,.mollL (5.2 mg/dL) and urine creatinine clearance level decreased to 0.18 mLis (11 mL/min). The patient developed signs of volume overload and electrolyte abnormalities, unresponsive to maximal doses of diuretics, and uremic symptoms progressed. Chronic hemodialysis was performed with marked improvement in her overall condition. Ganciclovir therapy was continued and pharmacokinetic studies were performed to evaluate the drug's clearance with dialysis. From the Division of Nephrology and Hypertension, Oregon Health Sciences University, Portland, OR; and the College of PharTfl11CY, Oregon State University, Corvallis, OR. Address reprint requests to Suzanne K. Swan, MD, Division of Nephrology and Hypertension, Oregon Health Sciences University, 3181 SW Sam Jackson Park Rd, Portland, OR 97201. © 1991 by the National Kidney Foundation, Inc. 0272-63861911/701-0012$3.0010

American Journal of Kidney Diseases, Vol XVII, No 1 (January), 1991: pp 69-72

69

70

SWAN ET AL

METHODS

RESULTS

Ganciclovir was administered as a 300 mg (5 mg!kg) intravenous infusion over 60 minutes. Serial venous samples were collected 0.5, 0.75, 2, 3, 4, 6, and 10 hours after the end of the infusion. The patient underwent routine hemodialysis (19 hours after receiving the 300 mg infusion) for 4 hours the following day, during which simultaneous paired arterial and venous blood samples were obtained at hourly intervals. A Lundia IC 3L (Gambro, Newport News, VA) dialyzer was used with an effective surface area of 0.8 m2 (Cuprophan, Newport News, VA) and measuring 10.5 jtm in thickness. A blood flow rate of 250 mLimin was maintained throughout the procedure. All samples were analyzed for ganciclovir concentration using high-performance liquid chromatography.3 Ganciclovir concentrations were expressed in jtg!mL-I jtg!mL corresponds to 4 jtmollL. Ganciclovir serum levels obtained after the infusion and before dialysis declined in a monoexponential manner, although a rapid distribution phase has been reported in other investigations. ,.SData were fit to a one-compartment model using nonlinear least squares regression (RStrip, MicroMath, Inc, Salt Lake City, UT, and PCNONLIN, Lexington, KY). Area under the serum-concentration-time curve (AUC) was determined by the linear trapezoidal rule and extrapolated to infinity (AUCO-... ) .6 Pharmacokinetic parameters were determined as follows : ganciclovir total body clearance (CL) = dose! AUCO-"" volume of distribution (Vd) = CLlk, and elimination half-life (t 1l2) = 0.693!k, where k is the terminal elimination phase rate constant. 6 The volume of distribution was assumed to be constant before and during dialysis. The extraction coefficient (EJ and dialysis clearance (CLt,d) were calculated to measure the effect of hemodialysis on ganciclovir disposition. The following equations were used: Eo = (A - V)!A, CLhd = Qp x E" and Qp = Qb (1 - Hct) , where A and V are the paired arterial and venous levels obtained during dialysis, Qp and Qb are the respective plasma and blood flow rates through the dialyzer, and Hct is the patient's hematocrit level. 7 .•

Figures 1 and 2 illustrate ganciclovir plasma concentration versus time curves before (postinfusion) and during hemodialysis, respectively. A ganciclovir peak level of 20 p.glmL occurred 45 minutes following completion of the infusion. Trough levels immediately preceding and following 4 hours of hemodialysis were 4 p.g/mL and 1.5 p.g/mL, respectively. Ganciclovir pharmacokinetic parameters were summarized as follows: Vd = 12.58 L (0.21 LI kg), CL = 35.52 mLimin (0.59 mLimin/kg), and t1/2 = 6.3 hours. The extraction coefficient for the 4-hour hemodialysis period was 0.29, which resulted in a dialysis clearance of 48 .30 mLimin. Plasma ganciclovir concentrations decreased by 62.5% during dialysis.

E c,

30

2c

o

~ C .,

s

20

s

u

c o

U

s

s 10

s s

0 0

2

4

6

8

10

12

Time (hours)

Fig 1. Plasma concentration versus time profile of ganciclovlr after administration of 5 mg of ganciclovlr per kg over a 1-hour Intravenous infusion. The patient's Ce • was 11 mL/mln.

DISCUSSION

Since its first description in 1964 by Rifkind et al, 9 CMV disease continues to plague renal transplantation recipients. Though numerous clinical trials have described ganciclovir's antiviral therapeutic efficacy, little is known about the drug's clearance and requirements for dosage adjustments in patients undergoing dialysis. Limited data for a small number of hemodialysis patients has been reported, yet variability in pharmacokinetic parameters exists. Compared with patients who have normal renal function, we observed a markedly elevated peak ganciclovir concentration. Vd was similar to the value reported by Lake et al. 7 However, Sommadossi et al reported greater, though highly variable, values for Vd in patients with renal failure (0.41 ± 1.5 Llkg).s Total body clearance in our patient was lower than average clearance of 1.20 ± 0.87 mLiminl kg for patients with significant renal impairment. 5 The large standard deviation represents significant variability in clearance values for this population. Our patient's clearance represented approximately 15 % of that observed in patients with normal renal function. 4 ,5 Previous investigators have evaluated the correlation between creatinine clearance and ganciclovir clearance in eight patients with varying degrees of renal function .5 Ganciclovir clearance was found to exceed creatinine clearance reflecting the additive role of tubular secretion in total body elimination of this antiviral agent. A linear regression equation was derived that estimates ganciclovir clearance (CI g) based on an estimate of

71

GANCICLOVIR PHARMACOKINETICS 4

• Fig 2. Plasma concentration versus time profile of ganciclovir during a 4-hour hemodialysis treatment. The patient underwent hemodialysis 19 hours following a 300 mg intravenous dose of ganciclovir.

creatinine clearance (Ccr), where Cl g = 0 .25 x Ccr) + 8.57 (r = 0.92, no P value was reported). S According to this equation, estimated ganciclovir clearance in our patient was 22 .3 mLimin compared with the actual clearance of 35.52 mLimin. One possible reason for this underestimation is that the equation does not fully take into account renal tubular secretion, which likely plays an important role in ganciclovir clearance. Therefore, the predictive value of this equation needs to be evaluated in a large number of patients before it can be used in the clinical setting. Hemodialysis effectively removes ganciclovir, which might be expected given its low molecular weight (255) and water solubility (3 mg/mL). Sommadossi et al suggest replacement of approximately 50% of the administered drug after dialysis because half of the administered drug was eliminated by a 4-hour dialysis treatment. 5 We agree with these recommendations because we observed a similar decrease in plasma levels in our patient after dialysis . However, an important factor affecting drug clearance during hemodialysis is the extent to which it is protein bound. Questions remain regarding protein-binding kinetics for ganciclovir in patients with normal renal function as well as those with renal failure. The effect of altered protein-binding secondary to other drugs or disease states on ganciclovir disposition is largely unknown. Additionally, therapeutic plasma concentrations await further delineation. Several IDso (dose of antimicrobial agent required to inhibit growth of 50% of specific isolates) values have been reported in the literature for specific strains of CMV

2

3

4

Time (hrs) and range from 0 .8 /Lmol/L (0.20 /Lg/mL) to 11 /LmolfL (2.75 /Lg/mL). 10.11 In a large series by Erice et aI, mean plasma peak concentrations of ganciclovir were lower in patients who improved (15.2/Lmo1/L = 3.8 p.g/mL) compared with those who did not improve (16.6 /LmollL = 4.15 /Lg/ mL) or who showed no progression of disease (21.5 /LmolfL = 5.38 /Lg/mLJ,2 Clearly, further study is needed to evaluate the in vitro sensitivity of CMV isolates to ganciclovir and its application to in vivo conditions. Our findings confirm those of other investigators and illustrate many of the pharmacological al· terations induced by the presence of renal failure and the effects of hemodialysis in a patient receiving ganciclovir. As further experience is obtained with the use of this antiviral agent, greater understanding of appropriate dosage adjustments for patients who have renal failure and require dialytic support will be gained. Presently, close monitoring of plasma concentrations of ganciclovir in patients who have renal dysfunction, particularly those undergoing dialysis , is prudent. ACKNOWLEDGMENT The authors would like to thank Susan J. Kohlhepp, PhD for her assistance with the HPLC methodology.

REFERENCES I . Harbison MA, DeBirolami PC, Jenkins RL, et al: Ganciclovir therapy of severe cytomegalovirus infections in solid organ transplant recipients. Transplantation 46:82-88, 1988 2. Erice A, Jordan MC, Chace BA, et al : Ganciclovir treat-

72 ment of cytomegalovirus disease in transplant recipients and other immunocompromised hosts . JAMA 257: 3082-3087, 1987 3. Sommadossi JP, Bevan R: High performance liquid chromatographic method for the determination of 9-(1 ,3-dihydroxy-2-propoxy-methyl)guanine in human plasma. J Chromatogr 414:429-433, 1987 4. Fletcher C, Sawchuk B, Chinnock B, et al: Human pharmacokinetics of antiviral drug DHPG . Clin Pharmacol Ther 40:281-286, 1986 5. Sommadossi Jp, Bevan R, Ling T, et al: Clinical pharmacokinetics of ganciclovir in patients with normal and impaired renal function. Rev Infect Dis 3:507-514, 1988 (suppl 10) 6. Gibaldi M, Perrier D: Pharmacokinetics (ed 2). New York, NY, Dekker, 1982

SWAN ET AL 7. LakeKD, FletcherCV, LoveKR, etal : Ganciclovirpharmacokinetics during renal impairment. Antimicrob Agents Chemother 32:1899-1900,1988 8. Gibson TP: Problems in designing hemodialysis drug studies . Pharmacotherapy 5:23-29, 1985 9. Rifkind D, Starzl TE, Marchioro TL, et al: Transplantation pneumonia. JAMA 11 :808-812 , 1964 10. Plotkin SA, Drew WL, Felsenslein D, et al: Sensitivity of clinical isolates of human cytomegalovirus 10 9-(l,3-dihydroxy-2-propoxymethyl)guanine. J Infect Dis 152:833-834, 1985 11. Fletcher CV, Balfour HH: Evaluation of ganciclovir for cytomegalovirus disease. DICP 23:5-11 , 1989