Differential Effect of IFN -2B on the Cytochrome P450 Enzyme System

2480 Vol. 8, 2480–2487, August 2002 Clinical Cancer Research

Advances in Brief Differential Effect of IFN␣-2b on the Cytochrome P450 Enzyme System: A Potential Basis of IFN Toxicity and Its Modulation by Other Drugs1

Mohammed Islam, Reginald F. Frye, Conclusion: These data suggest that strategies to mini- Thomas J. Richards, Ibraham Sbeitan, mize the impairment of CYP enzymes could alter the toxic- Sandra S. Donnelly, Paul Glue, ity profile of HDI and augment its therapeutic utility, and 2 that recognition of these potential interactions is important Sanjiv S. Agarwala, and John M. Kirkwood in the therapeutic application of IFNs. Melanoma Center, University of Pittsburgh Cancer Institute [M. I., T. J. R., I. S., S. S. D., S. S. A., J. M. K.], Department of Medicine, School of Medicine [J. M. K.], and Department of Pharmaceutical Introduction Sciences, School of Pharmacy and Center for Clinical Pharmacology IFNs are cellular proteins of 143–187 amino acids pro- [R. F. F.], University of Pittsburgh, Pittsburgh, Pennsylvania 15213, duced by stimulated immune and nonimmune cells that exhibit and Schering Plough Research Institute, Kennilworth, New Jersey antiviral, differentiating, antiproliferative, and immunomodula- 07033 [P. G.] tory functions. IFN␣-2b exhibits variable therapeutic responses when used in combination with various chemotherapies and Abstract cytokines in metastatic melanoma (1–3). However, no therapy Purpose: High-dose IFN␣-2b therapy (HDI) is the has yet demonstrated unequivocal effects on survival in patients standard of adjuvant therapy for patients with high-risk with metastatic disease. In the adjuvant setting, survival and melanoma, but toxicities of this regimen have limited its relapse-free interval were significantly prolonged in the pivotal application. IFNs affect cytochrome P450 (CYP) enzymes, E1684 trial of the Eastern Cooperative Oncology Group, and in which metabolize many endogenous (e.g., steroids, fatty ac- the most recent intergroup trial E1694 (4, 5). Despite the sig- 3 ids) and exogenous (e.g., drugs) substrates. No systematic nificant therapeutic gain associated with adjuvant HDI therapy, studies have been performed to evaluate the effect of HDI on toxicity and adverse effects have impeded adoption of this CYP enzymes. A significant inhibitory effect of HDI on CYP therapy by physicians and patients. Approximately 78% of enzymes would increase the potential for adverse drug re- patients receiving IFN therapy in E1684 experienced grade 3 actions and altered homeostasis through effects on hormone toxicity, and 24% discontinued therapy because of this toxicity metabolism. (6, 7). The management or prevention of toxicities associated Methods: To evaluate the potential effect of HDI on with HDI therapy is one of the greatest challenges in broadening CYP enzymes, 17 patients with high-risk melanoma were and improving the efficacy of this regimen. treated with HDI, and CYP enzyme activity was measured The CYP enzymes are a superfamily of heme-containing by administration of selectively metabolized probe drugs enzymes distributed widely throughout the body that are in- -over time (days ؊6, ؉1, ؉26, and ؉52 of HDI). Probe drugs volved in the synthesis and metabolism of endogenous sub and/or metabolites were quantified and used to derive in- strates including steroid hormones, fatty acids, and lipids, as dexes of enzyme activity. well as the metabolism of exogenous substrates such as drugs Results: The results indicate that HDI differentially and environmental chemicals. IFNs have been shown to de- impairs CYP-mediated metabolism, having no effect on crease the expression and activity of CYP enzymes in animal some enzymes (CYP2E1) and substantial effects on others models (8–11). Human data are less extensive and primarily (CYP1A2; median 60% decrease). A significant association limited to patients with hepatitis being treated with compara- was found between the magnitude of CYP inhibition and the tively low-dose IFNs (12–15), but the available data indicate a occurrence of side effects including fever and neurological detrimental effect on drug metabolism. However, the selectivity toxicity, which may form a novel basis of the underlying and magnitude of the effect on individual drug metabolizing pathophysiology of some IFN␣-2b-induced toxicity. enzymes is largely unknown. In addition, there are no data on the effect of HDI used to treat patients with melanoma on drug-metabolizing enzymes. A deleterious effect on the CYP enzyme system would have important consequences because of Received 11/16/01; revised 3/23/02; accepted 4/2/02. the increased potential for drug-induced adverse effects related The costs of publication of this article were defrayed in part by the to concomitant drug treatment (i.e., drug-cytokine interactions) payment of page charges. This article must therefore be hereby marked and also the potential for altering homeostasis through effects on advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported in part by NIH/NCRR/GCRC Grant #5M01 RR00056. 2 To whom requests for reprints should be addressed, at Department of Medicine, Division of Hematology/Oncology, University of Pittsburgh 3 The abbreviations used are: HDI, high-dose IFN␣-2b; AUC, area Medical Center, 200 Lothrop Street, Montefiore Hospital N-755, Pitts- under the plasma concentration-time curve; Cmax, maximum observed burgh, PA 15216. Phone: (412) 648-6571; Fax: (412) 648-6599; E-mail: serum concentration; CYP, cytochrome P450; IL, interleukin; Tmax, [email protected]. time to achieve maximum serum concentration. Downloaded from clincancerres.aacrjournals.org on October 2, 2021. © 2002 American Association for Cancer Research. Clinical Cancer Research 2481

Table 1 Schema of study designed to evaluate the differential effect of HDI therapy on the CYP enzyme system

Table 2 Five probe drugs regimen or “Pittsburgh mixture” and their doses, corresponding CYP, and activity measure Probe drug (dose) Enzyme Specimen Trait measure Reference Caffeine (100 mg) CYP1A2 Plasma, 8 hours Paraxanthine/caffeine 16 Mephenytoin (100 mg) CYP2C19 Urine, 0–8 hours 4ЈOH-mephenytoin recovery 47 Debrisoquine (10 mg) CYP2D6 Urine, 0–8 hours DBRRa 18 Chlorzoxazone (250 mg) CYP2E1 Plasma, 4 hours 6-OH-chlorzoxazone/chlorzoxazone 17 Dapsone (100 mg) CYP2C8/9b Urine, 0–8 hours DPRRc 22 NATd Plasma, 8 hours MAD/DDSe a Debrisoquine recovery ratio, calculated as the urinary recovery of 4-Hydroxydebrisoquine divided by the sum of 4-Hydroxydebrisoquine and debrisoquine. b May also involve CYP3A4/5 and CYP2E1. c Dapsone recovery ratio calculated as the urinary recovery of dapsone hydroxylamine divided by the sum of dapsone hydroxylamine and dapsone. d NAT, N-acetyltransferase. e MAD, monoacetyldapsone; DDS, dapsone.

endogenous substrates such as hormones. Thus, this study was of s.c. IFN␣-2b) as shown in Table 1. CYP enzyme activities conducted to examine the effects of acute and chronic high-dose were estimated in vivo using the “Pittsburgh mixture” approach, IFN␣-2b monotherapy on several important drug-metabolizing which involves the simultaneous oral administration of five CYP enzymes in vivo using known enzyme-selective probe drugs for which the metabolism has been well characterized. drugs (16–18). Each patient received the mixture of five drugs during each visit as shown in Table 2. We have shown previously that there is no Patients and Methods interaction between the mixture drugs at the doses given (20). Study Design. Seventeen patients with high-risk, re- Blood samples (20 ml) were obtained at 0, 4, and 8 h, and urine sected melanoma who were scheduled to undertake adjuvant was collected from 0 to8htomeasure the concentration of therapy with high-dose IFN␣-2b participated in this study after probe drugs and/or their metabolites. Plasma harvested by cen- providing written informed consent. Eligibility criteria included trifugation, and urine aliquots were stored frozen at Ϫ20°C until biopsy-proven high-risk melanoma, normal renal (serum creat- analyzed. The probe drugs and their metabolites were analyzed inine Ͻ2.0 mg/dl) and liver function (total bilirubin Ͻ1.5 mg/ by high-performance liquid chromatography for the determina- dl), and no history of allergy to sulfa drugs. All of the subjects tion of caffeine and paraxanthine, (21), 4-hydroxymephenytoin, were instructed to abstain from alcohol, caffeine, barbecued (20), debrisoquine, and 4-hydroxydebrisoquine (18), chlorzoxa- meat, and grapefruit or grapefruit juice consumption for at least zone and 6-hydroxychlorzoxazone (17), dapsone and N- 2 days before each visit (19). Patients with other underlying hydroxydapsone in urine, and dapsone and monoacetyldap- medical diseases taking drugs known to affect the CYP enzyme sone in plasma (22). Probe drug analyses were conducted system that could not be safely interrupted were excluded from such that all of the samples for a given patient were analyzed the study. within the same run so as to minimize within-subject varia- All of the patients received i.v. IFN-␣2b (INTRON A; tion because of analytical procedures. The interday coeffi- Schering Plough, Kenilworth, NJ) at a dose of 20 million units cients of variation for each of these assays were Ͻ15%. In (MU)/m2/day for 5 days/week ϫ 4 weeks (induction phase) addition, all of the assay procedures were cross-validated to followed by s.c. IFN-␣2b at a dose of 10 MU/m2/day for 3 ensure that no analytical interference would occur with si- days/week ϫ 48 weeks (maintenance phase). multaneous administration of these drugs. Phenotypic trait CYP Cocktail Studies. CYP enzyme activities were measures (Table 2) that serve as indexes of enzyme activity studied on day Ϫ6 (baseline), day 1 (first dose of i.v. IFN␣-2b), were calculated from the quantified drugs and metabolites as day 26 (last dose of i.v. IFN␣-2b), and day 52 (end of 1 month described previously (20).

Table 3 Inhibition of P450 isozyme activity, by time Mean change from baseline P450 isozyme Individual P Mean activity at baseline Day 1 Day 26 Day 52 CYP1A2 0.0004 0.99 Ϫ0.18a Ϫ0.53b Ϫ0.45b CYP2C19 0.001 127.22 ϩ2.71 Ϫ47.09b Ϫ29.53b DH 0.002 0.68 Ϫ0.00047 Ϫ0.15b Ϫ0.08c CYP2D6 0.004 0.59 Ϫ0.06b Ϫ0.15b Ϫ0.10a NAT 0.1 0.34 Ϫ0.0089 Ϫ0.0090 ϩ0.020 CYP2E1 0.2 1.22 Ϫ0.17 Ϫ0.03 Ϫ0.19 a ϭ 0.001 Յ P Ͻ 0.005. b ϭ P Ͻ 0.001. c ϭ 0.005 Յ P Ͻ 0.02.

IFN Pharmacokinetics Study. The pharmacokinetics of sion and correlation and to incidence of adverse events using IFN were determined in each subject on study days 1, 26, and logistic regression. 52. Blood samples were obtained before IFN administration and at 0.25, 0.5, 1, 2, 4, 8, 12, and 24 h after i.v. administration, or 4, 8, 10, 12, 16, and 24 h after s.c./i.v. administration. After Results clotting, the samples were centrifuged, and the resulting serum CYP Enzyme Activity. Study drugs were well tolerated was stored at Ϫ20°C until analyzed. IFN␣-2b concentrations by all of the subjects participating in this study. The summary were measured as described previously (23) using an electro- results of phenotypic measures for each probe drug determined chemiluminescence immunoassay with a limit of quantitation of after simultaneous administration of the five drug mixture at 4 IU/ml. baseline (day Ϫ6), day 1 (first dose of i.v. IFN␣-2b), day 26 Pharmacokinetic Data Analysis. Pharmacokinetic pa- (last dose of i.v. IFN␣-2b, and day 52 (end of one month of s.c. rameter estimates were calculated by noncompartmental analy- IFN␣-2b) are shown in Table 3. The omnibus P for testing all of sis using WinNonlin version 3.1 (Pharsight Corp., Mountain the enzyme profiles was P ϭ 0.002, and Ps for individual View, CA). The Cmax and Tmax were determined by visual isozymes are listed in the second column of Table 3 (rows are inspection of the data. The AUC was estimated by the linear ordered by statistical significance of individual isozyme pro- trapezoidal rule with extrapolation to infinity. files). As the individual time point comparisons in Table 3 Statistical Analysis. Changes in individual CYP enzyme indicate, activity of two CYP enzymes, 1A2 and 2D6, were activity profiles were examined using multivariate permutation found to be significantly inhibited immediately after the first methods. To control for multiple comparisons, an omnibus test IFN dose, whereas significant inhibition of CYP2C19 and di- was performed at level ␣ϭ0.05 to test the global null hypoth- hydrodial was first detected at day 26. Significant inhibition was esis that none of the enzyme activity estimates varied at any not found for either N-acetyltransferase or CYP2E1. Figs. 1 and subsequent time point from its baseline activity level. CYP 2 depict isozyme activity levels of each patient during the i.v. enzyme activity estimates were randomly permuted under the phase. The median percentage of changes of the trait measures null hypothesis in a multivariate Wilcoxon signed-rank test. observed during high-dose i.v. and s.c. IFN␣-2b treatment After rejection of the null hypothesis, Ps for null hypotheses of phases are shown in Fig. 3. no change in activity for individual enzymes were examined. On IFN Pharmacokinetics. Data from 15 patients were rejection of an individual null hypothesis, Ps were computed for available for pharmacokinetic analysis after i.v. IFN adminis- comparing enzyme activity at subsequent time points to activity tration and from 12 patients after s.c. administration. The Cmax at baseline. Changes from baseline at time points found to differ values [median (range)] were 1575 IU/ml (1,200–3,230 IU/ml) significantly were related to incidence of adverse events. Ad- and 2630 IU/ml (1,090–4,020 IU/ml) on study days 1 and 26, verse events with onset dates between day 7 and day 26 of i.v. respectively, and were 149 IU/ml (4–16 IU/ml) after s.c. ad- IFN administration were attributed to the i.v. phase of treatment; ministration on day 52. The AUC [median (range)] values were adverse events occurring between day 7 of the s.c. phase and the 2,469 IU ϫ h/ml (1,463–6,537), 7,247 IU ϫ h/ml (2,122– off-study date were attributed to the s.c. phase of treatment. The 17,262), and 2,254 IU ϫ h/ml (1,098–3,025) on days 1, 26, and following adverse events were included if deemed to be possibly 52, respectively. The Tmax after SC administration was ob- treatment-related: (a) all of the grade 2 or higher adverse events served at 8 h (range, 4–16). IFN␣-2b exposure as measured by in the gastrointestinal, cardiovascular, hematological, and neu- AUC was positively correlated with the percentage of inhibition rological systems; (b) all of the instances of fatigue (a neuro- of CYP1A2 at day 26, the last day of i.v. IFN␣-2b administra- logical event) of grade 1 or higher; (c) all of the instances of tion (r ϭ 0.68; P ϭ 0.01). No other enzyme inhibition data were neutropenia (a hematologic event) of grade 1 or higher; and (d) found to correlate with either AUC or Cmax. Neither AUC nor all of the flu-like symptoms of grade 1 or higher. Logistic Cmax was found to be associated with any type of adverse event regression analysis was used to relate incidence of adverse during i.v. or s.c. phase of IFN␣-2b administration. events to magnitude of CYP enzyme inhibition. IFN Cmax and Adverse Events and CYP Enzyme Activity. Adverse AUC data were related to CYP enzyme inhibition using regres- events during i.v. and s.c. phases of HDI were observed,

Fig. 1 Individual and summary boxplot data of caffeine (CYP1A2), debrisoquine (CYP2D6), and dapsone hydroxylation during IFN␣-2b treatment.

and the association with CYP activities was assessed. The subtype of toxicity, derived from logistic regression models number of patients who experienced each type of adverse with binary response variables indicating adverse events of event during the i.v. and s.c. phases is listed in Table 4. The each type, are presented in Table 4. Identical conclusions number of patients who experienced at least one adverse were also reached using absolute change from baseline as the event in a particular organ is separately recorded during i.v. covariate. For neurological toxicities, the estimated relative and s.c. phases (n ϭ 17). A single patient could have adverse risk for each 1% increase in inhibition of CYP1A2 is 1.1. The events of more than one subtype. Binary indicators for estimated relative risk for flu-like symptoms in relation to each type of adverse event were analyzed separately in rela- CYP2D6 inhibition is 1.1 for each 1% increase in CYP2D6 tion to each CYP enzyme that was found to exhibit a signif- inhibition. The estimated relative risk for fever in relation to icant level of inhibition (CYP1A2, 2D6/IV, and 2C19/SC). CYP2D6 inhibition is 1.2 for each 1% increase in CYP2D6 The Ps for the association of particular isozymes and each inhibition.

Fig. 2 Individual and summary box plot data of chlorzoxazone (CYP2E1) and mephenytoin (CYP2C19) activity during IFN␣-2b treatment.

survival benefit, administered at maximal tolerable dosages i.v. for 4 weeks and s.c. for 48 weeks. This has posed new challenges for the patient on HDI therapy and the treating physician because of the toxicities of HDI. We report here the first comprehensive in vivo data showing that HDI impairs CYP- mediated metabolism in a differential fashion in high-risk resected melanoma patients. The observed effects on CYP ranged from no change (CYP2E1) to substantial impairment (CYP1A2), and we have also shown an association between the magnitude of inhibition and the occurrence of adverse effects including fever and neurological effects. These observations are an important platform for improved understanding of adverse effects that are associated with HDI in patients with melanoma and may allow for the avoidance of toxicity in patients with reduced levels of CYP function on HDI in the future. Previous studies evaluating the effect of IFN treatment on drug metabolism have been conducted with the doses used to treat hepatitis (i.e., 3–6 million units three times per week) and Fig. 3 Summary of median percentage of decrease from baseline of in some cases after only a single dose. Even at these relatively individual CYP enzyme activities during HDI treatment. lower doses, it was shown that IFN treatment adversely affects drug metabolism. For example, Williams et al. (13) evaluated theophylline clearance in 5 patients with chronic active hepatitis B and 4 normal healthy volunteers. Theophylline clearance was Discussion assessed 1–2 weeks before and 20 h after a single i.m. injection It has been known for many years that viral infection, of IFN␣. Inhibition of theophylline clearance ranged from 33 to influenza vaccination, Bacillus Calmette-Gue´rin inoculation, 81% and was observed in 8 of the 9 subjects. Theophylline and fever are associated with impaired CYP-mediated metabo- clearance was also decreased in 7 patients with hepatitis C who lism of drugs such as theophylline, phenytoin, and aminopyrine were treated with IFN-␤ for 8 weeks (14). The disposition of (24–26). Such impairment of drug metabolism has been attrib- antipyrine, a probe of hepatic oxidative metabolism that is uted in part to release of endogenous IFNs (9–11, 27) and metabolized by multiple CYP enzymes, was evaluated in pa- partially corroborated in animal models. High-dose IFN␣2b tients receiving IFN with the results generally showing a de- therapy was approved 6 years ago by the United States Food and crease in antipyrine clearance of ϳ20% (range 5 to 47; Refs. 28, Drug Administration and regulatory authorities worldwide as 29) Theophylline, and to a lesser extent antipyrine, are predom- the first systemic adjuvant intervention capable of significant inately metabolized by CYP1A2; data on other CYP enzymes

Table 4 Adverse events by body system and relation to P450 isozyme inhibition IV Phase SQ Phase % CYP1A2 % CYP2D6 % CYP2C19 %DH System Subtype # Pts inhibition inhibition # Pts. inhibition inhibition Gastrointestinal 1 0.87 – 2 0.42 0.33 Nausea 0 ––0 –– Anorexia 0 ––1 0.22 0.18 Weight loss 0 ––2 0.42 0.33 Cardiovascular 0 ––0 –– Neurologic 13 0.026 0.14 3 0.99 0.26 Depression 1 ––1 0.89 0.89 Fatigue 12 0.083 0.074 2 0.90 0.11 Hematologic 12 0.080 0.20 3 0.30 0.97 Neutropenia 4 0.16 0.90 1 0.86 0.73 Thrombocytopenia 0 ––0 –– Anemia 2 0.43 0.022 0 –– Flu-like Symptoms 11 1.00 0.05 2 0.49 0.70 Fever 7 0.93 0.0088 1 0.56 0.63 Chills/rigor 6 0.32 0.87 0 ––

are lacking. Our in vivo data clearly show that IFN␣-2b therapy Table 5 Selected listing of clinically relevant drugs metabolized by has a differential effect on CYP-mediated metabolism. Activi- CYP enzymes ties of CYP1A2, CYP2D6, and CYP2C19 were decreased, CYP1A2 CYP2C19 CYP2D6 whereas CYP2E1 was not changed. The hydroxylation of dap- Amitriptyline Clomipramine Amitriptyline sone was also significantly decreased but more difficult to Clomipramine Diazepam Clomipramine interpret because multiple CYP enzymes including CYP2C8, Clozapine Imipramine Codeine CYP2C9, CYP2E1, and CYP3A4 contribute to the metabolism Imipramine Lansoprazole Desipramine Propanolol Omeprazole Doxepin of dapsone (30, 31). The clinical implication of our findings is Tacrine Phenytoin Fluoxetine that the clearance of drugs predominately metabolized by Theophylline Propanolol Haloperidol CYP1A2, CYP2D6, and CYP2C19 enzymes would be de- Warfarin Tolbutamide Imipramine creased, which may result in increased drug exposure and, Venlafaxine Metoprolol Nortriptyline therefore, increased potential for adverse effects. Thus, the Paroxetine drug-cytokine interaction observed during IFN␣-2b therapy Perphenazine might have significant clinical consequences. CYP1A2, Propanolol CYP2D6, and CYP2C19 are involved in the metabolism of Risperidone sedatives, narcotics, psychotherapeutics, antiepileptics, beta Thioridazine Timolol blockers, proton pump inhibitors, anesthetics, bronchodilators, Tramadol and antibiotics (11–14, 32–35). Therefore, caution is indicated when such agents of which the metabolism is inhibited by IFN are being used during and after HDI therapy. Selected represen- tatives of the classes of drugs metabolized by these CYP en- inhibition and IFN-induced adverse effects may form a new zymes are shown in Table 5. basis of the underlying pathophysiology of IFN␣-2b-induced The mechanism(s) by which IFN␣-2b inhibits hepatic CYP adverse effects. The associations of CYP2D6 inhibition with is not definitively known (36, 37). It has been suggested that flu-like symptoms, and CYP1A2 with neuropsychiatric symp- inhibition of CYP by IFN may require production of free radi- toms and fatigue are significant and demand additional study. cals by xanthine oxidase (38). More recently, Ghezzi et al. (39) The sample size for analyzing associations between enzyme provided additional evidence for this hypothesis by observing inhibition and incidence of adverse events is small, but not that free radical scavengers (N-acetylcystein and ␣-tocopherol) inordinately so for an exploratory study, because inhibition of a or xanthine oxidase inhibitors (allopurinol) can protect against single P450 isozyme was the only predictor variable used in IFN␣-2b-mediated loss of CYP activity. Because IFN␣-2b mod- each data analysis. Because only 2 patients experienced anemia ulates production of cytokines, it is reasonable to presume during the i.v. phase, the possible association in Table 4 between cytokine-mediated inhibition of the CYP enzyme system. We anemia and CYP2D6 inhibition in the i.v. phase may be a are not able to differentiate between direct and indirect IFN␣-2b statistical artifact. The 2 patients with anemia during the i.v. effects in the current study; however, earlier studies failed to phase had two of the four highest values for CYP2D6 inhibition. demonstrate a role for the cytokines tumor necrosis factor, IL-1, But for hypothesis generation the number of adverse events may IL-6, and nitric oxide in IFN␣-2b-mediated CYP enzyme inhi- be sufficient for both the neurological system and flu-like symp- bition (40, 41), and it may be therefore presumed that the effects toms. In addition, the link between CYP inhibition and fever is observed in this study are more likely direct effects of IFN␣-2b. plausible because it has been shown that CYP inhibitors aug- The significant association between the degree of CYP ment and CYP inducers attenuate lipopolysaccharide- and IL-

1-induced fever in mice (42, 43). The mechanism by which this ants of F1 hybrid lymphoma cells: Evidence consistent with trans gene occurs is thought to involve alternative pathways of arachidonic regulation. J. Immunol., 138: 2734–2738, 1987. acid metabolism. Cyclooxygenase enzymes metabolize arachi- 11. Del Prete, S. A., Maurer, L. H., O’Donnell, J., Forcier, R. J., and donic acid to proinflammatory eicosanoids (e.g., prostaglandins, LeMarbre, P. Combination chemotherapy with cisplatin, carmustine, dacarbazine and tamoxifen in malignant melanoma. Cancer Treat. Rep., thromboxanes, and so forth), whereas CYP enzymes produce 68: 1403–1405, 1984. anti-inflammatory epoxyeicosatrienoic acids and mono- 12. Jonkman, J. H. G., Nicholson, K. G., Farrow, P. R., Eckert, M., hydroxyeicosatetraenoic acids. The observed association be- Grasmeijer, G., Oosterhuis, B., De Noord, O. E., and Guentert, T. W. tween CYP1A2 inhibition and neurological toxicities may be Effects of a-interferon on theophylline pharmacokinetics and metabo- related to the altered metabolism of endogenous neurotransmit- lism. Br. J. Clin. Pharm., 27: 795–802, 1989. ters and neurohormones. CYP enzymes have been implicated in 13. Williams, S. J., Baird-Lambert, J. A., and Farrell, G. C. Inhibition both the formation (44) and metabolism (45, 46) of dopamine, of theophylline metabolism by interferon. Lancet, 2: 939–941, 1987. suggesting that modulation of CYP enzymes may alter neuro- 14. Okuno, H., Takasu, M., Kano, H., Seki, T., Shiozaki, Y., and Inoue, transmission. This is a reasonable hypothesis but warrants ad- K. Depression of drug-metabolizing activity in the human liver by interferon-b. Hepatology, 17: 65–69, 1993. ditional study. 15. Guerry, D., IV, Alexander, M. A., Herlyn, M. F., Zehngebot, L. M., Knowledge of the substrates, inhibitors, and inducers of Mitchell, K. F., Zmijewski, C. M., and Lusk, E. J. HLA-DR histocom- CYP enzymes is critical to the avoidance of drug-drug interac- patibility leukocyte antigens permit cultured human melanoma cells tions and unwitting exacerbation of toxicities associated with from early but not advanced disease to stimulate autologous lympho- IFN␣-2b therapy. A better understanding of the regulation of cytes. J. Clin. Investig., 73: 267–271, 1984. CYP enzymes and their interactions with cytokines such as 16. Streetman, D. S., Bertino, J. S., Jr., and Nafziger, A. N. Phenotyping IFN␣-2b may permit avoidance of toxicity. More importantly, a of drug-metabolizing enzymes in adults: a review of in-vivo cytochrome P450 phenotyping probes. Pharmacogenetics, 10(3): 187–216, 2000. better understanding of the role of the CYP enzyme system in 17. Frye, R. F., and Stiff, D. D. Determination of chlorzoxazone and terms of both exogenous and endogenous mediators of neuro- 6-hydroxychlorzoxazone in human plasma and urine by high-perfor- transmission will ultimately permit more rational therapies to be mance liquid chromatography. J. Chromatogr. B Biomed. Appl., 686: developed. 291–296, 1996. 18. Frye, R. F., and Branch, R. A. Improved high performance liquid chromatographic determination of debrisoquine and 4-hydroxydebriso- References quine in human urine following direct injection. J. Chromatogr. B 1. Kirkwood, J. M., Ernstoff, M. S., Guiliano, A., Gams, R., Robinson, Biomed. Appl., 677: 178–182, 1996. W. A., Costanzi, J., Pouillart, P., Speyer, J., Grimm, M., and Spiegel, R. 19. Santoli, D., Yang, Y-C., Clark, S. C., Kreider, B. L., Caracciolo, D., (Letter to the editor) Interferon a-2a and dacarbazine in melanoma. and Rovera, G. Synergistic and antagonistic effects of recombinant J. Natl. Cancer Inst., 82: 1062–1063, 1990. human interleukin (IL) 3. IL-1 ␣, granulocyte and macrophage colony 2. Marincola, F. M., White, D. E., Wise, A. P., and Rosenberg, S. A. stimulating factors (G-CSF and M-CSF) on the growth of GM-CSF- Combination therapy with interferon alfa-2a and interleukin-2 for the dependent leukemic cell lines. J. Immunol., 139: 3348–3354, 1987. treatment of metastic cancer. J. Clin. Oncol., 13: 1110–1122, 1995. 20. Frye, R. F., Matzke, G. R., Adedoyin, A., Porter, J. A., and Branch, 3. Legha, S. S., Ring, S., Bedikian, A., Plager, C., Eton, O., Buzaid, R. A. Validation of the five-drug “Pittsburgh cocktail” approach for A. C., and Papadopoulos, N. Treatment of metastatic melanoma with assessment of selective regulation of drug-metabolizing enzymes. Clin. combined chemotherapy containing cisplatin, vinblastine and dacarba- Pharmacol. Ther., 62: 365–376, 1997. zine (CVD) and biotherapy using interleukin-2 and interferon-␣. Ann. 21. Frye, R. F., Stiff, D. D., and Branch, R. A. A sensitive method for Oncol., 7: 827–835, 1996. the simultaneous determination of caffeine and its dimethylxanthine 4. Kirkwood, J. M., Strawderman, M. H., Ernstoff, M. S., Smith, T. J., metabolites in human plasma: Application to CYP1A2 phenotyping. J. Borden, E. C., and Blum, R. H. Interferon alfa-2b adjuvant therapy of Liq. Chromatogr., 21: 1161–1171, 1998. high-risk resected cutaneous melanoma: The Eastern Cooperative On- 22. May, D. G., Porter, J. A., Uetrecht, J. P., Wilkinson, G. R., and cology Group Trial EST 1684. J. Clin. Oncol., 14: 7–17, 1996. Branch, R. A. The contribution of N-hydroxylation and acetylation to 5. Kirkwood, J. M., Ibrahim, J. G., Sondak, V. K., Richards, J., Fla- dapsone pharmacokinetics in normal subjects. Clin. Pharmacol. Ther., herty, L. E., Ernstoff, M. S., Smith, T. J., Rao, U. N. M., Steele, M., and 48: 619–627, 1990. Blum, R. H. High- and low-dose interferon ␣-2b in high-risk melanoma: 23. Jen, J. F., Glue, P., Ezzel, F., Chung, C., Gupta, S. K., Jacobs, S., first analysis of intergroup trial E1690/S9111/C9190. J. Clin. Oncol., and Hajian, G. Population pharmacokinetic analysis of pegylated inter- 18: 2444–2458, 2000. feron alfa-2b and interferon alfa-2b in patients with chronic hepatitis C. 6. Cole, B. F., Gelber, R. D., Kirkwood, J. M., Goldhirsch, A., Barylak, Clin. Pharmacol. Ther., 69: 407–421, 2001. E., and Borden, E. A quality-of-life-adjusted survival analysis of inter- 24. Deuchars, K. L., and Ling, V. P-Glycoprotein and multidrug resist- feron alfa-2b adjuvant treatment for high-risk resected cutaneous mel- ance in cancer chemotherapy. Semin. Oncol., 16: 156–165, 1989. anoma: An Eastern Cooperative Oncology Group Study (E1684). 25. Houghton, A., Taormina, M. C., Ikeda, H., Watanabe, T., Oettgen, J. Clin. Oncol., 14: 2666–2673, 1996. H. F., and Old, L. J. Serological survey of normal humans for natural 7. Quesada, J. R., Talpaz, M., Rios, A., Kurzrock, R., and Gutterman, antibody to cell surface antigens of melanoma. Proc. Natl. Acad. Sci. J. U. Clinical toxicity of interferons in cancer patients: a review. J. Clin. USA, 77: 4260–4264, 1980. Oncol., 4: 234–243, 1986. 26. Irie, R. F., Jones, P. C., Morton, D. L., and Sidell, N. In vitro 8. Zarbo, R. J., Regezi, J. A., Lloyd, R. V., Crissman, J. D., and production of human antibody to a tumour-associated foetal antigen. Batsakis, J. G. HLA-DR antigens in normal, inflammatory, and neo- Br. J. Cancer, 44: 262–266, 1981. plastic salivary glands. Oral Surg. Oral Med. Oral Pathol., 64: 577–584, 27. Irie, R. F., Sze, L. L., and Saxton, R. E. Human antibody to OFA-I, 1987. a tumor antigen, produced in vitro by Epstein-Barr virus transformed 9. Morein, B. Immunopotentiating complexes. Acta Trop., 44(Suppl. human B-lymphoid cell lines. Proc. Natl. Acad. Sci. USA, 79: 5666– 12): 98–103, 1987. 5670, 1982. 10. Rembecki, R. M., Bennett, M., Kumar, V., and Potter, T. A. 28. Brockmeyer, N. H., Barthel, B., Mertins, L., and Goos, M. Changes Expression of hemopoietic histocompatibilty antigens on H-2-loss vari- of antipyrine pharmacokinetics during influenza and after administration

of interferon-␣ and -␤. Int. J. Clin. Pharmacol. Ther., 36: 309–311, 39. Ghezzi, P., Bianchi, M., Gianera, L., Landolfo, S., and Salmona, M. 1998. Role of reactive oxygen intermediates in the interfon-mediated depres- 29. Williams, S. J., and Farrell, G. C. Inhibition of antipyrine metabo- sion of hepatic drug metabolism and protective effect of N-acetylcys- lism by interferon. Br. J. Clin. Pharmacol., 22: 610–612, 1986. teine in mice. Cancer Res., 45: 3444–3447, 1985. 30. Mitra, A. K., Thummel, K. E., Kalhorn, T. F., Kharasch, E. D., 40. Gerdes, J., Lemke, H., Baisch, H., Wacker, H-H., Schwab, U., and Unadkat, J. D., and Slattery, J. T. Metabolism of dapsone to its hydrox- Stein, H. Cell cycle analysis of a cell proliferation-associated human ylamine by CYP2E1 in vitro and in vivo. Clin. Pharmacol. Ther., 58: nuclear antigen defined by the monoclonal antibody Ki-67. J. Immunol., 556–566, 1995. 133: 1710–1715, 1984. 31. Winter, H. R., Wang, Y., and Unadkat, J. D. CYP2C8/9 mediate 41. Urban, J. L., Kripke, M. L., and Schreiber, H. Stepwise immuno- dapsone N-hydroxylation at clinical concentrations of dapsone. Drug logic selection of antigenic variants during tumor growth. J. Immunol., Metab. Dispos., 28: 865–868, 2000. 137: 3036–3041, 1986. 32. Fossati, G., Taramelli, D., Balsari, A., Bogdanovich, A., Andreola, 42. Kozak, W., Archuleta, I., Mayfield, K. P., Kozak, A., Rudolph, K., G., and Parmiani, G. Primary but not metastatic human melanomas and Kluger, M. J. Inhibitors of alternative pathways of arachidonate expressing DR antigens stimulate autologous lymphocytes. Int. J. Can- metabolism differentially affect fever in mice. Am. J. Physiol., 275: cer, 33: 591–597, 1984. R1031–R1040, 1998. 33. Foung, S. K. H., Perkins, S., Raubitschek, A., Larrick, J., Lizak, G., Fishwald, D., Engelman, E. G., and Grumet, F. C. Rescue of human 43. Kozak, W., Kluger, M. J., Kozak, A., Wachulec, M., and Dokladny, monoclonal antibody production from an EBV-transformed B cell line K. Role of cytochrome P-450 in endogenous antipyresis. Am. J. Physiol. by fusion to a human-mouse hybridoma. J. Immunol. Methods, 70: Regul. Integr. Comp. Physiol., 279: R455–R460, 2000. 83–90, 1984. 44. Hiroi, T., Imaoka, S., and Funae, Y. Dopamine formation from 34. Rendic, S., and Di Carlo, F. I. Human cytochrome P450 enzymes: tyramine by CYP2D6. Biochem. Biophys. Res. Commun., 249: 838– a status report summarizing their reactions, substrates, inducers, and 843, 1998. inhibitors. Drug Metab. Rev. 29: 413–580, 1997. 45. Nissbrandt, H., Bergquist, F., Jonason, J., and Engberg, G. Inhibi- 35. Sarkar, M. A., and Jackson, B. J. Theophylline N-demethylations as tion of cytochrome P450 2E1 induces an increase in extracellular probes for P4501A1 and P4501A2. Drug Metab. Dispos., 22: 827–834, dopamine in rat substantia nigra: a new metabolic pathway? Synapse, 1994. 40: 294–301, 2001. 36. Morgan, E. T. Regulation of cytochrome p450 by inflammatory 46. Thompson, C. M., Capdevila, J. H., and Strobel, H. W. Recombi- mediators: why and how? Drug Metab. Dispos., 29: 207–212, 2001. nant cytochrome P450 2D18 metabolism of dopamine and arachidonic 37. Renton, K. W. Alteration of drug biotransformation and elimination acid. J. Pharmacol. Exp. Ther., 294: 1120–1130, 2000. during infection and inflammation. Pharmacol. Ther. 92: 147–163, 2001. 47. Wedlund, P. J., Aslanian, W. A., Jacqz, E., McAllister, C. B., 38. Moochhala, S. M., and Renton, K. W. A role for xanthine oxidase Wilkinson, G. R., and Branch, R. A. Mephenytoin hydroxylation defi- in the loss cytochrome P-450 evoked by interferon. Can. J. Surg., 69: ciency in Caucasians: frequency of a new oxidative drug metabolism 944–950, 1991. polymorphism. Clin. Pharmacol. Ther., 36: 773–780, 1984.

Downloaded from clincancerres.aacrjournals.org on October 2, 2021. © 2002 American Association for Cancer Research. Differential Effect of IFNα-2b on the Cytochrome P450 Enzyme System: A Potential Basis of IFN Toxicity and Its Modulation by Other Drugs

Mohammed Islam, Reginald F. Frye, Thomas J. Richards, et al.

Clin Cancer Res 2002;8:2480-2487.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/8/8/2480

Cited articles This article cites 46 articles, 16 of which you can access for free at: http://clincancerres.aacrjournals.org/content/8/8/2480.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/8/8/2480.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/8/8/2480. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.