SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. REFERENCES

Objective

Drug–drug interactions can limit the safety of colchicine for treating rheumatic diseases. Seven separate drug–drug interaction (DDI) studies were performed to elucidate the in vivo effects of concomitant treatment with colchicine and known inhibitors of cytochrome P450 3A4 (CYP3A4)/P-glycoprotein (cyclosporine, ketoconazole, ritonavir, clarithromycin, azithromycin, verapamil ER [extended release]), and diltiazem ER) on the pharmacokinetics of colchicine. The objective was to develop colchicine-dosing algorithms with improved safety.

Methods

All studies were open-label, non-randomized, single-center, one-sequence, two-period DDI experiments, using two 0.6-mg doses of colchicine, separated by a minimum 14-day washout period, followed by administration of the approved on-label regimen of known CYP3A4/P-glycoprotein inhibitors. Plasma concentrations of colchicine, but not the reference CYP3A4/P-glycoprotein inhibitors, were determined, and the pharmacokinetic parameters were calculated.

Results

The ratios of the maximum concentration and area under the curve from time 0 to infinity for colchicine plus CYP3A4/P-glycoprotein inhibitors versus colchicine alone were >125% across all studies, with the exception of studies involving azithromycin. Significant DDIs were present when single doses of colchicine were coadministered with most of the selected CYP3A4/P-glycoprotein inhibitors. Recommended colchicine dose reductions of 33–66% for the treatment of acute gout and 50–75% for prophylaxis were calculated for concomitant therapy with each agent, with the exception of no dose adjustment when colchicine is used in combination with azithromycin.

Conclusion

These studies provide quantitative evidence regarding drug interactions and necessary adjustments in the dose of colchicine if colchicine treatment is continued during therapy with multiple CYP3A4/P-glycoprotein inhibitors. We demonstrated the need for specific reductions in the dose of colchicine when it is used in combination with 2 broadly prescribed calcium channel blockers (verapamil ER and diltiazem ER) and that the dose of colchicine does not need to be adjusted when it is used in combination with azithromycin.

Despite the widespread use of colchicine for the treatment and prophylaxis of gout flares (1–5) and the treatment of familial Mediterranean fever (FMF) (6, 7), the pharmacokinetics of colchicine have not been extensively characterized. The importance of understanding colchicine drug–drug interactions (DDIs) is exemplified when one considers the clinical profile of the average patient with gout. Many patients with gout are obese and have multiple comorbid conditions, including hypertension, diabetes, metabolic syndrome, and renal insufficiency, that also require medication (8, 9). This situation may bring about a heightened potential for DDIs that increase colchicine drug exposure and subsequently increase the risk of colchicine toxicities (10–20).

In addition to the lack of understanding of colchicine pharmacokinetics and DDIs, colchicine metabolism has received little attention. Colchicine is reported to be metabolized by cytochrome P450 3A4 (CYP3A4) (21) and is also a substrate for the P-glycoprotein transporter (also known as multidrug transporter 1 and ATP-binding cassette B1 transporter). P-glycoprotein is associated with drug efflux from cells, and certain P-glycoprotein genetic variants are linked with instances of “drug resistance” in FMF (22, 23). Thus, in theory, systemic concentrations of colchicine may be altered when it is coadministered with inhibitors of CYP3A4 and/or P-glycoprotein (24, 25).

Analysis of the literature and the US Food and Drug Administration (FDA) Adverse Event Reporting System database has revealed apparent safety issues when colchicine is administered indiscriminately or concurrently with P-glycoprotein or CYP3A4 inhibitors, possibly resulting in life-threatening conditions and an increased number of serious adverse events (13). The fact that P-glycoprotein and CYP3A4 genetic loci are not far removed from each other (on chromosome 7q21.1 and chromosome 7q22.1, respectively) and could be coordinately regulated (26, 27) further complicates the understanding of the relative contributions of these proteins in colchicine metabolism. For example, induction of the pregnane X receptor (PXR) ligand, coded by NR1I2 (23) located at 3q12–q13.3 (28), increases the expression of both P-glycoprotein and CYP3A4, whereas down-regulation of PXR decreases the expression of both P-glycoprotein and CYP3A4 (29). P-glycoprotein and CYP3A4 exhibit similar regional distribution in the intestinal tract and liver (30–32). Likewise, substrates of P-glycoprotein are often, but not always, substrates of CYP3A4 and vice versa. Multiple drugs and xenobiotics coinduce both P-glycoprotein and CYP3A4, but not always in a predictable manner (32). Due to similar tissue expression profiles of CYP3A4 and P-glycoprotein, it currently is not possible to clarify the relative importance of each protein in the in vivo absorption, distribution, metabolism, and excretion of drugs such as the dual CYP3A4 and P-glycoprotein substrate colchicine (32).

Although there has been concern about colchicine and drug interactions, no published study has characterized the interaction between commonly used medications and colchicine. The series of studies presented here was designed to characterize the specific effects of P-glycoprotein or CYP3A4 inhibition on the pharmacokinetics of single-dose colchicine, with the objective of developing evidence-based colchicine dosing algorithms with improved safety.

Seven separate DDI studies were performed with single-dose colchicine to elucidate the in vivo effect of administering colchicine concomitantly with known inhibitors of CYP3A4/P-glycoprotein. Cyclosporine was assessed as a comparator, because its use has been associated with marked hyperuricemia and is linked with gout in recipients of major organ transplants. Cyclosporine and colchicine DDIs also can precipitate the rapid onset of colchicine-induced neuromyopathy (33). The macrolide antibiotic drug class (which includes clarithromycin and azithromycin) was examined, because this class of drugs is frequently prescribed. Moreover, since 1991, at least 60 deaths have been linked to colchicine in patients receiving concomitant clarithromycin (13), which is well-established as a strong inhibitor of CYP3A4 and P-glycoprotein. Two widely prescribed calcium channel blockers, diltiazem ER (extended release) and verapamil ER, were examined, as were ketoconazole and ritonavir, as examples of potent inhibitors of both CYP3A4 and P-glycoprotein (34). The cumulative results of this series of studies provide the first evidence-based treatment guidelines for colchicine dosing in the presence of CYP3A4/P-glycoprotein inhibitors.

SUBJECTS AND METHODS

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. REFERENCES

Eligibility criteria and clinical and laboratory evaluations.

Study eligibility criteria included the following: healthy, nonsmoking, 18–45 years of age, body mass index (BMI) 18–32 kg/m2, and no concomitant use of medications or other products that might interfere with the interpretation of the pharmacokinetic results of these studies. The general health of the subjects was confirmed by medical history, physical examination, clinical laboratory testing, and measurement of vital signs. Study flow and subject disposition are shown in Table 1.

Table 1. Study flow and disposition*
 Administration of CYP34A/P-gp inhibitorColchicine with CYP34A/P-gp inhibitorSubject disposition/demographics (sex, male/female; race, white/black/other)Reason for withdrawal
  • *

    In all studies, colchicine alone was administered as a single dose, with pharmacokinetics (PK) analysis performed on days 1–5, followed by a washout period from day 5 to day 14. All cytochrome P450 3A4/P-glycoprotein (CYP34A/P-gp) inhibitors were administered orally.

  • Twenty-four subjects were enrolled in each study.

Cyclosporine100 mg, day 15Single dose, PK analysis on days 15–18Completed, n = 23; PK samples, n = 23; sex, n = 10/14; race, n = 23/1/0Withdrawn by investigator due to out-of-range laboratory results, n = 1
Clarithromycin250 mg twice daily, days 22–29Single dose, PK analysis on days 29–33Completed, n = 23; PK samples, n = 23; sex, n = 11/13; race, n = 24/0/0Failed to take study drug as outpatient, n = 1
Azithromycin500 mg, day 15; 250 mg/day, days 16–19Single dose, PK analysis on days 19–22Completed, n = 21; PK samples, n = 21; sex, n = 10/14; race, n = 19/3/2Withdrew consent (personal reasons, n = 2; vomiting, n = 1)
Diltiazem ER240 mg, days 15–21Single dose, PK analysis on days 21–24Completed, n = 20; PK samples, n = 20; sex, n = 15/9; race, n = 19/3/2Failed to take study drug as outpatient, n = 2; vomiting, n= 1; withdrew consent for personal reasons, n = 1
Verapamil ER240 mg, days 15–19Single dose, PK analysis on days 19–22Completed, n = 24; PK samples, n = 24; sex, n = 6/18; race, n = 21/0/3Not applicable
Ketoconazole200 mg twice daily, days 15–19Single dose, PK analysis on days 19–22Completed, n = 24; PK samples, n = 24; sex, n = 7/17; race, n = 21/0/3Not applicable
Ritonavir100 mg twice daily, days 15–19Single dose, PK analysis on days 19–22Completed, n = 17; PK samples, n = 18; sex, n = 14/10; race, n = 22/2/0Failed inclusion criteria, n = 6; lost to followup, n = 1

DDI studies.

Cyclosporine.

In the DDI study with cyclosporine, colchicine (0.6 mg) was administered on day 1 and day 15, and cyclosporine (100 mg once) was administered on day 15.

Macrolide antibiotics (clarithromycin and azithromycin).

In the study involving clarithromycin, colchicine (0.6 mg) was administered before and after a 7-day regimen of clarithromycin 250 mg (administered twice daily on days 22–29). In the azithromycin pharmacokinetics study, azithromycin 500 mg (2 × 250 mg) was administered on day 15, and 250 mg (once daily) was administered on days 16–19.

Calcium channel blockers (diltiazem ER and verapamil ER).

In the study involving diltiazem ER, colchicine (0.6 mg) was administered on days 1 and 21, and diltiazem ER (240 mg once daily) was administered on days 15–21 (total of 7 doses). In the DDI study with verapamil ER, colchicine (0.6 mg) was administered on days 1 and 19, and verapamil ER (240 mg once daily) was administered on days 15–19.

Ketoconazole or ritonavir.

In the 2 DDI studies involving either ketoconazole or ritonavir, colchicine (0.6 mg) was administered on days 1 and 19, and ketoconazole (200 mg twice daily) or ritonavir (100 mg twice daily) was administered on days 15–19.

All 7 studies were open-label, non-randomized, single-center, one-sequence, two-period DDI studies. For the drugs other than colchicine, the doses were chosen to reflect those typically prescribed to treat common conditions in clinical practice. Subjects received two distinct 0.6-mg doses of colchicine (URL Pharma), separated by a minimum 14-day washout period. Each study protocol received ethics committee approval (PRACS Institute, Ltd./Cetero Research institutional review board). All subjects provided written informed consent prior to participation in the study, which was conducted in accordance with the US Code of Federal Regulations and International Conference on Harmonisation Guidelines for Good Clinical Practice and adhered to the ethics principles of the Declaration of Helsinki. Each of the studies was performed at a single study center (PRACS Institute, Ltd./Cetero Research).

In each of the studies, subjects checked into the clinical investigation facility on day –1. A single dose of colchicine was administered on the morning of day 1 following an overnight fast for at least 10 hours. Subjects remained confined for 24–48 hours (as defined by the individual protocol) and then returned daily on a nonconfined basis to provide blood samples for the pharmacokinetics studies.

Pharmacokinetics evaluations.

At each of the specified sampling times, a 6-ml blood sample was collected from each subject by direct venipuncture. Samples were collected immediately prior to the administration of each dose of colchicine, and at 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 8, 12, 24, 36, 48, 72, and 96 hours post-dose. Samples were centrifuged, and plasma was transferred to polypropylene tubes and stored at −20°C or colder until the time of sample analysis. Plasma concentrations of colchicine, but not the reference CYP3A4/P-glycoprotein inhibitors, were determined, and pharmacokinetic parameters were calculated from these data.

To determine the plasma concentrations of colchicine, extracts were analyzed by liquid chromatography–tandem mass spectrometry (API 5000 LC/MS/MS system with a TurbolonSpray source). Plasma concentrations of colchicine were ascertained by comparison with a standard curve (range 0.20–40 ng/ml). Standards and quality control samples were incorporated into each batch analysis, and the validity of each batch run was determined by assessing the accuracy of the standards and quality control samples. All samples for each subject were analyzed together. The lower limit of quantification for colchicine was 0.2 ng/ml.

The calculated pharmacokinetic parameters were as follows: maximum concentration (Cmax), observed time to Cmax (Tmax), area under the curve (AUC) from time 0 to the last measurable concentration (AUC0–t), AUC from time 0 to infinity (AUC0–∞), elimination rate constant (Kel), apparent volume of distribution (Varea/F), oral clearance (CL/F), and terminal elimination half-life (T1/2). Pharmacokinetic parameters from day 1 (colchicine alone) were compared with those obtained after administration of the second dose of colchicine combined with the drug under investigation for its potential to interact with colchicine. The Cmax and AUC0–t values are reported as ln-transformed geometric mean data, and T1/2 and Kel values are reported as nontransformed data using arithmetic means, according to FDA guidance (35).

RESULTS

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. REFERENCES

Characteristics of the subjects.

Of the 337 subjects who were screened, 168 were enrolled in the studies (24 subjects per study) and treated between November 2007 and September 2008. The mean age of the subjects was 26 years. Subject disposition within the study groups is outlined in Table 1.

Pharmacokinetics evaluations and adverse events.

The pharmacokinetic parameters evaluated in each of the DDI studies are summarized in Table 2. Adverse events are shown in Table 3. Colchicine was rapidly absorbed, regardless of the dosing condition. Across studies, plasma concentrations displayed a bimodal distribution following administration of a single dose; levels were below the lower limit of quantification 24–36 hours post-dose in most subjects. The geometric mean ratios of Cmax and AUC0–∞ for the test (colchicine plus CYP3A4/P-glycoprotein inhibitors) versus reference treatments (colchicine alone) across all studies were >125%, indicating that significant DDIs are present when colchicine is coadministered with selected CYP3A4/P-glycoprotein inhibitors (cyclosporine, macrolide antibiotics [clarithromycin and azithromycin], calcium channel blockers [diltiazem ER and verapamil ER], ketoconazole, and ritonavir). The geometric mean (with upper and lower bounds of 90% confidence interval) concentrations of colchicine, alone or following coadministration with the CYP3A4/P-glycoprotein inhibitors, are shown in Figure 1. The results of individual studies are presented below.

Table 2. Pharmacokinetic parameters of single-dose colchicine, alone and combined with a CYP34A/P-gp inhibitor*
Concomitant drug/ FDA classification/ parameterColchicine aloneColchicine plus concomitant drug% ratio (90% CI)
  • *

    Except where indicated otherwise, values are the arithmetic mean. CYP34A/P-gp = cytochrome P450 3A4/P-glycoprotein; FDA = US Food and Drug Administration; 90% CI = 90% confidence interval; Cmax = maximum concentration; AUC0–t = area under the curve from time 0 to the last measurable concentration; Kel = elimination rate constant; T1/2 = terminal elimination half-life; CL/F = oral clearance.

Cyclosporine/strong P-gp inhibitor   
 Cmax, ng/ml2.728.82324.17 (292.32–356.01)
 AUC0–t, ng/hour/ml12.5539.83317.48 (291.79–343.17)
 Kel0.1470.03826.07 (9.14–43)
 T1/2, hours6.7720.65104.43 (88–120.85)
 CL/F, liters/hour48.2413.4227.82 (14.78–40.87)
Clarithromycin/strong CYP3A4 inhibitor   
 Cmax, ng/ml2.848.44297.49 (277.65–317.33)
 AUC0–t, ng/hour/ml12.3741.95339.21 (314.64–363.78)
 Kel0.13240.029622.35 (5.28–39.43)
 T1/2, hours8.8930.31340.97 (277.7–404.24)
 CL/F, liters/hour46.812.0
Ketoconazole/strong CYP3A4 inhibitor   
 Cmax, ng/ml2.785.27189.52 (176.37–202.67)
 AUC0–t, ng/hour/ml11.9934.38286.75 (265.75–307.85)
 Kel0.14910.33222.29 (8.05–36.52)
 T1/2, hours6.2826.06415.24 (339.4–491.09)
 CL/F, liters/hour49.314.830.01 (17.81–42.22)
Ritonavir/strong CYP3A4 inhibitor   
 Cmax, ng/ml1.874.99267.08 (239.71–294.45)
 AUC0–t, ng/hour/ml8.4129.05345.32 (304.35–386.29)
 Kel0.16660.488131.83 (117.25–146.41)
 T1/2, hours5.1517.41338.42 (274.48–402.36)
 CL/F, liters/hour67.9318.5927.37 (13–41.74)
Verapamil ER/moderate CYP3A4 inhibitor   
 Cmax, ng/ml2.973.85129.72 (115.29–149.88)
 AUC0–t, ng/hour/ml13.0924.64199.29 (174.69–201.88)
 Kel0.14090.048034.05 (20.13–47.96)
 T1/2, hours4.317.17274.99 (239.97–310.02)
 CL/F, liters/hour43.9321.0147.81 (37.87–57.76)
Diltiazem ER/moderate CYP34A inhibitor   
 Cmax, ng/ml2.172.8129.03 (108.22–149.84)
 AUC0–t, ng/hour/ml10.0412.73176.67 (146.49–206.84)
 Kel0.15890.083852.76 (38.24–67.27)
 T1/2, hours5.5112.5226.66 (176.7–276.63)
 CL/F, liters/hour58.8834.758.94 (45.81–72.06)
Azithromycin/weak CYP34A inhibitor   
 Cmax, ng/ml2.743.05111.5 (94.01–128.99)
 AUC0–t, ng/hour/ml11.9817.16143.3 (124.85–161.75)
 Kel0.1470.142697.02 (79.03–115)
 T1/2, hours6.076.71110.44 (79.82–141.02)
 CL/F, liters/hour50.2435.0169.69 (57.58–81.8)
Table 3. Overall treatment-emergent adverse events (AEs)*
AE/MedDRA SOC/preferred termColchicine/ritonavirColchicine/ketoconazoleColchicine/azithromycinColchicine/clarithromycinColchicine/cyclosporineColchicine/diltiazemColchicine/verapamil
  • *

    Values are the number (%). Each study group comprised 24 subjects. SOC = system organ class.

Total subjects with AEs overall5 (20.8)10 (41.7)11 (45.8)14 (58.3)6 (25.0)14 (58.3)14 (58.3)
Cardiac disorders       
 Cardiac flutter001 (4.2)0000
 Palpitations001 (4.2)0000
Ear/labyrinth disorders       
 Hypoacusis0001 (4.2)000
Eye disorders       
 Dry eye000001 (4.2)0
 Eye irritation000001 (4.2)0
Gastrointestinal disorder       
 Upper abdominal pain1 (4.2)03 (12.5)1 (4.2)1 (4.2)01 (4.2)
 Constipation0000001 (4.2)
 Diarrhea002 (8.4)2 (8.4)001 (4.2)
 Dyspepsia000001 (4.2)0
 Nausea2 (8.4)1 (4.2)2 (8.4)4 (16.7)1 (4.2)3 (12.5)1 (4.2)
 Stomach discomfort001 (4.2)0000
 Vomiting1 (4.2)02 (8.4)001 (4.2)0
General disorders/administration site conditions       
 Chest pain0001 (4.2)01 (4.2)0
 Feeling hot0000001 (4.2)
 Vessel puncture site hematoma0001 (4.2)000
Infections and infestations       
 Bacteriuria00001 (4.2)00
 Gastroenteritis, viral1 (4.2)000000
Injury, poisoning/procedure complications       
 Contusion001 (4.2)0001 (4.2)
 Skin laceration0001 (4.2)000
Musculoskeletal and connective tissue disorders       
 Back pain001 (4.2)0001 (4.2)
 Musculoskeletal pain000001 (4.2)0
 Musculoskeletal stiffness001 (4.2)0000
 Pain in extremity0003 (12.5)000
Nervous system disorders       
 Dizziness02 (8.4)1 (4.2)1 (4.2)1 (4.2)1 (4.2)2 (8.4)
 Dysgeusia1 (4.2)000000
 Headache1 (4.2)7 (29.2)5 (20.8)7 (29.2)1 (4.2)10 (41.7)12 (50.0)
 Lethargy000001 (4.2)0
 Paresthesia001 (4.2)0000
 Sinus congestion0000001 (4.2)
 Somnolence001 (4.2)0000
 Syncope0002 (8.4)000
 Blurred vision0002 (8.4)000
Psychiatric disorders       
 Insomnia001 (4.2)0000
 Pollakiuria001 (4.2)0000
Reproductive system       
 Dysmenorrhea1 (4.2)02 (8.4)0000
 Polymenorrhea01 (4.2)00000
thumbnail image

Figure 1. Relative effect of cytochrome P450 3A4/P-glycoprotein (CYP3A4) inhibitors on colchicine pharmacokinetics, grouped according to the US Food and Drug Administration classification. Bars show the geometric means with upper and lower bounds of the 90% confidence interval (90% CI). Cmax = maximum concentration; AUC = area under the curve.

Download figure to PowerPoint

Effect of cyclosporine on the pharmacokinetics of colchicine.

The 100-mg capsule of cyclosporine was the highest-strength capsule form available, and because of its extensive side effect profile, a single dose was given (36). Cyclosporine, a strong CYP3A4/P-glycoprotein inhibitor, markedly increased colchicine total exposure, half-life, and peak concentration when it was coadministered with colchicine. The mean peak colchicine concentration increased by 270%, and the mean AUC0–t value increased by ∼260% when cyclosporine was coadministered with colchicine compared with administration of colchicine alone. The total apparent oral clearance decreased by 72% when colchicine was coadministered with cyclosporine compared with administration of colchicine alone (13.4 liters/hour versus 48.2 liters/hour).

Effect of macrolide antibiotics (clarithromycin and azithromycin).

The regimen of clarithromycin (a potent CYP3A4 inhibitor) at a dosage of 250 mg twice daily for 7 consecutive days was consistent with approved labeling and was calculated to achieve steady-state blood levels (37). When colchicine was coadministered with steady-state concentrations of clarithromycin, the mean maximum concentration of colchicine was ∼230% higher compared with administration of colchicine alone, and the mean total colchicine exposure (AUC0–t) increased by ∼280%. The total apparent oral clearance of colchicine was decreased by 75% when it was coadministered with clarithromycin compared with administration of colchicine alone (12.0 liters/hour versus 46.8 liters/hour). The terminal elimination half-life was 9 hours alone versus ∼30 hours when colchicine was administered with clarithromycin.

In the analysis of the pharmacokinetics of azithromycin, the regimen of 500 mg (2 × 250 mg) administered on day 15, followed by 250 mg daily for 4 days (days 16–19, for a total of 5 days) is the on-label dosage for most indications and is calculated to achieve study-state blood levels (38). Azithromycin, which is known to be a weak CYP3A4 inhibitor, minimally increased colchicine exposure when the 2 drugs were administered concomitantly. The mean maximum concentration of colchicine following cotreatment with multiple-dose azithromycin increased by ∼21% compared with administration of colchicine alone; similarly, mean total exposure was 57% higher. The observed time to maximum concentration (Tmax) was not affected by coadministration of azithromycin (1.5 hour and 1 hour with and without azithromycin, respectively). The total apparent oral clearance decreased by 30% when azithromycin was coadministered with colchicine compared with colchicine administered alone (35.0 liters/hour versus 50.2 liters/hour). The terminal elimination half-life of colchicine alone was ∼6.1 hours, compared with 6.7 hours when it was coadministered with azithromycin.

Effect of calcium channel blockers (diltiazem ER and verapamil ER).

Both diltiazem ER and verapamil ER inhibit CYP3A4. The approved on-label regimen of diltiazem ER (240 mg/day) was administered for 7 days, a period of time that is of adequate duration to achieve steady-state conditions (39). The mean maximum concentration of colchicine was minimally higher (44%) following cotreatment with colchicine and diltiazem ER compared with colchicine administered alone; the mean total colchicine exposure increased by ∼93%. There was no significant effect on the observed time to maximum concentration (1.5 hour and 1 hour with and without diltiazem ER, respectively). The total apparent oral clearance decreased by ∼40% when diltiazem ER was coadministered with colchicine compared with colchicine administered alone (34.7 liters/hour and 58.9 liters/hour, respectively). The terminal elimination half-life for colchicine was ∼6 hours, compared with ∼12 hours when it was coadministered with diltiazem ER. These data suggest that a minimal drug interaction occurs when diltiazem ER and colchicine are coadministered, and that, in general, a dose adjustment for colchicine is not required.

The effects of verapamil ER, a moderate inhibitor of CYP3A4 and P-glycoprotein, on colchicine pharmacokinetics were also examined. When colchicine was coadministered with multiple-dose verapamil ER, mean colchicine concentrations increased by ∼40% compared with administration of colchicine alone as a single dose; the mean total colchicine exposure increased by ∼103%. The observed time to maximum concentration was not affected by coadministration with verapamil ER (1.0 hours with and without verapamil ER coadministration). The total apparent oral clearance decreased by 52% when colchicine was coadministered with verapamil ER as compared with the total apparent oral clearance when colchicine was administered alone (21.0 liters/hour versus 43.9 liters/hour). The terminal elimination half-life of colchicine alone was 4.3 hours, versus ∼17 hours when it was coadministered with verapamil ER.

Effect of ketoconazole on the pharmacokinetics of colchicine.

The regimen of 200 mg of ketoconazole administered twice daily for 5 consecutive days was calculated to be of sufficient duration to achieve steady-state ketoconazole conditions according to approved ketoconazole labeling (40). Ketoconazole, a known strong CYP3A4/P-glycoprotein inhibitor, appreciably increased colchicine exposure when it was administered concomitantly with colchicine. The mean maximum concentration of colchicine was ∼100% higher when colchicine was coadministered with ketoconazole compared with colchicine administered alone. Total colchicine exposure increased by ∼210% when colchicine was coadministered with ketoconazole compared with administration of colchicine alone. The observed time to maximum concentration was not affected by coadministration with ketoconazole (1.0 hour with and without ketoconazole coadministration). The total apparent oral clearance decreased by 70% when the 2 agents were coadministered compared with administration of colchicine alone. The terminal elimination half-life for colchicine alone was ∼6.3 hours, compared with 26 hours when colchicine was coadministered with ketoconazole.

Effect of ritonavir on the pharmacokinetics of colchicine.

The widely prescribed boosting regimen of ritonavir, 100 mg twice daily, was administered for 5 consecutive days; this regimen is calculated to be of sufficient duration to achieve steady-state ritonavir concentrations (41). When colchicine and ritonavir (steady-state concentrations) were coadministered, total colchicine concentrations increased significantly. Specifically, the mean maximum concentration of colchicine was ∼185% higher, and the mean total colchicine exposure was ∼290% higher when colchicine was combined with ritonavir as compared with colchicine administered alone. The total apparent oral clearance decreased by 70% when colchicine was coadministered with ritonavir compared with colchicine administered alone (18.6 liters/hour versus 67.9 liters/hour). The terminal elimination half-life of colchicine was ∼5 hours, compared with ∼17 hours when it was coadministered with ritonavir.

DISCUSSION

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. REFERENCES

There are numerous reports in the literature of significant or lethal drug interactions between colchicine and inhibitors of CYP3A4/P-glycoprotein (10–20). What remains unknown from many of these reports is precisely which concurrent medication(s) promoted colchicine toxicity, and what role was played by metabolism-driven (CYP3A4) or transporter-driven (P-glycoprotein) DDIs. The present studies are the first to systematically examine the pharmacokinetics of various inhibitors of CYP3A4/P-glycoprotein and calculate their effects on colchicine metabolism. We utilized the preferred crossover design to assess changes in colchicine exposure (Cmax and AUC) in the same subject and under the same study conditions and assay methodology. These DDI studies form the foundation for the first evidence-based dosing recommendations and adjustments of colchicine dosing in the presence of CYP3A4/P-glycoprotein inhibitors (Table 4). The process used in the construction of this algorithm was based on pharmacokinetics principles. Prior studies showed divergent pharmacokinetic values for colchicine, most likely due to insensitive assays. The first step in the construction of this algorithm was to develop an extremely sensitive assay for the detection of colchicine in plasma (down to 0.20 ng/ml). Next, colchicine pharmacokinetic parameters for single-dose treatment (1), twice-daily dosing (42), and the “low” dose used for the treatment of early acute gout (1, 5) were established. The suggested colchicine dose reductions were calculated to achieve colchicine blood levels in patients receiving concomitant treatment with colchicine and a strong CYP3A4/P-glycoprotein inhibitor similar to those in patients not receiving a strong CYP3A4/P-glycoprotein inhibitor.

Table 4. Colchicine dosing modifications for concomitant therapy with CYP3A4/P-gp inhibitors*
Concomitant drug studied/ FDA classificationFDA recommendationDosing recommendations
Acute gout flareProphylaxis of gout flaresFamilial Mediterranean fever
  • *

    US Food and Drug Administration (FDA) recommendations were based on the current study. CYP34A/P-gp = cytochrome P450 3A4/P-glycoprotein.

  • Use higher dose if preferred practice is to give colchicine 0.6 mg orally twice daily for gout flare prophylaxis, and lower dose if preferred practice is to give colchicine 0.6 mg orally daily for gout flare prophylaxis.

  • Until further studies are performed, tacrolimus should be considered as a potentially strong P-gp inhibitor. The median inhibition concentration values for tacrolimus and cyclosporine are 0.74 μM and 1.3 μM, respectively (34).

  • §

    At this time, the FDA does not provide guidance on the definition of a “strong” P-gp inhibitor.

Cyclosporine/strong P-gp inhibitorRanolazine0.6 mg (1 tablet), 1 dose; dose to be repeated no earlier than 3 days0.3 mg once/day or 0.3 mg every other dayMaximum daily dose of 0.6 mg; may be given as 0.3 mg twice/day
Clarithromycin, ketoconazole, ritonavir/strong CYP34A inhibitors§Atazanavir, darunavir (with ritonavir), indinavir, itraconazole, lopinavir (with ritonavir), nefazodone, nelfinavir, ritonavir, saquinavir, telithromycin, tipranavir (with ritonavir)0.6 mg (1 tablet), 1 dose 0.3 mg twice/day to be repeated no earlier than 3 days0.3 mg once/day or 0.3 mg every other dayMaximum daily dose of 0.6 mg; may be given as 0.3 mg twice/day
Diltiazem, verapamil/moderate CYP34A inhibitorsAmprenavir, aprepitant, erythromycin, fluconazole, fosamprenavir, grapefruit juice1.2 mg (2 tablets), 1 dose; dose to be repeated no earlier than 3 days0.3 mg twice/day (or 0.6 mg once/day) or 0.3 mg once/dayMaximum daily dose of 1.2 mg; may be given as 0.6 mg twice/day
Azithromycin/weak CYP34A inhibitorAzithromycinNo dose reduction required; 1.2 mg (2 tablets), 1 dose, followed by 0.6 mg 1 hour laterNo dose reduction required; 0.6 mg twice/day or 0.6 mg once/dayNo dose reduction required; maximum daily dose of 2.4 mg; may be given as 0.6 mg twice/day

The FDA classification of CYP3A4 inhibitors (34) was used to allow consistent characterization between this series of studies and those that have been completed for other drugs. At the present time, the FDA has not classified P-glycoprotein inhibitors in a manner analogous to that used for the classification of cytochrome P450 enzymes (34); therefore, terminology such as “strong P-glycoprotein inhibitor” is used to characterize compounds such as cyclosporine. This classification approach is meant to be consistent with the description of drug interactions currently appearing in drug package inserts in the US.

The drugs selected for this series of pharmacokinetics studies were chosen for their prevalence of use, including the potential to be prescribed for patients with gout; clarithromycin is noteworthy due to the numerous deaths attributed to the combination of colchicine and this drug. The pharmacokinetics data presented here demonstrate clearly how these fatalities may have occurred with clarithromycin, because this strong inhibitor of CYP3A4 and P-glycoprotein dramatically increased colchicine exposure and prolonged the terminal half-life of colchicine. Importantly, dose adjustment of colchicine was not required when it was used concomitantly with the macrolide antibiotic azithromycin; thus, azithromycin was identified as a potential alternative to clarithromycin in patients requiring colchicine in the presence of an antibiotic with a suitable sensitivity profile.

The dose-adjustment information presented in this report reflects consideration of the interaction with the calcineurin inhibitor cyclosporine, which is important because of the high incidence and severity of cyclosporine-induced hyperuricemia and gout in patients undergoing major organ transplantation. Based on the colchicine dose adjustments that were calculated from these studies, we recommend reducing the dose of colchicine, if clinically required, for acute gout flares by two-thirds and reducing the dose for flare prophylaxis by three-fourths in patients requiring concomitant treatment with a strong P-glycoprotein inhibitor (e.g., cyclosporine). Although the calcineurin tacrolimus (median inhibition concentration [IC50] 0.74 μM) was not studied, we recommend that it should also be considered a strong P-glycoprotein inhibitor, because it is analogous to cyclosporine (IC50 1.3 μM) (34). For patients requiring concomitant treatment with a moderate CYP3A4 inhibitor, such as verapamil ER or diltiazem ER, colchicine dose reductions of one-third for an acute gout flare and one-half for gout flare prophylaxis are recommended. Because nifedipine has not been implicated as a significant inhibitor of P-glycoprotein or CYP3A4, it represents a potentially superior option for concomitant colchicine therapy compared with verapamil ER or diltiazem ER, but this possibility has not yet been tested directly.

The results of the current studies are the foundation of rational dosing guidance to help achieve the optimal therapeutic benefit of colchicine while limiting safety risks with drugs not yet studied for colchicine DDIs. However, grouping drugs into broad classifications for colchicine DDIs is not expected to be completely accurate. There may be variations among individual drugs within a given classification that do not conform to the general attributes of the class. Thus, drugs not specifically examined in this series may interact with colchicine differently from those used for these trials. Another limitation of the current studies is the lack of general consensus in the FDA classification of CYP3A4/P-glycoprotein inhibitors. There are several inconsistencies in this classification scheme. For example, erythromycin is classified as a moderate CYP3A4 inhibitor, but serious adverse events have been linked to the concomitant use of colchicine with erythromycin (10). Conversely, the FDA classifies grapefruit juice as a moderate inhibitor of CYP3A4, but our past studies failed to demonstrate an adverse interaction between colchicine and grapefruit juice (43). It should also be noted that these pharmacokinetics studies were not designed to assess specific relationships between the Cmax and/or AUC and adverse events experienced by healthy volunteers. These studies were, by design, short in duration and too small in terms of the number of subjects to calculate any correlation. Thus, direct statements about colchicine exposure and the incidence of adverse events cannot be made. Also, other DDIs reported in the literature (e.g., with disulfiram) (44) for which the mechanisms of action are not clear still need to be studied.

If the algorithm in Table 4 is used, then no other changes in the dosing of colchicine in clinical practice are required. Patients receiving colchicine for the treatment of acute gout flares should begin colchicine prophylaxis 12 hours after the initial flare. This recommendation is based on the pharmacokinetics data. Twelve hours after colchicine treatment is started for an acute gout flare, colchicine blood levels are ∼1 ng/ml. Similarly, patients receiving colchicine twice daily have the same colchicine blood level (i.e., 1 ng/ml) 12 hours after their previous colchicine dose (42). If the algorithm in Table 4 is used, colchicine blood levels in patients concomitantly receiving inhibitors of CYP3A4/P-glycoprotein will be comparable with those in patients not receiving inhibitors of CYP3A4/P-glycoprotein. Notably, FDA-approved instructions allow for colchicine treatment of an acute gout attack even when the patient is receiving colchicine for gout prophylaxis (45). As with all treatment modalities, the patient's total medical status should be evaluated prior to the administration of any treatments. Given the availability of other effective treatment modalities for acute gout, we advocate considering the use of alternate treatments for acute gout flares in patients who are already receiving established colchicine prophylaxis at an appropriate dose.

An interesting and significant finding that was revealed during early studies of colchicine centered around its biotransformation. Several pharmacokinetics studies previously conducted with various single-dose and multiple-dose colchicine regimens measured plasma concentrations of colchicine and its metabolites in order to assess the predominant uptake and metabolic pathways of colchicine. Of note, measurement of colchicine metabolites (2-demethylcolchicine, 3-demethylcolchicine, and 10-demethylcolchicine) following colchicine administration revealed that this pathway contributes very little to the overall degradation and removal of colchicine from the body. The scant plasma concentrations of the colchicine metabolites identify the importance of the excretion pattern of colchicine from the body. It would appear that the role of P-glycoprotein on absorption, distribution, and excretion of colchicine is more significant than the role of CYP3A4 in the overall metabolism of colchicine. Until a drug is identified as solely a P-glycoprotein or solely a CYP3A4 substrate, the distinction will be purely of academic interest.

Colchicine dosing recommendations for the treatment of acute gout have been based on both the Acute Gout Flare Receiving Colchicine Evaluation (AGREE) trial and pharmacokinetics data (1). The AGREE trial data collectively suggested that the primary therapeutic effect in acute gout flare was linked to achieving a target maximum concentration of ∼6 ng/ml (accomplished with both the low-dose regimen of 1.2 mg colchicine followed by a 0.6-mg dose 1 hour later, as well as by the high-dose colchicine regimen) (1). The AGREE trial data also suggested that colchicine-related adverse events in patients with acute gout flares were more closely linked to drug exposure than to the maximum concentration (1). In this context, when calculating the dosage of colchicine to treat acute gout flares while avoiding DDIs and limiting adverse effects, the goal should be achieving the optimal maximum concentration while reducing exposure.

We conclude that it is of paramount clinical importance to avoid toxicity due to unanticipated increased colchicine exposure from concomitant use of inhibitors of CYP3A4 or P-glycoprotein. Proper dose adjustments of colchicine will aid clinicians in the safe prescribing of colchicine. Avoidance of outdated high-dose colchicine regimens in the treatment of early acute gout flare, especially in the presence of inhibitors of CYP3A4 or P-glycoprotein, provides an additional safety margin (1). Other considerations that must be weighed by health care practitioners when prescribing colchicine for patients who are receiving other medications include the fact that these DDI studies were performed in young healthy individuals with normal renal and hepatic clearance (a standard protocol); the typical patient with gout, as a common example of someone for whom colchicine is prescribed, does not fit this profile. Further colchicine dose reductions should be considered in patients with renal or hepatic impairment (46) and those who are morbidly obese (9). Finally, when clinically appropriate, avoidance of the use of colchicine or the interacting drug is always the best way to prevent a drug interaction.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Terkeltaub had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. DiGiacinto, Kook, Davis.

Acquisition of data. Kook, Davis.

Analysis and interpretation of data. Terkeltaub, Furst, DiGiacinto, Kook, Davis.

ROLE OF THE STUDY SPONSOR

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. REFERENCES

The study sponsor, URL Pharma, designed the pharmacokinetics studies and collected and performed primary analyses of the data. The writing of the manuscript was done in part by the sponsor, with final editing and approval by the first author required. The study sponsor agreed with submission of the manuscript for publication. The content of the submitted manuscript, and publication of the manuscript were contingent upon the approval of the first author, this occurring after the final draft was jointly prepared by the authors and the sponsor and approved by the sponsor.

REFERENCES

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. REFERENCES